@ -1,6 +1,6 @@
/ *
XOWA: the XOWA Offline Wiki Application
Copyright ( C ) 2012 - 20 17 gnosygnu @gmail.com
Copyright ( C ) 2012 - 20 2 1 gnosygnu @gmail.com
XOWA is licensed under the terms of the General Public License ( GPL ) Version 3 ,
or alternatively under the terms of the Apache License Version 2.0 .
@ -13,80 +13,94 @@ 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.htmls.hrefs ; import gplx.* ; import gplx.xowa.* ; import gplx.xowa.htmls.* ;
import gplx.core.btries.* ; import gplx.core.primitives.* ;
public class Xoh_href_gui_utl {
public static String Html_extract_text ( String site , String page , String text_str ) {
byte [ ] text_bry = Bry_ . new_u8 ( text_str ) ;
int text_len = text_bry . length ;
int text_tid = Byte_ascii . To_a7_int ( text_bry [ 0 ] ) ;
switch ( text_tid ) {
case Text_tid_none : return "" ; // "0"
case Text_tid_text : return String_ . new_u8 ( text_bry , 2 , text_len ) ; // 2 to skip "1|"
case Text_tid_href : break ; // fall through to below
default : throw Err_ . new_unhandled ( text_tid ) ;
}
int href_bgn = 2 ; // 2 to skip "2|"
if ( Bry_ . Has_at_bgn ( text_bry , Xoh_href_ . Bry__file , href_bgn , text_len ) )
href_bgn + = Xoh_href_ . Len__file ; // skip "file://"
Byte_obj_val href_tid = ( Byte_obj_val ) href_trie . Match_bgn ( text_bry , href_bgn , text_len ) ;
if ( href_tid ! = null ) {
switch ( href_tid . Val ( ) ) {
case Href_tid_wiki : return site + String_ . new_u8 ( text_bry , href_bgn , text_len ) ;
case Href_tid_site : return String_ . new_u8 ( text_bry , href_bgn + 6 , text_len ) ; // +6 to skip "site/"
case Href_tid_anch : return site + "/wiki/" + page + String_ . new_u8 ( text_bry , href_bgn , text_len ) ;
}
}
return String_ . new_u8 ( text_bry , 2 , text_len ) ; // 2 to skip "2|"; handles "http://" text as well as any fall-thru from above
}
public static String Standardize_xowa_link ( String str ) {
byte [ ] bry = Bry_ . new_u8 ( str ) ;
int skip = Skip_start_of_xowa_link ( bry , bry . length , 0 ) ;
return skip = = 0 ? str : String_ . Mid ( str , skip ) ;
}
private static int Skip_start_of_xowa_link ( byte [ ] src , int src_len , int bgn ) {
if ( ! Bry_ . Has_at_bgn ( src , Xoh_href_ . Bry__file , bgn , src_len ) ) return bgn ; // does not start with "file://"
int pos = bgn + Xoh_href_ . Len__file ; // skip "file://"
Object tid_obj = href_trie . Match_bgn ( src , pos , src_len ) ;
if ( tid_obj = = null ) {
return bgn ; // if not a known xowa link, return original bgn;
}
switch ( ( ( Byte_obj_val ) tid_obj ) . Val ( ) ) {
case Href_tid_site : return pos ;
case Href_tid_wiki : return pos ;
case Href_tid_anch : return pos ;
default : throw Err_ . new_unhandled ( tid_obj ) ;
}
}
private static final byte Text_tid_none = 0 , Text_tid_text = 1 , Text_tid_href = 2 ;
private static final byte Href_tid_wiki = 1 , Href_tid_site = 2 , Href_tid_anch = 3 ;
private static final Btrie_slim_mgr href_trie = Btrie_slim_mgr . cs ( )
. Add_bry_byte ( Xoh_href_ . Bry__site , Href_tid_site )
. Add_bry_byte ( Xoh_href_ . Bry__wiki , Href_tid_wiki )
. Add_bry_byte ( Xoh_href_ . Bry__anch , Href_tid_anch )
;
}
/ *
NOTE_1:
. swt / mozilla treats text differently in href = "{text}" when content_editable = n ; occurs in LocationListener . changing
http: //a.org -> http://a.org does nothing
A - > file : ///A adds "file:///"
/ wiki / A - > file : ///wiki/A adds "file://"
Category: A - > Category : A noops ; Category is assumed to be protocol ?
//en.wiktionary.org/wiki/a -> file:///wiki/a strips out site name and prepends "file://"; no idea why
. so , to handle the above , the code does the following
http: //a.org -> http://a.org does nothing; nothing needed
A - > / wiki / A always prepend / wiki /
Category: A - > / wiki / Category : A always prepend / wiki /
//en.wiktionary.org/wiki/A -> /site/en.wiktionary.org/wiki/A always transform relative url to /site/
. the href will still come here as file : ///wiki/A or file:///site/en.wiktionary.org/wiki/A.
. however , the file : // can be lopped off and discarded and the rest of the href will fall into one of the following cases
. . / wiki /
. . / site /
. . / xcmd /
. . #
. . anything else - > assume to be really a file : // url; EX: file://C/dir/fil.txt -> C/dir/fil.txt
. the other advantage of this approach is that this proc can be reused outside of swt calls ; i . e . : it can parse both "file:///wiki/A" and "/wiki/A"
* /
package gplx.xowa.htmls.hrefs ;
import gplx.Bry_ ;
import gplx.Byte_ascii ;
import gplx.Err_ ;
import gplx.String_ ;
import gplx.core.btries.Btrie_slim_mgr ;
import gplx.core.primitives.Byte_obj_val ;
import gplx.gfui.kits.swts.Swt_html_utl ;
public class Xoh_href_gui_utl {
public static String Html_extract_text ( String site , String page , String text_str ) {
byte [ ] text_bry = Bry_ . new_u8 ( text_str ) ;
int text_len = text_bry . length ;
int text_tid = Byte_ascii . To_a7_int ( text_bry [ 0 ] ) ;
switch ( text_tid ) {
case Text_tid_none : return "" ; // "0"
case Text_tid_text : return String_ . new_u8 ( text_bry , 2 , text_len ) ; // 2 to skip "1|"
case Text_tid_href : break ; // fall through to below
default : throw Err_ . new_unhandled ( text_tid ) ;
}
String href_str = String_ . Mid ( String_ . new_u8 ( text_bry ) , 2 ) ;
href_str = Swt_html_utl . NormalizeSwtUrl ( href_str ) ;
if ( String_ . Has_at_bgn ( href_str , Xoh_href_ . Str__file ) )
href_str = Standardize_xowa_link ( href_str ) ; // skip "file://"
Byte_obj_val href_tid = ( Byte_obj_val ) href_trie . Match_bgn ( Bry_ . new_u8 ( href_str ) , 0 , href_str . length ( ) ) ;
if ( href_tid ! = null ) {
switch ( href_tid . Val ( ) ) {
case Href_tid_wiki : return site + href_str ;
case Href_tid_site : return String_ . Mid ( href_str , 6 ) ; // +6 to skip "site/"
case Href_tid_anch : return site + "/wiki/" + page + href_str ;
}
}
return href_str ;
}
public static String Standardize_xowa_link ( String str ) {
byte [ ] bry = Bry_ . new_u8 ( str ) ;
int skip = Skip_start_of_xowa_link ( bry , bry . length , 0 ) ;
return skip = = 0 ? str : String_ . Mid ( str , skip ) ;
}
private static int Skip_start_of_xowa_link ( byte [ ] src , int src_len , int bgn ) {
if ( ! Bry_ . Has_at_bgn ( src , Xoh_href_ . Bry__file , bgn , src_len ) ) return bgn ; // does not start with "file://"
int pos = bgn + Xoh_href_ . Len__file ; // skip "file://"
Object tid_obj = href_trie . Match_bgn ( src , pos , src_len ) ;
if ( tid_obj = = null ) {
if ( src_len - pos > 0 & & src [ pos ] = = Byte_ascii . Slash ) { // handle "file:///C:/dir/fil.png"
return pos + 1 ;
}
else {
return bgn ; // if not a known xowa link, return original bgn;
}
}
switch ( ( ( Byte_obj_val ) tid_obj ) . Val ( ) ) {
case Href_tid_site : return pos ;
case Href_tid_wiki : return pos ;
case Href_tid_anch : return pos ;
default : throw Err_ . new_unhandled ( tid_obj ) ;
}
}
private static final byte Text_tid_none = 0 , Text_tid_text = 1 , Text_tid_href = 2 ;
private static final byte Href_tid_wiki = 1 , Href_tid_site = 2 , Href_tid_anch = 3 ;
private static final Btrie_slim_mgr href_trie = Btrie_slim_mgr . cs ( )
. Add_bry_byte ( Xoh_href_ . Bry__site , Href_tid_site )
. Add_bry_byte ( Xoh_href_ . Bry__wiki , Href_tid_wiki )
. Add_bry_byte ( Xoh_href_ . Bry__anch , Href_tid_anch )
;
}
/ *
NOTE_1:
. swt / mozilla treats text differently in href = "{text}" when content_editable = n ; occurs in LocationListener . changing
http: //a.org -> http://a.org does nothing
A - > file : ///A adds "file:///"
/ wiki / A - > file : ///wiki/A adds "file://"
Category: A - > Category : A noops ; Category is assumed to be protocol ?
//en.wiktionary.org/wiki/a -> file:///wiki/a strips out site name and prepends "file://"; no idea why
. so , to handle the above , the code does the following
http: //a.org -> http://a.org does nothing; nothing needed
A - > / wiki / A always prepend / wiki /
Category: A - > / wiki / Category : A always prepend / wiki /
//en.wiktionary.org/wiki/A -> /site/en.wiktionary.org/wiki/A always transform relative url to /site/
. the href will still come here as file : ///wiki/A or file:///site/en.wiktionary.org/wiki/A.
. however , the file : // can be lopped off and discarded and the rest of the href will fall into one of the following cases
. . / wiki /
. . / site /
. . / xcmd /
. . #
. . anything else - > assume to be really a file : // url; EX: file://C/dir/fil.txt -> C/dir/fil.txt
. the other advantage of this approach is that this proc can be reused outside of swt calls ; i . e . : it can parse both "file:///wiki/A" and "/wiki/A"
* /
gnosygnu
3 years ago