@Whatamidoing-WMF is this still a problem ? It works for me and we didn't have many complaints in the mean time.

FYI: Seems the change missed the cut-off for the Safari 11 release.

Please think of T176300: Significant reflow on watchlist when filters are enabled when this gets done... combined, i'm sure they will account for a significant amount of 'misclicks' by people.

FYI, there's already been like 4 questions about this on WP:VP/T. Which is a LOT relatively speaking (esp. if you consider that most people won't find their way to WP:VP/T usually.

Seems that Editpage is still a bit of a mess. Not easy to modify the HTML, without reimplementing the entire form...
Simple workaround would be to use OutputPage->addBodyClasses, to add editor-wikieditor or something..

@Redrose64 where are you still seeing this though ? Maybe it's a template generating it, instead of core ?

Possibly related to Thumbor work ? Maybe @Gilles will know.

Another report: https://commons.wikimedia.org/wiki/Commons:Village_pump#An_odd_problem_with_my_old_vertical.2Fportrait_photos_taken_with_Canon_A650_IS

That patch only works if the old toolbar is enabled as well.. Since we are removing that soon, we should probably look at this again... We really need some unique CSS marker in the generated HTML, to denote that WikiEditor will be loading.

ping @Kolossos @simon04

Hmm, @Yurik might have tried to fix this on OSM some 10 months ago. Maybe the data that WIWOSM is using is not updating ?

I'll just sigh about me not having gotten around to this yet, 4 weeks after Wikimania :)

@dschwen, you might want to consider switching it to the new kubernetes grid

The problem with secondary coordinates is that you cannot know what they are about... Only that they are present somewhere on an article. For this example, you would get a 1000 POIs all pointing to the same article. Instead, we should look at pulling coordinates straight from wikidata, as detailed in [[phab:T35704]].

Those are two common misuses of the markup [..] It is also used for what the element actually exists for

That's why I said use of : not use of ; :

Wrong popups...

I did a tally:

Note that bullets next to floating items are inherently somewhat problematic in HTML/CSS. It's why we have Template flowlist on English Wikipedia.

Yeah, i'm guessing this is a bug caused by the epic reworking of how all the editsummaries are stores. T166732: Refactor comment storage in the database and abstract access in MediaWiki and T166733: Deploy refactored comment storage

" in the headline view and as the article header image when the articles were opened in the app"

If possible, could someone with OTRS access possibly make them 'decent' by overlaying with black squares or something and upload them here. Just to make sure we are all looking at the right data fields in the various APIs, instead of making guesses.

@Nthep just wondering; could you describe as exact as possible, what fragment of content (where exactly (screenshots would be cool), type (text/thumb), etc) was not up-to-date after the revert, what you tried to do in the app to check or rectify (closing, reopening, navigation actions, refresh buttons), and what amount of time you tried to do this ?

This project no longer exists btw. The code is in Google code archive, waiting for someone to export it to GitHub. I think there is also a clone here: https://github.com/librefonts/tharlon

This is no longer a problem

This is also browser specific behaviour isn't it ?

Nice work @mxn !

I'd just add a disclaimer that all avatars, meme, tokens and uploaded screenshots are exempt from stated licenses.

The coordinates parser function is within the scope of the GeoData extension.

That's not really the point, we can't go redesign extensions on the whim of a local consensus that didn't account for side effects for the technology stack. We (WMF, nor volunteer devs) don't have the resources for that.

Note that the parser function is not datastorage. The wikicode is the storage function and the first call to #coordinates on a page becomes a cached value in the GeoData tables.

So... wait, you are in a module, execute something, then string modify the result and return that ????

After https://gerrit.wikimedia.org/r/#/c/376036/ is live, these will be warnings and not errors, which will make this problem less disruptive.

@Jdlrobson no, this is OCG specific and not an issue with Proton/Electron.

I'm experiencing this myself, a user on WP:VP/T reported it too (see the dupe), and now another report on WP:HELPDESK.
https://en.wikipedia.org/wiki/Wikipedia:Help_desk#Help_me_disable_Beta_Feature:_Wikitext_syntax_highlighting

Do remember that 'on idle' is something other than machine performance. If you take an older machine, it won't really matter when you load something, the sheer size (of memory usage and graphics size) will likely do you under. Scroll performance especially will likely tank for those users.

I still have the problem...

Hmm, maybe a master vs. slave database race condition in the logic for "automatically enable all new beta features" ?

Yeah, width doesn't work for that. You need min-width. https://en.wikipedia.beta.wmflabs.org/w/index.php?title=Template:Wikipedia%27s_sister_projects&diff=369840&oldid=365653

I've gone a bit trough the issues in moment's bug tracker, and the reason they don't have this as a native property, is because the javascript dates don't have this (everything is unixtime related). Adding something like it could complicate calculations etc. which is not desirable.

Thanks for reporting, I accidentally broke this a couple of weeks ago.

@Nirzar Why not two buttons on either side of the diff, have it like caroussel/gallery pagination ? Seems much more aligned with mobile web UX concepts.

My ActualLivePreview script also needs a way to hook into the change event of the textarea, and it's missing this now.

Related or duplicate of T171374: CodeMirror shows ULS icon in wrong positions

state of this ?

FYI, this also happens when you click to position in the textarea, before CodeMirror had the chance to load. Especially common on slower machines, or on slow connections.

Note that popups broke during this weeks train T174724: [Regression] Page previews stopped working on wikis where jQuery 3 is not available, impact of that will need to be checked.

It's using jQuery 3 specific catch on deferred, but that isn't available on all wikis yet pending T124742: Upgrade to jQuery 3.

I have now tested this and the preview does indeed implement origin-when-cross-origin as directed by our meta referrer header.

Feedback from the field:

I think i'm personally in favour of moving it into either Kartographer, or a separate extension..

For now, Special:Nearby is still part of mobile frontend btw..