Subramanya Sastry has uploaded a new change for review.
https://gerrit.wikimedia.org/r/52630
Change subject: Used config to identify valid exts; Made pre & nowiki case
insensitive
......................................................................
Used config to identify valid exts; Made pre & nowiki case insensitive
* Used config information to identify valid extensions and removed
temporary hack.
* Made nowiki and pre tag matching case-insensitive.
- RTing of these in original case is not yet done -- this seems
a little bit messy and will require fixing any tag-comparison
that doesn't do normalized case comparison. Will also require
fixes to serializer. Can be done post ongoing refactoring.
* 1 wt2wt test worse off -- this is because of RTing in normalized
lower-case of nowiki tags.
Change-Id: Ie727b76926247d65abf4b9c3fe79afc32eb03827
---
M js/lib/mediawiki.Util.js
M js/lib/pegTokenizer.pegjs.txt
2 files changed, 27 insertions(+), 33 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/mediawiki/extensions/Parsoid
refs/changes/30/52630/1
diff --git a/js/lib/mediawiki.Util.js b/js/lib/mediawiki.Util.js
index 028bba6..74a451b 100644
--- a/js/lib/mediawiki.Util.js
+++ b/js/lib/mediawiki.Util.js
@@ -803,25 +803,6 @@
}
console.error(out.join(arguments[1]));
},
-
- // SSS FIXME: This should probably come from some config/api?
- // This is just a temporary hack for this patch
- installedExts: null,
-
- extensionInstalled: function(env, name) {
- // SSS FIXME: This is just a temporary hack for this patch.
- //
- // Once this info is available in the config, we should check
- // env.conf.wiki.extensionTags instead.
-
- if (!this.installedExts) {
- this.installedExts = this.arrayToHash([
- 'categorytree', 'charinsert', 'gallery',
'hiero', 'imagemap',
- 'inputbox', 'math', 'poem', 'syntaxhighlight',
'tag', 'timeline'
- ]);
- }
- return this.installedExts[name] === true;
- }
};
/**
diff --git a/js/lib/pegTokenizer.pegjs.txt b/js/lib/pegTokenizer.pegjs.txt
index fbeae09..3559e17 100644
--- a/js/lib/pegTokenizer.pegjs.txt
+++ b/js/lib/pegTokenizer.pegjs.txt
@@ -1251,6 +1251,12 @@
return [l].concat(ls);
}
+pre_tag_name =
+ tag:[prePRE]+ {
+ tag = tag.join('');
+ return tag.toLowerCase() === "pre" ? tag : null;
+ }
+
// An indented pre block that is surrounded with pre tags. The pre tags are
// used directly.
// XXX gwicke: check if the first line is not indented, and round-trip spaces;
@@ -1259,12 +1265,12 @@
pre_indent_in_tags
= & { return stops.inc('pre'); }
space+ // XXX: capture space for round-tripping
- "<pre"
+ "<" pre_tag_name
attribs:generic_attribute*
">"
l:inlineline
ls:(sol pre_indent_line)*
- "</pre>"
+ "</" pre_tag_name ">"
{
stops.dec('pre');
var ret = [ new TagTk( 'pre', attribs, { tsr: [pos0, pos0] } ) ];
@@ -1288,7 +1294,7 @@
*/
pre
= & { return stops.inc('pre'); }
- "<pre"
+ "<" pre_tag_name
attribs:generic_attribute*
endpos:(">" { return pos })
// MediaWiki <pre> is special in that it converts all pre content to plain
@@ -1296,14 +1302,14 @@
ts:( newlineToken
/ (htmlentity / [^&<]+)+
/ nowiki
- / !"</pre>" t2:(htmlentity / .) { return t2 })+
- ("</pre>" / eof) {
+ / !("</" pre_tag_name ">") t2:(htmlentity / .) { return t2 })+
+ ("</" pre_tag_name ">" / eof) {
stops.dec('pre');
// return nowiki tags as well?
return [ new TagTk( 'pre', attribs, { stx: 'html', tsr: [pos0, endpos]
} ) ]
.concat(flatten_stringlist(ts), [ new EndTagTk( 'pre', [],
{ tsr: [pos - 6, pos] } ) ]);
}
- / "</pre>" { stops.dec('pre'); return "</pre>"; }
+ / "</" pre_tag_name ">" { stops.dec('pre'); return "</pre>"; }
/ & { return stops.dec('pre'); }
/* -----------------------------------------------------------------------
@@ -1325,7 +1331,7 @@
var tagName = t.name.toLowerCase(),
dp = t.dataAttribs,
isHtmlTag = html5_tag_names[tagName] === true ||
html_old_names[tagName] === true,
- isInstalledExt = Util.extensionInstalled(pegArgs.env, tagName),
+ isInstalledExt = pegArgs.env.conf.wiki.isExtensionTag(tagName),
supportedTag = nativeParsoidExts[tagName] === true;
if (!isHtmlTag && !isInstalledExt && !supportedTag) {
@@ -1464,9 +1470,16 @@
* pre blocks would still remain in the grammar though, so overall handling it
* all here is cleaner.
*/
+
+nowiki_tag_name =
+ tag:[nowikNOWIK]+ {
+ tag = tag.join('');
+ return tag.toLowerCase() === "nowiki" ? tag : null;
+ }
+
nowiki
- = "<nowiki>" nc:nowiki_content "</nowiki>" {
- //console.warn( 'full nowiki return: ' + pp(nc));
+ = "<" nowiki_tag_name ">" nc:nowiki_content "</" nowiki_tag_name ">" {
+ // console.warn( 'full nowiki return: ' + pp(nc));
return [
new TagTk( 'span',
[
@@ -1483,8 +1496,8 @@
}
// nowiki fallback: source-based round-tripping of <nowiki />.
/ nw0:({return pos})
- "<nowiki" [ ]* "/>" {
- //console.warn('<nowiki/>');
+ "<" nowiki_tag_name [ ]* "/>" {
+ // console.warn('<nowiki/>');
return [
new SelfclosingTagTk('meta',
[new KV('typeof', 'mw:Placeholder')],
@@ -1498,8 +1511,8 @@
// of unbalanced nowiki tags that are treated as text.
/ ! { return stops.counters.pre > 0; }
nw0:({return pos})
- "<" "/"? "nowiki" [ ]* "/"? ">" {
- //console.warn('nowiki text');
+ "<" "/"? nowiki_tag_name [ ]* "/"? ">" {
+ // console.warn('nowiki text');
var nowiki = input.substring(nw0, pos);
return [
new TagTk( 'span', [
@@ -1528,7 +1541,7 @@
//console.warn('nested pre in nowiki');
return ["<pre"].concat(p0, p1, [">"], p2,
["</pre>"]).join('');
}
- / (!pre_break !"</nowiki>" c:(htmlentity / .) {
+ / (!pre_break !("</" nowiki_tag_name ">") c:(htmlentity / .) {
//console.warn('nowiki: single char' + c);
return c;
})
--
To view, visit https://gerrit.wikimedia.org/r/52630
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: Ie727b76926247d65abf4b9c3fe79afc32eb03827
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/extensions/Parsoid
Gerrit-Branch: master
Gerrit-Owner: Subramanya Sastry <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
