A WMF researcher mass-messaged a bunch of users with the wrong link, which is currently at RfD. Would someone with a bot account kindly generate a list from the WhatLinksHere (all NS) of User talk:BGerdemann (WMF) and replace [[Special:MyLanguage/Research: with [[m:Special:MyLanguage/Research:? Thanks. Aaron Liu (talk) 22:41, 29 October 2024 (UTC)
(edit conflict) There is only one bot approved to fix mass message errors, but the operator of that bot has been inactive for the last two months. Since there are 2,390 pages needing correction, I am filing a new BRFA. – DreamRimmer (talk) 02:24, 30 October 2024 (UTC)
If you can do it in 8 hours, sure, but I'm feeling like there's a queue thing here and would prefer to retain the old request. Aaron Liu (talk) 10:55, 30 October 2024 (UTC)
Since there is no active bot doing this, I would like to take it on as a generic task for my bot. I have done this before and am available most of the time to address any issues that may arise with mass messages. – DreamRimmer (talk) 11:03, 30 October 2024 (UTC)
There are 1802. Please search for insource:/\[\[Special:MyLanguage\/Research:Wikipedia Administrator Recruitment, Retention, and Attrition/. – DreamRimmer (talk) 15:46, 30 October 2024 (UTC)
I have fixed the remaining 32 pages. I always respect 'nobots' templates, but this was the best approach, as fixing them manually without a bot account would have triggered new message notifications. – DreamRimmer (talk) 05:56, 31 October 2024 (UTC)
Guideline at WP:CAT#T says that "templates are not articles, and thus do not belong in content categories." - Is it possible for an AWB wizard remove these templates from their eponymous categories?RachelTensions (talk) 02:33, 20 October 2024 (UTC)
Given the prefix index only shows 7 (8 including the parent page) pages, I’ll do this by hand, unless there are more pages that I’m not seeing. Working DoneGeardona (talk to me?) 12:00, 6 December 2024 (UTC)
Would it be possible for someone to edit the prose of the articles in Category:Chungha songs to use "Chung Ha" instead of "Chungha"? The parent article was moved after it was established that the artist's name is "Chung Ha" (with the space), not "Chungha".
Right now I'm focusing on the articles directly related to the subject, but there probably a bunch of other locations that refer to her as "Chungha" instead of "Chung Ha" as well.
Are you volunteering to run AWB as bot, or should I be recruiting another bot operator for this? Some semi-automated work on this to develop a good set of patterns would go a long way toward getting bot approval, I think. Dicklyon (talk) 16:18, 6 December 2024 (UTC)
Pages using |Bangladesh cricket=yes should be changed to |Asia=yes as the former task force has been merged into the latter one. (Pages are found in Category:Asia cricket task force articles)
Pages using |needs-photo=yes, |needs-image=yes, |photo-needed=yes and |image-requested=yes should be changed to |image-needed=yes as the former parameters are to be deprecated.
Hi, would it be possible to link instances of "Uprise RI", "UpriseRI", and "upriseri.com" (or "upriseri.com/") within article text and references (website, work, newspaper parameters). At least these searches: [6][7][8]. Thank you! Best, Bridget(talk)01:16, 15 December 2024 (UTC)
Till now I used a regex which checked after a pipe between [[ and ]]. I thought changing the page name might break some links. Should I lowercase all occurrences of at-large, even in page names? ~/Bunnypranav:<ping>04:05, 7 December 2024 (UTC)
Yes, go ahead, since there was a multi-RM discussion that moved all those "At-large" districts to "at-large". If there are any remaining articles with capitalized "At-large" in their titles, you should be able to find them (and maybe fix them) with this search. I don't see much to worry about, other than At-large itself. Dicklyon (talk) 06:38, 7 December 2024 (UTC)
Thanks! If you'd like to do more, with "At-large" in other contexts, take a look at this search, which is mostly in the title bar of infoboxes. Or these in links from templates such as Main and Further. Or these in a common text phrase. Dicklyon (talk) 16:42, 7 December 2024 (UTC)
Yes, make it all lowercase. By the way, I use the "subscribe" mechanism to keep track of conversations, so you don't need to ping me. Dicklyon (talk) 15:57, 15 December 2024 (UTC)
I am done with all three above. The last one below is empty (probably done already). I am not doing the below one due to possible errors. ~/Bunnypranav:<ping>13:23, 19 December 2024 (UTC)
I know that this splits the discussion, but if possible could you post this on my talk page. I can't do this for a day or two, but will definitely try my best. Thanks! ~/Bunnypranav:<ping>17:32, 7 December 2024 (UTC)
@Dicklyon Sorry for the delay, unfortunately I do not have access to a computer for the next couple of days. I will hopefully do them by the end of this weekend. If anyone else is able to do this, then go ahead. ~/Bunnypranav:<ping>10:01, 13 December 2024 (UTC)
@DreamRimmer: I don't have the time to check them all — there are over 700 in the affected nomination. It will include every item on the list that doesn't already have a CFDS tag. –LaundryPizza03 (dc̄) 06:40, 24 December 2024 (UTC)
Certain text, most notably "|collapsed=no", is considered redundant in the WikiProject Banner Shell. Can someone please remove this text from the articles in that category? Thanks! - OpalYosutebito (talk) 05:31, 24 December 2024 (UTC)
This search shows a cluster of about 60 articles with a word-splitting hyphen after a line break, when conventionally the hyphen should be before the line break. Can someone fix? Dicklyon (talk) 23:04, 24 December 2024 (UTC)
Agreed; these should probably be removed; I see little reason to make a table "skinny" by adding in awkward line breaks. Primefac (talk) 09:57, 27 December 2024 (UTC)
Some editor thought it would be good to make the columns less wide, and I wasn't going to argue about that, but it's fine by me if someone wants to fix it by just not doing the break at all. Dicklyon (talk) 09:08, 28 December 2024 (UTC)
Many of those same articles also have over-capitalized table headings and such, like I fixed in this one. I'm not sure how consistent the pattern is, but I've seen quite a few like that. Could someone do the case fixing along with the line-break fixing please? Dicklyon (talk) 09:14, 28 December 2024 (UTC)
@Dicklyon Done60 edits. Fixed the case errors in the totals section of the tables based on some sampling of what parameters exist. Random checking afterwards seems to show most of the case fixing is done. ~/Bunnypranav:<ping>14:02, 28 December 2024 (UTC)
Hey DreamRimmer, I saw this so I made the list for you (https://ctxt.io/2/AAB4sW0WEA, it contains the format you wanted, no pipes, without Talk: prefix and only the links to mental disorders on the page) on behalf of WhatamIdoing. Also, I don't believe this a task that necessitates AWB (there shouldn't be a need for a human to review), so I've coded a regular old bot script to do this based on the list (User:MolecularBot One-off Task #3) and will file a BRFA. MolecularPilot🧪️✈️05:32, 1 January 2025 (UTC)
I usually file BRFAs for tasks that involve editing at least 700-800 pages. For this task, I did not see a need for a BRFA, so I planned to use AWB or PAWS. Thank you for providing the list. I could have created it myself, but I was not sure if any pages needed to be excluded. I am perfectly fine with you doing this using your bot. – DreamRimmer (talk) 10:44, 1 January 2025 (UTC)
Please feel free to run this with AWB as there are only ~300 pages and we tend to not set up bot runs for less than 500. Primefac (talk) 13:22, 1 January 2025 (UTC)
Thank you, WhatamIdoing and thanks to DreamRimmer for handling this! It's really great to see another editor interested in psychology (it's my favorite class at school, and that's why I made the list for DR, because I know about the DSM). :) MolecularPilot🧪️✈️02:36, 2 January 2025 (UTC)
I am aware of TALKLEAD and have just made two edits. I don't see any issue with placing it at the top, but if putting it in the middle of other templates is a better approach, I am happy to do that. – DreamRimmer (talk) 17:26, 1 January 2025 (UTC)
Do you think this qualifies as #5 High-importance attention template, or as an addendum to #14 reliable source for medical articles? VanIsaac, GHTVcontrabout18:53, 1 January 2025 (UTC)
I'd guess that it fits in with #5 ("Notice") or #6 ("Warning"), but I will be satisfied with any location that seems plausible. WhatamIdoing (talk) 23:48, 1 January 2025 (UTC)
The following discussion has been moved here from WP:BOTREQ as such a task would require human confirmation and is not appropriate for a bot to do (but is the perfect use case for AWB). Thank you to everyone for your time! :) MolecularPilot🧪️✈️05:18, 1 January 2025 (UTC)
I'm still finding a lot of TV show articles following the format "Name of show was" even though MOS:TV has dictated "Name of show is" for ages. Could a bot be made that can find cases where "was" is still being used? Ten Pound Hammer • (What did I screw up now?) 19:13, 28 September 2024 (UTC)
@TenPoundHammer: in theory, yes. A lot of variations of the sentence would have to be considered, but it is possible. A few rare false positives may occur if the show's name was changed. This would make it borderline WP:CONTEXTBOT though. But if we make the bot good enough, then we may be able to run it. —usernamekiran (talk) 21:07, 28 September 2024 (UTC)
I have context concerns as well. This might be better for a database report, if only to see the scale/scope of the issue. Primefac (talk) 13:40, 29 September 2024 (UTC)
Sounds like a job for WP:AWB, where each change can be verified by a human before it is made. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:30, 11 November 2024 (UTC)
I wonder, though: If there are really a lot of these, then that's evidence that the community's practice is different from the guideline – and generally, in such cases, it's the guideline that needs updating. If we had an RFC saying something like "Should television shows that have long since ended be described in the present tense? For example, Captain Kangaroo was broadcast from 1955 to 1992. Should the first sentence say that it 'was' a television show, or that it still 'is' a television show?", I'm not certain that people would say that something that stopped being produced and aired 32 years ago is something that we should still describe in the present tense. WhatamIdoing (talk) 01:00, 2 January 2025 (UTC)
From what I gather from past discussions, as well as my own interpretation of MOS:PRESENT, is that if a show still exists then it "is" a program. I think it's worth checking to see how many of these exist before we start worrying about whether practice has drifted from procedure. Primefac (talk) 09:45, 2 January 2025 (UTC)
I disagree with the premise that this needs Human verification on a number of levels. At the most basic level, if we currently have pages that are in error (and we do, as I see these being manually changed and fixed quite often), and even if the bot changes all instances to "is" instead of "was", then that is still more correct than having "was". But we don't need to go that far, if a show is lost, it should be under Category:Lost television shows, so the bot can ignore those pages. If the page isn't there, then we have bigger issues than "is". Since this should not be a one-time run, but at max a weekly run, needing a Human instead of a bot to run this seems counter-productive (and most likely not going to happen). Gonnym (talk) 13:19, 2 January 2025 (UTC)
More "NFL Draft" over-capitalization fixes needed
This search shows about 157 articles needing more post-move cleanup, particularly needing Draft changed to lowercase draft in the lead sentence, all alike except for punctuation order. Other fixes are likely needed in the same articles, like I did here (or variations thereon). I'm happy to help develop or review regex patterns, though I'm still a relative amateur at such things. Dicklyon (talk) 21:38, 11 December 2024 (UTC)
I didn't know that linking it was also requested, as it was not mentioned that clearly in the first post. Will probably do another run soon ~/Bunnypranav:<ping>10:03, 13 December 2024 (UTC)
I was pretty ambiguous. Happy with whatever you want to do on this. We can discuss offline, or at your talk page, or whatever, if you want to volunteer to do more. Dicklyon (talk) 06:39, 14 December 2024 (UTC)
Yes, thanks. But while you're at it, you might see if there are more related fixes to piped links to do, like I did here. Or whatever else you see, if you have time to spend looking for more things to fix. If not, no worries. Dicklyon (talk) 22:40, 14 December 2024 (UTC)
@Bunnypranav: If you're still looking to contribute on this one, the four numbered literal strings above just need to have all occurrences of "Draft" replaced by "draft" (do 3 before 4). Dicklyon (talk) 11:16, 5 January 2025 (UTC)
Thank you! I was legitimately just thinking about doing that when creating the 2025 category archive, checked the history and didn't realize that got fixed today. Appreciate the effort you two! Utopes(talk / cont)18:12, 11 January 2025 (UTC)
A simple case fix in 800 articles
Would someone please do a quick replace run to downcase "Crash" in links found by insource:/\[\[Wall Street Crash of 1929/ (that is, piped or not)? Dicklyon (talk) 09:49, 14 January 2025 (UTC)
Quite a few articles have a two-line table heading where the second line is "Votes" and should be "votes" (e.g. for Total votes, Electoral votes, etc.). The pattern insource:/\<br\>Votes/ finds 97 of these. Some may have other related over-capping, but for many, or most, it's just this one error. Oh, and the br tag should have slash in it. Can someone take on replacing these? Dicklyon (talk) 10:20, 14 January 2025 (UTC)
I've been working on these issues for some time. I have settings files in AWB that target the capitalized versions of our articles for different countries, and there are a lot of them. Dawnseeker200017:25, 5 December 2024 (UTC)
Indeed, I see over 22,000 candidates here, and there are lots more. Perhaps best to divide and conquer, as you look pretty fully occupied? And yes I'd be happy to call and chat. Dicklyon (talk) 17:45, 5 December 2024 (UTC)
I don't think that is a good search parameter for candidates. If I understand the report correctly, it comes from the link part being incorrectly capitalized, not the text. VanIsaac, GHTVcontrabout00:30, 7 December 2024 (UTC)
My initial concern with overfitting the criterion is identifying a link to "Census", which can occur as the initial word of a sentence. Initiating a correction in that instance would be problematic, and I don't know if there are others as well. VanIsaac, GHTVcontrabout02:37, 7 December 2024 (UTC)
Well, yes, that's why we review edits in AWB, isn't it? To catch false positives and write better patterns to avoid them recurring, no? On that particular one, I see a few sentences like "Census day was ...", which seems like a peculiar place to link it, so I'd unlink that and link census somewhere else. Dicklyon (talk) 03:10, 7 December 2024 (UTC)
Here's a good search that finds about 7600 to fix. Just replace the search regex therein with $1 $2 census{{subst:lc:$3}}. It's US only, but could easily be modified for other country censuses. And if you're good with regexes you'll probably see ways to improve it. Dicklyon (talk) 23:14, 8 December 2024 (UTC)
And here's a search finding nearly 1000 different ones, to be replaced with $1 $2 census|$3census]]. Is there a better way to share patterns that doesn't rely on being able to run AWB or JWB? Dicklyon (talk) 23:31, 8 December 2024 (UTC)
@[[User:Dicklyon|Dicklyon] No, the process is very much manual and requires attention while using AWB. I will keep at it when I have the time (i.e. not now, due to WP:BUSY). Milo8505(talk)08:14, 16 January 2025 (UTC)
Mass message fix
Help! I sent a mass message yesterday to Wikipedia:Meetup/LA/Invites with an incorrect date in the header. I wrote "LA Wildfire edit-a-thons January 26 and February 3" - it should be "LA Wildfire edit-a-thons January 26 and February 2." Can someone correct it? Thank you! JSFarman (talk) 01:27, 22 January 2025 (UTC)
If someone wants to start on this, maybe try the first one, search for Latvia and search and replace this regex:
(Aizkraukle|Alūksne|Balvi|Bauska|Cēsis|Daugavpils|Dobele|Gulbene|Jēkabpils|Jelgava|Krāslava|Kuldīga|Liepāja|Limbaži|Ludza|Madona|Ogre|Preiļi|Rēzekne|Riga|Saldus|Talsi|Tukums|Valka|Valmiera|Ventspils) District
Hi. I just performed two moves from ambiguous titles to unambiguous titles, but I need help with all the links pointing to the ambiguous titles. All the links to Joan, Duchess of Brittany should go to Joan of Penthièvre instead (in many cases it is only a matter of removing the silly pipe because she is normally called Joan of Penthièvre anyway); and the links to Joanna of Flanders should go to Joan of Flanders, Countess of Montfort. After that I can turn the two pages into disambiguation pages. I will be very grateful for your help. Surtsicna (talk) 10:55, 25 January 2025 (UTC)
This was incredibly fast! Thank you! I have been editing Wikipedia for 16 years and have never used such tools. It may be time that I learn :) Surtsicna (talk) 11:19, 25 January 2025 (UTC)
Over-capped newspaper titles in refs generated by Trove
We get a lot of auto-generated refs from the Trove newspaper archive, but it capitalizes every word in the newspaper names, including "And". I've fixed at least a hundred of these. Fixing just these two will take several hundred edits:
So that's a quick task for AWB. Going forward, it would be useful if someone would do a run now and then to fix those and all these others (and whichever other ones we might add to the list on discovering them:
In all these the "And" (or two "And"s in couple of them) should be replaced by "and", whether the text is in a link or not. I've linked them here to demonstrate that links won't go red with these changes. Dicklyon (talk) 10:59, 15 January 2025 (UTC)
I can fix those by hand, but if you're developing AWB settings that should be able to find and fix these somewhat automatically, you might want to look at these before I make them disappear. Probably there are also more that I don't see right now, as I'm just looking at ones that currently have incoming links. Dicklyon (talk) 12:27, 16 January 2025 (UTC)
There's also a much bigger generalization of this to redirects with " And " in them. My Quarry query has found over 5000 of them, listed at User:Dicklyon/And. Some of these also have others words over-capitalized (typically "Of", "The", and such). From a quick sample, I'd say most do not have any incoming links, which is good. To find ones that do, it would be useful if each was tagged as "R from miscapitalization". Dicklyon (talk) 10:51, 17 January 2025 (UTC)
If an AWB run or genfix is happening, it could also usefully fix certain wikilinks. Trove offers a canned citation template, but it often links to a different newspaper, to an unrelated topic or to a disambiguation page. For example, the wikitext Trove offers for citing id 505 links to Sunday Times (a UK newspaper) rather than The Sunday Times (Sydney). Id 499 links to Referee, enlightening the reader about sporting umpires, rather than The Referee (newspaper). Id 248 links to dab Record rather than The Record (Melbourne). A list of the more common errors is in User:Certes/Trove. Such links added to Wikipedia before June 2024 have almost all been fixed, but new ones will be appearing and I am not aware of any methodical process to catch them. Certes (talk) 17:32, 17 January 2025 (UTC)
@Dicklyon, my current method for generating the list of fixes is very hard to scale to multiple pages. Could I get a version of https://quarry.wmcloud.org/query/89960 with all of the instances of the problem redirects occurring in plaintext and links? (for example, instead of showing Ashmore And Cartier Islands (the page), it would give me a list of pages that have Ashmore And Cartier Islands (the phrase) on them, etc for all 5057 pages). As for @Certes, my concern with running any kind of automated or semi-automated editor on that dataset is the same ambiguity that is the problem in the first place, if there is any way to find only the problematic citations, that would work fine. (I hope this isn't misunderstanding the request)Let me know if none of this makes sense Geardona (talk to me?) 17:53, 17 January 2025 (UTC)
I wonder if we could convince the good people at Trove to provide actual good wikilinks, using Certes's table.
I don't know a way to collect all the pages that contain all these strings, other than generating one at a time in JWB. But someone who is good with scripting might be able to augment JWB or AWB to take a list of searches to generate from. Dicklyon (talk) 22:33, 17 January 2025 (UTC)
I've tried asking Trove, who politely acknowledged my request but hadn't changed anything last time I looked. The erroneous links are in very specific strings. A typical example newspaper article is here. Click the bookmark icon with hover text "Cite" in the left column, then scroll down to "Wikipedia citation" to see {{cite news |url=http://nla.gov.au/nla.news-article149498474 |title=SHIPPING. |newspaper=[[Daily Telegraph]] |volume=III, |issue=71 |location=Tasmania, Australia |date=18 June 1883 |accessdate=18 January 2025 |page=2 |via=National Library of Australia}}, which our editors copy verbatim. Sadly, this doesn't contain the Trove ID which would distinguish it from other historic Australian papers called Daily Telegraph but the location usually provides a distinctive enough pattern, e.g. \{\{cite news \|url=http://nla\.gov\.au/nla\.news-article[^}]*\|newspaper=\[\[Daily Telegraph\]\][^}]* \|location=Tasmania, Australia[^}]* \|via=National Library of Australia\}\}. In a few awkward cases, the same vague location was used for two distinct papers with different articles. Those require checking the date but are edge cases. It's basically what I used to do manually each day as and when the errors appeared. Certes (talk) 23:33, 17 January 2025 (UTC)
Hello, I was pointed here for this task from Wikipedia:Bot requests, there is a need to remove leading and trailing spaces from the non-breaking space character ( ). If a space exists then you end up with 2 spaces in the rendered text and it negates the purpose of having a non-breaking space as a break can be made between the space and the non-breaking space. You should ignore the cases where a non-breaking space is used as a template parameter or a cell entry in a table. Keith D (talk) 17:04, 5 February 2025 (UTC)
This request seems wrongheaded on its face. The insertion of NBSP between spaces is the only way to have the wikitext interpreter render multiple spaces when needed, and without viewing the full page in a standard browser window, you would be unable to even make any sort of judgement about whether this might be a possibility, making it incompatible with standard operation in the AWB interface. I'm not even sure you could generate a target list sans a full database dump. This is exactly the sort of unthoughtful fiddling - making assumptions that standard workarounds are somehow errors - that pisses off editors, so I'd withdraw the request. VanIsaac, GHTVcontrabout04:30, 6 February 2025 (UTC)
OK I will withdraw this, but there is a need to clean-up such things as £3 which causes a problem of having 2 spaces and negates the use of the non-breaking space. Keith D (talk) 11:30, 6 February 2025 (UTC)
This is not a bad idea, but the request above is underspecified at best. What are the exact situations in which this construction is allowed, and in what situations is it not allowed? Figuring out which instances should be fixed, and how, and which should not, will probably require human judgement in many cases. In many cases, the right answer is to remove the nbsp, not the space. Here's a search in article space that currently returns 17,649 articles, some of which are probably false positives that should not be modified. – Jonesey95 (talk) 00:17, 8 February 2025 (UTC)
It has been decided this infobox should be deleted after converting its transclusions. As a part of the conversion, the category (Category:Cyrillic letters) and the short description ("Cyrillic letter") transcluded from the infobox should be added to articles that transcludes the template. Preferably, SD should not be added to articles that already have one, and the category should not be added to pages that are already in one of its subcats (as it is non-diffusing). Janhrach (talk) 20:54, 7 February 2025 (UTC)
I am sorry, I withdraw my request. I wanted to handle tranclusions of {{Infobox Cyrillic letter}} without any parameters (with empty parameters treated as if they weren't there; hence I used the potentially unclear wording "tranclusions [...] without any non-empty parameters") separately, but it is not necessary. Janhrach (talk) 17:08, 10 February 2025 (UTC)
I would like to request an AWB run to convert transclusions of {{Infobox Cyrillic letter}} to {{Infobox grapheme}} and {{Cyrillic alphabet sidebar}}. I have already done such conversions (semi-)manually by replacing {{Infobox Cyrillic letter|[...]}} with {{subst:User:Janhrach/Infobox Cyrillic letter|[...]}}. I think this method could be used to convert the rest, with help of AWB.
I have spotted some drawbacks, but I think I have handled them sufficiently. The title and audio params of Infobox_Cyrillic_letter are not supported by Infobox_grapheme; however, these appear to be unused (tracking categories: Category:Pages using Template:Infobox Cyrillic letter with param audio, Category:Pages using Template:Infobox Cyrillic letter with param title). There is no simple way to fill the letter param of Infobox_grapheme, and also the derived param is not converted into the proper format for fam1. The former drawback causes visible quirks only when the param derived is present; I have manually converted all transclusions containing derived and ensured the results are good. (See Category:Pages using Template:Infobox Cyrillic letter with param derived. The two drafts remain there because the characters in question are not in Unicode, and I am unsure if images are OK to be added to |letter=. Anyway, these are not going to be accepted because of unnotability.)
I assume that this is intentional, because these are in the User or Draft namespaces. I think the best thing to do is simply to remove the template transclusions – I think we should minimize messing with others' userspace, and as for the drafts, these are not going to be accepted because of unnotability, so we should not waste time on them. Janhrach (talk) 16:45, 17 February 2025 (UTC)
I think that's a rather flawed line of reasoning. If the template is already being replaced, why not just replace it everywhere? That way the person using the original template isn't struggling to figure out how to replace it themselves after it is removed/deleted. Helping another user isn't "messing with" their draft. Primefac (talk) 16:57, 17 February 2025 (UTC)
So? Who cares? If a template needs replacing, it should be replaced. It shouldn't matter if it's on a live article, a user sandbox, or the user page of someone who hasn't edited in 20 years. Primefac (talk) 17:17, 17 February 2025 (UTC)