Embeddable: Create core dbs in proper subdirectory

gplx.xowa.mediawiki/src/gplx/core/lists/HashByInt.java

@@ -13,3 +13,55 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.core.lists; import gplx.*; import gplx.core.*;
+import gplx.core.primitives.*;
+public class HashByInt {
+	private final    Ordered_hash hash = Ordered_hash_.New();
+	private final    Int_obj_ref tmp_key = Int_obj_ref.New_neg1();
+	public void Clear() {
+		hash.Clear();
+	}
+	public int Len() {
+		return hash.Len();
+	}
+	public Object Get_at_or_null(int idx) {
+		HashByIntItem item = (HashByIntItem)hash.Get_at(idx);
+		return item.val;
+	}
+	public Object Get_by_or_fail(int key) {
+		synchronized (tmp_key) {
+			HashByIntItem item = (HashByIntItem)hash.Get_by_or_fail(tmp_key.Val_(key));
+			return item.val;
+		}
+	}
+	public Object Get_by_or_null(int key) {
+		synchronized (tmp_key) {
+			HashByIntItem item = (HashByIntItem)hash.Get_by(tmp_key.Val_(key));
+			return item == null ? null : item.val;
+		}
+	}
+	public HashByInt Add_as_bry(int key, String val) {return Add(key, Bry_.new_u8(val));}
+	public HashByInt Add(int key, Object val) {
+		HashByIntItem item = new HashByIntItem(key, val);
+		hash.Add(Int_obj_ref.New(key), item);
+		return this;
+	}
+	public HashByInt Clone() {
+		HashByInt rv = new HashByInt();
+
+		int len = hash.Len();
+		for (int i = 0; i < len; i++) {
+			HashByIntItem item = (HashByIntItem)hash.Get_at(i);
+			rv.Add(item.key, item.val);
+		}
+		return rv;
+	}
+}
+class HashByIntItem {
+	public final    int key;
+	public final    Object val;
+	public HashByIntItem(int key, Object val) {
+		this.key = key;
+		this.val = val;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XomwEnv.java

@@ -13,3 +13,40 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.interwiki.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*; import gplx.xowa.mediawiki.includes.media.*;
+import gplx.xowa.mediawiki.includes.site.*;
+import gplx.xowa.mediawiki.languages.*;
+import gplx.xowa.langs.*;
+public class XomwEnv {
+	public byte[] Lang__align_end = Bry_.new_a7("right");
+	public int User__default__thumbsize = 220;
+
+	public int Global__wgSVGMaxSize = 5120;
+	public double Global__wgThumbUpright = .75d;
+	public int[] Global__wgThumbLimits = new int[] {120, 150, 180, 200, 250, 300};
+
+	public XomwEnv(Xol_lang_itm xoLang) {
+		this.language = new XomwLanguage(xoLang);
+		XomwSiteLookup siteLookup = new XomwXowaSiteLookup();
+		XomwInterwikiLookup interwikiLookup = new XomwInterwikiLookupAdapter(siteLookup);
+		byte[][] localInterwikis = new byte[0][]; // TODO.XO: pass in to XomwEnv or retrieve from datastore
+		this.mediaWikiServices = new XomwMediaWikiServices(this, interwikiLookup, language, localInterwikis);
+	}
+
+	public XomwLanguage Language() {return language;} private final    XomwLanguage language;
+	public XomwMagicWordMgr Magic_word_mgr() {return magic_word_mgr;} private final    XomwMagicWordMgr magic_word_mgr = new XomwMagicWordMgr();
+	public XomwMessageMgr Message_mgr() {return message_mgr;} private final    XomwMessageMgr message_mgr = new XomwMessageMgr();
+	public XomwFileFinder File_finder() {return file_finder;} private XomwFileFinder file_finder = new XomwFileFinderNoop();
+	public XomwMediaHandlerFactory MediaHandlerFactory() {return mediaHandlerFactory;} private final    XomwMediaHandlerFactory mediaHandlerFactory = new XomwMediaHandlerFactory();
+	public XomwMediaWikiServices MediaWikiServices() {return mediaWikiServices;} private final    XomwMediaWikiServices mediaWikiServices;
+
+	public XomwEnv File_finder_(XomwFileFinder v) {file_finder = v; return this;} 
+
+	public static XomwEnv NewTest() {return NewTestByApp(Xoa_app_fxt.Make__app__edit());}
+	public static XomwEnv NewTestByApp(Xoae_app app) {
+		return new XomwEnv(Xol_lang_itm_.Lang_en_make(app.Lang_mgr()));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpArray.java

@@ -13,3 +13,33 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+public class XophpArray {
+	public static boolean popBoolOrN(List_adp list)           {return Bool_.Cast(List_adp_.Pop_or(list, false));}
+	public static byte[] popBryOrNull(List_adp list)       {return (byte[])List_adp_.Pop_or(list, null);}
+	public static String[] array_keys_str(Ordered_hash array) {
+		int len = array.Len();
+		String[] rv = new String[len];
+		for (int i = 0; i < len; i++) {
+			 rv[i] = (String)array.Get_at(i);
+		}
+		return rv;
+	}
+	public static byte[][] array_keys_bry(Ordered_hash array) {
+		int len = array.Len();
+		byte[][] rv = new byte[len][];
+		for (int i = 0; i < len; i++) {
+			 rv[i] = (byte[])array.Get_at(i);
+		}
+		return rv;
+	}
+	public static boolean array_key_exists(int key, Ordered_hash array)    {return array.Has(key);}
+	public static boolean array_key_exists(String key, Ordered_hash array) {return array.Has(key);}
+	public static boolean array_key_exists(byte[] key, Ordered_hash array) {return array.Has(key);}
+	public static boolean array_is_empty(Ordered_hash array) {
+		return array.Len() == 0;
+	}
+	public static void unset(Ordered_hash array, Object key) {
+		array.Del(key);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpEncode.java

@@ -13,3 +13,9 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+public class XophpEncode {
+	public static byte[] rawurlencode(byte[] v) {
+		return gplx.langs.htmls.encoders.Gfo_url_encoder_.Php_rawurlencode.Encode(v);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpInvalidArgumentException.java

@@ -13,3 +13,6 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+public class XophpInvalidArgumentException extends Err { 	public XophpInvalidArgumentException(String text) {super(false, "", text, text);}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpMath.java

@@ -13,3 +13,15 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+public class XophpMath {
+	public static double round(double v, int places) {
+		if (places < 0) {	// -1 means round to 10; -2 means round to 100; etc..
+			int factor = (int)Math_.Pow(10, places * -1);
+			return ((int)(Math_.Round(v, 0) / factor)) * factor;	// EX: ((int)Round(123, 0) / 10) * 10: 123 -> 12.3 -> 12 -> 120
+		}
+		else {
+			return Math_.Round(v, places);
+		}
+	}					
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpPreg.java

@@ -13,3 +13,100 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import gplx.core.btries.*; import gplx.core.brys.*;
+import gplx.core.primitives.*;
+public class XophpPreg {
+	public static byte[][] split(Int_list list, byte[] src, int src_bgn, int src_end, byte[] dlm, boolean extend) {
+		// find delimiters
+		int dlm_len = dlm.length;
+		byte dlm_nth = dlm[dlm_len - 1];
+		int i = src_bgn;
+		list.Add(src_bgn);
+		while (true) {
+			if (i == src_end) break;
+			int dlm_end = i + dlm_len;
+			if (dlm_end <= src_end && Bry_.Eq(src, i, dlm_end, dlm)) {
+				if (extend) {
+					dlm_end = Bry_find_.Find_fwd_while(src, i, src_end, dlm_nth);
+				}
+				list.Add(i);
+				list.Add(dlm_end);
+				i = dlm_end;
+			}
+			else
+				i++;
+		}
+		list.Add(src_end);
+
+		// create brys
+		int rv_len = list.Len() - 1;
+		if (rv_len == 1) {
+			list.Clear();
+			return null;
+		}
+		if (list.Get_at(list.Len() - 2) == src_end) {	// if 2nd to last elem == src_end, then last item is Bry_.Empty; ignore it; EX: "a''" -> "a", "''" x> "a", "''", ""
+			rv_len--;
+		}
+		byte[][] rv = new byte[rv_len][];
+		for (i = 0; i < rv_len; i += 2) {
+			rv[i    ] = Bry_.Mid(src, list.Get_at(i + 0), list.Get_at(i + 1));
+			if (i + 1 == rv_len) break;
+			rv[i + 1] = Bry_.Mid(src, list.Get_at(i + 1), list.Get_at(i + 2));
+		}
+		list.Clear();
+		return rv;
+	}
+	public static Object match(Btrie_slim_mgr trie, Btrie_rv trv, byte[] src, int src_bgn, int src_end) {
+		trv.Match_bgn = -1;
+		int cur = src_bgn;
+		while (cur < src_end) {
+			byte b = src[cur];
+			Object o = trie.Match_at_w_b0(trv, b, src, cur, src_end);
+			if (o == null)
+				cur += gplx.core.intls.Utf8_.Len_of_char_by_1st_byte(b);
+			else {
+				trv.Match_bgn = cur;
+				return o;
+			}
+		}
+		return null;
+	}
+	
+	public static void replace(Bry_tmp bry, Bry_bfr tmp, Btrie_slim_mgr find_trie, Btrie_rv trv, byte[] repl_bry) {
+		byte[] src = bry.src;
+		int src_bgn = bry.src_bgn;
+		int src_end = bry.src_end;
+ 
+		int cur = src_bgn;
+		int prv = cur;
+		boolean dirty = false;
+
+		while (true) {
+			// eos
+			if (cur == src_end) {
+				if (dirty) {
+					tmp.Add_mid(src, prv, src_end);
+				}
+				break;
+			}
+
+			byte b = src[cur];
+			Object o = find_trie.Match_at_w_b0(trv, b, src, cur, src_end);
+			if (o == null) {
+				cur += gplx.core.intls.Utf8_.Len_of_char_by_1st_byte(b);
+			}
+			else {
+				dirty = true;
+				tmp.Add_mid(src, prv, cur);
+				tmp.Add(repl_bry);
+				cur = trv.Pos();
+				prv = cur;
+			}
+		}
+
+		if (dirty) {
+			bry.Set_by_bfr(tmp);
+		}
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpPregTest.java

@@ -13,3 +13,19 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XophpPregTest {
+	private final    XophpPregFxt fxt = new XophpPregFxt();
+	@Test  public void Basic()         {fxt.Test_split("a''b''c"          , "''", Bool_.Y, "a", "''", "b", "''", "c");}
+	@Test  public void Extend()        {fxt.Test_split("a'''b'''c"        , "''", Bool_.Y, "a", "'''", "b", "'''", "c");}
+	@Test  public void Eos()           {fxt.Test_split("a''"              , "''", Bool_.Y, "a", "''");}
+}
+class XophpPregFxt {
+	private final    gplx.core.primitives.Int_list rv = new gplx.core.primitives.Int_list();
+	public void Test_split(String src, String dlm, boolean extend, String... expd) {Test_split(src, 0, String_.Len(src), dlm, extend, expd);}
+	public void Test_split(String src, int src_bgn, int src_end, String dlm, boolean extend, String... expd) {
+		byte[][] actl = XophpPreg.split(rv, Bry_.new_u8(src), src_bgn, src_end, Bry_.new_u8(dlm), extend);
+		Gftest.Eq__ary(expd, String_.Ary(actl), "find_failed");
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpString.java

@@ -13,3 +13,149 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import gplx.core.btries.*;
+public class XophpString {
+	public static int strpos(byte[] src, byte find) {return strpos(src, find, 0, src.length);}
+	public static int strpos(byte[] src, byte find, int bgn, int end) {
+		return Bry_find_.Find_fwd(src, find, bgn, end);
+	}
+	public static byte[] substr(byte[] src, int bgn) {return substr(src, bgn, src.length);}
+	public static byte[] substr(byte[] src, int bgn, int len) {
+		int src_len = src.length;
+		if (bgn < 0) bgn = src_len + bgn; // handle negative
+		if (bgn < 0) bgn = 0;	// handle out of bounds; EX: ("a", -1, -1)
+		int end = len < 0 ? src_len + len : bgn + len;
+		if (end > src.length) end = src.length;; // handle out of bounds;
+		return Bry_.Mid(src, bgn, end);
+	}
+	public static byte substr_byte(byte[] src, int bgn) {return substr_byte(src, bgn, src.length);}
+	public static byte substr_byte(byte[] src, int bgn, int len) {
+		int src_len = src.length;
+		if (src_len == 0) return Byte_ascii.Null;
+		if (bgn < 0) bgn = src_len + bgn; // handle negative
+		if (bgn < 0) bgn = 0;	// handle out of bounds; EX: ("a", -1, -1)
+		int end = len < 0 ? src_len + len : bgn + len;
+		if (end > src.length) end = src.length;; // handle out of bounds;
+		return src[bgn];
+	}
+	public static int strspn_fwd__ary(byte[] src, boolean[] find, int bgn, int max, int src_len) {
+		if (max == -1) max = src_len;
+		int rv = 0;
+		for (int i = bgn; i < src_len; i++) {
+			if (find[src[i] & 0xFF] && rv < max) // PATCH.JAVA:need to convert to unsigned byte
+				rv++;
+			else
+				break;
+		}
+		return rv;
+	}
+	public static int strspn_fwd__byte(byte[] src, byte find, int bgn, int max, int src_len) {
+		if (max == -1) max = src_len;
+		int rv = 0;
+		for (int i = bgn; i < src_len; i++) {
+			if (find == src[i] && rv < max) 
+				rv++;
+			else
+				break;
+		}
+		return rv;
+	}
+	public static int strspn_fwd__space_or_tab(byte[] src, int bgn, int max, int src_len) {
+		if (max == -1) max = src_len;
+		int rv = 0;
+		for (int i = bgn; i < src_len; i++) {
+			switch (src[i]) {
+				case Byte_ascii.Space:
+				case Byte_ascii.Tab:
+					if (rv < max) {
+						rv++;
+						continue;
+					}
+					break;
+			}
+			break;
+		}
+		return rv;
+	}
+	public static int strspn_bwd__byte(byte[] src, byte find, int bgn, int max) {
+		if (max == -1) max = Int_.Max_value;
+		int rv = 0;
+		for (int i = bgn - 1; i > -1; i--) {
+			if (find == src[i] && rv < max) 
+				rv++;
+			else
+				break;
+		}
+		return rv;
+	}
+	public static int strspn_bwd__ary(byte[] src, boolean[] find, int bgn, int max) {
+		if (max == -1) max = Int_.Max_value;
+		int rv = 0;
+		for (int i = bgn - 1; i > -1; i--) {
+			if (find[src[i & 0xFF]] && rv < max)  // PATCH.JAVA:need to convert to unsigned byte
+				rv++;
+			else
+				break;
+		}
+		return rv;
+	}
+	public static int strspn_bwd__space_or_tab(byte[] src, int bgn, int max) {
+		if (max == -1) max = Int_.Max_value;
+		int rv = 0;
+		for (int i = bgn - 1; i > -1; i--) {
+			switch (src[i]) {
+				case Byte_ascii.Space:
+				case Byte_ascii.Tab:
+					if (rv < max) {
+						rv++;
+						continue;
+					}
+					break;
+			}
+			break;
+		}
+		return rv;
+	}
+	public static byte[] strtr(byte[] src, Btrie_slim_mgr trie, Bry_bfr tmp, Btrie_rv trv) {
+		boolean dirty = false;
+		int src_bgn = 0;
+		int src_end = src.length;
+		int i = src_bgn;
+
+		while (true) {
+			if (i == src_end) break;
+			byte b = src[i];
+			Object o = trie.Match_at_w_b0(trv, b, src, i, src_end);
+			if (o == null) {
+				if (dirty) {
+					tmp.Add_byte(b);
+				}
+				i++;
+			}
+			else {
+				if (!dirty) {
+					dirty = true;
+					tmp.Add_mid(src, 0, i);
+				}
+				tmp.Add((byte[])o);
+				i = trv.Pos();
+			}
+		}
+		return dirty ? tmp.To_bry_and_clear() : src;
+	}
+	public static byte[] strtr(byte[] src, byte find, byte repl) {
+		return Bry_.Replace(src, 0, src.length, find, repl);
+	}
+	public static byte[] str_replace(byte find, byte repl, byte[] src) {
+		return Bry_.Replace(src, 0, src.length, find, repl);
+	}
+	public static byte[] str_replace(byte[] find, byte[] repl, byte[] src) {
+		return Bry_.Replace(src, find, repl);
+	}
+	public static byte[] strstr(byte[] src, byte[] find) {
+		int pos = Bry_find_.Find_fwd(src, find);
+		return pos == Bry_find_.Not_found ? null : Bry_.Mid(src, pos, src.length);
+	}
+	public static int strlen(byte[] src) {return src.length;}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpStringTest.java

@@ -13,3 +13,73 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import org.junit.*; import gplx.core.tests.*; import gplx.core.btries.*;
+public class XophpStringTest {
+	private final    XophpStringFxt fxt = new XophpStringFxt();
+	@Test  public void Strspn_fwd__byte() {
+		fxt.Test_strspn_fwd__byte("aaaaab", Byte_ascii.Ltr_a, 0, -1, 5);	// basic
+		fxt.Test_strspn_fwd__byte("aaaaab", Byte_ascii.Ltr_a, 1, -1, 4);	// bgn
+		fxt.Test_strspn_fwd__byte("aaaaab", Byte_ascii.Ltr_a, 1,  2, 2);	// max
+	}
+	@Test  public void Strspn_fwd__space_or_tab() {
+		fxt.Test_strspn_fwd__space_or_tab("     a", 0, -1, 5);	// basic
+		fxt.Test_strspn_fwd__space_or_tab("     a", 1, -1, 4);	// bgn
+		fxt.Test_strspn_fwd__space_or_tab("     a", 1,  2, 2);	// max
+	}
+	@Test  public void Strspn_bwd__byte() {
+		fxt.Test_strspn_bwd__byte("aaaaab", Byte_ascii.Ltr_a, 5, -1, 5);	// basic
+		fxt.Test_strspn_bwd__byte("aaaaab", Byte_ascii.Ltr_a, 4, -1, 4);	// bgn
+		fxt.Test_strspn_bwd__byte("aaaaab", Byte_ascii.Ltr_a, 4,  2, 2);	// max
+	}
+	@Test  public void Strspn_bwd__space_or_tab() {
+		fxt.Test_strspn_bwd__space_or_tab("     a", 5, -1, 5);	// basic
+		fxt.Test_strspn_bwd__space_or_tab("     a", 4, -1, 4);	// bgn
+		fxt.Test_strspn_bwd__space_or_tab("     a", 4,  2, 2);	// max
+	}
+	@Test   public void Substr__bgn_is_neg() {
+		fxt.Test_substr("abcde"                   , -1, "e");
+		fxt.Test_substr("abcde"                   , -3, -1, "cd");
+	}
+	@Test   public void Strtr() {
+		fxt.Init_strtr_by_trie("01", "89", "02", "79");
+		fxt.Test_strtr_by_trie("abc"           , "abc");                 // found=none
+		fxt.Test_strtr_by_trie("ab_01_cd"      , "ab_89_cd");            // found=one
+		fxt.Test_strtr_by_trie("ab_01_cd_02_ef", "ab_89_cd_79_ef");      // found=many
+		fxt.Test_strtr_by_trie("01_ab"         , "89_ab");               // BOS
+		fxt.Test_strtr_by_trie("ab_01"         , "ab_89");               // EOS
+	}
+}
+class XophpStringFxt {
+	public void Test_strspn_fwd__byte(String src_str, byte find, int bgn, int max, int expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__int(expd, XophpString.strspn_fwd__byte(src_bry, find, bgn, max, src_bry.length));
+	}
+	public void Test_strspn_fwd__space_or_tab(String src_str, int bgn, int max, int expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__int(expd, XophpString.strspn_fwd__space_or_tab(src_bry, bgn, max, src_bry.length));
+	}
+	public void Test_strspn_bwd__byte(String src_str, byte find, int bgn, int max, int expd) {
+		Gftest.Eq__int(expd, XophpString.strspn_bwd__byte(Bry_.new_u8(src_str), find, bgn, max));
+	}
+	public void Test_strspn_bwd__space_or_tab(String src_str, int bgn, int max, int expd) {
+		Gftest.Eq__int(expd, XophpString.strspn_bwd__space_or_tab(Bry_.new_u8(src_str), bgn, max));
+	}
+	public void Test_substr(String src_str, int bgn, String expd) {Test_substr(src_str, bgn, String_.Len(src_str), expd);}
+	public void Test_substr(String src_str, int bgn, int len, String expd) {
+		Gftest.Eq__str(expd, XophpString.substr(Bry_.new_u8(src_str), bgn, len));
+	}
+	private Btrie_slim_mgr strtr_trie;
+	public void Init_strtr_by_trie(String... kvs) {
+		if (strtr_trie == null) strtr_trie = Btrie_slim_mgr.cs();
+		int len = kvs.length;
+		for (int i = 0; i < len; i += 2) {
+			strtr_trie.Add_str_str(kvs[i], kvs[i + 1]);
+		}
+	}
+	public void Test_strtr_by_trie(String src, String expd) {
+		Bry_bfr tmp = Bry_bfr_.New();
+		Btrie_rv trv = new Btrie_rv();
+		Gftest.Eq__str(expd, XophpString.strtr(Bry_.new_u8(src), strtr_trie, tmp, trv));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpUrl.java

@@ -13,3 +13,29 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+import gplx.core.net.*;
+public class XophpUrl {
+	public static final int 
+	  PHP_URL_SCHEME = 0
+	, PHP_URL_HOST = 3
+	, PHP_URL_PATH = 5
+	;
+	public static String parse_url(String url, int type) {
+		Gfo_url_parser parser = new Gfo_url_parser();
+		Gfo_url url_itm = parser.Parse(Bry_.new_u8(url));
+		switch (type) {
+			case PHP_URL_SCHEME:   return String_.new_u8(url_itm.Protocol_bry());
+			case PHP_URL_HOST:     return String_.new_u8(url_itm.Segs__get_at_1st());
+			case PHP_URL_PATH:
+				Bry_bfr bfr = Bry_bfr_.New();
+				byte[][] segs = url_itm.Segs();
+				int len = segs.length;
+				for (int i = 1; i < len; i++) {
+					bfr.Add_byte_slash().Add(segs[i]);
+				}
+				return bfr.To_str_and_clear();
+			default: throw Err_.new_unhandled_default(type);
+		}
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/XophpUtility.java

@@ -13,3 +13,33 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki; import gplx.*; import gplx.xowa.*;
+public class XophpUtility {
+	public static boolean empty(byte[] v)   {return v == null || v.length == 0;}
+	public static boolean empty(boolean v)     {return v == false;}
+	public static boolean empty(int v)      {return v == 0;}
+	public static boolean isset(byte[] v)   {return v != null;}
+	public static boolean isset(int v)      {return v != NULL_INT;}
+	public static boolean isset(double v)   {return v != NULL_DOUBLE;}
+	public static boolean istrue(int v)     {return v != NULL_INT;}
+	public static boolean isnumeric(byte[] src) {
+		if (src == null) return false;
+		int len = src.length;
+		for (int i = 0; i < len; i++) {
+			byte b = src[i];
+			switch (b) {
+				case Byte_ascii.Num_0: case Byte_ascii.Num_1: case Byte_ascii.Num_2: case Byte_ascii.Num_3: case Byte_ascii.Num_4:
+				case Byte_ascii.Num_5: case Byte_ascii.Num_6: case Byte_ascii.Num_7: case Byte_ascii.Num_8: case Byte_ascii.Num_9:
+					break;
+				default:
+					return false;
+			}
+		}
+		return true;
+	}
+
+	public static boolean is_null(int v) {return v == NULL_INT;}
+	public static final int NULL_INT = Int_.Max_value;
+	public static final double NULL_DOUBLE = Double_.MinValue;
+	public static final    byte[] NULL_BRY = null;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwDefines.java

@@ -13,3 +13,297 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwDefines {
+//		/**
+//		* @defgroup Constants MediaWiki constants
+//		*/
+//
+//		/**@{
+//		* Database related constants
+//		*/
+//		define( 'DBO_DEBUG', 1 );
+//		define( 'DBO_NOBUFFER', 2 );
+//		define( 'DBO_IGNORE', 4 );
+//		define( 'DBO_TRX', 8 ); // automatically start transaction on first query
+//		define( 'DBO_DEFAULT', 16 );
+//		define( 'DBO_PERSISTENT', 32 );
+//		define( 'DBO_SYSDBA', 64 ); // for oracle maintenance
+//		define( 'DBO_DDLMODE', 128 ); // when using schema files: mostly for Oracle
+//		define( 'DBO_SSL', 256 );
+//		define( 'DBO_COMPRESS', 512 );
+//		/**@}*/
+//
+//		/**@{
+//		* Valid database indexes
+//		* Operation-based indexes
+//		*/
+//		define( 'DB_SLAVE', -1 );     # Read from the slave (or only server)
+//		define( 'DB_MASTER', -2 );    # Write to master (or only server)
+//		/**@}*/
+//
+//		# Obsolete aliases
+//		define( 'DB_READ', -1 );
+//		define( 'DB_WRITE', -2 );
+
+	/**@{
+	* Virtual namespaces; don't appear in the page database
+	*/
+	public static final int NS_MEDIA = -2;
+	public static final int NS_SPECIAL = -1;
+	/**@}*/
+
+	/**@{
+	* Real namespaces
+	*
+	* Number 100 and beyond are reserved for custom namespaces;
+	* DO NOT assign standard namespaces at 100 or beyond.
+	* DO NOT Change integer values as they are most probably hardcoded everywhere
+	* see bug #696 which talked about that.
+	*/
+	public static final int NS_MAIN = 0;
+	public static final int NS_TALK = 1;
+	public static final int NS_USER = 2;
+	public static final int NS_USER_TALK = 3;
+	public static final int NS_PROJECT = 4;
+	public static final int NS_PROJECT_TALK = 5;
+	public static final int NS_FILE = 6;
+	public static final int NS_FILE_TALK = 7;
+	public static final int NS_MEDIAWIKI = 8;
+	public static final int NS_MEDIAWIKI_TALK = 9;
+	public static final int NS_TEMPLATE = 10;
+	public static final int NS_TEMPLATE_TALK = 11;
+	public static final int NS_HELP = 12;
+	public static final int NS_HELP_TALK = 13;
+	public static final int NS_CATEGORY = 14;
+	public static final int NS_CATEGORY_TALK = 15;
+
+//		/**
+//		* NS_IMAGE and NS_IMAGE_TALK are the pre-v1.14 names for NS_FILE and
+//		* NS_FILE_TALK respectively, and are kept for compatibility.
+//		*
+//		* When writing code that should be compatible with older MediaWiki
+//		* versions, either stick to the old names or define the new constants
+//		* yourself, if they're not defined already.
+//		*/
+//		define( 'NS_IMAGE', NS_FILE );
+//		define( 'NS_IMAGE_TALK', NS_FILE_TALK );
+//		/**@}*/
+//
+//		/**@{
+//		* Cache type
+//		*/
+//		define( 'CACHE_ANYTHING', -1 );  // Use anything, as long as it works
+//		define( 'CACHE_NONE', 0 );       // Do not cache
+//		define( 'CACHE_DB', 1 );         // Store cache objects in the DB
+//		define( 'CACHE_MEMCACHED', 2 );  // MemCached, must specify servers in $wgMemCacheServers
+//		define( 'CACHE_ACCEL', 3 );      // APC, XCache or WinCache
+//		/**@}*/
+//
+//		/**@{
+//		* Media types.
+//		* This defines constants for the value returned by File::getMediaType()
+//		*/
+//		// unknown format
+//		define( 'MEDIATYPE_UNKNOWN', 'UNKNOWN' );
+//		// some bitmap image or image source (like psd, etc). Can't scale up.
+//		define( 'MEDIATYPE_BITMAP', 'BITMAP' );
+//		// some vector drawing (SVG, WMF, PS, ...) or image source (oo-draw, etc). Can scale up.
+//		define( 'MEDIATYPE_DRAWING', 'DRAWING' );
+//		// simple audio file (ogg, mp3, wav, midi, whatever)
+//		define( 'MEDIATYPE_AUDIO', 'AUDIO' );
+//		// simple video file (ogg, mpg, etc;
+//		// no not include formats here that may contain executable sections or scripts!)
+//		define( 'MEDIATYPE_VIDEO', 'VIDEO' );
+//		// Scriptable Multimedia (flash, advanced video container formats, etc)
+//		define( 'MEDIATYPE_MULTIMEDIA', 'MULTIMEDIA' );
+//		// Office Documents, Spreadsheets (office formats possibly containing apples, scripts, etc)
+//		define( 'MEDIATYPE_OFFICE', 'OFFICE' );
+//		// Plain text (possibly containing program code or scripts)
+//		define( 'MEDIATYPE_TEXT', 'TEXT' );
+//		// binary executable
+//		define( 'MEDIATYPE_EXECUTABLE', 'EXECUTABLE' );
+//		// archive file (zip, tar, etc)
+//		define( 'MEDIATYPE_ARCHIVE', 'ARCHIVE' );
+//		/**@}*/
+//
+//		/**@{
+//		* Antivirus result codes, for use in $wgAntivirusSetup.
+//		*/
+//		define( 'AV_NO_VIRUS', 0 );  # scan ok, no virus found
+//		define( 'AV_VIRUS_FOUND', 1 );  # virus found!
+//		define( 'AV_SCAN_ABORTED', -1 );  # scan aborted, the file is probably immune
+//		define( 'AV_SCAN_FAILED', false );  # scan failed (scanner not found or error in scanner)
+//		/**@}*/
+//
+//		/**@{
+//		* Anti-synchronized flags
+//		* Was used by $wgAntiLockFlags, which was removed with 1.25
+//		* Constants kept to not have warnings when used in LocalSettings
+//		*/
+//		define( 'ALF_PRELOAD_LINKS', 1 ); // unused
+//		define( 'ALF_PRELOAD_EXISTENCE', 2 ); // unused
+//		define( 'ALF_NO_LINK_LOCK', 4 ); // unused
+//		define( 'ALF_NO_BLOCK_LOCK', 8 ); // unused
+//		/**@}*/
+//
+//		/**@{
+//		* Date format selectors; used in user preference storage and by
+//		* Language::date() and co.
+//		*/
+//		define( 'MW_DATE_DEFAULT', 'default' );
+//		define( 'MW_DATE_MDY', 'mdy' );
+//		define( 'MW_DATE_DMY', 'dmy' );
+//		define( 'MW_DATE_YMD', 'ymd' );
+//		define( 'MW_DATE_ISO', 'ISO 8601' );
+//		/**@}*/
+//
+//		/**@{
+//		* RecentChange type identifiers
+//		*/
+//		define( 'RC_EDIT', 0 );
+//		define( 'RC_NEW', 1 );
+//		define( 'RC_LOG', 3 );
+//		define( 'RC_EXTERNAL', 5 );
+//		define( 'RC_CATEGORIZE', 6 );
+//		/**@}*/
+//
+//		/**@{
+//		* Article edit flags
+//		*/
+//		define( 'EDIT_NEW', 1 );
+//		define( 'EDIT_UPDATE', 2 );
+//		define( 'EDIT_MINOR', 4 );
+//		define( 'EDIT_SUPPRESS_RC', 8 );
+//		define( 'EDIT_FORCE_BOT', 16 );
+//		define( 'EDIT_DEFER_UPDATES', 32 ); // Unused since 1.27
+//		define( 'EDIT_AUTOSUMMARY', 64 );
+//		/**@}*/
+//
+//		/**@{
+//		* Flags for Database::makeList()
+//		* These are also available as Database class constants
+//		*/
+//		define( 'LIST_COMMA', 0 );
+//		define( 'LIST_AND', 1 );
+//		define( 'LIST_SET', 2 );
+//		define( 'LIST_NAMES', 3 );
+//		define( 'LIST_OR', 4 );
+//		/**@}*/
+//
+//		/**
+//		* Unicode and normalisation related
+//		*/
+//		require_once __DIR__ . '/compat/normal/UtfNormalDefines.php';
+//
+//		/**@{
+//		* Hook support constants
+//		*/
+//		define( 'MW_SUPPORTS_PARSERFIRSTCALLINIT', 1 );
+//		define( 'MW_SUPPORTS_LOCALISATIONCACHE', 1 );
+//		define( 'MW_SUPPORTS_CONTENTHANDLER', 1 );
+//		define( 'MW_EDITFILTERMERGED_SUPPORTS_API', 1 );
+//		/**@}*/
+//
+//		/** Support for $wgResourceModules */
+//		define( 'MW_SUPPORTS_RESOURCE_MODULES', 1 );
+//
+//		/**@{
+//		* Allowed values for Parser::$mOutputType
+//		* Parameter to Parser::startExternalParse().
+//		* Use of Parser consts is preferred:
+//		* - Parser::OT_HTML
+//		* - Parser::OT_WIKI
+//		* - Parser::OT_PREPROCESS
+//		* - Parser::OT_MSG
+//		* - Parser::OT_PLAIN
+//		*/
+//		define( 'OT_HTML', 1 );
+//		define( 'OT_WIKI', 2 );
+//		define( 'OT_PREPROCESS', 3 );
+//		define( 'OT_MSG', 3 );  // b/c alias for OT_PREPROCESS
+//		define( 'OT_PLAIN', 4 );
+//		/**@}*/
+//
+//		/**@{
+//		* Flags for Parser::setFunctionHook
+//		* Use of Parser consts is preferred:
+//		* - Parser::SFH_NO_HASH
+//		* - Parser::SFH_OBJECT_ARGS
+//		*/
+//		define( 'SFH_NO_HASH', 1 );
+//		define( 'SFH_OBJECT_ARGS', 2 );
+//		/**@}*/
+//
+//		/**@{
+//		* Autopromote conditions (must be here and not in Autopromote.php, so that
+//		* they're loaded for DefaultSettings.php before AutoLoader.php)
+//		*/
+//		define( 'APCOND_EDITCOUNT', 1 );
+//		define( 'APCOND_AGE', 2 );
+//		define( 'APCOND_EMAILCONFIRMED', 3 );
+//		define( 'APCOND_INGROUPS', 4 );
+//		define( 'APCOND_ISIP', 5 );
+//		define( 'APCOND_IPINRANGE', 6 );
+//		define( 'APCOND_AGE_FROM_EDIT', 7 );
+//		define( 'APCOND_BLOCKED', 8 );
+//		define( 'APCOND_ISBOT', 9 );
+//		/**@}*/
+//
+//		/** @{
+//		* Protocol constants for wfExpandUrl()
+//		*/
+//		define( 'PROTO_HTTP', 'http://' );
+//		define( 'PROTO_HTTPS', 'https://' );
+//		define( 'PROTO_RELATIVE', '//' );
+//		define( 'PROTO_CURRENT', null );
+//		define( 'PROTO_CANONICAL', 1 );
+//		define( 'PROTO_INTERNAL', 2 );
+//		/**@}*/
+//
+//		/**@{
+//		* Content model ids, used by Content and ContentHandler.
+//		* These IDs will be exposed in the API and XML dumps.
+//		*
+//		* Extensions that define their own content model IDs should take
+//		* care to avoid conflicts. Using the extension name as a prefix is recommended,
+//		* for example 'myextension-somecontent'.
+//		*/
+//		define( 'CONTENT_MODEL_WIKITEXT', 'wikitext' );
+//		define( 'CONTENT_MODEL_JAVASCRIPT', 'javascript' );
+//		define( 'CONTENT_MODEL_CSS', 'css' );
+//		define( 'CONTENT_MODEL_TEXT', 'text' );
+//		define( 'CONTENT_MODEL_JSON', 'json' );
+//		/**@}*/
+
+	/**@{
+	* Content formats, used by Content and ContentHandler.
+	* These should be MIME types, and will be exposed in the API and XML dumps.
+	*
+	* Extensions are free to use the below formats, or define their own.
+	* It is recommended to stick with the conventions for MIME types.
+	*/
+	// wikitext
+	public static final String CONTENT_FORMAT_WIKITEXT = "text/x-wiki";
+	// for js pages
+	public static final String CONTENT_FORMAT_JAVASCRIPT = "text/javascript";
+	// for css pages
+	public static final String CONTENT_FORMAT_CSS = "text/css";
+	// for future use, e.g. with some plain-html messages.
+	public static final String CONTENT_FORMAT_TEXT = "text/plain";
+	// for future use, e.g. with some plain-html messages.
+	public static final String CONTENT_FORMAT_HTML = "text/html";
+	// for future use with the api and for extensions
+	public static final String CONTENT_FORMAT_SERIALIZED = "application/vnd.php.serialized";
+	// for future use with the api, and for use by extensions
+	public static final String CONTENT_FORMAT_JSON = "application/json";
+	// for future use with the api, and for use by extensions
+	public static final String CONTENT_FORMAT_XML = "application/xml";
+	/**@}*/
+
+//		/**@{
+//		* Max String length for shell invocations; based on binfmts.h
+//		*/
+//		define( 'SHELL_MAX_ARG_STRLEN', '100000' );
+//		/**@}*/
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwHtmlTemp.java

@@ -13,3 +13,9 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import gplx.core.btries.*;
+public class XomwHtmlTemp {
+	public final    Bry_bfr bfr = Bry_bfr_.New();
+	public final    Btrie_rv trv = new Btrie_rv();
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwHtml_expandAttributesTest.java

@@ -13,3 +13,31 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*;
+import gplx.xowa.mediawiki.includes.xohtml.*;
+public class XomwHtml_expandAttributesTest {
+	private final    XomwHtml_expandAttributesFxt fxt = new XomwHtml_expandAttributesFxt();
+	@Test   public void Basic()	{
+		fxt.Test__expand_attributes(" a=\"b\"", "a", "b");
+	}
+	@Test   public void NullVal()	{
+		fxt.Test__expand_attributes("", "a", null);
+	}
+}
+class XomwHtml_expandAttributesFxt {
+	private final    Bry_bfr bfr = Bry_bfr_.New();
+	private final    XomwHtmlTemp temp = new XomwHtmlTemp();
+	public void Test__expand_attributes(String expd, String... kvs) {
+		Xomw_atr_mgr atrs = new Xomw_atr_mgr();
+		int kvs_len = kvs.length;
+		for (int i = 0; i < kvs_len; i += 2) {
+			byte[] key = Bry_.new_a7(kvs[i]);
+			byte[] val = Bry_.new_a7(kvs[i + 1]);
+			Xomw_atr_itm itm = new Xomw_atr_itm(-1, key, val);
+			atrs.Add(itm);
+		}
+		XomwHtml.expandAttributes(bfr, temp, atrs);
+		Gftest.Eq__str(expd, bfr.To_str_and_clear());
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwLinker_NormalizeSubpageLink.java

@@ -13,3 +13,13 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwLinker_NormalizeSubpageLink {
+	public byte[] link;
+	public byte[] text;
+	public XomwLinker_NormalizeSubpageLink Init(byte[] link, byte[] text) {
+		this.link = link;
+		this.text = text;
+		return this;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwLinker_NormalizeSubpageLinkTest.java

@@ -13,3 +13,28 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XomwLinker_NormalizeSubpageLinkTest {
+	private final    XomwLinker_NormalizeSubpageLinkFxt fxt = new XomwLinker_NormalizeSubpageLinkFxt();
+	@Test  public void None()                {fxt.Test__normalize_subpage_link("A/B/C"          , "Z"             , ""    , "Z"          , "");}
+	@Test  public void Hash()                {fxt.Test__normalize_subpage_link("A/B/C"          , "/Y#Z"          , ""    , "A/B/C/Y#Z"  , "/Y#Z");}
+	@Test  public void Slash__basic()        {fxt.Test__normalize_subpage_link("A/B/C"          , "/Z"            , ""    , "A/B/C/Z"    , "/Z");}
+	@Test  public void Slash__slash()        {fxt.Test__normalize_subpage_link("A/B/C"          , "/Z/"           , ""    , "A/B/C/Z"    , "Z");}
+	@Test  public void Dot2__empty()         {fxt.Test__normalize_subpage_link("A/B/C"          , "../"           , ""    , "A/B"        , "");}
+	@Test  public void Dot2__many()          {fxt.Test__normalize_subpage_link("A/B/C"          , "../../Z"       , "z1"  , "A/Z"        , "z1");}
+	@Test  public void Dot2__trailing()      {fxt.Test__normalize_subpage_link("A/B/C"          , "../../Z/"      , ""    , "A/Z"        , "Z");}
+}
+class XomwLinker_NormalizeSubpageLinkFxt {
+	private final    XomwEnv env;
+	private final    XomwLinker mgr = new XomwLinker(new gplx.xowa.mediawiki.includes.linkers.XomwLinkRenderer(new XomwSanitizer()));
+	private final    XomwLinker_NormalizeSubpageLink normalize_subpage_link = new XomwLinker_NormalizeSubpageLink();
+	public XomwLinker_NormalizeSubpageLinkFxt() {
+		this.env = XomwEnv.NewTest();
+	}
+	public void Test__normalize_subpage_link(String page_title_str, String link, String text, String expd_link, String expd_text) {
+		mgr.normalizeSubpageLink(normalize_subpage_link, XomwTitle.newFromText(env, Bry_.new_u8(page_title_str)), Bry_.new_u8(link), Bry_.new_u8(text));
+		Gftest.Eq__str(expd_link, String_.new_u8(normalize_subpage_link.link));
+		Gftest.Eq__str(expd_text, String_.new_u8(normalize_subpage_link.text));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwLinker_SplitTrailTest.java

@@ -13,3 +13,25 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*; import gplx.core.btries.*; import gplx.xowa.mediawiki.includes.parsers.*;
+public class XomwLinker_SplitTrailTest {
+	private final    XomwLinker_SplitTrailFxt fxt = new XomwLinker_SplitTrailFxt();
+	@Test  public void Basic()                {fxt.Test__split_trail("abc def"          , "abc"             , " def");}
+	@Test  public void None()                 {fxt.Test__split_trail(" abc"             , null              , " abc");}
+}
+class XomwLinker_SplitTrailFxt {
+	private final    XomwLinker linker = new XomwLinker(new gplx.xowa.mediawiki.includes.linkers.XomwLinkRenderer(new XomwSanitizer()));
+	private final    Btrie_slim_mgr trie = Btrie_slim_mgr.cs();
+	public XomwLinker_SplitTrailFxt() {
+		String[] ary = new String[] {"a", "b", "c", "d", "e", "f"};
+		for (String itm : ary)
+			trie.Add_str_str(itm, itm);
+		linker.Init_by_wiki(XomwEnv.NewTest(), trie);
+	}
+	public void Test__split_trail(String trail_str, String expd_inside, String expd_trail) {
+		byte[][] split_trail = linker.splitTrail(Bry_.new_u8(trail_str));
+		Gftest.Eq__str(expd_inside, String_.new_u8(split_trail[0]));
+		Gftest.Eq__str(expd_trail , String_.new_u8(split_trail[1]));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMagicWord.java

@@ -13,3 +13,19 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwMagicWord {
+	public boolean case_match;
+	public byte[] name;
+	public XomwMagicWordSynonym[] synonyms;
+	public XomwMagicWord(byte[] name, boolean case_match, byte[][] synonyms_ary) {
+		this.name = name;
+		this.case_match = case_match;
+
+		int synonyms_len = synonyms_ary.length;
+		this.synonyms = new XomwMagicWordSynonym[synonyms_len];
+		for (int i = 0; i < synonyms_len; i++) {
+			synonyms[i] = new XomwMagicWordSynonym(name, case_match, synonyms_ary[i]);
+		}
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMagicWordArray.java

@@ -13,3 +13,362 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import gplx.core.btries.*; import gplx.core.primitives.*;
+public class XomwMagicWordArray {
+	private Btrie_slim_mgr fwd_trie;
+	private Btrie_bwd_mgr  bwd_trie;
+	private final    Btrie_rv trv = new Btrie_rv();
+//		private final    XomwMagicWordMgr magic_word_mgr;
+	public final    byte[][] names;
+
+//		/** @var array */
+//		private hash;
+
+//		private baseRegex;
+
+//		private regex;
+
+	public XomwMagicWordArray(XomwMagicWordMgr magic_word_mgr, byte[][] names) {
+//			this.magic_word_mgr = magic_word_mgr;
+		this.names = names;
+
+		// ASSUME: all magic words in a group have the same case sensitivity
+		for (byte[] name : names) {
+			XomwMagicWord word = magic_word_mgr.Get(name);
+			if (word == null) continue;
+			XomwMagicWordSynonym[] synonyms = word.synonyms;
+			int synonyms_len = synonyms.length;
+			for (int i = 0; i < synonyms_len; i++) {
+				XomwMagicWordSynonym synonym = synonyms[i];
+				switch (synonym.arg1_tid) {
+					case XomwMagicWordSynonym.Arg1__nil:
+					case XomwMagicWordSynonym.Arg1__end:
+						if (fwd_trie == null) fwd_trie = word.case_match ? Btrie_slim_mgr.cs() : Btrie_slim_mgr.ci_u8();
+						fwd_trie.Add_obj(synonym.text_wo_arg1, synonym);
+						break;
+					case XomwMagicWordSynonym.Arg1__bgn:
+						if (bwd_trie == null) bwd_trie = Btrie_bwd_mgr.c__(word.case_match);
+						bwd_trie.Add(synonym.text_wo_arg1, synonym);
+						break;
+					// ignore if mid / mix
+					case XomwMagicWordSynonym.Arg1__mid:
+					case XomwMagicWordSynonym.Arg1__mix:
+						Gfo_usr_dlg_.Instance.Warn_many("", "", "MagicWordArray: unsupported arg_1_tid: tid=~{0}", synonym.arg1_tid);
+						continue;
+				}
+			}
+		}
+	}
+
+//		/**
+//		* Add a magic word by name
+//		*
+//		* @param String name
+//		*/
+//		public function add(name) {
+//			this->names[] = name;
+//			this->hash = this->baseRegex = this->regex = null;
+//		}
+//
+//		/**
+//		* Add a number of magic words by name
+//		*
+//		* @param array names
+//		*/
+//		public function addArray(names) {
+//			this->names = array_merge(this->names, array_values(names));
+//			this->hash = this->baseRegex = this->regex = null;
+//		}
+//
+//		/**
+//		* Get a 2-d hashtable for this array
+//		* @return array
+//		*/
+//		public function getHash() {
+//			if (is_null(this->hash)) {
+//				global wgContLang;
+//				this->hash = [ 0 => [], 1 => [] ];
+//				foreach (this->names as name) {
+//					magic = MagicWord::get(name);
+//					case = intval(magic->isCaseSensitive());
+//					foreach (magic->getSynonyms() as syn) {
+//						if (!case) {
+//							syn = wgContLang->lc(syn);
+//						}
+//						this->hash[case][syn] = name;
+//					}
+//				}
+//			}
+//			return this->hash;
+//		}
+//
+//		/**
+//		* Get the super regex
+//		* @return array
+//		*/
+//		public function getBaseRegex() {
+//			if (is_null(this->baseRegex)) {
+//				this->baseRegex = [ 0 => '', 1 => '' ];
+//				foreach (this->names as name) {
+//					magic = MagicWord::get(name);
+//					case = intval(magic->isCaseSensitive());
+//					foreach (magic->getSynonyms() as i => syn) {
+//						// Group name must start with a non-digit in PCRE 8.34+
+//						it = strtr(i, '0123456789', 'abcdefghij');
+//						group = "(?P<{it}_{name}>" . preg_quote(syn, '/') . ')';
+//						if (this->baseRegex[case] === '') {
+//							this->baseRegex[case] = group;
+//						} else {
+//							this->baseRegex[case] .= '|' . group;
+//						}
+//					}
+//				}
+//			}
+//			return this->baseRegex;
+//		}
+//
+//		/**
+//		* Get an unanchored regex that does not match parameters
+//		* @return array
+//		*/
+//		public function getRegex() {
+//			if (is_null(this->regex)) {
+//				super = this->getBaseRegex();
+//				this->regex = [ '', '' ];
+//				if (this->baseRegex[0] !== '') {
+//					this->regex[0] = "/{super[0]}/iuS";
+//				}
+//				if (this->baseRegex[1] !== '') {
+//					this->regex[1] = "/{super[1]}/S";
+//				}
+//			}
+//			return this->regex;
+//		}
+//
+//		/**
+//		* Get a regex for matching variables with parameters
+//		*
+//		* @return String
+//		*/
+//		public function getVariableRegex() {
+//			return str_replace("\\1", "(.*?)", this->getRegex());
+//		}
+//
+//		/**
+//		* Get a regex anchored to the start of the String that does not match parameters
+//		*
+//		* @return array
+//		*/
+//		public function getRegexStart() {
+//			super = this->getBaseRegex();
+//			newRegex = [ '', '' ];
+//			if (super[0] !== '') {
+//				newRegex[0] = "/^(?:{super[0]})/iuS";
+//			}
+//			if (super[1] !== '') {
+//				newRegex[1] = "/^(?:{super[1]})/S";
+//			}
+//			return newRegex;
+//		}
+//
+//		/**
+//		* Get an anchored regex for matching variables with parameters
+//		*
+//		* @return array
+//		*/
+//		public function getVariableStartToEndRegex() {
+//			super = this->getBaseRegex();
+//			newRegex = [ '', '' ];
+//			if (super[0] !== '') {
+//				newRegex[0] = str_replace("\\1", "(.*?)", "/^(?:{super[0]})/iuS");
+//			}
+//			if (super[1] !== '') {
+//				newRegex[1] = str_replace("\\1", "(.*?)", "/^(?:{super[1]})/S");
+//			}
+//			return newRegex;
+//		}
+//
+//		/**
+//		* @since 1.20
+//		* @return array
+//		*/
+//		public function getNames() {
+//			return this->names;
+//		}
+//
+//		/**
+//		* Parse a match array from preg_match
+//		* Returns array(magic word ID, parameter value)
+//		* If there is no parameter value, that element will be false.
+//		*
+//		* @param array m
+//		*
+//		* @throws MWException
+//		* @return array
+//		*/
+//		public function parseMatch(m) {
+//			reset(m);
+//			while (list(key, value) = each(m)) {
+//				if (key === 0 || value === '') {
+//					continue;
+//				}
+//				parts = explode('_', key, 2);
+//				if (count(parts) != 2) {
+//					// This shouldn't happen
+//					// continue;
+//					throw new MWException(__METHOD__ . ': bad parameter name');
+//				}
+//				list(/* synIndex */, magicName) = parts;
+//				paramValue = next(m);
+//				return [ magicName, paramValue ];
+//			}
+//			// This shouldn't happen either
+//			throw new MWException(__METHOD__ . ': parameter not found');
+//		}
+
+	/**
+	* Match some text, with parameter capture
+	* Returns an array with the magic word name in the first element and the
+	* parameter in the second element.
+	* Both elements are false if there was no match.
+	*
+	* @param String text
+	*
+	* @return array
+	*/
+	public void matchVariableStartToEnd(byte[][] rv, byte[] src) {
+		int src_end = src.length;
+		if (src_end == 0) {
+			rv[0] = rv[1] = null;
+			return;
+		}
+
+		byte[] name = null;
+		int val_bgn = -1, val_end = -1;
+
+		// check fwd; EX: "thumb=$1"
+		if (fwd_trie != null) {
+			Object o = fwd_trie.Match_at(trv, src, 0, src_end);
+			if (o != null) {
+				XomwMagicWordSynonym syn = ((XomwMagicWordSynonym)o);
+				name = syn.magic_name;
+				val_bgn = trv.Pos();
+				val_end = src_end;
+
+				// if "nil", then must be full match; EX: "thumbx" does not match "thumb"
+				if (syn.arg1_tid == XomwMagicWordSynonym.Arg1__nil
+					&& syn.text_wo_arg1.length != src_end) {
+					rv[0] = rv[1] = null;
+					return;
+				}
+			}
+		}
+
+		// check bwd; EX: "$1px"
+		if (bwd_trie != null) {
+			Object o = bwd_trie.Match_at(trv, src, src_end - 1, -1);
+			if (o != null) {
+				XomwMagicWordSynonym syn = ((XomwMagicWordSynonym)o);
+				name = syn.magic_name;
+				val_bgn = 0;
+				val_end = src_end - syn.text_wo_arg1.length;
+			}
+		}
+
+		rv[0] = name;
+		rv[1] = val_end - val_bgn == 0 ? Bry_.Empty : Bry_.Mid(src, val_bgn, val_end);
+	}
+
+//		/**
+//		* Match some text, without parameter capture
+//		* Returns the magic word name, or false if there was no capture
+//		*
+//		* @param String text
+//		*
+//		* @return String|boolean False on failure
+//		*/
+//		public function matchStartToEnd(text) {
+//			hash = this->getHash();
+//			if (isset(hash[1][text])) {
+//				return hash[1][text];
+//			}
+//			global wgContLang;
+//			lc = wgContLang->lc(text);
+//			if (isset(hash[0][lc])) {
+//				return hash[0][lc];
+//			}
+//			return false;
+//		}
+//
+//		/**
+//		* Returns an associative array, ID => param value, for all items that match
+//		* Removes the matched items from the input String (passed by reference)
+//		*
+//		* @param String text
+//		*
+//		* @return array
+//		*/
+//		public function matchAndRemove(&text) {
+//			found = [];
+//			regexes = this->getRegex();
+//			foreach (regexes as regex) {
+//				if (regex === '') {
+//					continue;
+//				}
+//				matches = [];
+//				res = preg_match_all(regex, text, matches, PREG_SET_ORDER);
+//				if (res === false) {
+//					LoggerFactory::getInstance('parser')->warning('preg_match_all returned false', [
+//						'code' => preg_last_error(),
+//						'regex' => regex,
+//						'text' => text,
+//					]);
+//				} elseif (res) {
+//					foreach (matches as m) {
+//						list(name, param) = this->parseMatch(m);
+//						found[name] = param;
+//					}
+//				}
+//				res = preg_replace(regex, '', text);
+//				if (res === null) {
+//					LoggerFactory::getInstance('parser')->warning('preg_replace returned null', [
+//						'code' => preg_last_error(),
+//						'regex' => regex,
+//						'text' => text,
+//					]);
+//				}
+//				text = res;
+//			}
+//			return found;
+//		}
+//
+//		/**
+//		* Return the ID of the magic word at the start of text, and remove
+//		* the prefix from text.
+//		* Return false if no match found and text is not modified.
+//		* Does not match parameters.
+//		*
+//		* @param String text
+//		*
+//		* @return int|boolean False on failure
+//		*/
+//		public function matchStartAndRemove(&text) {
+//			regexes = this->getRegexStart();
+//			foreach (regexes as regex) {
+//				if (regex === '') {
+//					continue;
+//				}
+//				if (preg_match(regex, text, m)) {
+//					list(id,) = this->parseMatch(m);
+//					if (strlen(m[0]) >= strlen(text)) {
+//						text = '';
+//					} else {
+//						text = substr(text, strlen(m[0]));
+//					}
+//					return id;
+//				}
+//			}
+//			return false;
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMagicWordArrayTest.java

@@ -13,3 +13,50 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XomwMagicWordArrayTest {
+	private final    XomwMagicWordArrayFxt fxt = new XomwMagicWordArrayFxt();
+	@Test  public void Nil() {
+		fxt.Init__word(Bool_.Y, "img_nil", "nil");
+		fxt.Init__ary("img_nil");
+		fxt.Test__matchVariableStartToEnd("nil", "img_nil", "");
+		fxt.Test__matchVariableStartToEnd("nila", null, null);
+	}
+	@Test  public void Bgn() {
+		fxt.Init__word(Bool_.Y, "img_bgn", "bgn$1");
+		fxt.Init__ary("img_bgn");
+		fxt.Test__matchVariableStartToEnd("bgna", "img_bgn", "a");
+		fxt.Test__matchVariableStartToEnd("bgn", "img_bgn", "");
+	}
+	@Test  public void End() {
+		fxt.Init__word(Bool_.Y, "img_end", "$1end");
+		fxt.Init__ary("img_end");
+		fxt.Test__matchVariableStartToEnd("aend", "img_end", "a");
+		fxt.Test__matchVariableStartToEnd("end", "img_end", "");
+	}
+	@Test  public void Smoke() {
+		fxt.Init__word(Bool_.Y, "img_upright", "upright", "upright=$1", "upright $1");
+		fxt.Init__word(Bool_.Y, "img_width", "$1px");
+		fxt.Init__ary("img_upright", "img_width");
+
+		fxt.Test__matchVariableStartToEnd("upright=123", "img_upright", "123");
+		fxt.Test__matchVariableStartToEnd("123px", "img_width", "123");
+	}
+}
+class XomwMagicWordArrayFxt {
+	private final    XomwMagicWordMgr magic_word_mgr = new XomwMagicWordMgr();
+	private XomwMagicWordArray magic_word_ary;
+	public void Init__word(boolean cs, String word, String... synonyms) {
+		magic_word_mgr.Add(Bry_.new_u8(word), cs, Bry_.Ary(synonyms));
+	}
+	public void Init__ary(String... words) {
+		magic_word_ary = new XomwMagicWordArray(magic_word_mgr, Bry_.Ary(words));
+	}
+	public void Test__matchVariableStartToEnd(String src, String expd_name, String expd_val) {
+		byte[][] rv = new byte[2][];
+		magic_word_ary.matchVariableStartToEnd(rv, Bry_.new_u8(src));
+		Gftest.Eq__str(expd_name, rv[0], expd_name);
+		Gftest.Eq__str(expd_val , rv[1], expd_val);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMagicWordMgr.java

@@ -13,3 +13,14 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwMagicWordMgr {
+	private final    Hash_adp_bry hash = Hash_adp_bry.cs();
+	public void Add(byte[] name, boolean cs, byte[]... synonyms) {
+		XomwMagicWord mw = new XomwMagicWord(name, cs, synonyms);
+		hash.Add(name, mw);
+	}
+	public XomwMagicWord Get(byte[] name) {
+		return (XomwMagicWord)hash.Get_by(name);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMagicWordSynonym.java

@@ -13,3 +13,77 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwMagicWordSynonym {
+	public final    byte[] magic_name;
+	public final    boolean case_match;
+	public final    byte[] text;
+	public final    byte[] text_wo_arg1;
+	public final    byte arg1_tid;
+	public XomwMagicWordSynonym(byte[] magic_name, boolean case_match, byte[] text) {
+		this.magic_name = magic_name;
+		this.case_match = case_match;
+		this.text = text;
+		this.arg1_tid = Get_arg1_tid(text);
+		switch (arg1_tid) {
+			case Arg1__bgn:
+				text_wo_arg1 = Bry_.Mid(text, 2);
+				break;
+			case Arg1__end:
+				text_wo_arg1 = Bry_.Mid(text, 0, text.length - 2);
+				break;
+			default:
+				text_wo_arg1 = text;
+				break;
+		}
+	}
+
+	private static byte Get_arg1_tid(byte[] src) {
+		int len = src.length;
+		byte rv = Arg1__nil;
+		int cur = 0;
+		while (true) {
+			if (cur == len) break;
+			byte b = src[cur];
+			// "$" matched
+			if (b == Byte_ascii.Dollar) {
+				// "1" matched?
+				int nxt_pos = cur + 1;
+				if (nxt_pos < len && src[nxt_pos] == Byte_ascii.Num_1) {
+					// "$1" matched
+					if (cur == 0) {
+						rv = Arg1__bgn;
+					}
+					else if (cur == len - 2) {							
+						rv = rv == Arg1__nil ? Arg1__end : Arg1__mix;
+					}
+					else {
+						if (rv == Arg1__nil)
+							rv = Arg1__mid;
+						else if (rv == Arg1__mid)
+							rv = Arg1__mix;
+					}
+					cur += 2;
+					continue;
+				}
+				else {
+					cur += 1;
+					continue;
+				}
+			}
+			else {
+				cur += 1;
+				continue;
+			}
+		}
+		return rv;
+	}
+
+	public static final byte 
+	  Arg1__nil = 0 // EX: "thumb"
+	, Arg1__bgn = 1 // EX: "$1px"
+	, Arg1__end = 2 // EX: "thumb=$1"
+	, Arg1__mid = 3 // EX: "a$1b"
+	, Arg1__mix = 4 // EX: "a$1b$cc"
+	;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMediaWikiServices.java

@@ -13,3 +13,589 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import gplx.xowa.mediawiki.languages.*;
+import gplx.xowa.mediawiki.includes.interwiki.*;
+import gplx.xowa.mediawiki.includes.title.*;
+/**
+* MediaWikiServices is the service locator for the application scope of MediaWiki.
+* Its implemented as a simple configurable DI container.
+* MediaWikiServices acts as a top level factory/registry for top level services, and builds
+* the network of service objects that defines MediaWiki's application logic.
+* It acts as an entry point to MediaWiki's dependency injection mechanism.
+*
+* Services are defined in the "wiring" array passed to the constructor,
+* or by calling defineService().
+*
+* @see docs/injection.txt for an overview of using dependency injection in the
+*      MediaWiki code super.
+*/
+public class XomwMediaWikiServices {
+	// XO.MW.SKIP:remove global getInstance(). See XomwEnv
+	private final    XomwMediaWikiTitleCodec titleParser;
+	private final    XomwInterwikiLookup interwikiLookup;
+
+	public XomwEnv env;
+	public XomwMediaWikiServices(XomwEnv env, XomwInterwikiLookup interwikiLookup, XomwLanguage language, byte[][] localInterwikis) {
+		this.env = env;
+		this.interwikiLookup = interwikiLookup;
+		this.titleParser = new XomwMediaWikiTitleCodec(this, language, localInterwikis);
+	}
+
+//		/**
+//		* Replaces the global MediaWikiServices instance.
+//		*
+//		* @since 1.28
+//		*
+//		* @note This is for use in PHPUnit tests only!
+//		*
+//		* @throws MWException if called outside of PHPUnit tests.
+//		*
+//		* @param MediaWikiServices $services The new MediaWikiServices Object.
+//		*
+//		* @return MediaWikiServices The old MediaWikiServices Object, so it can be restored later.
+//		*/
+//		public static function forceGlobalInstance( MediaWikiServices $services ) {
+//			if ( !defined( 'MW_PHPUNIT_TEST' ) ) {
+//				throw new MWException( __METHOD__ . ' must not be used outside unit tests.' );
+//			}
+//
+//			$old = self::getInstance();
+//			self::$instance = $services;
+//
+//			return $old;
+//		}
+//
+//		/**
+//		* Creates a new instance of MediaWikiServices and sets it as the global default
+//		* instance. getInstance() will return a different MediaWikiServices Object
+//		* after every call to resetGlobalInstance().
+//		*
+//		* @since 1.28
+//		*
+//		* @warning This should not be used during normal operation. It is intended for use
+//		* when the configuration has changed significantly since bootstrap time, e.g.
+//		* during the installation process or during testing.
+//		*
+//		* @warning Calling resetGlobalInstance() may leave the application in an inconsistent
+//		* state. Calling this is only safe under the ASSUMPTION that NO REFERENCE to
+//		* any of the services managed by MediaWikiServices exist. If any service objects
+//		* managed by the old MediaWikiServices instance remain in use, they may INTERFERE
+//		* with the operation of the services managed by the new MediaWikiServices.
+//		* Operating with a mix of services created by the old and the new
+//		* MediaWikiServices instance may lead to INCONSISTENCIES and even DATA LOSS!
+//		* Any class implementing LAZY LOADING is especially prone to this problem,
+//		* since instances would typically retain a reference to a storage layer service.
+//		*
+//		* @see forceGlobalInstance()
+//		* @see resetGlobalInstance()
+//		* @see resetBetweenTest()
+//		*
+//		* @param Config|null $bootstrapConfig The Config Object to be registered as the
+//		*        'BootstrapConfig' service. This has to contain at least the information
+//		*        needed to set up the 'ConfigFactory' service. If not given, the bootstrap
+//		*        config of the old instance of MediaWikiServices will be re-used. If there
+//		*        was no previous instance, a new GlobalVarConfig Object will be used to
+//		*        bootstrap the services.
+//		*
+//		* @param String $quick Set this to "quick" to allow expensive resources to be re-used.
+//		* See SalvageableService for details.
+//		*
+//		* @throws MWException If called after MW_SERVICE_BOOTSTRAP_COMPLETE has been defined in
+//		*         Setup.php (unless MW_PHPUNIT_TEST or MEDIAWIKI_INSTALL or RUN_MAINTENANCE_IF_MAIN
+//		*          is defined).
+//		*/
+//		public static function resetGlobalInstance( Config $bootstrapConfig = null, $quick = '' ) {
+//			if ( self::$instance === null ) {
+//				// no global instance yet, nothing to reset
+//				return;
+//			}
+//
+//			self::failIfResetNotAllowed( __METHOD__ );
+//
+//			if ( $bootstrapConfig === null ) {
+//				$bootstrapConfig = self::$instance->getBootstrapConfig();
+//			}
+//
+//			$oldInstance = self::$instance;
+//
+//			self::$instance = self::newInstance( $bootstrapConfig, 'load' );
+//			self::$instance->importWiring( $oldInstance, [ 'BootstrapConfig' ] );
+//
+//			if ( $quick === 'quick' ) {
+//				self::$instance->salvage( $oldInstance );
+//			} else {
+//				$oldInstance->destroy();
+//			}
+//		}
+//
+//		/**
+//		* Salvages the state of any salvageable service instances in $other.
+//		*
+//		* @note $other will have been destroyed when salvage() returns.
+//		*
+//		* @param MediaWikiServices $other
+//		*/
+//		private function salvage( self $other ) {
+//			foreach ( this.getServiceNames() as $name ) {
+//				// The service could be new in the new instance and not registered in the
+//				// other instance (e.g. an extension that was loaded after the instantiation of
+//				// the other instance. Skip this service in this case. See T143974
+//				try {
+//					$oldService = $other->peekService( $name );
+//				} catch ( NoSuchServiceException $e ) {
+//					continue;
+//				}
+//
+//				if ( $oldService instanceof SalvageableService ) {
+//					/** @var SalvageableService $newService */
+//					$newService = this.getService( $name );
+//					$newService->salvage( $oldService );
+//				}
+//			}
+//
+//			$other->destroy();
+//		}
+//
+//		/**
+//		* Creates a new MediaWikiServices instance and initializes it according to the
+//		* given $bootstrapConfig. In particular, all wiring files defined in the
+//		* ServiceWiringFiles setting are loaded, and the MediaWikiServices hook is called.
+//		*
+//		* @param Config|null $bootstrapConfig The Config Object to be registered as the
+//		*        'BootstrapConfig' service.
+//		*
+//		* @param String $loadWiring set this to 'load' to load the wiring files specified
+//		*        in the 'ServiceWiringFiles' setting in $bootstrapConfig.
+//		*
+//		* @return MediaWikiServices
+//		* @throws MWException
+//		* @throws \FatalError
+//		*/
+//		private static function newInstance( Config $bootstrapConfig, $loadWiring = '' ) {
+//			$instance = new self( $bootstrapConfig );
+//
+//			// Load the default wiring from the specified files.
+//			if ( $loadWiring === 'load' ) {
+//				$wiringFiles = $bootstrapConfig->get( 'ServiceWiringFiles' );
+//				$instance->loadWiringFiles( $wiringFiles );
+//			}
+//
+//			// Provide a traditional hook point to allow extensions to configure services.
+//			Hooks::run( 'MediaWikiServices', [ $instance ] );
+//
+//			return $instance;
+//		}
+//
+//		/**
+//		* Disables all storage layer services. After calling this, any attempt to access the
+//		* storage layer will result in an error. Use resetGlobalInstance() to restore normal
+//		* operation.
+//		*
+//		* @since 1.28
+//		*
+//		* @warning This is intended for extreme situations only and should never be used
+//		* while serving normal web requests. Legitimate use cases for this method include
+//		* the installation process. Test fixtures may also use this, if the fixture relies
+//		* on globalState.
+//		*
+//		* @see resetGlobalInstance()
+//		* @see resetChildProcessServices()
+//		*/
+//		public static function disableStorageBackend() {
+//			// TODO: also disable some Caches, JobQueues, etc
+//			$destroy = [ 'DBLoadBalancer', 'DBLoadBalancerFactory' ];
+//			$services = self::getInstance();
+//
+//			foreach ( $destroy as $name ) {
+//				$services->disableService( $name );
+//			}
+//
+//			ObjectCache::clear();
+//		}
+//
+//		/**
+//		* Resets any services that may have become stale after a child process
+//		* returns from after pcntl_fork(). It's also safe, but generally unnecessary,
+//		* to call this method from the parent process.
+//		*
+//		* @since 1.28
+//		*
+//		* @note This is intended for use in the context of process forking only!
+//		*
+//		* @see resetGlobalInstance()
+//		* @see disableStorageBackend()
+//		*/
+//		public static function resetChildProcessServices() {
+//			// NOTE: for now, just reset everything. Since we don't know the interdependencies
+//			// between services, we can't do this more selectively at this time.
+//			self::resetGlobalInstance();
+//
+//			// Child, reseed because there is no bug in PHP:
+//			// https://bugs.php.net/bug.php?id=42465
+//			mt_srand( getmypid() );
+//		}
+//
+//		/**
+//		* Resets the given service for testing purposes.
+//		*
+//		* @since 1.28
+//		*
+//		* @warning This is generally unsafe! Other services may still retain references
+//		* to the stale service instance, leading to failures and inconsistencies. Subclasses
+//		* may use this method to reset specific services under specific instances, but
+//		* it should not be exposed to application logic.
+//		*
+//		* @note With proper dependency injection used throughout the codebase, this method
+//		* should not be needed. It is provided to allow tests that pollute global service
+//		* instances to clean up.
+//		*
+//		* @param String $name
+//		* @param boolean $destroy Whether the service instance should be destroyed if it exists.
+//		*        When set to false, any existing service instance will effectively be detached
+//		*        from the container.
+//		*
+//		* @throws MWException if called outside of PHPUnit tests.
+//		*/
+//		public function resetServiceForTesting( $name, $destroy = true ) {
+//			if ( !defined( 'MW_PHPUNIT_TEST' ) && !defined( 'MW_PARSER_TEST' ) ) {
+//				throw new MWException( 'resetServiceForTesting() must not be used outside unit tests.' );
+//			}
+//
+//			this.resetService( $name, $destroy );
+//		}
+//
+//		/**
+//		* Convenience method that throws an exception unless it is called during a phase in which
+//		* resetting of global services is allowed. In general, services should not be reset
+//		* individually, since that may introduce inconsistencies.
+//		*
+//		* @since 1.28
+//		*
+//		* This method will throw an exception if:
+//		*
+//		* - self::$resetInProgress is false (to allow all services to be reset together
+//		*   via resetGlobalInstance)
+//		* - and MEDIAWIKI_INSTALL is not defined (to allow services to be reset during installation)
+//		* - and MW_PHPUNIT_TEST is not defined (to allow services to be reset during testing)
+//		*
+//		* This method is intended to be used to safeguard against accidentally resetting
+//		* global service instances that are not yet managed by MediaWikiServices. It is
+//		* defined here in the MediaWikiServices services class to have a central place
+//		* for managing service bootstrapping and resetting.
+//		*
+//		* @param String $method the name of the caller method, as given by __METHOD__.
+//		*
+//		* @throws MWException if called outside bootstrap mode.
+//		*
+//		* @see resetGlobalInstance()
+//		* @see forceGlobalInstance()
+//		* @see disableStorageBackend()
+//		*/
+//		public static function failIfResetNotAllowed( $method ) {
+//			if ( !defined( 'MW_PHPUNIT_TEST' )
+//				&& !defined( 'MW_PARSER_TEST' )
+//				&& !defined( 'MEDIAWIKI_INSTALL' )
+//				&& !defined( 'RUN_MAINTENANCE_IF_MAIN' )
+//				&& defined( 'MW_SERVICE_BOOTSTRAP_COMPLETE' )
+//			) {
+//				throw new MWException( $method . ' may only be called during bootstrapping and unit tests!' );
+//			}
+//		}
+//
+//		/**
+//		* @param Config $config The Config Object to be registered as the 'BootstrapConfig' service.
+//		*        This has to contain at least the information needed to set up the 'ConfigFactory'
+//		*        service.
+//		*/
+//		public function __construct( Config $config ) {
+//			parent::__construct();
+//
+//			// Register the given Config Object as the bootstrap config service.
+//			this.defineService( 'BootstrapConfig', function() use ( $config ) {
+//				return $config;
+//			} );
+//		}
+//
+//		// CONVENIENCE GETTERS ////////////////////////////////////////////////////
+//
+//		/**
+//		* Returns the Config Object containing the bootstrap configuration.
+//		* Bootstrap configuration would typically include database credentials
+//		* and other information that may be needed before the ConfigFactory
+//		* service can be instantiated.
+//		*
+//		* @note This should only be used during bootstrapping, in particular
+//		* when creating the MainConfig service. Application logic should
+//		* use getMainConfig() to get a Config instances.
+//		*
+//		* @since 1.27
+//		* @return Config
+//		*/
+//		public function getBootstrapConfig() {
+//			return this.getService( 'BootstrapConfig' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return ConfigFactory
+//		*/
+//		public function getConfigFactory() {
+//			return this.getService( 'ConfigFactory' );
+//		}
+//
+//		/**
+//		* Returns the Config Object that provides configuration for MediaWiki core.
+//		* This may or may not be the same Object that is returned by getBootstrapConfig().
+//		*
+//		* @since 1.27
+//		* @return Config
+//		*/
+//		public function getMainConfig() {
+//			return this.getService( 'MainConfig' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SiteLookup
+//		*/
+//		public function getSiteLookup() {
+//			return this.getService( 'SiteLookup' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SiteStore
+//		*/
+//		public function getSiteStore() {
+//			return this.getService( 'SiteStore' );
+//		}
+
+	/**
+	* @since 1.28
+	* @return InterwikiLookup
+	*/
+	public XomwInterwikiLookup getInterwikiLookup() {
+		return interwikiLookup;
+	}
+
+//		/**
+//		* @since 1.27
+//		* @return StatsdDataFactory
+//		*/
+//		public function getStatsdDataFactory() {
+//			return this.getService( 'StatsdDataFactory' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return EventRelayerGroup
+//		*/
+//		public function getEventRelayerGroup() {
+//			return this.getService( 'EventRelayerGroup' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SearchEngine
+//		*/
+//		public function newSearchEngine() {
+//			// New engine Object every time, since they keep state
+//			return this.getService( 'SearchEngineFactory' )->create();
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SearchEngineFactory
+//		*/
+//		public function getSearchEngineFactory() {
+//			return this.getService( 'SearchEngineFactory' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SearchEngineConfig
+//		*/
+//		public function getSearchEngineConfig() {
+//			return this.getService( 'SearchEngineConfig' );
+//		}
+//
+//		/**
+//		* @since 1.27
+//		* @return SkinFactory
+//		*/
+//		public function getSkinFactory() {
+//			return this.getService( 'SkinFactory' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return LBFactory
+//		*/
+//		public function getDBLoadBalancerFactory() {
+//			return this.getService( 'DBLoadBalancerFactory' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return LoadBalancer The main DB load balancer for the local wiki.
+//		*/
+//		public function getDBLoadBalancer() {
+//			return this.getService( 'DBLoadBalancer' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return WatchedItemStore
+//		*/
+//		public function getWatchedItemStore() {
+//			return this.getService( 'WatchedItemStore' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return WatchedItemQueryService
+//		*/
+//		public function getWatchedItemQueryService() {
+//			return this.getService( 'WatchedItemQueryService' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return CryptRand
+//		*/
+//		public function getCryptRand() {
+//			return this.getService( 'CryptRand' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return CryptHKDF
+//		*/
+//		public function getCryptHKDF() {
+//			return this.getService( 'CryptHKDF' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return MediaHandlerFactory
+//		*/
+//		public function getMediaHandlerFactory() {
+//			return this.getService( 'MediaHandlerFactory' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return MimeAnalyzer
+//		*/
+//		public function getMimeAnalyzer() {
+//			return this.getService( 'MimeAnalyzer' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return ProxyLookup
+//		*/
+//		public function getProxyLookup() {
+//			return this.getService( 'ProxyLookup' );
+//		}
+//
+//		/**
+//		* @since 1.29
+//		* @return Parser
+//		*/
+//		public function getParser() {
+//			return this.getService( 'Parser' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return GenderCache
+//		*/
+//		public function getGenderCache() {
+//			return this.getService( 'GenderCache' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return LinkCache
+//		*/
+//		public function getLinkCache() {
+//			return this.getService( 'LinkCache' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return LinkRendererFactory
+//		*/
+//		public function getLinkRendererFactory() {
+//			return this.getService( 'LinkRendererFactory' );
+//		}
+//
+//		/**
+//		* LinkRenderer instance that can be used
+//		* if no custom options are needed
+//		*
+//		* @since 1.28
+//		* @return LinkRenderer
+//		*/
+//		public function getLinkRenderer() {
+//			return this.getService( 'LinkRenderer' );
+//		}
+
+	/**
+	* @since 1.28
+	* @return TitleFormatter
+	*/
+	public XomwMediaWikiTitleCodec getTitleFormatter() {
+		// return this.getService( 'TitleFormatter' );
+		return titleParser;
+	}
+
+	/**
+	* @since 1.28
+	* @return TitleParser
+	*/
+	public XomwMediaWikiTitleCodec getTitleParser() {
+		// return this.getService( 'TitleParser' );
+		return titleParser;
+	}
+
+//		/**
+//		* @since 1.28
+//		* @return \BagOStuff
+//		*/
+//		public function getMainObjectStash() {
+//			return this.getService( 'MainObjectStash' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return \WANObjectCache
+//		*/
+//		public function getMainWANObjectCache() {
+//			return this.getService( 'MainWANObjectCache' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return \BagOStuff
+//		*/
+//		public function getLocalServerObjectCache() {
+//			return this.getService( 'LocalServerObjectCache' );
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return VirtualRESTServiceClient
+//		*/
+//		public function getVirtualRESTServiceClient() {
+//			return this.getService( 'VirtualRESTServiceClient' );
+//		}
+//
+//		///////////////////////////////////////////////////////////////////////////
+//		// NOTE: When adding a service getter here, don't forget to add a test
+//		// case for it in MediaWikiServicesTest::provideGetters() and in
+//		// MediaWikiServicesTest::provideGetService()!
+//		///////////////////////////////////////////////////////////////////////////
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwMessageMgr.java

@@ -13,3 +13,12 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import gplx.xowa.mediawiki.languages.*;
+public class XomwMessageMgr {
+	private final    Hash_adp hash = Hash_adp_.New();
+	public void Add(String key, String val, XomwLanguage language) {
+		hash.Add(key, new XomwMessage(Bry_.new_u8(val), language));
+	}
+	public XomwMessage Get_by_str(String key) {return (XomwMessage)hash.Get_by(key);}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwNamespace.java

@@ -13,3 +13,492 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwNamespace {
+//		/**
+//		* These namespaces should always be first-letter capitalized, now and
+//		* forevermore. Historically, they could've probably been lowercased too,
+//		* but some things are just too ingrained now. :)
+//		*/
+//		private static $alwaysCapitalizedNamespaces = [ NS_SPECIAL, NS_USER, NS_MEDIAWIKI ];
+//
+//		/**
+//		* Throw an exception when trying to get the subject or talk page
+//		* for a given namespace where it does not make sense.
+//		* Special namespaces are defined in includes/Defines.php and have
+//		* a value below 0 (ex: NS_SPECIAL = -1 , NS_MEDIA = -2)
+//		*
+//		* @param int $index
+//		* @param String $method
+//		*
+//		* @throws MWException
+//		* @return boolean
+//		*/
+//		private static function isMethodValidFor($index, $method) {
+//			if ($index < NS_MAIN) {
+//				throw new MWException("$method does not make any sense for given namespace $index");
+//			}
+//			return true;
+//		}
+//
+//		/**
+//		* Can pages in the given namespace be moved?
+//		*
+//		* @param int $index Namespace index
+//		* @return boolean
+//		*/
+//		public static function isMovable($index) {
+//			global $wgAllowImageMoving;
+//
+//			$result = !($index < NS_MAIN || ($index == NS_FILE && !$wgAllowImageMoving));
+//
+//			/**
+//			* @since 1.20
+//			*/
+//			Hooks::run('NamespaceIsMovable', [ $index, &$result ]);
+//
+//			return $result;
+//		}
+//
+//		/**
+//		* Is the given namespace is a subject (non-talk) namespace?
+//		*
+//		* @param int $index Namespace index
+//		* @return boolean
+//		* @since 1.19
+//		*/
+//		public static function isSubject($index) {
+//			return !self::isTalk($index);
+//		}
+//
+//		/**
+//		* Is the given namespace a talk namespace?
+//		*
+//		* @param int $index Namespace index
+//		* @return boolean
+//		*/
+//		public static function isTalk($index) {
+//			return $index > NS_MAIN
+//				&& $index % 2;
+//		}
+//
+//		/**
+//		* Get the talk namespace index for a given namespace
+//		*
+//		* @param int $index Namespace index
+//		* @return int
+//		*/
+//		public static function getTalk($index) {
+//			self::isMethodValidFor($index, __METHOD__);
+//			return self::isTalk($index)
+//				? $index
+//				: $index + 1;
+//		}
+//
+//		/**
+//		* Get the subject namespace index for a given namespace
+//		* Special namespaces (NS_MEDIA, NS_SPECIAL) are always the subject.
+//		*
+//		* @param int $index Namespace index
+//		* @return int
+//		*/
+//		public static function getSubject($index) {
+//			# Handle special namespaces
+//			if ($index < NS_MAIN) {
+//				return $index;
+//			}
+//
+//			return self::isTalk($index)
+//				? $index - 1
+//				: $index;
+//		}
+//
+//		/**
+//		* Get the associated namespace.
+//		* For talk namespaces, returns the subject (non-talk) namespace
+//		* For subject (non-talk) namespaces, returns the talk namespace
+//		*
+//		* @param int $index Namespace index
+//		* @return int|null If no associated namespace could be found
+//		*/
+//		public static function getAssociated($index) {
+//			self::isMethodValidFor($index, __METHOD__);
+//
+//			if (self::isSubject($index)) {
+//				return self::getTalk($index);
+//			} elseif (self::isTalk($index)) {
+//				return self::getSubject($index);
+//			} else {
+//				return null;
+//			}
+//		}
+//
+//		/**
+//		* Returns whether the specified namespace exists
+//		*
+//		* @param int $index
+//		*
+//		* @return boolean
+//		* @since 1.19
+//		*/
+//		public static function exists($index) {
+//			$nslist = self::getCanonicalNamespaces();
+//			return isset($nslist[$index]);
+//		}
+//
+//		/**
+//		* Returns whether the specified namespaces are the same namespace
+//		*
+//		* @note It's possible that in the future we may start using something
+//		* other than just namespace indexes. Under that circumstance making use
+//		* of this function rather than directly doing comparison will make
+//		* sure that code will not potentially break.
+//		*
+//		* @param int $ns1 The first namespace index
+//		* @param int $ns2 The second namespace index
+//		*
+//		* @return boolean
+//		* @since 1.19
+//		*/
+//		public static function equals($ns1, $ns2) {
+//			return $ns1 == $ns2;
+//		}
+//
+//		/**
+//		* Returns whether the specified namespaces share the same subject.
+//		* eg: NS_USER and NS_USER wil return true, as well
+//		*     NS_USER and NS_USER_TALK will return true.
+//		*
+//		* @param int $ns1 The first namespace index
+//		* @param int $ns2 The second namespace index
+//		*
+//		* @return boolean
+//		* @since 1.19
+//		*/
+//		public static function subjectEquals($ns1, $ns2) {
+//			return self::getSubject($ns1) == self::getSubject($ns2);
+//		}
+
+	/**
+	* Returns array of all defined namespaces with their canonical
+	* (English) names.
+	*
+	* @param boolean $rebuild Rebuild namespace list (default = false). Used for testing.
+	*
+	* @return array
+	* @since 1.17
+	*/
+	private static XomwNamespacesById namespaces = null;
+	public static XomwNamespacesById getCanonicalNamespaces() {return getCanonicalNamespaces(false);}
+	public static XomwNamespacesById getCanonicalNamespaces(boolean rebuild) {
+		if (namespaces == null || rebuild) {
+//				global $wgExtraNamespaces, $wgCanonicalNamespaceNames;
+			namespaces = XomwSetup.wgCanonicalNamespaceNames.Clone();
+			namespaces.Add(XomwDefines.NS_MAIN, "");
+
+//				// Add extension namespaces
+//				$namespaces += ExtensionRegistry::getInstance()->getAttribute('ExtensionNamespaces');
+//				if (is_array($wgExtraNamespaces)) {
+//					$namespaces += $wgExtraNamespaces;
+//				}
+//				Hooks::run('CanonicalNamespaces', [ &$namespaces ]);
+		}
+		return namespaces;
+	}
+
+//		/**
+//		* Returns the canonical (English) name for a given index
+//		*
+//		* @param int $index Namespace index
+//		* @return String|boolean If no canonical definition.
+//		*/
+//		public static function getCanonicalName($index) {
+//			$nslist = self::getCanonicalNamespaces();
+//			if (isset($nslist[$index])) {
+//				return $nslist[$index];
+//			} else {
+//				return false;
+//			}
+//		}
+
+	/**
+	* Returns the index for a given canonical name, or NULL
+	* The input *must* be converted to lower case first
+	*
+	* @param String $name Namespace name
+	* @return int
+	*/
+	private static Hash_adp xNamespaces = null;
+	public static int getCanonicalIndex(byte[] name) {
+		if (xNamespaces == null) {
+			xNamespaces = Hash_adp_bry.cs();
+			XomwNamespacesById namespacesHash = getCanonicalNamespaces();
+			int len = namespacesHash.Len();
+			for (int i = 0; i < len; i++) {
+				XomwNamespaceItem item = (XomwNamespaceItem)namespacesHash.GetAtOrNull(i);
+				xNamespaces.Add(Bry_.Lcase__all(item.name), item);	// NOTE: MW does "strtolower($text)"; canonical namespaces are always ascii
+			}
+		}
+		XomwNamespaceItem xNs = (XomwNamespaceItem)xNamespaces.Get_by(name);
+		if (xNs != null) {
+			return xNs.id;
+		}
+		else {
+			return XomwNamespace.NULL_NS_ID;
+		}
+	}
+
+//		/**
+//		* Returns an array of the namespaces (by integer id) that exist on the
+//		* wiki. Used primarily by the api in help documentation.
+//		* @return array
+//		*/
+//		public static function getValidNamespaces() {
+//			static $mValidNamespaces = null;
+//
+//			if (is_null($mValidNamespaces)) {
+//				foreach (array_keys(self::getCanonicalNamespaces()) as $ns) {
+//					if ($ns >= 0) {
+//						$mValidNamespaces[] = $ns;
+//					}
+//				}
+//				// T109137: sort numerically
+//				sort($mValidNamespaces, SORT_NUMERIC);
+//			}
+//
+//			return $mValidNamespaces;
+//		}
+//
+//		/**
+//		* Can this namespace ever have a talk namespace?
+//		*
+//		* @param int $index Namespace index
+//		* @return boolean
+//		*/
+//		public static function canTalk($index) {
+//			return $index >= NS_MAIN;
+//		}
+//
+//		/**
+//		* Does this namespace contain content, for the purposes of calculating
+//		* statistics, etc?
+//		*
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function isContent($index) {
+//			global $wgContentNamespaces;
+//			return $index == NS_MAIN || in_array($index, $wgContentNamespaces);
+//		}
+//
+//		/**
+//		* Might pages in this namespace require the use of the Signature button on
+//		* the edit toolbar?
+//		*
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function wantSignatures($index) {
+//			global $wgExtraSignatureNamespaces;
+//			return self::isTalk($index) || in_array($index, $wgExtraSignatureNamespaces);
+//		}
+//
+//		/**
+//		* Can pages in a namespace be watched?
+//		*
+//		* @param int $index
+//		* @return boolean
+//		*/
+//		public static function isWatchable($index) {
+//			return $index >= NS_MAIN;
+//		}
+//
+//		/**
+//		* Does the namespace allow subpages?
+//		*
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function hasSubpages($index) {
+//			global $wgNamespacesWithSubpages;
+//			return !empty($wgNamespacesWithSubpages[$index]);
+//		}
+//
+//		/**
+//		* Get a list of all namespace indices which are considered to contain content
+//		* @return array Array of namespace indices
+//		*/
+//		public static function getContentNamespaces() {
+//			global $wgContentNamespaces;
+//			if (!is_array($wgContentNamespaces) || $wgContentNamespaces == []) {
+//				return [ NS_MAIN ];
+//			} elseif (!in_array(NS_MAIN, $wgContentNamespaces)) {
+//				// always force NS_MAIN to be part of array (to match the algorithm used by isContent)
+//				return array_merge([ NS_MAIN ], $wgContentNamespaces);
+//			} else {
+//				return $wgContentNamespaces;
+//			}
+//		}
+//
+//		/**
+//		* List all namespace indices which are considered subject, aka not a talk
+//		* or special namespace. See also XomwNamespace::isSubject
+//		*
+//		* @return array Array of namespace indices
+//		*/
+//		public static function getSubjectNamespaces() {
+//			return array_filter(
+//				XomwNamespace::getValidNamespaces(),
+//				'XomwNamespace::isSubject'
+//			);
+//		}
+//
+//		/**
+//		* List all namespace indices which are considered talks, aka not a subject
+//		* or special namespace. See also XomwNamespace::isTalk
+//		*
+//		* @return array Array of namespace indices
+//		*/
+//		public static function getTalkNamespaces() {
+//			return array_filter(
+//				XomwNamespace::getValidNamespaces(),
+//				'XomwNamespace::isTalk'
+//			);
+//		}
+//
+//		/**
+//		* Is the namespace first-letter capitalized?
+//		*
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function isCapitalized($index) {
+//			global $wgCapitalLinks, $wgCapitalLinkOverrides;
+//			// Turn NS_MEDIA into NS_FILE
+//			$index = $index == NS_MEDIA ? NS_FILE : $index;
+//
+//			// Make sure to get the subject of our namespace
+//			$index = self::getSubject($index);
+//
+//			// Some namespaces are special and should always be upper case
+//			if (in_array($index, self::$alwaysCapitalizedNamespaces)) {
+//				return true;
+//			}
+//			if (isset($wgCapitalLinkOverrides[$index])) {
+//				// $wgCapitalLinkOverrides is explicitly set
+//				return $wgCapitalLinkOverrides[$index];
+//			}
+//			// Default to the global setting
+//			return $wgCapitalLinks;
+//		}
+//
+//		/**
+//		* Does the namespace (potentially) have different aliases for different
+//		* genders. Not all languages make a distinction here.
+//		*
+//		* @since 1.18
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function hasGenderDistinction($index) {
+//			return $index == NS_USER || $index == NS_USER_TALK;
+//		}
+//
+//		/**
+//		* It is not possible to use pages from this namespace as template?
+//		*
+//		* @since 1.20
+//		* @param int $index Index to check
+//		* @return boolean
+//		*/
+//		public static function isNonincludable($index) {
+//			global $wgNonincludableNamespaces;
+//			return $wgNonincludableNamespaces && in_array($index, $wgNonincludableNamespaces);
+//		}
+//
+//		/**
+//		* Get the default content model for a namespace
+//		* This does not mean that all pages in that namespace have the model
+//		*
+//		* @since 1.21
+//		* @param int $index Index to check
+//		* @return null|String Default model name for the given namespace, if set
+//		*/
+//		public static function getNamespaceContentModel($index) {
+//			global $wgNamespaceContentModels;
+//			return isset($wgNamespaceContentModels[$index])
+//				? $wgNamespaceContentModels[$index]
+//				: null;
+//		}
+//
+//		/**
+//		* Determine which restriction levels it makes sense to use in a namespace,
+//		* optionally filtered by a user's rights.
+//		*
+//		* @since 1.23
+//		* @param int $index Index to check
+//		* @param User $user User to check
+//		* @return array
+//		*/
+//		public static function getRestrictionLevels($index, User $user = null) {
+//			global $wgNamespaceProtection, $wgRestrictionLevels;
+//
+//			if (!isset($wgNamespaceProtection[$index])) {
+//				// All levels are valid if there's no namespace restriction.
+//				// But still filter by user, if necessary
+//				$levels = $wgRestrictionLevels;
+//				if ($user) {
+//					$levels = array_values(array_filter($levels, function ($level) use ($user) {
+//						$right = $level;
+//						if ($right == 'sysop') {
+//							$right = 'editprotected'; // BC
+//						}
+//						if ($right == 'autoconfirmed') {
+//							$right = 'editsemiprotected'; // BC
+//						}
+//						return ($right == '' || $user->isAllowed($right));
+//					}));
+//				}
+//				return $levels;
+//			}
+//
+//			// First, get the list of groups that can edit this namespace.
+//			$namespaceGroups = [];
+//			$combine = 'array_merge';
+//			foreach ((array)$wgNamespaceProtection[$index] as $right) {
+//				if ($right == 'sysop') {
+//					$right = 'editprotected'; // BC
+//				}
+//				if ($right == 'autoconfirmed') {
+//					$right = 'editsemiprotected'; // BC
+//				}
+//				if ($right != '') {
+//					$namespaceGroups = call_user_func($combine, $namespaceGroups,
+//						User::getGroupsWithPermission($right));
+//					$combine = 'array_intersect';
+//				}
+//			}
+//
+//			// Now, keep only those restriction levels where there is at least one
+//			// group that can edit the namespace but would be blocked by the
+//			// restriction.
+//			$usableLevels = [ '' ];
+//			foreach ($wgRestrictionLevels as $level) {
+//				$right = $level;
+//				if ($right == 'sysop') {
+//					$right = 'editprotected'; // BC
+//				}
+//				if ($right == 'autoconfirmed') {
+//					$right = 'editsemiprotected'; // BC
+//				}
+//				if ($right != '' && (!$user || $user->isAllowed($right)) &&
+//					array_diff($namespaceGroups, User::getGroupsWithPermission($right))
+//				) {
+//					$usableLevels[] = $level;
+//				}
+//			}
+//
+//			return $usableLevels;
+//		}
+	public static final int NULL_NS_ID = XophpUtility.NULL_INT;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwNamespaceItem.java

@@ -13,3 +13,12 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwNamespaceItem {
+	public final    int id;
+	public final    byte[] name;
+	public XomwNamespaceItem(int id, byte[] name) {
+		this.id = id;
+		this.name = name;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwNamespacesById.java

@@ -13,3 +13,25 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import gplx.core.lists.*;
+public class XomwNamespacesById {
+	private HashByInt hash = new HashByInt();
+	public int Len() {return hash.Len();}
+	public byte[] GetNameOrNull(int id) {
+		XomwNamespaceItem item = (XomwNamespaceItem)hash.Get_by_or_null(id);
+		return item == null ? null : item.name;
+	}
+	public XomwNamespaceItem GetAtOrNull(int idx) {
+		return (XomwNamespaceItem)hash.Get_at_or_null(idx);
+	}
+	public XomwNamespacesById Add(int id, String text) {
+		hash.Add(id, new XomwNamespaceItem(id, Bry_.new_u8(text)));
+		return this;
+	}
+	public XomwNamespacesById Clone() {
+		XomwNamespacesById rv = new XomwNamespacesById();
+		rv.hash = hash.Clone();
+		return rv;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwNamespacesByName.java

@@ -13,3 +13,23 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwNamespacesByName {
+	private final    Ordered_hash hash = Ordered_hash_.New_bry();
+	public int Len() {return hash.Len();}
+	public int GetAsIdOrNullInt(byte[] name) {
+		XomwNamespaceItem item = (XomwNamespaceItem)hash.Get_by(name);
+		return item == null ? XophpUtility.NULL_INT : item.id;
+	}
+	public XomwNamespaceItem GetAtOrNull(int idx) {
+		return (XomwNamespaceItem)hash.Get_at(idx);
+	}
+	public void Add(byte[] name, XomwNamespaceItem item) {
+		hash.Add(name, item);
+	}
+	public XomwNamespacesByName Add(String name, int id) {
+		byte[] nameBry = Bry_.new_u8(name);
+		hash.Add(nameBry, new XomwNamespaceItem(id, nameBry));
+		return this;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwRawMessage.java

@@ -13,3 +13,55 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+/**
+* Variant of the Message cls.
+*
+* Rather than treating the message key as a lookup
+* value (which is passed to the MessageCache and
+* translated as necessary), a RawMessage key is
+* treated as the actual message.
+*
+* All other functionality (parsing, escaping, etc.)
+* is preserved.
+*
+* @since 1.21
+*/
+class XomwRawMessage { // : XomwMessage 
+//
+//		/**
+//		* Call the parent constructor, then store the key as
+//		* the message.
+//		*
+//		* @see Message::__construct
+//		*
+//		* @param String $text Message to use.
+//		* @param array $params Parameters for the message.
+//		*
+//		* @throws InvalidArgumentException
+//		*/
+//		public function __construct( $text, $params = [] ) {
+//			if ( !is_string( $text ) ) {
+//				throw new InvalidArgumentException( '$text must be a String' );
+//			}
+//
+//			parent::__construct( $text, $params );
+//
+//			// The key is the message.
+//			$this->message = $text;
+//		}
+//
+//		/**
+//		* Fetch the message (in this case, the key).
+//		*
+//		* @return String
+//		*/
+//		public function fetchMessage() {
+//			// Just in case the message is unset somewhere.
+//			if ( $this->message === null ) {
+//				$this->message = $this->key;
+//			}
+//
+//			return $this->message;
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwSanitizerTest.java

@@ -13,3 +13,164 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*; import gplx.core.btries.*; import gplx.xowa.mediawiki.includes.xohtml.*;
+public class XomwSanitizerTest {
+	private final    XomwSanitizerFxt fxt = new XomwSanitizerFxt();
+	@Test   public void Normalize__text()                  {fxt.Test__normalize_char_references("abc"                      , "abc");}
+	@Test   public void Normalize__dec()                   {fxt.Test__normalize_char_references(""                    , "");}
+	@Test   public void Normalize__dec__invalid()          {fxt.Test__normalize_char_references("	"                    , "	");}
+	@Test   public void Normalize__hex()                   {fxt.Test__normalize_char_references("ÿ"                   , "ÿ");}
+	@Test   public void Normalize__entity()                {fxt.Test__normalize_char_references("α"                  , "α");}
+	@Test   public void Normalize__entity__lt()            {fxt.Test__normalize_char_references("<"                     , "<");}
+	@Test   public void Normalize__entity__alias()         {fxt.Test__normalize_char_references("&רלמ;"                    , "‏");}
+	@Test   public void Normalize__amp()                   {fxt.Test__normalize_char_references("a&b"                      , "a&b");}
+	@Test   public void Normalize__invalid()               {fxt.Test__normalize_char_references("&(invalid);"              , "&(invalid);");}
+	@Test   public void Normalize__many() {
+		fxt.Test__normalize_char_references
+		( "a 	 b α c ÿ d &(invalid); e"
+		, "a 	 b α c ÿ d &(invalid); e"
+		);
+	}
+	@Test   public void Regex__domain() {
+		Xomw_regex_find_domain regex_domain = new Xomw_regex_find_domain();
+		// normal
+		fxt.Test__regex_domain_y(regex_domain, "https://a.org/bcd", "https:", "//a.org", "/bcd");
+		// trailing backslash
+		fxt.Test__regex_domain_y(regex_domain, "https://a.org/", "https:", "//a.org", "/");
+		// domain only
+		fxt.Test__regex_domain_y(regex_domain, "https://a.org", "https:", "//a.org", "");
+		// colon not found
+		fxt.Test__regex_domain_n(regex_domain, "https//a.org/bcd");
+		// host_bgn.eos
+		fxt.Test__regex_domain_n(regex_domain, "https:");
+		// host_bgn.//
+		fxt.Test__regex_domain_n(regex_domain, "https:a//");
+		// host_bgn.///
+		fxt.Test__regex_domain_n(regex_domain, "https:///a.org/b");
+	}
+	@Test   public void Regex__clean_url() {
+		Xomw_regex_escape_invalid regex = new Xomw_regex_escape_invalid();
+		// noop
+		fxt.Test__regex_escape_invalid(regex, "https://a.org/bcd", Bool_.N, "");
+		// symbols
+		fxt.Test__regex_escape_invalid(regex, "[]<>\"|", Bool_.Y, "%5B%5D%3C%3E%22%7C%7F");
+		// range: 00 - 32
+		fxt.Test__regex_escape_invalid(regex, "\t\n ", Bool_.Y, "%09%0A+");
+	}
+	@Test   public void Regex__ipv6_brack() {
+		Xomw_regex_ipv6_brack regex = new Xomw_regex_ipv6_brack();
+		// basic
+		fxt.Test__regex_ipv6_brack(regex, Bool_.Y, "//%5B0a.1b:12%5D:123");
+		// port: none
+		fxt.Test__regex_ipv6_brack(regex, Bool_.Y, "//%5Ba%5D");
+		// port: multiple
+		fxt.Test__regex_ipv6_brack(regex, Bool_.Y, "//%5Ba%5D:1:2:3");
+		// "//%5B" missing
+		fxt.Test__regex_ipv6_brack(regex, Bool_.N, "abc");
+		// ipv6: invalid
+		fxt.Test__regex_ipv6_brack(regex, Bool_.N, "//%5Ba!%5D:1");
+		// ipv6: 0-len
+		fxt.Test__regex_ipv6_brack(regex, Bool_.N, "//%5B%5D:1");
+		// port: invalid
+		fxt.Test__regex_ipv6_brack(regex, Bool_.N, "//%5Ba%5D:a");
+		// port: 0-len
+		fxt.Test__regex_ipv6_brack(regex, Bool_.N, "//%5Ba%5D:");
+	}
+	@Test   public void Decode() {
+		// dec
+		fxt.Test__decode_char_references("&#33;"           , "!");
+		// hex
+		fxt.Test__decode_char_references("&#x23;"          , "#");
+		// entity
+		fxt.Test__decode_char_references("&alpha;"         , "α");
+		// entity:lt
+		fxt.Test__decode_char_references("&lt;"            , "<");
+		// entity:rlm
+		fxt.Test__decode_char_references("&רלמ;"           , "‏");
+		// entity:invalid
+		fxt.Test__decode_char_references("&invalid;"       , "&invalid;");
+		// amp
+		fxt.Test__decode_char_references("a&b"             , "a&b");
+	}
+	@Test   public void Clean_url() {
+		// entity
+		fxt.Test__clean_url("http://a.org/b&amp;c"           , "http://a.org/b&c");
+		// entity: escape
+		fxt.Test__clean_url("http://a.org/b&quot;c"          , "http://a.org/b%22c");
+		// domain=n; make sure &quot; is changed, but not soft-hyphen
+		fxt.Test__clean_url("a&quot;­z"                      , "a%22­z");
+		// host: invalid idn
+		fxt.Test__clean_url("http://a᠆b.org/c᠆d"              , "http://ab.org/c᠆d");
+		// ipv6_brack
+		fxt.Test__clean_url("http://[0a.1b:12]:123/cd"       , "http://[0a.1b:12]:123/cd");
+	}
+	@Test   public void Merge_atrs() {
+		Xomw_atr_mgr src_atrs = new Xomw_atr_mgr();
+		Xomw_atr_mgr trg_atrs = new Xomw_atr_mgr();
+		Xomw_atr_mgr expd_atrs = new Xomw_atr_mgr();
+		String cls = "class";
+		// basic: k1 + k2
+		fxt.Test__merge_attributes(src_atrs.Clear().Add_many("k1", "v1"), trg_atrs.Clear().Add_many("k2", "v2"), expd_atrs.Clear().Add_many("k1", "v1", "k2", "v2"));
+		// overwrite: k1 + k1
+		fxt.Test__merge_attributes(src_atrs.Clear().Add_many("k1", "v1"), trg_atrs.Clear().Add_many("k1", "v1a"), expd_atrs.Clear().Add_many("k1", "v1a"));
+		// cls: many
+		fxt.Test__merge_attributes(src_atrs.Clear().Add_many(cls, "v1 v2"), trg_atrs.Clear().Add_many(cls, "v3 v4"), expd_atrs.Clear().Add_many(cls, "v1 v2 v3 v4"));
+		// cls: src.empty
+		fxt.Test__merge_attributes(src_atrs.Clear(), trg_atrs.Clear().Add_many(cls, "v1"), expd_atrs.Clear().Add_many(cls, "v1"));
+		// cls: ws
+		fxt.Test__merge_attributes(src_atrs.Clear().Add_many(cls, "  v1   v2  "), trg_atrs.Clear().Add_many(cls, "  v3   v4   "), expd_atrs.Clear().Add_many(cls, "v1 v2 v3 v4"));
+	}
+
+	@Test   public void normalizeWhitespace() {
+		fxt.Test_normalizeWhitespace("a\r\nb", "a b");
+		fxt.Test_normalizeWhitespace("a\rb", "a b");
+		fxt.Test_normalizeWhitespace("a\nb", "a b");
+		fxt.Test_normalizeWhitespace("a\tb", "a b");
+	}
+}
+class XomwSanitizerFxt {
+	private final    XomwSanitizer sanitizer = new XomwSanitizer();
+	private final    Bry_bfr tmp = Bry_bfr_.New();
+	public void Test__normalize_char_references(String src_str, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		sanitizer.normalizeCharReferences(tmp, Bool_.Y, src_bry, 0, src_bry.length);
+		Gftest.Eq__str(expd, tmp.To_str_and_clear());
+	}
+	public void Test__regex_domain_y(Xomw_regex_find_domain regex_domain, String src_str, String expd_prot, String expd_host, String expd_rest) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__bool(true, regex_domain.Match(src_bry, 0, src_bry.length), src_str);
+		Gftest.Eq__str(expd_prot, Bry_.Mid(src_bry, regex_domain.prot_bgn, regex_domain.prot_end));
+		Gftest.Eq__str(expd_host, Bry_.Mid(src_bry, regex_domain.host_bgn, regex_domain.host_end));
+		Gftest.Eq__str(expd_rest, Bry_.Mid(src_bry, regex_domain.rest_bgn, regex_domain.rest_end));
+	}
+	public void Test__regex_domain_n(Xomw_regex_find_domain regex_domain, String src_str) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__bool(false, regex_domain.Match(src_bry, 0, src_bry.length), src_str);
+	}
+	public void Test__regex_escape_invalid(Xomw_regex_escape_invalid regex, String src_str, boolean expd_rslt, String expd_str) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__bool(expd_rslt, regex.Escape(tmp, src_bry, 0, src_bry.length));
+		Gftest.Eq__str(expd_str, tmp.To_bry_and_clear());
+	}
+	public void Test__regex_ipv6_brack(Xomw_regex_ipv6_brack regex, boolean expd_rslt, String src_str) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__bool(expd_rslt, regex.Match(src_bry, 0, src_bry.length));
+	}
+	public void Test__decode_char_references(String src_str, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		sanitizer.decodeCharReferences(tmp, Bool_.Y, src_bry, 0, src_bry.length);
+		Gftest.Eq__str(expd, tmp.To_str_and_clear());
+	}
+	public void Test__clean_url(String src_str, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__str(expd, sanitizer.cleanUrl(src_bry));
+	}
+	public void Test__merge_attributes(Xomw_atr_mgr src, Xomw_atr_mgr trg, Xomw_atr_mgr expd) {
+		sanitizer.mergeAttributes(src, trg);
+		Gftest.Eq__ary__lines(expd.To_str(tmp), src.To_str(tmp), "merge_atrs");
+	}
+	public void Test_normalizeWhitespace(String src_str, String expd) {
+		Gftest.Eq__str(expd, sanitizer.normalizeWhitespace(Bry_.new_u8(src_str)), "merge_atrs");
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwSetup.java

@@ -13,3 +13,887 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+/**
+* Include most things that are needed to make MediaWiki work.
+*
+* This file is included by WebStart.php and doMaintenance.php so that both
+* web and maintenance scripts share a final set up phase to include necessary
+* files and create global Object variables.
+*
+* This program is free software; you can redistribute it and/or modify
+* it under the terms of the GNU General Public License as published by
+* the Free Software Foundation; either version 2 of the License, or
+* (at your option) any later version.
+*
+* This program is distributed in the hope that it will be useful,
+* but WITHOUT ANY WARRANTY; without even the implied warranty of
+* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+* GNU General Public License for more details.
+*
+* You should have received a copy of the GNU General Public License along
+* with this program; if not, write to the Free Software Foundation, Inc.,
+* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+* http://www.gnu.org/copyleft/gpl.html
+*
+* @file
+*/
+public class XomwSetup {
+//		/**
+//		* This file is not a valid entry point, perform no further processing unless
+//		* MEDIAWIKI is defined
+//		*/
+//		if ( !defined( 'MEDIAWIKI' ) ) {
+//			exit( 1 );
+//		}
+//
+//		$fname = 'Setup.php';
+//		$ps_setup = Profiler::instance()->scopedProfileIn( $fname );
+//
+//		// Load queued extensions
+//		ExtensionRegistry::getInstance()->loadFromQueue();
+//		// Don't let any other extensions load
+//		ExtensionRegistry::getInstance()->finish();
+//
+//		// Check to see if we are at the file scope
+//		if ( !isset( $wgVersion ) ) {
+//			echo "Error, Setup.php must be included from the file scope, after DefaultSettings.php\n";
+//			die( 1 );
+//		}
+//
+//		mb_internal_encoding( 'UTF-8' );
+//
+//		// Set various default paths sensibly...
+//		$ps_default = Profiler::instance()->scopedProfileIn( $fname . '-defaults' );
+//
+//		if ( $wgScript === false ) {
+//			$wgScript = "$wgScriptPath/index.php";
+//		}
+//		if ( $wgLoadScript === false ) {
+//			$wgLoadScript = "$wgScriptPath/load.php";
+//		}
+//
+//		if ( $wgArticlePath === false ) {
+//			if ( $wgUsePathInfo ) {
+//				$wgArticlePath = "$wgScript/$1";
+//			} else {
+//				$wgArticlePath = "$wgScript?title=$1";
+//			}
+//		}
+//
+//		if ( !empty( $wgActionPaths ) && !isset( $wgActionPaths['view'] ) ) {
+//			// 'view' is assumed the default action path everywhere in the code
+//			// but is rarely filled in $wgActionPaths
+//			$wgActionPaths['view'] = $wgArticlePath;
+//		}
+//
+//		if ( $wgResourceBasePath === null ) {
+//			$wgResourceBasePath = $wgScriptPath;
+//		}
+//		if ( $wgStylePath === false ) {
+//			$wgStylePath = "$wgResourceBasePath/skins";
+//		}
+//		if ( $wgLocalStylePath === false ) {
+//			// Avoid wgResourceBasePath here since that may point to a different domain (e.g. CDN)
+//			$wgLocalStylePath = "$wgScriptPath/skins";
+//		}
+//		if ( $wgExtensionAssetsPath === false ) {
+//			$wgExtensionAssetsPath = "$wgResourceBasePath/extensions";
+//		}
+//
+//		if ( $wgLogo === false ) {
+//			$wgLogo = "$wgResourceBasePath/resources/assets/wiki.png";
+//		}
+//
+//		if ( $wgUploadPath === false ) {
+//			$wgUploadPath = "$wgScriptPath/images";
+//		}
+//		if ( $wgUploadDirectory === false ) {
+//			$wgUploadDirectory = "$IP/images";
+//		}
+//		if ( $wgReadOnlyFile === false ) {
+//			$wgReadOnlyFile = "{$wgUploadDirectory}/lock_yBgMBwiR";
+//		}
+//		if ( $wgFileCacheDirectory === false ) {
+//			$wgFileCacheDirectory = "{$wgUploadDirectory}/cache";
+//		}
+//		if ( $wgDeletedDirectory === false ) {
+//			$wgDeletedDirectory = "{$wgUploadDirectory}/deleted";
+//		}
+//
+//		if ( $wgGitInfoCacheDirectory === false && $wgCacheDirectory !== false ) {
+//			$wgGitInfoCacheDirectory = "{$wgCacheDirectory}/gitinfo";
+//		}
+//
+//		if ( $wgEnableParserCache === false ) {
+//			$wgParserCacheType = CACHE_NONE;
+//		}
+//
+//		// Fix path to icon images after they were moved in 1.24
+//		if ( $wgRightsIcon ) {
+//			$wgRightsIcon = str_replace(
+//				"{$wgStylePath}/common/images/",
+//				"{$wgResourceBasePath}/resources/assets/licenses/",
+//				$wgRightsIcon
+//			);
+//		}
+//
+//		if ( isset( $wgFooterIcons['copyright']['copyright'] )
+//			&& $wgFooterIcons['copyright']['copyright'] === []
+//		) {
+//			if ( $wgRightsIcon || $wgRightsText ) {
+//				$wgFooterIcons['copyright']['copyright'] = [
+//					'url' => $wgRightsUrl,
+//					'src' => $wgRightsIcon,
+//					'alt' => $wgRightsText,
+//				];
+//			}
+//		}
+//
+//		if ( isset( $wgFooterIcons['poweredby'] )
+//			&& isset( $wgFooterIcons['poweredby']['mediawiki'] )
+//			&& $wgFooterIcons['poweredby']['mediawiki']['src'] === null
+//		) {
+//			$wgFooterIcons['poweredby']['mediawiki']['src'] =
+//				"$wgResourceBasePath/resources/assets/poweredby_mediawiki_88x31.png";
+//			$wgFooterIcons['poweredby']['mediawiki']['srcset'] =
+//				"$wgResourceBasePath/resources/assets/poweredby_mediawiki_132x47.png 1.5x, " .
+//				"$wgResourceBasePath/resources/assets/poweredby_mediawiki_176x62.png 2x";
+//		}
+//
+//		/**
+//		* Unconditional protection for NS_MEDIAWIKI since otherwise it's too easy for a
+//		* sysadmin to set $wgNamespaceProtection incorrectly and leave the wiki insecure.
+//		*
+//		* Note that this is the definition of editinterface and it can be granted to
+//		* all users if desired.
+//		*/
+//		$wgNamespaceProtection[NS_MEDIAWIKI] = 'editinterface';
+
+	/**
+	* The canonical names of namespaces 6 and 7 are, as of v1.14, "File"
+	* and "File_talk".  The old names "Image" and "Image_talk" are
+	* retained as aliases for backwards compatibility.
+	*/
+	public static final    XomwNamespacesByName wgNamespaceAliases = new XomwNamespacesByName()
+		.Add("Image", XomwDefines.NS_FILE)
+		.Add("Image_talk", XomwDefines.NS_FILE_TALK)
+		;
+
+//		/**
+//		* Initialise $wgLockManagers to include basic FS version
+//		*/
+//		$wgLockManagers[] = [
+//			'name' => 'fsLockManager',
+//			'class' => 'FSLockManager',
+//			'lockDirectory' => "{$wgUploadDirectory}/lockdir",
+//		];
+//		$wgLockManagers[] = [
+//			'name' => 'nullLockManager',
+//			'class' => 'NullLockManager',
+//		];
+//
+//		/**
+//		* Initialise $wgLocalFileRepo from backwards-compatible settings
+//		*/
+//		if ( !$wgLocalFileRepo ) {
+//			$wgLocalFileRepo = [
+//				'class' => 'LocalRepo',
+//				'name' => 'local',
+//				'directory' => $wgUploadDirectory,
+//				'scriptDirUrl' => $wgScriptPath,
+//				'scriptExtension' => '.php',
+//				'url' => $wgUploadBaseUrl ? $wgUploadBaseUrl . $wgUploadPath : $wgUploadPath,
+//				'hashLevels' => $wgHashedUploadDirectory ? 2 : 0,
+//				'thumbScriptUrl' => $wgThumbnailScriptPath,
+//				'transformVia404' => !$wgGenerateThumbnailOnParse,
+//				'deletedDir' => $wgDeletedDirectory,
+//				'deletedHashLevels' => $wgHashedUploadDirectory ? 3 : 0
+//			];
+//		}
+//		/**
+//		* Initialise shared repo from backwards-compatible settings
+//		*/
+//		if ( $wgUseSharedUploads ) {
+//			if ( $wgSharedUploadDBname ) {
+//				$wgForeignFileRepos[] = [
+//					'class' => 'ForeignDBRepo',
+//					'name' => 'shared',
+//					'directory' => $wgSharedUploadDirectory,
+//					'url' => $wgSharedUploadPath,
+//					'hashLevels' => $wgHashedSharedUploadDirectory ? 2 : 0,
+//					'thumbScriptUrl' => $wgSharedThumbnailScriptPath,
+//					'transformVia404' => !$wgGenerateThumbnailOnParse,
+//					'dbType' => $wgDBtype,
+//					'dbServer' => $wgDBserver,
+//					'dbUser' => $wgDBuser,
+//					'dbPassword' => $wgDBpassword,
+//					'dbName' => $wgSharedUploadDBname,
+//					'dbFlags' => ( $wgDebugDumpSql ? DBO_DEBUG : 0 ) | DBO_DEFAULT,
+//					'tablePrefix' => $wgSharedUploadDBprefix,
+//					'hasSharedCache' => $wgCacheSharedUploads,
+//					'descBaseUrl' => $wgRepositoryBaseUrl,
+//					'fetchDescription' => $wgFetchCommonsDescriptions,
+//				];
+//			} else {
+//				$wgForeignFileRepos[] = [
+//					'class' => 'FileRepo',
+//					'name' => 'shared',
+//					'directory' => $wgSharedUploadDirectory,
+//					'url' => $wgSharedUploadPath,
+//					'hashLevels' => $wgHashedSharedUploadDirectory ? 2 : 0,
+//					'thumbScriptUrl' => $wgSharedThumbnailScriptPath,
+//					'transformVia404' => !$wgGenerateThumbnailOnParse,
+//					'descBaseUrl' => $wgRepositoryBaseUrl,
+//					'fetchDescription' => $wgFetchCommonsDescriptions,
+//				];
+//			}
+//		}
+//		if ( $wgUseInstantCommons ) {
+//			$wgForeignFileRepos[] = [
+//				'class' => 'ForeignAPIRepo',
+//				'name' => 'wikimediacommons',
+//				'apibase' => 'https://commons.wikimedia.org/w/api.php',
+//				'url' => 'https://upload.wikimedia.org/wikipedia/commons',
+//				'thumbUrl' => 'https://upload.wikimedia.org/wikipedia/commons/thumb',
+//				'hashLevels' => 2,
+//				'transformVia404' => true,
+//				'fetchDescription' => true,
+//				'descriptionCacheExpiry' => 43200,
+//				'apiThumbCacheExpiry' => 0,
+//			];
+//		}
+//		/*
+//		* Add on default file backend config for file repos.
+//		* FileBackendGroup will handle initializing the backends.
+//		*/
+//		if ( !isset( $wgLocalFileRepo['backend'] ) ) {
+//			$wgLocalFileRepo['backend'] = $wgLocalFileRepo['name'] . '-backend';
+//		}
+//		foreach ( $wgForeignFileRepos as &$repo ) {
+//			if ( !isset( $repo['directory'] ) && $repo['class'] === 'ForeignAPIRepo' ) {
+//				$repo['directory'] = $wgUploadDirectory; // b/c
+//			}
+//			if ( !isset( $repo['backend'] ) ) {
+//				$repo['backend'] = $repo['name'] . '-backend';
+//			}
+//		}
+//		unset( $repo ); // no global pollution; destroy reference
+//
+//		$rcMaxAgeDays = $wgRCMaxAge / ( 3600 * 24 );
+//		if ( $wgRCFilterByAge ) {
+//			// Trim down $wgRCLinkDays so that it only lists links which are valid
+//			// as determined by $wgRCMaxAge.
+//			// Note that we allow 1 link higher than the max for things like 56 days but a 60 day link.
+//			sort( $wgRCLinkDays );
+//
+//			// @codingStandardsIgnoreStart Generic.CodeAnalysis.ForLoopWithTestFunctionCall.NotAllowed
+//			for ( $i = 0; $i < count( $wgRCLinkDays ); $i++ ) {
+//				// @codingStandardsIgnoreEnd
+//				if ( $wgRCLinkDays[$i] >= $rcMaxAgeDays ) {
+//					$wgRCLinkDays = array_slice( $wgRCLinkDays, 0, $i + 1, false );
+//					break;
+//				}
+//			}
+//		}
+//		// Ensure that default user options are not invalid, since that breaks Special:Preferences
+//		$wgDefaultUserOptions['rcdays'] = min(
+//			$wgDefaultUserOptions['rcdays'],
+//			ceil( $rcMaxAgeDays )
+//		);
+//		$wgDefaultUserOptions['watchlistdays'] = min(
+//			$wgDefaultUserOptions['watchlistdays'],
+//			ceil( $rcMaxAgeDays )
+//		);
+//		unset( $rcMaxAgeDays );
+//
+//		if ( $wgSkipSkin ) {
+//			$wgSkipSkins[] = $wgSkipSkin;
+//		}
+//
+//		$wgSkipSkins[] = 'fallback';
+//		$wgSkipSkins[] = 'apioutput';
+//
+//		if ( $wgLocalInterwiki ) {
+//			array_unshift( $wgLocalInterwikis, $wgLocalInterwiki );
+//		}
+//
+//		// Set default shared prefix
+//		if ( $wgSharedPrefix === false ) {
+//			$wgSharedPrefix = $wgDBprefix;
+//		}
+//
+//		// Set default shared schema
+//		if ( $wgSharedSchema === false ) {
+//			$wgSharedSchema = $wgDBmwschema;
+//		}
+//
+//		if ( !$wgCookiePrefix ) {
+//			if ( $wgSharedDB && $wgSharedPrefix && in_array( 'user', $wgSharedTables ) ) {
+//				$wgCookiePrefix = $wgSharedDB . '_' . $wgSharedPrefix;
+//			} elseif ( $wgSharedDB && in_array( 'user', $wgSharedTables ) ) {
+//				$wgCookiePrefix = $wgSharedDB;
+//			} elseif ( $wgDBprefix ) {
+//				$wgCookiePrefix = $wgDBname . '_' . $wgDBprefix;
+//			} else {
+//				$wgCookiePrefix = $wgDBname;
+//			}
+//		}
+//		$wgCookiePrefix = strtr( $wgCookiePrefix, '=,; +."\'\\[', '__________' );
+//
+//		if ( $wgEnableEmail ) {
+//			$wgUseEnotif = $wgEnotifUserTalk || $wgEnotifWatchlist;
+//		} else {
+//			// Disable all other email settings automatically if $wgEnableEmail
+//			// is set to false. - bug 63678
+//			$wgAllowHTMLEmail = false;
+//			$wgEmailAuthentication = false; // do not require auth if you're not sending email anyway
+//			$wgEnableUserEmail = false;
+//			$wgEnotifFromEditor = false;
+//			$wgEnotifImpersonal = false;
+//			$wgEnotifMaxRecips = 0;
+//			$wgEnotifMinorEdits = false;
+//			$wgEnotifRevealEditorAddress = false;
+//			$wgEnotifUseRealName = false;
+//			$wgEnotifUserTalk = false;
+//			$wgEnotifWatchlist = false;
+//			unset( $wgGroupPermissions['user']['sendemail'] );
+//			$wgUseEnotif = false;
+//			$wgUserEmailUseReplyTo = false;
+//			$wgUsersNotifiedOnAllChanges = [];
+//		}
+//
+//		if ( $wgMetaNamespace === false ) {
+//			$wgMetaNamespace = str_replace( ' ', '_', $wgSitename );
+//		}
+//
+//		// Default value is 2000 or the suhosin limit if it is between 1 and 2000
+//		if ( $wgResourceLoaderMaxQueryLength === false ) {
+//			$suhosinMaxValueLength = (int)ini_get( 'suhosin.get.max_value_length' );
+//			if ( $suhosinMaxValueLength > 0 && $suhosinMaxValueLength < 2000 ) {
+//				$wgResourceLoaderMaxQueryLength = $suhosinMaxValueLength;
+//			} else {
+//				$wgResourceLoaderMaxQueryLength = 2000;
+//			}
+//			unset( $suhosinMaxValueLength );
+//		}
+//
+//		// Ensure the minimum chunk size is less than PHP upload limits or the maximum
+//		// upload size.
+//		$wgMinUploadChunkSize = min(
+//			$wgMinUploadChunkSize,
+//			UploadBase::getMaxUploadSize( 'file' ),
+//			UploadBase::getMaxXophpUploadSize(),
+//			( wfShorthandToInteger(
+//				ini_get( 'post_max_size' ) ?: ini_get( 'hhvm.server.max_post_size' ),
+//				PHP_INT_MAX
+//			) ?: PHP_INT_MAX ) - 1024 // Leave some room for other POST parameters
+//		);
+
+	/**
+	* Definitions of the NS_ constants are in Defines.php
+	* @private
+	*/
+	public static XomwNamespacesById wgCanonicalNamespaceNames = new XomwNamespacesById()
+		.Add(XomwDefines.NS_MEDIA            ,  "Media")
+		.Add(XomwDefines.NS_SPECIAL          ,  "Special")
+		.Add(XomwDefines.NS_TALK             ,  "Talk")
+		.Add(XomwDefines.NS_USER             ,  "User")
+		.Add(XomwDefines.NS_USER_TALK        ,  "User_talk")
+		.Add(XomwDefines.NS_PROJECT          ,  "Project")
+		.Add(XomwDefines.NS_PROJECT_TALK     ,  "Project_talk")
+		.Add(XomwDefines.NS_FILE             ,  "File")
+		.Add(XomwDefines.NS_FILE_TALK        ,  "File_talk")
+		.Add(XomwDefines.NS_MEDIAWIKI        ,  "MediaWiki")
+		.Add(XomwDefines.NS_MEDIAWIKI_TALK   ,  "MediaWiki_talk")
+		.Add(XomwDefines.NS_TEMPLATE         ,  "Template")
+		.Add(XomwDefines.NS_TEMPLATE_TALK    ,  "Template_talk")
+		.Add(XomwDefines.NS_HELP             ,  "Help")
+		.Add(XomwDefines.NS_HELP_TALK        ,  "Help_talk")
+		.Add(XomwDefines.NS_CATEGORY         ,  "Category")
+		.Add(XomwDefines.NS_CATEGORY_TALK    ,  "Category_talk")
+	;
+
+//		/// @todo UGLY UGLY
+//		if ( is_array( $wgExtraNamespaces ) ) {
+//			$wgCanonicalNamespaceNames = $wgCanonicalNamespaceNames + $wgExtraNamespaces;
+//		}
+//
+//		// These are now the same, always
+//		// To determine the user language, use $wgLang->getCode()
+//		$wgContLanguageCode = $wgLanguageCode;
+//
+//		// Easy to forget to falsify $wgDebugToolbar for static caches.
+//		// If file cache or CDN cache is on, just disable this (DWIMD).
+//		if ( $wgUseFileCache || $wgUseSquid ) {
+//			$wgDebugToolbar = false;
+//		}
+//
+//		// We always output HTML5 since 1.22, overriding these is no longer supported
+//		// we set them here for extensions that depend on its value.
+//		$wgHtml5 = true;
+//		$wgXhtmlDefaultNamespace = 'http://www.w3.org/1999/xhtml';
+//		$wgJsMimeType = 'text/javascript';
+//
+//		// Blacklisted file extensions shouldn't appear on the "allowed" list
+//		$wgFileExtensions = array_values( array_diff( $wgFileExtensions, $wgFileBlacklist ) );
+//
+//		if ( $wgInvalidateCacheOnLocalSettingsChange ) {
+//			MediaWiki\suppressWarnings();
+//			$wgCacheEpoch = max( $wgCacheEpoch, gmdate( 'YmdHis', filemtime( "$IP/LocalSettings.php" ) ) );
+//			MediaWiki\restoreWarnings();
+//		}
+//
+//		if ( $wgNewUserLog ) {
+//			// Add a new log type
+//			$wgLogTypes[] = 'newusers';
+//			$wgLogNames['newusers'] = 'newuserlogpage';
+//			$wgLogHeaders['newusers'] = 'newuserlogpagetext';
+//			$wgLogActionsHandlers['newusers/newusers'] = 'NewUsersLogFormatter';
+//			$wgLogActionsHandlers['newusers/create'] = 'NewUsersLogFormatter';
+//			$wgLogActionsHandlers['newusers/create2'] = 'NewUsersLogFormatter';
+//			$wgLogActionsHandlers['newusers/byemail'] = 'NewUsersLogFormatter';
+//			$wgLogActionsHandlers['newusers/autocreate'] = 'NewUsersLogFormatter';
+//		}
+//
+//		if ( $wgPageLanguageUseDB ) {
+//			$wgLogTypes[] = 'pagelang';
+//			$wgLogActionsHandlers['pagelang/pagelang'] = 'PageLangLogFormatter';
+//		}
+//
+//		if ( $wgCookieSecure === 'detect' ) {
+//			$wgCookieSecure = ( WebRequest::detectProtocol() === 'https' );
+//		}
+//
+//		if ( $wgProfileOnly ) {
+//			$wgDebugLogGroups['profileoutput'] = $wgDebugLogFile;
+//			$wgDebugLogFile = '';
+//		}
+//
+//		// Backwards compatibility with old password limits
+//		if ( $wgMinimalPasswordLength !== false ) {
+//			$wgPasswordPolicy['policies']['default']['MinimalPasswordLength'] = $wgMinimalPasswordLength;
+//		}
+//
+//		if ( $wgMaximalPasswordLength !== false ) {
+//			$wgPasswordPolicy['policies']['default']['MaximalPasswordLength'] = $wgMaximalPasswordLength;
+//		}
+//
+//		// Backwards compatibility warning
+//		if ( !$wgSessionsInObjectCache ) {
+//			wfDeprecated( '$wgSessionsInObjectCache = false', '1.27' );
+//			if ( $wgSessionHandler ) {
+//				wfDeprecated( '$wgSessionsHandler', '1.27' );
+//			}
+//			$cacheType = get_class( ObjectCache::getInstance( $wgSessionCacheType ) );
+//			wfDebugLog(
+//				'caches',
+//				"Session data will be stored in \"$cacheType\" cache with " .
+//					"expiry $wgObjectCacheSessionExpiry seconds"
+//			);
+//		}
+//		$wgSessionsInObjectCache = true;
+//
+//		if ( $wgPHPSessionHandling !== 'enable' &&
+//			$wgPHPSessionHandling !== 'warn' &&
+//			$wgPHPSessionHandling !== 'disable'
+//		) {
+//			$wgPHPSessionHandling = 'warn';
+//		}
+//		if ( defined( 'MW_NO_SESSION' ) ) {
+//			// If the entry point wants no session, force 'disable' here unless they
+//			// specifically set it to the (undocumented) 'warn'.
+//			$wgPHPSessionHandling = MW_NO_SESSION === 'warn' ? 'warn' : 'disable';
+//		}
+//
+//		Profiler::instance()->scopedProfileOut( $ps_default );
+//
+//		// Disable MWDebug for command line mode, this prevents MWDebug from eating up
+//		// all the memory from logging SQL queries on maintenance scripts
+//		global $wgCommandLineMode;
+//		if ( $wgDebugToolbar && !$wgCommandLineMode ) {
+//			MWDebug::init();
+//		}
+//
+//		// Reset the global service locator, so any services that have already been created will be
+//		// re-created while taking into account any custom settings and extensions.
+//		MediaWikiServices::resetGlobalInstance( new GlobalVarConfig(), 'quick' );
+//
+//		if ( $wgSharedDB && $wgSharedTables ) {
+//			// Apply $wgSharedDB table aliases for the local LB (all non-foreign DB connections)
+//			MediaWikiServices::getInstance()->getDBLoadBalancer()->setTableAliases(
+//				array_fill_keys(
+//					$wgSharedTables,
+//					[
+//						'dbname' => $wgSharedDB,
+//						'schema' => $wgSharedSchema,
+//						'prefix' => $wgSharedPrefix
+//					]
+//				)
+//			);
+//		}
+//
+//		// Define a constant that indicates that the bootstrapping of the service locator
+//		// is complete.
+//		define( 'MW_SERVICE_BOOTSTRAP_COMPLETE', 1 );
+//
+//		// Install a header callback to prevent caching of responses with cookies (T127993)
+//		if ( !$wgCommandLineMode ) {
+//			header_register_callback( function () {
+//				$headers = [];
+//				foreach ( headers_list() as $header ) {
+//					list( $name, $value ) = explode( ':', $header, 2 );
+//					$headers[strtolower( trim( $name ) )][] = trim( $value );
+//				}
+//
+//				if ( isset( $headers['set-cookie'] ) ) {
+//					$cacheControl = isset( $headers['cache-control'] )
+//						? implode( ', ', $headers['cache-control'] )
+//						: '';
+//
+//					if ( !preg_match( '/(?:^|,)\s*(?:private|no-cache|no-store)\s*(?:$|,)/i', $cacheControl ) ) {
+//						header( 'Expires: Thu, 01 Jan 1970 00:00:00 GMT' );
+//						header( 'Cache-Control: private, max-age=0, s-maxage=0' );
+//						MediaWiki\Logger\LoggerFactory::getInstance( 'cache-cookies' )->warning(
+//							'Cookies set on {url} with Cache-Control "{cache-control}"', [
+//								'url' => WebRequest::getGlobalRequestURL(),
+//								'cookies' => $headers['set-cookie'],
+//								'cache-control' => $cacheControl ?: '<not set>',
+//							]
+//						);
+//					}
+//				}
+//			} );
+//		}
+//
+//		MWExceptionHandler::installHandler();
+//
+//		require_once "$IP/includes/compat/normal/UtfNormalUtil.php";
+//
+//		$ps_validation = Profiler::instance()->scopedProfileIn( $fname . '-validation' );
+//
+//		// T48998: Bail out early if $wgArticlePath is non-absolute
+//		foreach ( [ 'wgArticlePath', 'wgVariantArticlePath' ] as $varName ) {
+//			if ( $$varName && !preg_match( '/^(https?:\/\/|\/)/', $$varName ) ) {
+//				throw new FatalError(
+//					"If you use a relative URL for \$$varName, it must start " .
+//					'with a slash (<code>/</code>).<br><br>See ' .
+//					"<a href=\"https://www.mediawiki.org/wiki/Manual:\$$varName\">" .
+//					"https://www.mediawiki.org/wiki/Manual:\$$varName</a>."
+//				);
+//			}
+//		}
+//
+//		Profiler::instance()->scopedProfileOut( $ps_validation );
+//
+//		$ps_default2 = Profiler::instance()->scopedProfileIn( $fname . '-defaults2' );
+//
+//		if ( $wgCanonicalServer === false ) {
+//			$wgCanonicalServer = wfExpandUrl( $wgServer, PROTO_HTTP );
+//		}
+//
+//		// Set server name
+//		$serverParts = wfParseUrl( $wgCanonicalServer );
+//		if ( $wgServerName !== false ) {
+//			wfWarn( '$wgServerName should be derived from $wgCanonicalServer, '
+//				. 'not customized. Overwriting $wgServerName.' );
+//		}
+//		$wgServerName = $serverParts['host'];
+//		unset( $serverParts );
+//
+//		// Set defaults for configuration variables
+//		// that are derived from the server name by default
+//		// Note: $wgEmergencyContact and $wgPasswordSender may be false or empty String (T104142)
+//		if ( !$wgEmergencyContact ) {
+//			$wgEmergencyContact = 'wikiadmin@' . $wgServerName;
+//		}
+//		if ( !$wgPasswordSender ) {
+//			$wgPasswordSender = 'apache@' . $wgServerName;
+//		}
+//		if ( !$wgNoReplyAddress ) {
+//			$wgNoReplyAddress = $wgPasswordSender;
+//		}
+//
+//		if ( $wgSecureLogin && substr( $wgServer, 0, 2 ) !== '//' ) {
+//			$wgSecureLogin = false;
+//			wfWarn( 'Secure login was enabled on a server that only supports '
+//				. 'HTTP or HTTPS. Disabling secure login.' );
+//		}
+//
+//		$wgVirtualRestConfig['global']['domain'] = $wgCanonicalServer;
+//
+//		// Now that GlobalFunctions is loaded, set defaults that depend on it.
+//		if ( $wgTmpDirectory === false ) {
+//			$wgTmpDirectory = wfTempDir();
+//		}
+//
+//		// We don't use counters anymore. Left here for extensions still
+//		// expecting this to exist. Should be removed sometime 1.26 or later.
+//		if ( !isset( $wgDisableCounters ) ) {
+//			$wgDisableCounters = true;
+//		}
+//
+//		if ( $wgMainWANCache === false ) {
+//			// Setup a WAN cache from $wgMainCacheType with no relayer.
+//			// Sites using multiple datacenters can configure a relayer.
+//			$wgMainWANCache = 'mediawiki-main-default';
+//			$wgWANObjectCaches[$wgMainWANCache] = [
+//				'class'    => 'WANObjectCache',
+//				'cacheId'  => $wgMainCacheType,
+//				'channels' => [ 'purge' => 'wancache-main-default-purge' ]
+//			];
+//		}
+//
+//		Profiler::instance()->scopedProfileOut( $ps_default2 );
+//
+//		$ps_misc = Profiler::instance()->scopedProfileIn( $fname . '-misc1' );
+//
+//		// Raise the memory limit if it's too low
+//		wfMemoryLimit();
+//
+//		/**
+//		* Set up the timezone, suppressing the pseudo-security warning in PHP 5.1+
+//		* that happens whenever you use a date function without the timezone being
+//		* explicitly set. Inspired by phpMyAdmin's treatment of the problem.
+//		*/
+//		if ( is_null( $wgLocaltimezone ) ) {
+//			MediaWiki\suppressWarnings();
+//			$wgLocaltimezone = date_default_timezone_get();
+//			MediaWiki\restoreWarnings();
+//		}
+//
+//		date_default_timezone_set( $wgLocaltimezone );
+//		if ( is_null( $wgLocalTZoffset ) ) {
+//			$wgLocalTZoffset = date( 'Z' ) / 60;
+//		}
+//		// The part after the System| is ignored, but rest of MW fills it
+//		// out as the local offset.
+//		$wgDefaultUserOptions['timecorrection'] = "System|$wgLocalTZoffset";
+//
+//		if ( !$wgDBerrorLogTZ ) {
+//			$wgDBerrorLogTZ = $wgLocaltimezone;
+//		}
+//
+//		// initialize the request Object in $wgRequest
+//		$wgRequest = RequestContext::getMain()->getRequest(); // BackCompat
+//		// Set user IP/agent information for causal consistency purposes
+//		MediaWikiServices::getInstance()->getDBLoadBalancerFactory()->setRequestInfo( [
+//			'IPAddress' => $wgRequest->getIP(),
+//			'UserAgent' => $wgRequest->getHeader( 'User-Agent' ),
+//			'ChronologyProtection' => $wgRequest->getHeader( 'ChronologyProtection' )
+//		] );
+//
+//		// Useful debug output
+//		if ( $wgCommandLineMode ) {
+//			wfDebug( "\n\nStart command line script $self\n" );
+//		} else {
+//			$debug = "\n\nStart request {$wgRequest->getMethod()} {$wgRequest->getRequestURL()}\n";
+//
+//			if ( $wgDebugPrintHttpHeaders ) {
+//				$debug .= "HTTP HEADERS:\n";
+//
+//				foreach ( $wgRequest->getAllHeaders() as $name => $value ) {
+//					$debug .= "$name: $value\n";
+//				}
+//			}
+//			wfDebug( $debug );
+//		}
+//
+//		Profiler::instance()->scopedProfileOut( $ps_misc );
+//		$ps_memcached = Profiler::instance()->scopedProfileIn( $fname . '-memcached' );
+//
+//		$wgMemc = wfGetMainCache();
+//		$messageMemc = wfGetMessageCacheStorage();
+//		$parserMemc = wfGetParserCacheStorage();
+//
+//		wfDebugLog( 'caches',
+//			'cluster: ' . get_class( $wgMemc ) .
+//			', WAN: ' . ( $wgMainWANCache === CACHE_NONE ? 'CACHE_NONE' : $wgMainWANCache ) .
+//			', stash: ' . $wgMainStash .
+//			', message: ' . get_class( $messageMemc ) .
+//			', parser: ' . get_class( $parserMemc ) .
+//			', session: ' . get_class( ObjectCache::getInstance( $wgSessionCacheType ) )
+//		);
+//
+//		Profiler::instance()->scopedProfileOut( $ps_memcached );
+//
+//		// Most of the config is out, some might want to run hooks here.
+//		Hooks::run( 'SetupAfterCache' );
+//
+//		$ps_globals = Profiler::instance()->scopedProfileIn( $fname . '-globals' );
+//
+//		/**
+//		* @var Language $wgContLang
+//		*/
+//		$wgContLang = Language::factory( $wgLanguageCode );
+//		$wgContLang->initContLang();
+//
+//		// Now that variant lists may be available...
+//		$wgRequest->interpolateTitle();
+//
+//		if ( !is_object( $wgAuth ) ) {
+//			$wgAuth = new MediaWiki\Auth\AuthManagerAuthPlugin;
+//			Hooks::run( 'AuthPluginSetup', [ &$wgAuth ] );
+//		}
+//		if ( $wgAuth && !$wgAuth instanceof MediaWiki\Auth\AuthManagerAuthPlugin ) {
+//			MediaWiki\Auth\AuthManager::singleton()->forcePrimaryAuthenticationProviders( [
+//				new MediaWiki\Auth\TemporaryPasswordPrimaryAuthenticationProvider( [
+//					'authoritative' => false,
+//				] ),
+//				new MediaWiki\Auth\AuthPluginPrimaryAuthenticationProvider( $wgAuth ),
+//				new MediaWiki\Auth\LocalPasswordPrimaryAuthenticationProvider( [
+//					'authoritative' => true,
+//				] ),
+//			], '$wgAuth is ' . get_class( $wgAuth ) );
+//		}
+//
+//		// Set up the session
+//		$ps_session = Profiler::instance()->scopedProfileIn( $fname . '-session' );
+//		/**
+//		* @var MediaWiki\Session\SessionId|null $wgInitialSessionId The persistent
+//		* session ID (if any) loaded at startup
+//		*/
+//		$wgInitialSessionId = null;
+//		if ( !defined( 'MW_NO_SESSION' ) && !$wgCommandLineMode ) {
+//			// If session.auto_start is there, we can't touch session name
+//			if ( $wgPHPSessionHandling !== 'disable' && !wfIniGetBool( 'session.auto_start' ) ) {
+//				session_name( $wgSessionName ? $wgSessionName : $wgCookiePrefix . '_session' );
+//			}
+//
+//			// Create the SessionManager singleton and set up our session handler,
+//			// unless we're specifically asked not to.
+//			if ( !defined( 'MW_NO_SESSION_HANDLER' ) ) {
+//				MediaWiki\Session\PHPSessionHandler::install(
+//					MediaWiki\Session\SessionManager::singleton()
+//				);
+//			}
+//
+//			// Initialize the session
+//			try {
+//				$session = MediaWiki\Session\SessionManager::getGlobalSession();
+//			} catch ( OverflowException $ex ) {
+//				if ( isset( $ex->sessionInfos ) && count( $ex->sessionInfos ) >= 2 ) {
+//					// The exception is because the request had multiple possible
+//					// sessions tied for top priority. Report this to the user.
+//					$list = [];
+//					foreach ( $ex->sessionInfos as $info ) {
+//						$list[] = $info->getProvider()->describe( $wgContLang );
+//					}
+//					$list = $wgContLang->listToText( $list );
+//					throw new HttpError( 400,
+//						Message::newFromKey( 'sessionmanager-tie', $list )->inLanguage( $wgContLang )->plain()
+//					);
+//				}
+//
+//				// Not the one we want, rethrow
+//				throw $ex;
+//			}
+//
+//			if ( $session->isPersistent() ) {
+//				$wgInitialSessionId = $session->getSessionId();
+//			}
+//
+//			$session->renew();
+//			if ( MediaWiki\Session\PHPSessionHandler::isEnabled() &&
+//				( $session->isPersistent() || $session->shouldRememberUser() )
+//			) {
+//				// Start the PHP-session for backwards compatibility
+//				session_id( $session->getId() );
+//				MediaWiki\quietCall( 'session_start' );
+//			}
+//
+//			unset( $session );
+//		} else {
+//			// Even if we didn't set up a global Session, still install our session
+//			// handler unless specifically requested not to.
+//			if ( !defined( 'MW_NO_SESSION_HANDLER' ) ) {
+//				MediaWiki\Session\PHPSessionHandler::install(
+//					MediaWiki\Session\SessionManager::singleton()
+//				);
+//			}
+//		}
+//		Profiler::instance()->scopedProfileOut( $ps_session );
+//
+//		/**
+//		* @var User $wgUser
+//		*/
+//		$wgUser = RequestContext::getMain()->getUser(); // BackCompat
+//
+//		/**
+//		* @var Language $wgLang
+//		*/
+//		$wgLang = new StubUserLang;
+//
+//		/**
+//		* @var OutputPage $wgOut
+//		*/
+//		$wgOut = RequestContext::getMain()->getOutput(); // BackCompat
+//
+//		/**
+//		* @var Parser $wgParser
+//		*/
+//		$wgParser = new StubObject( 'wgParser', function () {
+//			return MediaWikiServices::getInstance()->getParser();
+//		} );
+//
+//		/**
+//		* @var Title $wgTitle
+//		*/
+//		$wgTitle = null;
+//
+//		Profiler::instance()->scopedProfileOut( $ps_globals );
+//		$ps_extensions = Profiler::instance()->scopedProfileIn( $fname . '-extensions' );
+//
+//		// Extension setup functions
+//		// Entries should be added to this variable during the inclusion
+//		// of the extension file. This allows the extension to perform
+//		// any necessary initialisation in the fully initialised environment
+//		foreach ( $wgExtensionFunctions as $func ) {
+//			// Allow closures in PHP 5.3+
+//			if ( is_object( $func ) && $func instanceof Closure ) {
+//				$profName = $fname . '-extensions-closure';
+//			} elseif ( is_array( $func ) ) {
+//				if ( is_object( $func[0] ) ) {
+//					$profName = $fname . '-extensions-' . get_class( $func[0] ) . '::' . $func[1];
+//				} else {
+//					$profName = $fname . '-extensions-' . implode( '::', $func );
+//				}
+//			} else {
+//				$profName = $fname . '-extensions-' . strval( $func );
+//			}
+//
+//			$ps_ext_func = Profiler::instance()->scopedProfileIn( $profName );
+//			call_user_func( $func );
+//			Profiler::instance()->scopedProfileOut( $ps_ext_func );
+//		}
+//
+//		// If the session user has a 0 id but a valid name, that means we need to
+//		// autocreate it.
+//		if ( !defined( 'MW_NO_SESSION' ) && !$wgCommandLineMode ) {
+//			$sessionUser = MediaWiki\Session\SessionManager::getGlobalSession()->getUser();
+//			if ( $sessionUser->getId() === 0 && User::isValidUserName( $sessionUser->getName() ) ) {
+//				$ps_autocreate = Profiler::instance()->scopedProfileIn( $fname . '-autocreate' );
+//				$res = MediaWiki\Auth\AuthManager::singleton()->autoCreateUser(
+//					$sessionUser,
+//					MediaWiki\Auth\AuthManager::AUTOCREATE_SOURCE_SESSION,
+//					true
+//				);
+//				Profiler::instance()->scopedProfileOut( $ps_autocreate );
+//				\MediaWiki\Logger\LoggerFactory::getInstance( 'authevents' )->info( 'Autocreation attempt', [
+//					'event' => 'autocreate',
+//					'status' => $res,
+//				] );
+//				unset( $res );
+//			}
+//			unset( $sessionUser );
+//		}
+//
+//		if ( !$wgCommandLineMode ) {
+//			Pingback::schedulePingback();
+//		}
+//
+//		$wgFullyInitialised = true;
+//
+//		Profiler::instance()->scopedProfileOut( $ps_extensions );
+//		Profiler::instance()->scopedProfileOut( $ps_setup );
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwTitleTest.java

@@ -13,3 +13,16 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XomwTitleTest {
+	private final    XomwTitleFxt fxt = new XomwTitleFxt();
+	@Test  public void Alphanum()           {fxt.Test__find_fwd_while_title("0aB"             , 3);}
+	@Test  public void Angle()              {fxt.Test__find_fwd_while_title("0a<"             , 2);}
+}
+class XomwTitleFxt {
+	public void Test__find_fwd_while_title(String src_str, int expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		Gftest.Eq__int(expd, XomwTitle.Find_fwd_while_title(src_bry, 0, src_bry.length, XomwTitle.Title_chars_valid()));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/XomwXml.java

@@ -13,3 +13,71 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*;
+public class XomwXml {
+	// Format an XML element with given attributes and, optionally, text content.
+	// Element and attribute names are assumed to be ready for literal inclusion.
+	// Strings are assumed to not contain XML-illegal characters; special
+	// characters (<, >, &) are escaped but illegals are not touched.
+	// ARGS: contents defaults to ""
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	public static void Element(Bry_bfr bfr, byte[] element, List_adp attribs, byte[] contents, boolean allow_short_tag) {
+		bfr.Add_byte(Byte_ascii.Angle_bgn).Add(element);
+		if (attribs.Len() > 0) {
+			Expand_attributes(bfr, attribs);
+		}
+		if (contents == null) {
+			bfr.Add_byte(Byte_ascii.Angle_end);
+		}
+		else {
+			if (allow_short_tag && contents == Bry_.Empty) {
+				bfr.Add_str_a7(" />");
+			}
+			else {
+				bfr.Add_byte(Byte_ascii.Angle_end);
+				bfr.Add_bry_escape_html(contents);
+				bfr.Add_byte(Byte_ascii.Angle_bgn).Add_byte(Byte_ascii.Slash).Add(element).Add_byte(Byte_ascii.Angle_end);
+			}
+		}
+	}
+	// Given an array of ('attributename' => 'value'), it generates the code
+	// to set the XML attributes : attributename="value".
+	// The values are passed to Sanitizer::encodeAttribute.
+	// Return null if no attributes given.
+	// @param array $attribs Array of attributes for an XML element
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	public static void Expand_attributes(Bry_bfr bfr, List_adp attribs) {
+		int attribs_len = attribs.Len();
+		for (int i = 0; i < attribs_len; i += 2) {
+			// XO.MW: $out .= " {$name}=\"" . Sanitizer::encodeAttribute( $val ) . '"';
+			bfr.Add_byte_space();
+			bfr.Add((byte[])attribs.Get_at(i));
+			bfr.Add_byte_eq().Add_byte_quote();
+			XomwSanitizer.encodeAttribute(bfr, (byte[])attribs.Get_at(i + 1));
+			bfr.Add_byte_quote();
+		}
+	}
+
+	// This opens an XML element
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	public static void Open_element(Bry_bfr bfr, byte[] element, List_adp attribs) {
+		bfr.Add_byte(Byte_ascii.Angle_bgn).Add(element);
+		Expand_attributes(bfr, attribs);
+		bfr.Add_byte(Byte_ascii.Angle_end);
+	}
+
+	// Shortcut to close an XML element
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	public static void Close_element(Bry_bfr bfr, byte[] element) {
+		bfr.Add_byte(Byte_ascii.Angle_bgn).Add_byte(Byte_ascii.Slash).Add(element).Add_byte(Byte_ascii.Angle_end);
+	}
+
+	// Same as Xml::element(), but does not escape contents. Handy when the
+	// content you have is already valid xml.
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	public static void Tags(Bry_bfr bfr, byte[] element, List_adp attribs, byte[] contents) {
+		Open_element(bfr, element, attribs);
+		bfr.Add(contents);
+		bfr.Add_byte(Byte_ascii.Angle_bgn).Add_byte(Byte_ascii.Slash).Add(element).Add_byte(Byte_ascii.Angle_end);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/content/XomwAbstractContent.java

@@ -13,3 +13,563 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.content; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.exception.*;
+import gplx.xowa.mediawiki.includes.parsers.*;
+/**
+* A content Object represents page content, e.g. the text to show on a page.
+* Content objects have no knowledge about how they relate to Wiki pages.
+*/
+
+/**
+* Base implementation for content objects.
+*
+* @ingroup Content
+*/
+public abstract class XomwAbstractContent implements XomwContent {
+	/**
+	* Name of the content model this Content Object represents.
+	* Use with CONTENT_MODEL_XXX constants
+	*
+	* @since 1.21
+	*
+	* @var String $model_id
+	*/
+	private int model_id;
+
+	/**
+	* @param String $modelId
+	*
+	* @since 1.21
+	*/
+	public XomwAbstractContent(int modelId) {
+		this.model_id = modelId;
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @see Content::getModel
+	*/
+	public int getModel() {
+		return this.model_id;
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @param String $modelId The model to check
+	*
+	* @throws MWException If the provided ID is not the ID of the content model supported by this
+	* Content Object.
+	*/
+	protected void checkModelID(int modelId) {
+		if (modelId != this.model_id) {
+			throw new XomwMWException(
+				"Bad content model: " +
+				"expected " + this.model_id +
+				"but got " + modelId
+			);
+		}
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @see Content::getContentHandler
+	*/
+	public XomwContentHandler getContentHandler() {
+		return XomwContentHandler.getForContent(this);
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @see Content::getDefaultFormat
+	*/
+	public String getDefaultFormat() {
+		return this.getContentHandler().getDefaultFormat();
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @see Content::getSupportedFormats
+	*/
+	public String[] getSupportedFormats() {
+		return this.getContentHandler().getSupportedFormats();
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @param String $format
+	*
+	* @return boolean
+	*
+	* @see Content::isSupportedFormat
+	*/
+	public boolean isSupportedFormat(String format) {
+		if (format == null) {
+			return true; // this means "use the default"
+		}
+
+		return this.getContentHandler().isSupportedFormat(format);
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @param String $format The serialization format to check.
+	*
+	* @throws MWException If the format is not supported by this content handler.
+	*/
+	public void checkFormat(String format) {
+		if (!this.isSupportedFormat(format)) {
+			throw new XomwMWException(
+				"Format " + format + " is not supported for content model " +
+				this.getModel()
+			);
+		}
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @param String $format
+	*
+	* @return String
+	*
+	* @see Content::serialize
+	*/
+	public String serialize(String format) {
+		// return this.getContentHandler().serializeContent(this, format);
+		throw Err_.new_unimplemented();
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @return boolean
+	*
+	* @see Content::isEmpty
+	*/
+	public boolean isEmpty() {
+		return this.getSize() == 0;
+	}
+
+	/**
+	* Subclasses @Override may this to implement (light weight) validation.
+	*
+	* @since 1.21
+	*
+	* @return boolean Always true.
+	*
+	* @see Content::isValid
+	*/
+	@gplx.Virtual public boolean isValid() {
+		return true;
+	}
+
+	/**
+	* @since 1.21
+	*
+	* @param Content that
+	*
+	* @return boolean
+	*
+	* @see Content::equals
+	*/
+	public boolean equals(XomwContent that) {
+		if (that == null) {
+			return false;
+		}
+
+		if (that == this) {
+			return true;
+		}
+
+		if (that.getModel() != this.getModel()) {
+			return false;
+		}
+
+		return this.getNativeData() == that.getNativeData();
+	}
+
+	/**
+	* Returns a list of DataUpdate objects for recording information about this
+	* Content in some secondary data store.
+	*
+	* This default implementation returns a LinksUpdate Object and calls the
+	* SecondaryDataUpdates hook.
+	*
+	* Subclasses @Override may this to determine the secondary data updates more
+	* efficiently, preferably without the need to generate a parser output Object.
+	* They should however make sure to call SecondaryDataUpdates to give extensions
+	* a chance to inject additional updates.
+	*
+	* @since 1.21
+	*
+	* @param Title $title
+	* @param Content $old
+	* @param boolean $recursive
+	* @param ParserOutput parserOutput
+	*
+	* @return DataUpdate[]
+	*
+	* @see Content::getSecondaryDataUpdates()
+	*/
+	// recursive=true
+//		public XomwDataUpdate[] getSecondaryDataUpdates(Title title, Content old,
+//			boolean recursive, ParserOutput parserOutput
+//		) {
+//			if (parserOutput == null) {
+//				parserOutput = this.getParserOutput(title, null, null, false);
+//			}
+//
+//			XomwDataUpdate[] updates = new XomwDataUpdate[] {
+//				new LinksUpdate(title, parserOutput, recursive)
+//															};
+//
+//			Hooks::run('SecondaryDataUpdates', [ $title, $old, $recursive, parserOutput, &$updates ]);
+//
+//			return updates;
+//		}
+
+	/**
+	* @since 1.21
+	*
+	* @return Title[]|null
+	*
+	* @see Content::getRedirectChain
+	*/
+	public XomwTitle[] getRedirectChain() {
+//			XomwTitle title = this.getRedirectTarget();
+//			if (title == null) {
+//				return null;
+//			}
+//			// recursive check to follow double redirects
+//			int recurse = XomwDefaultSettings.wgMaxRedirects;
+//
+//			List_adp titles = List_adp_.New_by_many(title);
+//			while (--recurse > 0) {
+//				XomwTitle newtitle = null;
+//				if (title.isRedirect()) {
+//					$page = WikiPage::factory(title);
+//					$newtitle = $page.getRedirectTarget();
+//				} else {
+//					break;
+//				}
+//				// Redirects to some special pages are not permitted
+//				if (Type_.Eq_by_obj(newtitle, typeof(XomwTitle)) && newtitle.isValidRedirectTarget()) {
+//					// The new title passes the checks, so make that our current
+//					// title so that further recursion can be checked
+//					title = newtitle;
+//					titles.Add(newtitle);
+//				} else {
+//					break;
+//				}
+//			}
+//
+//			return (XomwTitle[])titles.To_ary_and_clear(typeof(XomwTitle));
+		throw Err_.new_unimplemented();
+	}
+
+//		/**
+//		* Subclasses that implement redirects should override this.
+//		*
+//		* @since 1.21
+//		*
+//		* @return Title|null
+//		*
+//		* @see Content::getRedirectTarget
+//		*/
+//		public function getRedirectTarget() {
+//			return null;
+//		}
+//
+//		/**
+//		* @note Migrated here from Title::newFromRedirectRecurse.
+//		*
+//		* @since 1.21
+//		*
+//		* @return Title|null
+//		*
+//		* @see Content::getUltimateRedirectTarget
+//		*/
+//		public function getUltimateRedirectTarget() {
+//			$titles = this.getRedirectChain();
+//
+//			return $titles ? array_pop($titles) : null;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return boolean
+//		*
+//		* @see Content::isRedirect
+//		*/
+//		public function isRedirect() {
+//			return this.getRedirectTarget() != null;
+//		}
+//
+//		/**
+//		* This default implementation always returns $this.
+//		* Subclasses that implement redirects should override this.
+//		*
+//		* @since 1.21
+//		*
+//		* @param Title $target
+//		*
+//		* @return Content $this
+//		*
+//		* @see Content::updateRedirect
+//		*/
+//		public function updateRedirect(Title $target) {
+//			return $this;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return null
+//		*
+//		* @see Content::getSection
+//		*/
+//		public function getSection($sectionId) {
+//			return null;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return null
+//		*
+//		* @see Content::replaceSection
+//		*/
+//		public function replaceSection($sectionId, Content $with, $sectionTitle = '') {
+//			return null;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return Content $this
+//		*
+//		* @see Content::preSaveTransform
+//		*/
+//		public function preSaveTransform(Title $title, User $user, ParserOptions $popts) {
+//			return $this;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return Content $this
+//		*
+//		* @see Content::addSectionHeader
+//		*/
+//		public function addSectionHeader($header) {
+//			return $this;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return Content $this
+//		*
+//		* @see Content::preloadTransform
+//		*/
+//		public function preloadTransform(Title $title, ParserOptions $popts, $params = []) {
+//			return $this;
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @return Status
+//		*
+//		* @see Content::prepareSave
+//		*/
+//		public function prepareSave(WikiPage $page, $flags, $parentRevId, User $user) {
+//			if (this.isValid()) {
+//				return Status::newGood();
+//			} else {
+//				return Status::newFatal("invalid-content-data");
+//			}
+//		}
+//
+//		/**
+//		* @since 1.21
+//		*
+//		* @param WikiPage $page
+//		* @param ParserOutput parserOutput
+//		*
+//		* @return LinksDeletionUpdate[]
+//		*
+//		* @see Content::getDeletionUpdates
+//		*/
+//		public function getDeletionUpdates(WikiPage $page, ParserOutput parserOutput = null) {
+//			return [
+//				new LinksDeletionUpdate($page),
+//			];
+//		}
+
+	/**
+	* This default implementation always returns false. Subclasses @Override may 
+	* this to supply matching logic.
+	*
+	* @since 1.21
+	*
+	* @param MagicWord $word
+	*
+	* @return boolean Always false.
+	*
+	* @see Content::matchMagicWord
+	*/
+	@gplx.Virtual public boolean matchMagicWord(XomwMagicWord word) {
+		return false;
+	}
+
+//		/**
+//		* This super implementation calls the hook ConvertContent to enable custom conversions.
+//		* Subclasses may override this to implement conversion for "their" content model.
+//		*
+//		* @param String $toModel
+//		* @param String $lossy
+//		*
+//		* @return Content|boolean
+//		*
+//		* @see Content::convert()
+//		*/
+//		public function convert($toModel, $lossy = '') {
+//			if (this.getModel() == $toModel) {
+//				// nothing to do, shorten out.
+//				return $this;
+//			}
+//
+//			$lossy = ($lossy == 'lossy'); // String flag, convert to boolean for convenience
+//			$result = false;
+//
+//			Hooks::run('ConvertContent', [ $this, $toModel, $lossy, &$result ]);
+//
+//			return $result;
+//		}
+//
+//		/**
+//		* Returns a ParserOutput Object containing information derived from this content.
+//		* Most importantly, unless $generateHtml was false, the return value contains an
+//		* HTML representation of the content.
+//		*
+//		* Subclasses that want to control the parser output may override this, but it is
+//		* preferred to override fillParserOutput() instead.
+//		*
+//		* Subclasses that override getParserOutput() itself should take care to call the
+//		* ContentGetParserOutput hook.
+//		*
+//		* @since 1.24
+//		*
+//		* @param Title $title Context title for parsing
+//		* @param int|null $revId Revision ID (for {{REVISIONID}})
+//		* @param ParserOptions|null $options Parser options
+//		* @param boolean $generateHtml Whether or not to generate HTML
+//		*
+//		* @return ParserOutput Containing information derived from this content.
+//		*/
+//		public function getParserOutput(Title $title, $revId = null,
+//			ParserOptions $options = null, $generateHtml = true
+//		) {
+//			if ($options == null) {
+//				$options = this.getContentHandler().makeParserOptions('canonical');
+//			}
+//
+//			$po = new ParserOutput();
+//
+//			if (Hooks::run('ContentGetParserOutput',
+//				[ $this, $title, $revId, $options, $generateHtml, &$po ])) {
+//
+//				// Save and restore the old value, just in case something is reusing
+//				// the ParserOptions Object in some weird way.
+//				$oldRedir = $options.getRedirectTarget();
+//				$options.setRedirectTarget(this.getRedirectTarget());
+//				this.fillParserOutput($title, $revId, $options, $generateHtml, $po);
+//				$options.setRedirectTarget($oldRedir);
+//			}
+//
+//			Hooks::run('ContentAlterParserOutput', [ $this, $title, $po ]);
+//
+//			return $po;
+//		}
+//
+//		/**
+//		* Fills the provided ParserOutput with information derived from the content.
+//		* Unless $generateHtml was false, this includes an HTML representation of the content.
+//		*
+//		* This is called by getParserOutput() after consulting the ContentGetParserOutput hook.
+//		* Subclasses are expected to override this method (or getParserOutput(), if need be).
+//		* Subclasses of TextContent should generally override getHtml() instead.
+//		*
+//		* This placeholder implementation always throws an exception.
+//		*
+//		* @since 1.24
+//		*
+//		* @param Title $title Context title for parsing
+//		* @param int|null $revId Revision ID (for {{REVISIONID}})
+//		* @param ParserOptions $options Parser options
+//		* @param boolean $generateHtml Whether or not to generate HTML
+//		* @param ParserOutput &$output The output Object to fill (reference).
+//		*
+//		* @throws MWException
+//		*/
+//		protected function fillParserOutput(Title $title, $revId,
+//			ParserOptions $options, $generateHtml, ParserOutput &$output
+//		) {
+//			// Don't make abstract, so subclasses that override getParserOutput() directly don't fail.
+//			throw new MWException('Subclasses of AbstractContent must override fillParserOutput!');
+//		}
+	public abstract byte[] getTextForSearchIndex();
+
+	public abstract byte[] getWikitextForTransclusion();
+
+	public abstract byte[] getTextForSummary(int maxLength);
+
+	public abstract Object getNativeData();
+
+	public abstract int getSize();
+
+	public abstract XomwContent copy();
+
+	public abstract boolean isCountable(boolean hasLinks);
+
+	public abstract XomwParserOutput getParserOutput(XomwTitle title, int revId,
+		XomwParserOptions options, boolean generateHtml);
+
+	public abstract Object getSecondaryDataUpdates(XomwTitle title, XomwContent old,
+		boolean recursive, XomwParserOutput parserOutput);
+
+	public abstract XomwTitle getRedirectTarget();
+
+	public abstract XomwTitle getUltimateRedirectTarget();
+
+	public abstract boolean isRedirect();
+
+	public abstract XomwContent updateRedirect(XomwTitle target);
+
+	public abstract XomwContent getSection(String sectionId);
+
+	public abstract byte[] replaceSection(String sectionId, XomwContent with, String sectionTitle);
+
+	public abstract XomwContent preSaveTransform(XomwTitle title, Object user, XomwParserOptions parserOptions);
+
+	public abstract XomwContent addSectionHeader(byte[] header);
+
+	public abstract XomwContent preloadTransform(XomwTitle title, XomwParserOptions parserOptions, Object[] ary);
+
+	public abstract Object prepareSave(Object page, int flags, int parentRevId, Object user);
+
+	public abstract Object getDeletionUpdates(Object page,
+		XomwParserOutput parserOutput);
+
+	public abstract XomwContent convert(byte[] toModel, byte[] lossy);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/content/XomwContent.java

@@ -13,3 +13,511 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.content; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.parsers.*;
+/**
+* A content Object represents page content, e.g. the text to show on a page.
+* Content objects have no knowledge about how they relate to wiki pages.
+*/
+
+/**
+* Base interface for content objects.
+*
+* @ingroup Content
+*/
+public interface XomwContent {
+
+	/**
+	* @since 1.21
+	*
+	* @return String A String representing the content in a way useful for
+	*   building a full text search index. If no useful representation exists,
+	*   this method returns an empty String.
+	*
+	* @todo Test that this actually works
+	* @todo Make sure this also works with LuceneSearch / WikiSearch
+	*/
+	byte[] getTextForSearchIndex();
+
+	/**
+	* @since 1.21
+	*
+	* @return String|boolean The wikitext to include when another page includes this
+	* content, or false if the content is not includable in a wikitext page.
+	*
+	* @todo Allow native handling, bypassing wikitext representation, like
+	*  for includable special pages.
+	* @todo Allow transclusion into other content models than Wikitext!
+	* @todo Used in WikiPage and MessageCache to get message text. Not so
+	*  nice. What should we use instead?!
+	*/
+	byte[] getWikitextForTransclusion();
+
+	/**
+	* Returns a textual representation of the content suitable for use in edit
+	* summaries and log messages.
+	*
+	* @since 1.21
+	*
+	* @param int $maxLength Maximum length of the summary text.
+	*
+	* @return String The summary text.
+	*/
+	// DFLT:maxLength=250 
+	byte[] getTextForSummary(int maxLength);
+
+	/**
+	* Returns native representation of the data. Interpretation depends on
+	* the data model used, as given by getDataModel().
+	*
+	* @since 1.21
+	*
+	* @return mixed The native representation of the content. Could be a
+	*    String, a nested array structure, an Object, a binary blob...
+	*    anything, really.
+	*
+	* @note Caller must be aware of content model!
+	*/
+	Object getNativeData();
+
+	/**
+	* Returns the content's nominal size in "bogo-bytes".
+	*
+	* @return int
+	*/
+	int getSize();
+
+	/**
+	* Returns the ID of the content model used by this Content Object.
+	* Corresponds to the CONTENT_MODEL_XXX constants.
+	*
+	* @since 1.21
+	*
+	* @return String The model id
+	*/
+	int getModel();
+
+	/**
+	* Convenience method that returns the ContentHandler singleton for handling
+	* the content model that this Content Object uses.
+	*
+	* Shorthand for ContentHandler::getForContent( $this )
+	*
+	* @since 1.21
+	*
+	* @return ContentHandler
+	*/
+	XomwContentHandler getContentHandler();
+
+	/**
+	* Convenience method that returns the default serialization format for the
+	* content model that this Content Object uses.
+	*
+	* Shorthand for $this->getContentHandler()->getDefaultFormat()
+	*
+	* @since 1.21
+	*
+	* @return String
+	*/
+	String getDefaultFormat();
+
+	/**
+	* Convenience method that returns the list of serialization formats
+	* supported for the content model that this Content Object uses.
+	*
+	* Shorthand for $this->getContentHandler()->getSupportedFormats()
+	*
+	* @since 1.21
+	*
+	* @return String[] List of supported serialization formats
+	*/
+	String[] getSupportedFormats();
+
+	/**
+	* Returns true if $format is a supported serialization format for this
+	* Content Object, false if it isn't.
+	*
+	* Note that this should always return true if $format is null, because null
+	* stands for the default serialization.
+	*
+	* Shorthand for $this->getContentHandler()->isSupportedFormat( $format )
+	*
+	* @since 1.21
+	*
+	* @param String $format The serialization format to check.
+	*
+	* @return boolean Whether the format is supported
+	*/
+	boolean isSupportedFormat(String format);
+
+	/**
+	* Convenience method for serializing this Content Object.
+	*
+	* Shorthand for $this->getContentHandler()->serializeContent( $this, $format )
+	*
+	* @since 1.21
+	*
+	* @param String $format The desired serialization format, or null for the default format.
+	*
+	* @return String Serialized form of this Content Object.
+	*/
+	String serialize(String format);
+
+	/**
+	* Returns true if this Content Object represents empty content.
+	*
+	* @since 1.21
+	*
+	* @return boolean Whether this Content Object is empty
+	*/
+	boolean isEmpty();
+
+	/**
+	* Returns whether the content is valid. This is intended for local validity
+	* checks, not considering global consistency.
+	*
+	* Content needs to be valid before it can be saved.
+	*
+	* This default implementation always returns true.
+	*
+	* @since 1.21
+	*
+	* @return boolean
+	*/
+	boolean isValid();
+
+	/**
+	* Returns true if this Content objects is conceptually equivalent to the
+	* given Content Object.
+	*
+	* Contract:
+	*
+	* - Will return false if $that is null.
+	* - Will return true if $that === $this.
+	* - Will return false if $that->getModel() != $this->getModel().
+	* - Will return false if $that->getNativeData() is not equal to $this->getNativeData(),
+	*   where the meaning of "equal" depends on the actual data model.
+	*
+	* Implementations should be careful to make equals() transitive and reflexive:
+	*
+	* - $a->equals( $b ) <=> $b->equals( $a )
+	* - $a->equals( $b ) &&  $b->equals( $c ) ==> $a->equals( $c )
+	*
+	* @since 1.21
+	*
+	* @param Content $that The Content Object to compare to.
+	*
+	* @return boolean True if this Content Object is equal to $that, false otherwise.
+	*/
+	boolean equals(XomwContent that);
+
+	/**
+	* Return a copy of this Content Object. The following must be true for the
+	* Object returned:
+	*
+	* if $copy = $original->copy()
+	*
+	* - get_class($original) === get_class($copy)
+	* - $original->getModel() === $copy->getModel()
+	* - $original->equals( $copy )
+	*
+	* If and only if the Content Object is immutable, the copy() method can and
+	* should return $this. That is, $copy === $original may be true, but only
+	* for immutable content objects.
+	*
+	* @since 1.21
+	*
+	* @return Content A copy of this Object
+	*/
+	XomwContent copy();
+
+	/**
+	* Returns true if this content is countable as a "real" wiki page, provided
+	* that it's also in a countable location (e.g. a current revision in the
+	* main namespace).
+	*
+	* @since 1.21
+	*
+	* @param boolean|null $hasLinks If it is known whether this content contains
+	*    links, provide this information here, to avoid redundant parsing to
+	*    find out.
+	*
+	* @return boolean
+	*/
+	boolean isCountable(boolean hasLinks);
+
+	/**
+	* Parse the Content Object and generate a ParserOutput from the result.
+	* $result->getText() can be used to obtain the generated HTML. If no HTML
+	* is needed, $generateHtml can be set to false; in that case,
+	* $result->getText() may return null.
+	*
+	* @note To control which options are used in the cache key for the
+	*       generated parser output, implementations of this method
+	*       may call ParserOutput::recordOption() on the output Object.
+	*
+	* @param Title $title The page title to use as a context for rendering.
+	* @param int $revId Optional revision ID being rendered.
+	* @param ParserOptions $options Any parser options.
+	* @param boolean $generateHtml Whether to generate HTML (default: true). If false,
+	*        the result of calling getText() on the ParserOutput Object returned by
+	*        this method is undefined.
+	*
+	* @since 1.21
+	*
+	* @return ParserOutput
+	*/
+	// generateHtml = true
+	XomwParserOutput getParserOutput(XomwTitle title, int revId,
+		XomwParserOptions options, boolean generateHtml);
+
+	// TODO: make RenderOutput and RenderOptions super classes
+
+	/**
+	* Returns a list of DataUpdate objects for recording information about this
+	* Content in some secondary data store. If the optional second argument,
+	* $old, is given, the updates may model only the changes that need to be
+	* made to replace information about the old content with information about
+	* the new content.
+	*
+	* This default implementation calls
+	* $this->getParserOutput( $content, $title, null, null, false ),
+	* and then calls getSecondaryDataUpdates( $title, $recursive ) on the
+	* resulting ParserOutput Object.
+	*
+	* Subclasses may implement this to determine the necessary updates more
+	* efficiently, or make use of information about the old content.
+	*
+	* @note Implementations should call the SecondaryDataUpdates hook, like
+	*   AbstractContent does.
+	*
+	* @param Title $title The context for determining the necessary updates
+	* @param Content $old An optional Content Object representing the
+	*    previous content, i.e. the content being replaced by this Content
+	*    Object.
+	* @param boolean $recursive Whether to include recursive updates (default:
+	*    false).
+	* @param ParserOutput $parserOutput Optional ParserOutput Object.
+	*    Provide if you have one handy, to avoid re-parsing of the content.
+	*
+	* @return DataUpdate[] A list of DataUpdate objects for putting information
+	*    about this content Object somewhere.
+	*
+	* @since 1.21
+	*/
+	// DFLT: recursive = true
+	Object getSecondaryDataUpdates(XomwTitle title, XomwContent old,
+		boolean recursive, XomwParserOutput parserOutput);
+
+	/**
+	* Construct the redirect destination from this content and return an
+	* array of Titles, or null if this content doesn't represent a redirect.
+	* The last element in the array is the final destination after all redirects
+	* have been resolved (up to $wgMaxRedirects times).
+	*
+	* @since 1.21
+	*
+	* @return Title[]|null List of Titles, with the destination last.
+	*/
+	XomwTitle[] getRedirectChain();
+
+	/**
+	* Construct the redirect destination from this content and return a Title,
+	* or null if this content doesn't represent a redirect.
+	* This will only return the immediate redirect target, useful for
+	* the redirect table and other checks that don't need full recursion.
+	*
+	* @since 1.21
+	*
+	* @return Title|null The corresponding Title.
+	*/
+	XomwTitle getRedirectTarget();
+
+	/**
+	* Construct the redirect destination from this content and return the
+	* Title, or null if this content doesn't represent a redirect.
+	*
+	* This will recurse down $wgMaxRedirects times or until a non-redirect
+	* target is hit in order to provide (hopefully) the Title of the final
+	* destination instead of another redirect.
+	*
+	* There is usually no need @Override to the default behavior, subclasses that
+	* want to implement redirects @Override should getRedirectTarget().
+	*
+	* @since 1.21
+	*
+	* @return Title|null
+	*/
+	XomwTitle getUltimateRedirectTarget();
+
+	/**
+	* Returns whether this Content represents a redirect.
+	* Shorthand for getRedirectTarget() !== null.
+	*
+	* @since 1.21
+	*
+	* @return boolean
+	*/
+	boolean isRedirect();
+
+	/**
+	* If this Content Object is a redirect, this method updates the redirect target.
+	* Otherwise, it does nothing.
+	*
+	* @since 1.21
+	*
+	* @param Title $target The new redirect target
+	*
+	* @return Content A new Content Object with the updated redirect (or $this
+	*   if this Content Object isn't a redirect)
+	*/
+	XomwContent updateRedirect(XomwTitle target);
+
+	/**
+	* Returns the section with the given ID.
+	*
+	* @since 1.21
+	*
+	* @param String|int $sectionId Section identifier as a number or String
+	* (e.g. 0, 1 or 'T-1'). The ID "0" retrieves the section before the first heading, "1" the
+	* text between the first heading (included) and the second heading (excluded), etc.
+	*
+	* @return Content|boolean|null The section, or false if no such section
+	*    exist, or null if sections are not supported.
+	*/
+	XomwContent getSection(String sectionId);
+
+	/**
+	* Replaces a section of the content and returns a Content Object with the
+	* section replaced.
+	*
+	* @since 1.21
+	*
+	* @param String|int|null|boolean $sectionId Section identifier as a number or String
+	* (e.g. 0, 1 or 'T-1'), null/false or an empty String for the whole page
+	* or 'new' for a new section.
+	* @param Content $with New content of the section
+	* @param String $sectionTitle New section's subject, only if $section is 'new'
+	*
+	* @return String|null Complete article text, or null if error
+	*/
+	byte[] replaceSection(String sectionId, XomwContent with, String sectionTitle);
+
+	/**
+	* Returns a Content Object with pre-save transformations applied (or this
+	* Object if no transformations apply).
+	*
+	* @since 1.21
+	*
+	* @param Title $title
+	* @param User $user
+	* @param ParserOptions $parserOptions
+	*
+	* @return Content
+	*/
+	XomwContent preSaveTransform(XomwTitle title, Object user, XomwParserOptions parserOptions );
+
+	/**
+	* Returns a new WikitextContent Object with the given section heading
+	* prepended, if supported. The default implementation just returns this
+	* Content Object unmodified, ignoring the section header.
+	*
+	* @since 1.21
+	*
+	* @param String $header
+	*
+	* @return Content
+	*/
+	XomwContent addSectionHeader(byte[] header);
+
+	/**
+	* Returns a Content Object with preload transformations applied (or this
+	* Object if no transformations apply).
+	*
+	* @since 1.21
+	*
+	* @param Title $title
+	* @param ParserOptions $parserOptions
+	* @param array $prms
+	*
+	* @return Content
+	*/
+	XomwContent preloadTransform(XomwTitle title, XomwParserOptions parserOptions, Object[] ary);
+
+	/**
+	* Prepare Content for saving. Called before Content is saved by WikiPage::doEditContent() and in
+	* similar places.
+	*
+	* This may be used to check the content's consistency with global state. This function should
+	* NOT write any information to the database.
+	*
+	* Note that this method will usually be called inside the same transaction
+	* bracket that will be used to save the new revision.
+	*
+	* Note that this method is called before any update to the page table is
+	* performed. This means that $page may not yet know a page ID.
+	*
+	* @since 1.21
+	*
+	* @param WikiPage $page The page to be saved.
+	* @param int $flags Bitfield for use with EDIT_XXX constants, see WikiPage::doEditContent()
+	* @param int $parentRevId The ID of the current revision
+	* @param User $user
+	*
+	* @return Status A status Object indicating whether the content was
+	*   successfully prepared for saving. If the returned status indicates
+	*   an error, a rollback will be performed and the transaction aborted.
+	*
+	* @see WikiPage::doEditContent()
+	*/
+	Object prepareSave(Object page, int flags, int parentRevId, Object user);
+
+	/**
+	* Returns a list of updates to perform when this content is deleted.
+	* The necessary updates may be taken from the Content Object, or depend on
+	* the current state of the database.
+	*
+	* @since 1.21
+	*
+	* @param WikiPage $page The deleted page
+	* @param ParserOutput $parserOutput Optional parser output Object
+	*    for efficient access to meta-information about the content Object.
+	*    Provide if you have one handy.
+	*
+	* @return DataUpdate[] A list of DataUpdate instances that will clean up the
+	*    database after deletion.
+	*/
+	Object getDeletionUpdates(Object page,
+		XomwParserOutput parserOutput);
+
+	/**
+	* Returns true if this Content Object matches the given magic word.
+	*
+	* @since 1.21
+	*
+	* @param MagicWord $word The magic word to match
+	*
+	* @return boolean Whether this Content Object matches the given magic word.
+	*/
+	boolean matchMagicWord(XomwMagicWord word);
+
+	/**
+	* Converts this content Object into another content Object with the given content model,
+	* if that is possible.
+	*
+	* @param String $toModel The desired content model, use the CONTENT_MODEL_XXX flags.
+	* @param String $lossy Optional flag, set to "lossy" to allow lossy conversion. If lossy
+	* conversion is not allowed, full round-trip conversion is expected to work without losing
+	* information.
+	*
+	* @return Content|boolean A content Object with the content model $toModel, or false if
+	* that conversion is not supported.
+	*/
+	XomwContent convert(byte[] toModel, byte[] lossy);
+	// @todo ImagePage and CategoryPage interfere with per-content action handlers
+	// @todo nice&sane integration of GeSHi syntax highlighting
+	//   [11:59] <vvv> Hooks are ugly; make CodeHighlighter interface and a
+	//   config to set the class which handles syntax highlighting
+	//   [12:00] <vvv> And default it to a DummyHighlighter
+
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/content/XomwMessageContent.java

@@ -13,3 +13,156 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.content; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+//	<?php
+//	/**
+//	* Wrapper content Object allowing to handle a system message as a Content Object.
+//	*/
+//
+//	/**
+//	* Wrapper allowing us to handle a system message as a Content Object.
+//	* Note that this is generally *not* used to represent content from the
+//	* MediaWiki namespace, and that there is no MessageContentHandler.
+//	* MessageContent is just intended as glue for wrapping a message programmatically.
+//	*
+//	* @ingroup Content
+//	*/
+//	class MessageContent extends AbstractContent {
+//
+//		/**
+//		* @var Message
+//		*/
+//		protected $mMessage;
+//
+//		/**
+//		* @param Message|String $msg A Message Object, or a message key.
+//		* @param String[] $params An optional array of message parameters.
+//		*/
+//		public function __construct( $msg, $params = null ) {
+//			# XXX: messages may be wikitext, html or plain text! and maybe even something else entirely.
+//			parent::__construct( CONTENT_MODEL_WIKITEXT );
+//
+//			if ( is_string( $msg ) ) {
+//				$this->mMessage = wfMessage( $msg );
+//			} else {
+//				$this->mMessage = clone $msg;
+//			}
+//
+//			if ( $params ) {
+//				$this->mMessage = $this->mMessage->params( $params );
+//			}
+//		}
+//
+//		/**
+//		* Fully parse the text from wikitext to HTML.
+//		*
+//		* @return String Parsed HTML.
+//		*/
+//		public function getHtml() {
+//			return $this->mMessage->parse();
+//		}
+//
+//		/**
+//		* Returns the message text. {{-transformation is done.
+//		*
+//		* @return String Unescaped message text.
+//		*/
+//		public function getWikitext() {
+//			return $this->mMessage->text();
+//		}
+//
+//		/**
+//		* Returns the message Object, with any parameters already substituted.
+//		*
+//		* @return Message The message Object.
+//		*/
+//		public function getNativeData() {
+//			// NOTE: Message objects are mutable. Cloning here makes MessageContent immutable.
+//			return clone $this->mMessage;
+//		}
+//
+//		/**
+//		* @return String
+//		*
+//		* @see Content::getTextForSearchIndex
+//		*/
+//		public function getTextForSearchIndex() {
+//			return $this->mMessage->plain();
+//		}
+//
+//		/**
+//		* @return String
+//		*
+//		* @see Content::getWikitextForTransclusion
+//		*/
+//		public function getWikitextForTransclusion() {
+//			return $this->getWikitext();
+//		}
+//
+//		/**
+//		* @param int $maxlength Maximum length of the summary text, defaults to 250.
+//		*
+//		* @return String The summary text.
+//		*
+//		* @see Content::getTextForSummary
+//		*/
+//		public function getTextForSummary( $maxlength = 250 ) {
+//			return substr( $this->mMessage->plain(), 0, $maxlength );
+//		}
+//
+//		/**
+//		* @return int
+//		*
+//		* @see Content::getSize
+//		*/
+//		public function getSize() {
+//			return strlen( $this->mMessage->plain() );
+//		}
+//
+//		/**
+//		* @return Content A copy of this Object
+//		*
+//		* @see Content::copy
+//		*/
+//		public function copy() {
+//			// MessageContent is immutable (because getNativeData() returns a clone of the Message Object)
+//			return $this;
+//		}
+//
+//		/**
+//		* @param boolean|null $hasLinks
+//		*
+//		* @return boolean Always false.
+//		*
+//		* @see Content::isCountable
+//		*/
+//		public function isCountable( $hasLinks = null ) {
+//			return false;
+//		}
+//
+//		/**
+//		* @param Title $title Unused.
+//		* @param int $revId Unused.
+//		* @param ParserOptions $options Unused.
+//		* @param boolean $generateHtml Whether to generate HTML (default: true).
+//		*
+//		* @return ParserOutput
+//		*
+//		* @see Content::getParserOutput
+//		*/
+//		public function getParserOutput( Title $title, $revId = null,
+//			ParserOptions $options = null, $generateHtml = true ) {
+//			if ( $generateHtml ) {
+//				$html = $this->getHtml();
+//			} else {
+//				$html = '';
+//			}
+//
+//			$po = new ParserOutput( $html );
+//			// Message objects are in the user language.
+//			$po->recordOption( 'userlang' );
+//
+//			return $po;
+//		}
+//
+//	}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/content/XomwTextContentHandler.java

@@ -13,3 +13,138 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.content; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+/**
+* Base content handler implementation for flat text contents.
+*
+* @ingroup Content
+*/
+class XomwTextContentHandler extends XomwContentHandler { 
+	public XomwTextContentHandler() {super(XomwDefaultSettings.CONTENT_MODEL_TEXT, XomwDefines.CONTENT_FORMAT_TEXT);
+	}
+	public XomwTextContentHandler(int modelId, String... formats) {super(modelId, formats);
+	}
+
+//		/**
+//		* Returns the content's text as-is.
+//		*
+//		* @param Content $content
+//		* @param String  $format The serialization format to check
+//		*
+//		* @return mixed
+//		*/
+//		public function serializeContent( Content $content, $format = null ) {
+//			$this->checkFormat( $format );
+//
+//			return $content->getNativeData();
+//		}
+//
+//		/**
+//		* Attempts to merge differences between three versions. Returns a new
+//		* Content Object for a clean merge and false for failure or a conflict.
+//		*
+//		* All three Content objects passed as parameters must have the same
+//		* content model.
+//		*
+//		* This text-based implementation uses wfMerge().
+//		*
+//		* @param Content $oldContent The page's previous content.
+//		* @param Content $myContent One of the page's conflicting contents.
+//		* @param Content $yourContent One of the page's conflicting contents.
+//		*
+//		* @return Content|boolean
+//		*/
+//		public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
+//			$this->checkModelID( $oldContent->getModel() );
+//			$this->checkModelID( $myContent->getModel() );
+//			$this->checkModelID( $yourContent->getModel() );
+//
+//			$format = $this->getDefaultFormat();
+//
+//			$old = $this->serializeContent( $oldContent, $format );
+//			$mine = $this->serializeContent( $myContent, $format );
+//			$yours = $this->serializeContent( $yourContent, $format );
+//
+//			$ok = wfMerge( $old, $mine, $yours, $result );
+//
+//			if ( !$ok ) {
+//				return false;
+//			}
+//
+//			if ( !$result ) {
+//				return $this->makeEmptyContent();
+//			}
+//
+//			$mergedContent = $this->unserializeContent( $result, $format );
+//
+//			return $mergedContent;
+//		}
+//
+//		/**
+//		* Returns the name of the associated Content class, to
+//		* be used when creating new objects. Override expected
+//		* by subclasses.
+//		*
+//		* @since 1.24
+//		*
+//		* @return String
+//		*/
+//		protected function getContentClass() {
+//			return TextContent::class;
+//		}
+//
+//		/**
+//		* Unserializes a Content Object of the type supported by this ContentHandler.
+//		*
+//		* @since 1.21
+//		*
+//		* @param String $text Serialized form of the content
+//		* @param String $format The format used for serialization
+//		*
+//		* @return Content The TextContent Object wrapping $text
+//		*/
+//		public function unserializeContent( $text, $format = null ) {
+//			$this->checkFormat( $format );
+//
+//			$class = $this->getContentClass();
+//			return new $class( $text );
+//		}
+//
+//		/**
+//		* Creates an empty TextContent Object.
+//		*
+//		* @since 1.21
+//		*
+//		* @return Content A new TextContent Object with empty text.
+//		*/
+//		public function makeEmptyContent() {
+//			$class = $this->getContentClass();
+//			return new $class( '' );
+//		}
+//
+//		/**
+//		* @see ContentHandler::supportsDirectEditing
+//		*
+//		* @return boolean Default is true for TextContent and derivatives.
+//		*/
+//		public function supportsDirectEditing() {
+//			return true;
+//		}
+//
+//		public function getFieldsForSearchIndex( SearchEngine $engine ) {
+//			$fields = parent::getFieldsForSearchIndex( $engine );
+//			$fields['language'] =
+//				$engine->makeSearchFieldMapping( 'language', SearchIndexField::INDEX_TYPE_KEYWORD );
+//
+//			return $fields;
+//		}
+//
+//		public function getDataForSearchIndex( WikiPage $page, ParserOutput $output,
+//											SearchEngine $engine ) {
+//			$fields = parent::getDataForSearchIndex( $page, $output, $engine );
+//			$fields['language'] =
+//				$this->getPageLanguage( $page->getTitle(), $page->getContent() )->getCode();
+//			return $fields;
+//		}
+
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/content/XomwWikitextContentHandler.java

@@ -13,3 +13,136 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.content; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+/**
+* Content handler for wiki text pages.
+*
+* @ingroup Content
+*/
+class XomwWikitextContentHandler extends XomwTextContentHandler { 
+	public XomwWikitextContentHandler() {super(XomwDefaultSettings.CONTENT_MODEL_WIKITEXT, XomwDefines.CONTENT_FORMAT_WIKITEXT);
+	}
+
+//		protected function getContentClass() {
+//			return WikitextContent::class;
+//		}
+//
+//		/**
+//		* Returns a WikitextContent Object representing a redirect to the given destination page.
+//		*
+//		* @param Title $destination The page to redirect to.
+//		* @param String $text Text to include in the redirect, if possible.
+//		*
+//		* @return Content
+//		*
+//		* @see ContentHandler::makeRedirectContent
+//		*/
+//		public function makeRedirectContent( Title $destination, $text = '' ) {
+//			$optionalColon = '';
+//
+//			if ( $destination->getNamespace() == NS_CATEGORY ) {
+//				$optionalColon = ':';
+//			} else {
+//				$iw = $destination->getInterwiki();
+//				if ( $iw && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
+//					$optionalColon = ':';
+//				}
+//			}
+//
+//			$mwRedir = MagicWord::get( 'redirect' );
+//			$redirectText = $mwRedir->getSynonym( 0 ) .
+//				' [[' . $optionalColon . $destination->getFullText() . ']]';
+//
+//			if ( $text != '' ) {
+//				$redirectText .= "\n" . $text;
+//			}
+//
+//			$class = $this->getContentClass();
+//			return new $class( $redirectText );
+//		}
+//
+//		/**
+//		* Returns true because wikitext supports redirects.
+//		*
+//		* @return boolean Always true.
+//		*
+//		* @see ContentHandler::supportsRedirects
+//		*/
+//		public function supportsRedirects() {
+//			return true;
+//		}
+//
+//		/**
+//		* Returns true because wikitext supports sections.
+//		*
+//		* @return boolean Always true.
+//		*
+//		* @see ContentHandler::supportsSections
+//		*/
+//		public function supportsSections() {
+//			return true;
+//		}
+//
+//		/**
+//		* Returns true, because wikitext supports caching using the
+//		* ParserCache mechanism.
+//		*
+//		* @since 1.21
+//		*
+//		* @return boolean Always true.
+//		*
+//		* @see ContentHandler::isParserCacheSupported
+//		*/
+//		public function isParserCacheSupported() {
+//			return true;
+//		}
+//
+//		/**
+//		* Get file handler
+//		* @return FileContentHandler
+//		*/
+//		protected function getFileHandler() {
+//			return new FileContentHandler();
+//		}
+//
+//		public function getFieldsForSearchIndex( SearchEngine $engine ) {
+//			$fields = parent::getFieldsForSearchIndex( $engine );
+//
+//			$fields['heading'] =
+//				$engine->makeSearchFieldMapping( 'heading', SearchIndexField::INDEX_TYPE_TEXT );
+//			$fields['heading']->setFlag( SearchIndexField::FLAG_SCORING );
+//
+//			$fields['auxiliary_text'] =
+//				$engine->makeSearchFieldMapping( 'auxiliary_text', SearchIndexField::INDEX_TYPE_TEXT );
+//
+//			$fields['opening_text'] =
+//				$engine->makeSearchFieldMapping( 'opening_text', SearchIndexField::INDEX_TYPE_TEXT );
+//			$fields['opening_text']->setFlag( SearchIndexField::FLAG_SCORING |
+//											SearchIndexField::FLAG_NO_HIGHLIGHT );
+//			// Until we have full first-class content handler for files, we invoke it explicitly here
+//			$fields = array_merge( $fields, $this->getFileHandler()->getFieldsForSearchIndex( $engine ) );
+//
+//			return $fields;
+//		}
+//
+//		public function getDataForSearchIndex( WikiPage $page, ParserOutput $parserOutput,
+//											SearchEngine $engine ) {
+//			$fields = parent::getDataForSearchIndex( $page, $parserOutput, $engine );
+//
+//			$structure = new WikiTextStructure( $parserOutput );
+//			$fields['heading'] = $structure->headings();
+//			// text fields
+//			$fields['opening_text'] = $structure->getOpeningText();
+//			$fields['text'] = $structure->getMainText(); // overwrites one from ContentHandler
+//			$fields['auxiliary_text'] = $structure->getAuxiliaryText();
+//			$fields['defaultsort'] = $structure->getDefaultSort();
+//
+//			// Until we have full first-class content handler for files, we invoke it explicitly here
+//			if ( NS_FILE == $page->getTitle()->getNamespace() ) {
+//				$fields = array_merge( $fields,
+//						$this->getFileHandler()->getDataForSearchIndex( $page, $parserOutput, $engine ) );
+//			}
+//			return $fields;
+//		}
+
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/exception/XomwMWException.java

@@ -13,3 +13,8 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.exception; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class XomwMWException extends Err {
+	public XomwMWException(String msg) {super(true, "", "", msg);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/filerepo/file/XomwFileFinder.java

@@ -13,3 +13,7 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.filerepo.file; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.filerepo.*;
+public interface XomwFileFinder {
+	XomwFile Find_file(XomwTitle ttl);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/filerepo/file/XomwFileFinderMock.java

@@ -13,3 +13,19 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.filerepo.file; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.filerepo.*;
+import gplx.xowa.mediawiki.includes.parsers.*;
+public class XomwFileFinderMock implements XomwFileFinder {
+	private final    XomwEnv env;
+	public XomwFileFinderMock(XomwEnv env) {this.env = env;}
+	private final    Hash_adp_bry hash = Hash_adp_bry.cs();
+	public void Clear() {hash.Clear();}
+	public XomwFile Find_file(XomwTitle ttl) {
+		return (XomwFile)hash.Get_by(ttl.getPrefixedDBkey());
+	}
+	public void Add(String title, XomwFileRepo repo, int w, int h, byte[] mime) {
+		byte[] title_bry = Bry_.new_u8(title);
+		XomwLocalFile file = new XomwLocalFile(env, XomwTitle.newFromText(env, title_bry), repo, w, h, mime);
+		hash.Add_if_dupe_use_nth(title_bry, file);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/filerepo/file/XomwFileFinderNoop.java

@@ -13,3 +13,7 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.filerepo.file; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.filerepo.*;
+public class XomwFileFinderNoop implements XomwFileFinder {
+	public XomwFile Find_file(XomwTitle ttl) {return null;}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/interwiki/XomwInterwiki.java

@@ -13,3 +13,170 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.interwiki; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+/**
+* Value Object for representing interwiki records.
+*/
+public class XomwInterwiki {
+
+	/** @var String The interwiki prefix, (e.g. "Meatball", or the language prefix "de") */
+	private byte[] mPrefix;
+
+	/** @var String The URL of the wiki, with "1" as a placeholder for an article name. */
+	private byte[] mURL;
+
+	/** @var String The URL of the file api.php  */
+	private byte[] mAPI;
+
+	/** @var String The name of the database (for a connection to be established
+	*    with wfGetLB('wikiid'))
+	*/
+	private byte[] mWikiID;
+
+	/** @var boolean Whether the wiki is in this project */
+	private boolean mLocal;
+
+	/** @var boolean Whether interwiki transclusions are allowed */
+	private boolean mTrans;
+
+	public byte[] interwikiId;
+
+	public XomwInterwiki(byte[] prefix, byte[] url, byte[] api, byte[] wikiId, boolean local, boolean trans) {
+		this.mPrefix = prefix;
+		this.mURL = url;
+		this.mAPI = api;
+		this.mWikiID = wikiId;
+		this.mLocal = local;
+		this.mTrans = trans;
+	}
+
+	/**
+	* Check whether an interwiki prefix exists
+	*
+	* [@]deprecated since 1.28, use InterwikiLookup instead
+	*
+	* @param String prefix Interwiki prefix to use
+	* @return boolean Whether it exists
+	*/
+	public static boolean isValidInterwiki(XomwMediaWikiServices mws, byte[] prefix) {
+		return mws.getInterwikiLookup().isValidInterwiki(prefix);
+//			return MediaWikiServices::getInstance().getInterwikiLookup().isValidInterwiki(prefix);
+	}
+
+//		/**
+//		* Fetch an Interwiki Object
+//		*
+//		* @deprecated since 1.28, use InterwikiLookup instead
+//		*
+//		* @param String prefix Interwiki prefix to use
+//		* @return Interwiki|null|boolean
+//		*/
+//		public static function fetch(prefix) {
+//			return MediaWikiServices::getInstance().getInterwikiLookup().fetch(prefix);
+//		}
+//
+//		/**
+//		* Purge the cache (local and persistent) for an interwiki prefix.
+//		*
+//		* @param String prefix
+//		* @since 1.26
+//		*/
+//		public static function invalidateCache(prefix) {
+//			return MediaWikiServices::getInstance().getInterwikiLookup().invalidateCache(prefix);
+//		}
+//
+//		/**
+//		* Returns all interwiki prefixes
+//		*
+//		* @deprecated since 1.28, unused. Use InterwikiLookup instead.
+//		*
+//		* @param String|null local If set, limits output to local/non-local interwikis
+//		* @return array List of prefixes
+//		* @since 1.19
+//		*/
+//		public static function getAllPrefixes(local = null) {
+//			return MediaWikiServices::getInstance().getInterwikiLookup().getAllPrefixes(local);
+//		}
+
+	/**
+	* Get the URL for a particular title (or with 1 if no title given)
+	*
+	* @param String title What text to put for the article name
+	* @return String The URL
+	* @note Prior to 1.19 The getURL with an argument was broken.
+	*       If you if you use this arg in an extension that supports MW earlier
+	*       than 1.19 please wfUrlencode and substitute 1 on your own.
+	*/
+	// title=null
+	public byte[] getURL(byte[] title) {
+		byte[] url = this.mURL;
+		if (title != null) {
+			url = XophpString.str_replace(ARG_1, XomwGlobalFunctions.wfUrlencode(title), url);
+		}
+
+		return url;
+	}
+
+	/**
+	* Get the API URL for this wiki
+	*
+	* @return String The URL
+	*/
+	public byte[] getAPI() {
+		return this.mAPI;
+	}
+
+	/**
+	* Get the DB name for this wiki
+	*
+	* @return String The DB name
+	*/
+	public byte[] getWikiID() {
+		return this.mWikiID;
+	}
+
+	/**
+	* Is this a local link from a sister project, or is
+	* it something outside, like Google
+	*
+	* @return boolean
+	*/
+	public boolean isLocal() {
+		return this.mLocal;
+	}
+
+	/**
+	* Can pages from this wiki be transcluded?
+	* Still requires wgEnableScaryTransclusion
+	*
+	* @return boolean
+	*/
+	public boolean isTranscludable() {
+		return this.mTrans;
+	}
+
+	/**
+	* Get the name for the interwiki site
+	*
+	* @return String
+	*/
+	public byte[] getName(XomwEnv env) {
+//			XomwMessage msg = XomwGlobalFunctions.wfMessage(env, "interwiki-name-" + this.mPrefix).inContentLanguage();
+//
+//			return !msg.exists() ? Bry_.Empty : msg.text();
+            Tfds.Write(mPrefix);
+		return null;
+	}
+
+//		/**
+//		* Get a description for this interwiki
+//		*
+//		* @return String
+//		*/
+//		public function getDescription() {
+//			msg = wfMessage('interwiki-desc-' . this.mPrefix).inContentLanguage();
+//
+//			return !msg.exists() ? '' : msg.text();
+//		}
+	private static final    byte[] ARG_1 = Bry_.new_a7("$1");
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/interwiki/XomwInterwikiLookup.java

@@ -13,3 +13,40 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.interwiki; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+/**
+* Service interface for looking up Interwiki records.
+*
+* @since 1.28
+*/
+public interface XomwInterwikiLookup {
+	/**
+	* Check whether an interwiki prefix exists
+	*
+	* @param String $prefix Interwiki prefix to use
+	* @return boolean Whether it exists
+	*/
+	boolean isValidInterwiki(byte[] prefix);
+
+	/**
+	* Fetch an Interwiki Object
+	*
+	* @param String $prefix Interwiki prefix to use
+	* @return Interwiki|null|boolean
+	*/
+	XomwInterwiki fetch(byte[] prefix);
+
+	/**
+	* Returns all interwiki prefixes
+	*
+	* @param String|null $local If set, limits output to local/non-local interwikis
+	* @return String[] List of prefixes
+	*/
+	byte[][] getAllPrefixes(boolean local);
+
+//		/**
+//		* Purge the in-process and persistent Object cache for an interwiki prefix
+//		* @param String $prefix
+//		*/
+//		void invalidateCache(byte[] prefix);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/interwiki/XomwInterwikiLookupAdapter.java

@@ -13,3 +13,157 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.interwiki; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.site.*;
+public class XomwInterwikiLookupAdapter implements XomwInterwikiLookup {
+	/**
+	* @var SiteLookup
+	*/
+	private final    XomwSiteLookup siteLookup;
+
+	/**
+	* @var Interwiki[]|null associative array mapping interwiki prefixes to Interwiki objects
+	*/
+	private Ordered_hash interwikiMap = Ordered_hash_.New_bry();
+
+	public XomwInterwikiLookupAdapter (
+		XomwSiteLookup siteLookup
+		// Ordered_hash interwikiMap
+	) {
+		this.siteLookup = siteLookup;
+	}
+
+	/**
+	* See InterwikiLookup::isValidInterwiki
+	* It loads the whole interwiki map.
+	*
+	* @param String $prefix Interwiki prefix to use
+	* @return boolean Whether it exists
+	*/
+	public boolean isValidInterwiki(byte[] prefix) {
+		return XophpArray.array_key_exists(prefix, this.getInterwikiMap());
+	}
+
+	/**
+	* See InterwikiLookup::fetch
+	* It loads the whole interwiki map.
+	*
+	* @param String $prefix Interwiki prefix to use
+	* @return Interwiki|null|boolean
+	*/
+	public XomwInterwiki fetch(byte[] prefix) {
+		if (prefix == Bry_.Empty) {
+			return null;
+		}
+
+		if (!this.isValidInterwiki(prefix)) {
+			return null;
+		}
+
+		return (XomwInterwiki)this.interwikiMap.Get_by(prefix);
+	}
+
+	/**
+	* See InterwikiLookup::getAllPrefixes
+	*
+	* @param String|null $local If set, limits output to local/non-local interwikis
+	* @return String[] List of prefixes
+	*/
+	public byte[][] getAllPrefixes(boolean local) {
+		if (!local) {
+			XophpArray.array_keys_bry(this.getInterwikiMap());
+		}
+		List_adp res = List_adp_.New();
+		Ordered_hash hash = this.getInterwikiMap();
+		int len = hash.Len();
+		for (int i = 0; i < len; i++) {
+			XomwInterwiki interwiki = (XomwInterwiki)hash.Get_at(i);
+			if (interwiki.isLocal() == local) {
+				res.Add(interwiki.interwikiId);
+			}
+		}
+		return (byte[][])res.To_ary_and_clear(byte[].class);
+	}
+
+	//	/**
+	//	* See InterwikiLookup::invalidateCache
+	//	*
+	//	* @param String $prefix
+	//	*/
+	//	public function invalidateCache($prefix) {
+	//		if (!isset(this.interwikiMap[$prefix])) {
+	//			return;
+	//		}
+	//		$globalId = this.interwikiMap[$prefix].getWikiID();
+	//		unset(this.interwikiMap[$prefix]);
+	//
+	//		// Reload the interwiki
+	//		site = this.siteLookup.getSites().getSite($globalId);
+	//		interwikis = this.getSiteInterwikis(site);
+	//		this.interwikiMap = array_merge(this.interwikiMap, [ interwikis[$prefix] ]);
+	//	}
+
+	/**
+	* Load interwiki map to use as cache
+	*/
+	private Ordered_hash loadInterwikiMap() {
+		Ordered_hash interwikiMap = Ordered_hash_.New();
+		XomwSiteList siteList = this.siteLookup.getSites();
+		int len = siteList.Len();
+		for (int i = 0; i < len; i++) {
+			XomwSite site = siteList.GetAt(i);
+			XomwInterwiki[] interwikis = this.getSiteInterwikis(site);
+			// interwikiMap = array_merge(interwikiMap, interwikis);
+			for (XomwInterwiki interwiki : interwikis) {
+				interwikiMap.Add(interwiki.interwikiId, interwiki);
+			}
+		}
+		this.interwikiMap = interwikiMap;
+		return interwikiMap;
+	}
+
+	/**
+	* Get interwikiMap attribute, load if needed.
+	*
+	* @return Interwiki[]
+	*/
+	private Ordered_hash getInterwikiMap() {
+		if (this.interwikiMap == null) {
+			this.loadInterwikiMap();
+		}
+		return this.interwikiMap;
+	}
+
+	/**
+	* Load interwikis for the given site
+	*
+	* @param Site site
+	* @return Interwiki[]
+	*/
+	private XomwInterwiki[] getSiteInterwikis(XomwSite site) {
+		Ordered_hash interwikis = Ordered_hash_.New();
+		Ordered_hash hash = site.getInterwikiIds();
+		int len = hash.Len();
+		for (int i = 0; i < len; i++) {
+			String interwiki = (String)hash.Get_at(i);
+			String url = site.getPageUrl();
+			String path = null;
+			if (Type_.Eq_by_obj(site, XomwMediaWikiSite.class)) {
+				path = ((XomwMediaWikiSite)site).getFileUrl("api.php");
+			} else {
+				path = "";
+			}
+			boolean local = String_.Eq(site.getSource(), "local");
+			// TODO: How to adapt trans?
+			interwikis.Add(interwiki, new XomwInterwiki(
+				Bry_.new_u8(interwiki),
+				Bry_.new_u8(url),
+				Bry_.new_u8(path),
+				Bry_.new_u8(site.getGlobalId()),
+				local
+				, false
+			));
+		}
+		return (XomwInterwiki[])interwikis.To_ary_and_clear(XomwInterwiki.class);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/XomwArrayObject.java

@@ -13,3 +13,27 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+// bare-bones implementation of PHP ArrayObject
+// REF:http://php.net/manual/en/class.arrayobject.php
+public abstract class XomwArrayObject {
+	private final    Ordered_hash hash = Ordered_hash_.New();
+	public boolean offsetExists(Object key) {
+		return hash.Has(key);
+	}
+	public Object offsetGet(Object key) {
+		return hash.Get_by(key);
+	}
+	public void offsetUnset(Object key) {
+		hash.Del(key);
+	}
+	@gplx.Virtual public void offsetSet(int key, Object val) {
+		hash.Add(key, val);
+	}
+
+	public int count() {return hash.Len();}
+	public Object Get_at(int i) {return hash.Get_at(i);}
+	public void Add_or_update(Object val) {
+		hash.Add(hash.Count(), val);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/XomwGenericArrayObject.java

@@ -13,3 +13,219 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+/**
+* Extends ArrayObject and does two things:
+*
+* Allows for deriving cla+sses to easily intercept additions
+* and deletions for purposes such as additional indexing.
+*
+* Enforces the objects to be of a certain type, so this
+* can be replied upon, much like if this had true support
+* for generics, which sadly enough is not possible in PHP.
+*/
+public abstract class XomwGenericArrayObject extends XomwArrayObject {	/**
+	* Returns the name of an interface/class that the element should implement/extend.
+	*
+	* @since 1.20
+	*
+	* @return String
+	*/
+	abstract public Class<?> getObjectType();
+
+	/**
+	* @see SiteList::getNewOffset()
+	* @since 1.20
+	* @var integer
+	*/
+	protected int indexOffset = 0;
+
+	/**
+	* Finds a new offset for when appending an element.
+	* The super class does this, so it would be better to integrate,
+	* but there does not appear to be any way to do this...
+	*
+	* @since 1.20
+	*
+	* @return integer
+	*/
+	protected int getNewOffset() {
+		while (this.offsetExists(this.indexOffset)) {
+			this.indexOffset++;
+		}
+
+		return this.indexOffset;
+	}
+
+	/**
+	* Constructor.
+	* @see ArrayObject::__construct
+	*
+	* @since 1.20
+	*
+	* @param null|array $input
+	* @param int $flags
+	* @param String $iterator_class
+	*/
+	public XomwGenericArrayObject() {
+		//	if (input != null) {
+		//		int len = Array_.Len(input);
+		//		for (int i = 0; i < len; i++) {
+		//			Object val = Array_.Get_at(input, i);
+		//			this.offsetSet(i, val);
+		//		}
+		//	}
+	}
+
+	/**
+	* @see ArrayObject::append
+	*
+	* @since 1.20
+	*
+	* @param mixed $value
+	*/
+	public void append(Object val) {
+		this.setElement(XophpUtility.NULL_INT, val);
+	}
+
+	/**
+	* @see ArrayObject::offsetSet()
+	*
+	* @since 1.20
+	*
+	* @param mixed $index
+	* @param mixed $value
+	*/
+	@Override public void offsetSet(int index, Object val) {
+		this.setElement(index, val);
+	}
+
+	/**
+	* Returns if the provided value has the same type as the elements
+	* that can be added to this ArrayObject.
+	*
+	* @since 1.20
+	*
+	* @param mixed $value
+	*
+	* @return boolean
+	*/
+	protected boolean hasValidType(Object val) {
+		Class<?> cls = this.getObjectType();
+		return Type_.Eq_by_obj(val, cls);
+	}
+
+	/**
+	* Method that actually sets the element and holds
+	* all common code needed for set operations, including
+	* type checking and offset resolving.
+	*
+	* If you want to do additional indexing or have code that
+	* otherwise needs to be executed whenever an element is added,
+	* you can overload @see preSetElement.
+	*
+	* @since 1.20
+	*
+	* @param mixed $index
+	* @param mixed $value
+	*
+	* @throws InvalidArgumentException
+	*/
+	protected void setElement(int index, Object val) {
+		if (!this.hasValidType(val)) {
+			throw new XophpInvalidArgumentException(
+				"Can only add "	+ Type_.Canonical_name(this.getObjectType()) + " implementing objects to "
+				+ Type_.Type_by_obj(this) + "."
+			);
+		}
+
+		if (XophpUtility.is_null(index)) {
+			index = this.getNewOffset();
+		}
+
+		if (this.preSetElement(index, val)) {
+			super.offsetSet(index, val);
+		}
+	}
+
+	/**
+	* Gets called before a new element is added to the ArrayObject.
+	*
+	* At this point the index is always set (ie not null) and the
+	* value is always of the type returned by @see getObjectType.
+	*
+	* Should return a boolean. When false is returned the element
+	* does not get added to the ArrayObject.
+	*
+	* @since 1.20
+	*
+	* @param integer|String $index
+	* @param mixed $value
+	*
+	* @return boolean
+	*/
+	protected boolean preSetElement(int index, Object val) {
+		return true;
+	}
+
+	//	/**
+	//	* @see Serializable::serialize
+	//	*
+	//	* @since 1.20
+	//	*
+	//	* @return String
+	//	*/
+	//	public function serialize() {
+	//		return serialize(this.getSerializationData());
+	//	}
+	//
+	//	/**
+	//	* Returns an array holding all the data that should go into serialization calls.
+	//	* This is intended to allow overloading without having to reimplement the
+	//	* behavior of this super class.
+	//	*
+	//	* @since 1.20
+	//	*
+	//	* @return array
+	//	*/
+	//	protected function getSerializationData() {
+	//		return [
+	//			'data' => this.getArrayCopy(),
+	//			'index' => this.indexOffset,
+	//		];
+	//	}
+	//
+	//	/**
+	//	* @see Serializable::unserialize
+	//	*
+	//	* @since 1.20
+	//	*
+	//	* @param String $serialization
+	//	*
+	//	* @return array
+	//	*/
+	//	public function unserialize($serialization) {
+	//		$serializationData = unserialize($serialization);
+	//
+	//		foreach ($serializationData['data'] as $offset => $value) {
+	//			// Just set the element, bypassing checks and offset resolving,
+	//			// as these elements have already gone through this.
+	//			parent::offsetSet($offset, $value);
+	//		}
+	//
+	//		this.indexOffset = $serializationData['index'];
+	//
+	//		return $serializationData;
+	//	}
+
+	/**
+	* Returns if the ArrayObject has no elements.
+	*
+	* @since 1.20
+	*
+	* @return boolean
+	*/
+	@gplx.Virtual public boolean isEmpty() {
+		return this.count() == 0;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/XomwStringUtils.java

@@ -13,3 +13,356 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+import gplx.xowa.mediawiki.includes.libs.replacers.*;
+/**
+* A collection of static methods to play with strings.
+*/
+public class XomwStringUtils {
+//		/**
+//		* Test whether a String is valid UTF-8.
+//		*
+//		* The function check for invalid byte sequences, overlong encoding but
+//		* not for different normalisations.
+//		*
+//		* @note In MediaWiki 1.21, this function did not provide proper UTF-8 validation.
+//		* In particular, the pure PHP code path did not in fact check for overlong forms.
+//		* Beware of this when backporting code to that version of MediaWiki.
+//		*
+//		* @since 1.21
+//		* @param String $value String to check
+//		* @return boolean Whether the given $value is a valid UTF-8 encoded String
+//		*/
+//		static function isUtf8($value) {
+//			$value = (String)$value;
+//
+//			// HHVM 3.4 and older come with an outdated version of libmbfl that
+//			// incorrectly allows values above U+10FFFF, so we have to check
+//			// for them separately. (This issue also exists in PHP 5.3 and
+//			// older, which are no longer supported.)
+//			static $newPHP;
+//			if ($newPHP === null) {
+//				$newPHP = !mb_check_encoding("\xf4\x90\x80\x80", 'UTF-8');
+//			}
+//
+//			return mb_check_encoding($value, 'UTF-8') &&
+//				($newPHP || preg_match("/\xf4[\x90-\xbf]|[\xf5-\xff]/S", $value) === 0);
+//		}
+
+	private static final byte DELIMITER_EXPLODE__SEP = 0, DELIMITER_EXPLODE__BGN = 1, DELIMITER_EXPLODE__END = 2;
+	private static final    Btrie_slim_mgr delimiter_explode_trie = Btrie_slim_mgr.cs()
+		.Add_str_byte("|" , DELIMITER_EXPLODE__SEP)
+		.Add_str_byte("-{", DELIMITER_EXPLODE__BGN)
+		.Add_str_byte("}-", DELIMITER_EXPLODE__END)
+		;
+	/**
+	* Explode a String, but ignore any instances of the separator inside
+	* the given start and end delimiters, which may optionally nest.
+	* The delimiters are literal strings, not regular expressions.
+	* @param String $startDelim Start delimiter
+	* @param String $endDelim End delimiter
+	* @param String $separator Separator String for the explode.
+	* @param String $subject Subject String to explode.
+	* @param boolean $nested True iff the delimiters are allowed to nest.
+	* @return ArrayIterator
+	*/
+	// XO.MW: NOTE: function only used in two places; hard-coding (a) nested=true; (b) bgn="-{" end="}-" sep="|"
+	public static byte[][] delimiterExplode(List_adp tmp, Btrie_rv trv, byte[] src) {
+		// XO.MW.PORTED:entire proc rewritten; see PHP for source
+		int src_bgn = 0;
+		int src_end = src.length;
+
+		int depth = 0;
+		int cur = src_bgn;
+		int prv = cur;
+		while (true) {
+			// eos
+			if (cur == src_end) {
+				// add rest
+				tmp.Add(Bry_.Mid(src, prv, src_end));
+				break;
+			}
+
+			Object o = delimiter_explode_trie.Match_at(trv, src, cur, src_end);
+
+			// regular char; continue;
+			if (o == null) {
+				cur++;
+				continue;
+			}
+
+			// handle sep, bgn, end
+			byte tid = ((gplx.core.primitives.Byte_obj_val)o).Val();
+			switch (tid) {
+				case DELIMITER_EXPLODE__SEP:
+					if (depth == 0) {
+						tmp.Add(Bry_.Mid(src, prv, cur));
+						prv = cur + 1;
+					}
+					break;
+				case DELIMITER_EXPLODE__BGN:
+					depth++;
+					break;
+				case DELIMITER_EXPLODE__END:
+					depth--;
+					break;
+			}
+			cur = trv.Pos();
+		}
+		return (byte[][])tmp.To_ary_and_clear(byte[].class);
+	}
+
+//		/**
+//		* Perform an operation equivalent to `preg_replace()`
+//		*
+//		* Matches this code:
+//		*
+//		*     preg_replace("!$startDelim(.*?)$endDelim!", $replace, $subject);
+//		*
+//		* ..except that it's worst-case O(N) instead of O(N^2). Compared to delimiterReplace(), this
+//		* implementation is fast but memory-hungry and inflexible. The memory requirements are such
+//		* that I don't recommend using it on anything but guaranteed small chunks of text.
+//		*
+//		* @param String $startDelim
+//		* @param String $endDelim
+//		* @param String $replace
+//		* @param String $subject
+//		* @return String
+//		*/
+//		static function hungryDelimiterReplace($startDelim, $endDelim, $replace, $subject) {
+//			$segments = explode($startDelim, $subject);
+//			$output = array_shift($segments);
+//			foreach ($segments as $s) {
+//				$endDelimPos = strpos($s, $endDelim);
+//				if ($endDelimPos === false) {
+//					$output .= $startDelim . $s;
+//				} else {
+//					$output .= $replace . substr($s, $endDelimPos + strlen($endDelim));
+//				}
+//			}
+//
+//			return $output;
+//		}
+
+	/**
+	* Perform an operation equivalent to `preg_replace_callback()`
+	*
+	* Matches this code:
+	*
+	*     preg_replace_callback("!$startDelim(.*)$endDelim!s$flags", $callback, $subject);
+	*
+	* If the start delimiter ends with an initial substring of the end delimiter,
+	* e.g. in the case of C-style comments, the behavior differs from the model
+	* regex. In this implementation, the end must share no characters with the
+	* start, so e.g. `/*\/` is not considered to be both the start and end of a
+	* comment. `/*\/xy/*\/` is considered to be a single comment with contents `/xy/`.
+	*
+	* The implementation of delimiterReplaceCallback() is slower than hungryDelimiterReplace()
+	* but uses far less memory. The delimiters are literal strings, not regular expressions.
+	*
+	* @param String $startDelim Start delimiter
+	* @param String $endDelim End delimiter
+	* @param callable $callback Function to call on each match
+	* @param String $subject
+	* @param String $flags Regular expression flags
+	* @throws InvalidArgumentException
+	* @return String
+	*/
+	// XO.MW:flags not supported; goes directly to regex; also, flags of "i" will do case-insensitive
+	public static void delimiterReplaceCallback(Bry_bfr bfr, byte[] bgn, byte[] end, XomwReplacer callback,
+		byte[] src
+	) {
+		/* XO.MW.PORTED:
+			MW does following logic
+			* Run start/end regex on subject till no matches
+			* If start/end found, evaluate possible match (handling nesting)
+			* If match found, then pass find-replace pair to callback;
+			    find=substr(subject, outputPos, tokenOffset + tokenLength - outputPos)
+				replace=substr(subject, contentPos, tokenOffset - contentPos)				
+			* Also, unnecessary "overlapping" logic: bgn=ab;end=abc
+				$strcmp( $endDelim, substr( $subject, $tokenOffset, $endLength ) ) == 0
+		*/
+		int pos = 0;
+		int prv = 0;
+		int srcLen = src.length;
+		int bgnLen = bgn.length;
+		int endLen = end.length;
+		boolean foundStart = false;
+		boolean tokenTypeIsStart = false;
+
+		while (true) {
+			if (pos >= srcLen) {
+				bfr.Add_mid(src, prv, srcLen);
+				break;
+			}
+			if      (Bry_.Eq(src, pos, pos + bgnLen, bgn)) {
+				tokenTypeIsStart = true;
+			}
+			else if (Bry_.Eq(src, pos, pos + endLen, end)) {
+				tokenTypeIsStart = false;
+			}
+			else {
+				pos++;
+				continue;
+			}
+
+			if (tokenTypeIsStart) {
+				// Only move the start position if we haven't already found a start
+				// This means that START START END matches outer pair
+				// EX: "(a(b)" has match of "a(b"
+				if (!foundStart) {
+					// Found start
+					// Write out the non-matching section
+					bfr.Add_mid(src, prv, pos);
+					pos += bgnLen;
+					prv = pos;
+					foundStart = true;
+				} else {
+					// Move the input position past the *first character* of START,
+					// to protect against missing END when it overlaps with START
+					pos++;
+				}
+			} else { // elseif (tokenType == 'end')
+				if (foundStart) {
+					// Found match
+					callback.cb(bfr, src, prv, pos);
+					foundStart = false;
+				} else {
+					// Non-matching end, write it out
+					// EX: "a)b" -> "a)"
+					bfr.Add_mid(src, prv, pos + endLen);
+				}
+				pos += endLen;
+				prv = pos;
+			}
+		}
+	}
+
+	/**
+	* Perform an operation equivalent to `preg_replace()` with flags.
+	*
+	* Matches this code:
+	*
+	*     preg_replace("!$startDelim(.*)$endDelim!$flags", $replace, $subject);
+	*
+	* @param String $startDelim Start delimiter regular expression
+	* @param String $endDelim End delimiter regular expression
+	* @param String $replace Replacement String. May contain $1, which will be
+	*  replaced by the text between the delimiters
+	* @param String $subject String to search
+	* @param String $flags Regular expression flags
+	* @return String The String with the matches replaced
+	*/
+	// XO.MW:removed flags=''
+	public static void delimiterReplace(Bry_bfr bfr, byte[] startDelim, byte[] endDelim, byte[] replace, byte[] subject) {
+		XomwRegexlikeReplacer replacer = new XomwRegexlikeReplacer(replace);
+
+		delimiterReplaceCallback(bfr, startDelim, endDelim, replacer, subject);
+	}
+
+//		/**
+//		* More or less "markup-safe" explode()
+//		* Ignores any instances of the separator inside `<...>`
+//		* @param String $separator
+//		* @param String $text
+//		* @return array
+//		*/
+//		static function explodeMarkup($separator, $text) {
+//			$placeholder = "\x00";
+//
+//			// Remove placeholder instances
+//			$text = str_replace($placeholder, '', $text);
+//
+//			// Replace instances of the separator inside HTML-like tags with the placeholder
+//			$replacer = new DoubleReplacer($separator, $placeholder);
+//			$cleaned = StringUtils::delimiterReplaceCallback('<', '>', $replacer->cb(), $text);
+//
+//			// Explode, then put the replaced separators back in
+//			$items = explode($separator, $cleaned);
+//			foreach ($items as $i => $str) {
+//				$items[$i] = str_replace($placeholder, $separator, $str);
+//			}
+//
+//			return $items;
+//		}
+
+	/**
+	* More or less "markup-safe" str_replace()
+	* Ignores any instances of the separator inside `<...>`
+	* @param String $search
+	* @param String $replace
+	* @param String $text
+	* @return String
+	*/
+	public static void replaceMarkup(byte[] src, int src_bgn, int src_end, byte[] find, byte[] repl) {	// REF:/includes/libs/StringUtils.php|replaceMarkup
+		// XO.MW.PORTED: avoiding multiple regex calls / String creations
+		// $placeholder = "\x00";
+		//
+		// Remove placeholder instances
+		// $text = str_replace($placeholder, '', $text);
+		//
+		// Replace instances of the separator inside HTML-like tags with the placeholder
+		// $replacer = new DoubleReplacer($search, $placeholder);
+		// $cleaned = StringUtils::delimiterReplaceCallback('<', '>', $replacer->cb(), $text);
+		//
+		// Explode, then put the replaced separators back in
+		// $cleaned = str_replace($search, $replace, $cleaned);
+		// $text = str_replace($placeholder, $search, $cleaned);
+
+		// if same length find / repl, do in-place replacement; EX: "!!"  -> "||"
+		int find_len = find.length;
+		int repl_len = repl.length;
+		if (find_len != repl_len) throw Err_.new_wo_type("find and repl should be same length");
+
+		byte find_0 = find[0];
+		byte dlm_bgn = Byte_ascii.Angle_bgn;
+		byte dlm_end = Byte_ascii.Angle_end;
+		boolean repl_active = true;
+
+		// loop every char in array
+		for (int i = src_bgn; i < src_end; i++) {
+			byte b = src[i];
+			if (  b == find_0
+				&& Bry_.Match(src, i + 1, i + find_len, find, 1, find_len)
+				&& repl_active
+				) {
+				Bry_.Set(src, i, i + find_len, repl);
+			}
+			else if (b == dlm_bgn) {
+				repl_active = false;
+			}
+			else if (b == dlm_end) {
+				repl_active = true;
+			}
+		}
+	}
+
+//		/**
+//		* Escape a String to make it suitable for inclusion in a preg_replace()
+//		* replacement parameter.
+//		*
+//		* @param String $String
+//		* @return String
+//		*/
+//		static function escapeRegexReplacement($String) {
+//			$String = str_replace('\\', '\\\\', $String);
+//			$String = str_replace('$', '\\$', $String);
+//			return $String;
+//		}
+//
+//		/**
+//		* Workalike for explode() with limited memory usage.
+//		*
+//		* @param String $separator
+//		* @param String $subject
+//		* @return ArrayIterator|ExplodeIterator
+//		*/
+//		static function explode($separator, $subject) {
+//			if (substr_count($subject, $separator) > 1000) {
+//				return new ExplodeIterator($separator, $subject);
+//			} else {
+//				return new ArrayIterator(explode($separator, $subject));
+//			}
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/XomwStringUtilsTest.java

@@ -13,3 +13,62 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XomwStringUtilsTest {
+	private final    XomwStringUtilsFxt fxt = new XomwStringUtilsFxt();
+	@Test  public void Delimiter_explode() {
+		// basic
+		fxt.Test_delimiter_explode("a|b|c"                             , "a", "b", "c");
+		// empty
+		fxt.Test_delimiter_explode("|a||c|"                            , "", "a", "", "c", "");
+		// nest_1
+		fxt.Test_delimiter_explode("a|-{b|c}-|d"                       , "a", "-{b|c}-", "d");
+		// nest_many
+		fxt.Test_delimiter_explode("a|-{b-{c|d}-e}-|f"                 , "a", "-{b-{c|d}-e}-", "f");
+	}
+	@Test  public void Replace_markup() {
+		// basic
+		fxt.Test_replace_markup("a!!b"             , "!!", "||", "a||b");
+		// missing
+		fxt.Test_replace_markup("abcd"             , "!!", "||", "abcd");
+		// eos
+		fxt.Test_replace_markup("a!!"              , "!!", "||", "a||");
+		// ignore
+		fxt.Test_replace_markup("a!!b<!!>!!c"      , "!!", "||", "a||b<!!>||c");
+		// ignore asym_lhs
+		fxt.Test_replace_markup("a!!b<!!<!!>!!c"   , "!!", "||", "a||b<!!<!!>||c");
+		// ignore asym_lhs
+		fxt.Test_replace_markup("a!!b<!!>!!>!!c"   , "!!", "||", "a||b<!!>||>||c");	// NOTE: should probably be "!!>!!>", but unmatched ">" are escaped to "&gt;"
+	}
+	@Test   public void delimiterReplace() {
+		// basic
+		fxt.Test_delimiterReplace("/*", "*/", "a/*0*/c"         , "9", "a9c");
+		// overlapping; "/*/"
+		fxt.Test_delimiterReplace("/*", "*/", "a/*/0/*/c"       , "9", "a9c");
+		// dangling bgn; "/* /*"
+		fxt.Test_delimiterReplace("/*", "*/", "a/*0/*1*/c"      , "9", "a9c"); // fails if "a/*9c"
+		// dangling end; "*/ */"
+		fxt.Test_delimiterReplace("/*", "*/", "a/*0*/1*/c"      , "9", "a91*/c");
+	}
+}
+class XomwStringUtilsFxt {
+	public void Test_delimiter_explode(String src_str, String... expd) {
+		List_adp tmp = List_adp_.New();
+		gplx.core.btries.Btrie_rv trv = new gplx.core.btries.Btrie_rv();
+
+		byte[][] actl = XomwStringUtils.delimiterExplode(tmp, trv, Bry_.new_u8(src_str));
+		Gftest.Eq__ary(expd, actl, "src=~{0}", src_str);
+	}
+	public void Test_replace_markup(String src_str, String find, String repl, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		XomwStringUtils.replaceMarkup(src_bry, 0, src_bry.length, Bry_.new_a7(find), Bry_.new_a7(repl));
+		Gftest.Eq__str(expd, src_bry);
+	}
+	// byte[] startDelim, byte[] endDelim, byte[] replace, byte[] subject
+	public void Test_delimiterReplace(String bgn, String end, String src, String repl, String expd) {
+		Bry_bfr bfr = Bry_bfr_.New();
+		XomwStringUtils.delimiterReplace(bfr, Bry_.new_u8(bgn), Bry_.new_u8(end), Bry_.new_u8(repl), Bry_.new_u8(src));
+		Gftest.Eq__str(expd, bfr.To_str_and_clear());
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/replacers/XomwRegexlikeReplacer.java

@@ -13,3 +13,13 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs.replacers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.libs.*;
+public class XomwRegexlikeReplacer implements XomwReplacer {
+	private byte[] replace;
+	public XomwRegexlikeReplacer(byte[] replace) {
+		this.replace = replace;
+	}
+	public void cb(Bry_bfr bfr, byte[] src, int find_bgn, int find_end) {
+		bfr.Add(replace);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/libs/replacers/XomwReplacer.java

@@ -13,3 +13,11 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.libs.replacers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.libs.*;
+/**
+* Base class for "replacers", objects used in preg_replace_callback() and
+* StringUtils::delimiterReplaceCallback()
+*/
+public interface XomwReplacer {
+	void cb(Bry_bfr bfr, byte[] src, int find_bgn, int find_end);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/linkers/XomwLinkRenderer.java

@@ -13,3 +13,465 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.linkers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.langs.htmls.*;
+import gplx.xowa.mediawiki.includes.xohtml.*;
+/*	TODO.XO
+	* P7: $html = HtmlArmor::getHtml($text);
+	* P3: getLinkUrl [alternate urls? EX: mw/wiki/index.php/title?]
+	* P2: titleFormatter->getPrefixedText [depends on redlinks]
+	* P1: getLinkClasses [depends on redlinks]
+*/
+/**
+* Class that generates HTML <a> links for pages.
+*
+* @see https://www.mediawiki.org/wiki/Manual:LinkRenderer
+* @since 1.28
+*/
+public class XomwLinkRenderer {
+	/**
+	* Whether to force the pretty article path
+	*
+	* @var boolean
+	*/
+	private boolean forceArticlePath = false;
+
+	/**
+	* A PROTO_* constant or false
+	*
+	* @var String|boolean|int
+	*/
+	private boolean expandUrls = false;
+
+	/**
+	* @var int
+	*/
+	private int stubThreshold = 0;
+
+	/**
+	* @var TitleFormatter
+	*/
+//		private $titleFormatter;
+
+	/**
+	* @var LinkCache
+	*/
+//		private $linkCache;
+
+	/**
+	* Whether to run the legacy Linker hooks
+	*
+	* @var boolean
+	*/
+//		private boolean runLegacyBeginHook = true;
+
+	private final    XomwHtmlTemp htmlTemp = new XomwHtmlTemp();
+	private final    Xomw_atr_mgr attribs = new Xomw_atr_mgr();
+	private final    List_adp tmp_merge_deleted = List_adp_.New();
+	private final    XomwSanitizer sanitizer;
+//		/**
+//		* @param TitleFormatter $titleFormatter
+//		* @param LinkCache $linkCache
+//		*/
+	public XomwLinkRenderer(XomwSanitizer sanitizer) {	// TitleFormatter $titleFormatter, LinkCache $linkCache
+//			this.titleFormatter = $titleFormatter;
+//			this.linkCache = $linkCache;
+		this.sanitizer = sanitizer;
+	}
+
+	/**
+	* @param boolean $force
+	*/
+	public void setForceArticlePath(boolean force) {
+		this.forceArticlePath = force;
+	}
+
+	/**
+	* @return boolean
+	*/
+	public boolean getForceArticlePath() {
+		return this.forceArticlePath;
+	}
+
+	/**
+	* @param String|boolean|int $expand A PROTO_* constant or false
+	*/
+	public void setExpandURLs(boolean expand) {
+		this.expandUrls = expand;
+	}
+
+	/**
+	* @return String|boolean|int a PROTO_* constant or false
+	*/
+	public boolean getExpandURLs() {
+		return this.expandUrls;
+	}
+
+	/**
+	* @param int $threshold
+	*/
+	public void setStubThreshold(int threshold) {
+		this.stubThreshold = threshold;
+	}
+
+	/**
+	* @return int
+	*/
+	public int getStubThreshold() {
+		return this.stubThreshold;
+	}
+
+	/**
+	* @param boolean $run
+	*/
+//		public void setRunLegacyBeginHook(boolean run) {
+//			this.runLegacyBeginHook = run;
+//		}
+
+	/**
+	* @param LinkTarget $target
+	* @param String|HtmlArmor|null $text
+	* @param array $extraAttribs
+	* @param array $query
+	* @return String
+	*/
+	public void makeLink(Bry_bfr bfr,
+		XomwTitle target, byte[] text, Xomw_atr_mgr extraAttribs, Xomw_qry_mgr query) {
+//			$title = Title::newFromLinkTarget($target);	// does db lookup?
+		if (target.isKnown()) {
+			this.makeKnownLink(bfr, target, text, extraAttribs, query);
+		} else {
+			this.makeBrokenLink(bfr, target, text, extraAttribs, query);
+		}
+	}
+
+	/**
+	* Get the options in the legacy format
+	*
+	* @param boolean $isKnown Whether the link is known or broken
+	* @return array
+	*/
+//		private function getLegacyOptions($isKnown) {
+//			$options = [ 'stubThreshold' => this.stubThreshold ];
+//			if (this.forceArticlePath) {
+//				$options[] = 'forcearticlepath';
+//			}
+//			if (this.expandUrls === PROTO_HTTP) {
+//				$options[] = 'http';
+//			} elseif (this.expandUrls === PROTO_HTTPS) {
+//				$options[] = 'https';
+//			}
+//
+//			$options[] = $isKnown ? 'known' : 'broken';
+//
+//			return $options;
+//		}
+//
+//		private function runBeginHook(LinkTarget $target, &$text, &$extraAttribs, &$query, $isKnown) {
+//			$ret = null;
+//			if (!Hooks::run('HtmlPageLinkRendererBegin',
+//				[ $this, $target, &$text, &$extraAttribs, &$query, &$ret ])
+//			) {
+//				return $ret;
+//			}
+//
+//			// Now run the legacy hook
+//			return this.runLegacyBeginHook($target, $text, $extraAttribs, $query, $isKnown);
+//		}
+//
+//		private function runLegacyBeginHook(LinkTarget $target, &$text, &$extraAttribs, &$query,
+//			$isKnown
+//		) {
+//			if (!this.runLegacyBeginHook || !Hooks::isRegistered('LinkBegin')) {
+//				// Disabled, or nothing registered
+//				return null;
+//			}
+//
+//			$realOptions = $options = this.getLegacyOptions($isKnown);
+//			$ret = null;
+//			$dummy = new DummyLinker();
+//			$title = Title::newFromLinkTarget($target);
+//			if ($text !== null) {
+//				$realHtml = $html = HtmlArmor::getHtml($text);
+//			} else {
+//				$realHtml = $html = null;
+//			}
+//			if (!Hooks::run('LinkBegin',
+//				[ $dummy, $title, &$html, &$extraAttribs, &$query, &$options, &$ret ])
+//			) {
+//				return $ret;
+//			}
+//
+//			if ($html !== null && $html !== $realHtml) {
+//				// &$html was modified, so re-armor it as $text
+//				$text = new HtmlArmor($html);
+//			}
+//
+//			// Check if they changed any of the options, hopefully not!
+//			if ($options !== $realOptions) {
+//				$factory = MediaWikiServices::getInstance()->getLinkRendererFactory();
+//				// They did, so create a separate instance and have that take over the rest
+//				$newRenderer = $factory->createFromLegacyOptions($options);
+//				// Don't recurse the hook...
+//				$newRenderer->setRunLegacyBeginHook(false);
+//				if (in_array('known', $options, true)) {
+//					return $newRenderer->makeKnownLink($title, $text, $extraAttribs, $query);
+//				} elseif (in_array('broken', $options, true)) {
+//					return $newRenderer->makeBrokenLink($title, $text, $extraAttribs, $query);
+//				} else {
+//					return $newRenderer->makeLink($title, $text, $extraAttribs, $query);
+//				}
+//			}
+//
+//			return null;
+//		}
+
+	/**
+	* If you have already looked up the proper CSS classes using LinkRenderer::getLinkClasses()
+	* or some other method, use this to avoid looking it up again.
+	*
+	* @param LinkTarget $target
+	* @param String|HtmlArmor|null $text
+	* @param String $classes CSS classes to add
+	* @param array $extraAttribs
+	* @param array $query
+	* @return String
+	*/
+	public void makePreloadedLink(Bry_bfr bfr,
+		XomwTitle target, byte[] text, byte[] classes, Xomw_atr_mgr extraAttribs, Xomw_qry_mgr query) {
+		// XO.MW.HOOK: this.runBeginHook --> 'HtmlPageLinkRendererBegin', 'LinkBegin'
+
+		target = this.normalizeTarget(target);
+		byte[] url = this.getLinkUrl(target, query);
+		attribs.Clear();
+		attribs.Add(Gfh_atr_.Bry__href, url);	// XO.MW: add url 1st; MW does attribs["url", url] + attribs + extra_attribs
+		if (classes.length > 0)                 // XO.MW: do not bother adding if empty
+			attribs.Add(Gfh_atr_.Bry__class, classes);
+		byte[] prefixed_text = target.getPrefixedText();
+		if (prefixed_text != Bry_.Empty) {
+			attribs.Add(Gfh_atr_.Bry__title, prefixed_text);
+		}
+
+		this.mergeAttribs(attribs, extraAttribs); // XO.MW: changed to not always create another array
+
+		if (text == null) {
+			text = this.getLinkText(target);
+		}
+
+		this.buildAElement(bfr, target, text, attribs, true);
+	}		
+
+	/**
+	* @param LinkTarget $target
+	* @param String|HtmlArmor|null $text
+	* @param array $extraAttribs
+	* @param array $query
+	* @return String
+	*/
+	public void makeKnownLink(Bry_bfr bfr,
+		XomwTitle target, byte[] text, Xomw_atr_mgr extraAttribs, Xomw_qry_mgr query) {
+		byte[] classes = Bry_.Empty;
+		if (target.isExternal()) {
+			classes = Bry__classes__extiw;
+		}
+		byte[] colour = this.getLinkClasses(target);
+		if (colour != Bry_.Empty) {
+			classes = Bry_.Add(classes, Byte_ascii.Space_bry, colour); // XO.MW: also does "$classes ? implode(' ', $classes) : '',"
+		}
+
+		this.makePreloadedLink(bfr, 
+			target,
+			text,
+			classes,
+			extraAttribs,
+			query);
+	}
+
+	/**
+	* @param LinkTarget $target
+	* @param String|HtmlArmor|null $text
+	* @param array $extraAttribs
+	* @param array $query
+	* @return String
+	*/
+	public void makeBrokenLink(Bry_bfr bfr,
+		XomwTitle target, byte[] text, Xomw_atr_mgr extraAttribs, Xomw_qry_mgr query) {
+		// XO.MW.HOOK: Run legacy hook
+
+		// We don't want to include fragments for broken links, because they
+		// generally make no sense.
+		if (target.hasFragment()) {
+			target = target.createFragmentTarget(Bry_.Empty);
+		}
+		target = this.normalizeTarget(target);
+
+		if (!XophpUtility.isset(query.action) && target.getNamespace() != XomwDefines.NS_SPECIAL) {
+			query.action = Bry__action__edit;
+			query.redlink = 1;
+		}
+
+		byte[] url = this.getLinkUrl(target, query);
+		attribs.Clear();
+		attribs.Add(Gfh_atr_.Bry__href, url); // $attribs = ['href' => $url,] + this.mergeAttribs($attribs, $extraAttribs);
+		attribs.Add(Gfh_atr_.Bry__class, Bry__class__new);
+
+//			$prefixedText = this.titleFormatter->getPrefixedText($target);
+//			if ($prefixedText !== '') {
+//				// This ends up in parser cache!
+//				$attribs['title'] = wfMessage('red-link-title', $prefixedText)
+//					->inContentLanguage()
+//					->text();
+//			}
+
+		this.mergeAttribs(attribs, extraAttribs);
+
+		if (text == null) {
+			text = this.getLinkText(target);
+		}
+
+		this.buildAElement(bfr, target, text, attribs, false);
+	}
+
+	/**
+	* Builds the final <a> element
+	*
+	* @param LinkTarget $target
+	* @param String|HtmlArmor $text
+	* @param array $attribs
+	* @param boolean $isKnown
+	* @return null|String
+	*/
+	private void buildAElement(Bry_bfr bfr, XomwTitle target, byte[] text, Xomw_atr_mgr attribs, boolean isKnown) {
+		// XO.MW.HOOK:HtmlPageLinkRendererEnd
+
+		byte[] htmlBry = text;
+//			$html = HtmlArmor::getHtml($text);
+
+		// XO.MW.HOOK:LinkEnd
+
+		XomwHtml.rawElement(bfr, htmlTemp, Gfh_tag_.Bry__a, attribs, htmlBry);
+	}
+
+	/**
+	* @param LinkTarget $target
+	* @return String non-escaped text
+	*/
+	// XO.MW:SYNC:1.29; DATE:2017-01-31
+	private byte[] getLinkText(XomwTitle target) {
+		byte[] prefixed_text = target.getPrefixedText();
+		// If the target is just a fragment, with no title, we return the fragment
+		// text.  Otherwise, we return the title text itself.
+		if (prefixed_text == Bry_.Empty && target.hasFragment()) {
+			return target.getFragment();
+		}
+		return prefixed_text;
+	}
+
+	private byte[] getLinkUrl(XomwTitle target, Xomw_qry_mgr query) {
+		// TODO: Use a LinkTargetResolver service instead of Title
+//			$title = Title::newFromLinkTarget($target);
+//			if (this.forceArticlePath) {
+//				$realQuery = $query;
+//				$query = [];
+//			}
+//			else {
+//				$realQuery = [];
+//			}
+		byte[] url = target.getLinkURL(query, false, this.expandUrls);
+
+//			if (this.forceArticlePath && $realQuery) {
+//				$url = wfAppendQuery($url, $realQuery);
+//			}
+		return url;
+	}
+
+	/**
+	* Normalizes the provided target
+	*
+	* @todo move the code from Linker actually here
+	* @param LinkTarget $target
+	* @return LinkTarget
+	*/
+	private XomwTitle normalizeTarget(XomwTitle target) {
+		return XomwLinker.normaliseSpecialPage(target);
+	}
+
+	/**
+	* Merges two sets of attributes
+	*
+	* @param array $defaults
+	* @param array $attribs
+	*
+	* @return array
+	*/
+	private void mergeAttribs(Xomw_atr_mgr defaults, Xomw_atr_mgr attribs) {
+		// XO.MW: ignore; defaults is always non-null and empty; if attribs exists, it will be merged below
+		// if (!$attribs) {
+		//	return $defaults;
+		// }
+
+		// Merge the custom attribs with the default ones, and iterate
+		// over that, deleting all "false" attributes.
+		sanitizer.mergeAttributes(defaults, attribs);
+
+		// XO.MW.PORTED.BGN:MW removes "false" values; XO removes "null" values
+		//	foreach ($merged as $key => $val) {
+		//		# A false value suppresses the attribute
+		//		if ($val !== false) {
+		//			$ret[$key] = $val;
+		//		}
+		//	}
+		boolean deleted = false;
+		int len = attribs.Len();
+		for (int i = 0; i < len; i++) {
+			Xomw_atr_itm trg_atr = attribs.Get_at(i);
+			// A false value suppresses the attribute
+			if (trg_atr.Val() == null) {
+				tmp_merge_deleted.Add(trg_atr);
+				deleted = true;
+			}
+		}
+		if (deleted) {
+			len = tmp_merge_deleted.Len();
+			for (int i = 0; i < len; i++) {
+				Xomw_atr_itm atr = (Xomw_atr_itm)attribs.Get_at(i);
+				attribs.Del(atr.Key_bry());
+			}
+			tmp_merge_deleted.Clear();
+		}
+		// XO.MW.PORTED.END
+	}
+
+	/**
+	* Return the CSS classes of a known link
+	*
+	* @param LinkTarget $target
+	* @return String CSS class
+	*/
+	public byte[] getLinkClasses(XomwTitle target) {
+		// Make sure the target is in the cache
+//			$id = this.linkCache->addLinkObj($target);
+//			if ($id == 0) {
+//				// Doesn't exist
+//				return '';
+//			}
+
+//			if (this.linkCache->getGoodLinkFieldObj($target, 'redirect')) {
+			// Page is a redirect
+//				return 'mw-redirect';
+//			}
+//			elseif (this.stubThreshold > 0 && XomwNamespace::isContent($target->getNamespace())
+//				&& this.linkCache->getGoodLinkFieldObj($target, 'length') < this.stubThreshold
+//			) {
+			// Page is a stub
+//				return 'stub';
+//			}
+
+		return Bry_.Empty;
+	}
+	private static final    byte[] 
+	  Bry__classes__extiw = Bry_.new_a7("extiw")
+	, Bry__class__new = Bry_.new_a7("new")
+	, Bry__action__edit = Bry_.new_a7("edit")
+	;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/linkers/XomwLinkRendererTest.java

@@ -13,3 +13,21 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.linkers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+//	import org.junit.*;
+//	public class XomwLinkRendererTest {
+//		private final    XomwLinkRendererFxt fxt = new XomwLinkRendererFxt();
+/*
+Make_broken_link
+target.Has_fragment()
+*/
+//	}
+//	class XomwLinkRendererFxt {
+//		private final    XomwLinkRenderer wkr = new XomwLinkRenderer(new Xomw_parser());
+//		public void Test__parse(String src_str, String expd) {
+//			byte[] src_bry = Bry_.new_u8(src_str);
+//			wkr.Replace_external_links(new XomwParserCtx(), pbfr.Init(src_bry));
+//			if (apos) expd = gplx.langs.htmls.Gfh_utl.Replace_apos(expd);
+//			Tfds.Eq_str_lines(expd, pbfr.Rslt().To_str_and_clear(), src_str);
+//		}
+//	}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwImageHandler.java

@@ -13,3 +13,289 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*; import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+/*	XO.TODO:
+	* validateThumbParams
+*/
+// MEMORY:only one instance per wiki
+public abstract class XomwImageHandler extends XomwMediaHandler {	private final    Xomw_param_map paramMap = new Xomw_param_map();
+	public XomwImageHandler(byte[] key) {super(key);
+		paramMap.Add(Xomw_param_itm.Mw__img_width, Xomw_param_map.Type__handler, Xomw_param_itm.Name_bry__width);
+	}
+	/**
+	* @param File file
+	* @return boolean
+	*/
+	@Override public boolean canRender(XomwFile file) {
+		return (XophpUtility.istrue(file.getWidth()) && XophpUtility.istrue(file.getHeight()));
+	}
+
+	@Override public Xomw_param_map getParamMap() {
+		// XO.MW: defined above: "return [ 'img_width' => 'width' ];"
+		return paramMap;
+	}
+
+	@Override public boolean validateParam(int name_uid, byte[] val_bry, int val_int) {
+		if (name_uid == Xomw_param_itm.Name__width || name_uid == Xomw_param_itm.Name__height) {
+			if (val_int <= 0) {
+				return false;
+			}
+			else {
+				return true;
+			}
+		}
+		else {
+			return false;
+		}
+	}
+
+	@Override public byte[] makeParamString(Xomw_params_handler handlerParams) {
+		int width = 0;
+		if (XophpUtility.isset(handlerParams.physicalWidth)) {
+			width = handlerParams.physicalWidth;
+		}
+		else if (XophpUtility.isset(handlerParams.width)) {
+			width = handlerParams.width;
+		}
+		else {
+			throw Err_.new_wo_type("No width specified to makeParamString");
+		}
+
+		// Removed for ProofreadPage
+		// width = intval(width);
+		return Bry_.Add(Int_.To_bry(width), Xomw_lnki_wkr.Bry__px);
+	}
+
+//		public Xomw_param_map parseParamString(byte[] src) {
+//			int len = src.length;
+//			// XO.MW.REGEX: if (preg_match('/^(\d+)px/', str, m)) {
+//			if (    len > 0                                               // at least one char
+//				&&  Byte_ascii.Is_num(src[0]))                            // 1st char is numeric
+//			{
+//				pos = Bry_find_.Find_fwd_while_num(src, 1, len);          // skip numeric
+//				if (Bry_.Match(src, pos, len, Xomw_lnki_wkr.Bry__px)) {   // matches "px"
+//					Xomw_params_handler rv = new Xomw_params_handler();
+//					rv.width = Bry_.To_int_or(src, 0, pos, XophpUtility.NULL_INT);
+//					return rv;
+//				}
+//			}
+//			return null;
+//		}
+
+//		function getScriptParams(paramsVar) {
+//			return [ 'width' => paramsVar['width'] ];
+//		}
+
+	/**
+	* @param File image
+	* @param array paramsVar
+	* @return boolean
+	*/
+	@Override public boolean normaliseParams(XomwFile image, Xomw_params_handler handlerParams) {
+		byte[] mimeType = image.getMimeType();
+
+		if (!XophpUtility.isset(handlerParams.width)) {
+			return false;
+		}
+
+		if (!XophpUtility.isset(handlerParams.page)) {
+			handlerParams.page = 1;
+		}
+		else {
+			// handlerParams.page = intval(handlerParams.page);
+//				if (handlerParams.page > image.pageCount()) {
+//					handlerParams.page = image.pageCount();
+//				}
+//
+//				if (handlerParams.page < 1) {
+//					handlerParams.page = 1;
+//				}
+		}
+
+		int srcWidth = image.getWidth(handlerParams.page);
+		int srcHeight = image.getHeight(handlerParams.page);
+
+		if (XophpUtility.isset(handlerParams.height) && handlerParams.height != -1) {
+			// Height & width were both set
+			if (handlerParams.width * srcHeight > handlerParams.height * srcWidth) {
+				// Height is the relative smaller dimension, so scale width accordingly
+				handlerParams.width = fitBoxWidth(srcWidth, srcHeight, handlerParams.height);
+
+				if (handlerParams.width == 0) {
+					// Very small image, so we need to rely on client side scaling :(
+					handlerParams.width = 1;
+				}
+
+				handlerParams.physicalWidth = handlerParams.width;
+			} else {
+				// Height was crap, unset it so that it will be calculated later
+				handlerParams.height = XophpUtility.NULL_INT;
+			}
+		}
+
+		if (!XophpUtility.isset(handlerParams.physicalWidth)) {
+			// Passed all validations, so set the physicalWidth
+			handlerParams.physicalWidth = handlerParams.width;
+		}
+
+		// Because thumbs are only referred to by width, the height always needs
+		// to be scaled by the width to keep the thumbnail sizes consistent,
+		// even if it was set inside the if block above
+		handlerParams.physicalHeight = XomwFile.scaleHeight(srcWidth, srcHeight,
+			handlerParams.physicalWidth);
+
+		// Set the height if it was not validated in the if block higher up
+		if (!XophpUtility.isset(handlerParams.height) || handlerParams.height == -1) {
+			handlerParams.height = handlerParams.physicalHeight;
+		}
+
+		if (!this.validateThumbParams(handlerParams, srcWidth, srcHeight, mimeType)
+		) {
+			return false;
+		}
+
+		return true;
+	}
+
+	/**
+	* Validate thumbnail parameters and fill in the correct height
+	*
+	* @param int width Specified width (input/output)
+	* @param int height Height (output only)
+	* @param int srcWidth Width of the source image
+	* @param int srcHeight Height of the source image
+	* @param String mimeType Unused
+	* @return boolean False to indicate that an error should be returned to the user.
+	*/
+	// XO.MW.NOTE: MW passes w and h by ref, but only changes h; XO will pass handlerParams directly
+	private boolean validateThumbParams(Xomw_params_handler handlerParams, int srcWidth, int srcHeight, byte[] mimeType) {
+		int width = handlerParams.physicalWidth;
+		int height = handlerParams.physicalHeight;
+		// width = intval(width);
+
+		// Sanity check width
+		if (width <= 0) {
+			Gfo_usr_dlg_.Instance.Warn_many("", "", "validateThumbParams: Invalid destination width: width");
+
+			return false;
+		}
+		if (srcWidth <= 0) {
+			Gfo_usr_dlg_.Instance.Warn_many("", "", "validateThumbParams: Invalid source width: srcWidth");
+
+			return false;
+		}
+
+		height = XomwFile.scaleHeight(srcWidth, srcHeight, width);
+		if (height == 0) {
+			// Force height to be at least 1 pixel
+			height = 1;
+		}
+		handlerParams.height = height;
+
+		return true;
+	}
+
+//		/**
+//		* @param File image
+//		* @param String script
+//		* @param array paramsVar
+//		* @return boolean|MediaTransformOutput
+//		*/
+//		function getScriptedTransform(image, script, paramsVar) {
+//			if (!this.normaliseParams(image, paramsVar)) {
+//				return false;
+//			}
+//			url = wfAppendQuery(script, this.getScriptParams(paramsVar));
+//
+//			if (image.mustRender() || paramsVar['width'] < image.getWidth()) {
+//				return new ThumbnailImage(image, url, false, paramsVar);
+//			}
+//		}
+//
+//		function getImageSize(image, path) {
+//			MediaWiki\suppressWarnings();
+//			gis = getimagesize(path);
+//			MediaWiki\restoreWarnings();
+//
+//			return gis;
+//		}
+//
+//		/**
+//		* Function that returns the number of pixels to be thumbnailed.
+//		* Intended for animated GIFs to multiply by the number of frames.
+//		*
+//		* If the file doesn't support a notion of "area" return 0.
+//		*
+//		* @param File image
+//		* @return int
+//		*/
+//		function getImageArea(image) {
+//			return image.getWidth() * image.getHeight();
+//		}
+//
+//		/**
+//		* @param File file
+//		* @return String
+//		*/
+//		function getShortDesc(file) {
+//			global wgLang;
+//			nbytes = htmlspecialchars(wgLang.formatSize(file.getSize()));
+//			widthheight = wfMessage('widthheight')
+//				.numParams(file.getWidth(), file.getHeight()).escaped();
+//
+//			return "widthheight (nbytes)";
+//		}
+//
+//		/**
+//		* @param File file
+//		* @return String
+//		*/
+//		function getLongDesc(file) {
+//			global wgLang;
+//			pages = file.pageCount();
+//			size = htmlspecialchars(wgLang.formatSize(file.getSize()));
+//			if (pages === false || pages <= 1) {
+//				msg = wfMessage('file-info-size').numParams(file.getWidth(),
+//					file.getHeight()).paramsVar(size,
+//						'<span class="mime-type">' . file.getMimeType() . '</span>').parse();
+//			} else {
+//				msg = wfMessage('file-info-size-pages').numParams(file.getWidth(),
+//					file.getHeight()).paramsVar(size,
+//						'<span class="mime-type">' . file.getMimeType() . '</span>').numParams(pages).parse();
+//			}
+//
+//			return msg;
+//		}
+//
+//		/**
+//		* @param File file
+//		* @return String
+//		*/
+//		function getDimensionsString(file) {
+//			pages = file.pageCount();
+//			if (pages > 1) {
+//				return wfMessage('widthheightpage')
+//					.numParams(file.getWidth(), file.getHeight(), pages).text();
+//			} else {
+//				return wfMessage('widthheight')
+//					.numParams(file.getWidth(), file.getHeight()).text();
+//			}
+//		}
+//
+//		public function sanitizeParamsForBucketing(paramsVar) {
+//			paramsVar = parent::sanitizeParamsForBucketing(paramsVar);
+//
+//			// We unset the height parameters in order to let normaliseParams recalculate them
+//			// Otherwise there might be a height discrepancy
+//			if (isset(paramsVar['height'])) {
+//				unset(paramsVar['height']);
+//			}
+//
+//			if (isset(paramsVar['physicalHeight'])) {
+//				unset(paramsVar['physicalHeight']);
+//			}
+//
+//			return paramsVar;
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwImageHandlerTest.java

@@ -13,3 +13,49 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+import gplx.xowa.mediawiki.includes.parsers.*; import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+import gplx.xowa.mediawiki.includes.filerepo.*; import gplx.xowa.mediawiki.includes.filerepo.file.*;
+import gplx.xowa.mediawiki.languages.*;
+public class XomwImageHandlerTest {
+	private final    XomwImageHandler_fxt fxt = new XomwImageHandler_fxt();
+	@Before public void init() {
+		fxt.Init__file("A.png", 400, 200);
+	}
+	@Test   public void normaliseParams() {
+		// widthOnly; "Because thumbs are only referred to by width, the height always needs"
+		fxt.Test__normaliseParams(fxt.Make__handlerParams(200), fxt.Make__handlerParams(200, 100, 200, 100));
+	}
+}
+class XomwImageHandler_fxt {
+	private final    XomwImageHandler handler;
+	private final    XomwFileRepo repo = new XomwFileRepo(Bry_.new_a7("/orig"), Bry_.new_a7("/thumb"));
+	private final    XomwEnv env = XomwEnv.NewTest();
+	private XomwFile file;
+	public XomwImageHandler_fxt() {
+		this.handler = new XomwTransformationalImageHandler(Bry_.new_a7("test_handler"));
+	}
+	public Xomw_params_handler Make__handlerParams(int w) {return Make__handlerParams(w, XophpUtility.NULL_INT, XophpUtility.NULL_INT, XophpUtility.NULL_INT);}
+	public Xomw_params_handler Make__handlerParams(int w, int h, int phys_w, int phys_h) {
+		Xomw_params_handler rv = new Xomw_params_handler();
+		rv.width = w;
+		rv.height = h;
+		rv.physicalWidth = phys_w;
+		rv.physicalHeight = phys_h;
+		return rv;
+	}
+	public void Init__file(String title, int w, int h) {
+		this.file = new XomwLocalFile(env, XomwTitle.newFromText(env, Bry_.new_u8(title)), repo, w, h, XomwMediaHandlerFactory.Mime__image__png);
+	}
+	public void Test__normaliseParams(Xomw_params_handler prms, Xomw_params_handler expd) {
+		// exec
+		handler.normaliseParams(file, prms);
+
+		// test
+            Gftest.Eq__int(expd.width, prms.width);
+            Gftest.Eq__int(expd.height, prms.height);
+            Gftest.Eq__int(expd.physicalWidth, prms.physicalWidth);
+            Gftest.Eq__int(expd.physicalHeight, prms.physicalHeight);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwMediaHandler.java

@@ -13,3 +13,854 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*;
+import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+public abstract class XomwMediaHandler {
+	public byte[] Key() {return key;} private byte[] key;
+	public XomwMediaHandler(byte[] key) {
+		this.key = key;
+	}
+
+	private static final int TRANSFORM_LATER = 1;
+//		static final METADATA_GOOD = true;
+//		static final METADATA_BAD = false;
+//		static final METADATA_COMPATIBLE = 2; // for old but backwards compatible.
+//		/**
+//		* Max length of error logged by logErrorForExternalProcess()
+//		*/
+//		static final MAX_ERR_LOG_SIZE = 65535;
+//
+//		/**
+//		* Get a MediaHandler for a given MIME type from the instance cache
+//		*
+//		* @param String $type
+//		* @return MediaHandler|boolean
+//		*/
+//		static function getHandler($type) {
+//			return MediaWikiServices::getInstance()
+//				->getMediaHandlerFactory()->getHandler($type);
+//		}
+
+	/**
+	* Get an associative array mapping magic word IDs to parameter names.
+	* Will be used by the parser to identify parameters.
+	*/
+	public abstract Xomw_param_map getParamMap();
+
+	/**
+	* Validate a thumbnail parameter at parse time.
+	* Return true to accept the parameter, and false to reject it.
+	* If you return false, the parser will do something quiet and forgiving.
+	*
+	* @param String $name
+	* @param mixed $value
+	*/
+	public abstract boolean validateParam(int name_uid, byte[] val_bry, int val_int);
+
+	/**
+	* Merge a parameter array into a String appropriate for inclusion in filenames
+	*
+	* @param array paramsVar Array of parameters that have been through normaliseParams.
+	* @return String
+	*/
+	public abstract byte[] makeParamString(Xomw_params_handler handlerParams);
+
+//		/**
+//		* Parse a param String made with makeParamString back into an array
+//		*
+//		* @param String $str The parameter String without file name (e.g. 122px)
+//		* @return array|boolean Array of parameters or false on failure.
+//		*/
+//		abstract public function parseParamString($str);
+
+	/**
+	* Changes the parameter array as necessary, ready for transformation.
+	* Should be idempotent.
+	* Returns false if the parameters are unacceptable and the transform should fail
+	* @param File $image
+	* @param array $paramsVar
+	*/
+	public abstract boolean normaliseParams(XomwFile image, Xomw_params_handler handlerParams);
+
+//		/**
+//		* Get an image size array like that returned by getimagesize(), or false if it
+//		* can't be determined.
+//		*
+//		* This function is used for determining the width, height and bitdepth directly
+//		* from an image. The results are stored in the database in the img_width,
+//		* img_height, img_bits fields.
+//		*
+//		* @note If this is a multipage file, return the width and height of the
+//		*  first page.
+//		*
+//		* @param File|FSFile $image The image Object, or false if there isn't one.
+//		*   Warning, FSFile::getPropsFromPath might pass an FSFile instead of File (!)
+//		* @param String $path The filename
+//		* @return array|boolean Follow the format of PHP getimagesize() @gplx.Internal protected function.
+//		*   See https://secure.php.net/getimagesize. MediaWiki will only ever use the
+//		*   first two array keys (the width and height), and the 'bits' associative
+//		*   key. All other array keys are ignored. Returning a 'bits' key is optional
+//		*   as not all formats have a notion of "bitdepth". Returns false on failure.
+//		*/
+//		abstract function getImageSize($image, $path);
+//
+//		/**
+//		* Get handler-specific metadata which will be saved in the img_metadata field.
+//		*
+//		* @param File|FSFile $image The image Object, or false if there isn't one.
+//		*   Warning, FSFile::getPropsFromPath might pass an FSFile instead of File (!)
+//		* @param String $path The filename
+//		* @return String A String of metadata in php serialized form (Run through serialize())
+//		*/
+//		function getMetadata($image, $path) {
+//			return '';
+//		}
+//
+//		/**
+//		* Get metadata version.
+//		*
+//		* This is not used for validating metadata, this is used for the api when returning
+//		* metadata, since api content formats should stay the same over time, and so things
+//		* using ForeignApiRepo can keep backwards compatibility
+//		*
+//		* All core media handlers share a common version number, and extensions can
+//		* use the GetMetadataVersion hook to append to the array (they should append a unique
+//		* String so not to get confusing). If there was a media handler named 'foo' with metadata
+//		* version 3 it might add to the end of the array the element 'foo=3'. if the core metadata
+//		* version is 2, the end version String would look like '2;foo=3'.
+//		*
+//		* @return String Version String
+//		*/
+//		static function getMetadataVersion() {
+//			$version = [ '2' ]; // core metadata version
+//			Hooks::run('GetMetadataVersion', [ &$version ]);
+//
+//			return implode(';', $version);
+//		}
+//
+//		/**
+//		* Convert metadata version.
+//		*
+//		* By default just returns $metadata, but can be used to allow
+//		* media handlers to convert between metadata versions.
+//		*
+//		* @param String|array $metadata Metadata array (serialized if String)
+//		* @param int $version Target version
+//		* @return array Serialized metadata in specified version, or $metadata on fail.
+//		*/
+//		function convertMetadataVersion($metadata, $version = 1) {
+//			if (!is_array($metadata)) {
+//
+//				// unserialize to keep return parameter consistent.
+//				MediaWiki\suppressWarnings();
+//				$ret = unserialize($metadata);
+//				MediaWiki\restoreWarnings();
+//
+//				return $ret;
+//			}
+//
+//			return $metadata;
+//		}
+//
+//		/**
+//		* Get a String describing the type of metadata, for display purposes.
+//		*
+//		* @note This method is currently unused.
+//		* @param File $image
+//		* @return String
+//		*/
+//		function getMetadataType($image) {
+//			return false;
+//		}
+//
+//		/**
+//		* Check if the metadata String is valid for this handler.
+//		* If it returns MediaHandler::METADATA_BAD (or false), Image
+//		* will reload the metadata from the file and update the database.
+//		* MediaHandler::METADATA_GOOD for if the metadata is a-ok,
+//		* MediaHandler::METADATA_COMPATIBLE if metadata is old but backwards
+//		* compatible (which may or may not trigger a metadata reload).
+//		*
+//		* @note Returning self::METADATA_BAD will trigger a metadata reload from
+//		*  file on page view. Always returning this from a broken file, or suddenly
+//		*  triggering as bad metadata for a large number of files can cause
+//		*  performance problems.
+//		* @param File $image
+//		* @param String $metadata The metadata in serialized form
+//		* @return boolean
+//		*/
+//		function isMetadataValid($image, $metadata) {
+//			return self::METADATA_GOOD;
+//		}
+//
+//		/**
+//		* Get an array of standard (FormatMetadata type) metadata values.
+//		*
+//		* The returned data is largely the same as that from getMetadata(),
+//		* but formatted in a standard, stable, handler-independent way.
+//		* The idea being that some values like ImageDescription or Artist
+//		* are universal and should be retrievable in a handler generic way.
+//		*
+//		* The specific properties are the type of properties that can be
+//		* handled by the FormatMetadata class. These values are exposed to the
+//		* user via the filemetadata parser function.
+//		*
+//		* Details of the response format of this function can be found at
+//		* https://www.mediawiki.org/wiki/Manual:File_metadata_handling
+//		* tl/dr: the response is an associative array of
+//		* properties keyed by name, but the value can be complex. You probably
+//		* want to call one of the FormatMetadata::flatten* functions on the
+//		* property values before using them, or call
+//		* FormatMetadata::getFormattedData() on the full response array, which
+//		* transforms all values into prettified, human-readable text.
+//		*
+//		* Subclasses overriding this function must return a value which is a
+//		* valid API response fragment (all associative array keys are valid
+//		* XML tagnames).
+//		*
+//		* Note, if the file simply has no metadata, but the handler supports
+//		* this interface, it should return an empty array, not false.
+//		*
+//		* @param File $file
+//		* @return array|boolean False if interface not supported
+//		* @since 1.23
+//		*/
+//		public function getCommonMetaArray(File $file) {
+//			return false;
+//		}
+//
+//		/**
+//		* Get a MediaTransformOutput Object representing an alternate of the transformed
+//		* output which will call an intermediary thumbnail assist script.
+//		*
+//		* Used when the repository has a thumbnailScriptUrl option configured.
+//		*
+//		* Return false to fall back to the regular getTransform().
+//		* @param File $image
+//		* @param String $script
+//		* @param array $paramsVar
+//		* @return boolean|ThumbnailImage
+//		*/
+//		function getScriptedTransform($image, $script, $paramsVar) {
+//			return false;
+//		}
+
+	/**
+	* Get a MediaTransformOutput Object representing the transformed output. Does not
+	* actually do the transform.
+	*
+	* @param File $image The image Object
+	* @param String $dstPath Filesystem destination path
+	* @param String $dstUrl Destination URL to use in output HTML
+	* @param array $paramsVar Arbitrary set of parameters validated by $this->validateParam()
+	* @return MediaTransformOutput
+	*/
+	public XomwMediaTransformOutput getTransform(XomwFile image, byte[] dstPath, byte[] dstUrl, Xomw_params_handler handlerParams) {
+		return this.doTransform(image, dstPath, dstUrl, handlerParams, TRANSFORM_LATER);
+	}
+
+	/**
+	* Get a MediaTransformOutput Object representing the transformed output. Does the
+	* transform unless $flags contains self::TRANSFORM_LATER.
+	*
+	* @param File $image The image Object
+	* @param String $dstPath Filesystem destination path
+	* @param String $dstUrl Destination URL to use in output HTML
+	* @param array $paramsVar Arbitrary set of parameters validated by $this->validateParam()
+	*   Note: These parameters have *not* gone through $this->normaliseParams()
+	* @param int $flags A bitfield, may contain self::TRANSFORM_LATER
+	* @return MediaTransformOutput
+	*/
+	public XomwMediaTransformOutput doTransform(XomwFile image, byte[] dstPath, byte[] dstUrl, Xomw_params_handler handlerParams) {return doTransform(image, dstPath, dstUrl, handlerParams, 0);}
+	public abstract XomwMediaTransformOutput doTransform(XomwFile image, byte[] dstPath, byte[] dstUrl, Xomw_params_handler handlerParams, int flags);
+
+//		/**
+//		* Get the thumbnail extension and MIME type for a given source MIME type
+//		*
+//		* @param String $ext Extension of original file
+//		* @param String $mime MIME type of original file
+//		* @param array $paramsVar Handler specific rendering parameters
+//		* @return array Thumbnail extension and MIME type
+//		*/
+//		function getThumbType($ext, $mime, $paramsVar = null) {
+//			$magic = MimeMagic::singleton();
+//			if (!$ext || $magic->isMatchingExtension($ext, $mime) === false) {
+//				// The extension is not valid for this MIME type and we do
+//				// recognize the MIME type
+//				$extensions = $magic->getExtensionsForType($mime);
+//				if ($extensions) {
+//					return [ strtok($extensions, ' '), $mime ];
+//				}
+//			}
+//
+//			// The extension is correct (true) or the MIME type is unknown to
+//			// MediaWiki (null)
+//			return [ $ext, $mime ];
+//		}
+//
+//		/**
+//		* Get useful response headers for GET/HEAD requests for a file with the given metadata
+//		*
+//		* @param mixed $metadata Result of the getMetadata() function of this handler for a file
+//		* @return array
+//		*/
+//		public function getStreamHeaders($metadata) {
+//			return [];
+//		}
+
+	/**
+	* True if the handled types can be transformed
+	*
+	* @param File $file
+	* @return boolean
+	*/
+	@gplx.Virtual public boolean canRender(XomwFile file) {
+		return true;
+	}
+
+	/**
+	* True if handled types cannot be displayed directly in a browser
+	* but can be rendered
+	*
+	* @param File $file
+	* @return boolean
+	*/
+	public boolean mustRender(XomwFile file) {
+		return false;
+	}
+
+//		/**
+//		* True if the type has multi-page capabilities
+//		*
+//		* @param File $file
+//		* @return boolean
+//		*/
+//		public function isMultiPage($file) {
+//			return false;
+//		}
+//
+//		/**
+//		* Page count for a multi-page document, false if unsupported or unknown
+//		*
+//		* @param File $file
+//		* @return boolean
+//		*/
+//		function pageCount(File $file) {
+//			return false;
+//		}
+//
+//		/**
+//		* The material is vectorized and thus scaling is lossless
+//		*
+//		* @param File $file
+//		* @return boolean
+//		*/
+//		function isVectorized($file) {
+//			return false;
+//		}
+//
+//		/**
+//		* The material is an image, and is animated.
+//		* In particular, video material need not return true.
+//		* @note Before 1.20, this was a method of ImageHandler only
+//		*
+//		* @param File $file
+//		* @return boolean
+//		*/
+//		function isAnimatedImage($file) {
+//			return false;
+//		}
+//
+//		/**
+//		* If the material is animated, we can animate the thumbnail
+//		* @since 1.20
+//		*
+//		* @param File $file
+//		* @return boolean If material is not animated, handler may return any value.
+//		*/
+//		function canAnimateThumbnail($file) {
+//			return true;
+//		}
+//
+//		/**
+//		* False if the handler is disabled for all files
+//		* @return boolean
+//		*/
+//		function isEnabled() {
+//			return true;
+//		}
+//
+//		/**
+//		* Get an associative array of page dimensions
+//		* Currently "width" and "height" are understood, but this might be
+//		* expanded in the future.
+//		* Returns false if unknown.
+//		*
+//		* It is expected that handlers for paged media (e.g. DjVuHandler)
+//		* will override this method so that it gives the correct results
+//		* for each specific page of the file, using the $page argument.
+//		*
+//		* @note For non-paged media, use getImageSize.
+//		*
+//		* @param File $image
+//		* @param int $page What page to get dimensions of
+//		* @return array|boolean
+//		*/
+//		function getPageDimensions(File $image, $page) {
+//			$gis = $this->getImageSize($image, $image->getLocalRefPath());
+//			if ($gis) {
+//				return [
+//					'width' => $gis[0],
+//					'height' => $gis[1]
+//				];
+//			} else {
+//				return false;
+//			}
+//		}
+//
+//		/**
+//		* Generic getter for text layer.
+//		* Currently overloaded by PDF and DjVu handlers
+//		* @param File $image
+//		* @param int $page Page number to get information for
+//		* @return boolean|String Page text or false when no text found or if
+//		*   unsupported.
+//		*/
+//		function getPageText(File $image, $page) {
+//			return false;
+//		}
+//
+//		/**
+//		* Get the text of the entire document.
+//		* @param File $file
+//		* @return boolean|String The text of the document or false if unsupported.
+//		*/
+//		public function getEntireText(File $file) {
+//			$numPages = $file->pageCount();
+//			if (!$numPages) {
+//				// Not a multipage document
+//				return $this->getPageText($file, 1);
+//			}
+//			$document = '';
+//			for ($i = 1; $i <= $numPages; $i++) {
+//				$curPage = $this->getPageText($file, $i);
+//				if (is_string($curPage)) {
+//					$document .= $curPage . "\n";
+//				}
+//			}
+//			if ($document !== '') {
+//				return $document;
+//			}
+//			return false;
+//		}
+//
+//		/**
+//		* Get an array structure that looks like this:
+//		*
+//		* [
+//		*    'visible' => [
+//		*       'Human-readable name' => 'Human readable value',
+//		*       ...
+//		*    ],
+//		*    'collapsed' => [
+//		*       'Human-readable name' => 'Human readable value',
+//		*       ...
+//		*    ]
+//		* ]
+//		* The UI will format this into a table where the visible fields are always
+//		* visible, and the collapsed fields are optionally visible.
+//		*
+//		* The function should return false if there is no metadata to display.
+//		*/
+//
+//		/**
+//		* @todo FIXME: This interface is not very flexible. The media handler
+//		* should generate HTML instead. It can do all the formatting according
+//		* to some standard. That makes it possible to do things like visual
+//		* indication of grouped and chained streams in ogg container files.
+//		* @param File $image
+//		* @param boolean|IContextSource $context Context to use (optional)
+//		* @return array|boolean
+//		*/
+//		function formatMetadata($image, $context = false) {
+//			return false;
+//		}
+//
+//		/** sorts the visible/invisible field.
+//		* Split off from ImageHandler::formatMetadata, as used by more than
+//		* one type of handler.
+//		*
+//		* This is used by the media handlers that use the FormatMetadata class
+//		*
+//		* @param array $metadataArray Metadata array
+//		* @param boolean|IContextSource $context Context to use (optional)
+//		* @return array Array for use displaying metadata.
+//		*/
+//		function formatMetadataHelper($metadataArray, $context = false) {
+//			$result = [
+//				'visible' => [],
+//				'collapsed' => []
+//			];
+//
+//			$formatted = FormatMetadata::getFormattedData($metadataArray, $context);
+//			// Sort fields into visible and collapsed
+//			$visibleFields = $this->visibleMetadataFields();
+//			foreach ($formatted as $name => $value) {
+//				$tag = strtolower($name);
+//				self::addMeta($result,
+//					in_array($tag, $visibleFields) ? 'visible' : 'collapsed',
+//					'exif',
+//					$tag,
+//					$value
+//				);
+//			}
+//
+//			return $result;
+//		}
+//
+//		/**
+//		* Get a list of metadata items which should be displayed when
+//		* the metadata table is collapsed.
+//		*
+//		* @return array Array of strings
+//		*/
+//		protected function visibleMetadataFields() {
+//			return FormatMetadata::getVisibleFields();
+//		}
+//
+//		/**
+//		* This is used to generate an array element for each metadata value
+//		* That array is then used to generate the table of metadata values
+//		* on the image page
+//		*
+//		* @param array &$array An array containing elements for each type of visibility
+//		*   and each of those elements being an array of metadata items. This function adds
+//		*   a value to that array.
+//		* @param String $visibility ('visible' or 'collapsed') if this value is hidden
+//		*   by default.
+//		* @param String $type Type of metadata tag (currently always 'exif')
+//		* @param String $id The name of the metadata tag (like 'artist' for example).
+//		*   its name in the table displayed is the message "$type-$id" (Ex exif-artist).
+//		* @param String $value Thingy goes into a wikitext table; it used to be escaped but
+//		*   that was incompatible with previous practise of customized display
+//		*   with wikitext formatting via messages such as 'exif-model-value'.
+//		*   So the escaping is taken back out, but generally this seems a confusing
+//		*   interface.
+//		* @param boolean|String $param Value to pass to the message for the name of the field
+//		*   as $1. Currently this parameter doesn't seem to ever be used.
+//		*
+//		* Note, everything here is passed through the parser later on (!)
+//		*/
+//		protected static function addMeta(&$array, $visibility, $type, $id, $value, $param = false) {
+//			$msg = wfMessage("$type-$id", $param);
+//			if ($msg->exists()) {
+//				$name = $msg->text();
+//			} else {
+//				// This is for future compatibility when using instant commons.
+//				// So as to not display as ugly a name if a new metadata
+//				// property is defined that we don't know about
+//				// (not a major issue since such a property would be collapsed
+//				// by default).
+//				wfDebug(__METHOD__ . ' Unknown metadata name: ' . $id . "\n");
+//				$name = wfEscapeWikiText($id);
+//			}
+//			$array[$visibility][] = [
+//				'id' => "$type-$id",
+//				'name' => $name,
+//				'value' => $value
+//			];
+//		}
+//
+//		/**
+//		* Short description. Shown on Special:Search results.
+//		*
+//		* @param File $file
+//		* @return String
+//		*/
+//		function getShortDesc($file) {
+//			return self::getGeneralShortDesc($file);
+//		}
+//
+//		/**
+//		* Long description. Shown under image on image description page surounded by ().
+//		*
+//		* @param File $file
+//		* @return String
+//		*/
+//		function getLongDesc($file) {
+//			return self::getGeneralLongDesc($file);
+//		}
+//
+//		/**
+//		* Used instead of getShortDesc if there is no handler registered for file.
+//		*
+//		* @param File $file
+//		* @return String
+//		*/
+//		static function getGeneralShortDesc($file) {
+//			global $wgLang;
+//
+//			return htmlspecialchars($wgLang->formatSize($file->getSize()));
+//		}
+//
+//		/**
+//		* Used instead of getLongDesc if there is no handler registered for file.
+//		*
+//		* @param File $file
+//		* @return String
+//		*/
+//		static function getGeneralLongDesc($file) {
+//			return wfMessage('file-info')->sizeParams($file->getSize())
+//				->paramsVar('<span class="mime-type">' . $file->getMimeType() . '</span>')->parse();
+//		}
+
+	/**
+	* Calculate the largest thumbnail width for a given original file size
+	* such that the thumbnail's height is at most $maxHeight.
+	* @param int $boxWidth Width of the thumbnail box.
+	* @param int $boxHeight Height of the thumbnail box.
+	* @param int $maxHeight Maximum height expected for the thumbnail.
+	* @return int
+	*/
+	public static int fitBoxWidth(int boxWidth, int boxHeight, int maxHeight) {
+		double idealWidth = boxWidth * maxHeight / boxHeight;
+		int roundedUp = Math_.Ceil_as_int(idealWidth);
+		if (Math_.Round(roundedUp * boxHeight / boxWidth, 0) > maxHeight) {
+			return Math_.Floor_as_int(idealWidth);
+		} else {
+			return roundedUp;
+		}
+	}
+
+//		/**
+//		* Shown in file history box on image description page.
+//		*
+//		* @param File $file
+//		* @return String Dimensions
+//		*/
+//		function getDimensionsString($file) {
+//			return '';
+//		}
+//
+//		/**
+//		* Modify the parser Object post-transform.
+//		*
+//		* This is often used to do $parser->addOutputHook(),
+//		* in order to add some javascript to render a viewer.
+//		* See TimedMediaHandler or OggHandler for an example.
+//		*
+//		* @param Parser $parser
+//		* @param File $file
+//		*/
+//		function parserTransformHook($parser, $file) {
+//		}
+//
+//		/**
+//		* File validation hook called on upload.
+//		*
+//		* If the file at the given local path is not valid, or its MIME type does not
+//		* match the handler class, a Status Object should be returned containing
+//		* relevant errors.
+//		*
+//		* @param String $fileName The local path to the file.
+//		* @return Status
+//		*/
+//		function verifyUpload($fileName) {
+//			return Status::newGood();
+//		}
+//
+//		/**
+//		* Check for zero-sized thumbnails. These can be generated when
+//		* no disk space is available or some other error occurs
+//		*
+//		* @param String $dstPath The location of the suspect file
+//		* @param int $retval Return value of some shell process, file will be deleted if this is non-zero
+//		* @return boolean True if removed, false otherwise
+//		*/
+//		function removeBadFile($dstPath, $retval = 0) {
+//			if (file_exists($dstPath)) {
+//				$thumbstat = stat($dstPath);
+//				if ($thumbstat['size'] == 0 || $retval != 0) {
+//					$result = unlink($dstPath);
+//
+//					if ($result) {
+//						wfDebugLog('thumbnail',
+//							sprintf('Removing bad %d-byte thumbnail "%s". unlink() succeeded',
+//								$thumbstat['size'], $dstPath));
+//					} else {
+//						wfDebugLog('thumbnail',
+//							sprintf('Removing bad %d-byte thumbnail "%s". unlink() failed',
+//								$thumbstat['size'], $dstPath));
+//					}
+//
+//					return true;
+//				}
+//			}
+//
+//			return false;
+//		}
+//
+//		/**
+//		* Remove files from the purge list.
+//		*
+//		* This is used by some video handlers to prevent ?action=purge
+//		* from removing a transcoded video, which is expensive to
+//		* regenerate.
+//		*
+//		* @see LocalFile::purgeThumbnails
+//		*
+//		* @param array $files
+//		* @param array $options Purge options. Currently will always be
+//		*  an array with a single key 'forThumbRefresh' set to true.
+//		*/
+//		public function filterThumbnailPurgeList(&$files, $options) {
+//			// Do nothing
+//		}
+//
+//		/**
+//		* True if the handler can rotate the media
+//		* @since 1.24 non-static. From 1.21-1.23 was static
+//		* @return boolean
+//		*/
+//		public function canRotate() {
+//			return false;
+//		}
+//
+//		/**
+//		* On supporting image formats, try to read out the low-level orientation
+//		* of the file and return the angle that the file needs to be rotated to
+//		* be viewed.
+//		*
+//		* This information is only useful when manipulating the original file;
+//		* the width and height we normally work with is logical, and will match
+//		* any produced output views.
+//		*
+//		* For files we don't know, we return 0.
+//		*
+//		* @param File $file
+//		* @return int 0, 90, 180 or 270
+//		*/
+//		public function getRotation($file) {
+//			return 0;
+//		}
+//
+//		/**
+//		* Log an error that occurred in an external process
+//		*
+//		* Moved from BitmapHandler to MediaHandler with MediaWiki 1.23
+//		*
+//		* @since 1.23
+//		* @param int $retval
+//		* @param String $err Error reported by command. Anything longer than
+//		* MediaHandler::MAX_ERR_LOG_SIZE is stripped off.
+//		* @param String $cmd
+//		*/
+//		protected function logErrorForExternalProcess($retval, $err, $cmd) {
+//			# Keep error output limited (bug 57985)
+//			$errMessage = trim(substr($err, 0, self::MAX_ERR_LOG_SIZE));
+//
+//			wfDebugLog('thumbnail',
+//				sprintf('thumbnail failed on %s: error %d "%s" from "%s"',
+//						wfHostname(), $retval, $errMessage, $cmd));
+//		}
+//
+//		/**
+//		* Get list of languages file can be viewed in.
+//		*
+//		* @param File $file
+//		* @return String[] Array of language codes, or empty array if unsupported.
+//		* @since 1.23
+//		*/
+//		public function getAvailableLanguages(File $file) {
+//			return [];
+//		}
+//
+//		/**
+//		* On file types that support renderings in multiple languages,
+//		* which language is used by default if unspecified.
+//		*
+//		* If getAvailableLanguages returns a non-empty array, this must return
+//		* a valid language code. Otherwise can return null if files of this
+//		* type do not support alternative language renderings.
+//		*
+//		* @param File $file
+//		* @return String|null Language code or null if multi-language not supported for filetype.
+//		* @since 1.23
+//		*/
+//		public function getDefaultRenderLanguage(File $file) {
+//			return null;
+//		}
+//
+//		/**
+//		* If its an audio file, return the length of the file. Otherwise 0.
+//		*
+//		* File::getLength() existed for a long time, but was calling a method
+//		* that only existed in some subclasses of this class (The TMH ones).
+//		*
+//		* @param File $file
+//		* @return float length in seconds
+//		* @since 1.23
+//		*/
+//		public function getLength($file) {
+//			return 0.0;
+//		}
+//
+//		/**
+//		* True if creating thumbnails from the file is large or otherwise resource-intensive.
+//		* @param File $file
+//		* @return boolean
+//		*/
+//		public function isExpensiveToThumbnail($file) {
+//			return false;
+//		}
+//
+//		/**
+//		* Returns whether or not this handler supports the chained generation of thumbnails according
+//		* to buckets
+//		* @return boolean
+//		* @since 1.24
+//		*/
+//		public function supportsBucketing() {
+//			return false;
+//		}
+//
+//		/**
+//		* Returns a normalised paramsVar array for which parameters have been cleaned up for bucketing
+//		* purposes
+//		* @param array $paramsVar
+//		* @return array
+//		*/
+//		public function sanitizeParamsForBucketing($paramsVar) {
+//			return $paramsVar;
+//		}
+//
+//		/**
+//		* Gets configuration for the file warning message. Return value of
+//		* the following structure:
+//		*   [
+//		*     // Required, module with messages loaded for the client
+//		*     'module' => 'example.filewarning.messages',
+//		*     // Required, array of names of messages
+//		*     'messages' => [
+//		*       // Required, main warning message
+//		*       'main' => 'example-filewarning-main',
+//		*       // Optional, header for warning dialog
+//		*       'header' => 'example-filewarning-header',
+//		*       // Optional, footer for warning dialog
+//		*       'footer' => 'example-filewarning-footer',
+//		*       // Optional, text for more-information link (see below)
+//		*       'info' => 'example-filewarning-info',
+//		*     ],
+//		*     // Optional, link for more information
+//		*     'link' => 'http://example.com',
+//		*   ]
+//		*
+//		* Returns null if no warning is necessary.
+//		* @param File $file
+//		* @return array|null
+//		*/
+//		public function getWarningConfig($file) {
+//			return null;
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwMediaHandlerFactory.java

@@ -13,3 +13,49 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+// XO.MW:MW has registry and instance cache; XO only has instance
+// XO.MW:SYNC:1.29; DATE:2017-02-05
+public class XomwMediaHandlerFactory {
+	private final    Hash_adp_bry handlers = Hash_adp_bry.cs();
+
+	// XO.MW:SYNC:1.29; DATE:2017-02-05
+	public XomwMediaHandlerFactory() {
+		// Default, MediaWiki core media handlers
+//			'image/jpeg' => JpegHandler::class,
+		handlers.Add(Mime__image__png, new XomwTransformationalImageHandler(Mime__image__png));	// PngHandler
+//			'image/gif' => GIFHandler::class,
+//			'image/tiff' => TiffHandler::class,
+//			'image/webp' => WebPHandler::class,
+//			'image/x-ms-bmp' => BmpHandler::class,
+//			'image/x-bmp' => BmpHandler::class,
+//			'image/x-xcf' => XCFHandler::class,
+//			'image/svg+xml' => SvgHandler::class, // official
+//			'image/svg' => SvgHandler::class, // compat
+//			'image/vnd.djvu' => DjVuHandler::class, // official
+//			'image/x.djvu' => DjVuHandler::class, // compat
+//			'image/x-djvu' => DjVuHandler::class, // compat
+
+	}
+
+	// XO.MW:SYNC:1.29; DATE:2017-02-05
+	public XomwMediaHandler getHandler(byte[] type) {
+		return (XomwMediaHandler)handlers.Get_by(type);
+	}
+
+	public static byte[]
+	  Mime__image__jpeg                    = Bry_.new_a7("image/jpeg")
+	, Mime__image__png                     = Bry_.new_a7("image/png")
+	, Mime__image__gif                     = Bry_.new_a7("image/gif")
+	, Mime__image__tiff                    = Bry_.new_a7("image/tiff")
+	, Mime__image__webp                    = Bry_.new_a7("image/webp")
+	, Mime__image__x_ms_bmp                = Bry_.new_a7("image/x-ms-bmp")
+	, Mime__image__x_bmp                   = Bry_.new_a7("image/x-bmp")
+	, Mime__image__x_xcf                   = Bry_.new_a7("image/x-xcf")
+	, Mime__image__svg_xml                 = Bry_.new_a7("image/svg+xml")
+	, Mime__image__svg                     = Bry_.new_a7("image/svg")
+	, Mime__image__vnd_djvu                = Bry_.new_a7("image/vnd.djvu")
+	, Mime__image__x_djvu_dot              = Bry_.new_a7("image/x.djvu")
+	, Mime__image__x_djvu_dash             = Bry_.new_a7("image/x-djvu")
+	;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwMediaTransformOutput.java

@@ -13,3 +13,266 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.langs.htmls.*;
+import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*;
+public abstract class XomwMediaTransformOutput {
+	public XomwMediaTransformOutput(XomwFile file, byte[] url, byte[] path, int width, int height) {
+		this.file = file;
+		this.url = url;
+		this.width = width;
+		this.height = height;
+	}
+//		/** @var array Associative array mapping optional supplementary image files
+//		*  from pixel density (eg 1.5 or 2) to additional URLs.
+//		*/
+//		public $responsiveUrls = [];
+
+	/** @var File */
+	private final    XomwFile file;
+
+	/** @var int Image width */
+	protected final    int width;
+
+	/** @var int Image height */
+	protected final    int height;
+
+	/** @var String URL path to the thumb */
+	protected final    byte[] url;
+
+//		/** @var boolean|String */
+//		protected $page;
+//
+//		/** @var boolean|String Filesystem path to the thumb  */
+//		protected $path;
+//
+//		/** @var boolean|String Language code, false if not set */
+//		protected $lang;
+//
+//		/** @var boolean|String Permanent storage path  */
+//		protected $storagePath = false;
+
+
+	/**
+	* @return int Width of the output box
+	*/
+	public int getWidth() {
+		return this.width;
+	}
+
+	/**
+	* @return int Height of the output box
+	*/
+	public int getHeight() {
+		return this.height;
+	}
+
+//		/**
+//		* @return File
+//		*/
+//		public function getFile() {
+//			return $this->file;
+//		}
+//
+//		/**
+//		* Get the final extension of the thumbnail.
+//		* Returns false for scripted transformations.
+//		* @return String|boolean
+//		*/
+//		public function getExtension() {
+//			return $this->path ? FileBackend::extensionFromPath( $this->path ) : false;
+//		}
+//
+//		/**
+//		* @return String|boolean The thumbnail URL
+//		*/
+//		public function getUrl() {
+//			return $this->url;
+//		}
+//
+//		/**
+//		* @return String|boolean The permanent thumbnail storage path
+//		*/
+//		public function getStoragePath() {
+//			return $this->storagePath;
+//		}
+//
+//		/**
+//		* @param String $storagePath The permanent storage path
+//		* @return void
+//		*/
+//		public function setStoragePath( $storagePath ) {
+//			$this->storagePath = $storagePath;
+//			if ( $this->path === false ) {
+//				$this->path = $storagePath;
+//			}
+//		}
+
+	/**
+	* Fetch HTML for this transform output
+	*
+	* @param array $options Associative array of options. Boolean options
+	*     should be indicated with a value of true for true, and false or
+	*     absent for false.
+	*
+	*     alt          Alternate text or caption
+	*     desc-link    Boolean, show a description link
+	*     file-link    Boolean, show a file download link
+	*     custom-url-link    Custom URL to link to
+	*     custom-title-link  Custom Title Object to link to
+	*     valign       vertical-align property, if the output is an inline element
+	*     img-class    Class applied to the "<img>" tag, if there is such a tag
+	*
+	* For images, desc-link and file-link are implemented as a click-through. For
+	* sounds and videos, they may be displayed in other ways.
+	*
+	* @return String
+	*/
+	public abstract void toHtml(Bry_bfr bfr, Bry_bfr tmp, Xomw_params_mto options);
+
+//		/**
+//		* This will be overridden to return true in error classes
+//		* @return boolean
+//		*/
+//		public function isError() {
+//			return false;
+//		}
+//
+//		/**
+//		* Check if an output thumbnail file actually exists.
+//		*
+//		* This will return false if there was an error, the
+//		* thumbnail is to be handled client-side only, or if
+//		* transformation was deferred via TRANSFORM_LATER.
+//		* This file may exist as a new file in /tmp, a file
+//		* in permanent storage, or even refer to the original.
+//		*
+//		* @return boolean
+//		*/
+//		public function hasFile() {
+//			// If TRANSFORM_LATER, $this->path will be false.
+//			// Note: a null path means "use the source file".
+//			return ( !$this->isError() && ( $this->path || $this->path === null ) );
+//		}
+//
+//		/**
+//		* Check if the output thumbnail is the same as the source.
+//		* This can occur if the requested width was bigger than the source.
+//		*
+//		* @return boolean
+//		*/
+//		public function fileIsSource() {
+//			return ( !$this->isError() && $this->path === null );
+//		}
+//
+//		/**
+//		* Get the path of a file system copy of the thumbnail.
+//		* Callers should never write to this path.
+//		*
+//		* @return String|boolean Returns false if there isn't one
+//		*/
+//		public function getLocalCopyPath() {
+//			if ( $this->isError() ) {
+//				return false;
+//			} elseif ( $this->path === null ) {
+//				return $this->file->getLocalRefPath(); // assume thumb was not scaled
+//			} elseif ( FileBackend::isStoragePath( $this->path ) ) {
+//				$be = $this->file->getRepo()->getBackend();
+//				// The temp file will be process cached by FileBackend
+//				$fsFile = $be->getLocalReference( [ 'src' => $this->path ] );
+//
+//				return $fsFile ? $fsFile->getPath() : false;
+//			} else {
+//				return $this->path; // may return false
+//			}
+//		}
+//
+//		/**
+//		* Stream the file if there were no errors
+//		*
+//		* @param array $headers Additional HTTP headers to send on success
+//		* @return Status
+//		* @since 1.27
+//		*/
+//		public function streamFileWithStatus( $headers = [] ) {
+//			if ( !$this->path ) {
+//				return Status::newFatal( 'backend-fail-stream', '<no path>' );
+//			} elseif ( FileBackend::isStoragePath( $this->path ) ) {
+//				$be = $this->file->getRepo()->getBackend();
+//				return $be->streamFile( [ 'src' => $this->path, 'headers' => $headers ] );
+//			} else { // FS-file
+//				$success = StreamFile::stream( $this->getLocalCopyPath(), $headers );
+//				return $success ? Status::newGood() : Status::newFatal( 'backend-fail-stream', $this->path );
+//			}
+//		}
+//
+//		/**
+//		* Stream the file if there were no errors
+//		*
+//		* @deprecated since 1.26, use streamFileWithStatus
+//		* @param array $headers Additional HTTP headers to send on success
+//		* @return boolean Success
+//		*/
+//		public function streamFile( $headers = [] ) {
+//			$this->streamFileWithStatus( $headers )->isOK();
+//		}
+//
+//		/**
+//		* Wrap some XHTML text in an anchor tag with the given attributes
+//		*
+//		* @param array $linkAttribs
+//		* @param String $contents
+//		* @return String
+//		*/
+//		protected function linkWrap( $linkAttribs, $contents ) {
+//			if ( $linkAttribs ) {
+//				return Xml::tags( 'a', $linkAttribs, $contents );
+//			} else {
+//				return $contents;
+//			}
+//		}
+
+	/**
+	* @param String $title
+	* @param String|array $prms Query parameters to add
+	* @return array
+	*/
+	public void getDescLinkAttribs(List_adp attribs, byte[] title, List_adp prms) {
+		byte[] query = Bry_.Empty;
+//			if ( is_array( prms ) ) {
+//				$query = prms;
+//			} else {
+//				$query = [];
+//			}
+//			if ( $this->page && $this->page !== 1 ) {
+//				$query['page'] = $this->page;
+//			}
+//			if ( $this->lang ) {
+//				$query['lang'] = $this->lang;
+//			}
+//
+//			if ( is_string( prms ) && prms !== '' ) {
+//				$query = prms . '&' . wfArrayToCgi( $query );
+//			}
+
+		attribs.Clear();
+		attribs.Add_many(Gfh_atr_.Bry__href, this.file.getTitle().getLocalURL(query));
+		attribs.Add_many(Gfh_atr_.Bry__class, Bry__class__image);
+		if (title != null) {
+			attribs.Add_many(Gfh_atr_.Bry__title, title);
+		}
+	}
+
+	// Wrap some XHTML text in an anchor tag with the given attributes
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	protected void Link_wrap(Bry_bfr bfr, List_adp link_attribs, byte[] contents) {
+		if (link_attribs != null) {
+			XomwXml.Tags(bfr, Gfh_tag_.Bry__a, link_attribs, contents);
+		}
+		else {
+			bfr.Add(contents);
+		}
+	}
+	private static final    byte[] Bry__class__image = Bry_.new_a7("image");
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwThumbnailImage.java

@@ -13,3 +13,199 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.langs.htmls.*;
+import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*;
+// Media transform output for images
+public class XomwThumbnailImage extends XomwMediaTransformOutput {	private final    List_adp attribs = List_adp_.New(), link_attribs = List_adp_.New();
+	public XomwThumbnailImage(XomwFile file, byte[] url, byte[] path, int w, int h) {super(file, url, path, w, h);
+	}
+	/**
+	* Get a thumbnail Object from a file and parameters.
+	* If path is set to null, the output file is treated as a source copy.
+	* If path is set to false, no output file will be created.
+	* parameters should include, as a minimum, (file) 'width' and 'height'.
+	* It may also include a 'page' parameter for multipage files.
+	*
+	* @param File file
+	* @param String url URL path to the thumb
+	* @param String|boolean path Filesystem path to the thumb
+	* @param array parameters Associative array of parameters
+	*/
+	public XomwThumbnailImage(XomwFile file, byte[] url, byte[] path, Xomw_params_handler parameters) {super(file, url, path, parameters.width, parameters.height);
+//			defaults = [
+//				'page' => false,
+//				'lang' => false
+//			];
+//
+//			if (is_array(parameters)) {
+//				actualParams = parameters + defaults;
+//			} else {
+//				// Using old format, should convert. Later a warning could be added here.
+//				numArgs = func_num_args();
+//				actualParams = [
+//					'width' => path,
+//					'height' => parameters,
+//					'page' => (numArgs > 5) ? func_get_arg(5) : false
+//				] + defaults;
+//				path = (numArgs > 4) ? func_get_arg(4) : false;
+//			}
+
+//			this->file = file;
+//			this->url = url;
+//			this->path = path;
+
+		// These should be integers when they get here.
+		// If not, there's a bug somewhere.  But let's at
+		// least produce valid HTML code regardless.
+//			this->width = round(actualParams['width']);
+//			this->height = round(actualParams['height']);
+
+//			this->page = actualParams['page'];
+//			this->lang = actualParams['lang'];
+	}
+
+	/**
+	* Return HTML <img ... /> tag for the thumbnail, will include
+	* width and height attributes and a blank alt text (as required).
+	*
+	* @param array options Associative array of options. Boolean options
+	*     should be indicated with a value of true for true, and false or
+	*     absent for false.
+	*
+	*     alt          HTML alt attribute
+	*     title        HTML title attribute
+	*     desc-link    Boolean, show a description link
+	*     file-link    Boolean, show a file download link
+	*     valign       vertical-align property, if the output is an inline element
+	*     img-class    Class applied to the \<img\> tag, if there is such a tag
+	*     desc-query   String, description link query prms
+	*    @Override  width     Override width attribute. Should generally not set
+	*    @Override  height    Override height attribute. Should generally not set
+	*     no-dimensions      Boolean, skip width and height attributes (useful if
+	*                        set in CSS)
+	*     custom-url-link    Custom URL to link to
+	*     custom-title-link  Custom Title Object to link to
+	*     custom target-link Value of the target attribute, for custom-target-link
+	*     parser-extlink-*   Attributes added by parser for external links:
+	*          parser-extlink-rel: add rel="nofollow"
+	*          parser-extlink-target: link target, but overridden by custom-target-link
+	*
+	* For images, desc-link and file-link are implemented as a click-through. For
+	* sounds and videos, they may be displayed in other ways.
+	*
+	* @throws MWException
+	* @return String
+	*/
+	// Return HTML <img ... /> tag for the thumbnail, will include
+	// width and height attributes and a blank alt text (as required).
+	//
+	// @param array options Associative array of options. Boolean options
+	//     should be indicated with a value of true for true, and false or
+	//     absent for false.
+	//
+	//     alt          HTML alt attribute
+	//     title        HTML title attribute
+	//     desc-link    Boolean, show a description link
+	//     file-link    Boolean, show a file download link
+	//     valign       vertical-align property, if the output is an inline element
+	//     img-class    Class applied to the \<img\> tag, if there is such a tag
+	//     desc-query   String, description link query prms
+	//     override-width     Override width attribute. Should generally not set
+	//     override-height    Override height attribute. Should generally not set
+	//     no-dimensions      Boolean, skip width and height attributes (useful if
+	//                        set in CSS)
+	//     custom-url-link    Custom URL to link to
+	//     custom-title-link  Custom Title Object to link to
+	//     custom target-link Value of the target attribute, for custom-target-link
+	//     parser-extlink-*   Attributes added by parser for external links:
+	//          parser-extlink-rel: add rel="nofollow"
+	//          parser-extlink-target: link target, but overridden by custom-target-link
+	//
+	// For images, desc-link and file-link are implemented as a click-through. For
+	// sounds and videos, they may be displayed in other ways.
+	// XO.MW:SYNC:1.29; DATE:2017-02-03
+	@Override public void toHtml(Bry_bfr bfr, Bry_bfr tmp, Xomw_params_mto options) {
+		byte[] alt = options.alt;
+
+//			byte[] query = options.desc_query;
+
+		attribs.Clear();
+		attribs.Add_many(Gfh_atr_.Bry__alt, alt);
+		attribs.Add_many(Gfh_atr_.Bry__src, url);
+		boolean link_attribs_is_null = false;
+		if (!XophpUtility.empty(options.custom_url_link)) {
+			link_attribs.Clear();
+			link_attribs.Add_many(Gfh_atr_.Bry__href, options.custom_url_link);
+			if (!XophpUtility.empty(options.title)) {
+				link_attribs.Add_many(Gfh_atr_.Bry__title, options.title);
+			}
+			if (XophpUtility.empty(options.custom_target_link)) {
+				link_attribs.Add_many(Gfh_atr_.Bry__target, options.custom_target_link);
+			}
+			else if (XophpUtility.empty(options.parser_extlink_target)) {
+				link_attribs.Add_many(Gfh_atr_.Bry__target, options.parser_extlink_target);
+			}
+			if (XophpUtility.empty(options.parser_extlink_rel)) {
+				link_attribs.Add_many(Gfh_atr_.Bry__rel, options.parser_extlink_rel);
+			}
+		}
+		else if (!XophpUtility.empty(options.custom_title_link)) {
+//				byte[] title = options.custom_title_link;
+//				link_attribs.Clear();
+//				link_attribs.Add_many(Gfh_atr_.Bry__href, title.Get_link_url());
+//				byte[] options_title = options.title;
+//				link_attribs.Add_many(Gfh_atr_.Bry__title, XophpUtility.empty(options_title) ? title.Get_full_text() : options_title);
+		}
+		else if (!XophpUtility.empty(options.desc_link)) {
+//				link_attribs = this.getDescLinkAttribs(
+//					empty(options['title']) ? null : options['title'],
+//					$query
+//				);
+			link_attribs.Clear();
+			this.getDescLinkAttribs(link_attribs, 
+				XophpUtility.empty(options.title) ? null : options.title,
+				null);
+		}
+		else if (!XophpUtility.empty(options.file_link)) {
+//				link_attribs.Clear();
+//				link_attribs.Add_many(Gfh_atr_.Bry__href, file.Get_url());
+		}
+		else {
+			link_attribs_is_null = true;
+			if (!XophpUtility.empty(options.title)) {
+				attribs.Add_many(Gfh_atr_.Bry__title, options.title);
+			}
+		}
+
+		if (XophpUtility.empty(options.no_dimensions)) {
+			attribs.Add_many(Gfh_atr_.Bry__width, Int_.To_bry(width));
+			attribs.Add_many(Gfh_atr_.Bry__height, Int_.To_bry(height));
+		}
+		if (!XophpUtility.empty(options.valign)) {
+			attribs.Add_many(Gfh_atr_.Bry__style, Bry_.Add(Bry__vertical_align, options.valign));
+		}
+		if (!XophpUtility.empty(options.img_cls)) {
+			attribs.Add_many(Gfh_atr_.Bry__class, options.img_cls);
+		}
+		if (XophpUtility.isset(options.override_height)) {
+			attribs.Add_many(Gfh_atr_.Bry__class, options.override_height);
+		}
+		if (XophpUtility.isset(options.override_width)) {
+			attribs.Add_many(Gfh_atr_.Bry__width, options.override_height);
+		}
+
+		// Additional densities for responsive images, if specified.
+		// If any of these urls is the same as src url, it'll be excluded.
+//			$responsiveUrls = array_diff(this.responsiveUrls, [ this.url ]);
+//			if (!XophpUtility.empty($responsiveUrls)) {
+//				$attribs['srcset'] = Html::srcSet($responsiveUrls);
+//			}
+
+		// XO.MW.HOOK:ThumbnailBeforeProduceHTML
+		XomwXml.Element(tmp, Gfh_tag_.Bry__img, attribs, Bry_.Empty, Bool_.Y);
+		Link_wrap(bfr, link_attribs_is_null ? null : link_attribs, tmp.To_bry_and_clear());
+	}
+	private static final    byte[] Bry__vertical_align = Bry_.new_a7("vertical-align: ");
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/media/XomwTransformationalImageHandler.java

@@ -13,3 +13,597 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.media; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.filerepo.file.*;
+import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+public class XomwTransformationalImageHandler extends XomwImageHandler {	public XomwTransformationalImageHandler(byte[] key) {super(key);
+	}
+
+	/**
+	* @param File image
+	* @param array paramsVar Transform parameters. Entries with the keys 'width'
+	* and 'height' are the respective screen width and height, while the keys
+	* 'physicalWidth' and 'physicalHeight' indicate the thumbnail dimensions.
+	* @return boolean
+	*/
+	@Override public boolean normaliseParams(XomwFile image, Xomw_params_handler prms) {
+		if (!super.normaliseParams(image, prms)) {
+			return false;
+		}
+
+		// Obtain the source, pre-rotation dimensions
+		int srcWidth = image.getWidth(prms.page);
+		int srcHeight = image.getHeight(prms.page);
+
+		// Don't make an image bigger than the source
+		if (prms.physicalWidth >= srcWidth) {
+			prms.physicalWidth = srcWidth;
+			prms.physicalHeight = srcHeight;
+
+			// Skip scaling limit checks if no scaling is required
+			// due to requested size being bigger than source.
+			if (!image.mustRender()) {
+				return true;
+			}
+		}
+
+		return true;
+	}
+
+//		/**
+//		* Extracts the width/height if the image will be scaled before rotating
+//		*
+//		* This will match the physical size/aspect ratio of the original image
+//		* prior to application of the rotation -- so for a portrait image that's
+//		* stored as raw landscape with 90-degress rotation, the resulting size
+//		* will be wider than it is tall.
+//		*
+//		* @param array paramsVar Parameters as returned by normaliseParams
+//		* @param int rotation The rotation angle that will be applied
+//		* @return array (width, height) array
+//		*/
+//		public function extractPreRotationDimensions(paramsVar, rotation) {
+//			if (rotation == 90 || rotation == 270) {
+//				// We'll resize before rotation, so swap the dimensions again
+//				width = paramsVar['physicalHeight'];
+//				height = paramsVar['physicalWidth'];
+//			} else {
+//				width = paramsVar['physicalWidth'];
+//				height = paramsVar['physicalHeight'];
+//			}
+//
+//			return [ width, height ];
+//		}
+//
+	/**
+	* Create a thumbnail.
+	*
+	* This sets up various parameters, and then calls a helper method
+	* based on this.getScalerType in order to scale the image.
+	*
+	* @param File image
+	* @param String dstPath
+	* @param String dstUrl
+	* @param array paramsVar
+	* @param int flags
+	* @return MediaTransformError|ThumbnailImage|TransformParameterError
+	*/
+	@Override public XomwMediaTransformOutput doTransform(XomwFile image, byte[] dstPath, byte[] dstUrl, Xomw_params_handler prms, int flags) {
+//			if (!this.normaliseParams(image, paramsVar)) {
+//				return new TransformParameterError(paramsVar);
+//			}
+//
+//			// Create a parameter array to pass to the scaler
+		Xomw_params_scalar scalerParams = new Xomw_params_scalar();
+//			// The size to which the image will be resized
+		scalerParams.physicalWidth = prms.physicalWidth;
+		scalerParams.physicalHeight = prms.physicalHeight;
+//			'physicalDimensions' => "{paramsVar['physicalWidth']}x{paramsVar['physicalHeight']}",
+		// The size of the image on the page
+		scalerParams.clientWidth = prms.width;
+		scalerParams.clientHeight = prms.height;
+		// Comment as will be added to the Exif of the thumbnail
+//			'comment' => isset(paramsVar['descriptionUrl'])
+//				? "File source: {paramsVar['descriptionUrl']}"
+//				: '',
+		// Properties of the original image
+		scalerParams.srcWidth = image.getWidth();
+		scalerParams.srcHeight = image.getHeight();
+		scalerParams.mimeType = image.getMimeType();
+		scalerParams.dstPath = dstPath;
+		scalerParams.dstUrl = dstUrl;
+//			'interlace' => isset(paramsVar['interlace']) ? paramsVar['interlace'] : false,
+
+//			if (isset(paramsVar['quality']) && paramsVar['quality'] === 'low') {
+//				scalerParams['quality'] = 30;
+//			}
+
+		// For subclasses that might be paged.
+//			if (image.isMultipage() && isset(paramsVar['page'])) {
+//				scalerParams['page'] = intval(paramsVar['page']);
+//			}
+
+		// Determine scaler type
+//			scaler = this.getScalerType(dstPath);
+//
+//			if (is_array(scaler)) {
+//				scalerName = get_class(scaler[0]);
+//			} else {
+//				scalerName = scaler;
+//			}
+//
+//			wfDebug(__METHOD__ . ": creating {scalerParams['physicalDimensions']} " .
+//				"thumbnail at dstPath using scaler scalerName\n");
+
+		if (!image.mustRender() &&
+			scalerParams.physicalWidth == scalerParams.srcWidth
+			&& scalerParams.physicalHeight == scalerParams.srcHeight
+//				&& !isset(scalerParams['quality'])
+		) {
+
+			// normaliseParams (or the user) wants us to return the unscaled image
+//				wfDebug(__METHOD__ . ": returning unscaled image\n");
+
+			return this.getClientScalingThumbnailImage(image, scalerParams);
+		}
+
+//			if (scaler == 'client') {
+//				// Client-side image scaling, use the source URL
+//				// Using the destination URL in a TRANSFORM_LATER request would be incorrect
+//				return this.getClientScalingThumbnailImage(image, scalerParams);
+//			}
+//
+//			if (image.isTransformedLocally() && !this.isImageAreaOkForThumbnaling(image, paramsVar)) {
+//				global wgMaxImageArea;
+//				return new TransformTooBigImageAreaError(paramsVar, wgMaxImageArea);
+//			}
+//
+//			if (flags & self::TRANSFORM_LATER) {
+//				wfDebug(__METHOD__ . ": Transforming later per flags.\n");
+//				newParams = [
+//					'width' => scalerParams['clientWidth'],
+//					'height' => scalerParams['clientHeight']
+//				];
+//				if (isset(paramsVar['quality'])) {
+//					newParams['quality'] = paramsVar['quality'];
+//				}
+//				if (isset(paramsVar['page']) && paramsVar['page']) {
+//					newParams['page'] = paramsVar['page'];
+//				}
+//				return new XomwThumbnailImage(image, dstUrl, null, newParams);
+			return new XomwThumbnailImage(image, dstUrl, null, prms);
+//			}
+//
+//			// Try to make a target path for the thumbnail
+//			if (!wfMkdirParents(dirname(dstPath), null, __METHOD__)) {
+//				wfDebug(__METHOD__ . ": Unable to create thumbnail destination " .
+//					"directory, falling back to client scaling\n");
+//
+//				return this.getClientScalingThumbnailImage(image, scalerParams);
+//			}
+//
+//			// Transform functions and binaries need a FS source file
+//			thumbnailSource = this.getThumbnailSource(image, paramsVar);
+//
+//			// If the source isn't the original, disable EXIF rotation because it's already been applied
+//			if (scalerParams['srcWidth'] != thumbnailSource['width']
+//				|| scalerParams['srcHeight'] != thumbnailSource['height']) {
+//				scalerParams['disableRotation'] = true;
+//			}
+//
+//			scalerParams['srcPath'] = thumbnailSource['path'];
+//			scalerParams['srcWidth'] = thumbnailSource['width'];
+//			scalerParams['srcHeight'] = thumbnailSource['height'];
+//
+//			if (scalerParams['srcPath'] === false) { // Failed to get local copy
+//				wfDebugLog('thumbnail',
+//					sprintf('Thumbnail failed on %s: could not get local copy of "%s"',
+//						wfHostname(), image.getName()));
+//
+//				return new MediaTransformError('thumbnail_error',
+//					scalerParams['clientWidth'], scalerParams['clientHeight'],
+//					wfMessage('filemissing')
+//				);
+//			}
+//
+//			// Try a hook. Called "Bitmap" for historical reasons.
+//			/** @var mto MediaTransformOutput */
+//			mto = null;
+//			Hooks::run('BitmapHandlerTransform', [ this, image, &scalerParams, &mto ]);
+//			if (!is_null(mto)) {
+//				wfDebug(__METHOD__ . ": Hook to BitmapHandlerTransform created an mto\n");
+//				scaler = 'hookaborted';
+//			}
+//
+//			// scaler will return a MediaTransformError on failure, or false on success.
+//			// If the scaler is succesful, it will have created a thumbnail at the destination
+//			// path.
+//			if (is_array(scaler) && is_callable(scaler)) {
+//				// Allow subclasses to specify their own rendering methods.
+//				err = call_user_func(scaler, image, scalerParams);
+//			} else {
+//				switch (scaler) {
+//					case 'hookaborted':
+//						// Handled by the hook above
+//						err = mto.isError() ? mto : false;
+//						break;
+//					case 'im':
+//						err = this.transformImageMagick(image, scalerParams);
+//						break;
+//					case 'custom':
+//						err = this.transformCustom(image, scalerParams);
+//						break;
+//					case 'imext':
+//						err = this.transformImageMagickExt(image, scalerParams);
+//						break;
+//					case 'gd':
+//					default:
+//						err = this.transformGd(image, scalerParams);
+//						break;
+//				}
+//			}
+//
+//			// Remove the file if a zero-byte thumbnail was created, or if there was an error
+//			removed = this.removeBadFile(dstPath, (boolean)err);
+//			if (err) {
+//				// transform returned MediaTransforError
+//				return err;
+//			} elseif (removed) {
+//				// Thumbnail was zero-byte and had to be removed
+//				return new MediaTransformError('thumbnail_error',
+//					scalerParams['clientWidth'], scalerParams['clientHeight'],
+//					wfMessage('unknown-error')
+//				);
+//			} elseif (mto) {
+//				return mto;
+//			} else {
+//				newParams = [
+//					'width' => scalerParams['clientWidth'],
+//					'height' => scalerParams['clientHeight']
+//				];
+//				if (isset(paramsVar['quality'])) {
+//					newParams['quality'] = paramsVar['quality'];
+//				}
+//				if (isset(paramsVar['page']) && paramsVar['page']) {
+//					newParams['page'] = paramsVar['page'];
+//				}
+//				return new ThumbnailImage(image, dstUrl, dstPath, newParams);
+//			}
+//			return null;
+	}
+
+//		/**
+//		* Get the source file for the transform
+//		*
+//		* @param File file
+//		* @param array paramsVar
+//		* @return array Array with keys  width, height and path.
+//		*/
+//		protected function getThumbnailSource(file, paramsVar) {
+//			return file.getThumbnailSource(paramsVar);
+//		}
+//
+//		/**
+//		* Returns what sort of scaler type should be used.
+//		*
+//		* Values can be one of client, im, custom, gd, imext, or an array
+//		* of Object, method-name to call that specific method.
+//		*
+//		* If specifying a custom scaler command with [ Obj, method ],
+//		* the method in question should take 2 parameters, a File Object,
+//		* and a scalerParams array with various options (See doTransform
+//		* for what is in scalerParams). On error it should return a
+//		* MediaTransformError Object. On success it should return false,
+//		* and simply make sure the thumbnail file is located at
+//		* scalerParams['dstPath'].
+//		*
+//		* If there is a problem with the output path, it returns "client"
+//		* to do client side scaling.
+//		*
+//		* @param String dstPath
+//		* @param boolean checkDstPath Check that dstPath is valid
+//		* @return String|Callable One of client, im, custom, gd, imext, or a Callable array.
+//		*/
+//		abstract protected function getScalerType(dstPath, checkDstPath = true);
+
+	/**
+	* Get a ThumbnailImage that respresents an image that will be scaled
+	* client side
+	*
+	* @param File image File associated with this thumbnail
+	* @param array scalerParams Array with scaler paramsVar
+	* @return ThumbnailImage
+	*
+	* @todo FIXME: No rotation support
+	*/
+	private XomwThumbnailImage getClientScalingThumbnailImage(XomwFile image, Xomw_params_scalar scalerParams) {
+		Xomw_params_handler prms = new Xomw_params_handler();
+		prms.width = scalerParams.clientWidth;
+		prms.height = scalerParams.clientHeight;
+
+		return new XomwThumbnailImage(image, image.getUrl(), null, prms);
+	}
+
+//		/**
+//		* Transform an image using ImageMagick
+//		*
+//		* This is a stub method. The real method is in BitmapHander.
+//		*
+//		* @param File image File associated with this thumbnail
+//		* @param array paramsVar Array with scaler paramsVar
+//		*
+//		* @return MediaTransformError Error Object if error occurred, false (=no error) otherwise
+//		*/
+//		protected function transformImageMagick(image, paramsVar) {
+//			return this.getMediaTransformError(paramsVar, "Unimplemented");
+//		}
+//
+//		/**
+//		* Transform an image using the Imagick PHP extension
+//		*
+//		* This is a stub method. The real method is in BitmapHander.
+//		*
+//		* @param File image File associated with this thumbnail
+//		* @param array paramsVar Array with scaler paramsVar
+//		*
+//		* @return MediaTransformError Error Object if error occurred, false (=no error) otherwise
+//		*/
+//		protected function transformImageMagickExt(image, paramsVar) {
+//			return this.getMediaTransformError(paramsVar, "Unimplemented");
+//		}
+//
+//		/**
+//		* Transform an image using a custom command
+//		*
+//		* This is a stub method. The real method is in BitmapHander.
+//		*
+//		* @param File image File associated with this thumbnail
+//		* @param array paramsVar Array with scaler paramsVar
+//		*
+//		* @return MediaTransformError Error Object if error occurred, false (=no error) otherwise
+//		*/
+//		protected function transformCustom(image, paramsVar) {
+//			return this.getMediaTransformError(paramsVar, "Unimplemented");
+//		}
+//
+//		/**
+//		* Get a MediaTransformError with error 'thumbnail_error'
+//		*
+//		* @param array paramsVar Parameter array as passed to the transform* functions
+//		* @param String errMsg Error message
+//		* @return MediaTransformError
+//		*/
+//		public function getMediaTransformError(paramsVar, errMsg) {
+//			return new MediaTransformError('thumbnail_error', paramsVar['clientWidth'],
+//				paramsVar['clientHeight'], errMsg);
+//		}
+//
+//		/**
+//		* Transform an image using the built in GD library
+//		*
+//		* This is a stub method. The real method is in BitmapHander.
+//		*
+//		* @param File image File associated with this thumbnail
+//		* @param array paramsVar Array with scaler paramsVar
+//		*
+//		* @return MediaTransformError Error Object if error occurred, false (=no error) otherwise
+//		*/
+//		protected function transformGd(image, paramsVar) {
+//			return this.getMediaTransformError(paramsVar, "Unimplemented");
+//		}
+//
+//		/**
+//		* Escape a String for ImageMagick's property input (e.g. -set -comment)
+//		* See InterpretImageProperties() in magick/property.c
+//		* @param String s
+//		* @return String
+//		*/
+//		function escapeMagickProperty(s) {
+//			// Double the backslashes
+//			s = str_replace('\\', '\\\\', s);
+//			// Double the percents
+//			s = str_replace('%', '%%', s);
+//			// Escape initial - or @
+//			if (strlen(s) > 0 && (s[0] === '-' || s[0] === '@')) {
+//				s = '\\' . s;
+//			}
+//
+//			return s;
+//		}
+//
+//		/**
+//		* Escape a String for ImageMagick's input filenames. See ExpandFilenames()
+//		* and GetPathComponent() in magick/utility.c.
+//		*
+//		* This won't work with an initial ~ or @, so input files should be prefixed
+//		* with the directory name.
+//		*
+//		* Glob character unescaping is broken in ImageMagick before 6.6.1-5, but
+//		* it's broken in a way that doesn't involve trying to convert every file
+//		* in a directory, so we're better off escaping and waiting for the bugfix
+//		* to filter down to users.
+//		*
+//		* @param String path The file path
+//		* @param boolean|String scene The scene specification, or false if there is none
+//		* @throws MWException
+//		* @return String
+//		*/
+//		function escapeMagickInput(path, scene = false) {
+//			// Die on initial metacharacters (caller should prepend path)
+//			firstChar = substr(path, 0, 1);
+//			if (firstChar === '~' || firstChar === '@') {
+//				throw new MWException(__METHOD__ . ': cannot escape this path name');
+//			}
+//
+//			// Escape glob chars
+//			path = preg_replace('/[*?\[\]{}]/', '\\\\\0', path);
+//
+//			return this.escapeMagickPath(path, scene);
+//		}
+//
+//		/**
+//		* Escape a String for ImageMagick's output filename. See
+//		* InterpretImageFilename() in magick/image.c.
+//		* @param String path The file path
+//		* @param boolean|String scene The scene specification, or false if there is none
+//		* @return String
+//		*/
+//		function escapeMagickOutput(path, scene = false) {
+//			path = str_replace('%', '%%', path);
+//
+//			return this.escapeMagickPath(path, scene);
+//		}
+//
+//		/**
+//		* Armour a String against ImageMagick's GetPathComponent(). This is a
+//		* helper function for escapeMagickInput() and escapeMagickOutput().
+//		*
+//		* @param String path The file path
+//		* @param boolean|String scene The scene specification, or false if there is none
+//		* @throws MWException
+//		* @return String
+//		*/
+//		protected function escapeMagickPath(path, scene = false) {
+//			// Die on format specifiers (other than drive letters). The regex is
+//			// meant to match all the formats you get from "convert -list format"
+//			if (preg_match('/^([a-zA-Z0-9-]+):/', path, m)) {
+//				if (wfIsWindows() && is_dir(m[0])) {
+//					// OK, it's a drive letter
+//					// ImageMagick has a similar exception, see IsMagickConflict()
+//				} else {
+//					throw new MWException(__METHOD__ . ': unexpected colon character in path name');
+//				}
+//			}
+//
+//			// If there are square brackets, add a do-nothing scene specification
+//			// to force a literal interpretation
+//			if (scene === false) {
+//				if (strpos(path, '[') !== false) {
+//					path .= '[0--1]';
+//				}
+//			} else {
+//				path .= "[scene]";
+//			}
+//
+//			return path;
+//		}
+//
+//		/**
+//		* Retrieve the version of the installed ImageMagick
+//		* You can use PHPs version_compare() to use this value
+//		* Value is cached for one hour.
+//		* @return String|boolean Representing the IM version; false on error
+//		*/
+//		protected function getMagickVersion() {
+//			cache = MediaWikiServices::getInstance().getLocalServerObjectCache();
+//			method = __METHOD__;
+//			return cache.getWithSetCallback(
+//				'imagemagick-version',
+//				cache::TTL_HOUR,
+//				function () use (method) {
+//					global wgImageMagickConvertCommand;
+//
+//					cmd = wfEscapeShellArg(wgImageMagickConvertCommand) . ' -version';
+//					wfDebug(method . ": Running convert -version\n");
+//					retval = '';
+//					return = wfShellExec(cmd, retval);
+//					x = preg_match(
+//						'/Version: ImageMagick ([0-9]*\.[0-9]*\.[0-9]*)/', return, matches
+//					);
+//					if (x != 1) {
+//						wfDebug(method . ": ImageMagick version check failed\n");
+//						return false;
+//					}
+//
+//					return matches[1];
+//				}
+//			);
+//		}
+//
+//		/**
+//		* Returns whether the current scaler supports rotation.
+//		*
+//		* @since 1.24 No longer static
+//		* @return boolean
+//		*/
+//		public function canRotate() {
+//			return false;
+//		}
+//
+//		/**
+//		* Should we automatically rotate an image based on exif
+//		*
+//		* @since 1.24 No longer static
+//		* @see wgEnableAutoRotation
+//		* @return boolean Whether auto rotation is enabled
+//		*/
+//		public function autoRotateEnabled() {
+//			return false;
+//		}
+//
+//		/**
+//		* Rotate a thumbnail.
+//		*
+//		* This is a stub. See BitmapHandler::rotate.
+//		*
+//		* @param File file
+//		* @param array paramsVar Rotate parameters.
+//		*   'rotation' clockwise rotation in degrees, allowed are multiples of 90
+//		* @since 1.24 Is non-static. From 1.21 it was static
+//		* @return boolean|MediaTransformError
+//		*/
+//		public function rotate(file, paramsVar) {
+//			return new MediaTransformError('thumbnail_error', 0, 0,
+//				get_class(this) . ' rotation not implemented');
+//		}
+//
+//		/**
+//		* Returns whether the file needs to be rendered. Returns true if the
+//		* file requires rotation and we are able to rotate it.
+//		*
+//		* @param File file
+//		* @return boolean
+//		*/
+//		public function mustRender(file) {
+//			return this.canRotate() && this.getRotation(file) != 0;
+//		}
+//
+//		/**
+//		* Check if the file is smaller than the maximum image area for thumbnailing.
+//		*
+//		* Runs the 'BitmapHandlerCheckImageArea' hook.
+//		*
+//		* @param File file
+//		* @param array paramsVar
+//		* @return boolean
+//		* @since 1.25
+//		*/
+//		public function isImageAreaOkForThumbnaling(file, &paramsVar) {
+//			global wgMaxImageArea;
+//
+//			// For historical reasons, hook starts with BitmapHandler
+//			checkImageAreaHookResult = null;
+//			Hooks::run(
+//				'BitmapHandlerCheckImageArea',
+//				[ file, &paramsVar, &checkImageAreaHookResult ]
+//			);
+//
+//			if (!is_null(checkImageAreaHookResult)) {
+//				// was set by hook, so return that value
+//				return (boolean)checkImageAreaHookResult;
+//			}
+//
+//			srcWidth = file.getWidth(paramsVar['page']);
+//			srcHeight = file.getHeight(paramsVar['page']);
+//
+//			if (srcWidth * srcHeight > wgMaxImageArea
+//				&& !(file.getMimeType() == 'image/jpeg'
+//					&& this.getScalerType(false, false) == 'im')
+//			) {
+//				// Only ImageMagick can efficiently downsize jpg images without loading
+//				// the entire file in memory
+//				return false;
+//			}
+//			return true;
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwBlockLevelPass.java

@@ -13,3 +13,652 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+import gplx.langs.htmls.*;
+/**
+* This is the part of the wikitext parser which handles automatic paragraphs
+* and conversion of start-of-line prefixes to HTML lists.
+*/
+public class XomwBlockLevelPass {
+	private boolean DTopen = false;
+	private boolean inPre = false;
+	private int lastSection = LAST_SECTION_NONE;
+	private boolean linestart;
+//		private $text;
+	private final    Bry_bfr tmp = Bry_bfr_.New();
+	private final    Btrie_rv trv = new Btrie_rv();
+	private byte[] find_colon_no_links__before, find_colon_no_links__after;
+
+	// State constants for the definition list colon extraction
+	private static final int
+	  COLON_STATE_TEXT = 0
+	, COLON_STATE_TAG = 1
+	, COLON_STATE_TAGSTART = 2
+	, COLON_STATE_CLOSETAG = 3
+	, COLON_STATE_TAGSLASH = 4
+	, COLON_STATE_COMMENT = 5
+	, COLON_STATE_COMMENTDASH = 6
+	, COLON_STATE_COMMENTDASHDASH = 7
+	;
+
+	/**
+	* Make lists from lines starting with ':', '*', '#', etc.
+	*
+	* @param String $text
+	* @param boolean $linestart Whether or not this is at the start of a line.
+	* @return String The lists rendered as HTML
+	*/
+//		public static function doBlockLevels($text, $linestart) {
+//			$pass = new self($text, $linestart);
+//			return $pass->execute();
+//		}
+	public void doBlockLevels(XomwParserCtx pctx, XomwParserBfr pbfr, boolean linestart) {
+		this.linestart = linestart;
+		execute(pctx, pbfr, linestart);
+	}
+
+//		/**
+//		* Private constructor
+//		*/
+//		private function __construct($text, $linestart) {
+//			$this->text = $text;
+//			$this->linestart = $linestart;
+//		}
+
+	/**
+	* If a pre or p is open, return the corresponding close tag and update
+	* the state. If no tag is open, return an empty String.
+	* @return String
+	*/
+	private byte[] closeParagraph() {
+		byte[] result = Bry_.Empty;
+		if (this.lastSection != LAST_SECTION_NONE) {
+			result = tmp.Add(lastSection == LAST_SECTION_PARA ? Gfh_tag_.P_rhs : Gfh_tag_.Pre_rhs).Add_byte_nl().To_bry_and_clear(); // $result = '</' . $this->lastSection . ">\n";
+		}
+		this.inPre = false;
+		this.lastSection = LAST_SECTION_NONE;
+		return result;
+	}
+
+	/**
+	* getCommon() returns the length of the longest common substring
+	* of both arguments, starting at the beginning of both.
+	*
+	* @param String $st1
+	* @param String $st2
+	*
+	* @return int
+	*/
+	// getCommon() returns the length of the longest common substring
+	// of both arguments, starting at the beginning of both.
+	private int getCommon(byte[] st1, byte[] st2) {
+		int st1Len = st1.length, st2Len = st2.length;
+		int shorter = st1Len < st2Len ? st1Len : st2Len;
+
+		int i;
+		for (i = 0; i < shorter; ++i) {
+			if (st1[i] != st2[i]) {
+				break;
+			}
+		}
+		return i;
+	}
+
+	/**
+	* Open the list item element identified by the prefix character.
+	*
+	* @param String $char
+	*
+	* @return String
+	*/
+	private byte[] openList(byte c) {
+		byte[] result = this.closeParagraph();
+
+		if      (c == Byte_ascii.Star)
+			result = Bry_.Add(result, Bry_.new_a7("<ul><li>"));
+		else if (c == Byte_ascii.Hash)
+			result = Bry_.Add(result, Bry_.new_a7("<ol><li>"));
+		else if (c == Byte_ascii.Colon)
+			result = Bry_.Add(result, Bry_.new_a7("<dl><dd>"));
+		else if (c == Byte_ascii.Semic) {
+			result = Bry_.Add(result, Bry_.new_a7("<dl><dt>"));
+			this.DTopen = true;
+		}
+		else {
+			result = Bry_.new_a7("<!-- ERR 1 -->");
+		}
+
+		return result;
+	}
+
+	/**
+	* Close the current list item and open the next one.
+	* @param String $char
+	*
+	* @return String
+	*/
+	private byte[] nextItem(byte c) {
+		if (c == Byte_ascii.Star || c == Byte_ascii.Hash) {
+			return Bry_.new_a7("</li>\n<li>");
+		}
+		else if (c == Byte_ascii.Colon || c == Byte_ascii.Semic) {
+			byte[] close = Bry_.new_a7("</dd>\n");
+			if (this.DTopen) {
+				close = Bry_.new_a7("</dt>\n");
+			}
+			if (c == Byte_ascii.Semic) {
+				this.DTopen = true;
+				return Bry_.Add(close, Bry_.new_a7("<dt>"));
+			}
+			else {
+				this.DTopen = false;
+				return Bry_.Add(close, Bry_.new_a7("<dd>"));
+			}
+		}
+		return Bry_.new_a7("<!-- ERR 2 -->");
+	}
+
+	/**
+	* Close the current list item identified by the prefix character.
+	* @param String $char
+	*
+	* @return String
+	*/
+	private byte[] closeList(byte c) {
+		byte[] text = null;
+		if (c == Byte_ascii.Star) {
+			text = Bry_.new_a7("</li></ul>");
+		}
+		else if (c == Byte_ascii.Hash) {
+			text = Bry_.new_a7("</li></ol>");
+		}
+		else if (c == Byte_ascii.Colon) {
+			if (this.DTopen) {
+				this.DTopen = false;
+				text = Bry_.new_a7("</dt></dl>");
+			}
+			else {
+				text = Bry_.new_a7("</dd></dl>");
+			}
+		}
+		else {
+			return Bry_.new_a7("<!-- ERR 3 -->");
+		}
+		return text;
+	}
+
+	/**
+	* Execute the pass.
+	* @return String
+	*/
+	public void execute(XomwParserCtx pctx, XomwParserBfr pbfr, boolean linestart) {
+		// XO.PBFR
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src = src_bfr.Bfr();
+		int src_bgn = 0;
+		int src_end = src_bfr.Len();
+		Bry_bfr bfr = pbfr.Trg();
+		pbfr.Switch();
+
+		// XO.STATIC
+		if (block_chars_ary == null) {
+			synchronized (Type_.Type_by_obj(this)) {
+				block_chars_ary = Block_chars_ary__new();
+				openMatchTrie = Btrie_slim_mgr.ci_a7().Add_many_str
+				( "<table", "<h1", "<h2", "<h3", "<h4", "<h5", "<h6", "<pre", "<tr"
+				, "<p", "<ul", "<ol", "<dl", "<li", "</tr", "</td", "</th");
+				closeMatchTrie = Btrie_slim_mgr.ci_a7().Add_many_str
+				( "</table", "</h1", "</h2", "</h3", "</h4", "</h5", "</h6"
+				, "<td", "<th", "<blockquote", "</blockquote", "<div", "</div", "<hr", "</pre", "</p", "</mw:"
+				, XomwParser.MARKER_PREFIX_STR + "-pre"
+				, "</li", "</ul", "</ol", "</dl", "<center", "</center");
+				blockquoteTrie = Btrie_slim_mgr.ci_a7().Add_many_str("<blockquote", "</blockquote");
+				pre_trie = Btrie_slim_mgr.ci_a7().Add_str_int("<pre", PRE_BGN).Add_str_int("</pre", PRE_END);
+			}
+		}
+
+		// clear state
+		this.inPre = false;
+		this.lastSection = LAST_SECTION_NONE;
+		byte[] prefix2 = null;
+		bfr.Clear();
+
+		// Parsing through the text line by line.  The main thing
+		// happening here is handling of block-level elements p, pre,
+		// and making lists from lines starting with * # : etc.
+		byte[] lastPrefix = Bry_.Empty;
+		this.DTopen = false;
+		boolean inBlockElem = false;
+		int prefixLen = 0;
+		byte pendingPTag = PARA_STACK_NONE;
+		boolean inBlockquote = false;
+
+		// PORTED.SPLIT: $textLines = StringUtils::explode("\n", $text);
+		int lineBgn = src_bgn;
+		while (lineBgn < src_end) {
+			int lineEnd = Bry_find_.Find_fwd(src, Byte_ascii.Nl, lineBgn);
+			if (lineEnd == Bry_find_.Not_found)
+				lineEnd = src_end;
+
+			// Fix up linestart
+			if (!this.linestart) {
+				bfr.Add_mid(src, lineBgn, lineEnd);
+				this.linestart = true;
+				continue;
+			}
+			// * = ul
+			// # = ol
+			// ; = dt
+			// : = dd
+			int lastPrefixLen = lastPrefix.length;
+
+			// PORTED.BGN: preCloseMatch = preg_match('/<\\/pre/i', $oLine); preOpenMatch = preg_match('/<pre/i', $oLine);
+			int preCur = lineBgn;
+			boolean preCloseMatch = false;
+			boolean preOpenMatch = false;
+			while (true) {
+				if (preCur >= lineEnd)
+					break;
+				Object o = pre_trie.Match_at(trv, src, preCur, lineEnd);
+				if (o == null)
+					preCur++;
+				else {
+					int pre_tid = Int_.Cast(o);
+					if (pre_tid == PRE_BGN)
+						preOpenMatch = true;
+					else if (pre_tid == PRE_END)
+						preCloseMatch = true;
+					preCur = trv.Pos();
+				}
+			}
+			// PORTED.END
+
+			byte[] prefix = null, t = null;
+			// If not in a <pre> element, scan for and figure out what prefixes are there.
+			if (!this.inPre) {
+				// Multiple prefixes may abut each other for nested lists.
+				prefixLen = XophpString.strspn_fwd__ary(src, block_chars_ary, lineBgn, lineEnd, lineEnd); // strspn($oLine, '*#:;');
+				prefix = XophpString.substr(src, lineBgn, prefixLen);
+
+				// eh?
+				// ; and : are both from definition-lists, so they're equivalent
+				//  for the purposes of determining whether or not we need to open/close
+				//  elements.
+				// substr($inputLine, $prefixLength);
+				prefix2 = Bry_.Replace(prefix, Byte_ascii.Semic, Byte_ascii.Colon);
+				t = Bry_.Mid(src, lineBgn + prefixLen, lineEnd);
+				this.inPre = preOpenMatch;
+			}
+			else {
+				// Don't interpret any other prefixes in preformatted text
+				prefixLen = 0;
+				prefix = prefix2 = Bry_.Empty;
+				t = Bry_.Mid(src, lineBgn, lineEnd);
+			}
+
+			// List generation
+			byte[] term = null, t2 = null;
+			int commonPrefixLen = -1;
+			if (prefixLen > 0 && Bry_.Eq(lastPrefix, prefix2)) {
+				// Same as the last item, so no need to deal with nesting or opening stuff
+				bfr.Add(this.nextItem(XophpString.substr_byte(prefix, -1)));
+				pendingPTag = PARA_STACK_NONE;
+
+				if (prefixLen > 0 && prefix[prefixLen - 1] == Byte_ascii.Semic) {
+					// The one nasty exception: definition lists work like this:
+					// ; title : definition text
+					// So we check for : in the remainder text to split up the
+					// title and definition, without b0rking links.
+					term = t2 = Bry_.Empty;
+					if (this.findColonNoLinks(t, term, t2) != Bry_find_.Not_found) {
+						term = find_colon_no_links__before;
+						t2   = find_colon_no_links__after;
+						t = t2;
+						bfr.Add(term).Add(nextItem(Byte_ascii.Colon));
+					}
+				}
+			}
+			else if (prefixLen > 0 || lastPrefixLen > 0) {
+				// We need to open or close prefixes, or both.
+
+				// Either open or close a level...
+				commonPrefixLen = this.getCommon(prefix, lastPrefix);
+				pendingPTag = PARA_STACK_NONE;
+
+				// Close all the prefixes which aren't shared.
+				while (commonPrefixLen < lastPrefixLen) {
+					bfr.Add(this.closeList(lastPrefix[lastPrefixLen - 1]));
+					--lastPrefixLen;
+				}
+
+				// Continue the current prefix if appropriate.
+				if (prefixLen <= commonPrefixLen && commonPrefixLen > 0) {
+					bfr.Add(this.nextItem(prefix[commonPrefixLen - 1]));
+				}
+
+				// Open prefixes where appropriate.
+				if (Bry_.Len_gt_0(lastPrefix) && prefixLen > commonPrefixLen) {
+					bfr.Add_byte_nl();
+				}
+				while (prefixLen > commonPrefixLen) {
+					byte c = XophpString.substr_byte(prefix, commonPrefixLen, 1);
+					bfr.Add(this.openList(c));
+
+					if (c == Byte_ascii.Semic) {
+						// @todo FIXME: This is dupe of code above
+						if (findColonNoLinks(t, term, t2) != Bry_find_.Not_found) {
+							term = find_colon_no_links__before;
+							t2   = find_colon_no_links__after;
+							t = t2;
+							bfr.Add(term).Add(nextItem(Byte_ascii.Colon));
+						}
+					}
+					++commonPrefixLen;
+				}
+				if (prefixLen == 0 && Bry_.Len_gt_0(lastPrefix)) {
+					bfr.Add_byte_nl();
+				}
+				lastPrefix = prefix2;
+			}
+
+			// If we have no prefixes, go to paragraph mode.
+			if (0 == prefixLen) {
+				// No prefix (not in list)--go to paragraph mode
+				// @todo consider using a stack for nestable elements like span, table and div
+				int tLen = t.length;
+
+				// XO.MW.PORTED.BGN:
+				boolean openMatch = XophpPreg.match(openMatchTrie, trv, t, 0, tLen) != null;
+				boolean closeMatch = XophpPreg.match(closeMatchTrie, trv, t, 0, tLen) != null;
+				// XO.MW.PORTED.END
+				if (openMatch || closeMatch) {
+					pendingPTag = PARA_STACK_NONE;
+					// @todo bug 5718: paragraph closed
+					bfr.Add(this.closeParagraph());
+					if (preOpenMatch && !preCloseMatch) {
+						this.inPre = true;
+					}
+					int bqOffset = 0;
+					// PORTED:preg_match('/<(\\/?)blockquote[\s>]/i', t, $bqMatch, PREG_OFFSET_CAPTURE, $bqOffset)
+					while (true) {
+						Object o = XophpPreg.match(blockquoteTrie, trv, t, bqOffset, tLen);
+						if (o == null) { // no more blockquotes found; exit
+							break;
+						}
+						else {
+							byte[] bq_bry = (byte[])o;
+							inBlockquote = bq_bry[1] != Byte_ascii.Slash; // is this a close tag?
+							bqOffset = trv.Pos();
+						}
+					}
+					// PORTED:END
+					inBlockElem = !closeMatch;
+				}
+				else if (!inBlockElem && !this.inPre) {
+					if (XophpString.substr_byte(t, 0) == Byte_ascii.Space
+						&& (this.lastSection == LAST_SECTION_PRE || Bry_.Trim(t) != Bry_.Empty)
+						&& !inBlockquote
+					) {
+						// pre
+						if (this.lastSection != LAST_SECTION_PRE) {
+							pendingPTag = PARA_STACK_NONE;
+							bfr.Add(closeParagraph()).Add(Gfh_tag_.Pre_lhs);
+							this.lastSection = LAST_SECTION_PRE;
+						}
+						t = Bry_.Mid(t, 1);
+					}
+					else {
+						// paragraph
+						if (Bry_.Trim(t) == Bry_.Empty) {
+							if (pendingPTag != PARA_STACK_NONE) {
+								ParaStackAdd(bfr, pendingPTag);
+								bfr.Add_str_a7("<br />");
+								pendingPTag = PARA_STACK_NONE;
+								this.lastSection = LAST_SECTION_PARA;
+							}
+							else {
+								if (this.lastSection != LAST_SECTION_PARA) {
+									bfr.Add(this.closeParagraph());
+									this.lastSection = LAST_SECTION_NONE;
+									pendingPTag = PARA_STACK_BGN;
+								}
+								else {
+									pendingPTag = PARA_STACK_MID;
+								}
+							}
+						}
+						else {
+							if (pendingPTag != PARA_STACK_NONE) {
+								ParaStackAdd(bfr, pendingPTag);
+								pendingPTag = PARA_STACK_NONE;
+								this.lastSection = LAST_SECTION_PARA;
+							}
+							else if (lastSection != LAST_SECTION_PARA) {
+								bfr.Add(this.closeParagraph()).Add(Gfh_tag_.P_lhs);
+								this.lastSection = LAST_SECTION_PARA;
+							}
+						}
+					}
+				}
+			}
+			// somewhere above we forget to get out of pre block (bug 785)
+			if (preCloseMatch && this.inPre) {
+				this.inPre = false;
+			}
+			if (pendingPTag == PARA_STACK_NONE) {
+				bfr.Add(t);
+				if (prefixLen == 0) {
+					bfr.Add_byte_nl();
+				}
+			}
+
+			lineBgn = lineEnd + 1;
+		}
+
+		while (prefixLen > 0) {
+			bfr.Add(this.closeList(prefix2[prefixLen - 1]));
+			--prefixLen;
+			if (prefixLen > 0) {
+				bfr.Add_byte_nl();
+			}
+		}
+		if (this.lastSection != LAST_SECTION_NONE) {
+			bfr.Add(this.lastSection == LAST_SECTION_PARA ? Gfh_tag_.P_rhs : Gfh_tag_.Pre_rhs);
+			this.lastSection = LAST_SECTION_NONE;
+		}
+	}
+
+	/**
+	* Split up a String on ':', ignoring any occurrences inside tags
+	* to prevent illegal overlapping.
+	*
+	* @param String $str The String to split
+	* @param String &$before Set to everything before the ':'
+	* @param String &$after Set to everything after the ':'
+	* @throws MWException
+	* @return String The position of the ':', or false if none found
+	*/
+	private int findColonNoLinks(byte[] str, byte[] before, byte[] after) {
+		int len = str.length;
+		int colonPos = XophpString.strpos(str, Byte_ascii.Colon, 0, len);
+		if (colonPos == Bry_find_.Not_found) {
+			// Nothing to find!
+			return Bry_find_.Not_found;
+		}
+
+		int ltPos = XophpString.strpos(str, Byte_ascii.Angle_bgn, 0, len);
+		if (ltPos == Bry_find_.Not_found || ltPos > colonPos) {
+			// Easy; no tag nesting to worry about
+			// XOMW: MW passes before / after by reference; XO: changes member and depends on callers to update
+			find_colon_no_links__before = XophpString.substr(str, 0, colonPos);
+			find_colon_no_links__after = XophpString.substr(str, colonPos + 1);
+			return colonPos;
+		}
+
+		// Ugly state machine to walk through avoiding tags.
+		int state = COLON_STATE_TEXT;
+		int level = 0;
+		for (int i = 0; i < len; i++) {
+			byte c = str[i];
+
+			switch (state) {
+				case COLON_STATE_TEXT:
+					switch (c) {
+						case Byte_ascii.Angle_bgn:
+							// Could be either a <start> tag or an </end> tag
+							state = COLON_STATE_TAGSTART;
+							break;
+						case Byte_ascii.Colon:
+							if (level == 0) {
+								// We found it!
+								find_colon_no_links__before = XophpString.substr(str, 0, i);
+								find_colon_no_links__after = XophpString.substr(str, i + 1);
+								return i;
+							}
+							// Embedded in a tag; don't break it.
+							break;
+						default:
+							// Skip ahead looking for something interesting
+							colonPos = XophpString.strpos(str, Byte_ascii.Colon, i, len);
+							if (colonPos == Bry_find_.Not_found) {
+								// Nothing else interesting
+								return Bry_find_.Not_found;
+							}
+							ltPos = XophpString.strpos(str, Byte_ascii.Angle_bgn, i, len);
+							if (level == 0) {
+								if (ltPos == Bry_find_.Not_found || colonPos < ltPos) {
+									// We found it!
+									find_colon_no_links__before = XophpString.substr(str, 0, colonPos);
+									find_colon_no_links__after = XophpString.substr(str, colonPos + 1);
+									return i;
+								}
+							}
+							if (ltPos == Bry_find_.Not_found) {
+								// Nothing else interesting to find; abort!
+								// We're nested, but there's no close tags left. Abort!
+								i = len;	// break 2
+								break;
+							}
+							// Skip ahead to next tag start
+							i = ltPos;
+							state = COLON_STATE_TAGSTART;
+							break;
+						}
+					break;
+				case COLON_STATE_TAG:
+					// In a <tag>
+					switch (c) {
+						case Byte_ascii.Angle_end:
+							level++;
+							state = COLON_STATE_TEXT;
+							break;
+						case Byte_ascii.Slash:
+							// Slash may be followed by >?
+							state = COLON_STATE_TAGSLASH;
+							break;
+						default:
+							// ignore
+							break;
+					}
+					break;
+				case COLON_STATE_TAGSTART:
+					switch (c) {
+						case Byte_ascii.Slash:
+							state = COLON_STATE_CLOSETAG;
+							break;
+						case Byte_ascii.Bang:
+							state = COLON_STATE_COMMENT;
+							break;
+						case Byte_ascii.Angle_end:
+							// Illegal early close? This shouldn't happen D:
+							state = COLON_STATE_TEXT;
+							break;
+						default:
+							state = COLON_STATE_TAG;
+							break;
+					}
+					break;
+				case COLON_STATE_CLOSETAG:
+					// In a </tag>
+					if (c == Byte_ascii.Angle_end) {
+						level--;
+						if (level < 0) {
+							Gfo_usr_dlg_.Instance.Warn_many("", "", "Invalid input; too many close tags");
+							return Bry_find_.Not_found;
+						}
+						state = COLON_STATE_TEXT;
+					}
+					break;
+				case COLON_STATE_TAGSLASH:
+					if (c == Byte_ascii.Angle_end) {
+						// Yes, a self-closed tag <blah/>
+						state = COLON_STATE_TEXT;
+					}
+					else {
+						// Probably we're jumping the gun, and this is an attribute
+						state = COLON_STATE_TAG;
+					}
+					break;
+				case COLON_STATE_COMMENT:
+					if (c == Byte_ascii.Dash) {
+						state = COLON_STATE_COMMENTDASH;
+					}
+					break;
+				case COLON_STATE_COMMENTDASH:
+					if (c == Byte_ascii.Dash) {
+						state = COLON_STATE_COMMENTDASHDASH;
+					}
+					else {
+						state = COLON_STATE_COMMENT;
+					}
+					break;
+				case COLON_STATE_COMMENTDASHDASH:
+					if (c == Byte_ascii.Angle_bgn) {
+						state = COLON_STATE_TEXT;
+					}
+					else {
+						state = COLON_STATE_COMMENT;
+					}
+					break;
+			default:
+				throw Err_.new_wo_type("State machine error");
+			}
+		}
+		if (level > 0) {
+			Gfo_usr_dlg_.Instance.Warn_many("", "", "Invalid input; not enough close tags (level ~{0}, state ~{1})", level, state);
+			return Bry_find_.Not_found;
+		}
+		return Bry_find_.Not_found;
+	}
+
+	private static final byte
+	  LAST_SECTION_NONE = 0    // ''
+	, LAST_SECTION_PARA = 1    // p
+	, LAST_SECTION_PRE  = 2    // pre
+	;
+	private static final byte
+	  PARA_STACK_NONE = 0	// false
+	, PARA_STACK_BGN  = 1	// <p>
+	, PARA_STACK_MID  = 2	// </p><p>
+	;
+	private static final int PRE_BGN = 0, PRE_END = 1;
+	private static Btrie_slim_mgr pre_trie;
+	private static boolean[] block_chars_ary; 
+	private static boolean[] Block_chars_ary__new() {
+		boolean[] rv = new boolean[256];
+		rv[Byte_ascii.Star] = true;
+		rv[Byte_ascii.Hash] = true;
+		rv[Byte_ascii.Colon] = true;
+		rv[Byte_ascii.Semic] = true;
+		return rv;
+	}
+	private static Btrie_slim_mgr openMatchTrie, closeMatchTrie, blockquoteTrie;
+	private static void ParaStackAdd(Bry_bfr bfr, int id) {
+		switch (id) {
+			case PARA_STACK_BGN: bfr.Add_str_a7("<p>"); break;
+			case PARA_STACK_MID: bfr.Add_str_a7("</p><p>"); break;
+			default:              throw Err_.new_unhandled_default(id);
+		}
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwBlockLevelPassTest.java

@@ -13,3 +13,28 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+import gplx.xowa.mediawiki.includes.linkers.*;
+public class XomwBlockLevelPassTest {
+	private final    XomwBlockLevelPassFxt fxt = new XomwBlockLevelPassFxt();
+	@Test   public void Basic() {
+		fxt.Test__do_block_levels(String_.Concat_lines_nl_skip_last
+		( "a"
+		), String_.Concat_lines_nl_skip_last
+		( "<p>a"
+		, "</p>"
+		));
+	}
+}
+class XomwBlockLevelPassFxt {
+	private final    XomwBlockLevelPass block_level_pass = new XomwBlockLevelPass();
+	private final    XomwParserCtx pctx = new XomwParserCtx();
+	private final    XomwParserBfr pbfr = new XomwParserBfr();
+	private boolean apos = true;
+	public void Test__do_block_levels(String src, String expd) {
+		if (apos) expd = gplx.langs.htmls.Gfh_utl.Replace_apos(expd);
+		block_level_pass.doBlockLevels(pctx, pbfr.Init(Bry_.new_u8(src)), true);
+		Gftest.Eq__str(expd, pbfr.Rslt().To_str_and_clear());
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwLinkHolderArray.java

@@ -13,3 +13,767 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.langs.htmls.*;
+import gplx.xowa.mediawiki.includes.xohtml.*;
+import gplx.xowa.mediawiki.includes.linkers.*;
+/**
+* Holder of replacement pairs for wiki links
+*/
+public class XomwLinkHolderArray {
+	private final    XomwLinkHolderList internals = new XomwLinkHolderList();
+//		public $interwikis = [];
+	// private int size = 0;
+
+	private final    Bry_bfr tmp = Bry_bfr_.New();
+	private final    Xomw_atr_mgr extraAtrs = new Xomw_atr_mgr();
+	private final    Xomw_qry_mgr query = new Xomw_qry_mgr();
+	/**
+	* @var Parser
+	*/
+	private final    XomwParserIface parent;
+//		protected $tempIdOffset;
+
+	/**
+	* @param Parser $parent
+	*/
+	public XomwLinkHolderArray(XomwParserIface parent) {
+		this.parent = parent;
+	}
+
+	/**
+	* Reduce memory usage to reduce the impact of circular references
+	*/
+	//	public function __destruct() {
+	//		foreach ( $this as $name => $value ) {
+	//			unset( this.$name );
+	//		}
+	//	}
+
+	/**
+	* Don't serialize the parent Object, it is big, and not needed when it is
+	* a parameter to mergeForeign(), which is the only application of
+	* serializing at present.
+	*
+	* Compact the titles, only serialize the text form.
+	* @return array
+	*/
+	//	public function __sleep() {
+	//		foreach ( this.internals as &$nsLinks ) {
+	//			foreach ( $nsLinks as &$entry ) {
+	//				unset( $entry['title'] );
+	//			}
+	//		}
+	//		unset( $nsLinks );
+	//		unset( $entry );
+	//
+	//		foreach ( this.interwikis as &$entry ) {
+	//			unset( $entry['title'] );
+	//		}
+	//		unset( $entry );
+	//
+	//		return [ 'internals', 'interwikis', 'size' ];
+	//	}
+
+	/**
+	* Recreate the Title objects
+	*/
+	//	public function __wakeup() {
+	//		foreach ( this.internals as &$nsLinks ) {
+	//			foreach ( $nsLinks as &$entry ) {
+	//				$entry['title'] = Title::newFromText( $entry['pdbk'] );
+	//			}
+	//		}
+	//		unset( $nsLinks );
+	//		unset( $entry );
+	//
+	//		foreach ( this.interwikis as &$entry ) {
+	//			$entry['title'] = Title::newFromText( $entry['pdbk'] );
+	//		}
+	//		unset( $entry );
+	//	}
+
+//		/**
+//		* Merge another LinkHolderArray into this one
+//		* @param LinkHolderArray $other
+//		*/
+//		public function merge( $other ) {
+//			foreach ( $other->internals as $ns => $entries ) {
+//				this.size += count( $entries );
+//				if ( !isset( this.internals[$ns] ) ) {
+//					this.internals[$ns] = $entries;
+//				} else {
+//					this.internals[$ns] += $entries;
+//				}
+//			}
+//			this.interwikis += $other->interwikis;
+//		}
+//
+//		/**
+//		* Merge a LinkHolderArray from another parser instance into this one. The
+//		* keys will not be preserved. Any text which went with the old
+//		* LinkHolderArray and needs to work with the new one should be passed in
+//		* the $texts array. The strings in this array will have their link holders
+//		* converted for use in the destination link holder. The resulting array of
+//		* strings will be returned.
+//		*
+//		* @param LinkHolderArray $other
+//		* @param array $texts Array of strings
+//		* @return array
+//		*/
+//		public function mergeForeign( $other, $texts ) {
+//			this.tempIdOffset = $idOffset = this.parent->nextLinkID();
+//			$maxId = 0;
+//
+//			# Renumber @gplx.Internal protected links
+//			foreach ( $other->internals as $ns => $nsLinks ) {
+//				foreach ( $nsLinks as $key => $entry ) {
+//					$newKey = $idOffset + $key;
+//					this.internals[$ns][$newKey] = $entry;
+//					$maxId = $newKey > $maxId ? $newKey : $maxId;
+//				}
+//			}
+//			$texts = preg_replace_callback( '/(<!--LINK \d+:)(\d+)(-->)/',
+//				[ $this, 'mergeForeignCallback' ], $texts );
+//
+//			# Renumber interwiki links
+//			foreach ( $other->interwikis as $key => $entry ) {
+//				$newKey = $idOffset + $key;
+//				this.interwikis[$newKey] = $entry;
+//				$maxId = $newKey > $maxId ? $newKey : $maxId;
+//			}
+//			$texts = preg_replace_callback( '/(<!--IWLINK )(\d+)(-->)/',
+//				[ $this, 'mergeForeignCallback' ], $texts );
+//
+//			# Set the parent link ID to be beyond the highest used ID
+//			this.parent->setLinkID( $maxId + 1 );
+//			this.tempIdOffset = null;
+//			return $texts;
+//		}
+//
+//		/**
+//		* @param array $m
+//		* @return String
+//		*/
+//		protected function mergeForeignCallback( $m ) {
+//			return $m[1] . ( $m[2] + this.tempIdOffset ) . $m[3];
+//		}
+//
+//		/**
+//		* Get a subset of the current LinkHolderArray which is sufficient to
+//		* interpret the given text.
+//		* @param String $text
+//		* @return LinkHolderArray
+//		*/
+//		public function getSubArray( $text ) {
+//			$sub = new LinkHolderArray( this.parent );
+//
+//			# Internal links
+//			$pos = 0;
+//			while ( $pos < strlen( $text ) ) {
+//				if ( !preg_match( '/<!--LINK (\d+):(\d+)-->/',
+//					$text, $m, PREG_OFFSET_CAPTURE, $pos )
+//				) {
+//					break;
+//				}
+//				$ns = $m[1][0];
+//				$key = $m[2][0];
+//				$sub->internals[$ns][$key] = this.internals[$ns][$key];
+//				$pos = $m[0][1] + strlen( $m[0][0] );
+//			}
+//
+//			# Interwiki links
+//			$pos = 0;
+//			while ( $pos < strlen( $text ) ) {
+//				if ( !preg_match( '/<!--IWLINK (\d+)-->/', $text, $m, PREG_OFFSET_CAPTURE, $pos ) ) {
+//					break;
+//				}
+//				$key = $m[1][0];
+//				$sub->interwikis[$key] = this.interwikis[$key];
+//				$pos = $m[0][1] + strlen( $m[0][0] );
+//			}
+//			return $sub;
+//		}
+//
+//		/**
+//		* Returns true if the memory requirements of this Object are getting large
+//		* @return boolean
+//		*/
+//		public function isBig() {
+//			global $wgLinkHolderBatchSize;
+//			return this.size > $wgLinkHolderBatchSize;
+//		}
+
+	/**
+	* Clear all stored link holders.
+	* Make sure you don't have any text left using these link holders, before you call this
+	*/
+	public void clear() {
+		this.internals.Clear();//
+//			this.interwikis = [];
+		// this.size = 0;
+	}
+
+	/**
+	* Make a link placeholder. The text returned can be later resolved to a real link with
+	* replaceLinkHolders(). This is done for two reasons: firstly to avoid further
+	* parsing of interwiki links, and secondly to allow all existence checks and
+	* article length checks (for stub links) to be bundled into a single query.
+	*
+	* @param Title $nt
+	* @param String $text
+	* @param array $query [optional]
+	* @param String $trail [optional]
+	* @param String $prefix [optional]
+	* @return String
+	*/
+	public void makeHolder(Bry_bfr bfr, XomwTitle nt, byte[] text, byte[][] query, byte[] trail, byte[] prefix) {
+		if (nt == null) {
+			// Fail gracefully
+			bfr.Add_str_a7("<!-- ERROR -->").Add(prefix).Add(text).Add(trail);
+		}
+		else {
+			// Separate the link trail from the rest of the link
+//				list( $inside, $trail ) = Linker::splitTrail( $trail );
+			byte[] inside = Bry_.Empty;
+
+			XomwLinkHolderItem entry = new XomwLinkHolderItem
+				( nt
+				, tmp.Add_bry_many(prefix, text, inside).To_bry_and_clear()
+				, query);
+
+			if (nt.isExternal()) {
+				// Use a globally unique ID to keep the objects mergable
+//					$key = this.parent->nextLinkID();
+//					this.interwikis[$key] = $entry;
+//					$retVal = "<!--IWLINK $key-->{$trail}";
+			}
+			else {
+				int key = this.parent.nextLinkID();
+				this.internals.Add(key, entry);
+				bfr.Add(Bry__link__bgn).Add_int_variable(key).Add(Gfh_tag_.Comm_end).Add(trail);	// "<!--LINK $ns:$key-->{$trail}";
+			}
+		}
+	}
+
+	/**
+	* Replace <!--LINK--> link placeholders with actual links, in the buffer
+	*
+	* @param String $text
+	*/
+	public boolean replace(XomwParserBfr pbfr) {
+		return this.replaceInternal(pbfr);
+//			$this->replaceInterwiki( $text );
+	}
+	public byte[] replace(XomwParserBfr pbfr, byte[] text) {
+		boolean rv = this.replace(pbfr.Init(text));
+		return rv ? pbfr.Trg().To_bry_and_clear() : pbfr.Src().To_bry_and_clear();
+	}
+
+	/**
+	* Replace @gplx.Internal protected links
+	* @param String $text
+	*/
+	private boolean replaceInternal(XomwParserBfr pbfr) {
+		if (internals.Len() == 0) {
+			return false;
+		}
+
+		// SKIP:Replace_internals does db lookup to identify redlinks;
+//			global $wgContLang;
+//
+//			$colours = [];
+//			$linkCache = LinkCache::singleton();
+//			$output = this.parent->getOutput();
+		XomwLinkRenderer linkRenderer = this.parent.getLinkRenderer();
+//
+//			$dbr = wfGetDB( DB_REPLICA );
+//
+//			# Sort by namespace
+//			ksort( this.internals );
+//
+//			$linkcolour_ids = [];
+//
+//			# Generate query
+//			$lb = new LinkBatch();
+//			$lb->setCaller( __METHOD__ );
+//
+//			foreach ( this.internals as $ns => $entries ) {
+//				foreach ( $entries as $entry ) {
+//					/** @var Title $title */
+//					$title = $entry['title'];
+//					$pdbk = $entry['pdbk'];
+//
+//					# Skip invalid entries.
+//					# Result will be ugly, but prevents crash.
+//					if ( is_null( $title ) ) {
+//						continue;
+//					}
+//
+//					# Check if it's a static known link, e.g. interwiki
+//					if ( $title->isAlwaysKnown() ) {
+//						$colours[$pdbk] = '';
+//					} elseif ( $ns == NS_SPECIAL ) {
+//						$colours[$pdbk] = 'new';
+//					} else {
+//						$id = $linkCache->getGoodLinkID( $pdbk );
+//						if ( $id != 0 ) {
+//							$colours[$pdbk] = $linkRenderer->getLinkClasses( $title );
+//							$output->addLink( $title, $id );
+//							$linkcolour_ids[$id] = $pdbk;
+//						} elseif ( $linkCache->isBadLink( $pdbk ) ) {
+//							$colours[$pdbk] = 'new';
+//						} else {
+//							# Not in the link cache, add it to the query
+//							$lb->addObj( $title );
+//						}
+//					}
+//				}
+//			}
+//			if ( !$lb->isEmpty() ) {
+//				$fields = array_merge(
+//					LinkCache::getSelectFields(),
+//					[ 'page_namespace', 'page_title' ]
+//				);
+//
+//				$res = $dbr->select(
+//					'page',
+//					$fields,
+//					$lb->constructSet( 'page', $dbr ),
+//					__METHOD__
+//				);
+//
+//				# Fetch data and form into an associative array
+//				# non-existent = broken
+//				foreach ( $res as $s ) {
+//					$title = Title::makeTitle( $s->page_namespace, $s->page_title );
+//					$pdbk = $title->getPrefixedDBkey();
+//					$linkCache->addGoodLinkObjFromRow( $title, $s );
+//					$output->addLink( $title, $s->page_id );
+//					$colours[$pdbk] = $linkRenderer->getLinkClasses( $title );
+//					// add id to the extension todolist
+//					$linkcolour_ids[$s->page_id] = $pdbk;
+//				}
+//				unset( $res );
+//			}
+//			if ( count( $linkcolour_ids ) ) {
+//				// pass an array of page_ids to an extension
+//				Hooks::run( 'GetLinkColours', [ $linkcolour_ids, &$colours ] );
+//			}
+//
+//			# Do a second query for different language variants of links and categories
+//			if ( $wgContLang->hasVariants() ) {
+//				this.doVariants( $colours );
+//			}
+
+		// Construct search and replace arrays
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src = src_bfr.Bfr();
+		int src_bgn = 0;
+		int src_end = src_bfr.Len();
+		Bry_bfr bfr = pbfr.Trg();
+		pbfr.Switch();
+
+		int cur = src_bgn;
+		int prv = 0;
+		while (true) {
+			int link_bgn = Bry_find_.Find_fwd(src, Bry__link__bgn, cur, src_end);
+			if (link_bgn == Bry_find_.Not_found) {
+				bfr.Add_mid(src, prv, src_end);
+				break;
+			}
+			int key_bgn = link_bgn + Bry__link__bgn.length;
+			int key_end = Bry_find_.Find_fwd_while_num(src, key_bgn, src_end);
+			int link_key = Bry_.To_int_or(src, key_bgn, key_end, -1);
+			XomwLinkHolderItem item = internals.Get_by(link_key);
+
+//					$pdbk = $entry['pdbk'];
+//					$title = $entry['title'];
+//					$query = isset( $entry['query'] ) ? $entry['query'] : [];
+//					$key = "$ns:$index";
+//					$searchkey = "<!--LINK $key-->";
+//					$displayText = $entry['text'];
+//					if ( isset( $entry['selflink'] ) ) {
+//						$replacePairs[$searchkey] = Linker::makeSelfLinkObj( $title, $displayText, $query );
+//						continue;
+//					}
+//					if ( $displayText === '' ) {
+//						$displayText = null;
+//					} else {
+//						$displayText = new HtmlArmor( $displayText );
+//					}
+//					if ( !isset( $colours[$pdbk] ) ) {
+//						$colours[$pdbk] = 'new';
+//					}
+//					$attribs = [];
+//					if ( $colours[$pdbk] == 'new' ) {
+//						$linkCache->addBadLinkObj( $title );
+//						$output->addLink( $title, 0 );
+//						$link = $linkRenderer->makeBrokenLink(
+//							$title, $displayText, $attribs, $query
+//						);
+//					} else {
+//						$link = $linkRenderer->makePreloadedLink(
+//							$title, $displayText, $colours[$pdbk], $attribs, $query
+//						);
+//					}
+//
+//					$replacePairs[$searchkey] = $link;
+//				}
+//			}
+			bfr.Add_mid(src, prv, link_bgn);
+			linkRenderer.makePreloadedLink(bfr, item.Title(), item.Text(), Bry_.Empty, extraAtrs, query.Clear());
+			cur = key_end + Gfh_tag_.Comm_end_len;
+			prv = cur;
+		}
+//			$replacer = new HashtableReplacer( $replacePairs, 1 );
+//
+//			# Do the thing
+//			$text = preg_replace_callback(
+//				'/(<!--LINK .*?-->)/',
+//				$replacer->cb(),
+//				$text
+//			);
+		return true;
+	}
+
+//		/**
+//		* Replace interwiki links
+//		* @param String $text
+//		*/
+//		protected function replaceInterwiki( &$text ) {
+//			if ( empty( this.interwikis ) ) {
+//				return;
+//			}
+//
+//			# Make interwiki link HTML
+//			$output = this.parent->getOutput();
+//			$replacePairs = [];
+//			$linkRenderer = this.parent->getLinkRenderer();
+//			foreach ( this.interwikis as $key => $link ) {
+//				$replacePairs[$key] = $linkRenderer->makeLink(
+//					$link['title'],
+//					new HtmlArmor( $link['text'] )
+//				);
+//				$output->addInterwikiLink( $link['title'] );
+//			}
+//			$replacer = new HashtableReplacer( $replacePairs, 1 );
+//
+//			$text = preg_replace_callback(
+//				'/<!--IWLINK (.*?)-->/',
+//				$replacer->cb(),
+//				$text );
+//		}
+//
+//		/**
+//		* Modify this.internals and $colours according to language variant linking rules
+//		* @param array $colours
+//		*/
+//		protected function doVariants( &$colours ) {
+//			global $wgContLang;
+//			$linkBatch = new LinkBatch();
+//			$variantMap = []; // maps $pdbkey_Variant => $keys (of link holders)
+//			$output = this.parent->getOutput();
+//			$linkCache = LinkCache::singleton();
+//			$titlesToBeConverted = '';
+//			$titlesAttrs = [];
+//
+//			// Concatenate titles to a single String, thus we only need auto convert the
+//			// single String to all variants. This would improve parser's performance
+//			// significantly.
+//			foreach ( this.internals as $ns => $entries ) {
+//				if ( $ns == NS_SPECIAL ) {
+//					continue;
+//				}
+//				foreach ( $entries as $index => $entry ) {
+//					$pdbk = $entry['pdbk'];
+//					// we only deal with new links (in its first query)
+//					if ( !isset( $colours[$pdbk] ) || $colours[$pdbk] === 'new' ) {
+//						$titlesAttrs[] = [ $index, $entry['title'] ];
+//						// separate titles with \0 because it would never appears
+//						// in a valid title
+//						$titlesToBeConverted .= $entry['title']->getText() . "\0";
+//					}
+//				}
+//			}
+//
+//			// Now do the conversion and explode String to text of titles
+//			$titlesAllVariants = $wgContLang->autoConvertToAllVariants( rtrim( $titlesToBeConverted, "\0" ) );
+//			$allVariantsName = array_keys( $titlesAllVariants );
+//			foreach ( $titlesAllVariants as &$titlesVariant ) {
+//				$titlesVariant = explode( "\0", $titlesVariant );
+//			}
+//
+//			// Then add variants of links to link batch
+//			$parentTitle = this.parent->getTitle();
+//			foreach ( $titlesAttrs as $i => $attrs ) {
+//				/** @var Title $title */
+//				list( $index, $title ) = $attrs;
+//				$ns = $title->getNamespace();
+//				$text = $title->getText();
+//
+//				foreach ( $allVariantsName as $variantName ) {
+//					$textVariant = $titlesAllVariants[$variantName][$i];
+//					if ( $textVariant === $text ) {
+//						continue;
+//					}
+//
+//					$variantTitle = Title::makeTitle( $ns, $textVariant );
+//
+//					// Self-link checking for mixed/different variant titles. At this point, we
+//					// already know the exact title does not exist, so the link cannot be to a
+//					// variant of the current title that exists as a separate page.
+//					if ( $variantTitle->equals( $parentTitle ) && !$title->hasFragment() ) {
+//						this.internals[$ns][$index]['selflink'] = true;
+//						continue 2;
+//					}
+//
+//					$linkBatch->addObj( $variantTitle );
+//					$variantMap[$variantTitle->getPrefixedDBkey()][] = "$ns:$index";
+//				}
+//			}
+//
+//			// process categories, check if a category exists in some variant
+//			$categoryMap = []; // maps $category_variant => $category (dbkeys)
+//			$varCategories = []; // category replacements oldDBkey => newDBkey
+//			foreach ( $output->getCategoryLinks() as $category ) {
+//				$categoryTitle = Title::makeTitleSafe( NS_CATEGORY, $category );
+//				$linkBatch->addObj( $categoryTitle );
+//				$variants = $wgContLang->autoConvertToAllVariants( $category );
+//				foreach ( $variants as $variant ) {
+//					if ( $variant !== $category ) {
+//						$variantTitle = Title::makeTitleSafe( NS_CATEGORY, $variant );
+//						if ( is_null( $variantTitle ) ) {
+//							continue;
+//						}
+//						$linkBatch->addObj( $variantTitle );
+//						$categoryMap[$variant] = [ $category, $categoryTitle ];
+//					}
+//				}
+//			}
+//
+//			if ( !$linkBatch->isEmpty() ) {
+//				// construct query
+//				$dbr = wfGetDB( DB_REPLICA );
+//				$fields = array_merge(
+//					LinkCache::getSelectFields(),
+//					[ 'page_namespace', 'page_title' ]
+//				);
+//
+//				$varRes = $dbr->select( 'page',
+//					$fields,
+//					$linkBatch->constructSet( 'page', $dbr ),
+//					__METHOD__
+//				);
+//
+//				$linkcolour_ids = [];
+//				$linkRenderer = this.parent->getLinkRenderer();
+//
+//				// for each found variants, figure out link holders and replace
+//				foreach ( $varRes as $s ) {
+//					$variantTitle = Title::makeTitle( $s->page_namespace, $s->page_title );
+//					$varPdbk = $variantTitle->getPrefixedDBkey();
+//					$vardbk = $variantTitle->getDBkey();
+//
+//					$holderKeys = [];
+//					if ( isset( $variantMap[$varPdbk] ) ) {
+//						$holderKeys = $variantMap[$varPdbk];
+//						$linkCache->addGoodLinkObjFromRow( $variantTitle, $s );
+//						$output->addLink( $variantTitle, $s->page_id );
+//					}
+//
+//					// loop over link holders
+//					foreach ( $holderKeys as $key ) {
+//						list( $ns, $index ) = explode( ':', $key, 2 );
+//						$entry =& this.internals[$ns][$index];
+//						$pdbk = $entry['pdbk'];
+//
+//						if ( !isset( $colours[$pdbk] ) || $colours[$pdbk] === 'new' ) {
+//							// found link in some of the variants, replace the link holder data
+//							$entry['title'] = $variantTitle;
+//							$entry['pdbk'] = $varPdbk;
+//
+//							// set pdbk and colour
+//							$colours[$varPdbk] = $linkRenderer->getLinkClasses( $variantTitle );
+//							$linkcolour_ids[$s->page_id] = $pdbk;
+//						}
+//					}
+//
+//					// check if the Object is a variant of a category
+//					if ( isset( $categoryMap[$vardbk] ) ) {
+//						list( $oldkey, $oldtitle ) = $categoryMap[$vardbk];
+//						if ( !isset( $varCategories[$oldkey] ) && !$oldtitle->exists() ) {
+//							$varCategories[$oldkey] = $vardbk;
+//						}
+//					}
+//				}
+//				Hooks::run( 'GetLinkColours', [ $linkcolour_ids, &$colours ] );
+//
+//				// rebuild the categories in original order (if there are replacements)
+//				if ( count( $varCategories ) > 0 ) {
+//					$newCats = [];
+//					$originalCats = $output->getCategories();
+//					foreach ( $originalCats as $cat => $sortkey ) {
+//						// make the replacement
+//						if ( array_key_exists( $cat, $varCategories ) ) {
+//							$newCats[$varCategories[$cat]] = $sortkey;
+//						} else {
+//							$newCats[$cat] = $sortkey;
+//						}
+//					}
+//					$output->setCategoryLinks( $newCats );
+//				}
+//			}
+//		}
+//
+//		/**
+//		* Replace <!--LINK--> link placeholders with plain text of links
+//		* (not HTML-formatted).
+//		*
+//		* @param String $text
+//		* @return String
+//		*/
+//		public function replaceText( $text ) {
+//			$text = preg_replace_callback(
+//				'/<!--(LINK|IWLINK) (.*?)-->/',
+//				[ &$this, 'replaceTextCallback' ],
+//				$text );
+//
+//			return $text;
+//		}
+//
+//		/**
+//		* Callback for replaceText()
+//		*
+//		* @param array $matches
+//		* @return String
+//		* @private
+//		*/
+//		public function replaceTextCallback( $matches ) {
+//			$type = $matches[1];
+//			$key = $matches[2];
+//			if ( $type == 'LINK' ) {
+//				list( $ns, $index ) = explode( ':', $key, 2 );
+//				if ( isset( this.internals[$ns][$index]['text'] ) ) {
+//					return this.internals[$ns][$index]['text'];
+//				}
+//			} elseif ( $type == 'IWLINK' ) {
+//				if ( isset( this.interwikis[$key]['text'] ) ) {
+//					return this.interwikis[$key]['text'];
+//				}
+//			}
+//			return $matches[0];
+//		}
+
+
+//		private void Replace_internal__db() {
+//			// Generate query
+//			$lb = new LinkBatch();
+//			$lb->setCaller( __METHOD__ );
+//
+//			foreach ( $this->internals as $ns => $entries ) {
+//				foreach ( $entries as $entry ) {
+//					/** @var Title $title */
+//					$title = $entry['title'];
+//					$pdbk = $entry['pdbk'];
+//
+//					# Skip invalid entries.
+//					# Result will be ugly, but prevents crash.
+//					if ( is_null( $title ) ) {
+//						continue;
+//					}
+//
+//					# Check if it's a static known link, e.g. interwiki
+//					if ( $title->isAlwaysKnown() ) {
+//						$colours[$pdbk] = '';
+//					} elseif ( $ns == NS_SPECIAL ) {
+//						$colours[$pdbk] = 'new';
+//					} else {
+//						$id = $linkCache->getGoodLinkID( $pdbk );
+//						if ( $id != 0 ) {
+//							$colours[$pdbk] = $linkRenderer->getLinkClasses( $title );
+//							$output->addLink( $title, $id );
+//							$linkcolour_ids[$id] = $pdbk;
+//						} elseif ( $linkCache->isBadLink( $pdbk ) ) {
+//							$colours[$pdbk] = 'new';
+//						} else {
+//							# Not in the link cache, add it to the query
+//							$lb->addObj( $title );
+//						}
+//					}
+//				}
+//			}
+//			if ( !$lb->isEmpty() ) {
+//				$fields = array_merge(
+//					LinkCache::getSelectFields(),
+//					[ 'page_namespace', 'page_title' ]
+//				);
+//
+//				$res = $dbr->select(
+//					'page',
+//					$fields,
+//					$lb->constructSet( 'page', $dbr ),
+//					__METHOD__
+//				);
+//
+//				# Fetch data and form into an associative array
+//				# non-existent = broken
+//				foreach ( $res as $s ) {
+//					$title = Title::makeTitle( $s->page_namespace, $s->page_title );
+//					$pdbk = $title->getPrefixedDBkey();
+//					$linkCache->addGoodLinkObjFromRow( $title, $s );
+//					$output->addLink( $title, $s->page_id );
+//					$colours[$pdbk] = $linkRenderer->getLinkClasses( $title );
+//					// add id to the extension todolist
+//					$linkcolour_ids[$s->page_id] = $pdbk;
+//				}
+//				unset( $res );
+//			}
+//			if ( count( $linkcolour_ids ) ) {
+//				// pass an array of page_ids to an extension
+//				Hooks::run( 'GetLinkColours', [ $linkcolour_ids, &$colours ] );
+//			}
+//
+//			# Do a second query for different language variants of links and categories
+//			if ( $wgContLang->hasVariants() ) {
+//				$this->doVariants( $colours );
+//			}
+//		}
+	public void Test__add(XomwTitle ttl, byte[] capt) {
+		int key = parent.nextLinkID();
+		XomwLinkHolderItem item = new XomwLinkHolderItem(ttl, capt, Bry_.Ary_empty);
+		internals.Add(key, item);
+	}
+	private static final    byte[] Bry__link__bgn = Bry_.new_a7("<!--LINK ");
+}
+class XomwLinkHolderList {
+	private int ary_len = 0, ary_max = 128;
+	private XomwLinkHolderItem[] ary = new XomwLinkHolderItem[128];
+	public int Len() {return ary_len;}
+	public void Clear() {
+		ary_len = 0;
+		if (ary_max > 128)
+			ary = new XomwLinkHolderItem[128];
+	}
+	public void Add(int key, XomwLinkHolderItem item) {
+		if (key >= ary_max) {
+			int new_max = ary_max * 2;
+			ary = (XomwLinkHolderItem[])Array_.Resize(ary, new_max);
+			ary_max = new_max;
+		}
+		ary[key] = item;
+		ary_len++;
+	}
+	public XomwLinkHolderItem Get_by(int key) {return ary[key];}
+}
+class XomwLinkHolderItem {
+	public XomwLinkHolderItem(XomwTitle title, byte[] text, byte[][] query) {
+		this.title = title;
+		this.text = text;
+		this.query = query;
+	}
+	public XomwTitle Title() {return title;} private final    XomwTitle title;
+	public byte[] Text()      {return text;} private final    byte[] text;
+	public byte[] Pdbk()      {return title.getPrefixedDBkey();}
+	public byte[][] Query()   {return query;} private final    byte[][] query;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwLinkHolderArrayTest.java

@@ -13,3 +13,32 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+import gplx.xowa.mediawiki.includes.linkers.*;
+public class XomwLinkHolderArrayTest {
+	private final    XomwLinkHolderArrayFxt fxt = new XomwLinkHolderArrayFxt();
+	@Test   public void Replace__basic() {
+		fxt.Init__add("A", "a");
+		fxt.Test__replace("a <!--LINK 0--> b", "a <a href='/wiki/A' title='A'>a</a> b");
+	}
+}
+class XomwLinkHolderArrayFxt {
+	private final    XomwEnv env;
+	private final    XomwLinkHolderArray holders;
+	private final    XomwParserBfr pbfr = new XomwParserBfr();
+	private boolean apos = true;
+	public XomwLinkHolderArrayFxt() {
+		XomwParser parser = new XomwParser(XomwEnv.NewTest());
+		this.env = parser.Env();
+		this.holders = new XomwLinkHolderArray(parser);
+	}
+	public void Init__add(String ttl, String capt) {
+		holders.Test__add(XomwTitle.newFromText(env, Bry_.new_u8(ttl)), Bry_.new_u8(capt));
+	}
+	public void Test__replace(String src, String expd) {
+		if (apos) expd = gplx.langs.htmls.Gfh_utl.Replace_apos(expd);
+		holders.replace(pbfr.Init(Bry_.new_u8(src)));
+		Gftest.Eq__str(expd, pbfr.Rslt().To_str_and_clear());
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserBfr.java

@@ -13,3 +13,34 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class XomwParserBfr {	// manages 2 bfrs to eliminate multiple calls to new memory allocations ("return bfr.To_bry_and_clear()")
+	private final    Bry_bfr bfr_1 = Bry_bfr_.New(), bfr_2 = Bry_bfr_.New();
+	private Bry_bfr src, trg;
+	public XomwParserBfr() {
+		this.src = bfr_1;
+		this.trg = bfr_2;
+	}		
+	public Bry_bfr Src() {return src;}
+	public Bry_bfr Trg() {return trg;}
+	public Bry_bfr Rslt() {return src;}
+	public XomwParserBfr Init(byte[] text) {
+		// resize each bfr once by guessing that html_len = text_len * 2
+		int text_len = text.length;
+		int html_len = text_len * 2;
+		src.Resize(html_len);
+		trg.Resize(html_len);
+
+		// clear and add
+		src.Clear();
+		trg.Clear();
+		src.Add(text);
+		return this;
+	}
+	public void Switch() {
+		Bry_bfr tmp = src;
+		this.src = trg;
+		this.trg = tmp;
+		trg.Clear();
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserBfr_.java

@@ -13,3 +13,55 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class XomwParserBfr_ {
+	public static void Replace(XomwParserBfr pbfr, byte[] find, byte[] repl) {
+		// XO.PBFR
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src = src_bfr.Bfr();
+		int src_bgn = 0;
+		int src_end = src_bfr.Len();
+		Bry_bfr bfr = pbfr.Trg();
+
+		if (Replace(bfr, Bool_.N, src, src_bgn, src_end, find, repl) != null)
+			pbfr.Switch();
+	}
+	private static byte[] Replace(Bry_bfr bfr, boolean lone_bfr, byte[] src, int src_bgn, int src_end, byte[] find, byte[] repl) {
+		boolean dirty = false;
+		int cur = src_bgn;
+		boolean called_by_bry = bfr == null;
+
+		while (true) {
+			int find_bgn = Bry_find_.Find_fwd(src, find, cur);
+			if (find_bgn == Bry_find_.Not_found) {
+				if (dirty)
+					bfr.Add_mid(src, cur, src_end);
+				break;
+			}
+			if (called_by_bry) bfr = Bry_bfr_.New();
+			bfr.Add_mid(src, cur, find_bgn);
+			cur += find.length;
+			dirty = true;
+		}
+
+		if (dirty) {
+			if (called_by_bry)
+				return bfr.To_bry_and_clear();
+			else
+				return Bry_.Empty;
+		}
+		else {
+			if (called_by_bry) {
+				if (src_bgn == 0 && src_end == src.length)
+					return src;
+				else
+					return Bry_.Mid(src, src_bgn, src_end);
+			}
+			else {
+				if (lone_bfr)
+					bfr.Add_mid(src, src_bgn, src_end);
+				return null;
+			}
+		}
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserCtx.java

@@ -13,3 +13,18 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.parsers.lnkis.*;
+public class XomwParserCtx {
+	public XomwTitle Page_title() {return page_title;} private XomwTitle page_title;
+	public Xomw_image_params         Lnki_wkr__make_image__img_params = new Xomw_image_params();
+	public byte[][]                  Lnki_wkr__make_image__match_magic_word = new byte[2][];
+	public int[]                     Lnki_wkr__make_image__img_size = new int[2];
+	public Xomw_params_mto           Linker__makeImageLink__prms = new Xomw_params_mto();
+
+	public void Init_by_page(XomwTitle page_title) {
+		this.page_title = page_title;
+	}
+
+	public static final int Pos__bos = -1;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserIface.java

@@ -13,3 +13,15 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.xowa.mediawiki.includes.xohtml.*;
+import gplx.xowa.mediawiki.includes.linkers.*;
+public interface XomwParserIface {
+	int nextLinkID();
+	XomwParserOptions getOptions();
+	XomwLinkRenderer getLinkRenderer();
+
+	byte[] armorLinks(Bry_bfr trg, byte[] src, int src_bgn, int src_end);
+	Xomw_atr_mgr getExternalLinkAttribs(Xomw_atr_mgr atrs);
+	byte[] stripAltText(byte[] caption, XomwLinkHolderArray holders);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserOptions.java

@@ -13,3 +13,919 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class XomwParserOptions {
+	public XomwParserOptions() {
+		this.mThumbSize = 220;
+	}
+//		/**
+//		* Interlanguage links are removed and returned in an array
+//		*/
+//		private $mInterwikiMagic;
+//
+//		/**
+//		* Allow external images inline?
+//		*/
+//		private $mAllowExternalImages;
+//
+//		/**
+//		* If not, any exception?
+//		*/
+//		private $mAllowExternalImagesFrom;
+//
+//		/**
+//		* If not or it doesn't match, should we check an on-wiki whitelist?
+//		*/
+//		private $mEnableImageWhitelist;
+//
+//		/**
+//		* Date format index
+//		*/
+//		private $mDateFormat = null;
+//
+//		/**
+//		* Create "edit section" links?
+//		*/
+//		private $mEditSection = true;
+//
+//		/**
+//		* Allow inclusion of special pages?
+//		*/
+//		private $mAllowSpecialInclusion;
+//
+//		/**
+//		* Use tidy to cleanup output HTML?
+//		*/
+//		private $mTidy = false;
+//
+//		/**
+//		* Which lang to call for PLURAL and GRAMMAR
+//		*/
+//		private $mInterfaceMessage = false;
+//
+//		/**
+//		* Overrides $mInterfaceMessage with arbitrary language
+//		*/
+//		private $mTargetLanguage = null;
+//
+//		/**
+//		* Maximum size of template expansions, in bytes
+//		*/
+//		private $mMaxIncludeSize;
+//
+//		/**
+//		* Maximum number of nodes touched by PPFrame::expand()
+//		*/
+//		private $mMaxPPNodeCount;
+//
+//		/**
+//		* Maximum number of nodes generated by Preprocessor::preprocessToObj()
+//		*/
+//		private $mMaxGeneratedPPNodeCount;
+//
+//		/**
+//		* Maximum recursion depth in PPFrame::expand()
+//		*/
+//		private $mMaxPPExpandDepth;
+//
+//		/**
+//		* Maximum recursion depth for templates within templates
+//		*/
+//		private $mMaxTemplateDepth;
+//
+//		/**
+//		* Maximum number of calls per parse to expensive parser functions
+//		*/
+//		private $mExpensiveParserFunctionLimit;
+//
+//		/**
+//		* Remove HTML comments. ONLY APPLIES TO PREPROCESS OPERATIONS
+//		*/
+//		private $mRemoveComments = true;
+//
+//		/**
+//		* @var callable Callback for current revision fetching; first argument to call_user_func().
+//		*/
+//		private $mCurrentRevisionCallback =
+//			[ 'Parser', 'statelessFetchRevision' ];
+//
+//		/**
+//		* @var callable Callback for template fetching; first argument to call_user_func().
+//		*/
+//		private $mTemplateCallback =
+//			[ 'Parser', 'statelessFetchTemplate' ];
+//
+//		/**
+//		* @var callable|null Callback to generate a guess for {{REVISIONID}}
+//		*/
+//		private $mSpeculativeRevIdCallback;
+//
+//		/**
+//		* Enable limit report in an HTML comment on output
+//		*/
+//		private $mEnableLimitReport = false;
+//
+//		/**
+//		* Timestamp used for {{CURRENTDAY}} etc.
+//		*/
+//		private $mTimestamp;
+//
+//		/**
+//		* Target attribute for external links
+//		*/
+//		private $mExternalLinkTarget;
+//
+//		/**
+//		* Clean up signature texts?
+//		* @see Parser::cleanSig
+//		*/
+//		private $mCleanSignatures;
+//
+//		/**
+//		* Transform wiki markup when saving the page?
+//		*/
+//		private $mPreSaveTransform = true;
+//
+//		/**
+//		* Whether content conversion should be disabled
+//		*/
+//		private $mDisableContentConversion;
+//
+//		/**
+//		* Whether title conversion should be disabled
+//		*/
+//		private $mDisableTitleConversion;
+//
+//		/**
+//		* Automatically number headings?
+//		*/
+//		private $mNumberHeadings;
+
+	/**
+	* Thumb size preferred by the user.
+	*/
+	private int mThumbSize;
+
+//		/**
+//		* Maximum article size of an article to be marked as "stub"
+//		*/
+//		private $mStubThreshold;
+//
+//		/**
+//		* Language Object of the User language.
+//		*/
+//		private $mUserLang;
+//
+//		/**
+//		* @var User
+//		* Stored user Object
+//		*/
+//		private $mUser;
+//
+//		/**
+//		* Parsing the page for a "preview" operation?
+//		*/
+//		private $mIsPreview = false;
+//
+//		/**
+//		* Parsing the page for a "preview" operation on a single section?
+//		*/
+//		private $mIsSectionPreview = false;
+//
+//		/**
+//		* Parsing the printable version of the page?
+//		*/
+//		private $mIsPrintable = false;
+//
+//		/**
+//		* Extra key that should be present in the caching key.
+//		*/
+//		private $mExtraKey = '';
+//
+//		/**
+//		* Are magic ISBN links enabled?
+//		*/
+//		private $mMagicISBNLinks = true;
+//
+//		/**
+//		* Are magic PMID links enabled?
+//		*/
+//		private $mMagicPMIDLinks = true;
+//
+//		/**
+//		* Are magic RFC links enabled?
+//		*/
+//		private $mMagicRFCLinks = true;
+//
+//		/**
+//		* Function to be called when an option is accessed.
+//		*/
+//		private $onAccessCallback = null;
+//
+//		/**
+//		* If the page being parsed is a redirect, this should hold the redirect
+//		* target.
+//		* @var Title|null
+//		*/
+//		private $redirectTarget = null;
+//
+//		public function getInterwikiMagic() {
+//			return this.mInterwikiMagic;
+//		}
+//
+//		public function getAllowExternalImages() {
+//			return this.mAllowExternalImages;
+//		}
+//
+//		public function getAllowExternalImagesFrom() {
+//			return this.mAllowExternalImagesFrom;
+//		}
+//
+//		public function getEnableImageWhitelist() {
+//			return this.mEnableImageWhitelist;
+//		}
+//
+//		public function getEditSection() {
+//			return this.mEditSection;
+//		}
+//
+//		public function getNumberHeadings() {
+//			this.optionUsed( 'numberheadings' );
+//
+//			return this.mNumberHeadings;
+//		}
+//
+//		public function getAllowSpecialInclusion() {
+//			return this.mAllowSpecialInclusion;
+//		}
+//
+//		public function getTidy() {
+//			return this.mTidy;
+//		}
+//
+//		public function getInterfaceMessage() {
+//			return this.mInterfaceMessage;
+//		}
+//
+//		public function getTargetLanguage() {
+//			return this.mTargetLanguage;
+//		}
+//
+//		public function getMaxIncludeSize() {
+//			return this.mMaxIncludeSize;
+//		}
+//
+//		public function getMaxPPNodeCount() {
+//			return this.mMaxPPNodeCount;
+//		}
+//
+//		public function getMaxGeneratedPPNodeCount() {
+//			return this.mMaxGeneratedPPNodeCount;
+//		}
+//
+//		public function getMaxPPExpandDepth() {
+//			return this.mMaxPPExpandDepth;
+//		}
+//
+//		public function getMaxTemplateDepth() {
+//			return this.mMaxTemplateDepth;
+//		}
+//
+//		/* @since 1.20 */
+//		public function getExpensiveParserFunctionLimit() {
+//			return this.mExpensiveParserFunctionLimit;
+//		}
+//
+//		public function getRemoveComments() {
+//			return this.mRemoveComments;
+//		}
+//
+//		/* @since 1.24 */
+//		public function getCurrentRevisionCallback() {
+//			return this.mCurrentRevisionCallback;
+//		}
+//
+//		public function getTemplateCallback() {
+//			return this.mTemplateCallback;
+//		}
+//
+//		/** @since 1.28 */
+//		public function getSpeculativeRevIdCallback() {
+//			return this.mSpeculativeRevIdCallback;
+//		}
+//
+//		public function getEnableLimitReport() {
+//			return this.mEnableLimitReport;
+//		}
+//
+//		public function getCleanSignatures() {
+//			return this.mCleanSignatures;
+//		}
+//
+//		public function getExternalLinkTarget() {
+//			return this.mExternalLinkTarget;
+//		}
+//
+//		public function getDisableContentConversion() {
+//			return this.mDisableContentConversion;
+//		}
+//
+//		public function getDisableTitleConversion() {
+//			return this.mDisableTitleConversion;
+//		}
+
+	public int getThumbSize() {
+		// this.optionUsed( 'thumbsize' );
+
+		return this.mThumbSize;
+	}
+
+//		public function getStubThreshold() {
+//			this.optionUsed( 'stubthreshold' );
+//
+//			return this.mStubThreshold;
+//		}
+//
+//		public function getIsPreview() {
+//			return this.mIsPreview;
+//		}
+//
+//		public function getIsSectionPreview() {
+//			return this.mIsSectionPreview;
+//		}
+//
+//		public function getIsPrintable() {
+//			this.optionUsed( 'printable' );
+//
+//			return this.mIsPrintable;
+//		}
+//
+//		public function getUser() {
+//			return this.mUser;
+//		}
+//
+//		public function getPreSaveTransform() {
+//			return this.mPreSaveTransform;
+//		}
+//
+//		public function getDateFormat() {
+//			this.optionUsed( 'dateformat' );
+//			if ( !isset( this.mDateFormat ) ) {
+//				this.mDateFormat = this.mUser->getDatePreference();
+//			}
+//			return this.mDateFormat;
+//		}
+//
+//		public function getTimestamp() {
+//			if ( !isset( this.mTimestamp ) ) {
+//				this.mTimestamp = wfTimestampNow();
+//			}
+//			return this.mTimestamp;
+//		}
+//
+//		/**
+//		* Get the user language used by the parser for this page and split the parser cache.
+//		*
+//		* @warning: Calling this causes the parser cache to be fragmented by user language!
+//		* To avoid cache fragmentation, output should not depend on the user language.
+//		* Use Parser::getFunctionLang() or Parser::getTargetLanguage() instead!
+//		*
+//		* @note This function will trigger a cache fragmentation by recording the
+//		* 'userlang' option, see optionUsed(). This is done to avoid cache pollution
+//		* when the page is rendered based on the language of the user.
+//		*
+//		* @note When saving, this will return the default language instead of the user's.
+//		* {{int: }} uses this which used to produce inconsistent link tables (bug 14404).
+//		*
+//		* @return Language
+//		* @since 1.19
+//		*/
+//		public function getUserLangObj() {
+//			this.optionUsed( 'userlang' );
+//			return this.mUserLang;
+//		}
+//
+//		/**
+//		* Same as getUserLangObj() but returns a String instead.
+//		*
+//		* @warning: Calling this causes the parser cache to be fragmented by user language!
+//		* To avoid cache fragmentation, output should not depend on the user language.
+//		* Use Parser::getFunctionLang() or Parser::getTargetLanguage() instead!
+//		*
+//		* @see getUserLangObj()
+//		*
+//		* @return String Language code
+//		* @since 1.17
+//		*/
+//		public function getUserLang() {
+//			return this.getUserLangObj()->getCode();
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return boolean
+//		*/
+//		public function getMagicISBNLinks() {
+//			return this.mMagicISBNLinks;
+//		}
+//
+//		/**
+//		* @since 1.28
+//		* @return boolean
+//		*/
+//		public function getMagicPMIDLinks() {
+//			return this.mMagicPMIDLinks;
+//		}
+//		/**
+//		* @since 1.28
+//		* @return boolean
+//		*/
+//		public function getMagicRFCLinks() {
+//			return this.mMagicRFCLinks;
+//		}
+//		public function setInterwikiMagic( $x ) {
+//			return wfSetVar( this.mInterwikiMagic, $x );
+//		}
+//
+//		public function setAllowExternalImages( $x ) {
+//			return wfSetVar( this.mAllowExternalImages, $x );
+//		}
+//
+//		public function setAllowExternalImagesFrom( $x ) {
+//			return wfSetVar( this.mAllowExternalImagesFrom, $x );
+//		}
+//
+//		public function setEnableImageWhitelist( $x ) {
+//			return wfSetVar( this.mEnableImageWhitelist, $x );
+//		}
+//
+//		public function setDateFormat( $x ) {
+//			return wfSetVar( this.mDateFormat, $x );
+//		}
+//
+//		public function setEditSection( $x ) {
+//			return wfSetVar( this.mEditSection, $x );
+//		}
+//
+//		public function setNumberHeadings( $x ) {
+//			return wfSetVar( this.mNumberHeadings, $x );
+//		}
+//
+//		public function setAllowSpecialInclusion( $x ) {
+//			return wfSetVar( this.mAllowSpecialInclusion, $x );
+//		}
+//
+//		public function setTidy( $x ) {
+//			return wfSetVar( this.mTidy, $x );
+//		}
+//
+//		public function setInterfaceMessage( $x ) {
+//			return wfSetVar( this.mInterfaceMessage, $x );
+//		}
+//
+//		public function setTargetLanguage( $x ) {
+//			return wfSetVar( this.mTargetLanguage, $x, true );
+//		}
+//
+//		public function setMaxIncludeSize( $x ) {
+//			return wfSetVar( this.mMaxIncludeSize, $x );
+//		}
+//
+//		public function setMaxPPNodeCount( $x ) {
+//			return wfSetVar( this.mMaxPPNodeCount, $x );
+//		}
+//
+//		public function setMaxGeneratedPPNodeCount( $x ) {
+//			return wfSetVar( this.mMaxGeneratedPPNodeCount, $x );
+//		}
+//
+//		public function setMaxTemplateDepth( $x ) {
+//			return wfSetVar( this.mMaxTemplateDepth, $x );
+//		}
+//
+//		/* @since 1.20 */
+//		public function setExpensiveParserFunctionLimit( $x ) {
+//			return wfSetVar( this.mExpensiveParserFunctionLimit, $x );
+//		}
+//
+//		public function setRemoveComments( $x ) {
+//			return wfSetVar( this.mRemoveComments, $x );
+//		}
+//
+//		/* @since 1.24 */
+//		public function setCurrentRevisionCallback( $x ) {
+//			return wfSetVar( this.mCurrentRevisionCallback, $x );
+//		}
+//
+//		/** @since 1.28 */
+//		public function setSpeculativeRevIdCallback( $x ) {
+//			return wfSetVar( this.mSpeculativeRevIdCallback, $x );
+//		}
+//
+//		public function setTemplateCallback( $x ) {
+//			return wfSetVar( this.mTemplateCallback, $x );
+//		}
+//
+//		public function enableLimitReport( $x = true ) {
+//			return wfSetVar( this.mEnableLimitReport, $x );
+//		}
+//
+//		public function setTimestamp( $x ) {
+//			return wfSetVar( this.mTimestamp, $x );
+//		}
+//
+//		public function setCleanSignatures( $x ) {
+//			return wfSetVar( this.mCleanSignatures, $x );
+//		}
+//
+//		public function setExternalLinkTarget( $x ) {
+//			return wfSetVar( this.mExternalLinkTarget, $x );
+//		}
+//
+//		public function disableContentConversion( $x = true ) {
+//			return wfSetVar( this.mDisableContentConversion, $x );
+//		}
+//
+//		public function disableTitleConversion( $x = true ) {
+//			return wfSetVar( this.mDisableTitleConversion, $x );
+//		}
+//
+//		public function setUserLang( $x ) {
+//			if ( is_string( $x ) ) {
+//				$x = Language::factory( $x );
+//			}
+//
+//			return wfSetVar( this.mUserLang, $x );
+//		}
+//
+//		public function setThumbSize( $x ) {
+//			return wfSetVar( this.mThumbSize, $x );
+//		}
+//
+//		public function setStubThreshold( $x ) {
+//			return wfSetVar( this.mStubThreshold, $x );
+//		}
+//
+//		public function setPreSaveTransform( $x ) {
+//			return wfSetVar( this.mPreSaveTransform, $x );
+//		}
+//
+//		public function setIsPreview( $x ) {
+//			return wfSetVar( this.mIsPreview, $x );
+//		}
+//
+//		public function setIsSectionPreview( $x ) {
+//			return wfSetVar( this.mIsSectionPreview, $x );
+//		}
+//
+//		public function setIsPrintable( $x ) {
+//			return wfSetVar( this.mIsPrintable, $x );
+//		}
+//
+//		/**
+//		* Set the redirect target.
+//		*
+//		* Note that setting or changing this does not *make* the page a redirect
+//		* or change its target, it merely records the information for reference
+//		* during the parse.
+//		*
+//		* @since 1.24
+//		* @param Title|null $title
+//		*/
+//		function setRedirectTarget( $title ) {
+//			this.redirectTarget = $title;
+//		}
+//
+//		/**
+//		* Get the previously-set redirect target.
+//		*
+//		* @since 1.24
+//		* @return Title|null
+//		*/
+//		function getRedirectTarget() {
+//			return this.redirectTarget;
+//		}
+//
+//		/**
+//		* Extra key that should be present in the parser cache key.
+//		* @param String $key
+//		*/
+//		public function addExtraKey( $key ) {
+//			this.mExtraKey .= '!' . $key;
+//		}
+//
+//		/**
+//		* Constructor
+//		* @param User $user
+//		* @param Language $lang
+//		*/
+//		public function __construct( $user = null, $lang = null ) {
+//			if ( $user === null ) {
+//				global $wgUser;
+//				if ( $wgUser === null ) {
+//					$user = new User;
+//				} else {
+//					$user = $wgUser;
+//				}
+//			}
+//			if ( $lang === null ) {
+//				global $wgLang;
+//				if ( !StubObject::isRealObject( $wgLang ) ) {
+//					$wgLang->_unstub();
+//				}
+//				$lang = $wgLang;
+//			}
+//			this.initialiseFromUser( $user, $lang );
+//		}
+//
+//		/**
+//		* Get a ParserOptions Object for an anonymous user
+//		* @since 1.27
+//		* @return ParserOptions
+//		*/
+//		public static function newFromAnon() {
+//			global $wgContLang;
+//			return new ParserOptions( new User, $wgContLang );
+//		}
+//
+//		/**
+//		* Get a ParserOptions Object from a given user.
+//		* Language will be taken from $wgLang.
+//		*
+//		* @param User $user
+//		* @return ParserOptions
+//		*/
+//		public static function newFromUser( $user ) {
+//			return new ParserOptions( $user );
+//		}
+//
+//		/**
+//		* Get a ParserOptions Object from a given user and language
+//		*
+//		* @param User $user
+//		* @param Language $lang
+//		* @return ParserOptions
+//		*/
+//		public static function newFromUserAndLang( User $user, Language $lang ) {
+//			return new ParserOptions( $user, $lang );
+//		}
+//
+//		/**
+//		* Get a ParserOptions Object from a IContextSource Object
+//		*
+//		* @param IContextSource $context
+//		* @return ParserOptions
+//		*/
+//		public static function newFromContext( IContextSource $context ) {
+//			return new ParserOptions( $context->getUser(), $context->getLanguage() );
+//		}
+//
+//		/**
+//		* Get user options
+//		*
+//		* @param User $user
+//		* @param Language $lang
+//		*/
+//		private function initialiseFromUser( $user, $lang ) {
+//			global $wgInterwikiMagic, $wgAllowExternalImages,
+//				$wgAllowExternalImagesFrom, $wgEnableImageWhitelist, $wgAllowSpecialInclusion,
+//				$wgMaxArticleSize, $wgMaxPPNodeCount, $wgMaxTemplateDepth, $wgMaxPPExpandDepth,
+//				$wgCleanSignatures, $wgExternalLinkTarget, $wgExpensiveParserFunctionLimit,
+//				$wgMaxGeneratedPPNodeCount, $wgDisableLangConversion, $wgDisableTitleConversion,
+//				$wgEnableMagicLinks;
+//
+//			// *UPDATE* ParserOptions::matches() if any of this changes as needed
+//			this.mInterwikiMagic = $wgInterwikiMagic;
+//			this.mAllowExternalImages = $wgAllowExternalImages;
+//			this.mAllowExternalImagesFrom = $wgAllowExternalImagesFrom;
+//			this.mEnableImageWhitelist = $wgEnableImageWhitelist;
+//			this.mAllowSpecialInclusion = $wgAllowSpecialInclusion;
+//			this.mMaxIncludeSize = $wgMaxArticleSize * 1024;
+//			this.mMaxPPNodeCount = $wgMaxPPNodeCount;
+//			this.mMaxGeneratedPPNodeCount = $wgMaxGeneratedPPNodeCount;
+//			this.mMaxPPExpandDepth = $wgMaxPPExpandDepth;
+//			this.mMaxTemplateDepth = $wgMaxTemplateDepth;
+//			this.mExpensiveParserFunctionLimit = $wgExpensiveParserFunctionLimit;
+//			this.mCleanSignatures = $wgCleanSignatures;
+//			this.mExternalLinkTarget = $wgExternalLinkTarget;
+//			this.mDisableContentConversion = $wgDisableLangConversion;
+//			this.mDisableTitleConversion = $wgDisableLangConversion || $wgDisableTitleConversion;
+//			this.mMagicISBNLinks = $wgEnableMagicLinks['ISBN'];
+//			this.mMagicPMIDLinks = $wgEnableMagicLinks['PMID'];
+//			this.mMagicRFCLinks = $wgEnableMagicLinks['RFC'];
+//
+//			this.mUser = $user;
+//			this.mNumberHeadings = $user->getOption( 'numberheadings' );
+//			this.mThumbSize = $user->getOption( 'thumbsize' );
+//			this.mStubThreshold = $user->getStubThreshold();
+//			this.mUserLang = $lang;
+//		}
+//
+//		/**
+//		* Check if these options match that of another options set
+//		*
+//		* This ignores report limit settings that only affect HTML comments
+//		*
+//		* @param ParserOptions $other
+//		* @return boolean
+//		* @since 1.25
+//		*/
+//		public function matches( ParserOptions $other ) {
+//			$fields = array_keys( get_class_vars( __CLASS__ ) );
+//			$fields = array_diff( $fields, [
+//				'mEnableLimitReport', // only effects HTML comments
+//				'onAccessCallback', // only used for ParserOutput option tracking
+//			] );
+//			foreach ( $fields as $field ) {
+//				if ( !is_object( this.$field ) && this.$field !== $other->$field ) {
+//					return false;
+//				}
+//			}
+//			// Check the Object and lazy-loaded options
+//			return (
+//				this.mUserLang->equals( $other->mUserLang ) &&
+//				this.getDateFormat() === $other->getDateFormat()
+//			);
+//		}
+//
+//		/**
+//		* Registers a callback for tracking which ParserOptions which are used.
+//		* This is a private API with the parser.
+//		* @param callable $callback
+//		*/
+//		public function registerWatcher( $callback ) {
+//			this.onAccessCallback = $callback;
+//		}
+//
+//		/**
+//		* Called when an option is accessed.
+//		* Calls the watcher that was set using registerWatcher().
+//		* Typically, the watcher callback is ParserOutput::registerOption().
+//		* The information registered that way will be used by ParserCache::save().
+//		*
+//		* @param String $optionName Name of the option
+//		*/
+//		public function optionUsed( $optionName ) {
+//			if ( this.onAccessCallback ) {
+//				call_user_func( this.onAccessCallback, $optionName );
+//			}
+//		}
+//
+//		/**
+//		* Returns the full array of options that would have been used by
+//		* in 1.16.
+//		* Used to get the old parser cache entries when available.
+//		* @return array
+//		*/
+//		public static function legacyOptions() {
+//			return [
+//				'stubthreshold',
+//				'numberheadings',
+//				'userlang',
+//				'thumbsize',
+//				'editsection',
+//				'printable'
+//			];
+//		}
+//
+//		/**
+//		* Generate a hash String with the values set on these ParserOptions
+//		* for the keys given in the array.
+//		* This will be used as part of the hash key for the parser cache,
+//		* so users sharing the options with vary for the same page share
+//		* the same cached data safely.
+//		*
+//		* Extensions which require it should install 'PageRenderingHash' hook,
+//		* which will give them a chance to modify this key based on their own
+//		* settings.
+//		*
+//		* @since 1.17
+//		* @param array $forOptions
+//		* @param Title $title Used to get the content language of the page (since r97636)
+//		* @return String Page rendering hash
+//		*/
+//		public function optionsHash( $forOptions, $title = null ) {
+//			global $wgRenderHashAppend;
+//
+//			// FIXME: Once the cache key is reorganized this argument
+//			// can be dropped. It was used when the math extension was
+//			// part of core.
+//			$confstr = '*';
+//
+//			// Space assigned for the stubthreshold but unused
+//			// since it disables the parser cache, its value will always
+//			// be 0 when this function is called by parsercache.
+//			if ( in_array( 'stubthreshold', $forOptions ) ) {
+//				$confstr .= '!' . this.mStubThreshold;
+//			} else {
+//				$confstr .= '!*';
+//			}
+//
+//			if ( in_array( 'dateformat', $forOptions ) ) {
+//				$confstr .= '!' . this.getDateFormat();
+//			}
+//
+//			if ( in_array( 'numberheadings', $forOptions ) ) {
+//				$confstr .= '!' . ( this.mNumberHeadings ? '1' : '' );
+//			} else {
+//				$confstr .= '!*';
+//			}
+//
+//			if ( in_array( 'userlang', $forOptions ) ) {
+//				$confstr .= '!' . this.mUserLang->getCode();
+//			} else {
+//				$confstr .= '!*';
+//			}
+//
+//			if ( in_array( 'thumbsize', $forOptions ) ) {
+//				$confstr .= '!' . this.mThumbSize;
+//			} else {
+//				$confstr .= '!*';
+//			}
+//
+//			// add in language specific options, if any
+//			// @todo FIXME: This is just a way of retrieving the url/user preferred variant
+//			if ( !is_null( $title ) ) {
+//				$confstr .= $title->getPageLanguage()->getExtraHashOptions();
+//			} else {
+//				global $wgContLang;
+//				$confstr .= $wgContLang->getExtraHashOptions();
+//			}
+//
+//			$confstr .= $wgRenderHashAppend;
+//
+//			// @note: as of Feb 2015, core never sets the editsection flag, since it uses
+//			// <mw:editsection> tags to inject editsections on the fly. However, extensions
+//			// may be using it by calling ParserOption::optionUsed resp. ParserOutput::registerOption
+//			// directly. At least Wikibase does at this point in time.
+//			if ( !in_array( 'editsection', $forOptions ) ) {
+//				$confstr .= '!*';
+//			} elseif ( !this.mEditSection ) {
+//				$confstr .= '!edit=0';
+//			}
+//
+//			if ( this.mIsPrintable && in_array( 'printable', $forOptions ) ) {
+//				$confstr .= '!printable=1';
+//			}
+//
+//			if ( this.mExtraKey != '' ) {
+//				$confstr .= this.mExtraKey;
+//			}
+//
+//			// Give a chance for extensions to modify the hash, if they have
+//			// extra options or other effects on the parser cache.
+//			Hooks::run( 'PageRenderingHash', [ &$confstr, this.getUser(), &$forOptions ] );
+//
+//			// Make it a valid memcached key fragment
+//			$confstr = str_replace( ' ', '_', $confstr );
+//
+//			return $confstr;
+//		}
+//
+//		/**
+//		* Sets a hook to force that a page exists, and sets a current revision callback to return
+//		* a revision with custom content when the current revision of the page is requested.
+//		*
+//		* @since 1.25
+//		* @param Title $title
+//		* @param Content $content
+//		* @param User $user The user that the fake revision is attributed to
+//		* @return ScopedCallback to unset the hook
+//		*/
+//		public function setupFakeRevision( $title, $content, $user ) {
+//			$oldCallback = this.setCurrentRevisionCallback(
+//				function (
+//					$titleToCheck, $parser = false ) use ( $title, $content, $user, &$oldCallback
+//				) {
+//					if ( $titleToCheck->equals( $title ) ) {
+//						return new Revision( [
+//							'page' => $title->getArticleID(),
+//							'user_text' => $user->getName(),
+//							'user' => $user->getId(),
+//							'parent_id' => $title->getLatestRevID(),
+//							'title' => $title,
+//							'content' => $content
+//						] );
+//					} else {
+//						return call_user_func( $oldCallback, $titleToCheck, $parser );
+//					}
+//				}
+//			);
+//
+//			global $wgHooks;
+//			$wgHooks['TitleExists'][] =
+//				function ( $titleToCheck, &$exists ) use ( $title ) {
+//					if ( $titleToCheck->equals( $title ) ) {
+//						$exists = true;
+//					}
+//				};
+//			end( $wgHooks['TitleExists'] );
+//			$key = key( $wgHooks['TitleExists'] );
+//			LinkCache::singleton()->clearBadLink( $title->getPrefixedDBkey() );
+//			return new ScopedCallback( function () use ( $title, $key ) {
+//				global $wgHooks;
+//				unset( $wgHooks['TitleExists'][$key] );
+//				LinkCache::singleton()->clearLink( $title );
+//			} );
+//		}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwParserTest.java

@@ -13,3 +13,66 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*;
+public class XomwParserTest {
+	private final    XomwParserFxt fxt = new XomwParserFxt();
+	@Test  public void Basic() {
+		fxt.Test__parse(String_.Concat_lines_nl_skip_last
+		( "== heading_1 =="
+		, "para_1"
+		, "== heading_2 =="
+		, "para_2"
+		, "-----"
+		, "{|"
+		, "|-"
+		, "|a"
+		, "|}"
+		, "''italics''"
+		, "__TOC__"
+		, "[https://a.org b]"
+		, "[[A|abc]]"
+		, "https://c.org"
+		, "a »b«  !important c"
+		), String_.Concat_lines_nl_skip_last
+		( "<h2> heading_1 </h2>"
+		, "<p>para_1"
+		, "</p>"
+		, "<h2> heading_2 </h2>"
+		, "<p>para_2"
+		, "</p>"
+		, "<hr />"
+		, "<table>"
+		, ""
+		, "<tr>"
+		, "<td>a"
+		, "</td></tr></table>"
+		, "<p><i>italics</i>"
+		, "<!--MWTOC-->"
+		, "<a rel=\"nofollow\" class=\"external text\" href=\"https://a.org\">b</a>"
+		, "<a href=\"/wiki/A\" title=\"A\">abc</a>"
+		, "<a rel=\"nofollow\" class=\"external free\" href=\"https://c.org\">https://c.org</a>"
+		, "a&#160;»b«&#160; !important c"
+		, "</p>"
+		));
+	}		
+}
+class XomwParserFxt {
+	private final    XomwParser parser;
+	private final    XomwParserCtx pctx = new XomwParserCtx();
+	private final    XomwParserBfr pbfr = new XomwParserBfr();
+	public XomwParserFxt() {
+		Xoae_app app = Xoa_app_fxt.Make__app__edit();
+		Xowe_wiki wiki = Xoa_app_fxt.Make__wiki__edit(app);
+		this.parser = new XomwParser(XomwEnv.NewTestByApp(app));
+		parser.Init_by_wiki(wiki);
+		parser.Init_by_page(XomwTitle.newFromText(parser.Env(), Bry_.new_a7("Page_1")));
+		pctx.Init_by_page(XomwTitle.newFromText(parser.Env(), Bry_.new_a7("Page_1")));
+	}
+	public void Test__parse(String src_str, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		parser.internalParse(pbfr, pctx, src_bry);
+		parser.internalParseHalfParsed(pbfr, pctx, true, true);
+		Tfds.Eq_str_lines(expd, pbfr.Rslt().To_str_and_clear(), src_str);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwStripState.java

@@ -13,3 +13,334 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+public class XomwStripState {
+//		protected $prefix;
+//		protected $data;
+//		protected $regex;
+//
+//		protected $tempType, $tempMergePrefix;
+//		protected $circularRefGuard;
+//		protected $recursionLevel = 0;
+//
+//		static final UNSTRIP_RECURSION_LIMIT = 20;
+
+	private final    Btrie_slim_mgr trie = Btrie_slim_mgr.cs();
+	private final    Btrie_rv trv = new Btrie_rv();
+	private final    Bry_bfr tmp_1 = Bry_bfr_.New();
+	private final    Bry_bfr tmp_2 = Bry_bfr_.New();
+	private boolean tmp_2_used = false;
+	private int generalLen, nowikiLen;
+
+//		/**
+//		* @param String|null $prefix
+//		* @since 1.26 The prefix argument should be omitted, as the strip marker
+//		*  prefix String is now a constant.
+//		*/
+//		public function __construct($prefix = null) {
+//			if ($prefix !== null) {
+//				wfDeprecated(__METHOD__ . ' with called with $prefix argument' .
+//					' (call with no arguments instead)', '1.26');
+//			}
+//			this.data = [
+//				'nowiki' => [],
+//				'general' => []
+//			];
+//			this.regex = '/' . Parser::MARKER_PREFIX . "([^\x7f<>&'\"]+)" . Parser::MARKER_SUFFIX . '/';
+//			this.circularRefGuard = [];
+//		}
+//		public void Clear() {
+//			trie.Clear();
+//			generalLen = nowikiLen = 0;
+//			tmp_2_used = false;
+//		}
+
+	/**
+	* Add a nowiki strip item
+	* @param String $marker
+	* @param String $value
+	*/
+	public void addNoWiki(byte[] marker, byte[] val) {
+		this.addItem(TYPE_NOWIKI, marker, val);
+	}
+
+	/**
+	* @param String $marker
+	* @param String $value
+	*/
+	public void addGeneral(byte[] marker, byte[] val) {
+		this.addItem(TYPE_GENERAL, marker, val);
+	}
+
+	/**
+	* @throws MWException
+	* @param String $type
+	* @param String $marker
+	* @param String $value
+	*/
+	public void addItem(byte type, byte[] marker, byte[] val) {
+		// if (!preg_match(this.regex, $marker, $m)) {
+		//	throw new MWException("Invalid marker: $marker");
+		// }
+
+		// XO.MW:ported
+		// this.data[$type][$m[1]] = $value;
+		trie.Add_obj(marker, new XomwStripItem(type, marker, val));
+		if (type == TYPE_GENERAL)
+			generalLen++;
+		else
+			nowikiLen++;
+	}
+
+	/**
+	* @param String $text
+	* @return mixed
+	*/
+	public byte[] unstripGeneral(byte[] text) {
+		return this.unstripType(TYPE_GENERAL, text);
+	}
+
+	/**
+	* @param String $text
+	* @return mixed
+	*/
+	public byte[] unstripNoWiki(byte[] text) {
+		return this.unstripType(TYPE_NOWIKI, text);
+	}
+
+	/**
+	* @param String $text
+	* @return mixed
+	*/
+	public byte[] unstripBoth(byte[] text) {
+		//	$text = this.unstripType('general', $text);
+		//	$text = this.unstripType('nowiki', $text);
+		return this.unstripType(TYPE_BOTH, text);
+	}
+
+	public byte[] unstripType(byte tid, byte[] text) {
+		boolean dirty = unstripType(tid, tmp_1, text, 0, text.length);
+		return dirty ? tmp_1.To_bry_and_clear() : text;
+	}
+
+	// XOWA
+	public void unstripGeneral(XomwParserBfr pbfr) {unstripType(TYPE_GENERAL, pbfr);}
+	public void unstripNoWiki(XomwParserBfr pbfr)  {unstripType(TYPE_NOWIKI , pbfr);}
+	public void unstripBoth(XomwParserBfr pbfr)    {unstripType(TYPE_BOTH   , pbfr);}
+	private boolean unstripType(byte tid, XomwParserBfr pbfr) {
+		// XO.PBFR
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src = src_bfr.Bfr();
+		boolean dirty = unstripType(tid, pbfr.Trg(), src, 0, src_bfr.Len());
+		if (dirty)
+			pbfr.Switch();
+		return dirty;
+	}
+
+	/**
+	* @param String $type
+	* @param String $text
+	* @return mixed
+	*/
+	private boolean unstripType(byte tid, Bry_bfr trg, byte[] src, int src_bgn, int src_end) {
+		//	// Shortcut
+		//	if (!count(this.data[$type])) {
+		//		return $text;
+		//	}
+		// exit early if no items for type
+		if      ((tid & TYPE_GENERAL) == TYPE_GENERAL) {
+			if (generalLen == 0)
+				return false;
+		}
+		else if ((tid & TYPE_NOWIKI) == TYPE_NOWIKI) {
+			if (nowikiLen == 0)
+				return false;
+		}
+
+		// XO.MW:PORTED
+		//	$oldType = this.tempType;
+		//	this.tempType = $type;
+		//	$text = preg_replace_callback(this.regex, [ $this, 'unstripCallback' ], $text);
+		//	this.tempType = $oldType;
+		//	return $text;
+		int cur = src_bgn;
+		int prv = cur;
+		boolean dirty = false;
+		// loop over each src char
+		while (true) {
+			// EOS: exit
+			if (cur == src_end) {
+				if (dirty)	// add remainder if dirty
+					trg.Add_mid(src, prv, src_end);
+				break;
+			}
+
+			// check if current pos matches strip state
+			Object o = trie.Match_at(trv, src, cur, src_end);
+			if (o != null) {	// match
+				XomwStripItem item = (XomwStripItem)o;
+				byte item_tid = item.Type();
+				if ((tid & item_tid) == item_tid) {	// check if types match
+					// get bfr for recursion
+					Bry_bfr nested_bfr = null;
+					boolean tmp_2_release = false;
+					if (tmp_2_used) {
+						nested_bfr = Bry_bfr_.New();
+					}
+					else {
+						nested_bfr = tmp_2;
+						tmp_2_used = true;
+						tmp_2_release = true;
+					}
+
+					// recurse
+					byte[] item_val = item.Val();
+					if (unstripType(tid, nested_bfr, item_val, 0, item_val.length))
+						item_val = nested_bfr.To_bry_and_clear();
+					if (tmp_2_release)
+						tmp_2_used = false;
+
+					// add to trg
+					trg.Add_mid(src, prv, cur);
+					trg.Add(item_val);
+
+					// update vars
+					dirty = true;
+					cur += item.Key().length;
+					prv = cur;
+					continue;
+				}
+			}
+			cur++;
+		}
+		return dirty;
+	}
+
+	//	/**
+	//	* @param array $m
+	//	* @return array
+	//	*/
+	//	protected function unstripCallback($m) {
+	//		$marker = $m[1];
+	//		if (isset(this.data[this.tempType][$marker])) {
+	//			if (isset(this.circularRefGuard[$marker])) {
+	//				return '<span class="error">'
+	//					. wfMessage('parser-unstrip-loop-warning')->inContentLanguage()->text()
+	//					. '</span>';
+	//			}
+	//			if (this.recursionLevel >= self::UNSTRIP_RECURSION_LIMIT) {
+	//				return '<span class="error">' .
+	//					wfMessage('parser-unstrip-recursion-limit')
+	//						->numParams(self::UNSTRIP_RECURSION_LIMIT)->inContentLanguage()->text() .
+	//					'</span>';
+	//			}
+	//			this.circularRefGuard[$marker] = true;
+	//			this.recursionLevel++;
+	//			$value = this.data[this.tempType][$marker];
+	//			if ($value instanceof Closure) {
+	//				$value = $value();
+	//			}
+	//			$ret = this.unstripType(this.tempType, $value);
+	//			this.recursionLevel--;
+	//			unset(this.circularRefGuard[$marker]);
+	//			return $ret;
+	//		} else {
+	//			return $m[0];
+	//		}
+	//	}
+
+//		/**
+//		* Get a StripState Object which is sufficient to unstrip the given text.
+//		* It will contain the minimum subset of strip items necessary.
+//		*
+//		* @param String $text
+//		*
+//		* @return StripState
+//		*/
+//		public function getSubState($text) {
+//			$subState = new StripState();
+//			$pos = 0;
+//			while (true) {
+//				$startPos = strpos($text, Parser::MARKER_PREFIX, $pos);
+//				$endPos = strpos($text, Parser::MARKER_SUFFIX, $pos);
+//				if ($startPos === false || $endPos === false) {
+//					break;
+//				}
+//
+//				$endPos += strlen(Parser::MARKER_SUFFIX);
+//				$marker = substr($text, $startPos, $endPos - $startPos);
+//				if (!preg_match(this.regex, $marker, $m)) {
+//					continue;
+//				}
+//
+//				$key = $m[1];
+//				if (isset(this.data['nowiki'][$key])) {
+//					$subState->data['nowiki'][$key] = this.data['nowiki'][$key];
+//				} elseif (isset(this.data['general'][$key])) {
+//					$subState->data['general'][$key] = this.data['general'][$key];
+//				}
+//				$pos = $endPos;
+//			}
+//			return $subState;
+//		}
+//
+//		/**
+//		* Merge another StripState Object into this one. The strip marker keys
+//		* will not be preserved. The strings in the $texts array will have their
+//		* strip markers rewritten, the resulting array of strings will be returned.
+//		*
+//		* @param StripState $otherState
+//		* @param array $texts
+//		* @return array
+//		*/
+//		public function merge($otherState, $texts) {
+//			$mergePrefix = wfRandomString(16);
+//
+//			foreach ($otherState->data as $type => $items) {
+//				foreach ($items as $key => $value) {
+//					this.data[$type]["$mergePrefix-$key"] = $value;
+//				}
+//			}
+//
+//			this.tempMergePrefix = $mergePrefix;
+//			$texts = preg_replace_callback($otherState->regex, [ $this, 'mergeCallback' ], $texts);
+//			this.tempMergePrefix = null;
+//			return $texts;
+//		}
+//
+//		/**
+//		* @param array $m
+//		* @return String
+//		*/
+//		protected function mergeCallback($m) {
+//			$key = $m[1];
+//			return Parser::MARKER_PREFIX . this.tempMergePrefix . '-' . $key . Parser::MARKER_SUFFIX;
+//		}
+//
+//		/**
+//		* Remove any strip markers found in the given text.
+//		*
+//		* @param String $text Input String
+//		* @return String
+//		*/
+//		public function killMarkers($text) {
+//			return preg_replace(this.regex, '', $text);
+//		}
+//		public static final    String Str__marker_bgn = "\u007f'\"`UNIQ-";
+//		public static final    byte[] 
+//		  Bry__marker__bgn		= Bry_.new_a7(Str__marker_bgn)
+//		, Bry__marker__end		= Bry_.new_a7("-QINU`\"'\u007f")
+//		;
+	public static final byte TYPE_GENERAL = 1, TYPE_NOWIKI = 2, TYPE_BOTH = 3; 
+}
+class XomwStripItem {
+	public XomwStripItem(byte tid, byte[] key, byte[] val) {
+		this.tid = tid;
+		this.key = key;
+		this.val = val;
+	}
+	public byte Type() {return tid;} private final    byte tid;
+	public byte[] Key() {return key;} private final    byte[] key;
+	public byte[] Val() {return val;} private final    byte[] val;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/XomwStripStateTest.java

@@ -13,3 +13,30 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+public class XomwStripStateTest {
+	private final    XomwStripStateFxt fxt = new XomwStripStateFxt();
+	@Test   public void Basic() {
+		fxt.Init__add    (XomwStripState.TYPE_GENERAL, "\u007f'\"`UNIQ-key-1-QINU`\"'\u007f", "val-1");
+		fxt.Test__nostrip(XomwStripState.TYPE_NOWIKI , "a \u007f'\"`UNIQ-key-1-QINU`\"'\u007f b");
+		fxt.Test__unstrip(XomwStripState.TYPE_GENERAL, "a \u007f'\"`UNIQ-key-1-QINU`\"'\u007f b", "a val-1 b");
+		fxt.Test__unstrip(XomwStripState.TYPE_BOTH   , "a \u007f'\"`UNIQ-key-1-QINU`\"'\u007f b", "a val-1 b");
+	}
+	@Test   public void Recurse() {
+		fxt.Init__add    (XomwStripState.TYPE_GENERAL, "\u007f'\"`UNIQ-key-1-QINU`\"'\u007f", "val-1");
+		fxt.Init__add    (XomwStripState.TYPE_GENERAL, "\u007f'\"`UNIQ-key-2-QINU`\"'\u007f", "\u007f'\"`UNIQ-key-1-QINU`\"'\u007f");
+		fxt.Test__unstrip(XomwStripState.TYPE_GENERAL, "a \u007f'\"`UNIQ-key-2-QINU`\"'\u007f b", "a val-1 b");
+	}
+}
+class XomwStripStateFxt {
+	private final    XomwStripState stripState = new XomwStripState();
+	public void Init__add(byte tid, String marker, String val) {
+		stripState.addItem(tid, Bry_.new_u8(marker), Bry_.new_u8(val));
+	}
+	public void Test__nostrip(byte tid, String src) {Test__unstrip(tid, src, src);}
+	public void Test__unstrip(byte tid, String src, String expd) {
+		byte[] actl = stripState.unstripType(tid, Bry_.new_u8(src));
+		Gftest.Eq__str(expd, String_.new_u8(actl));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_output_type.java

@@ -13,3 +13,13 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class Xomw_output_type {
+	public static final byte 
+	  Tid__html       = 1   // like parse()
+	, Tid__wiki       = 2   // like preSaveTransform()
+	, Tid__preprocess = 3   // like preprocess()
+	, Tid__msg        = 3
+	, Tid__plain      = 4   // like extractSections() - portions of the original are returned unchanged.
+	;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_.java

@@ -13,3 +13,31 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+public class Xomw_regex_ {
+	public static int Find_fwd_while(Btrie_slim_mgr trie, Btrie_rv trv, byte[] src, int src_bgn, int src_end) {
+		int cur = src_bgn;
+		while (cur < src_end) {
+			byte b = src[cur];
+			Object o = trie.Match_at_w_b0(trv, b, src, cur, src_end);
+			if (o == null)
+				break;
+			else
+				cur += gplx.core.intls.Utf8_.Len_of_char_by_1st_byte(b);
+		}
+		return cur;
+	}
+	public static int Find_fwd_until(Btrie_slim_mgr trie, Btrie_rv trv, byte[] src, int src_bgn, int src_end) {
+		int cur = src_bgn;
+		while (cur < src_end) {
+			byte b = src[cur];
+			Object o = trie.Match_at_w_b0(trv, b, src, cur, src_end);
+			if (o == null)
+				cur += gplx.core.intls.Utf8_.Len_of_char_by_1st_byte(b);
+			else
+				break;
+		}
+		return cur;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_boundary.java

@@ -13,3 +13,25 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+public class Xomw_regex_boundary {	// THREAD.SAFE: trv is only for consistent interface
+	private final    Btrie_slim_mgr trie = Btrie_slim_mgr.cs();
+	private final    Btrie_rv trv = new Btrie_rv();
+	public Xomw_regex_boundary(Xomw_regex_space space) {
+		// naive implementation of is_boundary; ignore all ws and underscore
+		byte[][] ary = space.Ws();
+		for (byte[] bry : ary)
+			trie.Add_bry_byte(bry, Byte_.Zero);
+		ary = space.Zs();
+		for (byte[] bry : ary)
+			trie.Add_bry_byte(bry, Byte_.Zero);
+	}
+	public boolean Is_boundary_prv(byte[] src, int pos) {
+		if (pos == 0) return true; // BOS is true
+		int bgn = gplx.core.intls.Utf8_.Get_prv_char_pos0(src, pos);
+		byte b = src[bgn];
+		Object o = trie.Match_at_w_b0(trv, b, src, bgn, pos);
+		return o != null;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_parser.java

@@ -13,3 +13,87 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+public class Xomw_regex_parser {
+	private Bry_bfr tmp;	
+	public byte[][] Rslt() {return rslt;} private byte[][] rslt;
+	public Xomw_regex_parser Add_ary(String... ary) {return Set_or_add(Parse_ary(ary));}
+	private byte[][] Parse_ary(String... ary) {
+		if (tmp == null) tmp = Bry_bfr_.New();
+		int ary_len = ary.length;
+		byte[][] rv = new byte[ary_len][];
+		for (int i = 0; i < ary_len; i++) {
+			rv[i] = Compile_itm(tmp, Bry_.new_u8(ary[i]));
+		}
+		return rv;
+	}
+	public Xomw_regex_parser Add_rng(String bgn, String end) {return Set_or_add(Parse_rng(bgn, end));}
+	private byte[][] Parse_rng(String bgn, String end) {
+		if (tmp == null) tmp = Bry_bfr_.New();
+		byte[] bgn_bry = Compile_itm(tmp, Bry_.new_u8(bgn));
+		int bgn_val = gplx.core.intls.Utf16_.Decode_to_int(bgn_bry, 0);
+		byte[] end_bry = Compile_itm(tmp, Bry_.new_u8(end));
+		int end_val = gplx.core.intls.Utf16_.Decode_to_int(end_bry, 0);
+
+		int rv_len = end_val - bgn_val + 1;
+		byte[][] rv = new byte[rv_len][];
+		for (int i = 0; i < rv_len; i++) {
+			rv[i] = gplx.core.intls.Utf16_.Encode_int_to_bry(i + bgn_val);
+		}
+		return rv;
+	}
+	private Xomw_regex_parser Set_or_add(byte[][] val) {
+		rslt = rslt == null ? val : Bry_.Ary_add(rslt, val);
+		return this;
+	}
+	private static byte[] Compile_itm(Bry_bfr tmp, byte[] src) {
+		// parse each itm
+		int src_end = src.length;
+		int cur = 0;
+		int prv = cur;
+		boolean dirty = false;
+		while (true) {
+			// eos
+			if (cur == src_end) {
+				if (dirty)
+					tmp.Add_mid(src, prv, src_end);
+				break;
+			}
+
+			// look at byte
+			byte b = src[cur];
+			switch (b) {	// escape
+				case Byte_ascii.Backslash:
+					int nxt = cur + 1;
+					if (nxt >= src_end) throw Err_.new_wo_type("regex escape failed: no more chars left", "src", src, "pos", nxt);
+					byte nxt_byte = src[nxt];
+					switch (nxt_byte) {
+						case Byte_ascii.Ltr_s: // \s -> " "
+							src = Byte_ascii.Space_bry;
+							cur = src_end;
+							break;
+						case Byte_ascii.Ltr_x: // \ u -> utf8 sequence in hex-dec; EX: "\xc2\xad" -> new byte[] {194, 160}
+							// read next two bytes
+							dirty = true;
+							nxt++;
+							if (nxt + 2 > src_end) throw Err_.new_wo_type("utf8 escape failed: no more chars left", "src", src, "pos", nxt);
+							tmp.Add_byte((byte)gplx.core.encoders.Hex_utl_.Parse_or(src, nxt, nxt + 2, -1));
+							cur = nxt + 2;
+							prv = cur;
+							break;
+						default:
+							throw Err_.new_wo_type("regex escape failed: unknown char", "src", src, "pos", nxt);
+					}
+					break;
+				default: // handles ascii only
+					if (b > 127)
+						throw Err_.new_wo_type("regex compiled failed: unknown char", "src", src, "pos", cur);
+					cur++;
+					break;
+			}
+		}
+
+		// set item
+		return dirty ? tmp.To_bry_and_clear() : src;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_parser__tst.java

@@ -13,3 +13,28 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import org.junit.*; import gplx.core.tests.*;
+public class Xomw_regex_parser__tst {
+	private final    Xomw_regex_parser__fxt fxt = new Xomw_regex_parser__fxt();
+	@Test   public void Ary__space() {
+		fxt.Test__parse_ary(String_.Ary("\\s"), String_.Ary(" "));
+	}
+	@Test   public void Ary__utf8() {
+		fxt.Test__parse_ary(String_.Ary("\\xc2\\xa7", "\\xe0\\xb9\\x90"), String_.Ary("§", "๐"));
+	}
+	@Test   public void Rng__ascii() {
+		fxt.Test__parse_rng("a", "c", String_.Ary("a", "b", "c"));
+	}
+}
+class Xomw_regex_parser__fxt {
+	private final    Xomw_regex_parser parser = new Xomw_regex_parser();
+	public void Test__parse_ary(String[] ary, String[] expd) {
+		parser.Add_ary(ary);
+		Gftest.Eq__ary(expd, String_.Ary(parser.Rslt()));
+	}
+	public void Test__parse_rng(String bgn, String end, String[] expd) {
+		parser.Add_rng("a", "c");
+		Gftest.Eq__ary(expd, String_.Ary(parser.Rslt()));
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_space.java

@@ -13,3 +13,50 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+public class Xomw_regex_space {
+	private final    Btrie_slim_mgr trie = Btrie_slim_mgr.cs();
+	public Xomw_regex_space() {
+		byte[] space = Bry_.New_by_ints(32);
+		ws = new byte[][]
+		{ space
+		, Bry_.New_by_ints(9)
+		, Bry_.New_by_ints(10)
+		, Bry_.New_by_ints(13)
+		};
+		// Zs; REF:http://www.fileformat.info/info/unicode/category/Zs/list.htm
+		zs = new byte[][]
+		{ space
+		, Bry_.New_by_ints(194, 160)
+		, Bry_.New_by_ints(225, 154, 128)
+		, Bry_.New_by_ints(226, 128, 129)
+		, Bry_.New_by_ints(226, 128, 130)
+		, Bry_.New_by_ints(226, 128, 131)
+		, Bry_.New_by_ints(226, 128, 132)
+		, Bry_.New_by_ints(226, 128, 133)
+		, Bry_.New_by_ints(226, 128, 134)
+		, Bry_.New_by_ints(226, 128, 135)
+		, Bry_.New_by_ints(226, 128, 136)
+		, Bry_.New_by_ints(226, 128, 137)
+		, Bry_.New_by_ints(226, 128, 138)
+		, Bry_.New_by_ints(226, 128, 175)
+		, Bry_.New_by_ints(226, 129, 159)
+		, Bry_.New_by_ints(227, 128, 128)
+		};
+
+		byte[][] ary = ws;
+		for (byte[] bry : ary) {
+			trie.Add_bry_byte(bry, Byte_.Zero);
+		}
+		ary = zs;
+		for (byte[] bry : ary) {
+			trie.Add_bry_byte(bry, Byte_.Zero);
+		}
+	}
+	public byte[][] Ws() {return ws;} private byte[][] ws;
+	public byte[][] Zs() {return zs;} private byte[][] zs;
+	public int Find_fwd_while(Btrie_rv trv, byte[] src, int src_bgn, int src_end) {
+		return Xomw_regex_.Find_fwd_while(trie, trv, src, src_bgn, src_end);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/Xomw_regex_url.java

@@ -13,3 +13,26 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*;
+import gplx.core.btries.*;
+public class Xomw_regex_url {
+	private final    Btrie_slim_mgr trie;
+	public Xomw_regex_url(Xomw_regex_space regex_space) {
+		//       [^][<>"\\x00-\\x20\\x7F\|]
+		// REGEX:[^][<>"\\x00-\\x20\\x7F\p{Zs}]; NOTE: val is just a marker
+		this.trie = Btrie_slim_mgr.cs();
+		trie.Add_str_byte__many(Byte_.Zero, "[", "]", "<", ">", "\"");
+		for (byte i = 0; i < 33; i++) {
+			trie.Add_bry_byte(new byte[] {i}, Byte_.Zero);
+		}
+		trie.Add_bry_byte(Bry_.New_by_ints(127), Byte_.Zero);	// x7F
+
+		byte[][] zs_ary = regex_space.Zs();
+		for (byte[] zs : zs_ary) {
+			trie.Add_bry_byte(zs, Byte_.Zero);
+		}
+	}
+	public int Find_fwd_while(Btrie_rv trv, byte[] src, int src_bgn, int src_end) {
+		return Xomw_regex_.Find_fwd_until(trie, trv, src, src_bgn, src_end);
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/doubleunders/Xomw_doubleunder_data.java

@@ -13,3 +13,42 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.doubleunders; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+public class Xomw_doubleunder_data {
+	// XO.MW: MW stores these as mDoubleUnderscores in Parser
+	public boolean toc;
+	public boolean no_toc;
+	public boolean force_toc;
+
+	public boolean no_gallery;
+	public boolean force_gallery;
+
+	public boolean no_title_convert;
+	public boolean no_content_convert;
+
+	public boolean no_edit_section;
+	public boolean new_section_link;
+
+	public boolean static_redirect;
+
+	public boolean hidden_cat;
+
+	public boolean index;
+	public boolean no_index;
+
+	// XO.MW: MW stores these as member variables in Parser
+	public boolean show_toc;
+	public boolean force_toc_position;
+
+	public void Reset() {
+		toc = no_toc = force_toc =
+		no_gallery = force_gallery = 
+		no_title_convert = no_content_convert = 
+		no_edit_section = new_section_link = 
+		static_redirect = 
+		hidden_cat = index = no_index =
+			false;
+
+		show_toc = force_toc_position = false;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/doubleunders/Xomw_doubleunder_wkr.java

@@ -13,3 +13,190 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.doubleunders; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+import gplx.core.btries.*;
+import gplx.xowa.langs.*; import gplx.xowa.langs.kwds.*;
+public class Xomw_doubleunder_wkr {
+	private final    Btrie_slim_mgr trie = Btrie_slim_mgr.ci_u8();
+	private final    Btrie_rv trv = new Btrie_rv();
+	private Xomw_doubleunder_data data;
+	public void Init_by_wiki(Xomw_doubleunder_data data, Xol_lang_itm lang) {
+		this.data = data;
+		Reg(trie, lang.Kwd_mgr()
+		, Xol_kwd_grp_.Id_notoc
+		, Xol_kwd_grp_.Id_nogallery
+		, Xol_kwd_grp_.Id_forcetoc
+		, Xol_kwd_grp_.Id_toc
+		, Xol_kwd_grp_.Id_noeditsection
+		, Xol_kwd_grp_.Id_newsectionlink
+		, Xol_kwd_grp_.Id_hiddencat
+		, Xol_kwd_grp_.Id_index
+		, Xol_kwd_grp_.Id_noindex
+		, Xol_kwd_grp_.Id_staticredirect
+		, Xol_kwd_grp_.Id_notitleconvert
+		, Xol_kwd_grp_.Id_nocontentconvert
+		);
+	}
+	public void doDoubleUnderscore(XomwParserCtx pctx, XomwParserBfr pbfr) {
+		// XO.PBFR
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src = src_bfr.Bfr();
+		int src_bgn = 0;
+		int src_end = src_bfr.Len();
+		Bry_bfr bfr = pbfr.Trg();
+
+		data.Reset();
+
+		// XO.MW: MW does TOC before others; XO does it at the same time
+		// Now match and remove the rest of them
+		// XO.MW.BGN: $this->mDoubleUnderscores = $mwa->matchAndRemove( $text );
+		int cur = src_bgn;
+		int prv = cur;
+		boolean dirty = false;
+		while (true) {
+			// reached end; stop
+			if (cur == src_end) {
+				if (dirty) {
+					bfr.Add_mid(src, prv, src_end);
+				}
+				break;
+			}
+
+			// no match; keep searching
+			byte b = src[cur];
+			Object o = trie.Match_at_w_b0(trv, b, src, cur, src_end);
+			if (o == null) {
+				cur += gplx.core.intls.Utf8_.Len_of_char_by_1st_byte(b);
+				continue;
+			}
+
+			// if cs, ensure exact-match (trie is case-insensitive)
+			int kwd_end = trv.Pos();
+			Xomw_doubleunder_itm itm = (Xomw_doubleunder_itm)o;
+			if (itm.case_match && !Bry_.Match(src, cur, kwd_end, itm.val)) {
+				cur = kwd_end;
+				continue;
+			}
+
+			// match; replace __KWD__ with "" (or "<!--MWTOC-->" if __TOC__)
+			dirty = true;
+			bfr.Add_mid(src, prv, cur);
+			switch (itm.tid) {
+				case Xol_kwd_grp_.Id_toc:
+					// The position of __TOC__ needs to be recorded
+					boolean already_seen = !data.show_toc;
+					data.toc = true;
+					data.show_toc = true;
+					data.force_toc_position = true;
+
+					if (already_seen) { // Set a placeholder. At the end we'll fill it in with the TOC.
+						bfr.Add_str_a7("<!--MWTOC-->");
+					}
+					else { // Only keep the first one. XO.MW:ignore by not adding anything to bfr
+					}
+					break;
+				// XO.MW: MW adds boolean to hash_table; XO uses boolean props; note that "remove" is done by not adding to bfr
+				case Xol_kwd_grp_.Id_notoc:                data.no_toc = true; break;
+				case Xol_kwd_grp_.Id_nogallery:            data.no_gallery = true; break;
+				case Xol_kwd_grp_.Id_forcetoc:             data.force_toc = true; break;
+				case Xol_kwd_grp_.Id_noeditsection:        data.no_edit_section = true; break;
+				case Xol_kwd_grp_.Id_newsectionlink:       data.new_section_link = true; break;
+				case Xol_kwd_grp_.Id_hiddencat:            data.hidden_cat = true; break;
+				case Xol_kwd_grp_.Id_index:                data.index = true; break;
+				case Xol_kwd_grp_.Id_noindex:              data.no_index = true; break;
+				case Xol_kwd_grp_.Id_staticredirect:       data.static_redirect = true; break;
+				case Xol_kwd_grp_.Id_notitleconvert:       data.no_title_convert = true; break;
+				case Xol_kwd_grp_.Id_nocontentconvert:     data.no_content_convert = true; break;
+				default:                                   throw Err_.new_unhandled_default(itm.tid);
+			}
+			cur = kwd_end;
+			prv = cur;
+		}
+		// XO.MW.END: $this->mDoubleUnderscores = $mwa->matchAndRemove( $text );
+
+		if (data.no_toc && !data.force_toc_position) {
+			data.show_toc = false;
+		}
+
+		// XO.MW.EDIT: hidden_cat, index, noindex are used to add to tracking category
+
+		if (dirty)
+			pbfr.Switch();
+	}
+//		/**
+//		* Strip double-underscore items like __NOGALLERY__ and __NOTOC__
+//		* Fills this.mDoubleUnderscores, returns the modified text
+//		*
+//		* @param String $text
+//		*
+//		* @return String
+//		*/
+//		public function doDoubleUnderscore($text) {
+//
+//			# The position of __TOC__ needs to be recorded
+//			$mw = MagicWord::get('toc');
+//			if ($mw->match($text)) {
+//				this.mShowToc = true;
+//				this.mForceTocPosition = true;
+//
+//				# Set a placeholder. At the end we'll fill it in with the TOC.
+//				$text = $mw->replace('<!--MWTOC-->', $text, 1);
+//
+//				# Only keep the first one.
+//				$text = $mw->replace('', $text);
+//			}
+//
+//			# Now match and remove the rest of them
+//			$mwa = MagicWord::getDoubleUnderscoreArray();
+//			this.mDoubleUnderscores = $mwa->matchAndRemove($text);
+//
+//			if (isset(this.mDoubleUnderscores['nogallery'])) {
+//				this.mOutput->mNoGallery = true;
+//			}
+//			if (isset(this.mDoubleUnderscores['notoc']) && !this.mForceTocPosition) {
+//				this.mShowToc = false;
+//			}
+//			if (isset(this.mDoubleUnderscores['hiddencat'])
+//				&& this.mTitle->getNamespace() == NS_CATEGORY
+//			) {
+//				this.addTrackingCategory('hidden-category-category');
+//			}
+//			# (T10068) Allow control over whether robots index a page.
+//			# __INDEX__ always overrides __NOINDEX__, see T16899
+//			if (isset(this.mDoubleUnderscores['noindex']) && this.mTitle->canUseNoindex()) {
+//				this.mOutput->setIndexPolicy('noindex');
+//				this.addTrackingCategory('noindex-category');
+//			}
+//			if (isset(this.mDoubleUnderscores['index']) && this.mTitle->canUseNoindex()) {
+//				this.mOutput->setIndexPolicy('index');
+//				this.addTrackingCategory('index-category');
+//			}
+//
+//			# Cache all double underscores in the database
+//			foreach (this.mDoubleUnderscores as $key => $val) {
+//				this.mOutput->setProperty($key, '');
+//			}
+//
+//			return $text;
+//		}
+	private static void Reg(Btrie_slim_mgr trie, Xol_kwd_mgr mgr, int... ids) {
+		for (int id : ids) {
+			Xol_kwd_grp grp = mgr.Get_or_new(id);
+			Xol_kwd_itm[] itms = grp.Itms();
+			for (Xol_kwd_itm itm : itms) {
+				byte[] val = itm.Val();
+				trie.Add_obj(val, new Xomw_doubleunder_itm(id, grp.Case_match(), val));
+			}
+		}
+	}
+}
+class Xomw_doubleunder_itm {
+	public int tid;
+	public boolean case_match;
+	public byte[] val;
+	public Xomw_doubleunder_itm(int tid, boolean case_match, byte[] val) {
+		this.tid = tid;
+		this.case_match = case_match;
+		this.val = val;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/doubleunders/Xomw_doubleunder_wkr__tst.java

@@ -13,3 +13,38 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.doubleunders; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+import org.junit.*; import gplx.core.tests.*;
+public class Xomw_doubleunder_wkr__tst {
+	private final    Xomw_doubleunder_wkr__fxt fxt = new Xomw_doubleunder_wkr__fxt();
+	@Test  public void No_match()        {fxt.Test__parse("a b c"                                    , "a b c");}
+	@Test  public void Force_toc()       {fxt.Test__parse("a __FORCETOC__ b"                         , "a  b").Test__prop_y(fxt.data.force_toc);}
+	@Test  public void Toc()             {fxt.Test__parse("a __TOC__ b __TOC__ c"                    , "a <!--MWTOC--> b  c").Test__prop_y(fxt.data.toc, fxt.data.show_toc, fxt.data.force_toc_position);}
+	@Test  public void Notoc_only()      {fxt.Test__parse("a __NOTOC__ b"                            , "a  b").Test__prop_y(fxt.data.no_toc).Test__prop_n(fxt.data.show_toc);}	// show_toc is false
+	@Test  public void Notoc_w_toc()     {fxt.Test__parse("a __TOC__ b __NOTOC__ c"                  , "a <!--MWTOC--> b  c").Test__prop_y(fxt.data.toc, fxt.data.show_toc, fxt.data.force_toc_position);} // show_toc is true
+	@Test  public void Case_match()      {fxt.Test__parse("a __index__ b"                            , "a __index__ b");}
+}
+class Xomw_doubleunder_wkr__fxt {
+	private final    XomwParserCtx pctx = new XomwParserCtx();
+	private final    XomwParserBfr pbfr = new XomwParserBfr();
+	private final    Xomw_doubleunder_wkr wkr = new Xomw_doubleunder_wkr();
+	public Xomw_doubleunder_data data = new Xomw_doubleunder_data();
+	public Xomw_doubleunder_wkr__fxt() {
+		Xoae_app app = Xoa_app_fxt.Make__app__edit();
+		Xowe_wiki wiki = Xoa_app_fxt.Make__wiki__edit(app);
+		wkr.Init_by_wiki(data, wiki.Lang());
+	}
+	public Xomw_doubleunder_wkr__fxt Test__parse(String src_str, String expd) {
+		byte[] src_bry = Bry_.new_u8(src_str);
+		wkr.doDoubleUnderscore(pctx, pbfr.Init(src_bry));
+		Gftest.Eq__str(expd, pbfr.Rslt().To_str_and_clear(), src_str);
+		return this;
+	}
+	public  Xomw_doubleunder_wkr__fxt Test__prop_y(boolean... ary) {return Test__prop(Bool_.Y, ary);}
+	public  Xomw_doubleunder_wkr__fxt Test__prop_n(boolean... ary) {return Test__prop(Bool_.N, ary);}
+	private Xomw_doubleunder_wkr__fxt Test__prop(boolean expd, boolean... ary) {
+		for (boolean v : ary)
+			Gftest.Eq__bool(expd, v);
+		return this;
+	}
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/headings/Xomw_heading_cbk.java

@@ -13,3 +13,8 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.headings; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+public interface Xomw_heading_cbk {
+	void On_hdr_seen(XomwParserCtx pctx, Xomw_heading_wkr wkr);
+	void On_src_done(XomwParserCtx pctx, Xomw_heading_wkr wkr);
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/headings/Xomw_heading_cbk__html.java

@@ -13,3 +13,38 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.headings; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+public class Xomw_heading_cbk__html implements Xomw_heading_cbk {		
+	public Bry_bfr Bfr() {return bfr;} private Bry_bfr bfr; 
+	public Xomw_heading_cbk__html Bfr_(Bry_bfr bfr) {
+		this.bfr = bfr;
+		return this;
+	}
+	public void On_hdr_seen(XomwParserCtx pctx, Xomw_heading_wkr wkr) {
+		// add from txt_bgn to hdr_bgn; EX: "abc\n==A==\n"; "\n==" seen -> add "abc"
+		byte[] src = wkr.Src();
+		int hdr_bgn = wkr.Hdr_bgn(), txt_bgn = wkr.Txt_bgn();
+		if (hdr_bgn > txt_bgn)
+			bfr.Add_mid(src, txt_bgn, hdr_bgn);
+
+		// add "\n" unless BOS
+		if (hdr_bgn != XomwParserCtx.Pos__bos) bfr.Add_byte_nl();
+
+		// add <h2>...</h2>
+		int hdr_num = wkr.Hdr_num();
+		bfr.Add(Tag__lhs).Add_int_digits(1, hdr_num).Add(Byte_ascii.Angle_end_bry);	// <h2>
+		bfr.Add_mid(wkr.Src(), wkr.Hdr_lhs_end(), wkr.Hdr_rhs_bgn());
+		bfr.Add(Tag__rhs).Add_int_digits(1, hdr_num).Add(Byte_ascii.Angle_end_bry);	// </h2>
+	}
+	public void On_src_done(XomwParserCtx pctx, Xomw_heading_wkr wkr) {
+		// add from txt_bgn to EOS;
+		byte[] src = wkr.Src();
+		int txt_bgn = wkr.Txt_bgn(), src_end = wkr.Src_end();
+		if (txt_bgn != src_end)	// PERF: don't call Add_mid() if hdr is at end of EOS
+			bfr.Add_mid(src, txt_bgn, src_end);
+	}
+	private static final    byte[] 
+	  Tag__lhs = Bry_.new_a7("<h")
+	, Tag__rhs = Bry_.new_a7("</h")
+	;
+}

gplx.xowa.mediawiki/src/gplx/xowa/mediawiki/includes/parsers/headings/Xomw_heading_wkr.java

@@ -13,3 +13,110 @@ The terms of each license can be found in the source code repository:
 GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
 Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 */
+package gplx.xowa.mediawiki.includes.parsers.headings; import gplx.*; import gplx.xowa.*; import gplx.xowa.mediawiki.*; import gplx.xowa.mediawiki.includes.*; import gplx.xowa.mediawiki.includes.parsers.*;
+import gplx.core.btries.*; import gplx.xowa.langs.*;
+public class Xomw_heading_wkr {
+	private XomwParserCtx pctx;
+	private Xomw_heading_cbk cbk;
+	public byte[] Src()			{return src;} private byte[] src;
+	public int Src_end()		{return src_end;} private int src_end;
+	public int Txt_bgn()		{return txt_bgn;} private int txt_bgn;
+	public int Hdr_bgn()		{return hdr_bgn;} private int hdr_bgn;
+	public int Hdr_end()		{return hdr_end;} private int hdr_end;
+	public int Hdr_num()		{return hdr_num;} private int hdr_num;
+	public int Hdr_lhs_bgn()	{return hdr_lhs_bgn;} private int hdr_lhs_bgn;
+	public int Hdr_lhs_end()	{return hdr_lhs_end;} private int hdr_lhs_end;
+	public int Hdr_rhs_bgn()	{return hdr_rhs_bgn;} private int hdr_rhs_bgn;
+	public int Hdr_rhs_end()	{return hdr_rhs_end;} private int hdr_rhs_end;
+	public void doHeadings(XomwParserCtx pctx, XomwParserBfr pbfr, Xomw_heading_cbk__html cbk) {
+		Bry_bfr src_bfr = pbfr.Src();
+		byte[] src_bry = src_bfr.Bfr();
+		int src_end = src_bfr.Len();
+		cbk.Bfr_(pbfr.Trg());
+		pbfr.Switch();
+		Parse(pctx, src_bry, 0, src_end, cbk);
+	}
+	public void Parse(XomwParserCtx pctx, byte[] src, int src_bgn, int src_end, Xomw_heading_cbk cbk) {	// REF.MW: /includes/parser/Parser.php|doHeadings
+		// init members
+		this.pctx = pctx;
+		this.src = src;
+		this.src_end = src_end;
+		this.cbk = cbk;
+
+		// PORTED:
+		// for ( $i = 6; $i >= 1; --$i ) {
+		//   $h = str_repeat( '=', $i );
+		//   $text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
+		// }
+
+		// do loop
+		int pos = src_bgn;
+		this.txt_bgn = pos == XomwParserCtx.Pos__bos ? 0 : pos;
+		byte b = Byte_ascii.Nl;
+		while (true) {
+			int nxt = pos + 1;
+			// check if (a) cur is \n; (b) nxt is '='
+			if (	b == Byte_ascii.Nl
+				&&	nxt < src_end
+				&&	src[nxt] == Byte_ascii.Eq
+				) {
+				pos = Parse_hdr_nl(txt_bgn, pos, nxt + 1);
+				this.txt_bgn = pos;
+			}
+			else
+				++pos;
+
+			// EOS; add all text after last "==\n"
+			if (pos == src_end) {
+				cbk.On_src_done(pctx, this);
+				break;
+			}
+			b = src[pos];
+		}
+	}
+	private int Parse_hdr_nl(int txt_bgn, int nl_lhs, int pos) {
+		// calc lhs vars
+		this.hdr_bgn = nl_lhs;
+		this.hdr_lhs_bgn = nl_lhs == 0 ? 0 : nl_lhs + 1;	// set pos of 1st "="; note that "==" can be at BOS;
+		this.hdr_lhs_end = Bry_find_.Find_fwd_while(src, pos, src_end, Byte_ascii.Eq);
+
+		// calc rhs vars
+		int nl_rhs = Bry_find_.Find_fwd_or(src, Byte_ascii.Nl, hdr_lhs_end + 1, src_end, src_end);	// if no "\n", src_end is rest of text; EX: "\n==<text>EOS
+		this.hdr_end = nl_rhs;
+		this.hdr_rhs_end = Bry_find_.Find_bwd__skip_ws(src, nl_rhs, hdr_lhs_end);
+		this.hdr_rhs_bgn = Bry_find_.Find_bwd__skip(src, hdr_rhs_end - 1, hdr_lhs_end, Byte_ascii.Eq);
+
+		int hdr_lhs_len = hdr_lhs_end - hdr_lhs_bgn;
+		int hdr_rhs_len = hdr_rhs_end - hdr_rhs_bgn;
+
+		// handle rare situations like "\n====\n"
+		if (hdr_rhs_len == 0) {
+			int hdr_lhs_len_half = hdr_lhs_len / 2;
+			hdr_rhs_len = hdr_lhs_len - hdr_lhs_len_half;
+			hdr_lhs_len = hdr_lhs_len_half;
+			this.hdr_lhs_end = hdr_lhs_bgn + hdr_lhs_len;
+			this.hdr_rhs_bgn = hdr_lhs_end;
+		}
+
+		this.hdr_num = hdr_lhs_len < hdr_rhs_len ? hdr_lhs_len : hdr_rhs_len;
+
+		cbk.On_hdr_seen(pctx, this);
+		return nl_rhs;
+	}
+//		/**
+//		* Parse headers and return html
+//		*
+//		* @private
+//		*
+//		* @param String $text
+//		*
+//		* @return String
+//		*/
+//		public function doHeadings($text) {
+//			for ($i = 6; $i >= 1; --$i) {
+//				$h = str_repeat('=', $i);
+//				$text = preg_replace("/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text);
+//			}
+//			return $text;
+//		}
+}