Page MenuHomePhabricator

Pending (false positive?) translations on Gerrit to Toolforge tools
Open, Needs TriagePublic

Description

Looking at https://gerrit.wikimedia.org/r/#/dashboard/137, there are pending translations to Toolforge tools which are failing due to false positives? AFAIK, Raimond usually cleans such up for MW extensions but I'm not sure for tools.

So please if anyone knows the maintainers of these tools, he/she can just add them so they can review these translations and merge them? For a while now, they've been stacking on each other.

List of tools below

  • …/tools/crosswatch
  • …/tools/commons-mass-description
  • …/tools/weapon-of-mass-description
  • …/tools/map-of-monuments
  • wikimedia/portals
  • wikimedia/wikimania-scholarships
  • …/tools/commons-mass-description
  • …/tools/weapon-of-mass-description

See: https://lists.wikimedia.org/pipermail/wikitech-l/2019-April/092010.html (for discussion thread on Wikitech), thanks @Amire80 for asking me to file a task :)

Event Timeline

Restricted Application added a subscriber: Aklapper. · View Herald TranscriptApr 28 2019, 12:24 PM
D3r1ck01 updated the task description. (Show Details)Apr 28 2019, 12:29 PM

@abi_ is doing exports for these currently and is waiting for permissions (T222015) to be able to manually merge these patches.

It would also be nice if someone improved the checker to avoid these false positives to avoid wasting our time.

Krenair updated the task description. (Show Details)Apr 28 2019, 2:49 PM
Krenair added a subscriber: Krenair.
Urbanecm updated the task description. (Show Details)May 4 2019, 4:51 PM

@abi_ is doing exports for these currently and is waiting for permissions (T222015) to be able to manually merge these patches.

You can grant her abi the permissions ;). See T222015 for details on how to do that.

It would also be nice if someone improved the checker to avoid these false positives to avoid wasting our time.

They're not false positives in true sense, all contains HTML (if I haven't overlooked something). It's hard to detect "bad code", I guess...