Scribunto.Regex: Add only first match for mw.ustring.match, not all of them

100_core/src/gplx/langs/regxs/Regx_adp.java

@@ -20,6 +20,7 @@ public class Regx_adp {
 	@gplx.Internal protected Regx_adp(String regx) {Pattern_(regx);}
 	public String Pattern() {return pattern;} public Regx_adp Pattern_(String val) {pattern = val; Under_sync(); return this;} private String pattern;
 	public boolean Pattern_is_invalid() {return pattern_is_invalid;} private boolean pattern_is_invalid = false;
+	public Exception Pattern_is_invalid_exception() {return pattern_is_invalid_exception;} private Exception pattern_is_invalid_exception = null;
 	public Regx_match[] Match_all(String text, int bgn) {
 		int idx = bgn;
 		List_adp rv = List_adp_.New();
@@ -39,10 +40,11 @@ public class Regx_adp {
 	}
 		private Pattern under;
 	public Pattern Under() {return under;}
-	void Under_sync() {
+	private void Under_sync() {
 		try {under = Pattern.compile(pattern, Pattern.DOTALL | Pattern.UNICODE_CHARACTER_CLASS);}	// JRE.7:UNICODE_CHARACTER_CLASS; added during %w fix for en.w:A#; DATE:2015-06-10 
 		catch (Exception e) {	// NOTE: if invalid, then default to empty pattern (which should return nothing); EX:d:〆る generates [^]; DATE:2013-10-20
 			pattern_is_invalid = true;
+			pattern_is_invalid_exception = e;
 			under = Pattern.compile("", Pattern.DOTALL | Pattern.UNICODE_CHARACTER_CLASS);
 		}
 	}

400_xowa/src/gplx/xowa/xtns/scribunto/libs/Scrib_lib_ustring.java

@@ -107,11 +107,10 @@ public class Scrib_lib_ustring implements Scrib_lib {
 		Regx_match[] regx_rslts = regx_adp.Match_all(text, bgn);
 		int len = regx_rslts.length;
 		if (len == 0) return rslt.Init_null();	// return null if no matches found; EX:w:Mount_Gambier_(volcano); DATE:2014-04-02; confirmed with en.d:民; DATE:2015-01-30
+
+		// TOMBSTONE: add 1st match only; do not add all; PAGE:en.d:действительное_причастие_настоящего_времени DATE:2017-04-23
 		List_adp tmp_list = List_adp_.New();
-		for (int i = 0; i < len; i++) {
-			Regx_match match = regx_rslts[i];
-			AddCapturesFromMatch(tmp_list, match, text, regx_converter.Capt_ary(), true);
-		}
+		AddCapturesFromMatch(tmp_list, regx_rslts[0], text, regx_converter.Capt_ary(), true);
 		return rslt.Init_many_list(tmp_list);
 	}
 	private Scrib_lib_ustring_gsub_mgr[] gsub_mgr_ary = Scrib_lib_ustring_gsub_mgr.Ary_empty;
@@ -180,7 +179,9 @@ public class Scrib_lib_ustring implements Scrib_lib {
 	public static Regx_adp RegxAdp_new_(Xop_ctx ctx, String regx) {
 		Regx_adp rv = Regx_adp_.new_(regx);
 		if (rv.Pattern_is_invalid()) {
-			ctx.App().Usr_dlg().Warn_many("", "", "regx is invalid: regx=~{0} page=~{1}", regx, String_.new_u8(ctx.Page().Ttl().Page_db()));
+			// try to identify [z-a] errors; PAGE:https://en.wiktionary.org/wiki/Module:scripts/data;  DATE:2017-04-23
+			Exception exc = rv.Pattern_is_invalid_exception();
+			ctx.App().Usr_dlg().Warn_many("", "", "regx is invalid: regx=~{0} page=~{1} exc={2}", regx, ctx.Page().Ttl().Page_db(), Err_.Message_gplx_log(exc));
 		}
 		return rv;
 	}

400_xowa/src/gplx/xowa/xtns/scribunto/libs/Scrib_lib_ustring__match__tst.java

@@ -16,7 +16,7 @@ Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
 package gplx.xowa.xtns.scribunto.libs; import gplx.*; import gplx.xowa.*; import gplx.xowa.xtns.*; import gplx.xowa.xtns.scribunto.*;
 import org.junit.*; import gplx.xowa.xtns.scribunto.engines.mocks.*;
 public class Scrib_lib_ustring__match__tst {
-	private final Mock_scrib_fxt fxt = new Mock_scrib_fxt(); private Scrib_lib lib;
+	private final    Mock_scrib_fxt fxt = new Mock_scrib_fxt(); private Scrib_lib lib;
 	@Before public void init() {
 		fxt.Clear();
 		lib = fxt.Core().Lib_ustring().Init();
@@ -26,11 +26,11 @@ public class Scrib_lib_ustring__match__tst {
 		Exec_match("abcd"	, "x"				, 1, String_.Null_mark);			// empty
 		Exec_match("abcd"	, "a"				, 2, String_.Null_mark);			// bgn
 		Exec_match("abcd"	, "b(c)"			, 1, "c");							// group
-		Exec_match(" a b "	, "^%s*(.-)%s*$"	, 1, "a b;");						// trim; NOTE: changed from "a b" to "a b;"; DATE:2015-01-30
+		Exec_match(" a b "	, "^%s*(.-)%s*$"	, 1, "a b");						// trim; NOTE: changed back from "a b;" to "a b"; DATE:2017-04-23; changed from "a b" to "a b;"; DATE:2015-01-30
 		Exec_match("abcd"	, "a"				, 0, "a");							// handle 0; note that php/lua is super-1, but some modules pass in 0; ru.w:Module:Infocards; DATE:2013-11-08
 		Exec_match("abcd"	, "."				, -1, "d");							// -1
 		Exec_match("aaa"	, "a"				, 1, "a");							// should return 1st match not many
-		Exec_match("aaa"	, "(a)"				, 1, "a;a;a");						// should return all matches
+		Exec_match("aaa"	, "(a)"				, 1, "a");							// should return 1st match only; PAGE:en.d:действительное_причастие_настоящего_времени DATE:2017-04-23
 		Exec_match("a b"	, "%S"				, 1, "a");							// %S was returning every match instead of 1st; PAGE:en.w:Bertrand_Russell; DATE:2014-04-02
 		Exec_match(1		, "a"				, 1, String_.Null_mark);			// Module can pass raw ints; PAGE:en.w:Budget_of_the_European_Union; DATE:2015-01-22
 		Exec_match(""		, "a?"				, 1, "");							// no results with ? should return "" not nil; PAGE:en.d:民; DATE:2015-01-30