Seems to be related to references somehow.
Description
Details
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Open | None | T76456 Language Engineering tracker of trackers (tracking) | |||
Open | None | T112880 ContentTranslation alignment issues (tracking) | |||
Resolved | santhosh | T112253 CX alignment issue when translating the article [[HTML]] from Finnish |
Event Timeline
The reference section in that page has an unusual id for references
id is viitteet-malline
From aligner, mw.cx.getSourceSection('viitteet-malline') is retuning [] and there goes the alignment.
We have a sanity check
// Sanity check, id should be either a number or prefixed with mw if ( !id || ( isNaN( id ) && id.indexOf( 'mw' ) !== 0 ) ) { return $( [] ); } else { return $( document.getElementById( id ) ); }
Is there a reason to do this check now? @Nikerabbit?
The id is originated from https://fi.wikipedia.org/w/index.php?title=Malline:Viitteet&action=edit
I thought we added our own id's for everything we want to reference? For references it is important that we do not display html that is under user control.
I thought we added our own id's for everything we want to reference
We use the ids generated by parsoid. We no longer generate ids for sections
For references it is important that we do not display html that is under user control.
We need to display the references section. The wrong assumption about ids is making the published content broken for pages that use Malline:Viitteet - because the sanity check code breaks the correspondence of the sections
Change 241624 had a related patch set uploaded (by Santhosh):
Relax section id validations in mw.cx.getSourceSection and getTranslationSection
Change 241624 merged by jenkins-bot:
Relax section id validations in mw.cx.getSourceSection and getTranslationSection