This is a problem pretty unique to the failures endpoint, at the moment, since it's the only endpoint which is collapsing results from multiple test runs.
<rdar://problem/57334389>
Created attachment 383916 [details] Patch
Comment on attachment 383916 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=383916&action=review rs=me > Tools/resultsdbpy/resultsdbpy/controller/failure_controller.py:86 > + abort(404, description='No test runs found with the specified criteria') should it be 404 or something else. It might make it hard to distinguish incorrect urls from correct urls returning 404.
Comment on attachment 383916 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=383916&action=review >> Tools/resultsdbpy/resultsdbpy/controller/failure_controller.py:86 >> + abort(404, description='No test runs found with the specified criteria') > > should it be 404 or something else. It might make it hard to distinguish incorrect urls from correct urls returning 404. The error messages for this is different 'No test runs found with the specified criteria' vs 'The requested URL was not found on the server. If you entered the URL manually please check your spelling and try again.'. We clearly need a 400 family error, I don't see one that fits better....maybe 406 or 425? But you kind of have to think creatively for why those fit.
Created attachment 384200 [details] Patch for landing
Comment on attachment 384200 [details] Patch for landing Clearing flags on attachment: 384200 Committed r252815: <https://trac.webkit.org/changeset/252815>
All reviewed patches have been landed. Closing bug.