Skip to content

23.09.20

Compare
Choose a tag to compare
@adRn-s adRn-s released this 20 Sep 12:43
· 455 commits to develop since this release
  • New rule db-migras: loads production database snapshot with a proper reset of migration files. See source code and its help message for instructions on how to use it. It is a drop-in replacement to db rule when development version moved forward with changes into Django models that are not synchronized (yet) with production deployment that shall always run a version following github releases.
  • New 'Duties' model, to keep track in charge of whom are the responsibilities (both dry and wet processing of both short and long read sequencing.); so far it lists all users in database, but it's meant to be used only within the sequencing facility, and bioinformaticians.
  • New 'archived' field replaces the old obsolescence functionality that was broken in many if not all models. By default, all archived instances of models are filtered out at the Site Admin panel.
  • BUGFIX: Uploaded files appeared and disappeared just by switching sequenced to True.
  • BUGFIX: One API endpoint (/api/samples/?request_id=nnnn) was giving error 400 to non-staff users.
  • Renamed parkour_app/migrations to parkour_app/extras to avoid confusion with actual migrations. Renamed the corresponding test.py in there to test_migrations.py accordingly.
  • Added user email to mailed traceback when Django encounters any errors. This way we can contact users if they were experiencing a bug.
  • Added database from parkour-demo in JSON format under misc/ subfolder. Integrated all **/fixtures/*.json into it.
  • Updated fixtures with parkour-demo database, usefult to overwrite or further customize such entries.
  • Restored old makefile rules to save or load database in json. Do not use them with production data, BarcodeCounter bug is still in place, and it will be reset to 0 every time you use the json format. These rules are only meant to be helpers for the demo data which we prefer to have in JSON so that it's more robust to models' migrations.
  • Added tblib as a base dependency to have proper traceback when running tests in parallel.
  • Improvement: sweep rule won't remove current symlink targets anymore.