Commons:Bots/Work requests

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Shortcut: COM:BR · COM:BWR

Bot policy and list · Requests to operate a bot · Requests for work to be done by a bot  · Requests for batch uploads
Gnome-system-run.svg

Filing cabinet icon.svg
SpBot archives all sections tagged with {{Section resolved|1=~~~~}} after 7 days.

Cultural heritage monuments in Italy - overcategorization[edit]

Hi! I'm working on emptying Category:Cultural heritage monuments in Italy and I've noticed that a lot of items are already in its subcategories. Could someone remove the category from files which already have a subcategory of it? Cat-a-lot can find these cases of overcategorization, but (I think) cannot automatically remove the superfluous Category:Cultural heritage monuments in Italy; if I'm wrong please let me know how. Thank you very much, --Epìdosis 14:45, 19 January 2019 (UTC)

Ah, please do not remove Category:Cultural heritage monuments in Italy when the only subcategory is Category:Cultural heritage monuments in Italy with known IDs‎. --Epìdosis 14:50, 19 January 2019 (UTC)
For subcategory I also intend subcategory of subcategory and subcategory of subcategory of subcategory etc. --Epìdosis 14:56, 19 January 2019 (UTC)
BRFA filed --Kanashimi (talk) 23:47, 23 January 2019 (UTC)
@Epìdosis: I have do several edits. Please tell me if there is something wrong. Thank you. --Kanashimi (talk) 04:01, 19 February 2019 (UTC)
@Kanashimi: Edits like this are OK. Thank you very much, --Epìdosis 08:14, 19 February 2019 (UTC)
@Epìdosis: I have do more edits. If all seems OK, I will clean all the rest. --Kanashimi (talk) 11:07, 20 February 2019 (UTC)
@Kanashimi: These edits are all OK, you can go ahead. Thank you! --Epìdosis 11:15, 20 February 2019 (UTC)

Current events box[edit]

Is there someone who could make a bot that maintains a "current events" box, like on User:Alexis Jazz/Proposal incubator? (that box is static and only to illustrate the general idea, what exactly will be in it is up for debate)

A proposal to put such a box on top of the village pump would still have to be made, but there's no point if nobody can actually create it. Actually, any local VP could also adopt it, so if someone can create it it's unlikely to go completely unused. - Alexis Jazz ping plz 14:30, 30 January 2019 (UTC)

Internet Archive preservation of external links[edit]

System-search.svgSee also: Category:Uploads by Fæ with linkrot.

I am requesting for a bot operator to start archiving all external links on Wikimedia Commons using the InternetArchiveBot script as could be found here:

https://meta.wikimedia.org/wiki/InternetArchiveBot & https://tools.wmflabs.org/iabot/ github here https://github.com/cyberpower678/Cyberbot_II/tree/master/IABot (pages advised by Slowking4).

Maybe a bot operator could draw some inspiration from the page "User:Fæ/Wayback" developed by . Personally I'd prefer if the bot would try to archive links from the oldest files with external links on Wikimedia Commons to the newest, note that there exist a very large amount of files on Wikimedia Commons that have lost their "original website" and their licenses aren't as well documented anymore (see the top link). --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 10:16, 5 February 2019 (UTC)

Considering the size of this task, a proposal is needed for community discussion. -- (talk) 10:47, 5 February 2019 (UTC)
@: Sure, I'll have it live at "Commons:Village pump/Proposals" soon, thanks for clearing that up. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 11:04, 5 February 2019 (UTC)

The proposal has received a fair amount of support with no opposition and has heen open for quite a while, after it'll be closed I'd advise the Bot-operator to try and archive links based on the oldest first, but as there are +/- 25.000.000 (plus-minus twenty-five-million) media files 📁 with external links on Wikimedia Commons I'd probably say that 's suggestion of targeting high risk links first would be the wisest option. Anyhow as there's a huge backlog and linkrot has already been infecting a large number of files it would probably be best for someone who works a lot with external licenses like Fæ to do the directing (if they're interested, of course). Let's just hope that not more (unreviewed) licenses will be lost in the meantime, also maybe already (human- or bot-)reviewed licenses should go last(?), I'm not an expert. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 11:36, 18 February 2019 (UTC)

For some further precedent please also see the page 📃 "User:Fæ/Wayback". --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 11:37, 18 February 2019 (UTC)

size-files[edit]

Fixed-size files like: File:Pycnogonida_Nymphon_s_Sars_400px.png?redirect=no [now executed ... ] should be identified, checked, replaced & deleted. --Itu (talk) 22:24, 9 February 2019 (UTC)

I'm not sure how this can be automated. How would a bot be able to help identify these cases? --Schlurcher (talk) 12:44, 10 February 2019 (UTC)
A weak heuristic may be the \d+px\.{extension} format. --grin 17:17, 10 February 2019 (UTC)
If tasks can not be automated perfectly, they typically can get nearly or semi-automated. Instead: plain manual work will be tantalizing and more error-prone anyway.
First step is compiling a list of that files.
A synopis page should view the heuristic found original and size-derivates, so it can be checked by human eyes. And so on... --Itu (talk) 19:10, 10 February 2019 (UTC)
user:Túrelio is hunting such files (wasn't helpful for the demonstrating object). Maybe he will explain his concept for the work. --Itu (talk) 19:05, 11 February 2019 (UTC)

Finding images with captions[edit]

Examining the current API, I have a Pywikibot script running which takes all members of a category and checks if they have captions, but it's a stupidly slow linear way of doing it. The way I have is to:

  1. get the category member,
  2. ask for its pageid,
  3. turn that into a mid (by string adding M),
  4. then doing a wbgetentities from the Site using the mid

I'm thinking there should be a way to up front pull only files with captions (i.e. not "missing") from a category, without having to check every file separately. Anyone have some example code out there? -- (talk) 12:26, 13 February 2019 (UTC)