Set pydantic version less than 2.0 for all conda envs #160
Conversation
Pydantic 2.0 comes with major changes that we are not yet ready to support, and installs of it in environments that use `alchemiscale` are likely to break usage. This PR sets the version spec to <2.0 for pydantic in all conda envs.
Codecov ReportPatch coverage has no change and project coverage change:
Additional details and impacted files@@ Coverage Diff @@
## main #160 +/- ##
==========================================
- Coverage 82.62% 82.13% -0.49%
==========================================
Files 21 21
Lines 2676 2676
==========================================
- Hits 2211 2198 -13
- Misses 465 478 +13 ☔ View full report in Codecov by Sentry. |
|
Just FYI if you want to support pydantic~1 schema from pydantic>=2.0.0a @mattwthompson, @jhorton and I have been using this to import the v1 api from v2. from pydantic.v1 import BaseModel, FieldYou can then use the following guard to support both versions. try:
from pydantic.v1 import BaseModel, Field
except ImportError:
from pydantic import BaseModel, Field |
|
Oh gotcha, I suppose that would be the preferred approach to avoid holding back the rest of the stack? |
|
Trying out the import guard approach; will see if this breaks our tests. |
We tried putting in import guardrails, but tests fail in an unclear way. Will try to support pydantic 2.x at a later time.
|
Reverting to just supporting |
Pydantic 2.0 comes with major changes that we are not yet ready to support, and installs of it in environments that use
alchemiscaleare likely to break usage. This PR sets the version spec to <2.0 for pydantic in all conda envs.