G13 soon sorting is only set to update twice a week (I'm not even sure if that page is useful given the sheer size), the one seems to have run today. There have been way too many 500 errors from ORES servers from the past few days and I don't know what's causing them. – SD0001 (talk) 12:28, 27 September 2020 (UTC)[reply]
Thanks for the information, SD0001. I was just told about these pages by MusikAnimal because I couldn't understand why Wikipedia:Database reports/Stale drafts has been coming up empty for the past few weeks when it used to list hundreds of stale drafts every morning about 1 am (my time zone). Since your bot has been active, editors have been preemptively going in and deleting drafts when they reach the six month period, without waiting for the database report each morning. So, I've been doing it as well for the past 2 days.
It's confusing that there are now 4 or 5 pages/categories that list stale (or soon-to-be-stale) drafts and none of them are entirely complete as far as I can see. Since I've been looking at the lists for the past two days, I found that none of them list the exact same stale drafts and other editors keep tagging drafts that are older than 6 months that aren't on any of the lists! I guess it's because there are so many thousands of drafts at this point and user sandboxes, too. It's amazing that they can even categorize them all. Thank you again for cluing me in on the updating frequency. LizRead!Talk!20:39, 27 September 2020 (UTC)[reply]
I've been meaning to take a look again at these reports and make them more useful for people seeking to rescue worthy stuff. I've been told User:SDZeroBot/G13 Watch has been useful because of the excerpts. The plan is to add excerpts on both these reports, and reduce the scope of G13 soon to only include pages whose last edit date falls in a period of one day which is about (6 months minus 1 week) before current. – SD0001 (talk) 07:29, 28 September 2020 (UTC)[reply]
And I now notice here I wrote the same comment on this talk page and your user page! My apologies for being repetitive.
Since you are checking this bot's talk page, we've started this conversation and I've been looking through stale drafts recently, maybe I'll be more specific about what my comment above refers to: Why does Category:AfC G13 eligible soon submissions and User:SDZeroBot/G13 eligible sorting have different stale draft contents? Over the past couple of days, I keep finding G13 eligible pages that your bot has found that aren't listed in the G13 category. Are you drawing from different sources of data? It seems like the two pages would have similar content, with the category being larger and this specific bot page just listing drafts that had just expired from their six month period. Thanks for any insight you can offer. LizRead!Talk!00:49, 30 September 2020 (UTC)[reply]
Oh, I hope this will be considered "constructive criticism" but I have a comment about User:SDZeroBot/G13 eligible sorting. In the past hour, it just updated and posted eligible G13 stale drafts but it includes stale drafts that were "saved" during the day by an editor who made an edit to the page. So, instead of the last edit on some of the drafts being March 29, 2020, it is September 29, 2020. This is only happening to a few of them. But they still all get listed as eligible so I'm not sure what the "cutoff" time is for the bot to check their editing history. This is another reason not to use batch delete because it's necessary to check the edit history of each draft listed to see if it has been edited today.
Any way, sorry for all of the feedback...I usually work with categories but en.wiki has had a big lag time and the database report for empty categories hasn't been updated in a few days so I've been looking at drafts instead. LizRead!Talk!01:03, 30 September 2020 (UTC)[reply]
Liz, the AfC categories don't contain any drafts which don't have any AFC templates on them (which is technically natural since some template needs to be there that can apply the category). Anyway, a couple of days back I set up User:SDZeroBot/G13 eligible which IMO is a better format. I see now that User:SDZeroBot/G13 eligible sorting has stopped updating – oops, that was inadvertent, sorry if anyone was still looking at it. But I went ahead and stopped it for real to reduce the redundancy we already have (category, DB report, etc).
it just updated and posted eligible G13 stale drafts but it includes stale drafts that were "saved" during the day by an editor That must have been due to database replag. I amended the code so that it no longer relies solely on the database (which may be lagging), so this should not recur. Thanks for pointing that out. Any way, sorry for all of the feedback lol! I actually love to hear feedback. – SD0001 (talk) 11:25, 4 October 2020 (UTC)[reply]
The earlier version of User:SDZeroBot/G13 soon, where you could sort by date, was more useful to me than ordering them on the list by quality...I'm not even sure how you or a bot can quantify/judge draft quality, that is really quite impressive. The quality aspect is more useful to editors working in the AFC area, rather than my work which is deleting expired or inappropriate content (vandalism, duplicates, tests, spam). I hope editors in AFC are aware of your wonderful new pages and that it helps them prioritize their limited time to review drafts. What you've put together is really remarkable and I wish I understood more of the intelligence behind the topic sorting. LizRead!Talk!00:33, 6 October 2020 (UTC)[reply]
Doesn't the header of the page basically tell how it's done? It's nothing fancy – put the ones tagged as {{promising draft}} at the top, ones which are rejected, too short, declined as "blank" or as "test" at the bottom. Among the remaining, the sourced ones are put above the unsourced ones. Further tie-breakers include the ORES quality prediction (which uses machine learning, but it's an imperfect model that turns up lot of weird results), and then the page size (bigger ones come higher up). – SD0001 (talk) 12:07, 15 October 2020 (UTC) Now that I think of it, I can think of another metric which could be used in the future – the proportion of unreliable/deprecated/blacklisted sources used on the page, which can be detected using the way User:Headbomb/unreliable does it. – SD0001 (talk) 12:11, 15 October 2020 (UTC)[reply]
Suggestion
Hi, SD0001,
On User:SDZeroBot/G13 soon sorting page, could you make a list for the parent categories like Culture and Geography/Regions like you have for the STEM category (that includes all of the STEM subcategories)? It makes it much easier to browse when there is one large, parent category that contains all of the subcategories in it.
I realize that this is opposite to your original purpose of sorting G13s by topic but I'm using the page to look for expiring stale drafts in a more general way. If you don't want to incorporate this change, I understand. All of the new SDZeroBot pages are something that we are all getting used to. LizRead!Talk!16:47, 5 October 2020 (UTC)[reply]
Well the page is already more than 1MB in size, duplicating the entries in the parent categories would turn up the size over 2MB, essentially rendering the page unnavigable. The STEM category does not include the list items from STEM subcategories. – SD0001 (talk) 12:00, 15 October 2020 (UTC)[reply]
Another comment, I'm not sure you'll see, is there is a problem with User:SDZeroBot/G13 Watch. For example, it posts at 00:00 UTC on Oct. 11, drafts that expire on Oct. 11th. But by posting the drafts that expire at the end of the day that they expire, the drafts posted had been deleted when it was posted. So, it just contained a list of links to already deleted drafts.
Yes that's what G13 Watch is supposed to do. It is list of pages which have already been nominated for G13; used by a few folks like Calliopejen. (The way it's generated is quite different from the other pages – it's a bit unstable and prone to failures as seems to have happened today). – SD0001 (talk) 12:00, 15 October 2020 (UTC)[reply]
Repeated "draft moved from mainspace" edits
Something going on at Draft:Bamtang Games, the bot has thrice edited the date today. The first and third edits were wrong, as that's the date it was moved from draft to main. The 2nd edit in the middle was correct with the date I moved it from mainspace to draft. -- ferret (talk) 00:20, 20 October 2020 (UTC)[reply]
FWIW it was moved from mainspace to draftspace both in July and August. The repeated edits were because the bot first processed the July moves for trial. Later when approved for full run, it processed the August moves then the July moves. It was set up to replace any existing tag because there existed a lot of wrongly dated tags added by another editor. In any case, this task finished many days back so obviously this won't happen again. – SD0001 (talk) 06:55, 26 October 2020 (UTC)[reply]
further, it seems to be randomly tagging pages without any justification causing the timer to reset. See the following example, [1] supplied by 2pou (talk·contribs)/ It's several days later now. If it is continuing to do this, I shall block the bot. This is of importance to those few of us who work on the articles in this particular situation, which is essential part of the AfC process. DGG ( talk ) 00:58, 26 October 2020 (UTC)[reply]
but the approved task is having unfortunate side effects, in destroying the usefulness of Category:AfC G13 eligible soon submissions, a critical category used by several active reviewers. I'm blocking the bot pending further discussion. .Can tthe status of this cateogry before the task run be restored, or is the information lost? DGG ( talk ) 05:13, 26 October 2020 (UTC)[reply]
I think I understand,Before I unblock, I would like assurances that future entries into that category will proceed as before. If not , it would seem necessary to rework and reorganize the structures here in consultation with all those who would be affected . � DGG ( talk ) 17:47, 26 October 2020 (UTC)[reply]
Okay – I didn't anticipate the same person both expanding an article and nominating for deletion. Let me see what I can do. – SD0001 (talk) 04:58, 18 November 2020 (UTC)[reply]
I'm not sure rare anomalies of this sort are worth the coding. I think this particular one happens about 2 or 3 times a year. Anyway, as I see it, the notice confirms the system has recognized the request DGG ( talk ) 07:03, 18 November 2020 (UTC)[reply]
Delighted to learn that SDZeroBot has won the m:Coolest Tool Award in the Newcomer tools category! I don't know much about how this award is organised or decided, but thanks to everyone who recommended it or voted for it. – SD0001 (talk) 09:32, 12 December 2020 (UTC)[reply]
I didn't know about it before the CTAs. Hopefully this will help it reach more people, because it's such a great tool. --Zack (talk) 14:53, 17 December 2020 (UTC)[reply]
Having a sort category for Culture/Food and drink lets me find AfCs that are right in my wheelhouse, that I can often assess almost from my own knowledge base. Thanks for this! —valereee (talk) 16:09, 17 August 2021 (UTC)[reply]
SDZeroBot: AfD notifications failure - fixed
SDZeroBot's task "AfD notifications" failed to run per the configuration specified at Wikipedia:Bot activity monitor/Configurations. Detected only 0 edits in the last 24 hours, whereas at least 1 was expected. If/when the issue is fixed, please change the section title (e.g. append " - Fixed") or remove this section completely. When that is done, this notice will be reposted if the bot task is still broken or is re-broken. If your bot is behaving as expected, then you may want to modify the activity configuration instead. Or to unsubscribe from bot failure notifications, remove the |notify= parameter from the {{/task}} template. Thanks! – SDZeroBot (talk) 01:20, 24 August 2021 (UTC)[reply]
Once again, SDZeroBot has a hard time handling months that have a different number of days. Tonight, User:SDZeroBot/G13 eligible, which usually has 0 titles listed, has over 400 drafts it identifies as "eligible" for deletion. Except that they are not eligible, they'll be eligible on September 1st or September 2nd. But SDZeroBot thinks that every month has the same number of days. Because February had 28 days but August has 31 days, it is identifying drafts as being eligible before they are because, I guess, it thinks that August 30 is actually September 2.
This is less of a problem now than it is in February when there are fewer days and the bot just skips daily reports for three days in order to catch up. Then, DGG, who regularly scans the report to rescue promising drafts, is frustrated because SDZeroBot just jumps three days ahead and omits those days so he misses three days worth of draft reports and the drafts are simply deleted by patrolling admins.
This is a regular problem and has been since the bot's operation started a year ago. Since most months have 30 or 31 days, editors and admins who work with the G13 Eligible Soon list just adjust to the bot's quirks. But when there is a three day difference, this becomes a larger problem. I hate to nag you about this, because I'm so grateful for the work that the bot does, but you really need to find a way to fix this aberration, maybe by asking other technically-minded editors and admins who work with bots. If any of them have this talk page on their Watchlist and have any ideas for solutions, it would be greatly appreciated before we get to the end of September and this happens again. Thank you. LizRead!Talk!01:18, 31 August 2021 (UTC)[reply]
This seems to be a new issue. If I recall correctly, we were only having calendar-related issues with User:SDZeroBot/G13 soon. G13 eligible is more simple in that it just lists anything that's >6 months old. It doesn't assume anything about the number of days in a month (beyond what the programming language internally does).The issue appears to stem from JavaScript's datetime computation: > new Morebits.date() Morebits.date {_d: Tue Aug 31 2021 16:37:54 GMT+0530 (India Standard Time)}> new Morebits.date().subtract(6, 'months') Morebits.date {_d: Wed Mar 03 2021 16:38:08 GMT+0530 (India Standard Time)} // ???The 2nd result is bizarre.Thankfully, in SQL the date math makes more sense (quarry:query/58105):> SELECT NOW() 2021-08-31T11:13:31> SELECT NOW() - INTERVAL 6 MONTH 2021-02-28T11:13:46I'll shift the six-month-subtraction from the JS part to the SQL part. Thanks for the bug report!Regarding the date issues with G13 soon, I'll see if I can take a shot at them tonight. – SD0001 (talk) 11:17, 31 August 2021 (UTC)[reply]
Well, Explicit and I primarily work with the G13 Soon page although I check the G13 Eligible page daily in case we miss anything. It looks like SDZeroBot stopped reports yesterday and then started up again today. We already had a G13 Soon report for March 9th and then it started up today with a report for March 7th. It was actually nice to be a few days ahead, I know DGG probably appreciated this since he reviews the report ahead of the six month expiration date. Right now though, SDZeroBot has issued 3 reports for March 7th, this one on 8/28, and two today that look identical. Should we use the latest one because you say the dates were off by a day? Just ignore the previous March 7, March 8 and March 9 reports and then the bot will resume with a new March 8th report today?
I appreciate your speedy response to my query. The bot is reliable 99% of the time so when there is a glitch, it stands out. Thanks again! LizRead!Talk!23:12, 1 September 2021 (UTC)[reply]
There is something odd about that March 7th G13 Soon report, there are typically about ~200 drafts that expire daily and that list is over 400 drafts! Maybe there was a lot of AFC activity and drafts reviewed that particular day but it sure stands out. I rarely have seen a list with 400+ drafts, maybe once or twice before. LizRead!Talk!23:19, 1 September 2021 (UTC)[reply]
We already had a G13 Soon report for March 9th and then it started up today with a report for March 7th That again arises from the date math. From now on, this will not occur - rather, the bot will skip create multiple reports in a day or none at all, to keep pace with the calendar. That is, there would no report on next year's Sep 1 and Sep 2, and multiple reports at the end of Feb / start of March. So that there is exactly one report for each calendar day the drafts were last edited. Should we use the latest one because you say the dates were off by a day? They should all be correct, though yeah the latest one will have less entries as it skips the ones that have been edited in the last few days. – SD0001 (talk) 04:32, 2 September 2021 (UTC)[reply]
There are two reason I work ahead of the date.one is so I have a chance for some life in the intervals, or if not life outside WP, at least life for somthing else in WP, by taking a day or two away from the list; the other is that I only work with articles that have at least some possibilities, and they need at least a little thinking. I can delete at the rate of 2 or 3 a minute as fast as the script will run, but I cannot think at that speed. I try to catch two things for articles in fields I can handle--ones that were misreviewed and actually show or could easily show notability , and drafts which were never even looked at. Most of those are because they seem boring, but some are because they seem difficult. I hope now that we have temporarily caught up with the afc backlog, that 5 months from now, there will start being many fewer of these. (fwiw, as others have also found, reviewing quality during the rapid catchup drive seems to be the same quality as ordinary. I expected otherwise, and am vey glad my prediction was wrong.). DGG ( talk ) 06:02, 3 September 2021 (UTC)[reply]
It is extremely misleading for a bot to rate an article GA (even though such ratings are inadmissible), when that is a rating always decided by the community at the GAN process. I suggest it be removed entirely, possibly replaced by A class ratings. Aza24 (talk) 05:11, 28 December 2021 (UTC)[reply]
I suppose it would make sense to add a * mark to GA and FA ratings for clarification. The bot merely reports back whatever ratings it gets from ORES so I don't think removing it or replacing it with A-class would be appropriate. In any case, these are only used on WP:NPPSORT and WP:AFCSORT; the people who review those pages would know better than to take the predictions at face value. – SD0001 (talk) 10:54, 10 January 2022 (UTC)[reply]
It respects {{nobots}}, but not bots|deny=all and other variations. Will try to add the support for those when I get time. (I see you posted on GitHub as well - if you can file a pull request that'd be great!) – SD0001 (talk) 10:53, 10 January 2022 (UTC)[reply]
AfD notices when editor already notified
I don't know how common my approach is, but when I send something to AfD, I would use the standard template notice only for relatively new editors. If it's a regular with experience in the area, I'd just leave a short note about the AfD, I reckon that's less impersonal. Even with new editors, it's sometimes better to go for a handwritten note, like here, where the AfD is not about deletion at all, but about redirecting or merging. Well, it turns out that instead of making things a little bit better, my approach makes them worse, because the bot will invariably come along and slap on the template anyway. – Uanfala (talk)01:41, 10 January 2022 (UTC)[reply]
Hmm. I suppose it would be better for the bot to check if the talk page has a link to the AFD. It currently checks for text that looks like the standard section header for AfD notifications. So until the change is done in the bot, the nasty trick would be to place <!-- ==Nomination of [[Languages of the Indian subcontinent]] for deletion== --> on the talk page along with your message. – SD0001 (talk) 11:00, 10 January 2022 (UTC)[reply]
Ah I see. In that case, the QPQ check tool needs to be swapped with the DYK nominations count tool (or something similar) in the DYK toolbox. QPQs are now needed for a total of 5+ nominations made by an individual user instead of a total of 5+ DYK credits by an individual user. MrLinkinPark333 (talk) 20:21, 26 May 2022 (UTC)[reply]
SDZeroBot's task "AfD notifications" failed to run per the configuration specified at Wikipedia:Bot activity monitor/Configurations. Detected only 0 edits in the last 24 hours, whereas at least 1 was expected. If/when the issue is fixed, please change the section title (e.g. append " - Fixed") or remove this section completely. When that is done, this notice will be reposted if the bot task is still broken or is re-broken. If your bot is behaving as expected, then you may want to modify the task configuration instead. Or to unsubscribe from bot failure notifications, remove the |notify= parameter from the {{/task}} template. Thanks! – SDZeroBot (talk) 01:20, 17 June 2022 (UTC)[reply]
Good Article counter order
Wouldn't it make more sense to have the Good Article counter
list the count in reverse count. In my case I have 218 Good Articles from GANs, so it would be more logical to list 218
on top of the list with the most recent Good Article received, for example John Johnson (inventor) (2022-06-17).
The first Good Article I received of Conclusion of the American Civil War (2009-03-08) should be as number 1 (not # 218).
Can the counter be programmed to facilitate this counter number in a reverse numbering? Number 1 logically should be at the bottom and the highest number (i.e. #218) at the top of the list with the most recent Good Article received. How does that sound? --Doug Coldwell (talk) 17:46, 24 June 2022 (UTC)[reply]
SD0001, if this is not possible, could it be extended at least a bit further? I have proposed a bot that would need this information to function well. I was hoping this would be as simple as removing the current limit, which appears to be 500 rows. Mike Christie (talk - contribs - library) 13:25, 3 October 2022 (UTC)[reply]
@Mike Christie There are 4748 users who have written GAs – listing them all would make the page too big. SDZeroBot maintains the public database s54328__goodarticles_p on toolsdb (and updates the on-wiki list daily by querying it) – which does contain nominator data of all GAs. If your bot is hosted on Toolforge, you can also query this database directly instead of parsing the wiki page. – SD0001 (talk) 16:34, 3 October 2022 (UTC)[reply]
Thanks for the pointer to the database -- that's clearly the best option for me. I am a bit surprised that 4748 is too many records for a daily refresh; WP:GAN has about 1,000 lines and can get refreshed every twenty minutes. Anyway, thanks for the help. Mike Christie (talk - contribs - library) 16:58, 3 October 2022 (UTC)[reply]
Hi -- I'm working on a replacement for the GAN functions of Legobot (BRFA here) and I just noticed this page. Do you construct it by scanning the GA nominee category, or by reading WP:GAN? If the latter, would you be able to test if it works with User:ChristieBot/GAN existing format, which, if the bot is approved, will be the new contents of GAN? It's very similar to the existing format but it's not quite identical and I wanted to check that the change, if it happens, wouldn't break your code. If you build it by scanning the category, there's nothing to worry about, of course. Thanks. Mike Christie (talk - contribs - library) 14:07, 4 November 2022 (UTC)[reply]
Thank you for the very useful {{Database report}}. I'm working through my old Quarry queries which detect various anti-patterns in Wikipedia pages, and setting up those which are still relevant to run weekly. Everything is running smoothly and I've already fixed plenty of articles as a result. One suggestion: would it be sensible and easy to add a little technical information such as query duration? I'll be using this to tweak the run frequency (e.g. change quick jobs to daily and expensive ones to monthly). I can get a good estimate from Quarry, but it might highlight SQL inefficiencies for everyone's benefit. The time of last run (as in date+UTC rather than duration) might also be convenient, though it's easily obtained from page history. Thanks again, Certes (talk) 12:04, 19 November 2022 (UTC)[reply]
@Certes Done – the bot will now write query_runtime and last_updated params to its {{database report/footer}} call. Feel free to edit that template if you want to tweak the formatting. (It would have been good to put the timestamp just below the update button, but that area is created by the main template whose params are harder for bot to modify.) – SD0001 (talk) 05:53, 20 November 2022 (UTC)[reply]
The database report template kicks ass
@SD0001: This is something I have craved for a while. I am so glad you've made it! It works even better than if I'd made it myself. A couple questions:
Is it possible to specify a footer? That is, a control sequence that you can put at the end of a page, and have it not update past there? On something like Template:Signpost/Number of issues I have wrapped the whole thing in <onlyinclude> tags, so that transcluding the page just gives the number as the output, but this required some clever jiggery-pokery with the row template.
Is there a way to specify the offset from the interval? For example, if you want it to be every week, but indexed to the beginning of the month.
How often should I be running them? At Template:Signpost/Number_of_articles, for example, the number will only change once every couple of weeks when we publish... but I have it set to just run every day (because I don't want to have to manually run the thing every time I run the already-onerous publishing script). I don't want to put undue strain on the bot, though: is this an actual concern, or should I just feel free to set the interval to 1 by default? jp×g21:47, 27 January 2023 (UTC)[reply]
Would it be useful to have the report written to a subpage of the page where configuration exists? That way, it can be directly transcluded. This is already possible per this thread, though not documented!
Not really, but the check on whether to update is based on SDZeroBot's last edit to the page. So if you web-trigger an update on 1st of the month, then with |interval=7, subsequent updates will take place on 8th, 15th, and so on. Though I wouldn't be opposed to having a cron-based schedule option (without support for hours and minutes as the bot processes periodic updates just once a day), eg. |cron=*/7 * *.
Since the query runtime there is just 2-3 seconds, it's totally fine. Only issue would be the edit history pollution.
In the first two cases, the GA nominee template is in a section and not on the top of the talk page as per guidelines. In the third one, {{GAnominee}} is used instead of {{GA nominee}} – this can be easily handled in the bot's code so I've pushed up a fix. – SD0001 (talk) 06:46, 12 March 2023 (UTC)[reply]
Nomation of Abhi the Nomad for deletion
Personally, I don't think this page should be deleted. The subject has a relatively large internet presence and has millions of streams on some of his songs. While not all of his projects are particularly notable, I think it is important that this page remain entact. SaltieChips (talk) 13:35, 31 March 2023 (UTC)[reply]
G13 soon sorting not running?
Hello - I find the "SDZeroBot/G13 soon sorting" information quite helpful, but the information has not been updated since 27 March 2023. I checked the 'err' link, but it is blank. Is there a way to start this service running again? Thanks. DaffodilOcean (talk) 16:27, 9 April 2023 (UTC)[reply]
There seems to be a problem with several bots today and that includes SDZeroBot. Is this a system issue? I also ran into this with HaleBot. Thanks for any help you can provide. LizRead!Talk!01:32, 19 April 2023 (UTC)[reply]
This same problem happened with HaleBot and so bot operator Legoktm moved the bot over to a different "platform" because there is a problem with the "grid" (?). And now HaleBot issued its report. Maybe this would help with SDZeroBot. LizRead!Talk!04:05, 19 April 2023 (UTC)[reply]
This should be fixed now. I was in the middle of migrating away from the Grid when the supposed issue with grid occurred. Not sure if that was actually responsible. All SDZeroBot jobs now run on Kubernetes. – SD0001 (talk) 03:58, 21 April 2023 (UTC)[reply]
For some reason, SDZeroBot didn't issue its 00:00 UTC CSD G13 reports tonight. The ones I look at are User:SDZeroBot/G13 eligible and User:SDZeroBot/G13 soon. It's been active today. so the bot isn't down, it just seems to be having some issues with expiring draft reports. Thanks in advance for any solutions you can offer! LizRead!Talk!00:20, 7 May 2023 (UTC)[reply]
It seems to be a transient issue with the new kubernetes platform. The job just failed to even start up at the scheduled time - no error and no failure email. – SD0001 (talk) 12:04, 23 May 2023 (UTC)[reply]
Same old issue. Two days without reports. I figured that SDZeroBot would skip one G13 soon report because November had 30 days while May has 31 but now it's skipped two days. I'll add that I'm also seeing issues with other bots whose reports I rely on like AnomieBOT III, HaleBot and others. Things haven't been the same since the replag on s1. Even when it got "caught up", we never saw the backlog of pages/articles that should exist after 5 days with no reports. I'm not sure that everything was restored if they just patched up a solution. Luckily, with SDZeroBot, we have a 7 day lead time so missing reports aren't a dire situation until next week but I thought I'd let you know. Many thanks! LizRead!Talk!00:18, 2 June 2023 (UTC)[reply]
@Liz This isn't related to the replag or the the end-of-the-month disparity, but the same issue from 22 May - the new kubernetes platform just didn't start up the job! I have opened phab:T338006 regarding this. – SD0001 (talk) 06:25, 2 June 2023 (UTC)[reply]
Oh, okay, well, thank you for looking into this. It's odd that it would only affect some reports and not others. Thanks again. LizRead!Talk!20:42, 2 June 2023 (UTC)[reply]
Three days now without a G13 soon or G13 eligible report. I'm starting to get a little nervous. Though if I remember rightly, when the bot kicks in, it issues reports to catch up. Fingers crossed. LizRead!Talk!00:10, 3 June 2023 (UTC)[reply]
Okay, it's been 4 days now. We are working on the May 4th report and the last report that was issued was for May 7th. I made a comment on the phab ticket but it doesn't look like it is getting much response. Do you have any suggestions of what we could do without the report? There used to be a Stale Drafts Database report but I think that was discontinued after SDZeroBot's excellent work started. There is a category for G13 expiring drafts but, for some reason, it only seems to include about 60% of the drafts that appear in SDZeroBot's report. LizRead!Talk!00:18, 4 June 2023 (UTC)[reply]
@Liz I can still run the trigger reports manually - which I have done now. It created 3 reports - one for each day which was missed. – SD0001 (talk) 10:35, 4 June 2023 (UTC)[reply]
Whew! Disaster averted. Thank you very much. Although I have noticed that the G13 lists are shorter than in the past. I guess we have fewer new editors joining the project. Or some editor has swept through the Draft space with AWB correcting commas and we'll end up with 1,000 drafts expiring over one hour sometime during the summer. That's always fun when it happens. ;-) Thanks again for being available and your responsiveness. LizRead!Talk!16:56, 4 June 2023 (UTC)[reply]
The issue is that the platform is too busy at exactly midnight. I have changed the job schedules to run few minutes after midnight instead. – SD0001 (talk) 17:17, 5 June 2023 (UTC)[reply]
Was making use of Wikipedia:Database reports/Long pages/Talk to find long Talk pages that might benefit from archiving, but I had limited success because the report includes subpages, which may include existing archives, RfCs, etc.. Could we get a second, similar report that includes only Talk pages that are top level, not subpages? Thanks. -- Netoholic@06:24, 8 July 2023 (UTC)[reply]
Not easily. The bot doesn't know what section it's editing - there can be multiple reports in the same section, or no sections at all. – SD0001 (talk) 10:17, 28 July 2023 (UTC)[reply]
Yes, the issue is that this AfD includes 119 articles, each of which is supposed to appear on AFDSORT. Worse, most of these articles are sorted under multiple topics, so the nomination text is repeated 315 times. And as the nomination text is very long, AFDSORT going over the 2 MiB limit. – SD0001 (talk) 18:38, 25 August 2023 (UTC)[reply]
Looked into it but it's not clear what the issue is – bot is randomly stopping in the middle of the process with no errors logged. Will take a look again later today. – SD0001 (talk) 07:14, 3 October 2023 (UTC)[reply]
@SD0001: For NPP sorting pages (such as this), would it be possible to separate the creator and creator edit counts into two separate columns, so that reviewers can more easily sort to find more or less experienced authors (which should nominally correlate with more or less challenging reviews)? Thanks, I love your tool(s)! IazygesConsermonorOpus meum17:05, 15 October 2023 (UTC)[reply]
@Iazyges I've made it such that the creator field sorts by the edit count, rather than alphabetically. This is similar to how the class field actually sorts by the class (GA > B > C). As for splitting it into two columns, that would make the table wider - which is more of a problem nowadays as many folks are using vector-2022 which provides lesser horizontal space. – SD0001 (talk) 18:18, 15 October 2023 (UTC)[reply]
SDZeroBot irregularities
Hello SD0001,
When checking Pending AfC submissions and the draft I'm working on, Draft:V (programming language), I noticed it took a large dramatic drop. There were no edits of the draft, so the change seems likely the result of the algorithms used by the bot. The days in question are 27 September 2023 to 28 September 2023. I'm curious as to what could be the reasons or if it's anything to be concerned about.
Additionally, and this might not be under your area, but the "Prior AfD" listed is not to the present V language. What's linked to, is to a different programming language, from 2008. The present one in the draft is from 2019. These different languages, though sharing the same letter, need to be disconnected from each other. Thanks for your work and consideration in advance.Wukuendo (talk) 04:08, 1 November 2023 (UTC)[reply]
GANs tool: Epoch fail
Hi, I just stumbled over your GAN-counting tool and noticed that, at least for my results, the entry for Serious Sam: The First Encounter shows it as having passed GA on 1970-01-01 (0 epoch time) instead of 2023-10-30. I assume that the incorrect date is stored in some database, but I couldn't immediately find the origin. Curiously, I couldn't find comparable entries on the spot-checked results of a few other users. IceWelder [✉] 15:33, 4 February 2024 (UTC)[reply]
As far as I can tell the GA counts don't work for users with apostrophes in their usernames. This retrieves nothing, for example. Looking in the database it appears that the nominator for this record is stored as 'Tim O&'. Mike Christie (talk - contribs - library) 10:37, 14 April 2024 (UTC)[reply]
It's a problem with the signature rather than the username itself. Per diff, the user page link in the signature is written as User:Tim O'Doherty which MediaWiki is able to parse fine, but the bot's title parser discards the text after # thinking it to be a URL fragment. – SD0001 (talk) 13:56, 14 April 2024 (UTC)[reply]
I don't know if this would help, but I could give you a list of the correct usernames for the GAs in question. Or make the historical_GA_reviews table in ganfilter publicly readable so you could use it to fix these? Mike Christie (talk - contribs - library) 14:13, 14 April 2024 (UTC)[reply]
Will do. I can't check from where I am now but I think there were others "where nominator like '%&'" which it would be nice to fix too. Tim was the important one though since he's active so thanks for fixing that. Mike Christie (talk - contribs - library) 17:23, 17 April 2024 (UTC)[reply]
Yeah, it seems everyone with an apostrophe in the username also has a signature containing html entities. I have made a fix for that so html entities are decoded before username parsing. – SD0001 (talk) 18:35, 3 May 2024 (UTC)[reply]
I'm not sure what happened here. None of the drafts on this page showed up on the relevant User:SDZeroBot/G13 soon page for the day they should have. Things are a little off for the past few days as the editors who regularly deal with G13 drafts, me, Explicit and Hey man im josh, have not been available as much as usual. My mom is on hospice care so I work on Wikipedia when she is sleeping, Explicit, for some reason, has been absent from the project and Hey man im josh doesn't edit on the weekends. So, there has been a bit of a backlog on dealing with expiring drafts. But that's okay, they aren't urgent. But I still don't know why these drafts from Nov. 2/3 didn't show up on G13 soon page. Any ideas why they were "hidden" from us until the day after they expired?
There has been a trend though of a declining number of expiring drafts on the daily G13 soon lists (usually about ~150 drafts) but that has boomeranged in May 2024 because of a AFC backlog drive that occurred in Novembere 2023. So we are seeing higher numbers that we were a few weeks ago.
I had noticed that this G13 soon page stated that there were 338 drafts listed at the top of the page but there were clearly weren't that many drafts listed on this page. I often do a search on this page by hour (00:00, 01:00, etc.) and there was no way that 338 pages were included on this page given those numbers which were just 4 or 5 drafts/hour. Can you think of a reason why there would be 338 drafts listed at the top but not that many drafts included on the page and only show up the next day on the G13 eligible page for that date? I should say that I've only seen this kind of glitch happen a few times but it has happened occasionally so I thought I'd ask. LizRead!Talk!05:00, 5 May 2024 (UTC)[reply]
I just noticed that G13 eligible for this day says there are 217 drafts on the report but it only lists 5 drafts even though if you look at the page history for that day it shows a huge amount of content has been added. Why would this content be hidden from view? I see you are editing but haven't responded yet, should I go post on your personal user talk page? Thanks. LizRead!Talk!03:50, 6 May 2024 (UTC)[reply]
I had a busy weekend as I was travelling to attend the Wikimedia Hackathon and just got back. I'll take a look when I have some time. Sorry to hear about your mom, do take care of her. – SD0001 (talk) 13:23, 8 May 2024 (UTC)[reply]
Oh, that sounds like fun, did you have a good time? Despite my lack of programming skills, I attended some hackathons back in 2008-2012 and always enjoyed myself though I wasn't much help to my team.
Turns out that Explicit was only gone for a week so he is back on G13 patrol. But it still seems unusual that the drafts listed on those pages didn't match the counts for the page. Luckily, this doesn't happen often with dependable SDZeroBot but it does make me wonder what went wrong. Thanks for the reply. LizRead!Talk!03:53, 9 May 2024 (UTC)[reply]
If you ignore all of the "road" articles, which are restored drafts, all of the 20+ drafts should have appeared on a User:SDZeroBot/G13 soon list for 6/3 but for some reason, didn't show up on the daily list. I've looked over all of these drafts and they don't seem to have much in common. Do you know what SDZeroBot would leave these drafts off of the G13 soon list? Thanks. LizRead!Talk!01:36, 4 June 2024 (UTC)[reply]
Is it possible to avoid this error message when updating a report?
Please see this error message, which started appearing every once in a while a couple of weeks ago. It was happening all the time for a day or so, and then it mostly fixed itself. Is it possible for the bot to avoid updating the page when it encounters this timeout? It is better to have an old report than a completely missing one. Thanks for anything you can do. – Jonesey95 (talk) 02:29, 4 July 2024 (UTC)[reply]
I actually came here to whine about the same thing. Well, more precisely, to ask where you prefer feature requests go, so I could ask for a {{database report}} parameter to append "(timestamp) (errormessage)" instead of replacing whatever was already there. —Cryptic18:40, 23 July 2024 (UTC)[reply]
It was the unbalanced table tags in this version, which shouldn't happen. Not that I can think of a good way to fix it: the opening table tag is in Wikipedia:Database reports/Uncategorized templates/Header, which - correctly - only gets transcluded on the report page if the query is successful. I haven't experimented much with the header/row templates that {{database report}} supports, so maybe there's a better way to do it.Anyway, the reason this broke WP:DBR is because {{DBR row}} gets its "last updated" column by transcluding the report, with the expectation that the report page has a timestamp stuck inside a <onlyinclude> tag pair. Which is as horrifying as it sounds, but there's really no better way to do it - there's a magic word to get the timestamp of an arbitrary page, but it won't update unless the page it's used on is edited. The template can deal with reports that don't have that onlyinclude'd timestamp, or even most misformattings, but apparently can't handle misnested tables. —Cryptic20:57, 16 September 2024 (UTC)[reply]
Setting skip_table=y and using header_template and footer_template should probably work. The current setup is too complex for me to understand.I think H:LST would be better than the onlyinclude tag pair which is indeed horrific – in case of unexpected input, nothing would transclude with LST rather than everything. Also, using REVISIONTIMESTAMP does register a dependency in templatelinks – so the page it's used on will get purged when the other one is edited. – SD0001 (talk) 17:02, 18 September 2024 (UTC)[reply]
I couldn't get REVISIONTIMESTAMP to update on its own when I experimented with it - purging the page didn't work, and IIRC even null-editing it didn't. Transcluding a section would work better, yes, but migrating to it isn't worth coordinating with all the other bots, some of which are effectively unmaintained. This should keep the main table from breaking again; supporting {{DBR row}}'s foibles isn't worth the effort either. —Cryptic01:41, 19 September 2024 (UTC)[reply]
While pondering some arcane issues, it occurred to me that it might be possible and/or desirable to use database report templates for heavy lifting (e.g. as a foundation for other bots, or even for webapps elsewhere). This raises some vague security issues, e.g. if the page is vandalized the whole thing goes kablooey. Luckily, being a genius, I realized that this is precisely why WP:TPE exists. Would it be possible to give this bot TPE perms such that sensitive DBR pages could be TE-protected and the bot could roam happily? jp×g🗯️01:14, 7 July 2024 (UTC)[reply]
I think the other bots and webapps would do some validation to ensure the data looks as expected. The bot can edit through ECP protection which I think should be enough, especially if the bot-updated data is being transcluded from another hidden-away page. From what I've seen extendedconfirmed editors don't go randomly messing around in project or template namespaces. I'd prefer to see the first such vandalism to occur at ECP so it can be used as a justification for TPE. – SD0001 (talk) 08:00, 7 July 2024 (UTC)[reply]
Xtools error and SDZeroBot notifications
Howdy,
You recently left me a note on my talk page about Armenian–Azerbaijani cultural relations, linking to this Xtools page and saying that I had been a significant contributor. Xtools is, for some reason, attributing 40% of the page's content to me. This representation is incorrect; my contributions were to tag and remove copyvio, but the website appears to simply attribute all content that was revdel'd to the individual who made the next undeleted revision.
I'm not sure if there is anything you can do on your end, but the data for your notification pipeline does appear to have a (likely systematic) error.
Thanks for the note. There's indeed nothing much I can do about this. The data comes from WikiWho API (XTools is just another consumer of WikiWho data), which has a bug tracker here. – SD0001 (talk) 17:45, 21 October 2024 (UTC)[reply]
Suggestion - NPP sorting
A suggestion for improving the usability of the /NPP sorting. Currently, clicking on articles opens them in the same tab, which disrupts the workflow, particularly for those reviewing or managing multiple articles in succession. It would be more user-friendly if these links opened in a new tab by default. - The9ManTalk07:52, 27 October 2024 (UTC)[reply]
PetScan does this in its search results lists. The HTML emitted by the script has <a>...</a> elements with the usual attributes like href=title= etc. but also has the special attribute target="_blank" (see HTML 5.2 spec). --Redrose64 🌹 (talk) 23:28, 27 October 2024 (UTC)[reply]
Yes, that attribute can't be generated by wikilinks or external links. Nor does wikitext support manually using <a> tags. But it's possible to write a user script which appends that attribute on links in specific pages. – SD0001 (talk) 11:22, 28 October 2024 (UTC)[reply]
Hi SD0001. Would you be willing to add Wikipedia:New pages patrol/Header to the top of your bot's NPP sorting pages? I use the sorted pages quite a lot (thanks, by the way!) and would find it helpful to be able to quickly navigate back to the main NPP pages when I'm done with a particular list. – Joe (talk) 08:08, 25 November 2024 (UTC)[reply]
There have some intermittent issues with Toolforge today (phab:T380844) because of which it may have been skipped. I don't think there's any issue specific to that report. – SD0001 (talk) 16:26, 26 November 2024 (UTC)[reply]
SDZeroBot didn't issue its G13 reports tonight. It's published other pages so I'm not sure what the problem is with these ones. Thanks for manually kick things into gear. LizRead!Talk!02:03, 21 December 2024 (UTC)[reply]
Where's the query? This seems like something that should never take more than a few seconds to complete. (Or am I misunderstanding what went wrong?) —Cryptic15:43, 21 December 2024 (UTC)[reply]
Thank you for the update, User:SD0001. This is affecting other bots as well. I make use of GalaxyBot's reports and it just issued one at 18:00 UTC but hasn't produced one since and it issues hourly reports so I don't think the situation is resolved yet. But as long as WMF developers are aware of it and are working on the problem, it's a matter of having patience and reminding them if too much time passes without a solution. LizRead!Talk!21:23, 21 December 2024 (UTC)[reply]
هذه المقالة يتيمة إذ تصل إليها مقالات أخرى قليلة جدًا. فضلًا، ساعد بإضافة وصلة إليها في مقالات متعلقة بها. (يوليو 2023) هذه مقالة غير مراجعة. ينبغي أن يزال هذا القالب بعد أن يراجعها محرر مغاير للذي أنشأها؛ إذا لزم الأمر فيجب أن توسم المقالة بقوالب الصيانة المناسبة. يمكن أيضاً تق
Diócesis de Brno Dioecesis Brunen(sis) (en latín) Escudo de la diócesis Catedral de San Pedro y San PabloInformación generalIglesia católicaIglesia sui iuris latinaRito romanoSufragánea de arquidiócesis de OlomoucFecha de erección 5 de diciembre de 1777 (como diócesis)Bula de erección Inter plurimaSedeCatedral de San Pedro y San PabloCiudad sede BrnoDivisión administrativa región de Moravia MeridionalPaís República Checa República ChecaCuria diocesana Biskupstvi brnenske, ...
47°23′11″N 8°31′44″E / 47.3863169°N 8.5287591°E / 47.3863169; 8.5287591This article needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed.Find sources: Institute for Mathematical Research – news · newspapers · books · scholar · JSTOR (August 2021) (Learn how and when to remove this template message)This a...
dr. H.Harma ZaldiSp.B. FinaCSWakil Wali Kota Bukittinggi ke-3Masa jabatan13 Agustus 2010 – 13 Agustus 2015PendahuluIsmet AmzisPenggantiIrwandi Informasi pribadiLahir12 Desember 1952 (umur 70)Turawan, Tigo Koto, Rambatan, Tanah Datar, Sumatera BaratKebangsaanIndonesiaSuami/istridr. Rahmi Yetti Kamal, Sp.AAnakdr. M. Pramana K.H,Sp.OT.,M.Kes M. Irsyad K.H., S.KedM. Anshar K.H., S.KedAlma materUniversitas AndalasProfesiDokter spesialis bedahSunting kotak info • L ...
Romanian rock band This article needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed.Find sources: Iris Romanian band – news · newspapers · books · scholar · JSTOR (April 2015) (Learn how and when to remove this template message) IrisIris, 2010, the classic lineup, before the breakup into two bands. From left to right: Valter Popa, Ne...
Peta menunjukan lokasi Linamon Linamon adalah munisipalitas yang terletak di provinsi Lanao del Norte, Filipina. Pada tahun 2010, munisipalitas ini memiliki populasi sebesar 16.351 jiwa atau 3.105 rumah tangga. Pembagian wilayah Secara administratif Linamon terbagi menjadi 8 barangay, yaitu: Bosque Larapan Magoong Napo Poblacion Purakan Robocon Samburon Pranala luar Official website of Linamon, Lanao del Norte Diarsipkan 2006-05-22 di Wayback Machine. Philippine Standard Geographic Code Diars...
1969 film by Henry Levin This article is about the 1969 film. For other uses, see Desperado. This article needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed.Find sources: The Desperados – news · newspapers · books · scholar · JSTOR (June 2019) (Learn how and when to remove this template message) The DesperadosDirected byHenry LevinScreen...
This article appears to be slanted towards recent events. Please help improve the article. (March 2017) Bilateral relationsIndo-Sri Lankan relations India Sri Lanka Diplomatic missionHigh Commission of India, ColomboHigh Commission of Sri Lanka, New DelhiEnvoyIndian High Commissioner to Sri Lanka Gopal BaglaySri Lankan High Commissioner to India Milinda Moragoda President Maithripala Sirisena with Prime Minister Narendra Modi in February 2015 India–Sri Lanka relations, Indian-Sri Lankan rel...
Motor vehicle Saturn S seriesFirst generation Saturn SL1 (1995)OverviewManufacturerSaturn Corporation (General Motors)Production1990–2002Model years1991–2002AssemblyUnited States: Spring Hill, Tennessee (Spring Hill Manufacturing)Body and chassisClassCompact carLayoutTransverse front-engine, front-wheel drivePlatformGM Z platformChronologySuccessorSaturn Ion The Saturn S-series is a family of compact cars from the Saturn automobile company of General Motors. Saturn pioneered the bran...
This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these template messages) The neutrality of this article is disputed. Relevant discussion may be found on the talk page. Please do not remove this message until conditions to do so are met. (July 2021) (Learn how and when to remove this template message) This article needs additional citations for verification. Please help improve this article by adding citations to ...
Ancient Egyptian symbol of protection, royal power and good health This article is about the ancient Egyptian symbol. For the video game, see Eye of Horus (video game). The left wedjat eye, symbolizing the Eye of Horus The Eye of Horus, also known as left wedjat eye or udjat eye, specular to the Eye of Ra (right wedjat eye), is a concept and symbol in ancient Egyptian religion that represents well-being, healing, and protection. It derives from the mythical conflict between the god Horus with...
In this Chinese name, the family name is Teng. Teng HaiqingNative name滕海清BornMarch 2, 1909Jinzhai County, AnhuiDiedOctober 26, 1997AllegiancePeople’s Republic of ChinaBranchPeople's Liberation ArmyRankLieutenant generalKnown forInvolvement in the Inner Mongolia incident Teng Haiqing (Chinese: 滕海清) (March 2, 1909 – October 26, 1997) was a Chinese military officer and a politician. He was in charge of the massive purge in the Inner Mongolia incident. Biography Teng was ...
This article is about the editor for VM/CMS, and ports thereof. For the editor for the X Window System on Linux and Unix, see xedit (X11). An Informatics General computer programmer using XEDIT on an IBM 3279 terminal An early version of XEDIT from 1982, as displayed on a 3270 terminal emulator. XEDIT is a visual editor for VM/CMS using block mode IBM 3270 terminals. (Line-mode terminals are also supported.) [1] [2] XEDIT is much more line-oriented[3] than modern PC an...
American baseball player (1946–2019) Baseball player Leroy StantonStanton in 1978Right fielderBorn: (1946-04-10)April 10, 1946Latta, South Carolina, U.S.Died: March 13, 2019(2019-03-13) (aged 72)Florence County, South Carolina, U.S.Batted: RightThrew: RightProfessional debutMLB: September 10, 1970, for the New York MetsNPB: April 14, 1979, for the Hanshin TigersLast appearanceMLB: September 30, 1978, for the Seattle MarinersNPB: October 16,&...
Work played exclusively by the percussion section of a modern marching band This article is about marching hit patterns. For general hit patterns, see Drum beat. For single hits, see Drum stroke. The Sun Devil Marching Band marching to their pregame cadence Drum - Cadence A Drum - Cadence B Drum - Four Flams Snare drum cadences performed by the United States Navy Band Problems playing these files? See media help. In music, a drum cadence or street beat is a work played exclusively by the perc...
Supercoppa svizzera 2015 Competizione Supercoppa svizzera Sport Pallavolo Edizione 17ª Organizzatore SV Luogo Svizzera Partecipanti 2 Risultati Vincitore Losanna UC(3º titolo) Secondo Lugano Statistiche Miglior giocatore Marcelo Elgarten Cédric Hominal Incontri disputati 1 Cronologia della competizione 2012 2016 Manuale La Supercoppa svizzera 2015 si è svolta il 6 settembre 2015: al torneo hanno partecipato due squadre di club svizzere e la vittoria finale è a...
Strategi Solo vs Squad di Free Fire: Cara Menang Mudah!
