Cast pg_proc char columns to Text explicitly#126
Conversation
|
Not sure how to properly test these changes, since I don't know what realistic circumstances would cause the |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #126 +/- ##
=======================================
Coverage 89.81% 89.81%
=======================================
Files 86 86
Lines 4056 4056
Branches 841 841
=======================================
Hits 3643 3643
Misses 336 336
Partials 77 77 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
DanCardin
left a comment
There was a problem hiding this comment.
The test would probably be to do all postgres testing on asyncpg (since i believe im only doing psycopg3 right now, if anything?)
but adding a whole extra driver would be a more involved test refactor, so this is fine as-is!
|
Mind bumping the project patch version and adding it to the changelog, and i can release the merged PR directly? |
|
I applied those changes, and it works on my $ poetry run alembic check
[info ] Context impl PostgresqlImpl. [alembic.runtime.migration]
[info ] Will assume transactional DDL. [alembic.runtime.migration]
No new upgrade operations detected.$ poetry run alembic downgrade -1
[info ] Context impl PostgresqlImpl. [alembic.runtime.migration]
[info ] Will assume transactional DDL. [alembic.runtime.migration]
[info ] Running downgrade 00226 -> 00225, Remove unique constraint from the 'test' table. [alembic.runtime.migration]$ poetry run alembic upgrade head
[info ] Context impl PostgresqlImpl. [alembic.runtime.migration]
[info ] Will assume transactional DDL. [alembic.runtime.migration]
[info ] Running upgrade 00225 -> 00226, Remove unique constraint from the 'test' table. [alembic.runtime.migration] |
Seems like `char` columns can be wrapped as `bytes` values under certain circumstances. See DanCardin#125
3a30204 to
7915f4e
Compare
|
@YaraslauZhylko Thanks for confirming. @DanCardin done! |
Fixes #125
Seems like
charcolumns can be wrapped asbytesvalues under certain circumstances. See #125