Outdated when dealing with PUF

Legobot is currently editing and adding daily entries over Wikipedia:Possibly unfree files. However, it's adding pages from two weeks ago, instead of one week ago. The length was changed to close discussions from two weeks to one following this proposal. Can this bot be updated to reflect that? — ξxplicit 00:49, 27 June 2012 (UTC)

Edit. Is this an error with the bot or is it supposed to eliminate links to prior deletion discussions? I think this action should cease. Candleabracadabra (talk) 17:54, 22 September 2012 (UTC)

Thanks for notifying me of that, it seems to be a bug. I'll look into it and fix any other possible errors the bot might have made. LegoKontribsTalkM 22:23, 22 September 2012 (UTC)
Okay, thanks. Have a great weekend. Candleabracadabra (talk) 23:26, 22 September 2012 (UTC)

Task 20: Redirect CSD tagging

Just an FYI, but the task is a bit redundant to this page already updated by another bot: Wikipedia:Database reports/Broken redirects. -- Gogo Dodo (talk) 06:35, 25 September 2012 (UTC)

That is actually where the bot gets its information from. The BRFA has more information about how the bot works and the original request behind it. LegoKontribsTalkM 06:42, 25 September 2012 (UTC)
Ah, okay, I didn't read the BRFA that carefully. I missed the line that said that it picks it up from the list. I read the original request and I'm okay with it, but I don't quite see the urgency of clearing these things out. There being a backlog doesn't seem to be that big of a deal, but that's just my opinion.
On a different topic, I do see the discussion about having a custom CSD tag template for it, and I'm good with that, too, but one minor thing is that it breaks the automatic fill in of the deletion reason for the deleting admin. With the regular template, the reason is set to "G8: Redirect to a deleted or non-existent page", while your bot gives this "G8:". The picky admin in me has to reselect the deletion reason in the popup menu. I do see that there is a "bot" flag with {{db-redirnone}}, but the message is a little different and doesn't include a warning to the admin. -- Gogo Dodo (talk) 06:59, 25 September 2012 (UTC)
Is there a way we can modify the CSD template to include the full reason? Or does that need to be modified in a user-script like Twinkle? LegoKontribsTalkM 00:12, 27 September 2012 (UTC)
Nyttend let me know that he has fixed the template so it should provide the proper reason. LegoKontribsTalkM 02:25, 27 September 2012 (UTC)

Has-NFUR

Howdy. I recently noticed this edit. That then led me to here. It probably isn't a big deal, but wouldn't it have more appropriate to use the |file has rationale=yes as mentioned at the top of the TfD for ogg files instead of what was used?--Rockfang (talk) 23:46, 6 October 2012 (UTC)

Hm, thats a good point. I think it would make more sense to tag everything as file since all images are technically files as well. In the end it really won't make a difference (besides the 1 character) since once tagged users will rarely check it. I'm adjusting it right now to simply add |file has rationale=yes. Cheers, LegoKontribsTalkM 23:56, 6 October 2012 (UTC)
Well, turns out {{Non-free audio sample}} doesnt support image. I'll go back to image and bug Sfan00 about it. LegoKontribsTalkM 00:01, 7 October 2012 (UTC)

Problem on Talk:List of Unified Modeling Language tools.

Hi,

I have noticed a problem on Talk:List of Unified Modeling Language tools with your edit. It seems that the user has not been taken into account. Regards. Ftiercel (talk) 18:09, 8 October 2012 (UTC)

That's correct. The template it was converting it back to ({{oldafdfull}}) does not support that parameter, and as such, it was dropped. LegoKontribsTalkM 19:37, 8 October 2012 (UTC)

Bug: literal insertion of "subst" tokens

In this edit, Legobot added today's datestamp to a template, but somehow the {{subst:CURRENTMONTHNAME}} was added literally. The "subst" of the magic word (to insert the actual current month name) was performed later by another bot. During the intervening time, the actual wikicode markup is visible to readers. And if the second bot is delayed or doesn't happen (and some person edits this area later), the subst winds up inserting "current" for when the subst is finally done rather than "current" for when the CURRENTMONTHNAME word was added. DMacks (talk) 01:52, 9 October 2012 (UTC)

Ugh. This happens because you cant subst: things within ref tags. I've been tracking bugzilla:2700 for at least 4 years now. I'll code up a work-around right now. Thanks for bringing this to my attention. LegoKontribsTalkM 02:08, 9 October 2012 (UTC)
  Done LegoKontribsTalkM 02:20, 9 October 2012 (UTC)
Thanks for the speedy resolution! Not sure what sort of wiki-love a bot wants, but give yourself something nice:) DMacks (talk) 03:25, 9 October 2012 (UTC)

Adding of Category:Anime series to articles

Why is your bot doing this? This category is redundant (or a parent category) to those already on the category. Please stop this and reverse all the additions of this category. ···日本穣? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:06, 9 October 2012 (UTC)

I have temporarily halted the bot for now, the reason it started doing so was because of a request on Wikipedia:BOTR#Hard_categorization_of_anime_and_manga_articles. LegoKontribsTalkM 06:07, 9 October 2012 (UTC)
Ah, that makes sense. I'm not sure how complicated it would be, but could it sort them into the subcategories by decade based on the existing year categories (such as Category:1984 anime)? That would be more effective than filling the parent category. ···日本穣? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:10, 9 October 2012 (UTC)
I'm not sure if I understand what you're saying. Shouldn't all animes be in Category:Anime series? Should the year be used for sorting? LegoKontribsTalkM 06:59, 9 October 2012 (UTC)
Also, to try and keep the discussion in one place, I think it's best if we continue this at WP:BOTR LegoKontribsTalkM 07:01, 9 October 2012 (UTC)

I can't find it!...

I can't find the Instructions for adding this Bot to a talk page that I want to archive... User:HBC Archive Indexerbot has an Instructions section with code so I can set up the Archive Index for that Bot. Please tell me where the Instructions are so I can start adding this Bot to Talk pages that I am archiving. Cheers, Shearonink (talk) 15:31, 29 September 2012 (UTC)

Those instructions are still accurate, and should start indexing. I had planned on creating a new set of instructions for the new features, however I haven't had the time to do that yet. The bot hasn't run for the past few days with a few issues I have had, however I'll try and start a manual run on my personal machine. Thanks, LegoKontribsTalkM 21:57, 29 September 2012 (UTC)
Please also keep the logging from HBCI.--Oneiros (talk) 22:49, 1 October 2012 (UTC)
Should/will be at User:Legobot/Archive Indexer Log (see this history). I discovered an issue with an infinite loop and am debugging that when I can, however I have midterms this week so I am quite occupied. LegoKontribsTalkM 00:07, 2 October 2012 (UTC)
Thanks in advance; take your time.--Oneiros (talk) 05:34, 2 October 2012 (UTC)

Just an FYI, I've fixed the bug and have started a manual run of the bot, though it does have some catch-up work to do:

New pages: 23
Updated pages: 787

LegoKontribsTalkM 00:05, 13 October 2012 (UTC)

Bot has finished its run, and now I have scheduled it to update twice a day, at 00:00 and 12:00 UTC. Let me know if you have any problems/issues/questions/concerns. LegoKontribsTalkM 08:21, 13 October 2012 (UTC)

Legobot recreated a deleted page

Legobot recreated the page Value enhancement with this edit. Seems like a run of the mill edit conflict. Cheers. Hairhorn (talk) 19:35, 10 October 2012 (UTC)

Wow, that's a really rare occurrence. Not much I can do there since it was a split-second kind of thing. The bot grabs the text, process it (usually under 2 seconds) then puts the page again. Thanks for letting me know about it though, LegoKontribsTalkM 19:46, 10 October 2012 (UTC)
Yeah, I figured this was rare enough that it's not really worth worrying about; I checked my own logs and I've done this trick at least twice myself... Hairhorn (talk) 20:05, 10 October 2012 (UTC)

Capitalizing templates

Why does Legobot capitalize templates like {{cite}}, like it recently did on The Pirate Bay? This is 100% make-work, and is extremely irritating when I'm trying to look through recent changes to an article. Please make it stop doing this. —Kerfuffler  thunder
plunder
 
14:07, 11 October 2012 (UTC)

Looks like I was beaten to it. I came here to complain about the very same thing as it just inflates the differences and makes things difficult to see what is happening. Is there any agreement to capitalise these templates? Keith D (talk) 15:45, 11 October 2012 (UTC)
Sorry about that, I had that on my to-do list, just hadn't gotten around to fixing that. I've disabled it until I can write a patch. LegoKontribsTalkM 20:56, 11 October 2012 (UTC)
  Fixed If you take a look at this edit, Legobot dated the template correctly without capitalizing a bunch of templates. Additionally, I have created a stop page for this specific task at User:Legobot/Stop/II 2 so I don't have to manually disable it or have the bot blocked. This is mentioned in the edit summary for easy access as well. Cheers, LegoKontribsTalkM 17:46, 12 October 2012 (UTC)
Thanks for the quick response. —Kerfuffler  thunder
plunder
 
18:08, 12 October 2012 (UTC)
Thanks here as well. Keith D (talk) 18:56, 12 October 2012 (UTC)

Wikify

Just to let you know that Template:Wikify has been deprecated and shouldn't be used anymore. Thanks! Delsion23 (talk) 21:05, 14 October 2012 (UTC)

I assume you're referring to [1]. Legobot does not add tags like {{wikify}}, it merely adds a date like |date=October 2012. LegoKontribsTalkM 21:15, 14 October 2012 (UTC)

DEBORAH1111

Thank you for stoping by and leaving comment, I have added more refrence as you have mentioned. if more refrences are needed plz let me know — Preceding unsigned comment added by DEBORAH1111 (talkcontribs) 08:32, 17 October 2012 (UTC)

Please do not recreate a page

Please choose the option so that a page is not recreated if it is recently deleted. Example. Magog the Ogre (tc) 22:15, 16 October 2012 (UTC)

Hm, thats odd. Do you know what edit the bot made and what the summary was? I'm seeing if I can do anything on the bot's end to prevent that from happening again. Legoktm (talk) 22:18, 16 October 2012 (UTC)
  Fixed. I've (hopefully) patched it so it won't create the page if it doesn't exist upon putting the edit (done on the API side), however if you could still provide the information I asked for above that would be helpful. Thanks, Legoktm (talk) 22:38, 16 October 2012 (UTC)
At 20:17, 16 October 2012, I deleted the page. At 20:18, 16 October 2012, LegoBot recreated the page with the edit summary Bot: Updating license tag(s) with image has rationale=yes. Magog the Ogre (tc) 15:53, 17 October 2012 (UTC)
Thanks. Thinking about this logically, this happens because I'm pre-loading pages (50 at a time) to reduce requests on the API and increase processing speed. But this means I'm only checking if they exist anywhere from 0-4 minutes before editing them. Regardless, the fix I've implemented puts the check on it server side, and fixes the issue. Cheers, Legoktm (talk) 20:39, 19 October 2012 (UTC)

Recovering an image

Hello. I uploaded images of the dust jackets or covers of all eight novels written by J L Carr. Someone deleted one of them, the image of the cover of 'A Month in the Country', and I wasn't in a position to argue that it was fair use as I don't understand the arguments or technical language. I am delighted that your bot has updated the licence tags of some of the other images to argue that that the image has rationale=yes. Can you recover the deleted image so that all 8 novels are illustrated, or can I upload it again without someone deleting it? Please don't assume that I am expert, I am not, I just like to write what I know about. Thanks Ah1954 (talk) 21:09, 25 October 2012 (UTC)

  Done Restored and replaced into the article at A Month in the Country (novel). All of the tags should be fine now. VernoWhitney (talk) 21:21, 25 October 2012 (UTC)
Thanks VernoWhitney! Ah1954, if some of the images didn't get tagged yet and they do have a rationale, the bot will be doing a second more-comprehensive run (due to some weird way templates are used), which should pick up the rest. Legoktm (talk)

Bot broke ?

http://en.wikipedia.org/w/index.php?title=File:Iron_Tower_omnibus.jpg&action=history

Sfan00 IMG (talk) 09:44, 27 October 2012 (UTC)

It's a mwparserfromhell bug. I'll poke Earwig about it. Legoktm (talk) 09:47, 27 October 2012 (UTC)
  Fixed, by Earwig in this commit. Legoktm (talk) 02:25, 28 October 2012 (UTC)

Number of images tagged?

I'm a little curious if you know exactly how many images the bot tagged in this "rationale=yes" run. — CobraWiki ( jabber | stuff ) 15:56, 27 October 2012 (UTC)

I don't have a specific number, but I can get one by doing a database query. In the meantime, I estimate it was around 260k images by a rough edit count statistic. Legoktm (talk) 01:24, 28 October 2012 (UTC)

Bot is a little early

I have reverted Legobot's edit at WP:PUF, where it added a new day to the holding cell. It linked the log for October 21 today, but it should be linking October 20. It's a day too early, and it seems to have started doing that on October 25 UTC. — ξxplicit 01:03, 28 October 2012 (UTC)

Isn't it supposed to be 7 days behind? It was October 28th 00:01 UTC and it added a link to October 21st. 28-7=21. If thats not the intended function of the bot, I can change it though. Legoktm (talk) 01:07, 28 October 2012 (UTC)
The way the bot is running now, the link is being added at the beginning of its seventh day, only allowing a full six days of discussion. Discussions are having one day taken away from them. — ξxplicit 22:53, 1 November 2012 (UTC)
Ok, makes sense. I've adjusted it so it should be +1 more day now, starting with the next bot run. Legoktm (talk) 09:19, 2 November 2012 (UTC)

A barnstar for you!

  The Tireless Contributor Barnstar
Thank you for working on the task described at WP:Bots/Requests for approval/Legobot 26 Legolover26 (talk) 16:50, 6 November 2012 (UTC)

Blocked

I carefully unspeedied a set of "Section xx of the Constitution of Australia" and "yyy Corps (Germany)" redirects with red links because I think they are all valid redirects which need fixing rather than deleting. The bot promptly re-tagged them. I am prepared to discus this with an human being but I do accept a bot reversing my edits. — RHaworth (talk · contribs) 00:20, 13 November 2012 (UTC)

Makes sense. Would it be possible for the bot to be unblocked right now while a solution can be found for this problem? I have disabled the redirect tagging task, however Legobot has other things to do. Legoktm (talk) 00:22, 13 November 2012 (UTC)
Hi RHaworth. The original task (BRFA) was approved for Legobot to tag all broken redirects one a day (normally around midnight UTC). As such, the bot was not malfunctioning, the task merely needs to be enhanced. One of the advantages of Legobot tagging broken redirects is that they can be fixed by an editor faster than normal. If you look at the bots log of taggings, you'll note that many of the links are blue, because the redirect merely had to be fixed. Broken redirects are especially dangerous to readers, since they appear blue when used, however when the reader clicks, there is merely a redlink showing.
As a general rule, bots shouldn't be undoing/reverting a human editor's edits. I think the best solution to go from here, is for the bot to maintain a list of pages it's already tagged, and only tag if the bot hasn't already tagged it. This would prevent it from doing what just happened. This will probably cause some false negatives, however its best to be safe rather than sorry.
I'll also ask Nyttend to comment as well, since he was the one who originally requested the task. Thanks, Legoktm (talk) 00:39, 13 November 2012 (UTC)

Unblocked. — RHaworth (talk · contribs) 00:40, 13 November 2012 (UTC)

The solution is to retarget the redirects; removing the tags without changing the target isn't particularly helpful, and nor is changing the bot's programming. Nyttend (talk) 00:46, 13 November 2012 (UTC)

I fully agree. I have fixed the Australian ones and will do the German ones shortly. — RHaworth (talk · contribs) 01:17, 13 November 2012 (UTC)

Now fixed. In fact the bot was correct in every case. I should have investigated and actioned appropriately at the time instead of simply removing the tag. Any mods to the bot are very low priority. But why does it need to maintain a list? Surely it can simply check the edit history to see if it has been there before? — RHaworth (talk · contribs) 11:23, 13 November 2012 (UTC)

It could, however that would require more queries to the API than simply maintaining a list, something the bot already does. However, I think that Nyttend is right, that when a redirect has been tagged, rather than removing the tags, the responding administrator should simply try and fix them.
RHaworth, would it be alright if I turned the bot back on in it's original configuration? Wikipedia:Database reports/Broken redirects has piled up over the last few days. Thanks, Legoktm (talk) 19:57, 15 November 2012 (UTC)

Absolutely no objection. Sorry, I did not say that earlier. — RHaworth (talk · contribs) 23:45, 15 November 2012 (UTC)

Is there a way to automate this replacement, rather than simply flagging them as in this edit? -- Scray (talk) 01:05, 4 December 2012 (UTC)