jenkins-bot has submitted this change and it was merged.
Change subject: T122054: Strip nowiki spans from templated / extension content
......................................................................
T122054: Strip nowiki spans from templated / extension content
Change-Id: I06353fafac9c8a6efa063302727df5815842607f
---
M lib/wt2html/pp/cleanup.js
M tests/parserTests-blacklist.js
M tests/parserTests.txt
3 files changed, 21 insertions(+), 6 deletions(-)
Approvals:
Arlolra: Looks good to me, approved
jenkins-bot: Verified
diff --git a/lib/wt2html/pp/cleanup.js b/lib/wt2html/pp/cleanup.js
index 808b41a..fe82e16 100644
--- a/lib/wt2html/pp/cleanup.js
+++ b/lib/wt2html/pp/cleanup.js
@@ -50,6 +50,7 @@
if (!atTopLevel || !tplInfo || !DU.isElt(node)) {
return true;
}
+
// Cannot delete if:
// * it is the first node since that carries the transclusion
// information (typeof, data-mw). We could delete and migrate
@@ -87,6 +88,7 @@
if (!DU.isElt(node)) { return true; }
var dp = DU.getDataParsoid(node);
+ var next;
// Delete empty auto-inserted elements
if (dp.autoInsertedStart && dp.autoInsertedEnd &&
@@ -96,7 +98,7 @@
/^\s*$/.test(node.textContent)
)
)) {
- var next = node.nextSibling;
+ next = node.nextSibling;
if (node.firstChild) {
// migrate the ws out
node.parentNode.insertBefore(node.firstChild, node);
@@ -105,9 +107,9 @@
return next;
}
+ // Strip invalid data-mw attributes
var validDataMW = !!Object.keys(DU.getDataMw(node)).length;
if (!validDataMW) {
- // Strip it
DU.setDataMw(node, undefined);
}
@@ -139,11 +141,24 @@
//
// But, do not zero it out if the node has template encapsulation
// information. That will be disastrous (see T54638, T54488).
- if (dp.fostered && dp.dsr && !DU.isFirstEncapsulationWrapperNode(node))
{
+ if (dp.fostered && dp.dsr && !(tplInfo && tplInfo.first === node)) {
dp.dsr[0] = dp.dsr[1];
}
if (atTopLevel) {
+ // Strip nowiki spans from templated content
+ // but leave behind wrappers on root nodes (they have valid
about ids)
+ if (tplInfo && !DU.isTplOrExtToplevelNode(node) &&
+ /^mw:Nowiki$/.test(node.getAttribute('typeof'))) {
+ next = node.nextSibling;
+ DU.migrateChildren(node, node.parentNode, next);
+
+ // Replace the span with an empty text node.
+ // (better for perf instead of deleting the node)
+
node.parentNode.replaceChild(node.ownerDocument.createTextNode(''), node);
+ return next;
+ }
+
// Strip data-parsoid from templated content, where unnecessary.
if (env.discardDataParsoid ||
(tplInfo
diff --git a/tests/parserTests-blacklist.js b/tests/parserTests-blacklist.js
index 61c9637..a039543 100644
--- a/tests/parserTests-blacklist.js
+++ b/tests/parserTests-blacklist.js
@@ -381,7 +381,7 @@
add("wt2wt", "Ref: 5. body should accept generic wikitext", "A <ref>This is a
'''[[bolded link]]''' and this is a
{{echo|transclusion}}\n</ref>\n\n<references />");
add("wt2wt", "Ref: 6. indent-pres should not be output in ref-body", "A
<ref>foo\n bar\n baz\n</ref>\n\n<references />");
add("wt2wt", "Ref: 7. No p-wrapping in ref-body", "A
<ref>foo\nbar\nbaz\nbooz\n</ref>\n\n<references />");
-add("wt2wt", "Ref: 8. transclusion wikitext has lower precedence", "A <ref>foo
{{echo|</ref> B C}}\n\n<references />");
+add("wt2wt", "Ref: 8. transclusion wikitext has lower precedence", "A <ref>foo
<nowiki>{{</nowiki>echo|</ref> B C}}\n\n<references />");
add("wt2wt", "Ref: 9. unclosed comments should not leak out of ref-body", "A
<ref>foo <!----></ref> B C\n<references />");
add("wt2wt", "Ref: 10. Unclosed HTML tags should not leak out of ref-body", "A
<ref><b> foo </ref> B C\n\n<references />");
add("wt2wt", "Ref: 19. ref-tags with identical name encodings should get
identical indexes", "1 <ref name=\"a & b\">foo</ref> 2 <ref name=\"a &amp;
b\" />\n\n<references />");
@@ -1306,7 +1306,7 @@
add("html2wt", "Ref: 5. body should accept generic wikitext", "A <ref>This is
a '''[[bolded link]]''' and this is a
{{echo|transclusion}}\n</ref>\n\n<references />");
add("html2wt", "Ref: 6. indent-pres should not be output in ref-body", "A
<ref>foo\n bar\n baz\n</ref>\n\n<references />");
add("html2wt", "Ref: 7. No p-wrapping in ref-body", "A
<ref>foo\nbar\nbaz\nbooz\n</ref>\n\n<references />");
-add("html2wt", "Ref: 8. transclusion wikitext has lower precedence", "A
<ref>foo {{echo|</ref> B C<nowiki>}}</nowiki>\n<references />");
+add("html2wt", "Ref: 8. transclusion wikitext has lower precedence", "A
<ref>foo <nowiki>{{</nowiki>echo|</ref> B C<nowiki>}}</nowiki>\n<references
/>");
add("html2wt", "Ref: 9. unclosed comments should not leak out of ref-body", "A
<ref>foo <!----></ref> B C\n<references />");
add("html2wt", "Ref: 10. Unclosed HTML tags should not leak out of ref-body",
"A <ref><b> foo </ref> B C\n\n<references />");
add("html2wt", "Ref: 19. ref-tags with identical name encodings should get
identical indexes", "1 <ref name=\"a & b\">foo</ref> 2 <ref name=\"a &amp;
b\" />\n\n<references />");
diff --git a/tests/parserTests.txt b/tests/parserTests.txt
index adfceae..b710515 100644
--- a/tests/parserTests.txt
+++ b/tests/parserTests.txt
@@ -21610,7 +21610,7 @@
!! html
<p>A <span about="#mwt2" class="mw-ref" id="cite_ref-1" rel="dc:references"
typeof="mw:Extension/ref"
data-mw='{"name":"ref","body":{"id":"mw-reference-text-cite_note-1"},"attrs":{}}'><a
href="#cite_note-1"><span class="mw-reflink-text">[1]</span></a></span> B
C<span typeof="mw:Nowiki">}}</span></p>
<ol class="mw-references" typeof="mw:Extension/references" about="#mwt4"
data-mw='{"name":"references","attrs":{}}'>
-<li about="#cite_note-1" id="cite_note-1"><a href="#cite_ref-1"
rel="mw:referencedBy"><span class="mw-linkback-text">↑ </span></a> <span
id="mw-reference-text-cite_note-1" class="mw-reference-text">foo <span
typeof="mw:Nowiki"
data-parsoid='{"src":"{{","dsr":[12,14,0,0]}'>{{</span>echo|</span></li>
+<li about="#cite_note-1" id="cite_note-1"><a href="#cite_ref-1"
rel="mw:referencedBy"><span class="mw-linkback-text">↑ </span></a> <span
id="mw-reference-text-cite_note-1" class="mw-reference-text">foo
{{echo|</span></li>
</ol>
!!end
--
To view, visit https://gerrit.wikimedia.org/r/260395
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I06353fafac9c8a6efa063302727df5815842607f
Gerrit-PatchSet: 4
Gerrit-Project: mediawiki/services/parsoid
Gerrit-Branch: master
Gerrit-Owner: Subramanya Sastry <ssas...@wikimedia.org>
Gerrit-Reviewer: Arlolra <abrea...@wikimedia.org>
Gerrit-Reviewer: Cscott <canan...@wikimedia.org>
Gerrit-Reviewer: Subramanya Sastry <ssas...@wikimedia.org>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
