Last modified on 28 July 2014, at 07:36

Commons:Bots/Work requests

Bot policy and list · Requests to operate a bot
Requests for work to be done by a bot · Changes to allow localization  · Requests for batch uploads

Filing cabinet icon.svg

SpBot archives all sections tagged with {{Section resolved|1=~~~~}} after 1 day.

Links to GallicaEdit


A lot of files from the French National Library (Gallica) have been uploaded for Wikisource, but the source is often not properly given. The template {{gallica}} should be used, but it is not always the case. Could a bot replace the URL (which looks like by {{gallica|bpt6k5820787t}}? In addition, sometimes no URL was given. Could a list be made, so that the correct source be added manually later? Thanks in advance, Yann (talk) 11:57, 18 July 2014 (UTC)

Open proxy detectionEdit

Hello, I have tested bit around with my mobile phone and seen that it is easy to edit with an OP on commons (test). Anyone a idea how to write a open proxy finder? --Steinsplitter (talk) 11:15, 20 July 2014 (UTC)

Why would this be good for Commons (on the presumption that not all open proxies are being misused for spam, crime or vandalism and intelligent prevention or block-list inheritance might be better to automate detection of spam, crime or vandalism...)? -- (talk) 11:25, 20 July 2014 (UTC)
Because open proxy's are forbidden on commons and there is a huge abuser potential. For examle we block TOR atm with Extension:TorBlock.--Steinsplitter (talk) 11:36, 20 July 2014 (UTC)
Yes, if you presume I have read meta:NOP (which includes the Tor example), then we both know that "legitimate users[...] may freely use open proxies [...]" and "This policy is known to cause difficulty for some editors, who must use open proxies to circumvent censorship where they live". A corollary, is that blocking of every possible proxy is not desirable and policies probably ought to focus on known misused open proxies; by inheriting validated lists from elsewhere or based on evidence of a history of proxy range misuse here on Commons.
For this reason the common paradigm that "open proxies are Evil" is probably something that bot writers should probably avoid in preference to a slightly more nuanced interpretation of the meta policy. -- (talk) 11:41, 20 July 2014 (UTC)
Imho it would be at least helpful to tag all open proxy edits on a page. --Steinsplitter (talk) 11:43, 20 July 2014 (UTC)
Possibly, in the same way as we tag mobile platform edits. However I would be cautious if this might be interpreted such that our community presumed bad faith for all such edits, or it were to inevitably result in some administrators going on a blocking campaign as a preventative action. Backing off, happy to see where this goes. -- (talk) 11:48, 20 July 2014 (UTC)
I also don't know how to detect it (scanning ports or smithing like this?). --Steinsplitter (talk) 11:52, 20 July 2014 (UTC)
See Commons:Open proxy detection. -- (talk) 11:57, 20 July 2014 (UTC)
  • It's a tough question. Restating the obvious, for those facing censorship or being watched by their governments, use of open proxies may be an essential part of their personal security. We should be careful with those -- even identifying them on Commons may make them more likely to be investigated by the bad guys. On the other hand, an open proxy is a very useful tool for the spammer or vandal. We'd like to be able to crack down on them.
Generally, if you Google an IP address, you can see whether it is an open proxy or not, but I'm not at all sure you could teach a bot to do it -- there's a certain amount of judgement involved. Perhaps, if such a bot could be developed, it might report on a page accessible only to CUs -- that would solve the first problem and put the information in the hands of those that the community has trusted to take care of this kind of problem. .     Jim . . . . (Jameslwoodward) (talk to me) 12:58, 20 July 2014 (UTC)

The Film DailyEdit


Could a bot upload these DJVU files The Film Daily from Internet Archive? It can't be done by this tool because the size is near or over 100 MB. License is {{PD-US-not renewed}}, except this one, which is {{PD-US}}. Thanks, Yann (talk) 06:05, 25 July 2014 (UTC)

Is there an easy way to steal the Commons description from ia-uploader? Otherwise this looks like a manual copy-paste task cooperating with ChunkedUpload. --McZusatz (talk) 13:59, 25 July 2014 (UTC)
No idea about the internal of ia-uploader. User:Tpt knows. Downloading, and then uploading manually around 3 GB is out of question for me. That's why I am asking. Yann (talk) 14:25, 25 July 2014 (UTC)
Would you mind copy-pasting the Commons description from ia-uploader into the File:The name you want the file to have.djvu file description pages and send me all the links. I can do the upload pretty fast afterward, I think. --McZusatz (talk) 09:07, 26 July 2014 (UTC)
Like this File:The Film Daily, Jan-Jun 1932.djvu? The IA search engine is currently screwed. So I will do the rest when it is back. Regards, Yann (talk) 13:25, 26 July 2014 (UTC)
Yes, that is just right. --McZusatz (talk) 14:17, 26 July 2014 (UTC)
@McZusatz: Hi, Here is the list User:Yann/The Film Daily. Thanks for your help, Yann (talk) 16:00, 27 July 2014 (UTC)
Should be ✓ Done. --McZusatz (talk) 07:36, 28 July 2014 (UTC)