Skip to content

Update search_metadata.py ETC (sync from upstream: 9 commits / 64 files changed)#365

Merged
holta merged 9 commits intoiiab:masterfrom
janeczku:master
Mar 7, 2026
Merged

Update search_metadata.py ETC (sync from upstream: 9 commits / 64 files changed)#365
holta merged 9 commits intoiiab:masterfrom
janeczku:master

Conversation

@holta
Copy link
Copy Markdown
Member

@holta holta commented Mar 2, 2026

Fix TypeError when metadata provider returns None

asukavuuyn and others added 3 commits February 20, 2026 23:07
Fix TypeError when metadata provider returns None
Kobo: include PriorityTimestamp in PUT /state response
@holta holta added the enhancement New feature or request label Mar 2, 2026
jarynclouatre and others added 6 commits March 1, 2026 20:44
In series_list(), the SQLite query correctly orders results by
Series.sort, but a subsequent Python sorted() call (needed to
re-order after appending the "None" category entry) was using
Series.name as the sort key instead of Series.sort.

This caused series titles with leading articles (A, An, The) to
sort strictly alphabetically by the article rather than by the
meaningful word, e.g. "A Collins-Burke Mystery" appeared under
"A" instead of "C".

Fix by using Series.sort (with a fallback to Series.name if sort
is NULL) as the key in the Python re-sort, consistent with the
intent of the existing DB query.

Fixes #3583
@holta holta changed the title Update search_metadata.py (sync from upstream) Update search_metadata.py ETC (sync from upstream: 9 commits / 64 files changed) Mar 7, 2026
@holta holta requested review from chapmanjacobd and deldesir March 7, 2026 14:01
@holta
Copy link
Copy Markdown
Member Author

holta commented Mar 7, 2026

@Akatama @chapmanjacobd will a restructuring of this CI workflow plausibly/hopefully help avoid "false negatives" ?

(FWIW https://github.com/iiab/calibre-web/actions/runs/22796429132/job/66141089832?pr=365 below again appears to show an ongoing pattern, with PR's containing upstream commits somehow not being tested in a useful way :-)

image

@holta
Copy link
Copy Markdown
Member Author

holta commented Mar 7, 2026

(I'll go ahead and merge this experimentally — so we have another data point to try to confirm — allowing us to hopefully clean up this problematic CI workflow soon?)

@holta holta merged commit e27644b into iiab:master Mar 7, 2026
2 of 3 checks passed
@holta
Copy link
Copy Markdown
Member Author

holta commented Mar 7, 2026

so we have another data point to try to confirm

Confirmed! After merging this PR into our master branch:

image

cc: @Akatama @chapmanjacobd @deldesir let's work on this in coming days (this week!) if you have time to try to help rework this CI workflow, so that it properly avoids "False Negatives" (: 🙏

@chapmanjacobd
Copy link
Copy Markdown
Member

@holta
Copy link
Copy Markdown
Member Author

holta commented Mar 7, 2026

It must be something close to here: https://github.com/iiab/iiab/blob/c87f0add3d0c562ce7b764a87e3fdda1fce5e198/roles/calibre-web/tasks/install.yml#L92

Looks like you found the culprit... I think! (:

@holta
Copy link
Copy Markdown
Member Author

holta commented Mar 7, 2026

It must be something close to here: https://github.com/iiab/iiab/blob/c87f0add3d0c562ce7b764a87e3fdda1fce5e198/roles/calibre-web/tasks/install.yml#L92

How about we ignore var calibreweb_repo_url and have .github/workflows/iiab-install-smoke-test.yml set a new variable calibreweb_repo_dir in its /etc/iiab/local_vars.yml — that would contain the correct IIAB Calibre-Web code that needs CI testing:

  • Where "correct IIAB Calibre-Web code" means a speculative branch, containing the PR's commits speculatively merged into iiab/calibre-web's master branch, eventually destined for /usr/local/calibre-web-py3
  • If so, roles/calibre-web would of course need to ignore var calibreweb_repo_url in favor of calibreweb_repo_dir if the latter's found to be set — and then copy the contents of calibreweb_repo_dir into /usr/local/calibre-web-py3

Is this kind of approach or similar... reasonable?

cc: @chapmanjacobd @Akatama @deldesir

@chapmanjacobd
Copy link
Copy Markdown
Member

Makes sense to me! Just use uses: actions/checkout@v6 and then mv it to the location that it needs to be? Then it will be testing the actual code in the PR with no room for funny business!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants