feat: add backfill one off scripts #734
Conversation
these are one off python scripts to be run directly that will modify the database the first backfills the daily test rollups starting on a given day, and ending on a given date, starting for a given repo (it will backfill in repo id order) the second backfills the TestFlagBridge objects for all repos with the option to focus on a given repo, specified by repoid
Codecov ReportAttention: Patch coverage is
@@ Coverage Diff @@
## main #734 +/- ##
===========================================
- Coverage 98.09% 25.14% -72.95%
===========================================
Files 432 438 +6
Lines 36332 36543 +211
===========================================
- Hits 35640 9189 -26451
- Misses 692 27354 +26662
Flags with carried forward coverage won't be shown. Click here to find out more.
|
❌ 1 Tests Failed:
View the top 1 failed tests by shortest run time
To view individual test run time comparison to the main branch, go to the Test Analytics Dashboard |
|
Test Failures Detected: Due to failing tests, we cannot provide coverage reports at this time. ❌ Failed Test Results:Completed 1 tests with View the full list of failed testspytest
|
❌ 1 Tests Failed:
View the top 1 failed tests by shortest run time
To view individual test run time comparison to the main branch, go to the Test Analytics Dashboard |
| os.environ.setdefault("DJANGO_SETTINGS_MODULE", "django_scaffold.settings") | ||
| django.setup() |
There was a problem hiding this comment.
any reason this is duplicated from the main script? does this actually run twice when you import one of the child scripts?
There was a problem hiding this comment.
These scripts are meant to be run separately from the actual application running on a pod so we'd invoke it directly using python one_off_script.py and that means we have to setup django separately
| f"Updating test instances {start_repoid} {start_date} {end_date}", | ||
| extra=dict(start_repoid=start_repoid, start_date=start_date, end_date=end_date), |
There was a problem hiding this comment.
as you have all these fields in the extra as well, I would remove them from the log message, it only makes it harder to search for all the individual logs.
| / (obj.pass_count + obj.fail_count), | ||
| ) | ||
|
|
||
| rollup.save() |
There was a problem hiding this comment.
django has a bulk_create method as well, though I recommend to always define a reasonable chunk size.
| rollup.save() | ||
|
|
||
|
|
||
| def run_impl( |
There was a problem hiding this comment.
please give this a more descriptive name :-)
| f"Updating test instances {start_repoid} {start_date} {end_date}", | ||
| extra=dict(start_repoid=start_repoid, start_date=start_date, end_date=end_date), | ||
| ) | ||
| test_analytics_repos = get_test_analytics_repos(start_repoid) |
There was a problem hiding this comment.
you should probably also iterate over these in chunks, because you are also logging the whole list.
unless you know for certain there is only a small number of these?
There was a problem hiding this comment.
Yeah there aren't many Repository objects that would be returned from this function
| ).select_related("upload")[:1][0] | ||
|
|
||
| if first_test_instance is None: |
There was a problem hiding this comment.
I’m not 100% sure how this works with using the slicing syntax to define a limit, whether you actually get a list of appropriate length with None filled in?
I think this might just throw because you are indexing out of bounds if no such object exists?
| new_bridge = TestFlagBridge(test=test, flag=flags[flag_name]) | ||
| new_bridge.save() |
There was a problem hiding this comment.
again, bulk_create should be more appropriate here.
these are one off python scripts to be run directly that will modify the database
the first backfills the daily test rollups starting on a given day, and ending on a given date, starting for a given repo (it will backfill in repo id order)
the second backfills the TestFlagBridge objects for all repos with the option to focus on a given repo, specified by repoid