Fix bug with SQL query for analysis view

[ChrisPaulBennett]

Check List

• I have read CONTRIBUTING.md and added my name as a Code Contributor.
• Contains logically grouped changes (else tidy your branch by rebase).
• Does not contain off-topic changes (use other PRs for other changes).
• Applied any dependency changes to both setup.cfg (and conda-environment.yml if present).
• Tests are included (or explain why tests are not needed).
• CHANGES.md entry included if this is a change that can affect users
• Cylc-Doc pull request opened if required at cylc/cylc-doc/pull/XXXX.
• If this is a bug fix, PR should be raised against the relevant ?.?.x branch.

[codecov-commenter]

Codecov Report

Patch and project coverage have no change.

Comparison is base (3fdfafa) 76.95% compared to head (94ae9ce) 76.95%.

Additional details and impacted files

@@           Coverage Diff           @@
##            1.3.x     #483   +/-   ##
=======================================
  Coverage   76.95%   76.95%           
=======================================
  Files          12       12           
  Lines        1406     1406           
  Branches      233      233           
=======================================
  Hits         1082     1082           
  Misses        277      277           
  Partials       47       47           

Files Changed	Coverage Δ
cylc/uiserver/schema.py	70.80% <ø> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ChrisPaulBennett]

This bug was noticed in a code review (cylc/cylc-ui#1355)

[oliver-sanders]

Hi there, thanks for this fix.

You've merged master into this branch, but the PR is being raised against 1.3.x, as a result there are a bunch of unrelated commits from the 1.4.0 branch (master) showing on this PR.

The easiest way to proceed is probably to cherry-pick those commits onto 1.3.x e.g:

$ git fetch upstream
$ git checkout -b fix-sql-bug upstream/1.3.x
$ git cherry-pick 3e80eedb25a311c3ecc5e9fae5fe450d56fea930
$ git cherry-pick 6b3e99fcaa71f3cf945ef3fd9ff4c9ca529b1932
$ git cherry-pick 9b8b94d93a2f92c9d448629e70875ffb313b93c2

Check the diff is as expected:

$ git diff upstream/1.3.x

You can avoid re-opening this PR by pushing to origin/master e.g:

# override your remote master with this branch
$ git push -f origin HEAD:master

[MetRonnie]

Thanks, can you just extract commit 3e80eed? May have to create a feature branch and open a new PR

$ git checkout -b sql-fix upstream/1.3.x # (if the remote is called upstream)
$ git cherry-pick 3e80eedb25a311c3ecc5e9fae5fe450d56fea930

[MetRonnie]

LGTM, though I'm not sure how to test/reproduce the bug, may need @oliver-sanders to review too

[oliver-sanders]

I'm not sure how to test/reproduce the bug, may need @oliver-sanders to review too

I think this should do it, try viewing a workflow with:

• A single failed task script = false.
  • The analysis view should not list any times for this task (run status != 0).
• Then broadcast script = sleep 60 and re-trigger the task.
  • The task should now show in the analysis view.
  • The run time should be ~60s not ~30s (because the failed task has been excluded).

[MetRonnie]

Ok, tested, reproduced bug on 1.3.x but all good on this branch