Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix const validations #794

Merged
merged 1 commit into from Sep 17, 2019
Merged

Conversation

hmvp
Copy link
Contributor

@hmvp hmvp commented Sep 6, 2019

This fixes #620

NOTICE: pydantic is currently pushing towards V1 release,
see issue 576. Changes not required to release version 1
may be be delayed until after version 1 is released. If your PR is a bugfix for v0.32, please base off and target the v0.32.x branch.

Change Summary

Related issue number

Checklist

  • Unit tests for the changes exist
  • Tests pass on CI and coverage remains at 100%
  • Documentation reflects the changes where applicable
  • changes/<pull request or issue id>-<github username>.rst file added describing change
    (see changes/README.md for details)

@hmvp hmvp force-pushed the fix_const_validations branch 2 times, most recently from 4a6b5cf to f964990 Compare Sep 6, 2019
@codecov
Copy link

@codecov codecov bot commented Sep 6, 2019

Codecov Report

Merging #794 into master will not change coverage.
The diff coverage is 100%.

@@          Coverage Diff           @@
##           master   #794    +/-   ##
======================================
  Coverage     100%   100%            
======================================
  Files          17     17            
  Lines        2984   2847   -137     
  Branches      580    559    -21     
======================================
- Hits         2984   2847   -137
Impacted Files Coverage Δ
pydantic/fields.py 100% <100%> (ø) ⬆️
pydantic/error_wrappers.py 100% <100%> (ø) ⬆️
pydantic/types.py 100% <0%> (ø) ⬆️
pydantic/errors.py 100% <0%> (ø) ⬆️
pydantic/typing.py 100% <0%> (ø) ⬆️
pydantic/networks.py 100% <0%> (ø) ⬆️
pydantic/validators.py 100% <0%> (ø) ⬆️
pydantic/schema.py 100% <0%> (ø) ⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1b467da...e6d6389. Read the comment docs.

@hmvp
Copy link
Contributor Author

@hmvp hmvp commented Sep 6, 2019

Not sure if you wan't this in master or 0.32 so I submitted both

Copy link
Owner

@samuelcolvin samuelcolvin left a comment

looks like a good start, please can you also fix coverage.

pydantic/fields.py Outdated Show resolved Hide resolved
pydantic/fields.py Outdated Show resolved Hide resolved
HISTORY.rst Outdated Show resolved Hide resolved
tests/test_main.py Outdated Show resolved Hide resolved
@samuelcolvin
Copy link
Owner

@samuelcolvin samuelcolvin commented Sep 6, 2019

does this also resolve #793?

@hmvp hmvp force-pushed the fix_const_validations branch from f964990 to fb08c18 Compare Sep 7, 2019
@hmvp
Copy link
Contributor Author

@hmvp hmvp commented Sep 7, 2019

does this also resolve #793?

No, unfortunately it doesn't, and I don't really have a clue how to fix it. However it might not be a big issue for us since we initialize the constant fields with json compatible dicts instead of submodel instances

@hmvp hmvp force-pushed the fix_const_validations branch from fb08c18 to 43152ee Compare Sep 7, 2019
@hmvp
Copy link
Contributor Author

@hmvp hmvp commented Sep 7, 2019

does this also resolve #793?

No, unfortunately it doesn't, and I don't really have a clue how to fix it. However it might not be a big issue for us since we initialize the constant fields with json compatible dicts instead of submodel instances

Ok changed it up a bit more and now it should work for #793. At least in the most obvious cases.

It still fails if you initialize the default value with a BaseModel instance instead of a dict.

@hmvp hmvp force-pushed the fix_const_validations branch from 43152ee to 9260975 Compare Sep 7, 2019
Copy link
Owner

@samuelcolvin samuelcolvin left a comment

Looking good to me, just a few small things to change.

@tiangolo, please could you review this since you did most of the original work on schema.


m = Model()
assert m.a == [SubModel(b=1), SubModel(b=2), SubModel(b=3)]
assert m.b == [SubModel(b=4), SubModel(b=5), SubModel(b=6)]
Copy link
Owner

@samuelcolvin samuelcolvin Sep 9, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

would it also help to add m.schema() == {...} here?

Copy link
Contributor Author

@hmvp hmvp Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added, So this is interesting:

It seems that adding SubModel instances as default breaks in a lot of ways: Also the schema is now not json serializable, similar to the error as per #793 .

It might be a good idea to just document or add a warning of some sorts for that case and consider that unsupported. Of course not all use cases need json serializability and some use cases might want to set the default using a SubModel instance.

Copy link
Owner

@samuelcolvin samuelcolvin Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please create an issue about this.

try:
# Fails
Form(field1={'field': 1}, field2=[{'field': 1}])
except ValidationError as e:
Copy link
Owner

@samuelcolvin samuelcolvin Sep 9, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

use pytest.raises(ValidationError) as exc_info:.... exc_info.e.json()

Copy link
Contributor Author

@hmvp hmvp Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fixed

Form(field1={'field': 1}, field2=[{'field': 1}])
except ValidationError as e:
# This should not raise an Json error
assert json.dumps(e.json())
Copy link
Owner

@samuelcolvin samuelcolvin Sep 9, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this doesn't make sense, e.json() returns a string, this is effectively just json.dumps('foo') -> '"foo"'

Copy link
Contributor Author

@hmvp hmvp Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fixed

@samuelcolvin samuelcolvin requested review from tiangolo and removed request for tiangolo Sep 9, 2019
@samuelcolvin
Copy link
Owner

@samuelcolvin samuelcolvin commented Sep 17, 2019

Hi @hmvp would be great if you could work on fixing this. I'm trying to get a beta for version 1 out the door asap and I want to include this.

@hmvp
Copy link
Contributor Author

@hmvp hmvp commented Sep 17, 2019

Sure, I will try to get a fixed MR out today

@hmvp hmvp force-pushed the fix_const_validations branch from 9260975 to e6d6389 Compare Sep 17, 2019
@hmvp
Copy link
Contributor Author

@hmvp hmvp commented Sep 17, 2019

Pushed an update

if getattr(self.type_, '__origin__', None):
# type_ has been refined eg. as the type of a List and sub_fields needs to be populated
self.sub_fields = [self._create_sub_type(self.type_, '_' + self.name)]
# type_ has been refined eg. as the type of a List and sub_fields needs to be populated
Copy link
Owner

@samuelcolvin samuelcolvin Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've been working a lot on this part of the code in #803, I'm now sure this code will need more work.

But maybe it would be easier for me to fix it in #803.


m = Model()
assert m.a == [SubModel(b=1), SubModel(b=2), SubModel(b=3)]
assert m.b == [SubModel(b=4), SubModel(b=5), SubModel(b=6)]
Copy link
Owner

@samuelcolvin samuelcolvin Sep 17, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please create an issue about this.

@samuelcolvin samuelcolvin merged commit dd32a43 into samuelcolvin:master Sep 17, 2019
9 checks passed
@hmvp hmvp deleted the fix_const_validations branch Jan 6, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants