Wikidata:Requests for permissions/Bot/PintochBot 5
- 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) 20:46, 27 May 2019 (UTC)
PintochBot 5
[edit]PintochBot (talk • contribs • new items • new lexemes • SUL • Block log • User rights log • User rights • xtools)
Operator: Pintoch (talk • contribs • logs)
Task/s: normalize X username (P2002) to the preferred case (the one stated on the service)
Function details:
Twitter usernames are case-insensitive. Therefore, if I add X username (P2002)john_doe to an item and X username (P2002)John_Doe to another item, both links will get you the same profile, and this will not be reported as a violation of the distinct-values constraint (Q21502410). Each Twitter username has a canonical case: the way it is stated on the service. By normalizing Twitter ids to that case, the constraint system should surface these issues, potentially helping us to merge duplicate items or detect wrong Twitter ids more easily.
This bot task only affects the items where only one Twitter id is present: when there are more, manual review might be needed.
See the sample edits. − Pintoch (talk) 10:31, 24 May 2019 (UTC)
- Nice! We have a similar issue with DOI's (P356) - SourceMD is entering them all in upper-case, but they are often presented as lower-case, and some are shown as mixed case on the journal websites (eg. on https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.122.208101 the DOI is shown as 10.1103/PhysRevLett.122.208101). I don't know if there's a good source for the canonical form of these though. ArthurPSmith (talk) 14:42, 24 May 2019 (UTC)
- DOIs are formally case-insensitive, but there isn't really a "canonical" form, and a little awkwardly some publishers / data services have occasionally switched what approach they prefer. Internally it seems the system uses allcaps, but IME all-lowercase is more common for the way they're used in the wild, so we could make an argument either way. Andrew Gray (talk) 19:10, 24 May 2019 (UTC)
Support, looks good. I haven’t checked if the new values were correct indeed, but why wouldn’t they? —Tacsipacsi (talk) 17:49, 24 May 2019 (UTC)- Support Looks great, and it would be good to be able to roll this out for other similar identifiers (eg Facebook IDs). Andrew Gray (talk) 19:10, 24 May 2019 (UTC)
- I will approve the request in a couple of day provided no objections have been raised.--Ymblanter (talk) 19:56, 25 May 2019 (UTC)