Wikidata:Requests for permissions/Bot/JhsBot 3
The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
- Approved--Ymblanter (talk) 18:37, 6 September 2014 (UTC)[reply]
JhsBot 3[edit]
JhsBot (talk • contribs • new items • new lexemes • SUL • Block log • User rights log • User rights • xtools)
Operator: Jon Harald Søby (talk • contribs • logs)
Task/s: Automatically add descriptions and labels for Scandinavian (and other languages upon request).
Code: Pywikibot based (from Ladsgroup)
Function details: These are two separate scripts, but they work in almost the same way, so I'll just request them once here. The descriptions are added based on the nationality and occupation(s) of a person as listed on Wikidata. The labels are added for persons based on where they are from, and taken from the labels in their native language, but with several exceptions for weirdness (like royal names and so on). Jon Harald Søby (talk) 09:43, 2 September 2014 (UTC)[reply]
- Support - except for royal persons. The added labels and descriptions I have looked at are mostly OK. But names of royal persons are often translated and the bot doesn't catch that. I changed the Danish label for Prince Oskar of Prussia (Q61242) from "Oskar von Preußen" til "Oskar af Preussen" following the title of the red link in da:Wilhelm 2. af Tyskland, and the Danish label for Frederick II, Elector Palatine (Q61245) from "Friedrich II." to "Frederik 2. af Pfalz" following the title of the red link in da:Dorothea af Danmark (1520-1580). I think you should stay away from that kind of names. Regards, Dipsacus fullonum (talk) 09:46, 3 September 2014 (UTC)[reply]
- Yup, I agree. The current regex for labels to avoid is this:
([,(\d]| of | av | de | di | af | van | von | da |\.[a-zA-Z]|II|IV|VI|IX|XI|XV)
(I added 'von' after Prince Oskar of Prussia (Q61242) was edited). This will bypass quite a few names that would stay the same as well, such as Edwin van der Sar (Q482955), but it's better to be safe (as safe as possible, at least) than sorry. Jon Harald Søby (talk) 10:18, 3 September 2014 (UTC)[reply]
- Yup, I agree. The current regex for labels to avoid is this: