This page contains discussions that have been archived from Village pump (technical). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.
Ok, I don't see any "shrinking" in mobile view for the Backrooms article. What do you mean by that statement? Izno (talk) 00:00, 17 April 2024 (UTC)
When logged in, it’s normal. When logged out it is smaller,skinnier. Lines separating director, genre etc are gone. Words are squished together. Key limes (talk) 00:05, 17 April 2024 (UTC)
@Jon (WMF), I can reproduce this in a private window. hacks.less seems not to be loading for mobile right now? The styles show up when I plug ?debug=1 into the URL but not without. Izno (talk) 01:39, 17 April 2024 (UTC)
I need a bit more information for this one. Can someone upload a screenshot or a video of the issue? And just to double confirm we are talking about the mobile domain in an incognito window, right? e.g. https://en.m.wikipedia.org/wiki/Backrooms_(web_series)
Since Izno could fix this with debug=1 I'm currently inclined to think this is some kind of caching problem that can be fixed by ?action=purge but I want to make sure there isn't a bigger issue here.
Yes, that page looks correct now. I'm kind of wondering how this could have been caused by caching I guess, since the rules were added/removed at the same time, targeting the same ultimate element. Izno (talk) 20:48, 17 April 2024 (UTC)
Both of these articles feature a template ({{Excerpt|) that is meant to display the material entered on a third article (Meryl Davis and Charlie White). The section marked Competitive highlights displays properly, while the section marked Detailed results does not. The templates appear to be identical. Please advise, and thank you! Bgsu98(Talk)20:53, 17 April 2024 (UTC)
It's been a few years now that this hasn't worked User:TheDJ/Print options..... Should we not remove this from our Gadgets€preferences under gadgets? It doesn't come up often but a few times a year I see people inquiring about this at the help desk. Just wondering if there's routine testing of our gadgets to ensure they're still functioning? Moxy🍁 03:29, 17 April 2024 (UTC)
Load the mobile version of this album on a cell phone. The section title "Song overview" displays vertically along the left side of the page, over top of the professional ratings. This does not duplicate on a desktop machine whether you load the mobile version or not. Home Lander (talk) 23:02, 17 April 2024 (UTC)
Getting table captions to render with conditional statements
Per MOS:DTAB (and general accessibility best practices), all data tables must have a table caption. {{Linescore}} is a template that generates a data table and it ostensibly has a parameter Header that should generate one, but in reality, it does not. See Lee Richmond's perfect game where the HTML is:
line 82:<caption><span class="nowrap"> </span>
line 83:</caption>
rather than actually generating the caption "Line score for Lee Richmond's perfect game". I briefly made a fix for this that just removed some if/then statements and I previewed it on many pages (2001 World Series, Disco Demolition Night, 2009 Claxton Shield, box score (baseball)) and no problems appeared, but since there are 1,400+ transclusions, I wanted to be conservative so I reverted myself and came here to see if anyone thought there was a better solution or saw any problems with my approach. If anyone can debug this as-is, that would be great, but I can't and it seems like my solution actually works. ―Justin (koavf)❤T☮C☺M☯01:59, 18 April 2024 (UTC)
New posts not showing up as separate posts on a user talk
Myself and one other editor have posted on Jweaver28's user talk and I just realized our posts are showing up all mashed-together. Any subsequent posts will also show up all mashed together under the wrong header. Someone with more bandwidth than myself please take a look at User talk:Jweaver28 and the oddness going on at Unblock request for library wifi address. I've tried and tried to figure out how to fix it and have given up. Thanks, Shearonink (talk) 18:35, 18 April 2024 (UTC)
Something got messed up with the unblock request template on that page, and it had an unclosed <nowiki> tag. I fixed the markup: [1]Matma Rextalk18:46, 18 April 2024 (UTC)
Thank you! Those darn unclosed tags...At first I thought it might be some stray }} but that obvi wasn't the issue. Thanks again, Shearonink (talk) 19:12, 18 April 2024 (UTC)
Sometimes, a blocked user doesn't read the instructions properly and screws up the Wikimarkup when attempting to request unblock. --Redrose64 🌹 (talk) 21:15, 18 April 2024 (UTC)
User page doesn't contain a language
Resolved
Hello, I have 2 user pages, one in Arabic and another one in English, but the problem is that when I click on the languages tab on any of them it says "No languages yet"
6 weeks ago I did a deep dive for 1-2 days helping to debug, reproduce, identify the root case and help formulate a fix for the following issue. [2] This issue affects any user using webauthn aka 2FA security tokens across wikimedia. @Reedy (WMF) was very responsive and back-ported a fix from another repository. The fix was committed March 5.
my issue is more about getting help from WMF to deploy the fix that is ready to go. maybe this is better for village pump WMF Tonymetz💬21:08, 19 April 2024 (UTC)
Patches are primarily deployed by volunteer developers not "the WMF". There are currently over 70 tasks open for OATHAUTH alone, most of which are significantly older. This isn't anything specific to the English Wikipedia, so none of our noticeboards are the right venue. You could email the mailing list that has a team of volunteers that look over technical matters relating to MediaWiki software and interface - this team can be reached at mediawiki-llists.wikimedia.org. — xaosfluxTalk22:25, 19 April 2024 (UTC)
Patches are primarily deployed by volunteer developers not "the WMF". It might be more accurate to say that patches are auto-deployed by the weekly train for most repos. And a combination of volunteers and wmf do both the patch writing and the patch approvals. What's needed here is a patch approval (+2). Which can be a lot of brain power, because it involves wrapping your head around the ticket and the code, reviewing the patch code, and loading up the patch code in a localhost environment and testing it. But sadly, many repos are backlogged or do not have active maintainers or do not have any wmf team currently assigned to them. –Novem Linguae (talk) 22:58, 19 April 2024 (UTC)
Is there more information on how to apply to be a maintainer? I had assumed we were blocked by WMF. I’m happy to help contribute as a maintainer if it would help. Tonymetz💬23:26, 19 April 2024 (UTC)
First, you'd need to write a bunch of patches in gerrit for the repo you want to apply for. Do you have a gerrit account yet? Then you'd also want to get the support of the existing maintainers. Seems like an existing maintainer is a little frustrated with you in the linked phab ticket, so would need to mend that. Finally, when those two things are looking good, you'd ask the community for +2 of that repo via a Phabricator ticket. Example. –Novem Linguae (talk) 23:57, 19 April 2024 (UTC)
They don't quite make sense to me and don't seem to be telling me anything useful about the edit. Anyone care to explain? Thanks, Shearonink (talk) 19:38, 18 April 2024 (UTC)
Please always say which page or code a question is about. I guess it's about User:Penny(Cassopeia) where you used {{Portal image banner}}. Like many image-displaying templates, it takes raw file names as parameters and not image code: {{Portal image banner|Zinnia elegans with Bombus 01.JPG|...}}. PrimeHunter (talk) 19:24, 20 April 2024 (UTC)
Process template parameter in Lua before MediaWiki does its thing
Could there be a stopgap solution for the Graph module until its replacement is developed?
It has been 1 year since Graphs were disabled due to security vulnerabilities in the extension. The update plan as of now is to completely replace the Graph extension, because the other solutions to upgrade it were found to not be feasible. This process has only just started to gather the members necessary and will not start in earnest until July, per the update posted. This extension will then obviously take time to develop. In the meantime, there's 18k articles, per Category:Pages with disabled graphs, with data that is completely inaccessible unless you view the page source. We could either put the data in a table so at least the data is there, in an expando if there are many rows. We could even have a link to an external graphing service that graphs it visually on a separate page (Quickchart.io being an example that works simply, simply putting the data in JSON format in the URL.). Thoughts? MarkiPoli (talk) 15:48, 19 April 2024 (UTC)
The database on which this query was executed has a synchronization delay with the wiki. This can be caused by maintenance or incident on database, and should be resolved soon.
The modifications that was made in last 20 hours on the wiki are not taken into account in results bellow.
Just bellowing here to make sure that someone in charge is aware of this, and working to resolve it. Every time I've checked in for the past several hours, the lag has grown longer; there is no sign as yet that it's started to get shorter again. I searched for a Phabricator, and couldn't find any. Would like to have some idea of the meaning of "soon" in this case. wbm1058 (talk) 14:08, 17 April 2024 (UTC)
"FWIW, this is a data corruption issue. Last time it happened on sanitarium hosts the whole system went down for a week. I'm trying to figure out if I can avoid re-cloning the whole host (which would take a lot of time, potentially weeks) and only reclone the corrupted table but there is no easy way to do this AFAICS." wbm1058 (talk) 10:51, 22 April 2024 (UTC)
The top listings are from yesterday, the 15th. I have made two deletes today that are not showing up there. The change is usually instantaneous if I reload the grid page. — Maile (talk) 18:41, 16 April 2024 (UTC)
And now it's up to over 4 hour lag and my Quarry and bot reports are not updated. I hope this is not a maintenance tasks that takes several days. Work will just pile up. Thanks for pointing out the message from yesterday. I assume this will affect database and bot reports and not article content, right? LizRead!Talk!22:03, 16 April 2024 (UTC)
I think that's a good idea. Do you know how to do that? This might affect even more processes than we've mentioned above. — Maile (talk) 22:36, 16 April 2024 (UTC)
@Maile66:This page is a good start. You should follow the instructions there to create a Phabricator account, then you can create a bug report. If you need me to, I can, and I can give you the link. If you or Liz does, please put the link in this thread, I'd like to keep track of it. thetechie@enwiki: ~/talk/$22:44, 16 April 2024 (UTC)
Thanks I added my issue. Seems I already had a Phabricator account. It's been such a long time since I was over there, I had forgetten. — Maile (talk) 23:39, 16 April 2024 (UTC)
Yes, my primary participation is subscribing and following problems not filing reports. The discussion there goes over my head once it stops being English and starts into code. LizRead!Talk!01:15, 17 April 2024 (UTC)
As of today, April 24, the issue has returned. I have filed Phabricator Task T363077, hoping this resolve it once and for all. — Maile (talk) 21:03, 21 April 2024 (UTC)
This issue cannot be resolve[d] once and for al[l]. There will be spurts of replag lasting a few hours to a day from time to time, for the foreseeable future. * Pppery *it has begun...21:10, 21 April 2024 (UTC)
And my understanding is that these Phabricator tasks are not helpful - with only one exception I can think of replag has been due to deliberate database maintenance that the people involved are aware of the impact of, and it will resolve itself when its done. * Pppery *it has begun...21:15, 21 April 2024 (UTC)
Well, looks like I was mistaken and this report was helpful, and this replag instance was a real standalone bug not an expected consequence of database maintenance. So, I guess you can report it, but sometimes you will get "this is a known issue that the DBAs are working on" ("DBA" is an acronym for database administrator). * Pppery *it has begun...21:29, 21 April 2024 (UTC)
Thank you for all this insight. I have previously come across "this is a known issue ... " and usually just back off on those types. — Maile (talk) 21:41, 21 April 2024 (UTC)
If you go to ?action=info for any page, you will see a table with various statistics, including two lines about how many people are watching the page, e.g.:
I am finding that the ratio shown here is not at all unusual for older articles, but the first line gets more attention from editors. They think "hundreds of editors are watching this page", when they should be thinking "almost nobody is watching this page". Is there a way we could remove/hide the irrelevant number from this info page? Or should we just change MediaWiki:Pageinfo-watchers to something like "Total number of watchlists (includes inactive editors)"? WhatamIdoing (talk) 04:53, 10 April 2024 (UTC)
I’m one of those who ‘think "hundreds of editors are watching this page", when they should be thinking "almost nobody is watching this page"’ ... I think we should have that “second line” added to these pages (or replaced the “first line”):
I bet someone here could write a .css script that would blank that out, or rename it to something like "This is the wrong line – ignore it". I checked a bunch of Special:Random pages, and most of them showed no data, due to there being too few people watching the pages. WhatamIdoing (talk) 21:11, 11 April 2024 (UTC)
For hiding that row on the page information page on Wikipedia, try #mw-pageinfo-watchers { display: none; } in one of the .css files. –Novem Linguae (talk) 22:43, 11 April 2024 (UTC)
Thanks, @Novem Linguae. That seems to have worked for me. That suggests that iff we ever decided that we wanted to do that globally, it could be done in (e.g.,) global.css. I'm going to try this out for a while. I suspect that I'll like it. WhatamIdoing (talk) 22:32, 12 April 2024 (UTC)
If it's on <30 watchlists, no number is given for either item (the second item is simply suppressed).
If it's on ≥30 watchlists, an exact number is given for both items.
If the second number is zero, it says "There may or may not be a watching user visiting recent edits" instead of a second number. WhatamIdoing (talk) 23:53, 10 April 2024 (UTC)
Is there a way we could remove/hide the irrelevant number from this info page? I would not support removing this. Presenting both numbers, and letting the user decide which they want or need, seems like an acceptable status quo here. The less than 30 thing for non admins is for security reasons. Admins can see both numbers at all times. The linked phab ticket mentions changing the second message to mention 30 days explicitly. I could get behind a change like that. –Novem Linguae (talk) 02:05, 11 April 2024 (UTC)
Under what circumstances do you think it would it be useful for to you to know that the watchlists associated with 1,991 mostly inactive (and sometimes actually dead) include the defunct Wikipedia:WikiProject Contents?
The number of active editors presently watching that page is a single-digit number. I can understand why that number would be useful to know, but not why the first has practical value. WhatamIdoing (talk) 02:56, 11 April 2024 (UTC)
As far as I can tell, pages can be watched actively, without the user being considered active there - such as through email or syndication. — xaosfluxTalk08:45, 11 April 2024 (UTC)
I've never spent much time reading about the internal mechanism, but I believe that it works like this:
If you have some pages on your watchlist, and then get permanently locked out of your account, you are "a watcher" for all of those pages forever, even though you can't actually watch anything from that account.
If you have some pages on your watchlist, go to Special:Watchlist, close the tab without clicking on any link or visiting an article at all, and then get permanently locked out of your account, you are counted as "an active watcher" for all the pages on your entire watchlist for the next 30 days (including pages that did not have any changes made, so they weren't listed at Special:Watchlist on the day that you visited that page).
This means that there are 9 active editors with this page on their watchlist, of which an unknown number – but it is quite possibly zero – actually looked at the page in question during the last month.
So Xaosflux says, yes, there may only 9 editors who have that page on their watchlists and actually went to Special:Watchlist at any point during the last 30 days, but maybe a few more people also get e-mail messages about changes to articles on their watchlist, so the "9" might be a slight undercount.
The scenario you describe – visiting the watchlist but not checking every page – is certainly common. The "9" is probably an overcount, if the goal is to know whether anyone actually checked the specific page. WhatamIdoing (talk) 02:54, 15 April 2024 (UTC)
You're correct. The count uses the same "first unseen timestamp" data used for highlighting of unseen edits on the history page. If a revision older than the configured age would be highlighted, the user is not counted as having visited recent edits. Viewing the watchlist or history page doesn't reset that, while visiting the page itself will and viewing old revisions or diffs may. Anomie⚔11:41, 15 April 2024 (UTC)
Thanks. I have corrected the errors I introduced last week to Help:Watchlist ("Number of page watchers: 3,644; Number of page watchers who visited recent edits: 29; Number of editors who noticed my error: 0").
The "visited recent edits" should probably be changed to "visited the page recently". I think most editors will interpret this as "checked a diff" (i.e., the edit) instead of "clicked on the article" (e.g., displayed the page via Special:Random). WhatamIdoing (talk) 16:42, 15 April 2024 (UTC)
Personally, I would interpret "visited the page recently" to mean "visited the current version of the page recently". I think the natural assumption is that "page" without a qualifier refers to the current version. isaacl (talk) 16:59, 15 April 2024 (UTC)
I think referring to a time-bound period does change the natural assumption to having viewed the page as it appeared at the time of viewing during the specified period. I don't think it would be interpreted as having checked a diff. isaacl (talk) 16:56, 20 April 2024 (UTC)
I like the idea of having clarity.
BTW, I'm finding that I'm happy with the script @Novem Linguae wrote for me. I'm still sometimes startled to see that there's only the one line, but it gives me the information that I want, and if for some reason I need to see the number of inactive accounts, it's only one incognito/private window away. WhatamIdoing (talk) 19:11, 22 April 2024 (UTC)
Why does the talk page visual editor not have a "cite" button?
Am i missing something, or is it impossible to automatically cite urls from the talk page visual editor? Yes, you can use the source editor, but is there a specfic reason why this is? Is there any way to get a more featured editor on the talk page visual editor? MarkiPoli (talk) 12:14, 16 April 2024 (UTC)
Not sure if it's intentional, but the keyboard shortcut Ctrl+Shift+K (Cmd+Shift+K on macOS) still works for this. the wub"?!"13:17, 16 April 2024 (UTC)
Yeah, i noticed that too. Its still annoying not having a full featured editor exactly the same as the main page editor MarkiPoli (talk) 09:39, 17 April 2024 (UTC)
There's some things that're not allowed there because they don't work well in list-indented discussions, but mostly it's because of space constraints for the toolbar. The compromise is that the things left out are more on the power-user end (e.g. citations are very rarely used in discussions), and we hope that power-users will be able to use keyboard shortcuts and sequence triggers for these things. (You can press ctrl/cmd + ? to see all the options here. Or type \ if you're in visual mode for a nifty command-palette...) DLynch (WMF) (talk) 16:04, 17 April 2024 (UTC)
If you want to make a citation so that others can easily see it and potentially copy paste it into the article if useful (if its part of a discussion and can't be immediately put in the article for some reason i.e. its contentious). Also, a gaping flaw that's worse in my opinion is there's no real way to even insert templates at all in the replier (it literally says that wikitext is not allowed in the visual reply editor when it doesnt say this in the article editor). Is there any way at all to insert templates in the visual talk page editor without switching to source? MarkiPoli (talk) 19:49, 22 April 2024 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
You can nominate your favorite tools for the fifth edition of the Coolest Tool Award. Nominations will be open until May 10.
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 23 April. It will be on non-Wikipedia wikis and some Wikipedias from 24 April. It will be on all wikis from 25 April (calendar). [4][5]
Future changes
This is the last warning that by the end of May 2024 the Vector 2022 skin will no longer share site and user scripts/styles with old Vector. For user-scripts that you want to keep using on Vector 2022, copy the contents of Special:MyPage/vector.js to Special:MyPage/vector-2022.js. There are more technical details available. Interface administrators who foresee this leading to lots of technical support questions may wish to send a mass message to your community, as was done on French Wikipedia. [6]
Cursor keeps teleporting to the very start of the box I'm editing whenever I press shift
Is there any way to make the cursor stop teleporting to the beginning of the line or paragraph if I press shift? Shadow311 (talk) 00:47, 20 April 2024 (UTC)
@Shadow311, if you didn't have the GoogleTrans gadget enabled, then it might be phab:T316838. Please ping me if that's the case.
If you do have GoogleTrans enabled, and you need to keep it enabled, then there's a chance that you could bypass this bug, in the Reply tool and New Topic tool only (e.g., not in the visual editor, which you don't seem to be using) by going to Special:Preferences#mw-prefsection-editing-discussion and turning off "Enable editing tools in source mode". You'd lose the whole toolbar, including the button to @ ping someone. WhatamIdoing (talk) 22:52, 22 April 2024 (UTC)
Is there a convenient functionality to merge/extend Lua arrays?
I know I could iterate over the second array and append the values at the end of the first array, but I thought "hang on, there must be a closed function/method like Python's list.extend defined somewhere"... Am I right? Alexis Coutinho (talk) [ping me]20:30, 22 April 2024 (UTC)
I had checked Module:TableTools to no avail. I guess I could use the "union" functions from Module:Set, but that seems like overkill as those are not optimized for arrays. I think such simple function should be added to Module:TableTools. Ideally, it should actually be in the standard Scribuntu table library, alongside table.concat... Alexis Coutinho (talk) [ping me]22:26, 22 April 2024 (UTC)
That's one of the functions in the Lua standard library, not added by Scribunto. Your other options besides the modules are 1) iterating over table 2 and table.inserting into table 1, or 2) doing it with indices. Izno (talk) 23:11, 22 April 2024 (UTC)
When I click on an IP address's contributions and click on the IP information window, it shows The IP information could not be retrieved. even though it works on other wikis, such as the Simple English Wikipedia.
This issue can only affect users who use iPhones and not on computers, so is anybody who uses this beta feature experiencing this issue when using it on their iPhone (including me)? Codename Noreste 🤔 La Suma03:34, 22 April 2024 (UTC)
Does the tool work on the Simple English Wikipedia when on your iPhone?
I don't have an iPhone, so I cannot personally test this. The tool displays the The IP information could not be retrieved. error on any kind of "generic" error, which doesn't have a defined message. This could suggest that the request has not been getting through to the servers. Dreamy Jazztalk to me | my contributions10:06, 22 April 2024 (UTC)
Smells like an API error. Are you using a browser on your phone, or the official Wikipedia iOS app? Are you still getting the error right now? Is it one IP only (please share which one if you can), or all IPs you try? –Novem Linguae (talk) 11:00, 22 April 2024 (UTC)
Can you provide a screenshot so that we can see your skin, screen size, what page you're on, if you're logged in, and what IP you're trying to view? Also, does the problem go away in WP:SAFEMODE? –Novem Linguae (talk) 20:33, 22 April 2024 (UTC)
The screenshot is here. About the IP addresses, this do not matter since the error applies to every IP addresses' contribution pages, and for safemode, the issue still persists. Codename Noreste 🤔 La Suma22:47, 22 April 2024 (UTC)
@Xaosflux: Which guideline? "IP information" refers to "information obtained through this tool", so nothing covered by the guidelines is disclosed in the screenshot, no? Nardog (talk) 10:21, 23 April 2024 (UTC)
Oops, too many screens open, the screenshot with the successful results would have been the problem, but that is in phab not here. — xaosfluxTalk10:25, 23 April 2024 (UTC)
Thanks. I've filed phab:T363118 for you to help get dev attention. They may be able to search server logs to help see what the bad API query is. Please reply to devs if they ask you questions there since they are the most likely to be able to debug this. –Novem Linguae (talk) 23:55, 22 April 2024 (UTC)
To your question, my answer is that I edit from the Safari browser in my iPhone and I don't have iCloud+, so iCloud Private Relay is not enabled for my iPhone. Codename Noreste 🤔 La Suma00:21, 23 April 2024 (UTC)
I think SoY's question is if you get a block notice when you try to edit articles on this device. SoY suspects that you might be editing from a blocked IP range and wants to rule it out. –Novem Linguae (talk) 00:22, 23 April 2024 (UTC)
I got a revdel request for removal from page history of some relatively innocuous adolescent pranks from a couple of years ago. The requester is concerned that it could be found by search engines. I haven't been able to replicate it, and I seriously doubt that search engines care about history like that. In order to reassure the requester that they can quit worrying about it, can somebody clarify whether and to what extent article histories are or aren't indexed or considered by search engines? Acroterion(talk)13:49, 23 April 2024 (UTC)
Nearly everything that starts with /w/ is disallowed to be indexed by search engines in our robots.txt (see what is robots.txt). E.g. google site:en.wikipedia.org inurl:/w/index.php. There could be theoretically ways to circumvent this, but I haven't found any Google results for old versions and diffs. Jack who built the house (talk) 14:07, 23 April 2024 (UTC)
Diffs and old revisions also have noindex in the HTML so Google and other compliant search engines will not index them if they come across them in other ways at Wikimedia sites. PrimeHunter (talk) 14:44, 23 April 2024 (UTC).
Thanks, that's what I thought. I'll reassure them, since their request really doesn't meet revdel criteria. It's the kind of silly thing I'd have done at 16 if we'd had an editable online encyclopedia. Acroterion(talk)15:10, 23 April 2024 (UTC)
Early access to the dark mode (mobile web, logged-in)
Hi everyone, as announced in November, the Web team at the Wikimedia Foundation is working on dark (sometimes also called night) mode. Now, we have released the feature for logged-in users of advanced mobile mode across all wikis for testing purposes. But don't worry, the new feature is not disruptive! (See the "known limitations" section below.) It's just important for us to work together with you before we release this feature to a wider audience. Our goals for the early rollout are to:
Show what we've built very early. The earlier you are involved, the more your voices will be reflected in the final version
Get your help with flagging bugs, issues, and requests
Work with technical editors to adjust various templates and gadgets to the dark mode
Go to the project page and the FAQ page to see more information about the basics of this project.
Known limitations of the initial release
Currently, dark mode is only available on mobile, for logged-in users who have opted into advanced mode, as an opt-in feature.
Gadgets may initially not work well with dark mode and may have to be updated.
Our first goal is making dark mode work on articles. Special pages, talk pages, and other namespaces have not been updated to work in dark mode yet. We have temporarily disabled dark mode on some of these pages.
What we would like you to do (the broad community)
If you have questions - ask us! Also, where appropriate, consider linking to the Recommendations for dark mode compatibility on Wikimedia wikis. We would like to emphasize that the recommendations may evolve. For this reason, we are not suggesting to create your local wiki copies of recommendations. At some point, the copy could become different from the original version.
What we would like you to do (template editors, interface admins, technical editors)
When most bugs are solved, we'll be able to make the dark mode available for readers on both desktop and mobile. To make this happen, we need to work together with you on reporting and solving the problems.
To turn it on, use the mobile website and go to the settings part of your menu and opt into advanced mode, if you haven't already. Then, set the color to dark. (Later, we will be allowing the device preferences to set dark mode automatically).
Next, go to different articles and look for issues:
If you have noticed an issue with a template but do not know how to fix it
Would it solve the problem if the STV candidate template used the shortname in all cases? — Martin (MSGJ · talk) 20:24, 22 April 2024 (UTC)
I tried to do that with this edit, but ended up breaking the links in all cases. Could you see a means to include the shortname linked if there is an article, not linked if there is not? Iveagh Gardens (talk) 06:50, 23 April 2024 (UTC)
Thanks, that does fix it, although I'm not sure the existence of a page is required. It can be useful for once-off microparties that have become defunct, as with Seniors Solidarity at 2009 Fingal County Council election, to have an entry there, without needing a page. In ten years, if The Irish People is no longer registered, it might not even warrant a redirect. Iveagh Gardens (talk) 06:48, 23 April 2024 (UTC)
Anyone know why this is happening?
This isn't really a problem, per se, but I'm just curious if anyone has a similar thing going on. When I go to Template talk:Did you know (and only there; I've tried this on other talk pages of templates as well and all of the other namespaces work fine), my toolbox on the left side of my screen (I use vector legacy dark mode) turns white. I tried switching to light mode, and nothing happens. Is anyone experiencing something similar?
USS Triton (SSRN-586) is currently showing as cascade-protected, claiming that it is transcluded on the main page and various related pages. While it is linked from Wikipedia:Main Page/Tomorrow, it does not appear to be transcluded anywhere (and pages that are just linked shouldn't be cascade-protected). Is this a known bug of some sort? :Jay8g [V•T•E] 00:25, 24 April 2024 (UTC)
Request: To provide /update pictorial, video or tutorial help on how to use Template:Citation#Quote.
The policy regarding these types of quotes is briefly covered at WP:FOOTQUOTE.
Description: In the visual editor, under the 'cite' button, you first put in the basic parameters (sometimes just a URL is sufficient to get started), then scroll down the list of fields to find the one named 'Quote', which is a text box you can enter the quote into. (This is too verbose does not seem to save first time readers from confusion)
Why needed:
as a help in verification if needed
WP:VERIFY states "..Readers must be able to check that any of the information within Wikipedia articles is not just made up. .."
When we refer to academic sources many times those are not easily accessible to readers / copy editors and also who wish to verify. Specially in contentious topic area providing original quote from the source remains easier for transparency and verification and saves a lot of misunderstanding, stress and time.
When some user urged me to provide quotes from sources it took me months to understand what that user was urging about and how to do that although it is very easily accessible in visual editor.
When I also urge other editors I find they are confused what I am requesting all about.
I do have strong perception that updating related help pages with pictorial, video or tutorial help can save lot of hassle.
Idk if this is the right place to seek help in this respect, if not then please guide me where to make this request.
Also, quotations are not sufficient. If an editor were willing to put false information in the article, then that editor could just as easily put false information in the quotation field of the template. WhatamIdoing (talk) 22:58, 22 April 2024 (UTC)
Thanks for introducing me to WP:RSC and also share your concerns and part of it is certainly matter for consideration under project WP:REFCHECK. I wish to discuss these concerns further at WT:REFCHECK after couple of months since presently I am going bit busy.
This appears to be coming from the Android app. I can't find the option to try it out, even in the Alpha version, but at first glance this seems to be a privacy violation. We've never publicly logged pages that people were just viewing (or previewing an edit for) but decided not to edit. Is it being disclosed, up front and clearly, that the choice not to edit will be visible to the whole world? Suffusion of Yellow (talk) 21:03, 17 April 2024 (UTC)
@DBrant (WMF): How can I try out this feature? All I see under "suggested edits" is "Article descriptions", "Image captions", and "Image tags". And is my assumption correct, that we are logging when someone decides not to edit a page? If so, are they told that their username, along with the page they were viewing, is being publicly logged? Suffusion of Yellow (talk) 21:36, 17 April 2024 (UTC)
Alright, for some reason it doesn't work for me on the enwiki or testwiki, but I found it on the desktop site at es:Special:Homepage. I was told that Your answers improve future suggestions. but for all I would know, that just means I'm training an AI or something. There was no indication that by clicking "Submit", this log entry would be created, and my name would be publicly recorded. This feature needs to be disabled. Suffusion of Yellow (talk) 23:08, 17 April 2024 (UTC)
This is a "growthexperiment" log type. Pings to @Trizek (WMF) and KStoller-WMF: who can hopefully point to the documentation. Notably, @Suffusion of Yellow: this doesn't say it is about "viewing" but about "reviewing" so I'm assuming there was an affirmative user action performed here other than reading. — xaosfluxTalk00:07, 18 April 2024 (UTC)
You're asked if an AI-selected image should be in an article. If you click "Yes", you're given an opportunity to add it. If you click "No", you're asked why, but not told that your "No" is going to be made public. This is equivalent to logging every time someone previews an edit, and then abandons it. Suffusion of Yellow (talk) 00:17, 18 April 2024 (UTC)
I think that is a bit of a stretch. Patrolling essentially is similar in nature and we don't warn from the interface that those actions are publicly logged either. I do see however that the barrier to entry is significantly lower here, so some additional caution might indeed be warranted. —TheDJ (talk • contribs) 08:28, 18 April 2024 (UTC)
Thanks for the ping! The Android app is indeed using the same underlying API as the Growth team, but their implementation is a little different: Android/Image_Recommendations I'll follow up with that team and ask them to review this discussion. Thanks, - KStoller-WMF (talk) 16:51, 18 April 2024 (UTC)
@DBrant (WMF) is under the weather, I think I've coordinated with each of you before at some point, but I am Jaz the PM for the apps.
"Reviewed an image suggestion" means that someone determined that an Machine suggested image for an article is either adequate for an article, and the subsequently adds that image OR the image is not adequate (rejected). The reason it is important to track is so that we can remove bad recommendations from the stack that we share with the Growth experiments. Do we need to publicly log this information? Probably not, I am investigating why we are showing rejecting a suggestion publicly, if there isn't a super compelling reason, I will ensure we suppress the public log but still send the API call that is necessary for the feature.
As far as accessing the feature, you'll need to be:
Logged In
Set your app primary language as one where you have more than 50 unreverted edits
If you follow these steps and are still unable to see the feature, would you kindly file a task and screen record the steps you are taken, because that would be a bug. The feature is available in both the Beta version of the app and the production version.
The requirement of having over 50 unreverted edits for the language wiki you are using the feature in is the implementation difference that @KStoller-WMF. Under the hood and the flow of the feature is the same, however who has access to it is different. The Growth implementation is available to new editors on select wikis and we use the same API. New editors do not have access to this suggested edit task on the app.
The onboarding and guidance for the feature on Desktop is not the same as on Android. For example, on this screen we do let users know that their answers improve future suggestions.
Once DBrant is back in office (fingers crossed he's feeling better tomorrow) our team will update this thread about removing the rejection log. JTanner (WMF) (talk) 19:05, 18 April 2024 (UTC)
Thanks. I don't like putting using my main password on my phone, so I had used an alternate account with only a few edits. Honestly I spoke too quickly when I said "this feature needs to be disabled". So long as the messaging is changed to be more like MediaWiki:Thanks-confirmation2 ("Publicly send thanks?") I don't think the log needs to be removed. The problem with "Your answers improve future suggestions" is that it sounds like I'm training an AI ("select all squares containing Sarah Connor"), or taking an anonymous survey. Suffusion of Yellow (talk) 19:42, 18 April 2024 (UTC)
Thanks @Suffusion of Yellow this is helpful, digging into the code we were able to determine the rejection log is coming from the Growth Experiments API, which means the rejection log publishes for the Growth Experiment implementation as well. We are going to coordinate with them on next steps to see what makes sense regarding the public log. @ARamadan-WMF will follow up with the link to any phab tickets should they be created to make the rejection log private.
In the meantime, I've created task T362935 to update the copy in the UI for the app should the decision be to keep it public. I've subscribed you to it @Suffusion of Yellow, I welcome all of you to subscribe and comment on the proposed language change. JTanner (WMF) (talk) 20:37, 18 April 2024 (UTC)
This is Amal Ramadan, I am Sr. Movment Communications Specialist supporting the Mobile Applications team; I wanted to provide you with an update: both the Growth and the Apps teams are going to discontinue publicly logging rejections. If you're interested in following the progress of this change, please subscribe to ticket T363002.
Thank you for initiating and engaging in this conversation.
Hi @ARamadan-WMF another question. If they add a suggested image, the tag "#suggestededit-add-image-top" is included in the edit summary (example). Are both logging and tagging really needed? It seems like overkill (maybe?). If both are not needed, then I suggest using the edit summary as it is more transparent to editors than it being buried in logs. S0091 (talk) 18:26, 23 April 2024 (UTC)
Actually, we are aware of this, and actively working on making the change; the work progress on this can be found in this ticket T360164, I tried to sign you up, but couldn't find your account. You can follow to stay updated instead.
Google will sometimes prioritize amp links in mobile search results so I see them being passed around. At least it's easy to find the original URL. Skynxnex (talk) 02:35, 24 April 2024 (UTC)
These query strings are for server-side tracking, usually they may be safely removed. I'm sure that I've seen a bot removing them. I don't think that VE can be blamed at all. --Redrose64 🌹 (talk) 16:52, 24 April 2024 (UTC)
Easier way to mark an edit as minor
The minor edit button is very small and not very mobile or laptop trackpad friendly. Its too bothersome for me to actually use it. I do mark these edits as minor in the edit summary with the letter "m". Leaving "m" in the edit summary should tag the edit as a minor edit. This would encourage use of the minor edit button for minor edits and help declutter the watchlist (for those that filter minor edits from their watchlists). How easy would this be to from a technical standpoint? Schierbecker (talk) 18:33, 22 April 2024 (UTC)
I can't help with the change you're proposing, but in case it helps, there are actually some more accessible ways to toggle the "minor edit" checkbox: you can click (or tap) its textual label as well, not just the box; or on a laptop, while writing the edit summary, you can press Tab and then Space to move to and toggle the checkbox. This works with all other checkboxes on Wikipedia too, and also in most other websites and apps. Matma Rextalk21:10, 22 April 2024 (UTC)
Ticking the minor button is a "just trust me" move. Just trust me: it's minor, and nobody with the default watchlist settings needs to pay any attention to this... WhatamIdoing (talk) 19:55, 24 April 2024 (UTC)
What am I missing (translcudes redlink that doesn't appear in source text)
Yeah it's building the link by adding 'Deputy' to the wikilinks. If you put Test in office2 it will try to use Deputy Test. But why does it appear as transcoded? Wikilinks aren't transcoding, and I can't see why it's happening in the infobox code. -- LCU ActivelyDisinterested«@» °∆t°19:15, 24 April 2024 (UTC)
There are many layers there, it looks like this is coming from a nested template, that is then using a hack via {{Linkless exists}} to attempt to see if that page exists. In short, you can ignore it or develop an article for that title. — xaosfluxTalk19:47, 24 April 2024 (UTC)
The blocking IP addresses essay was linked from WP:AN recently (for reasons unrelated) and it had me thinking:
How will temporary accounts affect the potential blocking of sensitive IP addresses (i.e. IPs assigned to major government organizations and also 'may be a good idea to notify the committee' IPs like major corporations or ones with technical implications like WMF/WEduF IPs)? To quote that essay, it currently says "If you block an IP address in any of the following ranges, you are required to immediately notify the Wikimedia Foundation Communications Committee."
Is the procedure here going to be that every admin is supposed to check the IP address of temporary accounts to see if it's sensitive when blocking? Or does it not matter too much?
Huh. Didn't pause to consider what was done currently with actual accounts using these IPs... right so it's one account per cookie, even if they use the exact same IP... so that means, blocking a temporary account would only affect that device (until their cache was cleared) and autoblock the IP(IPs?) used (I'm not actually sure how long autoblocks keep autoblocking IPs for, but maybe that's intentional).
Not sure if that has public relations implications, I guess not. On that note, maybe if it turns out to be a problem to have temporary accounts (that will be mandatorily/automatically created on editing) using sensitive IPs being blocked then MediaWiki:Block-autoblock-exemptionlist could be used to mitigate that - looks like a lot of WMF/etc IPs are there already. – 2804:F1...7D:5C91 (talk) 22:26, 24 April 2024 (UTC)
I am pretty sure you will only be able to range block temporary accounts by blocking their ranges. So you will still bump into the standard "are you blocking a sensitive range?" Izno (talk) 22:25, 24 April 2024 (UTC)
I'll stop replying so much, but: The explanatory essay currently says if you block an IP in the ranges you should inform the committee, not if you block the ranges. This is mostly echoed in the actual policy WP:BP#IP address blocks.
Though as SoY said, this would already happen with an account (and only a CU with reason to check would know), so maybe this is a non-problem. – 2804:F1...7D:5C91 (talk) 22:42, 24 April 2024 (UTC)
And the explanatory essay was also written two decades ago. We can adjust as necessary. This is pretty low down on the list of "how do we deal with temporary accounts when that hits us". Izno (talk) 00:42, 25 April 2024 (UTC)
Why isn't the auto-archive working on this article talk page
Someone please with more mental bandwidth than I have right now, take a look at Talk:Lucille Ball and fix it? Why isn't the bot creating and then archiving to a Page 2? (and when I try to use the archive code-thingy I have installed, Archive 1 instead of 2 comes up...) Thanks, Shearonink (talk) 22:04, 24 April 2024 (UTC)
The archiving is setup to to archive anything older than 30 days, with a max archive size of 100k, and leaving at least one thread on the talk page (so it doesn't appear blank). There been nothing to archive since April 2022, and the current archive is still smaller than 100k so there's no need for a second archive yet. -- LCU ActivelyDisinterested«@» °∆t°22:45, 24 April 2024 (UTC)
Diffs: <archive>, <ANI>. The bot claims it archived 5 sections, and it did do that, but it removed 6 sections (one that had just been created). The post was recreated by the author, so there's no active problem - but is that a problem that happens with bots, or just this one? Or is this just a 'once in a blue moon' event? – 2804:F14:8092:9F01:61D4:EA42:137D:5C91 (talk) 02:58, 25 April 2024 (UTC)
Bots can detect edit conflicts, by setting baserevid or basetimestamp, see the API documentation. But if a bot only takes a fraction of a second between loading the base revision and making the edit, they can probably "get away" with not doing this before anyone notices a problem; what are the chances of an edit slipping in the gap? That edit was nine seconds after the previous one, so perhaps something was slowing down the bot. Suffusion of Yellow (talk) 03:33, 25 April 2024 (UTC)
I filtered the contributions of the bot a bit, for the last 50000 edits or so (I lost count, it was 5000 contribs at a time), found 4 other instances:
No supported authentication methods available when trying to log in to toolforge
I normally log in to toolforge using PuTTY and WinSCP, and a year or two ago I set it up (I don't remember how, except that it was a pain to do) to use a passphrase that checks a local key file on my PC. This has been working fine ever since. Today I got a message from WinSCP on connecting saying that the host key had changed (I forget the exact message). After a bit of research I trusted the host and connected successfully, using the passphrase as usual; I can see the usual directory structure for the tool I run. I then tried PuTTY and got the same message, said yes, and now the host won't let me connect. When I use the saved session I get "No supported authentication methods available" and then "Server refused our key". This surprised me because WinSCP is perfectly happy with the connection. Any help on how to get connected again would be much appreciated. Mike Christie (talk - contribs - library) 01:02, 25 April 2024 (UTC)
That article was already in that category before your changes. It is applied automatically by Template:GLS Presidents. The reason it is listed first is because that template is used before any of the other categories are defined — Martin (MSGJ · talk) 13:49, 25 April 2024 (UTC)
Well that would explain why I didn't see it was in that category, nobody would expect it to be listed before the dates. Is it common for templates to do that? It strikes me as unnecessary and confusing. — Preceding unsigned comment added by DuncanHill (talk • contribs)
WP:TEMPLATECAT says: "However, it is recommended that articles not be placed in ordinary content categories using templates in this way." But there are still many templates doing it. Some of them have a parameter for Wikipedia:Category suppression but not Template:GLS Presidents. You could use {{Suppress categories|{{GLS Presidents}}}} and manually place the category where you want it. PrimeHunter (talk) 14:50, 25 April 2024 (UTC)
@PrimeHunter: I tried that and got "template loop detected". @Jonesey95: I'd happily add them to the category, but haven't the faintest idea how to remove the category assignment code from the template. Something to do with includeonly or noinclude? DuncanHill (talk) 15:01, 25 April 2024 (UTC)
@Gonnym: so remove all of <includeonly>{{main other|[[Category:Presidents of the Geological Society of London]]}}</includeonly><noinclude>
[[Category:Presidents of the Geological Society of London| ]]
</noinclude>? DuncanHill (talk) 15:24, 25 April 2024 (UTC)
yes if you want to also remove the template itself from that category. If you want to keep the template in the category leave the noinclude section. Gonnym (talk) 15:26, 25 April 2024 (UTC)
{{Suppress categories|{{GLS Presidents}}}} is code which can be placed in an article using a template which automatically adds categories. Then the template wouldn't have to be edited, but the code would be needed in all articles where the category is unwanted. PrimeHunter (talk) 16:12, 25 April 2024 (UTC)
Redlinked category on js page
The latest run of Special:WantedCategories features a nonsense Category:+, generated by the use of [[Category:+]] text in User:HovigTheEditor/common.js — but since I don't have "editing other users' js pages" privileges I can't fix it. And while I don't particularly understand what the code is there for, it might very well be there for a perfectly valid purpose, but would still have to be coded in a way that it isn't causing the js page to become directly filed in a redlinked category itself since the text is obviously not intended to actually categorize things as plus-sign per se. So could somebody with the necessary privileges look into fixing this? Bearcat (talk) 18:00, 25 April 2024 (UTC)
Earlier today, I went to look up the article on Sonia Sotomayor, a current justice of the US Supreme Court. But to my surprise, as I started typing her name into the search bar, the search algorithm didn't seem to recognize her as a possible destination. By the time I had "Sonia So" typed in, the first suggestion was Sonia Soto (a two-sentence stub about a translator), followed by imperfect matches for the search term; once I'd typed in "Sonia Soto", it only suggested the translator; I didn't get pointed toward Sotomayor's article until I'd typed in her entire name. My first assumption was that the article title had a diacritic or something that would cause Sonia Sotomayor to be a redirect rather than the article title, but that turned out not to be the case.
I was curious whether this was a more widespread problem, so next I searched for Antonin Scalia, and had similar issues (when I typed "Antonin Sca", the first suggestion was Antonio Scarfoglio). Trying again, I searched for Joe Biden; when I typed "Joe Bi", the first suggestion was instead Joe Biden Supreme Court candidates. At this point, my suspicion was that the search algorithm was choosing to suppress the articles of political BLPs specifically - but then I found two counterexamples soon afterward. Bill Clinton showed up successfully as the first suggestion for "Bill C", whereas Reese Witherspoon was nowhere to be found when I typed "Reese With" into the search bar. Searching some other topics shows that it's not limited to biographical articles either: I had similar problems trying to search the articles for All Eyez on Me, Chemistry, and Norway, as well as for more niche topics like alkaline noodles. The search problem isn't universal, but it's affected the majority of the titles I've checked, across various obscurity levels and topic areas.
I've tested this on the desktop site and on the app, and found the same issue occurring on both platforms. I haven't been able to identify any clear commonalities between articles that are affected by this search issue. Does anyone know what might be going on? ModernDayTrilobite (talk • contribs) 15:05, 25 April 2024 (UTC)
FWIW, I am unable to reproduce any of the above problems. When I type the first few characters of the names of any of those people or articles, I get the expected article as one of the first two results (i.e. "chem" shows "Chemistry" as the second result). – Jonesey95 (talk) 16:28, 25 April 2024 (UTC)
For reference, the hardware I've tested on includes Google Chrome on Windows 7, and the Wikipedia app on an iPhone 13 Mini (running iOS 17.4.1). I just now tested the Wikipedia mobile website (using Safari on the same iPhone) and reproduced the issue there as well. ModernDayTrilobite (talk • contribs) 17:37, 25 April 2024 (UTC)
I'm getting the same behavior both in my Firefox Wikipedia search bar and on the desktop site. Some pages, like Abbott Elementary, don't get suggested unless the full title is typed in with correct capitalization. Jak86 (talk)(contribs) 18:33, 25 April 2024 (UTC)
The bug apparently depends on the datacenter you connect to, which depends on your geographical location. Using the WikimediaDebug extension, I can reproduce the problem when connecting to eqiad, but not when connecting to codfw. I filed this as T363516. Matma Rextalk18:59, 25 April 2024 (UTC)
I see the correct results for 'Sonia Sotomay' now, when connecting to either eqiad or codfw, and when using either rest.php or api.php. I think you might have looked when the index was in the middle of rebuilding, and rest.php may be cached more heavily. Can you try again in a private / incognito browser window? Matma Rextalk21:25, 25 April 2024 (UTC)
I only browse on incognito anyways, but I closed it and reopened it and tried it again on the link I used above, still no results
edit: It's working now... guess it was just more cache.
If any template coding gurus here have the time or desire to assist at Template talk:GTA table#Is this table getting too wide?, it would be greatly appreciated. We have run into a minor formatting concern, and for the most part, have agreed on a new format that will solve the problem. Unfortunately, we need some technical assistance to move forward. Thank you. -- GoneIn60 (talk) 20:41, 26 April 2024 (UTC)
It is to avoid urls from overflowing the viewport and infoboxes etc. This only applies to direct links, with no display contents and without wikicode [] surrounding them. Because we know so little about the length and context of a link (think 140 continuous characters for a link component sometimes), we have to be a bit more forceful with their linebreaking. Normal 'external' wikicode links break on word boundaries, because in those situations people generally have been more deliberate with placement and for display names, words are generally a lot smaller. —TheDJ (talk • contribs) 17:23, 27 April 2024 (UTC)
@RoySmith well from what you highlighted, the css style seems to be applied to the url link text and not the url itself otherwise the url won't work nor even be accessible because no gap or space are allowed in the url . The word-break:break-all; in css means that it will adjust every items in their html container once they reach the maximum width limits, thus preventing them from overflowing, as you highlighted here,<a>https://peterbondspac e.com/ </a> this was treated as the link text and not the url itself Thisasia(Talk)17:34, 27 April 2024 (UTC)
Using Firefox 125.0.2 and Windows 10, when logged out, I see the offending wrapping in the example of the original post. This also occurs if I switch skin to legacy Vector (by inserting &useskin=vector into the URL), Cologne Blue or Modern. But if I switch skin to MonoBook (in a similar manner), I then get no wrapping. It's therefore a skin thing. --Redrose64 🌹 (talk) 18:05, 27 April 2024 (UTC)
Is anyone else also seeing this bug on Flag of Russia? There are two major headers saying "User:CheezDeez ON TOP" that appear inside the infobox (this user is also blocked for sockpuppetry, but I don't think they have ever edited this article). I can't find anything in the wikitext that might produce this text, and the bug also doesn't show up in preview mode either. Liu1126 (talk) 13:41, 28 April 2024 (UTC)
I was reverting all these edits when it was happening (in this specific case, Special:Diff/1221172528), but it seems this article didn't refresh when I undid the changes. I gave the page a null edit, and the content is fixed again now. Aidan9382(talk)13:44, 28 April 2024 (UTC)
Hi! I noticed the same issue in the Haaretz article (Template:Literal translation) and the Tiger article (Template:MirrorH). I see you've already reverted the vandalistic changes to both templates, but the articles I linked don't appear to have been refreshed yet. Could you refresh them as well, or otherwise let me know how I could do that? Thanks! Wavevari (talk) 15:17, 28 April 2024 (UTC)
@Tonymetz: "request desktop site" is a feature in some mobile browsers. I don't know whether it's supposed to work with Wikipedia. The normal way to choose the desktop version of Wikipedia is the "Desktop" link at the bottom of pages in the mobile version. The link is made by MediaWiki and not the browser. PrimeHunter (talk) 17:47, 28 April 2024 (UTC)
Thanks I’ve subscribed. 11 years let’s keep our fingers crossed. Thanks for the tip on the desktop link I see it now. It had been cut off by my iPad keyboard bar. Tonymetz💬22:06, 28 April 2024 (UTC)
The four pages have one things in common: inside the {{User:ClueBot III/ArchiveThis}}, the first part of the |archiveprefix= parameter does not begin with the name of the page that is being archived. For three of the four, this is because a page move occurred at some point in the recent past and the archiving config wasn't amended to suit; in the case of User talk:Spiderjiu it's because the archiving was set up badly. The other main archiving bot, lowercase sigmabot III (talk·contribs), guards agains this by refusing to archive a page when the first part of the |archive= parameter does not begin with the name of the page that is being archived. --Redrose64 🌹 (talk) 17:27, 28 April 2024 (UTC)
@K6ka has since updated the docs to advise editors to change the template when moving the page (I'm not overly optimistic editors will heed that warning). Local Variable (talk) 23:26, 28 April 2024 (UTC)
Lua date formats
Hello everyone, does anyone know how to get Lua accepted two date format at the same time? That is. YYYY-MM-DAY and DD-MM-YYYY. I have tried my all to get them both work together at the same time in lua module Wikipedia but that's impossible for me. Only one type of date format can only work at the same time. Thisasia(Talk)06:40, 27 April 2024 (UTC)
Accepting a date like 01-02-2020 would not be a good idea at Wikipedia where some would think January 2, 2020 while others would see 1 February 2020. Where is this needed? Johnuniq (talk) 08:09, 27 April 2024 (UTC)
alright thanks for telling me that, I'm build a new dynamic and multipurpose infobox in lua by integrating all functionality in one Template, but the current programed date format is in something like 01-02-2020 so which date format is well acceptable for Wikipedia? Thisasia(Talk)09:43, 27 April 2024 (UTC)
Module:Date accepts a variety of dates formats, for example, 2024-04-27 or 27 April 2024 or April 27, 2024 (all allowed at Wikipedia) and only allows valid dates. Johnuniq (talk) 03:47, 28 April 2024 (UTC)
seconding YYYY-MM-DD which is not only the most logical but also sorts properly and is an international standard jp×g🗯️19:00, 28 April 2024 (UTC)
Why are we going through this again when MOS:DATE was settled years ago? We don't use all-numeric dates except in certain special situations, and even then, only CCYY-MM-DD is permitted. If you want that changing, WT:DATE is the place to do it but (i) make sure that you inform WP:VPP; (ii) be prepared for massive kickback. --Redrose64 🌹 (talk) 22:07, 28 April 2024 (UTC)
After updating the template styling of my user page, I managed to get most things as I wanted except for a gap between the "Start tab" template and the bordered box that comprises the rest of the page. Is there a way I could put the bordered box into the template, or make an adjustment to how they are configured, to remove the gap between them? -CoolieCoolster (talk) 23:26, 28 April 2024 (UTC)
@CoolieCoolster i have seen your page, are you referring to the gap between the "start tab" and your userboxes section, if so then I can help you with that or can you be more specific and explain properly? You may discuss with me on my talk page if I didn't respond here on time. Thisasia(Talk)05:23, 29 April 2024 (UTC)
I meant the thin white line that appears between the template that creates the tabs and the section that encloses the rest of the page's content. As I reused the format on the other three tabs, each one has the same white gap. -CoolieCoolster (talk) 05:26, 29 April 2024 (UTC)
Pretty not much clear, but see what I found so far.
there is a huge gap between the start tab and your userboxes section may be making it look unsatisfying since you applied the same background color to both of them
there is also a thin line that appears at the start tab when scrolled from either left or right. But except that I couldn't figure anything else.
I'm guessing the first issue you described is what it is; both halves using the same color is intentional, as ideally I'd like to find a way for the content of the page to be within the box connected directly to the tabs, but doing so while also having a border around the page seems to be a conundrum. -CoolieCoolster (talk) 06:36, 29 April 2024 (UTC)
Alright I do have solution for that. If you do permit me i will edit your page directly, and after that if that's not what you described we can find further. But first of all, take a look at the top of this page, it has similar start tab as yours and the start tab isn't separated from the rest of the page unlike yours that splits, check and tell if this is actually your point. Thisasia(Talk)07:02, 29 April 2024 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
Example of a talk page with the new design, in French.
The appearance of talk pages changed for the following wikis: Azerbaijani Wikipedia, Bengali Wikipedia, German Wikipedia, Persian Wikipedia, Hebrew Wikipedia, Hindi Wikipedia, Indonesian Wikipedia, Korean Wikipedia, Dutch Wikipedia, Portuguese Wikipedia, Romanian Wikipedia, Thai Wikipedia, Turkish Wikipedia, Ukrainian Wikipedia, Vietnamese Wikipedia. These wikis participated to a test, where 50% of users got the new design, for one year. As this test gave positive results, the new design is deployed on these wikis as the default design. It is possible to opt-out these changes in user preferences ("Show discussion activity"). The deployment will happen at all wikis in the coming weeks. [9]
Kartographer maps can use an alternative visual style without labels, by using mapstyle="osm". This wasn't working in previews, creating the wrong impression that it wasn't supported. This has now been fixed. [18]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 30 April. It will be on non-Wikipedia wikis and some Wikipedias from 1 May. It will be on all wikis from 2 May (calendar). [19][20]
How does one correct a citation when the source is listed dozens of times?
In the entry on lyricist Robert Hunter, I noticed that at some point an editor must've accidentally cited another source that's already cited 22 times in the article. There are only three 2015 sources used, so perhaps someone inadvertantly used the wrong one. At any rate I tried to fix it using the visual editor and, while it corrected the citation, it left it with the wrong footnote number still. So I undid my failed correction and left a note on the article's Talk page. Correcting simple sources that are only cited once is pretty straightforward, but when they're used dozens of times it's another (way too intimidating) matter. Or am I simply not able to find "the easy way"? Peterh6658 (talk) 23:22, 29 April 2024 (UTC)
Thanks, Duncan! I hope I wasn't out of line with that crack about the process being way too intimidating. (Another Maxwell Smart catchphrase.) I'm still a relatively new Wikipedian, and can never remember the best shortcut to help me remember that ref name technique. Peterh6658 (talk) 23:51, 29 April 2024 (UTC)
You can make this change in the visual editor too, but it's not very intuitive. There are two ways you could do it:
After you select the reference marker (i.e. the "[5]"), instead of clicking "Edit" in the menu, you can just delete it (press Delete/Backspace on your keyboard), and then copy-paste the right reference marker in its place (i.e. the "[2]").
Or, after you click "Edit" in the reference menu, you can click on "Change reference type" in the bottom-left corner, which will open the "Replace citation" menu – in that menu, go to "Re-use" and choose the correct citation from the list.
I am looking to source free images for articles on plant genera that are lacking them. Therefore I was wondering whether it was possible to configure PETScan only to return pages which do not contain images. The closest thing I thought of was to show images from the metadata in the output and then scroll through them, looking for those for which no image is presented. However, that is getting progressively tedious. Does anyone know a more direct way to do this? Felix QW (talk) 09:33, 30 April 2024 (UTC)
Not an answer to your question, but the search facility may help you. A search for plant hastemplate:"automatic taxobox" insource:/image *= * returns 16,436 results with an image parameter in the taxobox. Note, it's not restricted for genera and some may have empty taxobox image parameters. Then a search for plant hastemplate:"automatic taxobox" -insource:/image *= * finds 1,298 results without the image parameter, most of which appear to be genera. The search can probably be refined. — Jts1882 | talk10:57, 30 April 2024 (UTC)
Thank you for the interesting suggestion! It does seem to me that empty taxobox image parameters are quite common, but then again, it is not like I am not going to ever finish this task. PETScan has the additional advantage of alphabetical sorting, so it is consistent across searches, and I can exclude pages linked from my userspace list of taxa for which I have already searched for images without any success. Therefore, a PETScan-based method seems preferable to me at the moment. Felix QW (talk) 12:33, 30 April 2024 (UTC)
Brilliant, thank you! I don't know how I missed that. By the way, thank you very much for all your work on plant articles! Felix QW (talk) 07:18, 1 May 2024 (UTC)
Curious ephemeral js error
I turned on the "display js errors" gadget a few days ago. Today I got an error something like "unhandled exception missing ) for argument list line 23". Since the gadget pop-up fades away quickly, I refreshed the page a few times to try and select and copy the error. I then took a look at the source: line 23 is (for me) }];});});</script>. After looking at this I tried again to copy the error, but it had gone.
I compared the script which runs from line 6 to 23 before an after the error with an on-line diff, no apparent difference.
I got it too, so I filed phab:T363701. It happened whenever the CentralNotice banner for the U4C voting was displayed. But it stopped at some point even when I got the banner. I wonder if something was backported and fixed it. Nardog (talk) 22:44, 29 April 2024 (UTC)
Thanks! Do you know what counts as a view? According to this tool, this video had very few views, in spite of being used in an article with 30k daily viewers. Does it really mean that very few users choose to watch (admittedly hard-to-watch) video? Alaexis¿question?14:20, 30 April 2024 (UTC)
According to GlobalUsage that file isn't used anywhere; I suspect you wanted to look up the stats for this file instead (GlobalUsage), which has more views: [21]. The autocomplete in the mediaviews tool is a bit fiddly sometimes, and apparently doesn't find files that have been renamed, like this one, if you use the redirect name when searching. I didn't know that before. Matma Rextalk14:51, 30 April 2024 (UTC)
This reminds me that if we ever make videos 'preload' by default, these stats will break, because every pageview will preload the first bytes of a file... —TheDJ (talk • contribs) 17:12, 30 April 2024 (UTC)
Almost any character can appear in a URL, especially after ?title=. It may be better to accept anything except whatever delimits a URL in the relevant context, e.g. for parsing {{cite web|url=http...}} use [^|}]* or for space-delimited [http... Description of page] try \\S*. The former may pick up unwanted spaces unless we complicate it as something like .*?(?=\\s*[|}]) (grab any characters, but as few as possible, as long as optional spaces then | or } follows). Certes (talk) 00:14, 30 April 2024 (UTC)
OK, building on that idea, this solution is imperfect, but it will work in most cases. I'm not using it for critical purposes (log files, counting) so YMMV:
Out of interest I fed this into ChatGPT (I know...!). Here's its response, for what it's worth (minor tweaks for formatting) -
This regular expression (regex) is designed to match URLs that point to the domain "wikisophia.org" under specific conditions, while excluding URLs that are part of certain contexts or formats. Let's break down this regex to understand what it is specifically targeting:
1. Negative Lookbehind Assertions:
- `(?<!/)` ensures that the match does not directly follow a forward slash `/`. This might be used to avoid matching URLs that are part of a larger URL path or parameter.
- `(?<!\\?url=)` ensures that the match does not directly follow the query parameter string `?url=`. This helps in avoiding matching URLs that are query parameters.
2. URL Scheme:
- `https?://` matches both "http://" and "https://". The `s?` means that the character "s" is optional.
3. Domain Name:
- The part `(([a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9\\-]*[a-zA-Z0-9])[.])*` is somewhat complex but essentially matches sequences of alphanumeric characters separated by hyphens (not starting or ending with a hyphen) and followed by periods. This part matches subdomains as well, as it repeats zero or more times due to the `*` operator after the group.
- `([A-Za-z0-9]|[A-Za-z0-9][A-Za-z0-9\\-]*[A-Za-z0-9])` ensures the domain name or subdomain consists of alphanumeric characters, possibly with hyphens (but not starting or ending with them), much like the earlier subdomain matcher.
4. Specific Domain:
- `wikisophia[.]org` is very straightforward and matches the domain "wikisophia.org". The square brackets around the period `.` are technically not needed here, as they are used to escape special regex characters (though `.` would typically be escaped directly with a backslash).
5. Trailing URL Path and Parameters Exclusions:
- `[^\\s\\]|}{<]*` This part of the pattern is intended to consume any characters that are not whitespace, slashes, or certain special characters like `|`, `}`, `{`, `<`. However, there are syntax issues and potential misinterpretations in this segment:
- `[^\\s\\]` seems intended to exclude backslashes and whitespace (though `\\s` would be invalid in most regex flavors and might be an error).
- The character set `[^\\s\\]|}{<]` is followed by a `*` quantifier, meaning it matches zero or more of any character that is not a space, slash, or one of the special characters listed. This part could be attempting to match the rest of the URL while excluding specific trailing characters but is likely erroneous as presented.
Conclusion
The regex seems intended to match "wikisophia.org" URLs that do not follow a slash or the specific query parameter `?url=`. However, there are syntax issues that might cause it to behave unpredictably, particularly with the last part concerning non-whitespace and non-special characters, which appears to be syntactically incorrect. This regex should be reviewed for errors and tested thoroughly in its intended context to ensure it functions as expected.
Today, when I clicked the button "upload a non free file", it just reload the page, please fix this bug or glitch as soon as possible. SleepingJoe (talk) 04:30, 1 May 2024 (UTC)
Please if I may ask, when uploaded Files with Special:Upload which directly is it going to stay, is it wikimedia common or Wikipedia itself? because I have always thought that files can only be uploaded through wikimedia. Thisasia(Talk)10:53, 1 May 2024 (UTC)
In this discussion at my Talk page, I made an assertion (diff) that traces of a revdel can always be seen in the revision history, with revisions in strikeout type, and the like. But now, I wonder if my response there was accurate. Is there such a thing as, for lack of a better term, what I will call a stealth revdel—that is, a revdel which wipes some revisions from the history, without leaving any evidence of the removal visible to non-admins? If that exists, what is it called, and is there a page that covers this? Mathglot (talk) 11:13, 1 May 2024 (UTC)
Selective (un)deletion, i.e. deleting the page and undeleting all except certain revisions. If an admin really wanted to hide it, they'd have to then delete or revdel the log entries of the deletion and undeletion too. Anomie⚔11:35, 1 May 2024 (UTC)
The thing being discussed in that discussion was some template vandalism, which was of course on a different page transcluded onto that one. Some varieties of oversight can remove edits without much trace, though not something like this purported situation. -- zzuuzz(talk)11:42, 1 May 2024 (UTC)
No, I do mean suppression oversight (and 'hiding'). It's not a thing to focus on here, but OS can make some disappearances hard to track. -- zzuuzz(talk)11:48, 1 May 2024 (UTC)
so maybe I meant oversighted? It's idle curiosity in this instance but I am pretty sure I have noticed stuff disappear before, especially identifying personal information. In fact, I distinctly remember telling an admin that somebody's address was in an article edit history, and this got removed, at lrast as far as a plain vanilla editor could see.
Which is not what this was, granted; it would have been more of an unfortunate paste error or perhaps vandalism. Elinruby (talk) 12:22, 1 May 2024 (UTC)
I seem to recall days when there was no revision deletion feature we have now, and the way to hide select revisions was to move the page to another name, delete it, restore okay revisions, and move it back. I assume it's possible still. Nardog (talk) 13:25, 1 May 2024 (UTC)
There are certainly beansy ways to cause obfuscation - that are best left unlisted here. Just replace "can always..." with "can generally...." and most use cases will be covered. — xaosfluxTalk14:11, 1 May 2024 (UTC)
Thanks, all; this has been informative. And I got the read-between-the-lines stuff, so needn't go further. Cheers! Mathglot (talk) 00:35, 3 May 2024 (UTC)
I know it's possible to delete summaries of log entries... including for the revdel log... and for the revdel log summary deletion log. I think you can set up an infinite regress this way. jp×g🗯️07:53, 3 May 2024 (UTC)
Caption on Image wikiMarkup?
Please may i ask if caption is possible inside the image wikiMarkup? Because since it does have the argument for sizing like for example: [[File:Example.jpg|100px]] then i suppose I'd probably have an argument for caption and even perhaps, more arguments beyond? Thisasia(Talk)12:31, 3 May 2024 (UTC)
A new template to count the number of internal links
Hi, I searched for a template that returns the number of internal links, but couldn't find any so I created this Template:InternalLinkCounter now. Please take a look and improve it if you like. I just wanted the community to be aware of this new template. Thanks! ⇒ AramTalk23:39, 3 May 2024 (UTC)
"The time allocated for running scripts has expired."
Hi. I created Portal:Washington, D.C., which contains some random elements. Most of the time, the page loads properly, just a bit slow. However, sometimes the page will only load some of the elements, and returns "The time allocated for running scripts has expired.". It's weird how the elements that don't load are static, and elements that do load are random. How can I fix this? Thanks. '''[[User:CanonNi]]''' (talk|contribs) 11:09, 4 May 2024 (UTC)
CanonNi, looking at the profiling data the lua time usage was around 8 seconds (out of 10). It says: MediaWiki\Extension\Scribunto\Engines\LuaSandbox\LuaSandboxCallback::match 2960 ms 37.7% MediaWiki\Extension\Scribunto\Engines\LuaSandbox\LuaSandboxCallback::find 1580 ms 20.1% MediaWiki\Extension\Scribunto\Engines\LuaSandbox\LuaSandboxCallback::sub 1220 ms 15.5% So it looks like some expensive match operations are being run. — Qwerfjkltalk11:17, 4 May 2024 (UTC)
Hi, Every time I try and thank editors either on here or on Simple Wiki I'm getting the following error message:
Thank action failed (error code: internal_api_error_JobQueueError). Please try again.
I recently noticed that {{ambox}}, when viewed using the night mode beta feature for advanced mobile web users, has a dark gray left border, not a colored one like when viewed using light mode. This is despite the relevant TemplateStyles specifying the color #36c () as the default left border color, with other colors such as #b32424 (), #f28500 (), and #fc3 () being specified for other ambox types. Does anyone know why this is? Thanks, Andumé (talk) 05:06, 4 May 2024 (UTC)
WMF has added some !importants to their dark style sheet probably (or maybe just deeply nested selectors it doesn't need) and currently doesn't have sufficiently granular settings to turn that off... Ignoring also that no-one has actually looked to see what the color for ambox should be. (Could be the same, could be different, etc.) Izno (talk) 16:22, 4 May 2024 (UTC)
What links here says that a page transcludes itself, which should be an error
What transcludes Template:Requested move notice shows ~4 pages, including a self-transclusion, which should be impossible. When I edit the template to make it transclude itself, I see the error (in preview mode):
Warning: This page calls Template:Requested move notice which causes a template loop (an infinite recursive call).
But it's not populating Category:Pages with template loops. Why is this spurious transclusion report happening, and how do I suppress it? My bot looks up all the pages that transclude this template, and is confused by this spurious transclusion. wbm1058 (talk) 14:30, 3 May 2024 (UTC)
Self transclusions occur when a template uses the MW Lua library getContent on the page it's transcluded on. This happens for example with Module:Citation/CS1 (which hunts for the "use XXX date" templates in the page source, among others). Izno (talk) 16:20, 3 May 2024 (UTC)
It's common and allowed for pages to transclude themselves. A template loop error only occurs if the transcluded version of the page also tries to transclude itself. PrimeHunter (talk) 16:46, 3 May 2024 (UTC)
This is very confusing. These pages compare identical yet one "transcludes itself" while the other does not. I'll assume that neither actually transcludes itself on a first-level basis, but the documentation only "transcludes itself" via the template transcluding the documentation, in other words, "cascading transclusion", or "circular transclusion" where the template transcludes the documentation while the documentation is simultaneously transcluding the template, and there is no way to differentiate regular transclusion from cascading/circular transclusion, so in order to work around this bizarre behavior, I'll need to make a special, kludgey edit to my bot's source code to specifically remove one specific page from the list of transcluded pages it gets from its "what transcludes this page" API call. I guess I've been kind of aware of this strange behavior for a while, but I think this is the first time I've found myself needing to actively code a work-around for it. – wbm1058 (talk) 17:59, 3 May 2024 (UTC)
Many templates can cause a page to transclude itself. In this case it's caused by {{Bot use warning}} because it uses {{pagetype}} which examines the code of the page to identify certain page types. Module:Pagetype uses getContent to do it and as Izno said, this gives a self transclusion. Just work around it if needed as you say. It's not broken and doesn't require "fixing". Template:Requested move notice/doc/sandbox also transcludes itself. Maybe it had not been registered in link tables after your move.[22]PrimeHunter (talk) 21:19, 3 May 2024 (UTC)
I thought every page transcluded itself, to be honest. Every (article) I've looked at closely was 'transcluded' by both itself and its talk page. – 2804:F1...6E:399F (talk) 20:30, 3 May 2024 (UTC)
Sometimes it feels like that to me too. I don't think the system always worked that way. At some point, something changed, I think, and I've yet to figure out exactly when, what, or why. – wbm1058 (talk) 21:16, 3 May 2024 (UTC)
@Wbm1058: You are indeed correct that it hasn't always been this way; it started happening in April 2019. To find this I searched the village pump archives for "Transcludes itself" and found this discussion; my comment there pointed to the former thread. Graham87 (talk) 08:00, 5 May 2024 (UTC)
Riddle me this. Compare and contrast. My initial example, limited to template namespace, followed by a similar example (my bot uses both):
Gah! Well, thanks for solving my puzzle! No, it was a problem. My bot's console report was saying:
339 : Template:Requested move notice => NOT FOUND !!! Failed to remove subject notice
The transclusion made the bot think there was a {{Requested move notice}} sitting on the template – not unheard of that someone would request moving it, as it's been moved multiple times – the bot tried and failed to remove the template it thought was there, and thus reported the problem to me. THIS EDIT solved my problem by making the bot happy again. I added {{Bot use warning}} on 22 November 2023; guess I'd overlooked the bot's messages about this for months? Hey, I didn't realize that adding that template would change this template in a significant way! Who would have thunk that template needed to use {{pagetype}}, I never would have guessed that! {{Wikipedia:Article alerts/Bot use warning}} manages to give its warning without finding the {{pagetype}} first! Why would anyone ever need to find the {{pagetype}} to warn ppl that bots use their template? I'm just not gonna bother keeping the warning there, as it's just too much trouble and Murphy's Law says another WP:IAR editor is born every minute so resistance is futile. wbm1058 (talk) 22:22, 3 May 2024 (UTC)
Can't delete page
Trying to delete Dheeraj Kher. Get following error:
To avoid creating high replication lag, this transaction was aborted because the write duration (6.239844083786) exceeded the 3 second limit. If you are changing many items at once, try doing multiple smaller operations instead.
[063023b1-4d1e-462e-9306-ffba965548d1] 2024-05-05 17:23:44: Fatal exception of type "Wikimedia\Rdbms\DBTransactionSizeError"
Tried more than once. Also tried deleting another page, but didn't wait for error. Deletiions are usually almost instantaneous.--Bbb23 (talk) 17:26, 5 May 2024 (UTC)
Possible but unlikely - I tried to delete it and it worked for me, and ran into several other instance of this error in a recent deletion spree but trying again always worked. * Pppery *it has begun...18:08, 5 May 2024 (UTC)
I know (obviously, because I did say "be consistent with" instead of "conform to"). But how does it help to have our own interface contradicting our own MoS? — Mikhail Ryazanov (talk) 08:52, 4 May 2024 (UTC)
Yes they are contained in the interface see list. I do not think we should localize these for only those with their interface language set to English. You could make a case for this in a feature request which if accepted would "improve" this for all projects. — xaosfluxTalk10:10, 4 May 2024 (UTC)
Some other languages already have them "localized" (such as "bit/s" in Finnish, "b/s" in French, "б/с" in Macedonian). Do you propose changing the defaults such that all the languages where these messages haven't been "translated" will automatically switch to "bit/s"? Although this would be nice ("bit/s" is at least defined by the international standardIEC 60027, whereas "bps" is not standard at all), doing so globally might be perhaps more difficult than redefining these messages just for English at translatewiki. — Mikhail Ryazanov (talk) 11:17, 4 May 2024 (UTC)
My suggestion is that if the root of these is not the best possible international version of this, it be fixed at the root message level, not with a project-specific local override. I don't think "translating" them from the root English to English at translatewiki is the fix there - but you could give it a try. Fixing at the root reduces local technical debt and would share the improvement with all mediawiki projects. — xaosfluxTalk12:17, 4 May 2024 (UTC)
Follow on, inconsistent translatewiki work can make these sloppy - for example I can't imagine why in Portuguese the kilobits should be "kb/s" while the megabits are "Mbps". Likely because someone wanted to fix one instance and isn't even aware of the other ones - but if it was all in an agreeable root style that may have been avoidable. — xaosfluxTalk12:28, 4 May 2024 (UTC)
OK, would you then suggest requesting to change them directly in MediaWiki, as it seems from WP:BUG, or follow what System messages says: "General changes that could benefit other wikis can be submitted to translatewiki.net"? — Mikhail Ryazanov (talk) 23:16, 4 May 2024 (UTC)
I've never had a lot of luck getting English-->English translations in translatewiki coming over, that's why I suggested the root message itself. — xaosfluxTalk00:44, 6 May 2024 (UTC)
Whether other projects follow English Wikipedia's MoS or not is their decision entirely. Even if you want it to, any conseus on that here has no effect. I am listing this as a reason to oppose extending your steward rights next year xaosflux, very disrepectful behaviour of you. Snævar (talk) 11:33, 4 May 2024 (UTC)
Sorry we are in disagreement. I certainly don't think that article style guidelines on this project should be forced on other projects. To be clear my suggestion is just that this would best start as an upstream discussion - which could certainly result in a WONTFIX. — xaosfluxTalk12:24, 4 May 2024 (UTC)
Tech News: 2024-19
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
Example of a talk page with the new design, in French.
The appearance of talk pages changed for all wikis, except for Commons, Wikidata and most Wikipedias (a few have already received this design change). You can read the detail of the changes on Diff. It is possible to opt-out these changes in user preferences ("Show discussion activity"). The deployment will happen at remaining wikis in the coming weeks. [23][24]
Interface admins now have greater control over the styling of article components on mobile with the introduction of the SiteAdminHelper. More information on how styles can be disabled can be found at the extension's page. [25]
Wikimedia Enterprise has added article body sections in JSON format and a curated short description field to the existing parsed Infobox. This expansion to the API is also available via Wikimedia Cloud Services. [26]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 7 May. It will be on non-Wikipedia wikis and some Wikipedias from 8 May. It will be on all wikis from 9 May (calendar). [27][28]
When you look at the Special:Log page, the first view is labelled "All public logs", but it only shows some logs. This label will now say "Main public logs". [29]
How is one supposed to add stub tags using the VisualEditor? If they're added as templates at the bottom of the page, they'll incorrectly be placed above the categories. (Previously asked at the Help Desk but didn't get a response.) --Paul_012 (talk) 11:51, 6 May 2024 (UTC)
Ah you mean, ‘not according to the Manual of Style of English Wikipedia”, which speciefies an order that is not logical for the data, nor for anyone who reads the page. —TheDJ (talk • contribs) 15:41, 6 May 2024 (UTC)
It's logical for the reader to consistently find content categories, which encourage them to browse related topics, before stub categories, which are there to encourage expansion of similar stubs. --Paul_012 (talk) 17:53, 6 May 2024 (UTC)
No it’s not. That’s, at most, potentially handy for people who do stub sorting (hardly anyone). For most people it makes much more sense to have elements in the order in which they appear in the rendered article. ;) —TheDJ (talk • contribs) 20:15, 6 May 2024 (UTC)
Yes: we've been able to use the linktarget alternative for a while now. Today's news, according to the Phabricator ticket, is that they are now ready to drop the columns. Certes (talk) 21:33, 30 April 2024 (UTC)
Category:Proposed deletion as of 27 April 2024 is showing up on the latest run of Special:WantedCategories as a non-empty redlink with two pages in it, but is completely empty when I actually look at it — and even a temporary undeletion and redeletion failed to resolve either its emptiness as itself, or its non-emptiness on the report. I have seen cases like this before, which typically related to deleted content that was somehow leaving behind a phantom categorylinks row — IIRC, it was usually resolved by undeleting and then redeleting the phantom articles, but I don't know how to figure out what was in this category prior to its deletion on Saturday. Could somebody look into this? Thanks. Bearcat (talk) 15:00, 7 May 2024 (UTC)
Well, yes, if you want all of them; you could also look at the history of User:DumbBOT/ProdSummary. But restoring and deleting them all, instead of just the two that broke the count, is reasonably likely to break another. —Cryptic19:52, 7 May 2024 (UTC)
The revisions are in the revision SQL table, but not in the recentchanges SQL table. Quarry. Smells like a core bug related to saving an edit (as opposed to a core bug related to the recentchanges page). –Novem Linguae (talk) 17:36, 4 May 2024 (UTC)
Although I have no evidence to suggest it, this could possibly be a less serious bug in the recentchanges view which hides deleted revisions from Quarry, which I often lazily think of as an actual table. Certes (talk) 20:27, 4 May 2024 (UTC)
That one Works for me in the normal WL UI: mb Microphysidae 12:17:28 +14 AnomieBOT talk contribs block (Dating maintenance tags: {{Cn}}) rollback thank. — xaosfluxTalk10:45, 7 May 2024 (UTC)
My edits are visible, bot edits are not. If I allow edits by bots it appears, and indeed the last edit was from a bot. But if not, shouldn't the previous (human) edit appear? (Chronologically, it should) --cyclopiaspeak!12:17, 7 May 2024 (UTC)
Ok, it seems that, if bots are filtered, and the last edit is a bot, then no edits are shown from that page. Is that expected behaviour? I would expect that filtering bots would give me the last non-bot edit, not ignoring the page until there is a non-bot edit.--cyclopiaspeak!12:19, 7 May 2024 (UTC)
That is normal behavior; if you set any filter that hides the current revision the entry will not display on the filtered watchlist. This isn't specific to bots, if you hide "my edits" for example and you are the most recent, then it won't be on the watchlist display even if others also edited. — xaosfluxTalk12:26, 7 May 2024 (UTC)
Would it be possible to add something below the "Changes shown" heading, along the lines of 'If you enable any of these options, you should also enable "Expand watchlist to show all changes, not just the most recent" otherwise you may miss out on some changes'. --Redrose64 🌹 (talk) 20:22, 7 May 2024 (UTC)
Openzim.org is a third party website. Did this used to be a wiki tool with a web interface? Do you recall who the maintainer is? Is there a wikipedia namespsce or user namespsce wiki page with general info about it? The answer to this one will probably be "figure out who made the tool then ask them to fix it". –Novem Linguae (talk) 19:03, 6 May 2024 (UTC)
When I moved the {{coord}} template into the infobox, the infobox appeared. I don't know if that was the best fix or why the infobox completely fails to appear when a map is specified without coordinates in {{Infobox military conflict}}. I have reported the problem on the template's talk page. – Jonesey95 (talk) 01:03, 9 May 2024 (UTC)
The original svg (the rendered pngs are fine) file is misrendered on Android Chrome.
Namely, the text "SOUTHERN OCEAN" on the top is not shown at all. In the (identically worded) texts at right and lower left, the word "SOUTHERN" misses the "E" (there is a space there instead). The texts "ANTARCTIC PENINSULA" and "KONG HÅKON VII HAV" also miss several letters. The text "Antarctic Circle 66°34'S" is completely missing.
Could you reproduce this? I remember seeing editors discuss this somewhere (probably here, on VPT, but I'm not sure) several months ago, however they referred to the pre-rendered pngs, not to the original svgs. Where could be the problem? Janhrach (talk) 17:15, 9 May 2024 (UTC)
In the references for Special:Permalink/1223039613, one of the NY Times refs (#34) has a nice little padlock showing it is behind a paywall, but another one (#41) has a humungous padlock. Anybody know what's going on there? RoySmith(talk)14:56, 9 May 2024 (UTC)
Note sure WHERE it is, but the large one has padding: 8px 18px 8px 0; around it, without which it would be the same size. — xaosfluxTalk15:38, 9 May 2024 (UTC)
CS1 is synced only irregularly these days, but no quicker than once every 3 months. Someone may request a sync at the related section if they wish. Izno (talk) 16:43, 9 May 2024 (UTC)
I make no comment on such things. As I said, you can ask to have the modules synced now rather than later, or you can ask for this change to be made now rather than later. Izno (talk) 16:52, 9 May 2024 (UTC)
What's all this sync, sync timing, stuff? If someone has a fix they can just open an edit request and ask for the page to be updated, no? — xaosfluxTalk17:23, 9 May 2024 (UTC)
@Xaosflux, given the quantity of pages it is used on, the visibility of those transclusions (multiple hundreds per page), and the complexity of the module, sandboxes are (/used to be) synced only about once a quarter. This is to give opportunity for objection and bug identification for new features and to keep the edit history for bug fixes in the live modules to a minimum. Izno (talk) 17:44, 9 May 2024 (UTC)
Thanks everybody for the swift analysis. I've opened T364561 to track this; let's see what the folks who maintain Common.css say before we start patching it locally. RoySmith(talk)16:35, 9 May 2024 (UTC)
@RoySmith Your ticket was closed because both Common.css and Module:Citation/CS1/styles.css are local on-wiki content. In any case Common.css doesn't need to be patched, the fix would be entirely contained within the CS1 css file. --Ahecht (TALK PAGE)16:43, 9 May 2024 (UTC)
A recent OpenStreetMap changeset added Serbian place names (but did not change the primary names) for some Manhattan locations.
Some places such as Flatiron District still displaying the correct English name on Wikipedia have name:en defined on OpenStreetMap while other place names now displayed in Serbian such as Hell's Kitchen are missing name:en and only have name defined.
The above names are all displayed in English when viewed directly on openstreetmap.org.
I'm seeing the same issue on simple.wikipedia.org OpenStreetMap maps.
I don't know if this is an OpenStreetMap issue, a Wikipedia issue, both, or neither. I'm happy to file a bug report on Phabricator if needed, but I figured I would start here. Daniel Quinlan (talk) 21:33, 7 May 2024 (UTC)
Minor bug upon clicking "Publish changes" button: When I made this edit, I was presented with a captcha because I was allegedly adding an external URL. But the contents of the edit show I did no such thing, nor did I even alter an existing URL. 2605:A601:AAF7:3700:A1D7:26C1:E273:28CF (talk) 21:22, 8 May 2024 (UTC)
This isn't related to the captcha, but your edit set off a filter for 'Possible dead link replacement' which is rather odd: Special:AbuseLog/37687384.
Search for the added_links and removed_links variables.
edit: I mean, the filter isn't what caused the captcha, but the added_links and removed_links variables are what Wikipedia thinks was added and removed, pretty much. – 2804:F1...09:F464 (talk) 22:23, *edited: 22:35, 8 May 2024 (UTC)
A captcha requirement is based on external links in the expanded wikitext after transcluding all templates and so on. Something may have changed in a template, a module, a Wikidata item, or possibly the parser itself. Maybe something changed in the processing of news: links. The article includes [news://comp.os.vms/ comp.os.vms Usenet group] with two slashes in news://. It produces [removed to avoid captcha problems]. Special:AbuseLog/37687384 says:
All external links added in the edit (added_links)
[
0 => 'news:comp.os.vms',
1 => 'news:%2F'
]
%2F is percent-encoding of a slash. The abuse log also says:
All external links removed in the edit (removed_links)
[
0 => 'news:',
1 => 'news:',
(many similar lines omitted)
63 => 'news:'
]
Oh, that sounds like a bug. I only suspected a software change had happened shortly before the article edit. This is worse. PrimeHunter (talk) 11:39, 9 May 2024 (UTC)
Maybe I read too quickly. I saw the word "invalid" and skimmed the rest. If this is happening with valid news: links that's probably a different issue. Suffusion of Yellow (talk) 16:05, 10 May 2024 (UTC)
Maybe, but is there a reason for uploading a large video to English Wikipedia? I don't think that a huge non-free video could ever meet the WP:NFC requirements and a free file would fit on Commons just as well. Or is it a free-only-in-the-US file? Jo-Jo Eumerus (talk) 08:33, 5 May 2024 (UTC)
Only public domain in the U.S. I've been working on silent-era films (including non-U.S. works), but a 100 MB limit on a feature-length video means extremely noticeable artifacts. hinnk (talk) 08:51, 5 May 2024 (UTC)
It's T364288. There's more detail attached to the ticket, but I'd tried bigChunkedUpload.js for this file and got an error. hinnk (talk) 23:27, 7 May 2024 (UTC)
The MediaWiki API supports chunked uploads. Whether our File Upload Wizard script makes use of that ability I don't know. Anomie⚔11:54, 5 May 2024 (UTC)
An update: chunked uploading does work here, but only for files up to a certain size. This issue is happening because async uploads are only enabled on Commons (and some test wikis). That means that a tool like commons:User:Rillke/bigChunkedUpload.js will fail if the job takes too long. Having tried a few uploads of various sizes, files under ~1GB are fine, files over a few GB always fail, and files ~1GB or more can succeed if async is disabled. There's a Phabricator ticket for enabling async uploads outside of Commons. hinnk (talk) 19:01, 10 May 2024 (UTC)
Rollback issue
I have the preference enabled to have a confirmation prompt when using rollback. The problem is that the confirmation prompt doesn't actually do anything. I'll say "no" to if you wish to use rollback and the rollback will still happen. This has happened three times. Clovermoss🍀(talk)19:58, 10 May 2024 (UTC)
Okay so after some trial and error I realized that this probably has to do with my gadget settings having both open a user's contributions after a rollback and confirmation prompts checked, as one notes that it won't work if the other is enabled. From a useability perspective it might be useful to have some sort of popup that doesn't actually let you have both boxes enabled but maybe that's just because I apparently can't follow directions. Clovermoss🍀(talk)20:15, 10 May 2024 (UTC)
So this is not quite the same issue as the last time, but we seem to be having THURSDAY issues with the < symbol in the preloaded edit summary when editing a section. When editing (for example) User talk:Primefac § Test section <1> the edit summary should give /* Test section <1> */ but instead gives /* Test section */. Oddly enough, it shows properly for User talk:Primefac § Test section >2<. This is obviously just a made-up example to prove my point, but it's messing with the anchors at WP:ARC (specifically the "Arbitrator opinion" section). What gives? Primefac (talk) 15:59, 10 May 2024 (UTC)
@Primefac does appear to be a bug, can you point to an edit that was made with that sort of automatic edit summary in the past, where this did work as you expect? — xaosfluxTalk16:52, 10 May 2024 (UTC)
I'm not seeing one in your last 2000 edits - just trying to determine if this is something that broke recently (i.e. a THURSDAY thing) or if it is just something that has never worked or been broken for a long time. — xaosfluxTalk17:03, 10 May 2024 (UTC)
I'm not sure that it was working on 4 April based on that diff -- from Xaosflux's comment on the task, the summaries appear to work when using DiscussionTools, and that linked diff is tagged as using the ReplyTool. The diff immediately before that one (Special:Diff/1217280353) wasn't made using DiscussionTools, and doesn't include the LT/GT symbols. All the best, —a smart kitten[meow]17:36, 10 May 2024 (UTC)
The 6 March diff was using the temporary alternatives to </> («/»), before the previous bug had been fixed. I'm currently Ctrl+F-ing through the history of A/R/C, and so far I'm back to September 2023 with (seemingly) the same issue still present. All the best, —a smart kitten[meow]17:46, 10 May 2024 (UTC)
"never worked" is a fine answer, and it is still a bug - just not a THURSDAY bug (i.e. we're not looking for a change that someone just recently made that caused it to break). — xaosfluxTalk19:14, 10 May 2024 (UTC)
The current behaviour seems entirely correct to me - If the page "foo" contains either a link or a template wrapped in includeonly tags then the page "foo" does not actually contain that link or template. Only pages that transclude "foo" would have the link or template.
I don't see how this could be changed in the manner you suggest or why such a change would be beneficial - you would need the parser to render wikitext that it has explicitly been told not to render (and hope that doing so produces something sensible rather than a broken mess), then add entries to the linktables for links that don't actually exist. This seems likely to break all manner of things.
Your more complex "correct" regex does not capture every possible way of invoking the template, and doesn't actually capture any uses the simple case missed. You can call templates using all kinds of weird wikitext constructions, e.g. {{Dele{{void}}ted template}}, {{Deleted {{lc:{{NAMESPACE:Template:Cite}}}}}}. 86.23.109.101 (talk) 21:25, 10 May 2024 (UTC)
I believe I made my case for why my proposal is beneficial already. But, I'll elaborate further. When WLH is used in a deletion venue, what people want to know is where the nominated page is used. That includes links/transclusions between includeonly tags because, if the page is deleted, those will be affected too. I bet that 9 out of 10 frequent XfD participants will find the current behavior both surprising and undesirable.
Also, I never argued that my regex was "correct". That's why I called it "an attempt". In any case, you're making my point for me. Since regex won't work in the general case, we need something more powerful. Nickps (talk) 21:59, 10 May 2024 (UTC)
How is "the list of links only shows links that actually exist" and "the list of transclusions only shows places where a page is actually transcluded" surprising and undesirable behaviour? The whole point of includeonly tags is to stop stuff from being parsed when viewed directly.
I think you've missed that this behaviour isn't actually specific to includeonly tags, there are all manner of template coding schemes that can cause a template to call another template only in specific scenarios, the output can depend the parameters you pass it, where it is used, whether it is substituted, what the time is...
The point I was trying (and failing) to make RE regex is that the simplest, naïve approach to finding uses of the template found 6 uses. Your more comprehensive regex found 6 uses. While you can invoke templates using namespaces with arbitrary capitalisation and you can add whitespace and you could make the template name out of wikitext constructions these are rare to see in actual usage and the simple approach is good enough. 86.23.109.101 (talk) 23:47, 10 May 2024 (UTC)
The whole point of includeonly tags is to stop stuff from being parsed when viewed directly and the whole point of using WLH in XfD is to know all the pages that use a specific page in any way. Excluding everything between includeonly tags makes this intuition fail. You said, the simple approach is good enough. But if you read the TfD I linked above you'll see that the "simple approach" and the one tried first was not to use a regex. It was to trust WLH. In that case, no, the simple approach is not good enough. It utterly fails at what it's supposed to do. Nickps (talk) 00:03, 11 May 2024 (UTC)
I don't think we can. The template itself (which effectively excludes the includeonly section) doesn't transclude the page; the text it evaluates to does. It's a specific case of a more general insoluble problem. For example, a template Foo which transcludes [[{{{1}}}]] will, when called as {{Foo|Bar}}, transclude Bar, but as Template:Foo doesn't mention Bar there's no way to list Bar amongst its incoming links. Certes (talk) 21:01, 10 May 2024 (UTC)
But your Template:Foo does not link to Bar at all. It's just that the result of the transclusion of {{Foo|bar}} is a link to Bar. My motivating example is different. {{Expand}} directly links to {{Deleted template}}. The test to tell those two cases apart is simple. Just remove the includeonly tags. If you do that then {{Expand}} will show up in Special:WhatLinksHere/Template:Deleted template but {{Foo}} will not show up in Special:WhatLinksHere/Bar. And that's expected. What's unexpected is that when we add the includeonly tags back, the first link disappears. Why? It's clearly still there. Nickps (talk) 23:45, 10 May 2024 (UTC)
I don't think there is a way to achieve the OP's general request of finding all templates that use a template. If the template is called from inside a conditional if statement or similar, it won't appear in the "What links here" list either. There are no doubt more cases. – Jonesey95 (talk) 23:58, 10 May 2024 (UTC)
And that's wrong too if you ask me. If the link is spelled out in the source code of a given page then it should count for WLH. If we wrap a link in an if statement as in {{#if:{{{1|}}}|[[Foo]]}}, it shouldn't just disappear from the list. Nickps (talk) 00:32, 11 May 2024 (UTC)
Well, it's called "What links here" and the page doesn't link there. I don't know what else to tell you. Searching is necessary in some cases. – Jonesey95 (talk) 03:52, 11 May 2024 (UTC)
If the issue is semantics, then either rename "What links here" or add a "What links here including in noincude tags". At the end of the day, when you search for usages of a template (or any link), you are more than likely interested in all places it is used at, including those places. Gonnym (talk) 08:11, 11 May 2024 (UTC)
That phab task has nothing to do with what you are proposing here. That phab task is asking that articles should track where the categories on them are coming from and include that information in the category page, basically just adding metadata to the existing entries.
You are asking for template pages to parse content they have been told not to parse, somehow deal with missing parameters, figure out which templates "could" be used (even though this will vary with the parameters you provide) then add entries to the link tables for links and transclusions that don't actually exist. Certes has already tried to explain this to you once, but it is not possible to figure out an exhaustive list of every template that another template could call, as this depends upon how it is used. As a simple example, a template that contains the code {{cite {{{1}}}}} could end up evaluating to any of the citation templates ({{cite web}}{{cite book}}{{cite news}}) or an infinite number of redlinked template (e.g. {{cite foo}}) depending upon what someone puts into parameter 1. 86.23.109.101 (talk) 13:12, 11 May 2024 (UTC)
Nope, I never said anything about missing parameters. Your example is irrelevant. After all, there's nothing that says that parameters can only appear between includeonly tags. All I'm asking is that if for example, a template contains <includeonly>{{cite web}}</includeonly> then that template should appear in Special:Whatlinkshere/Template:cite web. Nickps (talk) 13:34, 11 May 2024 (UTC)
Let me turn that question back to you. How is {{cite {{{1}}}}} handled outside includeonly? This is a common usecase since not all templates wrap their code in includeonly (because they don't have to) and it works just fine. Why do you think that I ever asked we change that? Nickps (talk) 13:36, 11 May 2024 (UTC)
The data for WhatLinksHere is made when a page is rendered and actual links and transclusions are encountered during the rendering. It doesn't matter which code was used to produce the links. Code in <includeonly>...</includeonly> is ignored during rendering so no links are encountered. Your requested change would require something like a second rendering process where <includeonly>...</includeonly> is not ignored. It would be expensive and I'm pretty sure the developers would reject it as not worth the cost, and probably also as more likely to cause confusion than solve problems. Many users would be confused if WhatLinksHere reports links and transclusions which aren't actually in the page itself but may be in other pages which transclude it. PrimeHunter (talk) 22:26, 11 May 2024 (UTC)
It would be expensive. Per WP:PERF I can say that I don't care about that at all. What I suggest would be useful for XfD and that's all I care about. Now I agree with you that my proposal could potentially be confusing but that's easy to fix, just have the links explicitly marked as includeonly, like the phabricator issue suggests. Alternatively, you can make the feature opt-in so those links only appear for users that know they want them. Nickps (talk) 22:37, 11 May 2024 (UTC)
Can anyone think/suggest how to reliably archive a docx document from a website?
Archived the download load link, that's just a redirect but archive.org saves the redirect destination as well. Find where the reditect went and you have your archive link. -- LCU ActivelyDisinterested«@» °∆t°23:15, 11 May 2024 (UTC)
ListeriaBot
I have written a SPARQL query that works perfectly on Wikidata query service but ListeriaBot seems unable to execute it and generate a table from it unless I delete the big optional clause. Please see here the template syntax and a link to the query on WDQS where you can run and verify the results (see the Query here too. I just wanted to test the water here and see if someone have had a similar experience before I contact the bot operator who might have their hands full. Assem Khidhr (talk) 04:06, 12 May 2024 (UTC)
I edited Wikipedia:Articles for deletion/Cloverland, Washington prior to its closure, but probably didn't get something on the correct line. When it closed, it didn't show up at my AFC stats page. I think I erred on where I signed. Tried to correct it after it closed, with no luck. Other than that, it looks like others have also edited the closed template, and possibly re-opened it. I think others need to have a look at the status of this AFD. Thanks. — Maile (talk) 17:45, 12 May 2024 (UTC)
I'm trying to do some upkeep with many of the neglected articles on Zoroastrianism. When I visit WP:Zoroastrianism (or any other page containing the template), it seems the pages by quality template is broken. Thus, I am unable to review the assessment status of relevant articles. Is this problem on the end of WP 1.0? For what reason would the server would return a broken list? Doughbo (talk) 21:47, 8 May 2024 (UTC)
If you're sure about this, then I think your first step should be to "reactivate" WikiProject Zoroastrianism. As part of that you would update Template:WikiProject Zoroastrianism by changing "inactive" to "main", which would repopulate all the categories, which should make everything work again — Martin (MSGJ · talk) 20:34, 9 May 2024 (UTC)
Thanks. I'm still getting to know things around here, like which actions are appropriate, so I appreciate the help. Doughbo (talk) 15:58, 10 May 2024 (UTC)
Thank you very much! I see that the list is now functioning properly on the template page. On the Assessment page, however, the old, broken list is still present. What is needed in order to replicate the new list? If possible, I'd like to also include a line for unassessed articles. Doughbo (talk) 22:57, 11 May 2024 (UTC)
Thank you very much! I think I'm equipped from here on out. I've subscribed to the thread, so I'll check on any updates from there. Doughbo (talk) 21:53, 12 May 2024 (UTC)
OK, if I use Ctrl++ to zoom in one step (to 110%), the left-hand image appears. If I zoom in four more steps (making five in all), to 170%, the second and third appear. By this time, the images are approaching the "natural" size of the image, which is 394 × 532 pixels; my guess is that there's something funny about the jpeg encoding that doesn't allow reduction beyond a certain limit. --Redrose64 🌹 (talk) 15:14, 11 May 2024 (UTC)
There may be something broken with the full-size version of this image that browsers can cope with (which is why that can be displayed), while the software that's supposed to generate smaller versions can't. It may be enough just to download the affected images, open them in an image editing program, save a new copy, and upload that copy as a new version of each affected image (but I believe you need to be autopatrolled on Commons for that). Rummskartoffel16:52, 11 May 2024 (UTC)
Size 100px
MediaWiki appears unable to scale it to any size. Only the original upload works. HTML for the "small" 220px version:
Oddly, for me with Chrome 124.0.6367.158, looking at the diff for the initial post in visual diff mode displays all three images in the diff, but not in the page being displayed below the diff (I only see the leftmost image). (On Firefox 125.0.3, only the leftmost image is displayed in the visual diff and the page displayed below.) isaacl (talk) 17:24, 11 May 2024 (UTC)
Also when I do a simple reload via Ctrl-R on Chrome, all three images are displayed. (I see all three after posting a comment in this section; presumably the equivalent of a simple reload is being done.) If I do a force reload of the entire page via Shift-Ctrl-R on Chrome, just the leftmost image is shown. For me, page zoom doesn't cause the images to display/not display. isaacl (talk) 17:31, 11 May 2024 (UTC)
With both simple reload and force reload, I see three network requests being made. The first is for Nasseraldinshah3.png and it succeeds. The second two are for 330px-Nasseraldinshah3.png and 150px-Nasseraldinshah3.png, and they fail with a 429 code, "Too many requests". (As previously noted, the three images are displayed after the simple reload.) isaacl (talk) 18:05, 11 May 2024 (UTC)
I don't know. I think I understand better now why a simple reload causes the images to appear: since the original file URL (which was successfully retrieved) appears in the srcset for each of the images in the original post (but, as PrimeHunter stated, it doesn't get included for the 100px version), the browser will just use that to refresh the page, rather than try to match the file with the most appropriate pixel density based on the physical display (since with a simple refresh, the browser will try to avoid loading the images again, so there's no need to do a match if any of the images in the set are already available). My device probably has a different pixel density than the other commenters, such that changing the zoom level isn't causing the original image to be a match to be loaded. isaacl (talk) 22:58, 11 May 2024 (UTC)
I've uploaded a corrected version, but due to the 429, it will probably take a while (not sure how long) before we can see the effect of that. —TheDJ (talk • contribs) 09:16, 13 May 2024 (UTC)
It's fixed now. I processed it with the png-fix-IDAT-windowsize utility, to correct the invalid checksum it contained. —TheDJ (talk • contribs) 15:30, 13 May 2024 (UTC)
Migrating SQL data from one toolforge db to another
I run two tools on toolforge, both of which have SQL databases. I'd like to consolidate them under one of the tools and would like to migrate the data in one of the databases to the other. I could do this with a script to extract the data to a file, then connect to the other location and insert the data, but that's a lot more trouble than insert into <newdb>.<newtable> select * from <olddb>.<oldtable>, which is how I'd do it in SQL Server. For that I would need to be simultaneously connected to both databases at the SQL prompt, with write permission on the new database at the same time as read permission on the existing database, which is not currently public. Is there a way to do this at the MariaDB SQL prompt or do I have to just write that script? Mike Christie (talk - contribs - library) 12:42, 11 May 2024 (UTC)
As far as I know, most Toolforge tools' databases are still on the same server. But probably neither tool's database user has read access to the other's. You might try asking for help via one of the channels listed at wikitech:Help:Cloud Services communication. Anomie⚔13:33, 11 May 2024 (UTC)
I could do this with a script to extract the data to a file, then connect to the other location and insert the data. This is how I would do it. Can probably do everything you need from the command line. –Novem Linguae (talk) 16:59, 13 May 2024 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
On Wikisource there is a special page listing pages of works without corresponding scan images. Now you can use the new magic word __EXPECTWITHOUTSCANS__ to exclude certain pages (list of editions or translations of works) from that list. [34]
If you use the user-preference "Show preview without reloading the page", then the template-page feature "Preview page with this template (what's this?)" will now also work without reloading the page. [35]
Kartographer maps can now specify an alternative text via the alt= attribute. This is identical in usage to the alt= attribute in the image and gallery syntax. An exception for this feature is wikis like Wikivoyage where the miniature maps are interactive. [36]
The old Guided Tour for the "New Filters for Edit Review" feature has been removed. It was created in 2017 to show people with older accounts how the interface had changed, and has now been seen by most of the intended people. [37]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 14 May. It will be on non-Wikipedia wikis and some Wikipedias from 15 May. It will be on all wikis from 16 May (calendar). [38][39]
The Special:Search results page will now use CSS flex attributes, for better accessibility, instead of a table. If you have a gadget or script that adjusts search results, you should update your script to the new HTML structure. [40]
Future changes
In the Vector 2022 skin, main pages will be displayed at full width (like special pages). The goal is to keep the number of characters per line large enough. This is related to the coming changes to typography in Vector 2022. Learn more. [41]
Two columns of the pagelinks database table (pl_namespace and pl_title) are being dropped soon. Users must use two columns of the new linktarget table instead (lt_namespace and lt_title). In your existing SQL queries:
Replace JOIN pagelinks with JOIN linktarget and pl_ with lt_ in the ON statement
Below that add JOIN pagelinks ON lt_id = pl_target_id
Template:PanoViewer isn't protected, feel free to adjust it. That external tool isn't supported by us here on the English Wikipedia, and has sparse documentation on where to report problems. You can try to open a bug on it similar to phab:T354949. — xaosfluxTalk15:33, 13 May 2024 (UTC)
Preview warning: Page using Template:Infobox settlement with unknown parameter "pushpin_mapsize1"
Preview warning: Page using Template:Infobox settlement with unknown parameter "pushpin_map_caption1"
Preview warning: Page using Template:Infobox settlement with unknown parameter "pushpin_map1"
It was there when I arrived on the page and I have zero idea about what brings that up. (and I'm supposed to be working and I'm starting to run late on the job, so really no time to get into anything as mysterious as that.)
Anyone would know how to get rid of that, please? If no (very) simple fix is possible, thanks for letting me know whom I can ask who would be willing to sort it out. Have a good day. Pueblo89 (talk) 13:20, 14 May 2024 (UTC)
May 2024 Accessibility for Reading Update: appearance menu, upcoming font size changes and customization, dark mode, and more
The Appearance menu with the corresponding parts: 1. Typography 2. Dark mode 3. Page width options
Hi everyone! This is an update on the current work of the Web team on the Accessibility for Reading initiative that introduces changes to the Vector 2022 and Minerva skins. It improves readability, and allows everyone, both logged-out and logged-in users, to customize reading-focused settings. It will also increase the default font size for logged-out users of wikis and introduce dark mode. Customization will be done through the new Appearance menu, which will contain options for customizing 1. typography (font size, line height, and paragraph spacing combined), 2. color scheme (dark/light), and 3. page width.
1. Appearance menu, typography improvements, and font size changes for logged-out users
In December 2023, we introduced the "Accessibility for Reading" beta feature. It adds the Appearance menu which allows logged-in users to choose different font sizes and page widths. This makes it possible for people to read in a font size that they prefer and customize their experience. Since the introduction of the menu, we have studied its usage and performed usability testing:
No significant issues were observed, but based on the feedback, we have changed the copy and behavior of the menu (for example, we have renamed the "night" color scheme to "dark").
The majority of users who interact with the feature opt for a font size that is larger than the current default. This confirms our hypothesis that most users prefer a larger font size. We previously saw this in the findings from the community prototype testing.
Based on this data, we plan on increasing the default font size for all readers. The current (smaller) default will be kept for logged-in users and as one of the options so that anyone can return to the font size used in the past.
2. Dark mode
Dark mode is a highly-requested reading feature which was the top wish on last year’s Community Wishlist. We will bring dark mode to the desktop and mobile websites. Currently, dark mode is available to logged-in users on the mobile site (Minerva skin only). To access this feature, you must first be opted into advanced mode for mobile. Then, you can select "dark" from the list of color options. ("Automatic" will follow the preference of your device.) See the more detailed message about this change.
3. Page width options
With the introduction of Vector 2022, we made available the option for all readers and editors to switch their view from limited to full with a toggle in the bottom right corner of the screen.
That limited/full width toggle will be moved to the Appearance menu instead of being in the corner of the screen.
We will be also switching some pages to full width by default, including the main page.
Coming up soon!
Over the next weeks, we will be preparing for significant deployments related to this project:
First, we plan on introducing dark mode to the desktop website as a beta feature.
Then, we will bring the Appearance menu for all users and change the default font size for all logged-out readers.
Finally, after a period of testing, we plan on bringing dark mode into the Appearance menu for all users on the desktop and mobile websites.
The above message says: The majority of users who interact with the feature opt for a font size that is larger than the current default. This confirms our hypothesis that most users prefer a larger font size. Leaving aside the selection bias problems in that statement, the phabricator link appears to indicate the opposite, that 44.9% of sessions eventually opt for small (default: font=0) font size; 43.6% of sessions eventually opt for standard (font=1) font size; 11.5% of sessions eventually opt for large (font=2) font size. What am I missing? – Jonesey95 (talk) 14:34, 13 May 2024 (UTC)
Hey @Jonesey95. The selection bias part needs to be addressed by someone else than me, but to your point about the link indicating the opposite: Standard + Large (larger font size) > Small (default). SGrabarczuk (WMF) (talk) 16:36, 13 May 2024 (UTC)
That's what I was missing: the small size was the default. I confused the word "standard" with "default", I suppose. – Jonesey95 (talk) 17:07, 13 May 2024 (UTC)
Hey @Jonesey95 - wanted to leave a quick note on the selection bias question. In terms of the selection bias, we recognize that there is a very specific audience that will be opted into our beta feature, which is a downside of any beta feature data to begin with. This is generally an audience of active logged-in users who are interested in trying out new features in the first place (they are quite different from the average reader). For some features, that might be a sufficient proxy. For others, it might not make as much sense. In general, we try to get data from multiple different sources in order to account for this. In this case, we did prototype testing on a subset of logged-in users across languages, user testing with logged-out readers, as well as the beta feature launch. We've also launched the feature as the default for logged-out users on a small set of wikis to see if the data we're seeing there resembles what we saw on beta. These should be sufficient predictors of the behavior in production, but it's important to measure and double-check. Once we have a wider release, we'll do a similar comparison as well. OVasileva (WMF) (talk) 14:41, 14 May 2024 (UTC)
Anchoring new sections to set location on talk page
As I've formatted my user talk page in a manner that encloses all sections in a table, I was wondering if there is a way to have new talk page sections added at a specific point on a talk page, rather than simply at the end. Specifically, a means that would work with the default 'New section' button above the talk page, rather than the addition of a second, custom 'New section' wikilink or button on the talk page itself. -CoolieCoolster (talk) 01:35, 15 May 2024 (UTC)
If you remove the closing |}</div> tags, new sections will effectively go inside your custom frame. On the other hand, you might get people grumbling about this being a Linter error. Framing talk pages like this isn't well supported. Matma Rextalk02:40, 15 May 2024 (UTC)
CORS auth
Hello. I'm trying to edit Wikidata from Wikipedia (ro.wp to be precise) using a user script, but I am unable to do it under my user. The code from ro:MediaWiki:Gadget-wikidata-description.js works, it obtains an edit token and does the change, but in the history my IP appears. I made sure I am logged in both wikis. Can anyone give me an example of working code or some pointer on how to do the edits as logged-in user? Thanks! Strainu (talk) 08:59, 14 May 2024 (UTC)
mw.ForeignApi is the supported interface for accessing sister site APIs. Fetching a token manually isn't required, postWithEditToken() will do it for you. – SD0001 (talk) 09:08, 14 May 2024 (UTC)
I do get this message "We are unable to validate your login credentials. Please contact your institution for assistance. Please note, Referring URL authentication may have been prevented by antivirus or privacy control software" but I'm not convinced that's the issue. But if others are able to login, it must be. Thanks. Doug Wellertalk09:34, 11 May 2024 (UTC)
I don't have the issue when I can see my credential in the upper right hand corner, and then search for Wikipedia:TWL . I can then see the library when I click on database access. Maybe the message is an artifact. -- Ancheta Wis (talk | contribs)09:50, 11 May 2024 (UTC)
The error message matches one at EBSCO, which is just one of the partners with TWL. I doubt it's TWL you can't log into. Nardog (talk) 10:18, 11 May 2024 (UTC)
@Nardog@Ancheta Wis That makes sense. I can login to the WL but not use it to search for articles. I can click on Access Collection - but I'm puzzled by the fact that when I click on Brill it looks as though it just takes me to their web page, and if I click on a books offers it for sale. I've never clicked on Brill before so don't know if that's what it should do. Doug Wellertalk12:14, 11 May 2024 (UTC)
Hello Doug Weller! I'm having the same problem, can get to the Library page but as soon as I use the search I'm taken to the EBSCO Error page you describe. Others also reporting this problem, like here. I use the Library all the time, use it to access ProQuest as an invaluable source for referencing. Hope this is just a temporary glitch. Take care, LooksGreatInATurtleNeck (talk) 22:37, 11 May 2024 (UTC)
Hi, I'm encountering a weird bug (it's not the first time it's happened). When I type text in a long page in visual source editing (the 2017 wikitext mode), I'm getting some funky behaviour where what I'm actually typing ends up being multiple lines below my cursor. The video attached is this occurring on this page, but it sometimes happens on other such long pages.
Smells like a code highlighter userscript/gadget bug. Are you able to reproduce when adding ?safemode=1 / &safemode=1 to the URL? This will turn off gadgets and user scripts. I suspect you will not be able to reproduce it. Next step after that will be tracking down which code highlighter userscript/gadget and reporting the bug on their talk page. –Novem Linguae (talk) 08:07, 15 May 2024 (UTC)
For some reason, if I go to the page history of any enwiki page, the undo button now says "cin gbere le" instead of "undo", as shown in the screenshot here. It just started happening today all of a sudden. Anyone else had this problem?
Short answer, don't use en-gb. Long answer - its upstream vandalism at translatewiki of the MediaWiki:Editundo message in the en-gb language. Language (Warning: Selecting a language other than 'en - English' will prevent you from seeing localized parts of the interface on the English Wikipedia, and you may see inaccurate external translations.): — xaosfluxTalk09:04, 16 May 2024 (UTC)
The upstream vandalism seems to have been cleared, but now you will have to wait for sync. Or just change to en. — xaosfluxTalk09:09, 16 May 2024 (UTC)
I see! Lol. Indeed, I have my interface language set to en-gb. Though I'm not sure how a protected page like that got vandalised. I guess it's unprotected on the translatewiki. — AP 499D25(talk)09:16, 16 May 2024 (UTC)
It's not quite the most aggressive action by the Provisional Change to EN Brigade that I've ever seen, but it's up there. DuncanHill (talk) 10:09, 16 May 2024 (UTC)
Non-English en-gb messages are nearly always somebody editing the wrong Translatewiki messages when trying to translate another language. Here translatewiki:User:Umar Ahmad2345 was trying to make Nupe (nup) which sounds amusingly like noob. It's not an option in our preferences but maybe it's coming. Incubator has a Nupe Wikipedia. PrimeHunter (talk) 11:09, 16 May 2024 (UTC)
Search results have a thumbnail next to them, or a gray placeholder if there's no suitable image in the article. Since today, those placeholders are bigger than regular thumbnails, which also places them uncomfortably close to the article text. It appears to be caused by this CSS rule:
.searchResultImage .searchResultImage-thumbnail > div {
padding: 0.5em;
}
ChristieBot, which manages GA nominations, has been crashing when trying to transclude the GA review to Talk:Israeli invasion of the Gaza Strip (2023–present). The error is "Page en:Special:Log is not supported due to namespace restriction". The page is extended-confirmed protected. The bot does have extended-confirmed rights as part of the bot group rights, but presumably is missing some other related permission. I can change the code to not crash and simply log the error, but I'm at work and won't be able to touch the code for a few hours. If anyone can tell me what the problem is I would appreciate it. Thanks. Mike Christie (talk - contribs - library) 12:48, 16 May 2024 (UTC)
Is this something I can set myself, and if so, where? (I'm not an admin.) I thought that a bot just needed the same user rights as a non-bot editor would need to edit a given page. Mike Christie (talk - contribs - library) 13:46, 16 May 2024 (UTC)
How is your bot logging in? Through the webui, or the api? If the API are you using a consumer grant or BotPasswords? — xaosfluxTalk13:51, 16 May 2024 (UTC)
I don't have access to toolforge from where I am now, and since it was set up eighteen months ago I don't recall the details offhand, but as I recall I have a config set up in the toolforge shell account that handles the bot authentication. I think that means it uses BotPasswords? Mike Christie (talk - contribs - library) 14:04, 16 May 2024 (UTC)
@Mike Christie You'd have to check your PyWikiBot user-config.py file. If it has an authenticate[] = line you're using oauth, if it has a password_file = line you're using BotPasswords. --Ahecht (TALK PAGE)14:08, 16 May 2024 (UTC)
I don't recall getting oauth to work so I think it must be BotPasswords, which I remember dealing with. (If SD0001 is available and has a moment they may be able to look and confirm as they are a maintainer for that bot.) Mike Christie (talk - contribs - library) 14:19, 16 May 2024 (UTC)
It's using BotPassword. It should be able to edit the page if the appropriate grant is checked as mentioned above. But from the error message, I am wondering if there is a code issue as well - the bot seems to be literally trying to edit Special:Log which of course isn't possible. – SD0001 (talk) 17:58, 16 May 2024 (UTC)
That flag is now checked but it hasn't fixed the issue. The issue only occurs when trying to edit that page, using the same code that it uses for transcluding GA reviews on other talk pages, so I can't see why Special:Log is in the picture. There's certainly no explicit attempt by the bot to try to edit the special page. The error happens when it attempts to add text to the Page object's text field: in the traceback I see self.botMayEdit() is called, in pywikibot/page/_basepage.py, which eventually pops out to the Special:Log error for some reason. Later I will have time to add some code to trap the error so at least it'll finish its run, even if it can't do that particular transclusion. I'll do that this evening unless someone has any further ideas. Thanks for the help so far. Mike Christie (talk - contribs - library) 19:04, 16 May 2024 (UTC)
This appears to be an error in pywikibot, I can confirm it also occurs for me.
Full traceback
--------------------------------------------------------------------------- UnsupportedPageError Traceback (most recent call last) Cell In[1], line 4 2 site = pywikibot.Site('wikipedia:en') 3 page = pywikibot.Page(site, 'Talk:Israeli invasion of the Gaza Strip (2023–present)') ----> 4 page text = 'foo' File /srv/paws/lib/python3.10/site-packages/pywikibot/page/_basepage.py:576, in BasePage.text(self, value) 571 """Update the current (edited) wikitext. 572 573 :param value: New value or None 574 """ 575 try: --> 576 self.botMayEdit() # T262136, T267770 577 except Exception as e: 578 # dry tests aren't able to make an API call 579 # but are rejected by an Exception; ignore it then. 580 if not str(e).startswith('DryRequest rejecting request:'): File /srv/paws/lib/python3.10/site-packages/pywikibot/page/_basepage.py:1146, in BasePage.botMayEdit(self) 1131 """ 1132 Determine whether the active bot is allowed to edit the page. 1133 (...) 1143 user cnfig file (user-config.py), or using page.put(force=True). 1144 """ 1145 if not hasattr(self, '_bot_may_edit'): -> 1146 self _bot_may_edit = self._check_bot_may_edit() 1147 return self _bot_may_edit File /srv/paws/lib/python3.10/site-packages/pywikibot/page/_basepage.py:1163, in BasePage._check_bot_may_edit(self, module) 1161 username = self.site.username() 1162 try: -> 1163 templates = self.templatesWithParams() 1164 except (NoPageError, IsRedirectPageError, SectionError): 1165 return True File /srv/paws/lib/python3.10/site-packages/pywikibot/page/_page.py:78, in Page.templatesWithParams(self) 62 """Return templates used on this Page. 63 64 The templates are extracted by :meth:`raw_extracted_templates`, (...) 74 :rtype: list of (pywikibot.page.Page, list) 75 """ 76 # WARNING: may not return all templates used in particularly 77 # intricate cases such as template substitution ---> 78 titles = {t.title() for t in self.templates()} 79 templates = self raw_extracted_templates 80 # backwards-compatibility: convert the dict returned as the second 81 # element into a list in the format used by old scripts File /srv/paws/lib/python3.10/site-packages/pywikibot/page/_basepage.py:1612, in BasePage.templates(self, content) 1609 del self _templates 1611 if not hasattr(self, '_templates'): -> 1612 self _templates = set(self.itertemplates(content=content)) 1614 return list(self._templates) File /usr/lib/python3.10/_collections_abc.py:330, in Generator.__next__(self) 326 def __next__(self): 327 """Return the next item from the generator. 328 When exhausted, raise StopIteration. 329 """ --> 330 return self.send(None) File /srv/paws/lib/python3.10/site-packages/pywikibot/tools/collections.py:279, in GeneratorWrapper.send(self, value) 276 if not hasattr(self, '_started_gen'): 277 # start the generator 278 self _started_gen = self generator --> 279 return next(self._started_gen) File /srv/paws/lib/python3.10/site-packages/pywikibot/data/api/_generators.py:625, in QueryGenerator.generator(self) 622 self normalized = {} 624 try: --> 625 yield from self._extract_results(resultdata) 626 except RuntimeError: 627 break File /srv/paws/lib/python3.10/site-packages/pywikibot/data/api/_generators.py:567, in QueryGenerator._extract_results(self, resultdata) 565 """Extract results from resultdata.""" 566 for item in resultdata: --> 567 result = self.result(item) 568 if self _namespaces and not self._check_result_namespace(result): 569 continue File /srv/paws/lib/python3.10/site-packages/pywikibot/data/api/_generators.py:736, in PageGenerator.result(self, pagedata) 734 elif ns == Namespace.CATEGORY: 735 p = pywikibot.Category(p) --> 736 update_page(p, pagedata, self.props) 737 return p File /srv/paws/lib/python3.10/site-packages/pywikibot/data/api/_generators.py:983, in update_page(page, pagedict, props) 966 def update_page(page: pywikibot Page, 967 pagedict: dict[str, Any], 968 props: Iterable[str] | None = None) -> None: 969 """ 970 Update attributes of Page object *page*, based on query data in *pagedict*. 971 (...) 981 supported yet 982 """ --> 983 _update_pageid(page, pagedict) 984 _update_contentmodel(page, pagedict) 986 props = props or [] File /srv/paws/lib/python3.10/site-packages/pywikibot/data/api/_generators.py:882, in _update_pageid(page, pagedict) 880 raise InvalidTitleError(f"{page}: {pagedict['invalidreason']}") 881 if int(pagedict['ns']) < 0: --> 882 raise UnsupportedPageError(page) 883 raise RuntimeError(f"Page {pagedict['title']} has neither 'pageid'" 884 " nor 'missing' attribute") UnsupportedPageError: Page en:Special:Log is not supported due to namespace restriction.
Hi. One of my pastimes is trying to empty Category:Harv and Sfn no-target errors. Today I found Military courts of the United Kingdom, which has multiple uses of {{sfn|Armed Forces Act|2006}} which are meant to link to {{UK-LEG|title=Armed Forces Act 2006|path=ukpga/2006/52 |ref={{harvid|Armed Forces Act|2006}}}} but don't. I can't work out how to fix the no-target errors and get the link from sfn to citation template to work. Any help will be appreciated, thank you. DuncanHill (talk) 10:22, 16 May 2024 (UTC)
{{UK-LEG}} does not know about |ref= so does not create an anchor ID. Without an anchor ID, {{sfn}} has nothing to link to. Your options are to modify {{UK-LEG}} or implement an appropriate solution from the list of possible solutions at :Category:Harv and Sfn template errors § Resolving errors: {{wikicite}} (best) or {{anchor}} (not so best).
I don't think the talk page will get the comment any light, so I am putting it here. It does not require any discussion but a confirmation from technical person that it won't cause any issue. Thanks, ExclusiveEditorNotify Me!10:51, 17 May 2024 (UTC)
Nothing you change in a "Template:XXXXXX/doc" page is going to break anything on other pages; worst case is it confuses someones that reads the documentation. — xaosfluxTalk12:49, 17 May 2024 (UTC)
Main Page always full width
I Can't edit the Main Talk page so I'm posting here.
The main page (and it's talk page) seems to be stuck in full width mode even when I click on the full width toggle button in the bottom right corner. Other pages seem unaffected.
Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:125.0) Gecko/20100101 Firefox/125.0
In the Vector 2022 skin, main pages will be displayed at full width (like special pages). The goal is to keep the number of characters per line large enough. This is related to the coming changes to typography in Vector 2022. Learn more. [44]– robertsky (talk) 06:00, 18 May 2024 (UTC)
Need help fixing the "second" sticky row in this table. With rowspan used in the first column, only the first row in rowgroup is sticky, the rest of the rows in individual rowgroups aren't. I assume it is a "sticky-col2" issue from Template:COVID-19 pandemic data/styles.css. Can it be fixed so it displays all rows in the second column unbroken when horizontally scrolling? Qwerty284651 (talk) 19:04, 18 May 2024 (UTC)
Native Dark Mode Message
I have been using the dark mode gadget for a while now and I quite like it. I tried the native dark mode but I do not like it that much, as it is quite buggy and has a fair few contrast issues, see the phab board. Every time I load a page in namespaces outside of article space I get the message that informs me that I have two dark modes enabled and I need to choose one. When I go to disable the non-gadget dark mode it is not able to be disabled, are there any scripts to suppress this message, it is getting quite annoying. Thanks, v/r - Seawolf35T--C00:03, 19 May 2024 (UTC)
I use this search query to see pages tagged for speedy deletion. For several days it has returned BBL Drizzy as a result even though that page is not tagged for deletion. The search results say it was last edited at 01:12 on 13 May 2024, which corresponds with Special:Diff/1223572833 (now in the history of Draft:BBL Drizzy). Purging or null-editing either the article or the draft will make it disappear from the results temporarily, but so far it's always come back shortly thereafter. Today I tried deleting and undeleting the pages, but this didn't help either. Any suggestions? (This query brings up strange results too, for what it's worth.) Extraordinary Writ (talk) 03:01, 19 May 2024 (UTC)
I don't experience this. BBL Drizzy is not a result for the first query, and the last query only brings up BBL Drizzy (last edited 19:04, 18 May 2024). – 2804:F1...7F:938A (talk) 03:15, 19 May 2024 (UTC)
I've been seeing some outdated search results also related particularly to a couple page moves that got done between when the search was crafted and when my fix was made. Particularly, in this search I'm seeing Draft:Peter the Great Interrogating the Tsarevich Alexei Petrovich at Peterhof and Draft:Ellen Webster Palmer when I hit one or another of the data centers (which I assume is at least part of the problem), but at least once I've gotten a search not to return those two, because neither are drafts any longer nor do either carry the problem that makes them show up in this search. Izno (talk) 20:36, 19 May 2024 (UTC)
The Appearance menu and new default standard font size will be available for logged-out users
Hi everyone! We are the Wikimedia Foundation Web team.
We work on making it easier to read Wikimedia projects as part of the objective "Reading and media experience" of the current year’s annual plan. To achieve this goal, we have introduced the "Accessibility for Reading" beta feature. It adds a menu which works on the Vector 2022 skin and allows logged-in users to choose different font sizes and color schemes based on individual needs.
The menu introduces a new Standard font setting. It slightly increases the size and height of the font. It was selected based on multiple sources. You will find more information on this in the section "About the new Standard font setting". As we announced last week, we are now ready to begin bringing some of these feature out of beta and making them available to more people.
What will change
We are now ready to make the new Appearance menu available for logged-out and logged-in users.
At the same time, we will also make the Standard option the new default for logged-out users only.
If no breaking technical issues are found, we plan on making this change within the next three weeks.
Later, this menu will also include the option to select dark mode, which for the time being will remain a beta feature. For more information, check out our project page.
About the menu
The new menu will allow logged-in and logged-out users to set preferences for:
Text size and line height (available now as the beta feature): Users will be able to choose between the Small (current default), Standard (recommended for better accessibility), and Large options. Selecting an option will change both the font size and line height of the text.
Dark mode (available now as the beta feature): Users will be able to choose to see the site in night mode on a permanent basis, or select an "automatic" setting which will set day or night mode based on the device or browser preferences.
Content width (previously available as a toggle button): We have moved the content width toggle from an icon at the bottom of the page to a labeled radio button in the new menu. It will work exactly the same as the toggle. The previous toggle button will no longer be available.
This menu has been tested as a beta feature by logged-in users across wikis as well as in user testing with readers. Based on the findings of these tests, we changed the menu to improve discoverability and ease of use, and to accommodate gadget compatibility.
The menu will appear to the right of the page, immediately under the Tools menu if that has been pinned. Unlike the Tools menu, the Appearance menu is pinned by default, but can be unpinned. Once unpinned, it collapses under an icon at the top of the page.
About the new Standard font setting
The "Small" option is the current default. We will be changing this default to "Standard" for logged-out users, while keeping "Small" as the default for logged-in users. The "Standard" and "Large" options were built and tested based on the following:
Academic studies and recommendations for the best average font size for the majority of readers. These recommendations stated that our current size is too small for the majority of people to read comfortably. This means that on average, people read more slowly, strain their eyes while reading, or have difficulty clearly seeing the text. Increasing the font size by default improves these issues for all users, including users who might not have sufficient time to spend adjusting a setting via the appearance menu or browser. Information density is also important, which is why we wanted to increase font size without sacrificing information density. We have achieved this by changing not only font size, but also line height and paragraph spacing.
Designs submitted by more than 630 Wikipedians from across 13 wikis of different languages, scripts, and sizes. The majority (~450) of these users opted for a font size that was larger than the default. "Standard" represents the average of the most popular cluster of responses (15-20 pixels). "Large" represents the need for an even larger option, as represented by the cluster of sizes between 21-26 pixels. You can read more on how we included volunteers in the process and landed on these options.
Beta feature usage showed that the majority of users who interact with the feature at least once opt for a font size that is larger than the current default.
Our works so far and next steps
Logged-in users will remain with the "small" setting for the time being as their default, but can change to any other setting at any time. In a few months, we will study how many logged-in users switch to standard and start a conversation on whether it makes sense for logged-in users to make the switch as well. From the early data from the beta feature, 55% of sessions who interacted with the feature chose to use a setting that was standard or larger.
If you'd like to help, we have a few simple requests for you:
Try out the new menu. Is anything confusing? Do you understand all the labels and how the menu works?
Try out the small, standard, and large sizes, the color schemes, and the width toggle. Reach out to us if you notice any bugs, or have questions or concerns.
I noticed this at Aaron Swartz and Web feed: links to the article on the web feed format RSS (link here > RSS <) are not displaying on Wikipedia articles
The link displays and works normally for me in all five tested browsers. It sounds like something in your browser is messing with the link, maybe an extension for RSS feeds. PrimeHunter (talk) 20:08, 20 May 2024 (UTC)
Resolved Thanks, didn't even think that text on Wikipedia would be targeted by a content blocker. 1Blocker "Block Annoyances" setting responsible. PK-WIKI (talk) 20:21, 20 May 2024 (UTC)
Dark mode is available in Vector 2022 as a beta feature
Show what we've built very early. The earlier you are involved, the more your voices will be reflected in the final version
Get your help with flagging bugs, issues, and requests
Work with technical editors to adjust various templates and gadgets to the dark mode
Known limitations
Dark mode is only available for logged-in users: on desktop as a beta feature, and on mobile in the advanced mode.
Gadgets may initially not work well with dark mode and may have to be updated.
Our first goal is making dark mode work on articles. Special pages, talk pages, and other namespaces (including Wikipedia) have not been updated to work in dark mode yet. We have temporarily disabled dark mode on some of these pages.
What we would like you to do
Our request to you is exactly the same as previously:
I don't know if this was the right move. Tons of pages still look like garbage (shoutout Special:Watchlist)—the theme is definitely not ready for use by non-technical editors. It wouldn't be too bad if you had made a seperate beta option for it—because you've clumped it in with Accessibility for Reading a couple people have gone on the Discord confused. Snowmanonahoe (talk·contribs·typos) 23:54, 15 May 2024 (UTC)
"Fixing" it is as simple as setting your personal chosen theme to light mode.
As for tons of pages, I just got separate word that OOUI interfaces don't support light mode (currently/ever?), which is why Watchlist has light elements. Perhaps it shouldn't be displaying as dark. Izno (talk) 00:08, 16 May 2024 (UTC)
The point is for editors to find these things. Might as well drop logged users in and let them make the two button presses to find a different, less-painful choice. Izno (talk) 00:41, 16 May 2024 (UTC)
Exactly, there is no magic fairy dust for some of this, just grunt work to fix the pages and templates that never had to account for a darkmode. —TheDJ (talk • contribs) 14:56, 16 May 2024 (UTC)
We do have a long-term solution for OOUI fixes that is in development right now. We hope to have it ready within the next couple of weeks. In the meantime however, pages like Watchlist and should be displaying in light mode so this is a bug. We're tracking it in phab:T365084 and should have a fix out later today. OVasileva (WMF) (talk) 07:36, 16 May 2024 (UTC)
@SGrabarczuk (WMF): Is there really a plan to take over the right 1/4 of screen with a "settings" panel every time someone opens a page in a new private tab, or clears cookies, or switches to a new language, or doesn't realize what the "hide" button does? Or is that just so in-your-face during the testing phase? Suffusion of Yellow (talk) 02:01, 16 May 2024 (UTC)
Hey @Suffusion of Yellow - thanks for the question. In general, we think this is the most effective way to launch the new menu and be able to inform users about it. It removes the necessity to add additional modals or notices that say "New menu available!" or "Dark mode available", which would otherwise show on every pageview. So, the current plan is to keep the menu open by default for the wider release as well for all logged-in and logged-out users. Then, after the release, we can look at the data to gauge whether we need to keep it open as default, or collapse, and when. OVasileva (WMF) (talk) 07:40, 16 May 2024 (UTC)
Why can't it be on the left side, with the TOC? Or better yet, just put a plain-text link at the top? I mean, registered users have already been able to find the "preferences" link for 20 years now. Right when I open up testwiki:LOREM IPSUM (or any page with a TOC) on my phone in a new private tab, then switch to the desktop site, about half my screen is whitespace, and the text is squished into a narrow little strip in the middle. It's unusable until I hide the settings column
People are reading a website, not installing software. There shouldn't be a "setup" stage before they can get down to the business of looking up that one little fact. The defaults should at least be usable. Suffusion of Yellow (talk) 17:11, 16 May 2024 (UTC)
Hi @Suffusion of Yellow - we generally have most personalized tools (page tools, the user menu, preferences) on the right side of the page, which is why we chose this location. It would have been confusing to show a menu on the left side of the page, next to the content-related ToC, only to then collapse it into the right side. Hope that makes sense.
In terms of the defaults - I couldn't agree with you more. We want our defaults to be as usable as possible (and also to provide an easy opportunity for customization). We are currently in the process of defining and setting defaults for all three areas in the menu. These default will become available as the features themselves are ready. For example, the default for dark mode will be "automatic", which will follow the user's device settings. OVasileva (WMF) (talk) 06:46, 21 May 2024 (UTC)
@SGrabarczuk (WMF):https://night-mode-checker.wmcloud.org/ appears to only be for the mobile version. Is there is desktop report? Also, the beta is only for Vector 2022. Many editors still use legacy vector or other skins, will the new feature be moved to any other skin besides Vector 2022? How does the new feature compare to the dark mode gadget that is currently available? RudolfRed (talk) 03:31, 16 May 2024 (UTC)
I recently noticed that the CSS style background:transparent seems to turn the text color dark gray when combined with the class thumbinner and viewed through the beta vector night mode. Does anyone know why this happens? Andumé (talk) 20:12, 20 May 2024 (UTC)
In general however, if you define a background color, you should ALWAYS define a text color and vice versa, if you want to avoid problems with night/dark mode. —TheDJ (talk • contribs) 10:00, 21 May 2024 (UTC)
In dark mode, there is some blanket CSS for "it has a background but is relying on inherited color" which obviously doesn't work in a lot of places. Izno (talk) 17:13, 21 May 2024 (UTC)
Where is the code for an info-box? Where does it go?
I'm trying to create some info-boxes on a small hobby wiki. It had been running fine on "freewiki.in" but that domain disappeared last year. I've re-created it on a domain I control using MediaWiki v1.41.1. I've put up all the article text from .XML backups, but all the formatting is gone.
The software manual at https://www.mediawiki.org/wiki/Global_templates/Taxonomy explains how to use {{ambox}} but not the code required for it to work. It sends me to Wikipedia, to https://en.wikipedia.org/wiki/Module:Message_box.
That, finally, has 600+ lines of code.
Where does this code go? Somewhere on the wiki? Somewhere on the server in one of the .PHP files? There's gotta be a step-by-step procedure somewhere but I can't find it. And I'm not about to do my experimenting here on Wikipedia...
Answers can be here, if this is of general interest. If not, I'd be perfectly happy getting help at my talk page. SandyJax (talk) 13:05, 21 May 2024 (UTC)
If you preview code, e.g. an infobox call, then the bottom of the window shows all transcluded templates and modules at "Templates used in this preview". It can be a long list, and the list can change if the same template is called with different parameters or circumstances (e.g. the namespace it's called from). Some of the infobox styling in the English Wikipedia is in MediaWiki:Common.css which is not listed at "Templates used in this preview" since it's not transcluded but used in another way. PrimeHunter (talk) 18:30, 21 May 2024 (UTC)
It is not malfunctioning, it is doing what it is supposed to do, to notify everyone about the ongoing discussion. But I can see how this can be disruptive to the reading experience when it is being used multiple times within a short space, i.e. at Atlanta Braves#Braves Hall of Fame. – robertsky (talk) 03:45, 22 May 2024 (UTC)
I am quite happy that Wikipedia uses the IPA as a pronunciation guide. I know the IPA moderately well, so its use is convenient for me. Promoting the IPA in Wikipedia might also encourage people to learn it, which probably might not be a bad idea.
I think that the use of the IPA in Wikipedia might often or consitently make some errors. I don’t see these errors being made in Wiktionary though. The IPA there seems to be largely fine.
the approximant rhotic (also the approximant alveolar)
/ɹ/
The English language uses primarily for a rhotic the approximant alveolar which in the IPA is represented as: /ɹ/.
In Wikipedia, this rhotic seems to always get represented by /r/, which actually represents the rolled “r”, as the “r” in the Spanish word “perro”.
I’ve suspected that this error is deliberate in that maybe the people setting up the IPA transcriptions are concerned that an upside down “r” might confuse people. Although, this doesn’t make very good sense since much of the rest of the IPA will still confuse people who are not familiar with it.
These might be described as the “basic a” vowels: /ɑ/, /a/, /æ/
/a/ and /æ/ are similar and are like the “a” in “cat”.
/a/ is more open and might be more typical for Received Pronunciation or a British English. /æ/ is similar sounding but just with an articulation that is a bit more close (closed jaw) and is probably more typical for General American English or Standard Canadian English. Distinguishing between these two is not too much of a big deal, I don’t think for general pronunciation purposes.
The /ɑ/, described as a back, open, unrounded vowel, is like the “o” in the English word “top”. This vowel is more of what is considered an “a” vowel in languages such as Spanish, Italian, and other languages. This /ɑ/ though is significantly different from the /a/ and /æ/.
I’ve seen on a number of occasions in Wikipedia in which words with the /ɑ/ sound were given IPA transcriptions of /a/. I went looking for examples in Wikipedia of this today, but had difficulty finding this.
Here’s an example from Wiktionary where the /ɑ/ sound is written as /a/.
Pronunciation
(UK) IPA(key): /ˈɒn.tʊ.ɹɑːʒ/, /ˈɑ̃ː.tʊ.ɹɑːʒ/
(US) IPA(key): /ˈɑn.tə.ɹɑʒ/
I finally managed to find an example of the /ɑ/ sound being written as /a/.
I noticed in my search for words with an /ɑ/, for words that have this sound in General American English or Standard Canadian English, that same sound in Received Pronunciation often gets pronounced as /ɒ/, which is a very open “o” and pronunciation given in Wikipedia may tend to favor Received Pronunciation.
the schwa and the caret
This is an issue that probably would have to be directed towards the IPA Association or something like that, and not Wikipedia. But, I’ll mention it here anyways.
Apparently, any phonetics course, most phonetics courses, will tell you that no word in English ends with a caret. So, take the word “the” or “California”. Wiktionary will transcribe these into the IPA as :
It’s typical to see IPA transcriptions of the words “the” and “California” as ending with the schwa, /ə/.
Transcribing the final vowel in the words “the” and “California” as /ə/ to me seems wrong. Maybe my hearing sense is off. Maybe the articulation is /ə/ but to me sounds like /ʌ/ or /ɐ/.
Give it a try? The /ə/ or schwa is the vowel sound you more or less get between the “p” and “l” in “apple” in General American and probably Received Pronunciation. Or find some online IPA chart with audio pronunciation to find out how /ə/ is pronounced.
Now try sticking that sound into the word “the” or at the end of the word “California”. Maybe I’m not getting this right, but it consistently sounds wrong! Like weird!
Maybe when speaking quickly these sounds will reduce to an /ə/ sound. But even when spoken quickly, these words can often just sound like they end with a quick /ʌ/ or /ɐ/.
I used to think that “the” and “California” should be transcribed as /ðʌ/ and /ˈkalɪfɔɹnjʌ/, with the the /ʌ/ at the end of the word. This makes auditory sense to me.
More recently though, when realizing that / ʌ/ and /ɐ/ sound very similar, that for that sound, if the preceeding consonant is more forward, then maybe the more forward of /ʌ/ and /ɐ/ would actually be articulation being used, which is the /ɐ/. So, the word “the” would more likely be transcribed as:
/ðɐ/
With “California”, this might be different. The /j/ sound before the final vowel, it’s more central so would it make the final vowel sound /ɐ/ or /ʌ/? Maybe the /ʌ/ sound fills an entire linear area that connects the more middle /ɐ/ and the more back /ʌ/, in which case it could get kind of complicated or difficult to figure out where the articulation of this sound is taking place exactly.
I suspect that if the /ʌ/ sound is preceeded by a more back consonant, such as in the English word “gut”, then the /ʌ/ sound will take more of a back /ʌ/ articulation, as you get with the Wiktionary transcription of “gut”.
This is not a technical inquiry. The appropriate venue is the talk page of the IPA key for each language, though I strongly suggest you read the Handbook of the IPA and learn the difference between a phone and a phoneme and the concept of narrowness first. Nardog (talk) 06:57, 22 May 2024 (UTC)
trying to add new GeoJSON map to virgin river article - centering is getting messed up
{{mapframe|from=Virgin River (Utah).map|text=Route of the Virgin River and the North and East Forks|frame=yes|zoom=7|frame-coord|coord=37.174167,-113.326111}}
The problem is that every time I preview my proposed change the center of the map is completely off. It's like it's pulling the coordinates of the mouth from wikidata.org and using that vs using the more central coordinate that I'm using.
Assuming that that's the case idk that I necessarily want to change it. Ideally I'd like to just overwrite it for that one map. But if that's not possible I guess I could change the wikidata.org entries altho https://www.wikidata.org/wiki/Q1676970 shows multiple coordinate locations. Would just the first one need to be changed? I'm a little hesitant to save experimental changes like that. I'd rather Preview my changes before saving them but, assuming my hypothesis is correct then I'd prob need save it all the same, even if only to save just to preview TerraFrost (talk) 12:04, 22 May 2024 (UTC)
You are not setting |frame-coord= so it's retrieiving the coordinates from Wikidata. You have |frame-coord| which is acting as a positional parameter, equivalent to |1=frame-coord. You want to use |frame-coord={{Coord|37.174167|N|113.326111|W}} which will allow you to shift the centre. — Jts1882 | talk14:11, 22 May 2024 (UTC)
Edit Patrol on Android Wikpedia mobile app is now available on English Wikipedia and all wikis!
We invite you to check out a demonstration of the feature and explore more on our project page. If you have tried the feature, we would appreciate any feedback on our discussion page, or your thoughts in response any of the following questions:
What changes or additions would make this feature more useful to you?
Would you be interested in using this feature to patrol multiple language wikis at once?
The feature is currently available to users with rollback rights: would you like to see this feature evolve into something that is available to a broader user group?
Facing a new issue today regarding the IP Information Tool where access to some of the non-admin information does not show up in the Special:Contributions page for an IP as it did before. That page now only shows "Version", "Active blocks", and "Contributions". Oddly, the country (not city) location still pops up on the watchlist preview, even though it does not show on the Contributions page. Anyone else seeing similar? Best, CMD (talk) 08:00, 15 May 2024 (UTC)
The IP information drop down has been returning no information for me other than the version (IPv4 vs. IPv6), local block info and contribs for the last two days. It's like it can't access any of the data from whichever database it draws from. Every other field simply states "not available".-- Ponyobons mots21:43, 15 May 2024 (UTC)
And regarding my odd note on watchlist preview, this only happens sometimes, with even different edits from the same IP showing me country location in one instance but not another. CMD (talk) 01:50, 16 May 2024 (UTC)
Further learning, I can refresh the same IP contributions page repeatedly and sometimes it will show me the country, sometimes it will show no access. CMD (talk) 03:25, 16 May 2024 (UTC)
Thanks, I'm able to reproduce. According to https://phabricator.wikimedia.org/T363118#9804312, "not available" is case #2. This means Spur/Maxmind doesn't have that data. Looking at these fields, they're all populated by Spur. Maxmind and Spur have different coverage, so we may have a location for an IP but no other data for it.. So that one is not a bug and they have no plans to patch it, it looks like. –Novem Linguae (talk) 15:56, 16 May 2024 (UTC)
I don't understand how that data was available consistently for all IPs, then suddenly is missing for nearly all of them. It renders the entire IP contribs drop down useless. Oh well.-- Ponyobons mots16:13, 16 May 2024 (UTC)
Someone just commented in the ticket that they think it's happening way more than it used to and they think something is broken. So it may be a bug after all. Here's hoping the devs figure it out. –Novem Linguae (talk) 16:47, 16 May 2024 (UTC)
If I refresh sometimes, sometimes, the location info magically just appears. So the data is there. Sometimes. This is all on the same IP page. The one listed just above if I refresh does show the location 1 time out of 4 or so. Canterbury Tailtalk15:58, 21 May 2024 (UTC)
My script (I use Vector 2022) stopped working this week
Link classifier missing in Vector 2022
The Link Classifier which if I recall is maintained by Anomie has disappeared from my Vector (2022) skin. I use this frequently and have just now noticed it, so I is likely something very recent. I switched back to Vector (2010) just to check and it was still available there. Any suggestions about how to get this back in Vector (2022)? older ≠ wiser21:29, 20 May 2024 (UTC)
WMF has recently changed the software so that in Vector 2022 personal Javascript/CSS are loaded only from Special:MyPage/vector-2022.js and Special:MyPage/vector-2022.css. You are loading it only in Special:MyPage/vector.js. You will need to copy whatever you want from there to the vector-2022 version, or to your Common.js at Special:MyPage/common.js (which is the best place for it - scripts are/should be responsible for loading where they should; styles may vary more so they may not be as obvious to move). Remove it from vector.js if you add it to common.js. Izno (talk) 22:06, 20 May 2024 (UTC)
Not sure whether it is related but I am using one of the user scripts (do not have time now to search which one) which, in particular, highlights in pink pages proposed for deletion. The highlighting is gone today (checked on two different devices, different browsers), which gives me a HUGE headache for CFD handling. Ymblanter (talk) 08:35, 21 May 2024 (UTC)
Prompt for edit summary on rollback from watchlist?
Hi there, long time fan, first time caller...
I very much like having a link on my watchlist page that will allow me to perform rollback while prompting me for a custom edit summary. My JS pages were a bit of a mess, but I believe I'd been using User:Nageh/rollbackSum.js for this purpose.
Earlier this week the 'sum' links provided by that script disappeared for me. Nageh (talk·contribs) hasn't edited since 2014 and, while I tried a few other scripts that I found, none of them seemed to re-add such a link to my watchlist either. Has something changed that makes this no longer possible? The ability to add a custom edit summary when doing rollback from my watchlist is very valuable to me, so it would be a shame if the upshot was that this simply is no longer possible. Thanks for your help! DonIago (talk) 03:43, 24 May 2024 (UTC)
Thanks Izno. I've tried copying a version of rollback.js to my common.js page with no evident change at this time. DonIago (talk) 16:12, 24 May 2024 (UTC)
The line above it has no ; and misspells importScript with a lowercase s. This may or may not cause issues. Izno (talk) 16:32, 24 May 2024 (UTC)
Thanks Izno! I'm not sure which of the changes I made per your notes fixed it, but it seems to be fixed! DonIago (talk) 16:55, 24 May 2024 (UTC)
Issue with template
Hi there. There seems to be some kind of issue with {{Population WD}}. I would like to use it to get the bare population figure. At first, {{population WD|qid=Q559227|show=value}} was not working, and I was getting the figure plus a "<span" at the end. Primefac has solved it, at least partially. But there still seem to be some issues, as can be seen here. Thanks in advance for the help, Alavense (talk) 08:19, 22 May 2024 (UTC)
My guess is that the {{first word}} is cutting off the coding to show the "edit at Wikidata" icon, but I have a) no idea why it's doing that, and b) no idea how to fix it. Primefac (talk) 08:37, 22 May 2024 (UTC)
And, yes, nbsp is not considered one of the ASCII whitespace characters by some regular expression engines, including, according to what we've seen here, Lua's %s. — jmcgnh(talk)(contribs)09:19, 22 May 2024 (UTC)
There should not be any regexes that treat a nbsp as whitespace. For a start, its value is outside the 0x00-0x7F range; second, whilst it's encoded as 0xA0 under Unicode, some character sets use the same value for a different, non-blank, printing character - for example, in Code page 437, it's used for the "á" character. Third, ASCII whitespace is defined as characters 0x09 through 0x0c inclusive plus 0x20. --Redrose64 🌹 (talk) 19:40, 22 May 2024 (UTC)
@Redrose64 That's a very outdated point of view. Regex engines these days generally operate on Unicode and understand Unicode characters, or at least have an option to do so, and "whitespace" often means all Unicode whitespace, not ASCII whitespace. Matma Rextalk21:16, 22 May 2024 (UTC)
The HTML spec is very careful to call it out as ASCII whitespace every time that term is used, precisely because just saying whitespace often means Unicode whitespace. Matma Rextalk15:40, 23 May 2024 (UTC)
@Jmcgnh Lua's %s should actually match non-breaking spaces, if I understand the documentation correctly.
I think the problem in the original code is that in [ %s] the entity isn't interpreted, so it finds the first occurrence of &, n, b, etc. It will probably work if you do just [%s]. Matma Rextalk21:14, 22 May 2024 (UTC)
Oh, good, we're getting some people involved who know more than me. @Redros64, Matma Rex, and Izno: As I read it, {{First word}} already supplies %s as the default separator AND the results show that it is not treating nbsp as a member of %s. If I were going to spend more time on this, I'd first get rid of {{First word}} to see how many results are coming back - as an experiment. If there are multiple results (not just one string with multiple values in it), I'd argue that the job of selecting what to return needs to either be pushed down into WikidataIB or maybe do like the show=year branch and use Ustring|gsub to pull out the desired part of the data.
Primefac's current fix provides the correct defaults, but if someone wants to specify show=value|noicon=FALSE it would be best if the result didn't contain stray markup tokens. I checked for how show=year|noicon=FALSE would behave and it seems to get it right for Alavense's test case. I wish I understood a little better how that Ustring|gsub expression matching worked. As I look at it, it shouldn't be working for this case. If I could figure that out better, I'd be in favor of getting rid of {{first word}} and using a Ustring call for show=value as well as for <show=year>. — jmcgnh(talk)(contribs)02:34, 23 May 2024 (UTC)
I don't know much, I just looked up the docs :)
As I read it, {{First word}} already supplies %s as the default separator AND the results show that it is not treating nbsp as a member of %s.
I had a look at the output of {{#invoke:WikidataIB|...}} in Special:ExpandTemplates, and I can at least explain that – it's because the WikidataIB module isn't emitting an actual nbsp character, but rather it's emitting the HTML entity. The string matching functions just see that as a sequence of &, n, b, and so on. So you would need to make {{First word}} look for that sequence, instead of individual characters, but it has no way to do that.
You could probably implement this by invoking Module:String directly – but at this point I would take a step back, and see if there's some way to just output the bare population figure from Wikidata directly, instead of using a big template and applying another big template on top of it to discard most of its results, because that how you end up with pages that bump into parser limits. I'm not very familiar with this stuff, but there must surely be one. Matma Rextalk15:31, 23 May 2024 (UTC)
@Matma Rex: I agree with your suggested direction and thanks for being so clear about the confusion I was exhibiting between Ustring characters and HTML character entities. On a cursory look at available templates or modules to use instead of WikidataIB, I was not coming up with anything obvious. I will return to study this in greater detail, including trying to understand all the parameters passed to WikidataIB, as time allows. I do conclude that {{First word}} is simply the wrong template to use in this situation. — jmcgnh(talk)(contribs)20:52, 24 May 2024 (UTC)
Differentiation of subheading font sizes (or more accurately, lack thereof)
I've just used subheadings for the first time, and was somewhat dismayed to see hardly any noticeable differentiation between Subheadings 2, 3, and 4. Couldn't the subheadings be differentiated more?
Yes, I think many of us got confused at first and pointed out that h3 looks more prominent than h2, but there doesn't seem to be a consensus to change the styles and eventually we got used to it. I hope it doesn't lead too many readers astray. Certes (talk) 12:04, 22 May 2024 (UTC)
How could there be any editorial disagreement about the need for a technical fix to remedy this situation, @Certes? Of course it could lead readers astray!
It's not like the discussion going on elsewhere at the Village Pump on the topic of COI guidelines, with many senior editors all over the field as to just what COI is and what should be required of editors in self-reporting it. Augnablik (talk) 12:42, 22 May 2024 (UTC)
@Certes, I hope it didn’t seem to you in my reply to you above that I was unhappy with you personally. I was just reacting in surprise — okay, shock — that Wiki editors who deal with this sort of thing wouldn’t have very easily come to consensus about the need to make h2 and h3 look different from each other … for the sake of readers. Augnablik (talk) 16:27, 22 May 2024 (UTC)
I wasn't at all insulted or offended. I agree with you that h2 should be more prominent relative to h3. I think the problem is that h3 is bolder than h2, which can give the illusion that it is more important despite not being larger. Of course, it's easy for the regulars to put together a bit of CSS to make it look how we prefer it, but we're a tiny minority of the audience: we're both thinking of the casual reader who can't be expected to jump through that hoop for each site they visit. Certes (talk) 19:58, 22 May 2024 (UTC)
I have a sort of "hello, world" MediaWiki coding change I'd like to submit as my first-ever contribution to MediaWiki, and to get myself started and oriented with the system for submitting coding changes. If all goes well with that, I hope to follow that up with a more substantial change sometime, hopefully soon, after that. I already have a Wikimedia developer account, with accounts on MediaWiki and Wikitech. My usernames there, including my SSH access (shell) username, are the same as my English Wikipedia username. Now mw:Gerrit/Tutorial#Configure Git is telling me I need to have my "own Gerrit username". Is this a name which is unique to Gerrit, and not used anywhere else, such as the Toolforge? Also, I see on the Gerrit settings page a "Username" (is that the same as Git's "own Gerrit username"?), "Full name", and "Display name" – how are each of these used? Which of these names are used for the CREDITS page, the list that's updated by updateCredits.php? wbm1058 (talk) 20:35, 17 May 2024 (UTC)
Your Gerrit username is more properly your developer account username, which is the same as on Toolforge and other places. For those three fields, username is what you log in as, I don't think Full name is used anywhere, and display name is how your name appears on the Gerrit UI. updateCredits.php seems to parse "git log", so the name used is whatever shows up in Git, which usually is the same as one of the above but not necessarily. * Pppery *it has begun...20:52, 17 May 2024 (UTC)
Full name is actually what's used for sign-ins via web UIs (Wikitech, Toolsadmin). "Username" is only used for SSH access (toolforge / git review). The CREDITS page uses the name from git log, which you'd set through the git config --global user.name command. – SD0001 (talk) 21:22, 17 May 2024 (UTC)
Thanks. Further complicating naming matters, I see there is an "LDAP" (Lightweight Directory Access Protocol) username. See wikitech:SRE/LDAP. Per wikitech:SRE/LDAP/Renaming users, "We do not rename users (Developer accounts) anymore. It can (and has) lead to various problems and errors all over the many separate systems which consume Developer accounts as their local databases and authentication methods will get out of sync." So I guess I'm stuck for now with the name I have (not that I want to change it). But a reason for proceeding cautiously here. I don't want to stumble into doing something irreversible that I wish I'd done differently later, after I figured out what I was actually doing, rather than signing up for it by trial and error. I don't recall seeing the mw:Developer account page before, and I think I created mine before the Create a Wikimedia developer account form was created. Today I just ran into the Bitu Identity Manager, which shows me "My LDAP properties". (see wikitech:IDM). Phabricator says my LDAP User is "Unknown". I don't know if there's a way I can personally make it known, or whether it being unknown is a problem. – wbm1058 (talk) 22:14, 17 May 2024 (UTC)
I think theres basically 2 logins. #1 is the oauth / centralauth / all wikis but wikitech one. #2 is ldap / gerrit / toolforge / wikitech. Lots of synonyms here. I forget if phabricator is one of those two, or a third one. –Novem Linguae (talk) 22:26, 17 May 2024 (UTC)
I suppose Phabricator is probably bilingual. Obviously I sign into it using my #1 because my #2 is unknown to the phabulous Phabricator. On the other hand, there must be some developers using it who may not have a #1. – wbm1058 (talk) 22:43, 17 May 2024 (UTC)
Yes, phab supports sign-in via either of the two. You can still link phab with the other account through Settings > External accounts. – SD0001 (talk) 06:21, 18 May 2024 (UTC)
Facepalm I looked at that screen twice and all I saw was date & time settings! Thanks! Now my account is linked with all (two) available providers. – wbm1058 (talk) 10:45, 18 May 2024 (UTC)
When asked why he called the new software, 'git', British slang meaning 'a rotten person', Torvalds said 'I'm an egotistical bastard, so I name all my projects after myself. First Linux, now git.' Ha! wbm1058 (talk) 12:00, 18 May 2024 (UTC)
More from the "I figured out what was happening only after it already happened" department. Wikimedia Code Reviewhttps://gerrit.wikimedia.org/r/settings says I registered @ Monday, May 13, 2024, 9:08:55 PM UTC-04:00 ... what? I don't recall doing anything specific to "register" there last Monday. What was I doing at that time? I thought per mw:Gerrit/Tutorial I had to configure Git in order to register for Gerrit, but here I am already registered for Gerrit, and I haven't configured Git yet! O I C. I think I was looking at a previous code review related to the task I'd decided to work on, when I noticed "Sign up" and "Sign in" links on the upper right corner of that page. Clicking "Sign up" took me to this new IDM "Create account" page to create a Wikimedia developer account. Hey, I thought, I think I already have one of those that I needed for Wikitech/Toolforge. So I left that page, and clicked "Sign in". Voila, my Wikitech password got me in. I thought I had simply logged into Gerrit, not registered for it. What I didn't realize was that the "Bitu Identity Manager" would not only sign me in, but register me as well! wbm1058 (talk) 16:12, 18 May 2024 (UTC)
SSH keys
I already have SSH keys set up for Toolforge at Toolsadmin which I use on PuTTY and WinSCP but not directly from the Windows command prompt.
mw:SSH keys seems to indicate that I can't use my Toolsadmin SSH but will need another one, set up from the Windows command prompt. Correct?
Also, regarding configuring Git personal information. The guide says "You should have to do this only once." Is that literally true, or does it mean once on my desktop and once on my laptop, if I have two machines that I might want to submit code from? wbm1058 (talk) 18:06, 18 May 2024 (UTC)
@Wbm1058 You can reuse the same SSH key across multiple projects (in this case toolsadmin and Gerrit). The tutorial assumes that you have not setup the keys before.
My desktop is still running Windows 7. I know, I know, long in the tooth, but I'm proud to have kept it going for 14 years and would like to make it to 15. It still works for me, for the most part. I've downloaded the production version MediaWiki 1.41.1 and have it running for debugging. I generated my SSH keys with PuTTYgen, since the Windows 7 command prompt does not support the ssh command. I suppose I'd need to use PuTTY on that machine to connect to Gerrit, as I use it to connect to the Toolforge bastion. I think I can figure that out; haven't found documentation on how to use Gerrit on a Windows 7 machine. I haven't set up SSH on my Windows 10 laptop yet (only do Toolforge from my desktop). I don't know how to copy my keys from PuTTY to the required location on Windows 10. Might be easier to generate new keys on Windows 10. I have an ssh-rsa key for Toolforge access; the documentation says to use the ed25519 type for optimum security and performance. Can I use different SSH keys on each machine? wbm1058 (talk) 16:53, 19 May 2024 (UTC)
As a matter of fact, you are kind of expected to use different keys per user per machine. That’s why all the ssh settings of toolforge and Gerrit allow you to add multiple public keys. —TheDJ (talk • contribs) 16:58, 19 May 2024 (UTC)
What TheDJ said, you are expected different SSH keys across machines. However, if you are using one machine, you can reuse the key across multiple things (I have mine on Github/Gitlab/Toolforge and Gerrit as well as a bunch of private services). Sohom (talk) 18:15, 19 May 2024 (UTC)
OK, thanks! New state-of-the-art ed25519 keys generated and installed for my Windows 10 laptop (which MSFT tells me will be unsupported after next year, and my hardware is too old to run Windows 11(.
"This clones the entire MediaWiki core repository, synced to the master branch, into a sub-directory named mediawiki"
I previously installed MediaWiki 1.40.1 on my laptop last November at the Toronto wikiconference, by downloading the then-current version from mw:Download, and successfully installed that, for testing.
I want to overwrite my previous 1.40.1 installation with the new files I just git got.
The standard mediawiki directory holds core and data sub-directories.
It doesn't appear that the git download includes any data. It appears to be a core directory, which includes some extra files that aren't part of the mw:Download version. Why don't the instructions say to download to a sub-directory named core rather than a sub-directory named mediawiki?
If using Git, export the files into a clean location, and then copy the old customized files into the new location as described in the previous section.
You will also need to install some external PHP libraries using Composer or a provided collection maintained for the Wikimedia wiki farm. More details on installing and updating external libraries can be found in the Git download documentation
So, for some reason, although I can test using 1.40.1 without having any PHP problems, I'll need to figure out this "Composer" thing in order to do development testing.
I think all mediawiki unit tests are passing on php 8.1. Not sure about php 8.2. May want to switch to 8.1 to prevent hard to diagnose bugs. –Novem Linguae (talk) 00:45, 20 May 2024 (UTC)
mw:Manual:Installation requirements#PHP does not mention "Composer". It says "MediaWiki only requires PHP extensions that are enabled in PHP by default." I assume that's why I didn't get this error when I ran update.php on my desktop, where I'm just running the official MediaWiki releases. "If your hosting provider provides a basic LAMP environment without these, you may need to install or enable these manually."
I don't think submodules is what you need. The php maintenance/run.php update.php script just wants you to composer install in the mediawiki directory, that should download the required directories. Sohom (talk) 14:34, 20 May 2024 (UTC)
c:\php\mediawiki>composer update --no-dev
Composer could not find a composer.json file in c:\php\mediawiki
To initialize a project, please create a composer.json file. See https://getcomposer.org/basic-usage
I'm going to reboot my machine and try again. Composer install warned me that I might have a PATH problem. wbm1058 (talk) 15:56, 20 May 2024 (UTC)
Judging by your previous comments, you're just not in the directory that Composer expects – try going to c:\php\mediawiki\core, where you (and Composer) should find the composer.json file. Matma Rextalk16:09, 20 May 2024 (UTC)
Yes, thanks, that was it. Once again the instructions misled me: "then run composer update --no-dev from your MediaWiki core directory." It's still running. This step takes significant time! wbm1058 (talk) 16:20, 20 May 2024 (UTC)
I think it worked. The console log is long, with a pile of "failed to download" warning messages, but it appears to have successfully worked around all of them. Here's the end of the log, showing just the last 3 of many installations:
- Installing wikimedia/timestamp (v4.1.1): Cloning 138f3099b4 from cache
- Installing wikimedia/xmp-reader (0.9.1): Cloning 8338d67969 from cache
- Installing zordius/lightncandy (v1.2.6): Cloning b451f73e8b from cache
44 package suggestions were added by new dependencies, use `composer suggest` to see details.
Generating optimized autoload files
11 packages you are using are looking for funding.
Use the `composer fund` command to find out more!
> MediaWiki\Composer\ComposerVendorHtaccessCreator::onEvent
No security vulnerability advisories found.
c:\php\mediawiki\core>php maintenance/run.php update.php
Error: The MinervaNeue skin cannot be loaded. Check that all of its files are installed properly.
PHP Fatal error: Error Loading extension. Unable to open file C:\php\mediawiki\core/skins/MinervaNeue/skin.json: filemtime(): stat failed for C:\php\mediawiki\core/skins/MinervaNeue/skin.json in C:\php\mediawiki\core\includes\registration\MissingExtensionException.php on line 96
Fatal error: Error Loading extension. Unable to open file C:\php\mediawiki\core/skins/MinervaNeue/skin.json: filemtime(): stat failed for C:\php\mediawiki\core/skins/MinervaNeue/skin.json in C:\php\mediawiki\core\includes\registration\MissingExtensionException.php on line 96
Sigh. I didn't need no extensions when testing changes to the official release core page-moving functions. Now I do. – wbm1058 (talk) 18:04, 20 May 2024 (UTC)
You shouldn't need any extensions to run MediaWiki core. It's only trying to load the MinervaNeue skin, because you have a line in your LocalSettings.php like wfLoadSkin('MinervaNeue'); (maybe you copied it from your previous installation?). You can remove it or comment it out if you don't want it.
If you do want it, then you can install the skin using Git similarly to how you installed MediaWiki, just replacing the path in the git clone command: instead of mediawiki/core, use mediawiki/skins/MinervaNeue (and make sure to put it in the skins directory, where MediaWiki is looking for it). Similarly for all other skins and extensions. Matma Rextalk18:13, 20 May 2024 (UTC)
Thanks! I just used the default LocalSettings.php that came with the release-version installation. Figuring I only need one skin, I just copied the folder from my backup of the release version, and commented out the others. That did the trick, and the database update looks like it ran successfully. – wbm1058 (talk) 20:28, 20 May 2024 (UTC)
MediaWiki internal error.
Original exception: [6d2f7f0574eb09bb447c9687] /index.php/Main_Page Error: Class "ResourceLoaderSkinModule" not found
Backtrace:
from C:\php\mediawiki\core\includes\ResourceLoader\ResourceLoader.php(417)
The ResourceLoaderSkinModule class was recently renamed (gerrit:994854). It seems that you've just updated your MediaWiki to a version that doesn't have it any more, but one of your skins is an older version that is still using it. This will occasionally happen with skins and extensions.
If you've already installed the skin from Git, running git pull in the affected skin's repository should fix it. If you haven't, it's probably best if you do :) but you can also remove it from LocalSettings.php, or download the latest master snapshot from SkinDistributor/ExtensionDistributor. Matma Rextalk21:51, 20 May 2024 (UTC)
Done. Success. I have a development environment which seems to be operating identically with the official-release environment I just replaced. Tomorrow I move to the next step. Make minor "hello, world!" type changes in two files, and then figure out how to submit them for review. FYI, the project I'm working on, where I hope to make more substantial enhancements soon, is discussed on my talk: User talk:Wbm1058#My MediaWiki core developers thread. – wbm1058 (talk) 01:40, 21 May 2024 (UTC)
mw:Local development quickstart
It's just that the way you have cloned the repo is unconventional. mediawiki/core should be cloned to a directory named "mediawiki", not "core". I think you will have a lot easier time going through mw:Local development quickstart instead, which unfortunately isn't advertised more prominently. You may want to discard the existing mediawiki install and use the quick start. You have already done step 1, so start with step 2. It's easy! – SD0001 (talk) 16:40, 20 May 2024 (UTC)
That "quick start" page has stuff I've already done previously, and insufficient info about stuff I needed to do.
I've had an "official release" version installed on my laptop since November. It makes a lot of sense to me to have started that way, because as complicated as installing the official release was, installing the developers' version is way more complicated yet.
I've had PHP installed directly on Windows for over a decade now. My bots use that.
There was no need for me to update my php.ini file to load the required PHP extensions. I already did that a long time ago. The problem is that this "composer" system basically "ignores" that, it seems to me.
It just says to "use Git" to clone the core, as if that's easy. No it wasn't. See above for what steps I went through to figure it out.
I installed SQLite already last year; I don't need to do that again.
I already knew how to start my server, though I was told by someone to use "localhost:80" in Boston in 2019, not "localhost:4000". I don't know whether it matters which number I use there.
What I need is a "quickstart" page for upgrading from an "official release" version to the developers' version. – wbm1058 (talk) 17:28, 20 May 2024 (UTC)
1 and 2. yes I said as much - you have already done step 1. 3. There's no need to configure git before cloning repos. The steps from mw:Gerrit/Tutorial#Configure_Git are only to prepare for pushing changes. 4. composer mw-install:sqlite is not for installing sqlite (not required). Instead it initialises MediaWiki itself with sqlite as the db backend. This is a shortcut to avoid going through the web installer. 5. Running something on port 80 is only advisable on linux. On Windows/macOS, it's better to use a non-reserved port (> 1023) to avoid issues with firewalls. What I need is a "quickstart" page for upgrading from an "official release" version to the developers' version. The easiest way to do that is to delete or forget about the release version and setup the developers' version from scratch. – SD0001 (talk) 17:56, 20 May 2024 (UTC)
What I did need to do before cloning was download and install Git, and setup SSH to use it. None of that was necessary to download and install the release version, which makes installing the release version an easier task.
Oh I see. a shortcut to avoid going through the web installer... well, now I know. I was wondering how that was done. But now, water under the bridge. I'd like to keep the database I started last year, for sentimental reasons, and just upgrade. I think upgrading should be easier than installing from scratch every time.
I've noted that my server runs really slowly on my system. Wrote that off as bloatware overwhelming my 14-year old processor, but, now that you mention it, could be a symptom of firewall intervention. I'll switch to port 4000 and see whether that runs faster.
It looks like the directory where git-review got installed is not in your %PATH% environment variable. First of all, try closing the command-line window and opening it again, and retry – maybe it is just seeing outdated env variables.
If that doesn't help, you'll need to change that env variable, the option to do that is somewhere in your operating system settings. Add the directory where git-review is, probably something like like C:/Python310/Scripts/ (that's where it is on my machine). Note that you need to open a new command-line window every time for the changes to take effect when testing.
(If I recall, there's a checkbox to do that automatically when you install Python, which you may have left unchecked. You can also try reinstalling Python and finding that checkbox.) Matma Rextalk15:04, 21 May 2024 (UTC)
Yes, I checked the box on the Python installation. Closing my command window and opening a new one did the trick, of course. The instructions could explicitly say to do that. Thanks again for all your help – wbm1058 (talk) 15:20, 21 May 2024 (UTC)
That directory has a .gitreview file in it, with the following content:
I don't see the "found origin push URL" and "fetching commit hook" that the instructions said I should see. – wbm1058 (talk) 15:47, 21 May 2024 (UTC)
The instructions are probably outdated and everything is probably okay. git-review is being somewhat actively developed, and the example output in mw:Gerrit/Tutorial#Setting_up_git-review has a 2019 date. If it didn't work, you'll find out later if you get an error complaining about a missing Change-Id line. Matma Rextalk20:57, 21 May 2024 (UTC)
I don't know whether that's a problem or not, but I proceeded as if it wasn't.
Successfully rebased and updated refs/heads/T12814-hello.
c:\php\mediawiki\core>git review -R
wbm1058@gerrit.wikimedia.org: Permission denied (publickey).
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.
If you get a Permission denied (publickey). fatal: Could not read from remote repository., review the instructions at mw:SSH keys#Add SSH Private key to use with Git to make sure your ssh agent is running and your identity is added. If you close your Git Bash shell, you will be signed out and need to re-follow these instructions each time.
c:\php\mediawiki\core>eval `ssh-agent`
'eval' is not recognized as an internal or external command,
operable program or batch file.
Is there a way to run git bash from the Windows cmd prompt?
Bill@Mobile-laptouch MINGW64 ~
$ ssh-add ~/.ssh/id_ed25519
Enter passphrase for /c/Users/Bill/.ssh/id_ed25519:
Identity added: /c/Users/Bill/.ssh/id_ed25519 (wbm1058-wikipedia@yahoo.com)
Then I went back to Windows command window, tried "review" again, and still got the "permission denied" error. – wbm1058 (talk) 20:28, 21 May 2024 (UTC)
I haven't seen this exact error before, and your setup is rather different from mine, so this is a guess, but: the eval `ssh-agent` command actually works by setting environment variables, which everything else on your system can read to access the SSH agent. Maybe you just need to close and reopen the command prompt window again, so that it sees the new env variables? Matma Rextalk21:02, 21 May 2024 (UTC)
"Is there a way to run git bash from the Windows cmd prompt?" You can run something like "C:\Program Files\Git\usr\bin\bash.exe" from the command prompt, and it will run, but this is unlikely to do anything good. It will probably be confused about text encodings, file paths, and ANSI escape codes. (On Windows 10, they're slightly less incompatible.) Can you say why you want to do it? Matma Rextalk21:15, 21 May 2024 (UTC)
(edit conflict) I just tried doing it from my git bash window instead of command prompt...
Yeah, just doing everything from the Git Bash window is probably the best way to have the least amount of weird issues. Matma Rextalk21:16, 21 May 2024 (UTC)
Matma Rex, responding to your request at code review to edit another file. Per the instructions at mw:Gerrit/Tutorial#Rebasing, I clicked on the REBASE button in Gerrit's web interface, and it responded with a "Confirm rebase" box with radio buttons for "Rebase on top of the master branch" or "Rebase on a specfic change, ref, or commit". I'm unclear on which I should choose, and whether to check the "Allow rebase with conflicts" box. "It's best to make rebase updates a separate patch, so that your code reviewers have an easy time seeing what changes you've made."
"Hard reset and checkout the change with this command: (BEWARE: git review -d performs a hard reset that destroys all local changes. Stash or commit changes first which you wish to preserve!)" – what is meant by "stash" a change? That term only appears that one place on that page; it's not defined there. – wbm1058 (talk) 21:22, 22 May 2024 (UTC)
Don't I want to amend / add a file to my existing commit rather than add a second commit to my existing branch? Branches and commits, I'm still not comfortable with the terminology and how they fit together.
Funny, when I entered 'git' to get the usage and list of "common git commands" it didn't list "stash". – wbm1058 (talk) 00:46, 23 May 2024 (UTC)
For gerrit and git review, yes. For patchset 2 or higher, you'll want to git stage, then git commit --amend, then git review -R –Novem Linguae (talk) 02:13, 23 May 2024 (UTC)
Ack, though I see the term "stage" several times in mw:Gerrit/Tutorial, I don't see any explicit "git stage" command. Oh (looking it up), that's a synonym for "git add".
I think if you do git review without the -R, it will auto rebase. But i was taught not to auto rebase, and just to do it manually when needed. –Novem Linguae (talk) 03:05, 23 May 2024 (UTC)
I think a few things on that tutorial page are outdated – both Gerrit and git-review have had some changes related to rebases since it was written, and they mostly removed the footguns that this page tries to teach you to avoid. I'll have to read it more carefully and update it. Generally, you shouldn't have to think about rebasing, or do anything to avoid it, or click the "Rebase" button, unless you see an error somewhere telling you that there is a merge conflict in your change. Matma Rextalk15:48, 23 May 2024 (UTC)
Thanks, that makes sense. In baseball, a runner taking a big lead off first base sometimes needs to rebase when he sees a pitch conflict (pitch going to first base rather than home plate). Ha! wbm1058 (talk) 17:14, 23 May 2024 (UTC)
I removed all of the bad and outdated advice from that section: [46][47][48][49] and added some good advice instead: [50]. Hope that helps explain it. I concede that it is not entirely unlike baseball. Matma Rextalk23:14, 24 May 2024 (UTC)
"Publish" is a little too quick on the draw
I've been editing the Houseboat article and once again when I tried to review my changes before publishing a few minutes ago, they got published — with no summary, because I wanted to be reminded of the things I'd done in my edits. This has happened a few times before as well.
And because these edits were major changes, including the addition of new information along with a supporting citation and the removal of some existing text for greater clarity, I'll probably get a finger wag from a senior editor about this. 🫤
Aha. Thanks ... but I wonder why we have to opt in about this, given that we're not supposed to send blank edit summaries? Augnablik (talk) 12:45, 22 May 2024 (UTC)
I recall there being a discussion about this somewhere - if my memory serves me correctly, the idea of not having this on by default is to avoid there being an extra layer of friction that may discourage newer good faith editors.
In any case, there is no formal obligation to provide an edit summary; rather, the edit should be explained in some manner, or be self-explanatory, something which can be achieved after the fact through talk page discussion. WindTempos(talk • contribs)12:59, 22 May 2024 (UTC)
If you have accidentally made major changes without a good edit summary and want to fix that, the standard way is to make a Dummy edit. —Kusma (talk) 13:08, 22 May 2024 (UTC)
There are several keyboard actions that will trigger a save - in no particular order, they include:
Pressing Alt+⇧ Shift+S at any time
Pressing ↵ Enter when the focus is on one of: the edit summary; the "This is a minor edit" or "Watch this page" checkboxes; the Publish changes button
Pressing Space when the focus is on the Publish changes button
In {{Infobox road}} the above value passed on to Infobox image appears to be interpreted as if it had a white backgound, resulting in white corners/edges or a white box around the sign (it's always a white box, but sometimes the visual impact is otherwise). In dark mode this looks pretty bad. However it doesn't happen to next/previous images, or the junction images. See for example Garden_State_Parkway, note the white box around the sign at the top, and the grey around CR 508 (both in dark mode).
I beleive that CSS or {{Infobox}} are creating somthing relating to above that is interfering with the gadget, and providing a light background, where is should be inherited from an outer div.
I'm using chrome on android, and can't see any reason it should believe my system language is Chinese. This appears to be an error with whatever is doing the switching. -- LCU ActivelyDisinterested«@» °∆t°17:50, 25 May 2024 (UTC)
The traditional Chinese file is different, notice 'Wyzwolone Ukrainskie terytoria' that shows in the legend I see but not in the traditional Chinese file. -- LCU ActivelyDisinterested«@» °∆t°18:33, 25 May 2024 (UTC)
Indeed, I should have looked closer before posting. It seems it's not even using this version of the file, because some of it (ex. the exact chinese legend next to the red dot, that is one line long and then a hyphen) does not appear at all in any of the actual languages of the file. — Alien333 (what I did & why I did it wrong) 18:43, 25 May 2024 (UTC)
I'm in the process of revdel'ing personal attacks from edit summaries of 2601:85:C100:2770:51F5:457E:4CD9:606B (talk·contribs) made by undoing an editor's contributions on many pages. I have the mass rollback tool, so that was no problem. But now I have to revdel them one by one since they are on different pages. Is there a way that they can all be lumped together for revdel? – Muboshgu (talk) 19:05, 25 May 2024 (UTC)
Track listing/Professional ratings overlap for music albums
Speaking from the average viewer's POV (i.e., not logged in), I noticed when I visited the album pages for Ohio Players and Follow the Lights that the right side of the Track listing overlaps with the left side of the Professional ratings, with the text visibly overlapping (superimposed) as well. This was observed in Safari on a Mac mini, but not in Firefox. I'm merely reporting this, and not trying to fix it myself. Peterh6658 (talk) 22:49, 25 May 2024 (UTC)
I'm looking to search for article with titles matching '[number][ordinal][space]Road', but for all my trying I can't get it to work. I get intitle and using []{} for a certain number of character but whatever I put together doesn't work. Any help would be appreciated. -- LCU ActivelyDisinterested«@» °∆t°17:40, 25 May 2024 (UTC)
That's autocorrect it does the same for me with hastemplate becoming 'has template'. I'm apparently cursed when it comes to regex, I'd swear I had tried the exact same search. Thanks for the help. -- LCU ActivelyDisinterested«@» °∆t°18:22, 25 May 2024 (UTC)
My brain hurts
I started to answer the question above, but quickly ran into something I don't understand. Regex searches ignore the indexes, so it makes sense to do a composite search which includes a term that hits the index and then a regex to filter that down. But when I try intitle:Road intitle:/[0-9]+/ (link) I get results that include London Inner Ring Road. That doesn't match the regex. H:BOOLEAN says Search terms are implicitly joined by AND but that doesn't seem to be what's happening here. RoySmith(talk)15:20, 26 May 2024 (UTC)
Hello, regarding List of cult films: K, it seems like the listings The Killer and The Killing, both which neighbor listings that have longer titles starting with Killer or Killing, wind up at the bottom of each respective group when sorting alphabetically, when they should come first, having nothing after that keyword. Am I doing something wrong? Is there a way to fix it? See for yourself here. EDIT: I guess I fixed it by adding two spaces after each title, but this feels like too much of a hack. Is there a proper way to do this? Erik (talk | contrib) (ping me)15:58, 23 May 2024 (UTC)
It probably should be using {{sort}} rather than {{sortname}}. The latter makes the sort key "Killer, The" because it is intended to be used with the name of a person. Johnuniq (talk) 00:43, 24 May 2024 (UTC)
@Erik: The problem is that {{sortname}} automatically inserts a comma. {{sortname|The|Killer}} is sorted as "Killer, The". The comma sorts after the space in "Killer Nun" while the title should logically have been sorted before. In [51] you avoided the issue by placing spaces at the end in {{sortname|The|Killer }} so it sorts as "Killer , The". It works in the current implementation where spaces aren't stripped but it's not a pretty solution and other editors can easily remove the spaces without knowing the consequence. You could use optional sort key in {{sortname}} to give "Killer" as the full sort key with no comma or "The". Should {{sortname}} be changed to omit the automatic comma? Probably not. It's mostly used for people and often mixed with manually written sort keys like "Doe, John". Omitting the comma when the template is used would mess that up. PrimeHunter (talk) 20:03, 24 May 2024 (UTC)
Omitting the automatic comma wouldn't even fix this case where "Killer The" would still sort after "Killer Nun". A possibility would be for {{sortname}} to add a special case where {{sortname|The|Anything}} ignores "The" and just sorts as "Anything". "The" is unlikely to be the first name of a person which should be included in the sort key. PrimeHunter (talk) 20:17, 24 May 2024 (UTC)
Thank you! I will use this approach instead. Is there an easy way to identify problems like The King? On my laptop, I was just eyeballing the difference between the order in the code and the sorted order, but The King was too far down for me to see shift in order. Would like to be able to check all the pages and fix the order and sorting where needed. Erik (talk | contrib) (ping me)00:03, 27 May 2024 (UTC)
I haven't done it before but here are three ideas which all require something:
Zoom far out (Ctrl+- in Windows browsers) to view more rows at the same time.
Use the sort arrow on a previewed version where the first rows have been removed or commented out.
Copy-paste the sorted and unsorted rendered table with your browser and make a diff with some tool, e.g. someting at Google:online diff.
The latest run of Special:WantedCategories features a weird cluster of "Decade in nothing" redlinks that are being autogenerated by {{Sport clubs (dis)established in the YYY0s category header}} because the template appears to be failing to import a variable. The template itself does not appear to have been recently edited at all, so it's very likely a module doing this, but as I don't have module-editing privileges I wouldn't be able to fix it myself even if I could find the issue, so the following categories are going to need somebody else to look into them.
@Gonnym, I think I figured out the root cause. In Special:Diff/1225161437, the pipe | was incorrectly left in the usages of {{lcfirstletter}}. That is, {{lcfirst:|{{{sport}}}}} instead of {{lcfirst:{{{sport}}}}}. Feel free to re-implement the switch back to magic words, I don't think this template relies on special features of templates {{ucfirstletter}} and {{lcfirstletter}} for skipping non-letter characters. —andrybak (talk) 14:21, 25 May 2024 (UTC)
Hi. I have been fixing up climbing articles on Wikipedia (e.g. alpine climbing, mixed climbing, and ice climbing). I know that per 'Template:Sidebar', that the sidebar doesn't appear on mobile (which makes sense to me), but is there any way for the main image used in the top section of the sidebar to appear in mobile view (i.e. a qualifier in the sidebar image to say whether it should appear in mobile view)? The images that I have put into the top of the sidebar are good ones and have helpful captions that I think a mobile reader would enjoy? thanks. Aszx5000 (talk) 11:33, 27 May 2024 (UTC)
Do not use sidebars to provide meaningful images, is really the correct response, and in fact at some point they will not be available at all (somewhere in my backlog of work). If you want an image to display, you should use either a thumb image or an infobox, and of course both must fit the requirements for the main image. Izno (talk) 15:47, 27 May 2024 (UTC)
Normally, mobile (source, not visual) diff displays removed text in yellow and added text in blue in the same block of text. In this said diff, the edited paragraph is (confusingly) displayed twice, one copy indicating removed text, the other one added text. Could you reproduce this? Is this a bug or intended behavior? Thank you. Janhrach (talk) 07:52, 26 May 2024 (UTC)
Are you trying to use the mobilefrontend view with vector-2022, leading you to getting the desktop-type vector-2022 output? — xaosfluxTalk12:12, 26 May 2024 (UTC)
@Xaosflux, mobile diffs now use the same machinery as normal diffs. And no, it is not appearing as expected, the paragraph beginning with This battle was fought between various north-western highland clans from the lands of Ross, against the Earl of Ross and his followers. is repeated. Izno (talk) 16:20, 27 May 2024 (UTC)
It's worth saying that this is a hard problem in general. You'd probably need to look at the code to see what's going on, but it's likely that the diff treats headings as special, then trying to identify the two paragraphs as derivatives may not be easy. All the best: RichFarmbrough18:53, 27 May 2024 (UTC).
Why has my map broken?
Hi all
I spent ages building a map to help visualise the countries where WikiGap has been organised (a project to help close the gender gap on Wikipedia). The map was transcluded at the bottom of the main Wikigap page however it seems to be broken, no map is shown, the transcluded page doesn't appear to have been changed so I think it might be a technical change somewhere? If anyone knows how to fix it, please I'd love some help, its way beyond my technical ability.
Thanks as ever PrimeHunter, ah, that's a shame, well I guess its just permanently broken and it will be some time until a replacement is made, I wonder if I can do it through Kartographer or something. John Cummings (talk) 20:57, 27 May 2024 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
Several bugs related to the latest updates to the UploadWizard on Wikimedia Commons have been fixed. For more information, see T365107 and T365119.
In March 2024 a new addPortlet API was added to allow gadgets to create new portlets (menus) in the skin. In certain skins this can be used to create dropdowns. Gadget developers are invited to try it and give feedback.
Some CSS in the Minerva skin has been removed to enable easier community configuration. Interface editors should check the rendering on mobile devices for aspects related to the classes: .collapsible, .multicol, .reflist, .coordinates, .topicon. Further details are available on replacement CSS if it is needed.
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 28 May. It will be on non-Wikipedia wikis and some Wikipedias from 29 May. It will be on all wikis from 30 May (calendar). [54][55]
When you visit a wiki where you don't yet have a local account, local rules such as edit filters can sometimes prevent your account from being created. Starting this week, MediaWiki takes your global rights into account when evaluating whether you can override such local rules. [56]
This started sometime between midnight UTC last night and 10:30 UTC today, Monday 27 May. I use Edge on Win 11, Monobook, Navigation popups enabled, and the Use a black background with green text gadget. When I point my mouse at "diff" on my watchlist now the pop up/preview/whatever you call it is now very much harder to read, the change is "highlighted" by shewing the removed text in black on a grey background then the added text in black on a brown background. Please help! DuncanHill (talk) 11:09, 27 May 2024 (UTC)
@DuncanHill This is an unintended side effect of a change I made so that Navigation popups work in the new Vector dark mode. I've revised that change so it shouldn't affect the black background with green text gadget, will see about getting it merged. the wub"?!"21:33, 27 May 2024 (UTC)
Some VisualEditor features don't work with cell formatting templates which add a pipe as part of the table syntax. The code said | rowspan="3" {{yes}}. VisualEditor thinks the cell content is rowspan="3" {{yes}}, but {{yes}} adds both cell attributes and a pipe which separates them from the actual cell content yes. It produces: style="background:#9EFF9E;vertical-align:middle;text-align:center;" class="table-yes"|Yes. You just have to use the source editor for some things. PrimeHunter (talk) 11:27, 28 May 2024 (UTC)
Relocate mobile/desktop button
I would like to relocate the mobile view/desktop button to the top of the page (preferably, next to talk or in same line with the page title in the far right corner) for ease of access instead of having to scroll to the bottom each time to switch between mobile and desktop mode. Can this be done? Qwerty284651 (talk) 00:18, 28 May 2024 (UTC)
@Qwerty284651: I don't think that it can be relocated without a MediaWiki change, for which see WP:BUGS (it's worth noting that the Cologne Blue desktop skin already has the "Mobile view" link near the upper right, see e.g. VPT under Cologne Blue). But I'm pretty sure that somebody (not myself) can write some Javascript that will create an additional link in the desired place (the actual Javascript will need to be varied according to the skin). Once it's in place, you can then either leave the existing link alone, or hide it with some CSS:
/* (i) hide the "Mobile view" link when in desktop; (ii) hide the "Desktop" link when in mobile view */ul#footer-infoli#footer-places-mobileview,ul#f-listli#mobileview,ul#footer-placesli#footer-places-mobileview,ul#footer-placesli#footer-places-desktop-toggle{display:none;}
This goes in your CSS, as it's skin-universal - the first two selectors are for Modern and MonoBook respectively, the third is for all other current skins, the fourth is for mobile.
Don't activate it until the replacement link has been set up, otherwise you will lose the ability to toggle the views. --Redrose64 🌹 (talk) 09:56, 28 May 2024 (UTC)
@Redrose64, looking at the Cologne Blue skin, I see the whole bottom section (languages, last modified, cats, other links) of the page was moved to the top.
Where can I find the CSS and JS pages for the Cologne Blue skin or any skin currently available, for that matter?
Same here. I only use the mobile version for editing tables in VE, although it can be finicky when switching between selecting cells and editing them. Qwerty284651 (talk) 14:20, 28 May 2024 (UTC)
Mobile view Swiss flag
There is a bit of an issue on mobile view in when this template: "Switzerland" (or its variants) is used, the flag will appear stretched horizontally. Also, it’s to be noted this template: "" does not have this issue. —TwinBoo (talk) 11:35, 28 May 2024 (UTC)
The minimum width of a flag on mobile is 23px, because without a minimum width, flag icons risk going to 0x0 inside tables and other common elements when on mobile, due to lack of available width. —TheDJ (talk • contribs) 05:45, 29 May 2024 (UTC)
But can we make it keep the correct width-height ratio in all circumstances? I added a table row only saying 23px instead of 23×16px, but othwerwise identical to the distorted first row. This displays correctly for me as 23×23px in mobile. It apparently takes three things simultaneously to be distorted in mobile: flagicon, wikitext specifying both height and width, and an empty link= which omits a link on the image. PrimeHunter (talk) 11:49, 29 May 2024 (UTC)
Issue with cast template visuals, but only on Chrome
Was editing the page for Ezra (2023) and noticed that on Google Chrome, there is a huge issue with spacing. Wish I could show screenshots to detail what I see. But underneath 'Cast', I see a large, multi-paragraph jump of just blank, white space. It nearly takes up the whole screen. I do not see this on mobile, nor do I see it on Microsoft Edge -- just Google Chrome. Unsure how to fix this as I'm newer to Wiki. Is anyone seeing the same thing as me, and do they potentially know how to fix? Squiddyonwiki (talk) 02:02, 29 May 2024 (UTC)
I see that {{cast listing}} sets the style declaration min-width:min(25em,100%) - is Chrome interpreting that as expected? First off, the min(...) function is in CSS Values and Units Module Level 4, which is still at the W3C Working Draft stage - not all browser vendors will impement it yet. Second, the spec says
For all three functions, the argument calculations can resolve to any <number>, <dimension>, or <percentage>, but must have a consistent type or else the function is invalid; the result’s type will be the consistent type.
min() is implemented by most of the browsers worth discussing. I think your observation about types is perhaps interesting. However, you are allowed to mix units in these min/max/clamp functions (see examples earlier in the page), and percentage is in a sense a unit...? I also know that CSS columns have a complex formula for how wide they are laid out and the 100% may be interfering with that. And that Chrome only recently (a couple years ago) made their CSS columns implementation much closer to fully specification compliant. Izno (talk) 15:18, 29 May 2024 (UTC)
That investigation just above aside, there is a possibility this was caused by a change that WMF made because they're thinking about how to make tables display appropriately when the width of the page is too narrow, based on your description of what you saw. The task for that is phab:T366119 and an initial bug report about suboptimal display was phab:T366119, after which there was some adjustment to how it functions - namely restricting this functionality to wikitable class tables. (There is still some things being learned there it looks like - though Isarra sorted most of this a while ago in Timeless....) Izno (talk) 15:33, 29 May 2024 (UTC)
That sounds like what I was seeing. What's weird, though, is that I'm on a Macbook now, still Chrome, and it looks normal. But on a Dell the display is what's bugged. Maybe it's based on the size of the laptop's screen? But unsure, because on my Dell on Microsoft Edge it was fine. Either way, I've never had to do a bug report but I will try to figure that out. Squiddyonwiki (talk) 15:38, 29 May 2024 (UTC)
OpenStreetMaps in Croatian? Is it just me?
The OpenStreetMap maps in the infoboxes at Radio City Music Hall and Carnegie Hall (and probably other locations) seem to be in Serbo-Croatian; at least all of the place names are spelled phonetically according to Serbo-Croatian rules. Is this true for everyone, or is there some weird setting in my Preferences I need to change? If it's true for everyone, can it be fixed? Surely OpenStreetMap has an English-language version of their map of Manhattan. —Mahāgaja · talk11:28, 30 May 2024 (UTC)
I see this, too. If I click through to get a large map, it affects all of Manhattan but not the other boroughs, New Jersey, or (as far as I can tell) any other place. rbrwr±12:41, 30 May 2024 (UTC)
Since yesterday, page previews of links haven't worked for me. My "Enable page previews" option is turned on. Can anyone help? Thanks! LoganP25 (talk) 20:31, 29 May 2024 (UTC)
There's nothing to the right of "Mobile view" for me. Previews work when I'm logged out, but are broken again when I log back in. Previews don't work at the link you sent, and still don't after turning them off and on again. Navigation popups is disabled. Quite the enigma lol. LoganP25 (talk) 22:21, 29 May 2024 (UTC)
My skin is Vector legacy, and I'm browsing on Google Chrome. Previews do work in other wikis, I hadn't thought to check that, but not in the English wikipedia. LoganP25 (talk) 22:47, 29 May 2024 (UTC)
@LoganP25 Hello. I also had the same problem here, but fortunately I made it! So I would like to share my solution to you as well.
The replies that you made to the questions of @PrimeHunter about having the link of "Edit preview settings" at the bottom of pages, the matter of "whether or not login/logout has an affect on page preview", "Navigation popups" options disabled or not along with all the other factors you mentioned matches exactly with mine. But I solved it in a different way.
Firstly, go to the "Appearence" in your "Preferences" & check the box of the "Enable page previews" (if its already checked then follow the next steps). Now, go to the "User profile", click on "Set your global preferences". On the "Appearance", there you should enable the (another) page preview option (both of the boxes should be checked). Save it & again return to your (original) "Appearance" in your "Preferences" just to make sure that, two things are there — the "Enable page previews" is not only checked but also grayed out & the "Set a local exception for this global preference" is unchecked.
Glad you finally worked it out. It sounds like a bug if you have to set a global preference to make a local preference work. I haven't heard of that before and didn't expect it. PrimeHunter (talk) 14:15, 30 May 2024 (UTC)
Production table formatting for correct text alignment
An experienced editor informed me that a table format - used over years with no negative feedback on numerous articles to display crop production data - "crashes" the text aligning to it in the right text margin. The editor says it works fine in Chrome, but doesn't on any browsers tried on a Windows PC.
The original table format parameters are
{|class="wikitable"style="float:right; clear:left; width:14em; text-align:center;"|-!colspan=2|Mango* production – 2022|-!style="background:#ddf; width:75%;"| Country!style="background:#ddf; width:25%;"| <small>millions of [[tonne]]s</small>|-
The editor made a change based on a "wikitable floatright" parameter plus internal table formatting (column alignment left or right) using:
{|class="wikitable floatright"style="width:15em; margin: 20px 20px 0px 20px"|-!colspan=2|Mango* production – 2022|-!style="background:#ddf;"| Country!style="background:#ddf;"| <small>millions of [[tonne]]s</small>|-
@Zefr: The floatright class, when used on a table, is equivalent to setting the float:right;clear:right;margin:000.5em0.5em; styles. Notice two things: (i) the clear is for right not for left; (ii) it doesn't affect text alignment within the cells. You shouldn't need to explicitly set margin values if you are also using the floatright class. For an axample of floatright without other styles, see the "Distribution of locomotives, March 1974" table in British Rail Class 24#Operation. --Redrose64 🌹 (talk) 16:52, 30 May 2024 (UTC)
Replag
Another lengthy replag (~3 hours) on a Thursday night. Is there any place to go to find out when the estimated fix-it time will occur? Or is it just wait and see? LizRead!Talk!02:40, 31 May 2024 (UTC)
Seems to have fixed itself. The highest I've seen it go was around 11 or 12 hours, so this one was not nearly as bad as it's been in the past. --rchard2scout (talk) 13:44, 31 May 2024 (UTC)
The first one of these has been launched. Please @ping me if there are any reported issues. Follow up on the general concept of template gadgets is welcome at WP:IANB. — xaosfluxTalk15:29, 31 May 2024 (UTC)
Font size bug - mobile
Hello, I have question. For quite some time now, The font size is not correctly shown when I want to see it on visual editing (i mostly work with filmografies of actors) - basically, I want to change font size from 100 to 80, and its not working at all. It stays same. The whole filmography also gets bigger sometimes for some reason when I want to edit it only by visual editing. I remember there was a discussion about this long time ago but I dont know if there was any improvement. I work on mobile, so I dont know whether pc users have this problem as well. If there is any sollution or way how to make it work, feel free to comment your suggestions. Thx in advance. Kesseder (talk) 17:55, 31 May 2024 (UTC)
I'm running into the recurrence of a problem I've brought here a few times before, where an incategory search of Category:Living people for drafts or user sandbox pages that shouldn't be in it offers up several pages that aren't actually pages in the category but cross-namespace redirects to real articles, which resulted from an editor adding categories to the page while it was still in draft or user space, and then moving the page into mainspace shortly afterward. That's a relatively common occurrence which doesn't usually cause any serious problems, but every once in a while it causes situations like this, where for some reason the redirect lingers in the categories despite not actually having any categories on it.
In the past I've usually resolved this by redoing the move: I would move the page back overtop the redirect, wrap the categories in the {{draft categories}} wrapper to disable them, and then move the page back into articlespace again, which usually worked to get the redirect out of the category — but of the four such redirects currently showing up in the search, I've already done that on two of them, meaning even that trick isn't working anymore.
Additionally, I need to note that there are four phantom redirects currently showing up in the search, yet the results counter at the top of the page is claiming that there are eight pages, meaning that there are four more phantoms I can't even find.
Because Category:Living people is a megacategory with millions of pages in it, it's not a category I can search manually for draft or user pages — it has to be done as an incategory search, which means I can't just work around permanent speed bumps and need the tool to be clean. It might also be related to the replag problem that's been discussed above, because even the normal categorized drafts or userpages that I pull out of the category the normal way are taking longer than usual to clear from the search as well — but could somebody look into whether there's any way to fix it? Thanks. Bearcat (talk) 17:20, 31 May 2024 (UTC)
Replag is currently around five seconds, so that's not it. It wouldn't affect searches on the live site anyway.If I query the database for members of Category:Living people that are in namespaces user, draft, or their corresponding talk, I get no results. All four of the current false positives in search (Draft:Amre Hamcho, User:Miminity/Yoshinobu Yamakawa, User:Minmarion/sandbox/voice, User:WikiOriginal-9/David Hinds (gridiron football)) were moved within the last 24 hours. And neither a null edit nor an actual edit (both to User:Miminity/Yoshinobu Yamakawa) helped. Taken together, these lead me to guess - with absolutely no knowledge whatsoever of how the search backend works, mind you - that it's only the search indices that are out of date.I suppose it's worth checking back in a day or so, without your move/edit/move workaround, to see if they clear up themselves. Failing that, I can build a database report for you, which might be more convenient than the search. (That is, unless you're worried about search results, too, not just what appears on the category page.) Any other categories to check besides Category:Living people? —Cryptic18:39, 31 May 2024 (UTC)
I just got a new Windows 11 laptop. When I start using Wikipedia, it shrinks the display area (but not the very top menu bar} to about 2/3 of the available width, then below the normal desktop view it shows a sort-of mobile view of the same page. See screenshot to the right. Wikipedia still looks fine on the old laptop.
Is this a bug or some obscure configuration problem? Aymatth2 (talk) 13:46, 1 June 2024 (UTC)
Thanks. I thought it was something dumb like that. Taking out the old script fixes the problem. Both laptops are fairly up to date Windows 11, but there must be some difference in their configuration. Aymatth2 (talk) 15:03, 1 June 2024 (UTC)
Broken parsing/HTML generation near math tags
Look at the vertical spacing here. The source code has a single blank line to separate the paragraphs, as usual, but the result is terrible. Moreover, examining the HTML DOM in both Firefox and Chrome reveals that the end on the first paragraph ("In fact...") is actually outside any paragraph tags, but there's an empty <p></p> between "...Killing form)." and "This cross-product..." (which apparently creates that huge gap). A similar situation can be seen in the next section as well (before "Alternatively, one may...").
I've noticed that vertical spacing around <math> some time ago became noticeable broken in some cases (seemingly regardless of <math display="block"> or : <math> and blank lines before/after). Initially I was expecting that somebody else will notice that quickly and repair, then I forgot and was lazy to investigate, but this outstandingly wrong occurrence made me think that it must be at least reported... — Mikhail Ryazanov (talk) 09:41, 2 June 2024 (UTC)
Now it seems to me that even though the "legacy" skin didn't have so obvious problems, they were still present, but partially hidden because paragraphs didn't have bottom padding added (for what reason?) in the new skin. Please also take a look at some tests with comments that I've made in my sandbox:
Examining the DOM again reveals that the generated HTML is not correct (or recent versions of both Firefox and Chrome can't parse it properly, which is unlikely). — Mikhail Ryazanov (talk) 01:11, 3 June 2024 (UTC)
Thanks for the screenshot. So "the result is terrible" means a few mm extra whitespace. Your link [57] uses <math display="block">. We add whitespace to that with code in MediaWiki:Common.css:
/* Make <math display="block"> be left aligned with one space indent for * compatibility with style conventions */.mwe-math-fallback-image-display,.mwe-math-mathml-display{margin-left:1.6em!important;margin-top:0.6em;margin-bottom:0.6em;}
Paragraph spacing is the subject of much recent discussion in phab:T362939 after the WMF changed paragraphs to use padding rather than margin. I would guess this is the issue you have identified. They intend to fix their previous decision making. (Apparently a math-specific one is phab:T361273 but that probably should be closed duplicate.)
OK, phab:T182041 is apparently the root cause (made more obvious by phab:T362939 changes), reported many years ago, but nobody really cares... Would it be too difficult to use <span> instead of <div> to make it work properly? — Mikhail Ryazanov (talk) 04:07, 3 June 2024 (UTC)
Recent change somewhere in Wikipedia/Wikimedia has broken chart labels in Module:Chart
Something changed recently and now all charts generated by Module:Chart have broken and the top text label no longer displays properly. As you can even see on the help page for the charts, the top label of the chart has half the text chopped in half. I can't track down why this would have happened. Does anyone know where this could come from? It didn't use to do this a few weeks ago and nothing has changed on the page itself so it must be inherited from elsewhere on wikipedia. I've tested with Firefox, Safari and Chrome and all seem to have this issue so it doesn't appear to be limited to certain browser versions. Looking at archives on archive.org for some pages, it was still working as of the beginning of May but haven't found any page archives within the last few weeks yet. Here's one example of what it should look like. And here's what that section looks like currently. Ergzay (talk) 06:41, 3 June 2024 (UTC)
Through some asking questions I found that apparently a "overflow-x: auto;" media query was added globally across wikipedia, something that is very wrong to do. Anyone know where the code is for this so a bug report can be filed? Ergzay (talk) 07:52, 3 June 2024 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
It is now possible for local administrators to add new links to the bottom of the site Tools menu without JavaScript. Documentation is available. [58]
The message name for the definition of the tracking category of WikiHiero has changed from "MediaWiki:Wikhiero-usage-tracking-category" to "MediaWiki:Wikihiero-usage-tracking-category". [59]
The new version of MediaWiki will be on test wikis and MediaWiki.org from 4 June. It will be on non-Wikipedia wikis and some Wikipedias from 5 June. It will be on all wikis from 6 June (calendar). [61][62]
Future changes
Next week, on wikis with the Vector 2022 skin as the default, logged-out desktop users will be able to choose between different font sizes. The default font size will also be increased for them. This is to make Wikimedia projects easier to read. Learn more.
Are images transfers to WikiData working? (repost from help desk)
I use Wiki Shoot Me to take photos for Wikipedia while I’m traveling. Typically you can identify articles that need photos by looking for yellow dots indicating Wikipedia articles and larger red circles near by indicating WikiData items without photos (as previously these seemed to sync). In two cases recently I noticed articles with CC-licensed images placed correctly in the lead and appropriately sized that did not have their photos synced with WikiData: National Hotel (Q65056276) and Wet Mountain Valley (Q7989973). I know the page image is being picked up because they display correctly on Special:Nearby using their coordinates: National Hotel and Wet Mountain Valley. It looks like they’re just not making it to WikiData. - Scarpy (talk) 05:46, 4 June 2024 (UTC)
I'm not aware of any bot or tool which scrapes Wikipedia articles for suitable images to import into Wikidata. This may be an idea worth exploring, but there would be plenty of false positives. For example, on an article about an artist, if we do not have an image of the artist it is common to include a picture of one of their well known works instead. This would not be suitable for importing to Wikidata, and I can't think of a reliable way for an automatic process to detect these. Perhaps a semi-automated tool (which makes suggestions but requires manual review) is the way to go? I suggest you post at Wikidata:Project chat for a more informed response. — Martin (MSGJ · talk) 07:50, 4 June 2024 (UTC)
See this edit and User talk:Awesometd#nowrap. The nowrap attribute on a td element, already deprecated in HTML 4 (December 1997), was marked as obsolete in HTML 5 (October 2014). The user says that they are copying its use from other pages, so does anybody know where in Wikipedia such usage is recommended or even suggested? --Redrose64 🌹 (talk) 10:30, 28 May 2024 (UTC)
I doubt it's suggested anywhere. A few cases may have been added long ago and some users just copy what they saw in other articles. The user is right that it's used in 2024 F1 season. Unsurprisingly it's also in previous seasons. It's common to start such pages with a copy-paste from another season. PrimeHunter (talk) 11:16, 28 May 2024 (UTC)
There's many more obsolete attributes still being used in tables, such as align or bgcolor. If we truly want to get rid of them, the solution would probably be to extend the Linter extension, so that they'll be listed at Special:LintErrors. That's probably a discussion to be had over at WT:LINT. --rchard2scout (talk) 07:56, 29 May 2024 (UTC)
One could easily ask as well where in Wikipedia such usage is deprecated or discouraged? I'm pretty sure the amount of editors that read the HTML5 instructions prior to editing articles is rather low. You can't just assume that everyone is always aware of what parameters have become obsolete.
I am a regular editor of the Formula 1 Wikiproject and I remember us starting to use that parameter because it is more practical and intuïtive than the nowrap template and takes up less memory. The fact that it never produced any technical issue, nor was there any message that it is obsolete. This is litterally the first time anyone give an indication there is a problem. Why isn't this advertised more to the relevant WikiProjects? Like another person pointed out here, if these things would be flagged as LintErrors they would not be used. But I do wonder why such a simple, well working parameter was made obsolete.
We are well-intentioned people, so I'm sure that if you invite a couple of editors from the relevant WikiProjects, explain the issue and tell us what the correct CSS code is, per the HTML5 documentatien's recommendation, we'll set out to deal with those obsolete parameters. As a side note, I think the Superbike article has even more issues, like the usage of external links.Tvx1 21:53, 3 June 2024 (UTC)
This one is actually pretty easy to switch, we have a CSS class nowrap that you can change whatever templates to use instead. Izno (talk) 23:31, 3 June 2024 (UTC)
This applies the class to one specific cell. It can also be applied to a whole row at once; or to the entire table. Doing those isn't such a good idea, you may cause excessive sideways scrolling. --Redrose64 🌹 (talk) 13:55, 4 June 2024 (UTC)
Yes, it does the same thing, but (i) it's more to type; (ii) less easily remembered; (iii) more easily broken e.g. by omitting or mistyping that hyphen; (iv) more difficult to apply cascading styles to. In general, class= is preferred over style=.
As for "why such a simple, well working parameter was made obsolete", it's part of the overall plan for HTML, going right back to the mid-1990s, that HTML should concern itself only with semantics, and leave styling to style sheets. Accordingly attributes that have no semantic meaning and affect only the style - other than style= itself, were first deprecated and then made obsolete; similarly with elements like <font>...</font> that affect only the style and have no semantic meaning. --Redrose64 🌹 (talk) 05:46, 5 June 2024 (UTC)
@Ianmacm, if you're using different devices, different networks, proxies, VPNs, or any combination of these, then you might be affected by an IP range block. The "thank" links are hidden in such cases. —andrybak (talk) 18:43, 5 June 2024 (UTC)
That's useful. Maybe something along these lines could be added to H:THANK, as I read through this page and could not figure out why the thank link did not show up.--♦IanMacM♦(talk to me)19:22, 5 June 2024 (UTC)
Cite error: Invalid ref tag; no text was provided for refs named
Scrub it, I managed to guess the answer. I have to give the refs a different name in the definition to that used in the text. I expect someone thought that was clever, especially when combined with not mentioning it on the help page. DuncanHill (talk) 10:32, 6 June 2024 (UTC)
The problem is an undocumented feature in {{Excerpt}}. It adds the page name to ref names it's transcluding. Economy of India says <ref name="India_kapur"/> and defines it elsewhere with <ref name="India_kapur">...</ref>. That works fine in the article itself but {{Excerpt|Economy of India}} only transcludes the reuse and changes it to <ref name="Economy of India India_kapur"/>. If you manually copy the definition then you have to change name="India_kapur" to name="Economy of India India_kapur", as you found out. The purpose of the name change is to avoid a clash with an existing reference using the same name in the article calling {{Excerpt}}. VisualEditor often creates the same ref names so it can easily happen for completely unrelated references. It should certainly be documented in the template, and maybe mentioned as a possible cause in Help:Cite errors/Cite error references no text. PrimeHunter (talk) 11:58, 6 June 2024 (UTC)
Thanks. VisualEditor refnames are a menace, when people copy-paste lumps of articles they often produce clashes. Excerpt generates both undefined refname errors and harv/sfn no-target errors. I can see how adding the name can stop clashes, but it needs to be mentioned on the help page for the error messages. DuncanHill ::(talk) 12:05, 6 June 2024 (UTC)
It looks like {{Excerpt}}does try to handle the case where the ref is defined elsewhere in the page. But for some reason it's failing in that particular case. Anomie⚔12:09, 6 June 2024 (UTC)
Looks like the "some reason" is that those four refs are defined inside an infobox parameter. When the reference fix-up runs the infobox is still present, so it sees no need to fix anything. Then, later, the code strips out the infobox and along with it the definitions of those four refs. Anomie⚔12:30, 6 June 2024 (UTC)
Gråbergs Gråa Sång, the user appears to have added a very large number of invisible Unicode characters. Specifically it's the Combining Grapheme Joiner character hundreds of times. Credit to the user script w:de:Benutzer:Schnark/js/diff which shows invisible characters as well as their names. — Qwerfjkltalk14:04, 6 June 2024 (UTC)
In Wikipedia:Templates for discussion/Log/2024 May 22#Template:Edit semi-protected a reason some editors, including myself, are opposed to merging is that the merged template will no longer be able to work properly if used on an unprotected page (which can happen in the case of an unprotected WP:ARBECR page, for example). SilverLocust put it like this: If these were all redirected to one template, then there would be a loss of functionality unless someone knows how to tell a module not merely which wrapper is invoking a module (since there would only be one merged wrapper), but rather which redirect is being used to transclude the wrapper that invokes the module (and I don't think that is possible). So, is that right? Or is there a way to detect which redirect is used and merge the templates without any loss of functionality? Nickps (talk) 14:07, 2 June 2024 (UTC)
@Nickps The module could use getContent() to get the text of the current page and then search it for one of the redirect templates. --Ahecht (TALK PAGE)02:30, 3 June 2024 (UTC)
I would oppose that as confusing and probably inefficient. Editors expect it to make no difference whether a redirect is used. If we really want a certain "redirect" to behave differently then don't make it a redirect but a wrapper which passes a certain parameter. PrimeHunter (talk) 02:47, 3 June 2024 (UTC)
Yes, and that's exactly what has been done (but is proposed to be undone by merging the templates). Certes (talk) 13:56, 3 June 2024 (UTC)
@Certes It's no less efficient than what every single citation template is currently doing (using getContent() to get the text of the current page and then searching the text for date format templates). --Ahecht (TALK PAGE)16:59, 6 June 2024 (UTC)
Some JavaScript script loads, and on e.g. Syncopation, clicking on the JS'd play buttons just load forever now. Disabling JS enables normal playback, and page previews don't have this issue. Aaron Liu (talk) 20:57, 6 June 2024 (UTC)
WikidataInfo
Hi!!!
In it.wiki there's a gadget that show me a link to the wikidata item. Is possible to have it also in other wikis? Is there a way to activate tha same preferences on all wikimedia projects?
Thanks for help Esc0fans -and my 12 points go to...08:08, 7 June 2024 (UTC)
I'm getting this error pretty frequently lately when opening my watchlist. Maybe a third of the time over the past several days. Anyone know what this is? — Rhododendritestalk \\ 01:32, 1 June 2024 (UTC)
Thanks. I do have a very large watchlist (17k pages), but this issue started happening all of a sudden a few days ago and the list has barely changed in recent weeks. I don't think I've ever once seen the error before then. It's also inconsistent. If I just refresh a few times, it'll display. — Rhododendritestalk \\ 13:44, 1 June 2024 (UTC)
But the software and servers change all the time. If you are up to those numbers, even half percent of change in performance on that side can easily push you over an edge. —TheDJ (talk • contribs) 14:07, 1 June 2024 (UTC)
Thanks. Frankly, though, I don't want to change my watchlist. It's huge, yeah, and includes a ton of e.g. deleted pages, but I like being able to see when something is recreated or when someone edits a page from way-back-when that probably shouldn't be edited anymore. Edits to inactive pages sneak through too easily sometimes. I'm just kind of surprised that I've had a massive watchlist for years (I became an active editor playing with counter-vandalism tools and AWB, so built up a huge watchlist early) and it's never caused a problem. Of all the things that use memory on Wikipedia, it's volunteers' watchlists that need to be limited? — Rhododendritestalk \\ 14:29, 3 June 2024 (UTC)
Almost everything is limited. It’s just that most people don't know about that because they hardly ever run into those limits. —TheDJ (talk • contribs) 17:23, 3 June 2024 (UTC)
It's more common on other projects (where RC tables are larger as well though), but I don't think 17K pages should be close to where these problems would start to occur (at least when you only select to display 50 or 100 pages, maybe). 1234qwer1234qwer418:19, 7 June 2024 (UTC)
The likelihood of a timeout can vary depending on how much load the servers are under. More load, more likelihood of a timeout. Watchlists are actually the single most database intensive feature on Wikipedia, probably by a fairly wide margin. On smaller wikis, adjusting the time period to search can help a lot but on a wiki the size of Wikipedia not so much. (Essentially there are two methods of calculating the watchlist depending on if the number of changes in the time period being search is smaller or larger than the amount of entries in your watchlist. On Wikipedia you'd probably have to set that super low before it made a difference because so many edits are happening all the time). Most of the other filters (including total number of results to show) don't make much of a difference most of the time, although there might be edge cases where they matter. Bawolff (talk) 00:04, 4 June 2024 (UTC)
Thanks for this background. Did not know they were the most database intensive feature, but I guess that makes sense. So I take this to mean the only way to fix it is to remove pages from the list? — Rhododendritestalk \\ 00:07, 4 June 2024 (UTC)
Rhododendrites, you could remove some pages from your watchlist and put links to those pages somewhere in your userspace, then use Special:RecentChangesLinked on that page as a kind of pseudo-watchlist. That'd allow you to split your watchlist into smaller chunks for the servers to process, at the cost of part of your watchlist becoming public. Rummskartoffel16:28, 6 June 2024 (UTC)
I get this off-and-on, started about a week ago. No particular time of day afaict. My watchlist has been in the 25-26k range for ages, so it's not down to any change on my part. DuncanHill (talk) 16:47, 6 June 2024 (UTC)
Help needed with collapsible infobox section templates
On the Ryzen article (permalink), I added Template:Collapsed infobox section begin and Template:Collapsed infobox section end to some parts of the Template:Infobox CPU to make it so they are put into collapsed boxes. The cache one came out fine aside from odd indenting which looks like it can be fixed using an additional parameter for CISB, however the transistor count one is not working as intended. While the transistor count entries are indeed in the box, the core count and extension parameters are also somehow in the collapsed template even though the CISB template is placed below them. What could be causing this? Is there some issue with Template:Infobox CPU? I turned on the syntax highlighter and couldn't see anything that sticks out, but my knowledge of template programming is quite minimal.
I tried enabling parameter "hide_subheadings=yes" for the infobox, and also tried adding "div=yes" parameters to the CISB templates (after seeing an editor remove it for lint error), neither of which have rectified the issue. — AP 499D25(talk)08:11, 7 June 2024 (UTC)
@AP 499D25: The display order of infobox fields is not determined by the order in the call but by the coding of the infobox template. The call can use any order and the order is ignored. The {{{name}}} box at the right of Template:Infobox CPU shows that the field right before transistors is numinstructions. Ryzen doesn't use that field so we go back one more field to extensions. That's where {{Collapsed infobox section begin}} should be placed if you only want transistors collapsed. I have reorganized the call accordingly.[65] Is that the result you want? I didn't actually have to move down the extensions parameter. Since the call order is ignored, I could instead have moved up {{Collapsed infobox section begin}} to wherever extensions is placed in the call, but my call layout is easier to understand and less likely to be broken later. PrimeHunter (talk) 16:49, 7 June 2024 (UTC)
I see what you are talking about. I didn't know the collapsed section begin template could affect what's above it. Indeed, that's the outcome I was looking for. I am happy with the reordering option as it is straightforward to see how the code works. Thanks! — AP 499D25(talk)23:59, 7 June 2024 (UTC)
Quarry - now fails with error
Greetings,
Earlier today, I ran this query to find Orphan articles. It previously ran Ok. First error View 'enwiki_p.pagelinks' references invalid table(s) or column(s) or function(s) or definer/invoker of view lack rights to use them. I then logged off, waited a while, logged back in, then Second error Unknown column 'pl.pl_title' in 'on clause' .
This afternoon, same (second) error, so I ran this query instead. Same error Unknown column 'pl.pl_title' in 'on clause' .
I have no idea how to fix, so I am reporting here and hoping a Quarry expert will be able to correct. Regards, JoeNMLC (talk) 19:55, 7 June 2024 (UTC)
I may be good a cloning a quarry that works (changing month/year), but have zero knowledge of how to fix. Above is the broken one. If someone here can fix, that would be great. Thanks. JoeNMLC (talk) 20:55, 7 June 2024 (UTC)