To run a bot on the English Wikipedia, you must first get it approved. Follow the instructions below to add a request. If you are not familiar with programming consider asking someone else to run a bot for you.
If your task could be controversial (e.g. most bots making non-maintenance edits to articles and most bots posting messages on user talk pages), seek consensus for the task. Common places to start include WP:Village pump (proposals) and the talk pages of the relevant policies, guidelines, templates, and/or WikiProjects. Link to this discussion in your request for approval.
You will need to create an account for your bot if you haven't already done so. Click here when logged in to create the account, linking it to yours. (If you do not create the bot account while logged in, it is likely to be blocked as a possible sockpuppet or unauthorised bot until you verify ownership)
Create a userpage for your bot, linking to your userpage (this is commonly done using the ((bot)) template) and describing its functions. You may also include an 'emergency shutoff button'.
II
Filing the application
easy-brfa.js can be used for quickly filing BRFAs. It checks for a bunch of filing mistakes automatically! It's recommended for experienced bot operators, but the script can be used by anyone.
Enter your bot's user name in the box below and click the button. If this is a request for an additional task, put a task number as well (e.g. BotName 2).
Complete the questions on the resulting page and save it.
Your request must now be added to the correct section of the main approvals page: Click here and add ((BRFA)) to the top of the list, directly below the comment line.
For an additional task request: use ((BRFA|bot name|task number|Open))
III
During the approvals process
During the process, an approvals group member may approve a trial for your bot (typically after allowing time for community input), and AnomieBOT will move the request to this section.
Run the bot for the specified number of edits/time period, then add ((Bot trial complete)) to the request page. It helps if you also link to the bot's contributions, and comment on any errors that may have occurred.
AnomieBOT will move the request to the 'trial complete' section by moving the ((BRFA)) template that applies to your bot
If you feel that your request is being overlooked (no BAG attention for ~1 week) you can add ((BAG assistance needed)) to the page. However, please do not use it after every comment!
At any time during the approvals process, you may withdraw your request by adding ((BotWithdrawn)) to your bot's approval page.
IV
After the approvals process
After the trial edits have been reviewed and enough time has passed for any more discussion, a BAG member will approve or deny the request appropriately.
For approved requests: The request will be listed here. If necessary, a bureaucrat will flag the bot within a couple of days and you can then run the task fully (it's best to wait for the flag, to avoid cluttering recent changes). If the bot already has a flag, or is to run without one, you may start the task when ready.
For denied/expired/withdrawn requests: The request will be listed at the bottom of the main BRFA page in the relevant section.
A task is a task that is performed on multiple Wikipedias (link).
Discussion
Comment to BAG: Please note the sordid history of this editor's bot activity. It is/was blocked for good reasons. – Jonesey95 (talk) 19:10, 16 February 2024 (UTC)[reply]
This editor had problems with misspelling and inconsistency in the past. I have concerns that this pattern will continue, evidenced by invalid use of <ins>...</ins> tags and an unclosed <p> tag in the original post here (I have correct both in this edit), as well as misspelling of "ACCES_DATE" and the proposal to use the YYYY-MM-DD format in articles where another date format may be preferred. – Jonesey95 (talk) 15:27, 17 February 2024 (UTC)[reply]
@Jonesey95: I changed it. The date of the population will only be a year, and the cells will be displayed via the tooltip. --Dušan Kreheľ (talk) 19:46, 20 February 2024 (UTC)[reply]
My question is this: the bot's first BRFA was approved, and later revoked. This task is nearly identical to that task. What is different about this task that should allow it to be approved when what appears to be the same task was already revoked? Primefac (talk) 08:38, 17 February 2024 (UTC)[reply]
@Primefac: Request n. I is a multi-task with several sub-tasks and with a looser specification. Unlike that task, this task determines the boundaries of this task in detail. That task was not canceled due to the reason that "this sub-task of request n. I" makes errors. Dušan Kreheľ (talk) 19:07, 17 February 2024 (UTC)[reply]
I remember reviewing the original BRFA. I never closed because, regardless of whether the implementation was technically correct (and I remember being a bit unsure of that), I was a bit uncomfortable as to whether the bot operator can comply with the requirements WP:BOTPOL places on operators. Things like WP:BOTCOMM and WP:BOTISSUE for instance. Therefore I'm also unsure about approving this bot task. Perhaps other BAG feel differently here. ProcrastinatingReader (talk) 13:12, 20 February 2024 (UTC)[reply]
This might be a dumb question, but if every page indicating that it is missing coords is already on Wikipedia, why do we have this template? Why not just make it a wrapper/redirect to ((coord))? I know that MSGJ asked for this, and I trust their judgment, but this seems like a lot of work if it's that trivial of an issue. Primefac (talk) 14:26, 7 February 2024 (UTC)[reply]
@Primefac: ((coord missing)) contains 93,000 pages. Only 20,000 of them have coordinates on Wikidata. However, if you'd like to update the templates so this request would be moot, that's fine with me. GoingBatty (talk) 14:39, 7 February 2024 (UTC)[reply]
Yup, as I said... dumb question. Apparently I can't read, and (for some reason) thought that was also the transclusion count. Primefac (talk) 14:46, 7 February 2024 (UTC)[reply]
Do you mean that if coordinates are on Wikidata then we can just change the behaviour of ((coord missing)) so that it uses those coordinates instead of populating the maintenance category? Yes, that could be possible. The only disadvantages I can think of:
It will still say "coord missing" in the wikicode which could be misleading
The template may not be placed in the usual place (e.g. at the top) that the coord template is usually placed.
— Martin (MSGJ · talk) 15:05, 7 February 2024 (UTC)[reply]
I'd be happy to support this, with one proviso, that the ((coord)) template be changed so that transclusion from Wikidata was gated via a "source=wikidata" parameter, so that it was clear on the enwiki side why the edit was made, and where the coordinates were being pulled from. So ((coord missing|Name of region)) would become ((coord|source=wikidata|display=title)). — The Anome (talk) 17:02, 7 February 2024 (UTC)[reply]
That parameter is not required by the template, so are you just putting that in for explanatory purposes and the template will just ignore it? — Martin (MSGJ · talk) 17:07, 7 February 2024 (UTC)[reply]
Putting it there for explanatory and tracking purposes, so we can see how it got there and why. Right now, the template should ignore it, but might also be useful to let us add extra semantics to ((coord)). For example: right now, if you add ((coord|display=title)) to an article that doesn't have coordinates on Wikidata, it blows up in an uninformative and mystifying way. Having this extra parameter might let the error message be improved, and also let such erroneous articles be put in a tracking category.
From my viewpoint as a maintainer of the ((coord missing)) ecosystem, the more tracking metadata we have, the better. I'm committed to a long-term transition to Wikidata as the master source of geodata, but it's going to be a long process, and the more we can smooth the transition by having backward and forward compatibility during the transition process, the better. — The Anome (talk) 17:09, 7 February 2024 (UTC)[reply]
Okay thanks for the explanation. I support this approach. — Martin (MSGJ · talk) 17:27, 7 February 2024 (UTC)[reply]
Thanks. I'm glad you're stepping up for this; ecosystems made out of multiple single-purpose bots that use wikitext, templates and categories for orchestration are generally more robust that those which rely on complex multi-purpose bots. — The Anome (talk) 17:30, 7 February 2024 (UTC)[reply]
@The Anome and MSGJ: Updated this bot request to change to ((coord|source:wikidata|display=title)) (e.g. this edit). (Note that |source=wikidata generates an error.) GoingBatty (talk) 19:44, 7 February 2024 (UTC)[reply]
Thanks all, I noticed the lack of "where is this data coming from" earlier but got distracted and never came back to express my concern; this seems to deal with the issue in a suitable manner. Primefac (talk) 21:07, 7 February 2024 (UTC)[reply]
What sort of error? source:wikidata seems unhelpful for populating a tracking category (if desired) — Martin (MSGJ · talk) 21:38, 7 February 2024 (UTC)[reply]
If the source parameter gives an error, the template can and should be changed to fix that. I've put forward a proposal on this on the template's talk page. — The Anome (talk) 21:40, 7 February 2024 (UTC)[reply]
@MSGJ, @The Anome: Go to 2024 Istanbul church shooting, click "Edit source", change source:wikidata to source=wikidata, then click "Show preview", and you'll see a message at the top right of the article next to the coordinates which states Coordinates: Parameter: "source=" should be "source:". GoingBatty (talk) 22:30, 7 February 2024 (UTC)[reply]
Understood. I also support The Anome's proposals to change these strange parameters — Martin (MSGJ · talk) 08:51, 8 February 2024 (UTC)[reply]
I'm happy to use any new parameters in this request. GoingBatty (talk) 14:37, 8 February 2024 (UTC)[reply]
@The Anome and MSGJ: If it's the template deciding when it pulls data from WikiData (presumably when no parameters are supplied and data is available on WD), then wouldn't it make sense for the template/module to add things into a tracking cat, rather than it be a parameter. Because the parameter might not be accurate as it's not the source of truth, right? e.g.:
One could write ((Coord|57|18|22|N|4|27|32|W|source=wikidata)), so the actual data isn't shown via WD but the parameter is still there
One could write ((Coord)), and the actual data is from WD but the source parameter hasn't been supplied, so is not in a tracking cat
On hold. pending fix to the ((coord)) template (assuming that's the consensus here). Please deactivate the template/ping when the issue is resolved. ProcrastinatingReader (talk) 12:05, 15 February 2024 (UTC)[reply]
Function details: Goes through pages marked with Template:Expand language, and sorts them by topic, based on categories that the page is in (e.g. "Living people").
Discussion
Sort where? Could you please give an example of an edit this bot would make? Primefac (talk) 08:02, 22 April 2024 (UTC)[reply]
My bot would not categorize all articles. It would only do the ones where the topic is obvious, e.g. articles in Category:Living people. Numberguy6 (talk) 13:13, 30 April 2024 (UTC)[reply]
Approved for trial (100 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete.Primefac (talk) 11:33, 3 May 2024 (UTC)[reply]
Function details: This script searches for pages linking to "[year] NFL Draft" and "List of [players] in the NFL Draft" pages, which are all now redirects using "NFL draft" (note capitalization) per this RfC and related discussions. The script then corrects the capitalization of these links, including within wikilinks, ((Main)) templates, ((See also)) templates, and ((Further)) templates. (It does not make any cosmetic-only edits, such as only changing [[2024 NFL Draft|2024]] to [[2024 NFL draft|2024]].) I initially found 46,276 pages that link to one or more of these redirects and would likely be edited by this task initially. I plan to do an initial run with this list, editing 1,000 pages daily until the list is exhausted. (I'm open to other number suggestions—I just don't want to flood watchlists and whatnot too quickly.) Finally, here's an example diff in my sandbox.
The script then corrects the capitalization of these links, including within wikilinks: To be clear, would [[2024 NFL Draft|NFL Draft]] become [[2024 NFL draft|NFL draft]] too?—Bagumba (talk) 03:23, 23 March 2024 (UTC)[reply]
@Bagumba: As of now, yes. The actual procedure behind it is that the exact word "draft" in any capitalization within a wikilink's text becomes lowercase. Here's the code behind this and an example. Bsoyka (t • c • g) 03:28, 23 March 2024 (UTC)[reply]
As the well-attended RfC confirmed that "NFL Draft" should be lowercase in titles and prose, changing the "NFL Draft" redirect links and any piped text in articles seems to be a well-defined, low-risk task that is non-cosmetic. As someone doing this semi-manually on JWB, I hope this bot can be approved for this high-volume task.—Bagumba (talk) 03:39, 23 March 2024 (UTC)[reply]
More NFL Draft changes (downcasings) are needed in links to "List of * in the NFL Draft". It looks like there are 634 such over-capped redirects, with many links to each. This could be added here, or in a later bot request, perhaps with all the other league drafts added. I hope this first round goes well so we can advance to the rest (most of the other were approved via RM discussions, so didn't specially talk about "in text" as the NFL RFC did, but that's how it's done, so I don't think there will be any further controversy). Dicklyon (talk) 23:47, 24 March 2024 (UTC)[reply]
@Dicklyon: Good point. Including this search, I have now found 46,276 pages to edit initially, which I've updated above. Updated code coming soon to support this additional change. Bsoyka (t • c • g) 02:03, 25 March 2024 (UTC)[reply]
Cosmetics question: You say above "(It does not make any cosmetic-only edits, such as only changing 2024 to 2024.)" I had suggested you not edit articles where that would be the only change, as that would be "purely cosmetic", but it's still worth doing in articles where you're doing other changes, to help with the tracking of links to miscapitalized redirects, right? So you'd go ahead and make that edit when other non-cosmetic changes are being made in the same article? Or not? Dicklyon (talk) 05:38, 25 March 2024 (UTC)[reply]
@Dicklyon: To clarify, yes, cosmetic changes will be made when other non-cosmetic edits are done on the same page. The bot just shouldn't make any edits that are solely cosmetic changes, in line with your suggestion. Bsoyka (t • c • g) 21:04, 25 March 2024 (UTC)[reply]
Thanks for clarifying. Dicklyon (talk) 10:35, 26 March 2024 (UTC)[reply]
Maybe faster is better. If there's no adverse reaction after a few days at 1000 per day, I'd suggest cranking up to 3000 per day, to get it done in a little over 2 weeks. Dicklyon (talk) 10:26, 27 March 2024 (UTC)[reply]
Bsoyka, these may have already been addressed by hand, but curious if this would capture the use of "Draft" (unlinked) in a section heading? As an example, which has been fixed, see the headings in Washington Commanders draft history. « Gonzo fan2007(talk) @ 20:07, 3 April 2024 (UTC)[reply]
As of now, this does not address that. I feel like that's likely a bit more complicated to find and fix without false positives. Bsoyka (t • c • g) 01:45, 4 April 2024 (UTC)[reply]
Agree that it's riskier, but also less frequent. Section names can be lumped in with manual exercise of dealing with unlinked "NFL Draft" references in prose.—Bagumba (talk) 03:13, 4 April 2024 (UTC)[reply]
Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete.Primefac (talk) 09:48, 22 April 2024 (UTC)[reply]
It looks like Bsoyka hasn't edited for a few days. Here's hoping he's back soon and still willing to work on this. Dicklyon (talk) 03:15, 24 April 2024 (UTC)[reply]
I'm still around, but very busy in real life at the moment. I plan to try to get this up and running in around the next week, maybe two, whenever I find a free moment. Bsoyka (t • c • g) 18:52, 24 April 2024 (UTC)[reply]
Function details: This is a replacement for EranBot (BRFA 3) to report potential copyright issues found in articles and drafts from CopyPatrol to PageTriage (shown in Special:NewPagesFeed). When a report is added to CopyPatrol, check if the page is in PageTriage (action=pagetriagelist). If so, report the revision (action=pagetriagetagcopyvio). The bot does not edit.
Notes:
The code has been completely rewritten, migrating from Python 2 to Python 3.11 and from the XML-RPC-based API that EranBot uses to the new Turnitin Core API (TCA). CopyPatrol is also being migrated from Toolforge tools (EranBot is running on the deprecated Grid Engine.) and database to WMCS Cloud VPS instances.
Once this bot starts running, even for a trial, EranBot will be disabled.
We're currently waiting for the WMF to sign an agreement with Turnitin before we receive production TCA access. (required before trial)
Discussion
On hold. pending the TCA access. Otherwise looks fine to me, replacing functionality of existing bot. Can you drop a note on WT:NPP/R and WT:AFC please, alerting them of this BRFA. Also, is this handover being coordinated with the operator of EranBot? ProcrastinatingReader (talk) 11:30, 20 February 2024 (UTC)[reply]
The handover was authorized by User:ערן at phab:T334265. JJMC89 and myself already have access to the eranbot account on Toolforge and will be able to stop/restart it as needed. — MusikAnimaltalk 15:59, 20 February 2024 (UTC)[reply]
Yes the handover is co-ordindated. Note that EranBot bot is used for multiple tasks beside copyright violations (outside enwiki), and the handover is specifically for copyvio aspect. Having a dedicated user for copyright violation bot seems to be more correct than a general bot, and thanks for JJMC89 and MusikAnimal drive the migration and the rewrite effort. Eran (talk) 16:31, 20 February 2024 (UTC)[reply]
@ProcrastinatingReader: I dropped pointers to this BRFA on those two pages. It would be helpful to have trial approval and copyviobot in advance of getting TCA access to cut down on any wait or downtime. (We're coming up on the Grid Engine shutdown soon.) — JJMC89 (T·C) 23:33, 5 March 2024 (UTC)[reply]
JJMC89, I think PR wants you to have access before granting a trial. Primefac (talk) 20:58, 10 March 2024 (UTC)[reply]
Yeah, as Primefac said, I was waiting for this to be ready to begin before approving and requesting a crat give the bot the appropriate perms.
Not that it matters too much either way, but it seems like the delay time on TCA access is much greater than BAG delay time, and it's better practice not to assign userrights to bots-in-trial until they're ready to commence trial imminently. In this case, it seems like you're blocked on other things for an indeterminate period of time, before you're able to begin trial.
If you really want it sooner and Primefac is happy with that, then sure. Otherwise yeah, since there's no objections from users working in this area and there is the consent of the existing bot operator, I'm happy to tick this when you're ready to begin. ProcrastinatingReader (talk) 15:26, 12 March 2024 (UTC)[reply]
Also a poke to @Diannaa and Moneytrees: as some people involved in copyright issues, in case you have any comments, or know of anyone who might. ProcrastinatingReader (talk) 11:34, 20 February 2024 (UTC)[reply]
Approved for trial (30 days). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Please report back after this period on the bot's progress. Primefac (talk) 12:42, 13 March 2024 (UTC)[reply]
Not the Bot Operator but this link is what I think you mean. Nobody (talk) 12:43, 30 April 2024 (UTC)[reply]
The trial has been running for three weeks so far. I haven't encountered any issues. Feel free to review Special:Log/CopyPatrolBot at your convenience. — JJMC89 (T·C) 00:48, 1 May 2024 (UTC)[reply]
Function details: Based on the configuration at User:SDZeroBot/Gadgets-sync-config.json, the bot checks each remote page for new edits, and raises an edit request to the corresponding local page.
This is intended to allow locally hosting a copy of gadgets from other wikis. Hosting locally significantly improves performance due to ResourceLoader minification and caching. It's also somewhat better from a security standpoint as local interface admins can audit changes before syncing them.
Doing a one-time copy without a setup to track upstream edits generally doesn't work. It results in unmaintained forks which miss out on bug fixes and new features added in the upstream.
Discussion
Approved for trial (5 edits or 30 days, whichever happens first). Please provide a link to the relevant contributions and/or diffs when the trial is complete. I feel like a trial here is mostly going to be a proof of concept, but a very useful one if it behaves as intended, so it's not a high edit count (i.e. the time delay is mainly so it doesn't last forever). Primefac (talk) 20:57, 10 March 2024 (UTC)[reply]
@User:SD0001 Would the bot generate two requests if a change was made and subsequently reverted upstream within a short period of time. (ping to reply pls) Sohom (talk) 04:21, 5 April 2024 (UTC)[reply]
@Sohom Datta Depends on how short the period of time is. If less than 1 hour, then no. – SD0001 (talk) 09:27, 5 April 2024 (UTC)[reply]
I think that is a acceptable margin. (I would have personally gone for intervals of 1 day, but that's WP:BIKESHED) Sohom (talk) 15:55, 7 April 2024 (UTC)[reply]
Time filed: 03:19, Saturday, February 17, 2024 (UTC)
Function overview: Automatically populate sandboxes for modules listing disambiguation templates, set index templates or soft redirect templates; and if necessary issue edit requests to update the main modules.
The bot uses the following algorithm. This has not changed since task 1.
The bot constructs a list of disambiguation templates, set index templates or soft redirect templates. The list includes template redirects. It does this by iterating through all templates in a specific category (the -cat option). Non-templates are ignored, as are templates in the bot's exclusion list (the -exclude option). It then retrieves redirects for all of these templates.
It formats the list of templates and template redirects as a Lua table, and saves the result in a module sandbox (the -data-page-sandbox option). Saving is skipped if the module's content would not change. You can see sample output from the bot here.
If the sandbox module was updated, and if its new content is different from that of the main module (the -data-page option), then the bot adds an edit request to update the main module to the relevant talk page (the -data-talk-page option). There is a sample edit request here.
The bot will be used to update the following modules:
Approved for trial (one-time run). Please provide a link to the relevant contributions and/or diffs when the trial is complete. – SD0001 (talk) 05:58, 17 February 2024 (UTC)[reply]
Edit period(s): Checks the daily TfD subpage once an hour for new discussions, actually edits a few times a week. Also a large one time addition for all past discussion.
Estimated number of pages affected: Maybe 2000-2500 when applying boxes to discussions retroactively (one time only). When in continuous use maybe a dozen a week.
Function details: The bot consists of 3 parts. One script that goes through all Wikipedia:Templates for discussion/Log subpages and scrapes what templates were discussed (linked in ((Tfd links))), the result and the link to the discussion. This information is added to a 14 MB python dictionary. Relists are ignored. Part two goes through all the same pages and adds an ((Other TfDs)) box if one of the templates has been discussed some other time (either previously or later) as determined by the dictionary. There are up to ~3000 discussions out of ~56,000 that could get such a box, but the real number will be less due to identical mass noms and templates nominated more than twice being double counted. The third script would check the TfD subpage of the day for new discussions each hour and adds them to the dictionary. If any discussion needs a box it adds one there and updates previous discussions to include a link to the new one. This script isn't written yet, but will mostly consist of parts from the other two scripts. These user pages have a small number of successful test edits: 1, 2, and 3.
Discussion
Linking older discussions from new ones is fine, but I don't think modifying older discussions to link to the newer ones is a good idea – the archive headers say to not modify the closed discussion. – SD0001 (talk) 20:59, 28 December 2023 (UTC)[reply]
@SD0001 On AfD the corresponding feature uses Special:Prefixindex. This results in links to future discussions being automatically added. This is useful and in no way alters the contents of the discussion. I believe we should do the same here. Trialpears (talk) 21:08, 28 December 2023 (UTC)[reply]
I agree with SD0001 on this matter; past discussions do not necessarily need to have a link to further-past discussions as all discussions in question are already closed; in other words I see no tangible benefit (though as I stated in the BOTREQ I think this is a good idea going forward). Primefac (talk) 08:22, 31 December 2023 (UTC)[reply]
On the other hand why not? Retroactive is not the sort of task that I would see as actively worth doing, but if someone wants to do it then more power to them. "Please do not modify it" on closed discussions has never been an absolute rule. I would have the template clearly state it was added retroactively, but the addition is IMO useful and causes little harm. * Pppery *it has begun... 05:02, 11 January 2024 (UTC)[reply]
It's also two different sets of code, but I see your point. Primefac (talk) 19:00, 11 January 2024 (UTC)[reply]
Approved for trial. Please provide a link to the relevant contributions and/or diffs when the trial is complete. I want this to be run in two separate batches. One batch will be "50 edits or 14 days, whichever comes first" and will only deal with new nominations. The second batch will be 25 edits to historical log pages to see proof of concept. Ideally if you could run them that way so the contributions are easier to split, that would be great. Feel free to start with the second batch first since it's not time-dependent. Primefac (talk) 19:00, 11 January 2024 (UTC)[reply]
A user has requested the attention of the operator. Once the operator has seen this message and replied, please deactivate this tag. (user notified) Any updates? Primefac (talk) 20:11, 18 February 2024 (UTC)[reply]
Function overview: This task checks the Top 25 Report page frequently to see if the current report has updated. If it was updated, then it will go through all pages in the new report and add or update the Template:Top 25 Report template on their talk pages.
Function details: This task first checks the page Wikipedia:Top 25 Report to see if the transcluded link was modified. (This should mean that the report was updated.) If it has, then it uses the first revision of the transcluded page, which is always a basic list, to get a list of article talk pages to modify. It then goes through each talk page, updating the Template:Top 25 Report template if it exists and adding it if not. As for exclusion compliance, I have not added that feature in yet.
Discussion
The Top 25 report is updated weekly. Why does this task need to run twice a day? Primefac (talk) 09:08, 14 June 2023 (UTC)[reply]
I wanted to ensure that the template is added quickly. I've changed it to daily, and if it should be longer then you can tell me. Capsulecap (talk • contribs) 14:24, 14 June 2023 (UTC)[reply]
Additionally, some reports (including the one for last week) are finished late, and do not get added until later on. I wanted to ensure that the pages on the report get the template on their talk page. If the next report is done on time, then the maintainers of the report will replace the transclusion to the late report with the new one less than a week after the old report replaced the one before it. I agree that twice a day was a bit too excessive. Daily should be fine. Capsulecap (talk • contribs) 14:28, 14 June 2023 (UTC)[reply]
Capsulecap is right about this. And task need to run twice a day.--BabbaQ (talk) 15:58, 14 June 2023 (UTC)[reply]
@Capsulecap: Hi. What would happen if the same article comes in top 25 report again, say with a gap of four months? —usernamekiran (talk) 17:21, 21 June 2023 (UTC)[reply]
If that happens, then there will be no difference from if it was featured twice with more than a four month gap. There is nothing that says to do anything different for pages on T25 which are featured multiple times in a small timespan, and pages like Talk:ChatGPT feature multiple such examples. Capsulecap (talk • contribs) 23:56, 21 June 2023 (UTC)[reply]
Approved for trial (1 days). Please provide a link to the relevant contributions and/or diffs when the trial is complete. I'm trying to wrap my head around what's this bot supposed to do exactly, so I'm going to approve it for a one-time run of 1 day. This should give me (and perhaps others) a better idea of what this is about. Headbomb {t · c · p · b} 17:34, 2 July 2023 (UTC)[reply]
@Headbomb: Although I did a trial run, the bot made test edits with numerous errors. I have fixed the code causing these issues, and will (with permission) restart the trial when the next report comes in. Capsulecap (talk • contribs) 19:51, 3 July 2023 (UTC)[reply]
@Capsulecap: can you link to the results nonetheless? Headbomb {t · c · p · b} 21:46, 3 July 2023 (UTC)[reply]
See edits 4 through 29. Note that the newest three edits were a test run for a fix to something which happened in Talk:Elemental (2023 film), and that many incorrect edits were caused by other editors modifying talk pages to add the template before the test run was done. Although the bot will not add redundant templates assuming that nobody adds the top 25 placement before it, I am considering adding redundancy protection. One problem — the one on the page about the Titan submarine incident — was one I didn't think of, as the talk page was moved with the main page, causing the top 25 report template to be placed on a redirect instead of the actual talk page. This is a problem I am working on fixing, as I have noticed that "current events" pages that show up on the report often frequently get moved. The bot also ended up creating the page "Talk:Errible things in Russia, the North Atlantic and HBO have the most attention this week.", but I fixed the source issue and tagged the page for CSD. few of the edits are fine, and most would be fine if there was redundancy protection or if the top 25 templates didn't already have the week in there. One question, though — since the bot will run daily, and people wouldn't need to modify top 25 templates anymore — should I implement redundancy protection? Capsulecap (talk • contribs) 02:46, 4 July 2023 (UTC)[reply]
"Should I implement redundancy protection" I would say that's a good idea, regardless of how often it comes into play. Headbomb {t · c · p · b} 02:52, 4 July 2023 (UTC)[reply]
I just finished implementing the redundancy protection along with the redirect traversal stuff. The bot should work just fine now. Do I have to redo the trial? Capsulecap (talk • contribs) 04:35, 4 July 2023 (UTC)[reply]
Trial complete.See 21 most recent contributions. Out of the 25 pages in the June 25th to July 1st edition, 21 pages were correctly edited, two pages (Talk:Money in the Bank (2023) and Talk:Titan submersible implosion) were not edited because of unexpected and likely erroneous formatting in the report's first revision (a space was in place of the usual tab after those two pages' titles), and two pages were not edited as they already had this week in their templates. For context on those two pages which didn't get the template on accident, the first revision of the report is always an imported set of tab delimited data — in this case, spaces were in place of tabs for the names of those two articles. The bot created two new talk pages on accident, which I quickly tagged for CSD. Capsulecap (talk • contribs) 05:48, 7 July 2023 (UTC)[reply]
Update: I've come up with a solution to this problem and will be implementing and testing it soon. This is the last issue which I will have to fix. Capsulecap (talk • contribs) 16:26, 7 July 2023 (UTC)[reply]
Approved for extended trial (25 edits or 7 days). Please provide a link to the relevant contributions and/or diffs when the trial is complete. One week's worth, or 25 edits, whichever you need. Headbomb {t · c · p · b} 17:16, 7 July 2023 (UTC)[reply]
Trial complete. See See 25 most recent contributions. This time, I verified that all edits the bot would make would be correct on a script that had editing commented out. They were all good edits, so I ran the full script. All 25 pages on the report had the template added or changed on their talk pages. Capsulecap (talk • contribs) 01:57, 15 July 2023 (UTC)[reply]
Most seemed fine, but there was this that stood out.
I noticed that and didn't pay much attention to it as it was merely cosmetic. Since that was considered problematic, I'll get to fixing that and keeping the collapse as the last edit. Capsulecap (talk • contribs) 14:57, 21 July 2023 (UTC)[reply]
For testing you can revert to a prior state and unleash the bot on it. Headbomb {t · c · p · b} 16:17, 21 July 2023 (UTC)[reply]
Well... the collapsed stuff is handled correctly, but now it's inconsistent the other way around. It should list the ranks when they're there, or omit them when they're not.
Or, probably a better idea, update old listings to list the ranks, e.g. [3]. You might need some discussion before though. Headbomb {t · c · p · b} 20:18, 21 July 2023 (UTC)[reply]
I think it's a good idea to retroactively add the rankings to the templates, but I'm not sure of where to obtain consensus for that, and it would either require a bot task or lots of manual work. The other way you listed is probably easier, but causes inconsistency between pages. Something else I thought of is a Lua module that automatically grabs the placements, but I'm not sure if such a thing is supported. Capsulecap (talk • contribs) 20:30, 21 July 2023 (UTC)[reply]
What if it deleted what was there first, then re-added the template with all dates and ranks? In the same edit that is. Headbomb {t · c · p · b} 20:46, 21 July 2023 (UTC)[reply]
It could work, but I think I would have to submit a separate bot task for that. A separate (and much simpler) approach would be to add a "ranks" parameter that does nothing to the bot category. If set to yes, then the bot will add ranks when it updates the report. Otherwise or if unset, the bot will only add the date. This maintains consistency within talk pages, but not between talk pages; the latter would require consensus strongly towards either using ranks or not. Capsulecap (talk • contribs) 21:03, 21 July 2023 (UTC)[reply]
Approved for extended trial (25 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Indeed, consistency within talk pages is usually a lesser threshold to clear. I'm giving you trial for that (make sure to include a mix of both types of edits), but if you want to have that (should we always rank things) discussion first, you can also wait for consensus to emerge before trialing. Headbomb {t · c · p · b} 21:11, 21 July 2023 (UTC)[reply]
Yes, I am. Have taken a long break from editing but I never canceled this bot project. Capsulecap (talk • contribs) 15:44, 21 October 2023 (UTC)[reply]
Yes. As I said a long time ago, I wasn't really maintaining activity onwiki or paying attention to this page. I've decided to come back to wikipedia at some point in the near future (within 1-2 months) but I can add the features to the project. Thanks for reaching out. Capsulecap (talk • contribs) 22:27, 5 January 2024 (UTC)[reply]
@Primefac: I've added support for that ranks attribute, as is seen in the 5 most recent bot edits, all of which are to my test pages. I've also created support for converting preexisting top25 templates of the alternate form into regular form top 25s, and made it so top 25 report templates longer than 800 bytes are collapsed. I will test the bot once the report is switched to the current week, which should be on Saturday or Sunday. Capsulecap (talk • contribs) 01:34, 6 January 2024 (UTC)[reply]
@Primefac: Status: Current run had too many problems to use as a final trial. Fixed 2 bugs here. Should be ready to be released any week now. Capsulecap (talk • contribs) 03:27, 15 January 2024 (UTC)[reply]
@Primefac: Status: Apparently what I said above was wrong. The bot works fine on the average page but I forgot to put the ranks parameter on new templates. That has since been fixed. However, there are several larger problems that I discovered on this run — the bot broke another bot's template that was split over two lines (how rude of it!), and the entries in the report were changed from the first entry, necessitating me to delete templates from two talk pages. For some reason it also ignored the page Franz Beckenbauer in the first revision, but that shouldn't be a problem once I manage to switch over to using the newest revision with tables. Also created a list at User:CapsuleBot/Todo. Capsulecap (talk • contribs) 02:36, 20 January 2024 (UTC)[reply]
Function details: Go through Category:Unassessed articles (only deals with articles already tagged as belonging to a project). If an unassessed article is rated as a stub by ORES, tag the article as a stub. Example
Discussion
Note: This bot appears to have edited since this BRFA was filed. Bots may not edit outside their own or their operator's userspace unless approved or approved for trial. AnomieBOT⚡ 00:10, 28 March 2023 (UTC)[reply]
^. Also, may potentially be a CONTEXTBOT; see Wikipedia:Stub: There is no set size at which an article stops being a stub.EpicPupper (talk) 23:04, 30 March 2023 (UTC)[reply]
The Bot run only affects unassessed articles rated as stubs by mw:ORES. The ORES ratings for stubs are very reliable (some false negatives – which wouldn't be touched under this proposal – but no false positives). Hawkeye7(discuss) 00:03, 31 March 2023 (UTC)[reply]
Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Sounds reasonable as ORES is usually good for assessing stub articles as such. – SD0001 (talk) 11:41, 1 April 2023 (UTC)[reply]
Y Bot run with 50 edits. No problems reported. Diffs: [4]. Hawkeye7(discuss) 00:42, 18 April 2023 (UTC)[reply]
Comment: Some behavior I found interesting is that the bot is reverting start-class classifications already assigned by a human editor, and overriding those with stub-class. [5] and [6]EggRoll97(talk) 03:28, 18 May 2023 (UTC)[reply]
This should not be happening. Frostly (talk) 03:58, 18 May 2023 (UTC)[reply]
The question is: what should be happening? The article were flagged because some of the projects were not assessed. Should the Bot (1) assess the unassessed ones as stubs and ignore the assessed ones or (2) align the unassessed ones with the ones that are assessed? Hawkeye7(discuss) 04:21, 18 May 2023 (UTC)[reply]
Per recent consensus assessments should be for an entire article, not per WikiProject. The bot should amend the template to use the article wide code. If several projects have different assessments for an article it should leave it alone. Frostly (talk) 05:03, 18 May 2023 (UTC)[reply]
@Hawkeye7: Courtesy ping, I've manually fixed up the edits where the bot replaced an assessment by a human editor. 6 edits total to be fixed out of 52 total edits. EggRoll97(talk) 07:16, 18 May 2023 (UTC)[reply]
((BAG assistance needed)) This has been waiting for over 2 months since the end of the trial, and over 4 months since the creation of the request. Given the concerns expressed that the bot operator has since fixed, an extended trial may be a good idea here. EggRoll97(talk) 05:19, 8 August 2023 (UTC)[reply]
My apologies. I have been very busy. Should I run the new Bot again with a few more edits? Hawkeye7(discuss) 18:57, 15 October 2023 (UTC)[reply]
Approved for extended trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. – SD0001 (talk) 19:10, 15 October 2023 (UTC)[reply]
Yes. I wrote the bot using my C# API, and due to a necessary upgrade here, my dotnet environment got ahead of the one on the grid. I could neither build locally and run on the grid nor on build on the grid. (I could have run the trial locally but would not have been able to deploy to production.) There is currently a push to move bots onto Kubernetes containers, but there was no dotnet build pack available. The heroes on Toolforge have now provided one for dotnet, and I will be testing it when I return from vacation next week. If all goes well I will finally be able to deploy the bot and run the trial at last. See phab:T311466 for details. Hawkeye7(discuss) 22:54, 31 December 2023 (UTC)[reply]
A user has requested the attention of the operator. Once the operator has seen this message and replied, please deactivate this tag. (user notified) Primefac (talk) 20:10, 18 February 2024 (UTC)[reply]
Work was done in January and some changes made on Toolforge. Will resume the trial run when I get a chance. Hawkeye7(discuss) 23:33, 18 February 2024 (UTC)[reply]
Function details: Using the AmputatorBot API, replaces AMP links with canonical equivalents. This task runs on all pages with citation templates which have URL parameters (e.g. ((cite news)), ((cite web)), etc).
Discussion
Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete.Primefac (talk) 10:27, 8 March 2023 (UTC)[reply]
Just noting that I'm working on this but it may take some time. EpicPupper (talk) 23:01, 30 March 2023 (UTC)[reply]
Been a bit busy IRL, but will get to this soon. Frostly (talk) 20:33, 25 June 2023 (UTC)[reply]
Primefac, my apologies, this flew off my radar. I'll work on setting up the bot on Toolforge this month and should have the results soon. — Frostly (talk) 06:13, 5 January 2024 (UTC)[reply]
I've made progress on uploading to Toolforge; just fixing a few bugs. — Frostly (talk) 03:36, 4 February 2024 (UTC)[reply]
There's been a few issues with the API that have been difficult to tackle; I think pivoting to self-hosting the API backend on Toolforge is a good solution (working on that this month). — Frostly (talk) 22:41, 21 February 2024 (UTC)[reply]
Function details: ((Wikisource author)) recently was updated to allow for a |lang= parameter to link directly to non-English versions of wikisource for an author. A similar template, ((Wikisourcelang)), links to a generic search on said language wiki for said author. This task will change ((Wikisourcelang|<lang>|otherstuff)) into a ((Wikisource author|lang=<lang>|otherstuff)) call.
Discussion
((BAG assistance needed)) valid request not attended by any BAG members for almost two months. —usernamekiran (talk) 23:02, 29 June 2023 (UTC)[reply]
Approved for trial (25 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete.
This seems pretty straightforward. Let's go to trial.
Trial complete.Edits. As a note, I did not run genfixes just to make the proposed change more obvious, but if this task does proceed I will be running genfixes alongside them.
Piotrus, I think this request is a little more convoluted than initially requested. Languages such as de do not use an "author" prefix (see e.g. Adolph Friedrich Johann Riedel and his corresponding page on de Wikisource), but I can't figure out which languages it holds to. I am not necessarily seeing a specific pattern between what languages do and do not. My thoughts are of two possibilities - run this task only for languages where the proposed change has the intended effect, or just scrap this BRFA and do these changes manually. Primefac (talk) 12:52, 4 July 2023 (UTC)[reply]
@Primefac I think we can run it for some languages that we can determine now, it shouldn't be that hard as long as it is consistent for each language (ex. German never uses, Polish always uses it, etc.). We could create a list for all languages that wikisource exists on, or just run it for now for some editions that are the biggest (ex. the ones with interwikis here). I did some checks and it seems it's pretty consisten - just a wikisource naming convention. Note that depending on the language, the "author" prefix is different - Polish is "autor". Swedish is "Författare", etc. In the end, what we need to fix is not the outgoing links but the text on our side. Consider this case, similar to the German one you quote, where we improved the language or our template but messed the link: before, diff, after. Since the links work, can we just figure out the way to change the wording in the template but retain the same link as before? The older template was able to do it, somehow, seems we are introducing a new error somehow? Piotr Konieczny aka Prokonsul Piotrus| reply here 04:48, 5 July 2023 (UTC)[reply]
If you wouldn't mind making a list of which languages use the Author (in whatever language) prefix, I can hard-code their use into the template so that there isn't any issue.
This wasn't a problem before because ((wikisource author)) only linked to to the English version so no translation or odd coding was necessary. As mentioned in the original discussion, ((wikisource lang)) just links to a general search (which does sometimes turn up the author page directly) and thus does not require the "Author:" prefix. Primefac (talk) 08:06, 5 July 2023 (UTC)[reply]
On hold. Just for now, while we deal with actual template issues. Primefac (talk) 08:31, 5 July 2023 (UTC)[reply]
@Primefac See talk, is this helpful? Those are most larger Wiki source projects, should be enough to get most of our stuff sorted out. We can take a look at what, if anything, is left after dealing with those languages? Piotr Konieczny aka Prokonsul Piotrus| reply here 07:06, 7 July 2023 (UTC)[reply]
Should do, thanks for that. Going to keep this on hold for a bit longer, there's a TFD for merging all of these together and I might be able to enact these proposed changes during the merge process. Primefac (talk) 08:13, 7 July 2023 (UTC)[reply]
Somewhat stalled, been rather busy myself and it doesn't look like anyone has started work on the template merger. I think I might have cleared my on-wiki plate somewhat (touch wood) so I'll see about prioritising the merger. Primefac (talk) 10:51, 22 September 2023 (UTC)[reply]
Approved requests
Bots that have been approved for operations after a successful BRFA will be listed here for informational purposes. No other approval action is required for these bots. Recently approved requests can be found here (edit), while old requests can be found in the archives.
Bots that have been denied for operations will be listed here for informational purposes for at least 7 days before being archived. No other action is required for these bots. Older requests can be found in the Archive.
These requests have either expired, as information required by the operator was not provided, or been withdrawn. These tasks are not authorized to run, but such lack of authorization does not necessarily follow from a finding as to merit. A bot that, having been approved for testing, was not tested by an editor, or one for which the results of testing were not posted, for example, would appear here. Bot requests should not be placed here if there is an active discussion ongoing above. Operators whose requests have expired may reactivate their requests at any time. The following list shows recent requests (if any) that have expired, listed here for informational purposes for at least 7 days before being archived. Older requests can be found in the respective archives: Expired, Withdrawn.
