Subramanya Sastry has uploaded a new change for review. (
https://gerrit.wikimedia.org/r/379430 )
Change subject: Detect misnested tags that have different behavior in HTML5 vs
HTML4
......................................................................
Detect misnested tags that have different behavior in HTML5 vs HTML4
* This will make a different when Tidy is replaced by Remex and
will also make Parsoid rendering more compatible with Tidy.
Bug: T176363
Change-Id: I93d779eba9b7738c309b8a8a4a89f337bb4ac168
---
M lib/utils/DOMUtils.js
M lib/wt2html/pp/handlers/linter.js
M tests/mocha/linter.js
3 files changed, 76 insertions(+), 9 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/mediawiki/services/parsoid
refs/changes/30/379430/1
diff --git a/lib/utils/DOMUtils.js b/lib/utils/DOMUtils.js
index e073bde..5de498f 100644
--- a/lib/utils/DOMUtils.js
+++ b/lib/utils/DOMUtils.js
@@ -1244,6 +1244,12 @@
return next;
},
+ hasFollowingContent: function(node) {
+ return !DU.isBody(node) && (
+ DU.nextNonSepSibling(node) ||
DU.hasFollowingContent(node.parentNode)
+ );
+ },
+
numNonDeletedChildNodes: function(node) {
var n = 0;
var child = node.firstChild;
diff --git a/lib/wt2html/pp/handlers/linter.js
b/lib/wt2html/pp/handlers/linter.js
index 8e95440..bd6fb5b 100644
--- a/lib/wt2html/pp/handlers/linter.js
+++ b/lib/wt2html/pp/handlers/linter.js
@@ -17,6 +17,24 @@
var Util = require('../../../utils/Util.js').Util;
var Consts = require('../../../config/WikitextConstants.js').WikitextConstants;
+var tagsWithChangedMisnestingBehavior;
+function getTagsWithChangedMisnestingBehavior() {
+ if (!tagsWithChangedMisnestingBehavior) {
+ tagsWithChangedMisnestingBehavior = new Set();
+ Consts.HTML.HTML5Tags.forEach(function(t) {
+ if (Consts.Sanitizer.TagWhiteList.has(t) &&
+ !Consts.HTML.BlockTags.has(t) &&
+ !Consts.HTML.FormattingTags.has(t) &&
+ !Consts.HTML.VoidTags.has(t)
+ ) {
+ tagsWithChangedMisnestingBehavior.add(t);
+ }
+ });
+ }
+
+ return tagsWithChangedMisnestingBehavior;
+}
+
/*
* Log Transclusion with more than one parts
* Ex - {{table-start}}
@@ -172,16 +190,22 @@
templateInfo: templateInfo,
params: { name: cNodeName },
};
- var adjNode = getNextMatchingNode(c, c);
- if (adjNode) {
- var adjDp = DU.getDataParsoid(adjNode);
- if (!adjDp.tmp) {
- adjDp.tmp = {};
+ if
(getTagsWithChangedMisnestingBehavior().has(c.nodeName) &&
+ DU.hasFollowingContent(c)
+ ) {
+ env.log('lint/html5-misnesting', lintObj);
+ } else {
+ var adjNode = getNextMatchingNode(c, c);
+ if (adjNode) {
+ var adjDp = DU.getDataParsoid(adjNode);
+ if (!adjDp.tmp) {
+ adjDp.tmp = {};
+ }
+ adjDp.tmp.linted = true;
+ env.log('lint/misnested-tag', lintObj);
+ } else if (DU.hasLiteralHTMLMarker(dp)) {
+ env.log('lint/missing-end-tag',
lintObj);
}
- adjDp.tmp.linted = true;
- env.log('lint/misnested-tag', lintObj);
- } else if (DU.hasLiteralHTMLMarker(dp)) {
- env.log('lint/missing-end-tag', lintObj);
}
}
diff --git a/tests/mocha/linter.js b/tests/mocha/linter.js
index f1ad362..d206801 100644
--- a/tests/mocha/linter.js
+++ b/tests/mocha/linter.js
@@ -600,4 +600,41 @@
return expectEmptyResults(wt, { tweakEnv: tweakEnv });
});
});
+
+ describe('HTML5 MISNESTED TAGS', function() {
+ it('should not trigger html5 misnesting if there is no
following content', function() {
+ return parseWT('<del>foo\nbar').then(function(result) {
+ result.should.have.length(1);
+ result[0].should.have.a.property("type",
"missing-end-tag");
+ result[0].should.have.a.property("params");
+ result[0].params.should.have.a.property("name",
"del");
+ });
+ });
+ it('should trigger html5 misnesting correctly', function() {
+ return parseWT('<del>foo\n\nbar').then(function(result)
{
+ result.should.have.length(1);
+ result[0].should.have.a.property("type",
"html5-misnesting");
+ result[0].dsr.should.deep.equal([ 0, 8, 5, 0 ]);
+ result[0].should.have.a.property("params");
+ result[0].params.should.have.a.property("name",
"del");
+ });
+ });
+ it('should trigger html5 misnesting for span', function() {
+ return
parseWT('<span>foo\n\nbar').then(function(result) {
+ result.should.have.length(1);
+ result[0].should.have.a.property("type",
"html5-misnesting");
+ result[0].dsr.should.deep.equal([ 0, 9, 6, 0 ]);
+ result[0].should.have.a.property("params");
+ result[0].params.should.have.a.property("name",
"span");
+ });
+ });
+ it('should not trigger html5 misnesting for formatting tags',
function() {
+ return
parseWT('<small>foo\n\nbar').then(function(result) {
+ result.should.have.length(1);
+ result[0].should.have.a.property("type",
"missing-end-tag");
+ result[0].should.have.a.property("params");
+ result[0].params.should.have.a.property("name",
"small");
+ });
+ });
+ });
});
--
To view, visit https://gerrit.wikimedia.org/r/379430
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I93d779eba9b7738c309b8a8a4a89f337bb4ac168
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/services/parsoid
Gerrit-Branch: master
Gerrit-Owner: Subramanya Sastry <ssas...@wikimedia.org>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
