Page MenuHomePhabricator

TemplateData: Add a tracking category to all templates that don't have a TemplateData block
Open, LowestPublic

Description

No idea how we'd do this.


Version: master
Severity: enhancement

Details

Reference
bz70242

Event Timeline

bzimport raised the priority of this task from to High.Nov 22 2014, 3:41 AM
bzimport added a project: TemplateData.
bzimport set Reference to bz70242.

Maybe a Special:PagesWithoutProps would be more useful...? Definitely more efficient than a tracking category.

(In reply to Kunal Mehta (Legoktm) from comment #1)

Maybe a Special:PagesWithoutProps would be more useful...? Definitely more
efficient than a tracking category.

Could do. From an editor workflow, however, people would probably want something that integrates more closely with the backlog categories we currently use. I'm not sure.

... It is related to T87444, which was merged here as a dupe, but I was talking about showing the most transcluded pages. On projects with a lot of templates this may help editors prioritize the work.

matmarex removed a subscriber: matmarex.Apr 7 2015, 6:19 PM
Pppery awarded a token.Jan 9 2017, 3:26 AM
Deskana lowered the priority of this task from High to Lowest.Feb 23 2018, 4:30 PM
Krinkle removed a subscriber: Krinkle.Feb 23 2018, 9:26 PM

... It is related to T87444, which was merged here as a dupe, but I was talking about showing the most transcluded pages. On projects with a lot of templates this may help editors prioritize the work.

This seems like it'd be a fairly expensive query, and one that doesn't need to be in real-time, either. If implemented in production, it'd probably best as a cached query page, with a cheap real-time check that crosses out entries that now have template data.

As a simpler short-term solution, one could start with the existing "Mostlinkedtemplates" special page, and just cross out ones that have templatedata. This could even be done on Quarry or Toolforge.

But, unfortunately, that'd currently require inefficient workarounds with API queries followed by a labs-db query because the querypage tables aren't available on public replicas yet - per T65782.