Support complex numbers #1331

changhc · 2024-06-12T21:14:18Z

Change Summary

Implement a validator and a serialiser for complex numbers.

As discussed in pydantic/pydantic#8555, since there is no official representation for complex numbers in JSON, I propose to express complex numbers as dictionaries with two keys, real and imag, with floating point values. Please find examples in the newly added test suites in this PR.

Note that function str_as_complex in shared.rs is not yet implemented as I need to discuss with the team if we want to support this. For example, we might want to accept strings like 1+2j as a valid complex number, as long as these strings follow the format described in python's documentation. This implementation, however, will be a bit tricky. Using regular expression is the simplest solution, but it is going to be a bit slow since building regex is rather costly. Parsing strings using the crate num_complex will make the implementation very neat, but num_complex is more tolerant in terms of string format, which means we will need to handle strings not allowed in python and it is going to be problematic. The safest solution is probably to do exactly the same as how cpython parses strings. I think that can be done in a separate PR for easier reviews.

Related issue number

pydantic/pydantic#8555

Checklist

Unit tests for the changes exist
Documentation reflects the changes where applicable
Pydantic tests pass with this pydantic-core (except for expected changes)
My PR is ready to review, please add a comment including the phrase "please review" to assign reviewers

codecov · 2024-06-12T21:53:02Z

Codecov Report

Attention: Patch coverage is 82.12560% with 37 lines in your changes missing coverage. Please review.

Project coverage is 89.55%. Comparing base (ab503cb) to head (8919589).
Report is 148 commits behind head on main.

Files	Patch %	Lines
src/serializers/infer.rs	0.00%	15 Missing ⚠️
src/serializers/type_serializers/complex.rs	83.87%	10 Missing ⚠️
src/input/return_enums.rs	38.46%	8 Missing ⚠️
src/input/input_string.rs	80.00%	1 Missing ⚠️
src/serializers/ob_type.rs	50.00%	1 Missing ⚠️
src/serializers/shared.rs	0.00%	1 Missing ⚠️
src/validators/complex.rs	97.67%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1331      +/-   ##
==========================================
- Coverage   90.21%   89.55%   -0.67%     
==========================================
  Files         106      111       +5     
  Lines       16339    17537    +1198     
  Branches       36       41       +5     
==========================================
+ Hits        14740    15705     +965     
- Misses       1592     1812     +220     
- Partials        7       20      +13

Files	Coverage Δ
python/pydantic_core/core_schema.py	`94.76% <100.00%> (-0.01%)`	⬇️
src/errors/types.rs	`99.44% <100.00%> (+<0.01%)`	⬆️
src/input/input_abstract.rs	`42.85% <ø> (-27.39%)`	⬇️
src/input/input_json.rs	`90.50% <100.00%> (+1.57%)`	⬆️
src/input/input_python.rs	`97.40% <100.00%> (+0.19%)`	⬆️
src/validators/mod.rs	`96.06% <100.00%> (+0.03%)`	⬆️
src/input/input_string.rs	`47.80% <80.00%> (+9.56%)`	⬆️
src/serializers/ob_type.rs	`82.97% <50.00%> (+0.86%)`	⬆️
src/serializers/shared.rs	`78.07% <0.00%> (-1.14%)`	⬇️
src/validators/complex.rs	`97.67% <97.67%> (ø)`
... and 3 more

... and 34 files with indirect coverage changes

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6e96b85...8919589. Read the comment docs.

codspeed-hq · 2024-06-12T21:58:51Z

CodSpeed Performance Report

Merging #1331 will not alter performance

_{Comparing changhc:implement-complex (8919589) with main (863640b)}

Summary

✅ 155 untouched benchmarks

sydney-runkle

@changhc,

Great work thus far, thanks!!

As discussed in pydantic/pydantic#8555, since there is no official representation for complex numbers in JSON, I propose to express complex numbers as dictionaries with two keys, real and imag, with floating point values. Please find examples in the newly added test suites in this PR.

I agree with this approach. @davidhewitt what do you think? I suppose we could have support for either this or the 1+2j format (maybe with a config flag...)

Note that function str_as_complex in shared.rs is not yet implemented as I need to discuss with the team if we want to support this. For example, we might want to accept strings like 1+2j as a valid complex number, as long as these strings follow the format described in python's documentation.

As mentioned on the pydantic PR, I'd be keen to support this.

This implementation, however, will be a bit tricky. Using regular expression is the simplest solution, but it is going to be a bit slow since building regex is rather costly.

Perhaps we can speed things up by using the rust regex crate?

The safest solution is probably to do exactly the same as how cpython parses strings. I think that can be done in a separate PR for easier reviews.

Probably a question for @davidhewitt here.

changhc · 2024-06-13T18:05:24Z

Thanks!

I've looked into the rust regex crate. It has rather limited functionality, specifically lookahead, which means we will need to handle some corner cases ourselves or create some not so straightforward logic. Another crate fancy_regex has better support for regex, but introducing another dependency on a non-native library makes me a bit hesitant unless it's really necessary or beneficial enough. With regex, my biggest concern is whether or not we can capture exactly what is accepted by python. I'm concerned about accidentally accepting/rejecting some strings that are actually rejected/accepted by the complex class in python, which might bring users unexpected problems.

If the team thinks it's okay with regex (and maybe you don't feel like having this cumbersome dictionary representation at all,) I can definitely look into that.

davidhewitt · 2024-06-18T09:14:53Z

What if we avoided using either a regex or num_complex entirely, and we used the Python logic to build complex instances? This is basically what we do already for Decimal, and it seems like a reasonable design to repeat again IMO.

changhc · 2024-06-18T11:10:10Z

That sounds good to me. Then I'll work on that. On the other hand, what do you think about the dict representation I proposed and implemented in this PR? Should we still keep it or should we only accept strings like `1+2j`?

sydney-runkle · 2024-06-18T17:47:11Z

That sounds good to me. Then I'll work on that. On the other hand, what do
you think about the dict representation I proposed and implemented in this
PR? Should we still keep it or should we only accept strings like 1+2j?

We'll chat with the team in our open source sync tomorrow and get back to you :).

sydney-runkle · 2024-06-20T15:07:48Z

That sounds good to me. Then I'll work on that. On the other hand, what do
you think about the dict representation I proposed and implemented in this
PR? Should we still keep it or should we only accept strings like 1+2j?

Let's go with the string representation for now, and if the feature lands with some popularity, we can add support for the dictionary style. For now, if folks really need that dict type serialization for complex numbers, they can use a custom serializer.

changhc · 2024-06-22T22:43:59Z

I've updated the implementation. There is one test case failing with pypy, and I think that's because of some bugs in pypy. How do you usually handle discrepancy between pypy and python? Can we ignore this case for now and simply add a note in the documentation?

Another question regarding test-pydantic-integration. How do I handle this mutual dependency between pydantic and pydantic-core so that this test can pass?

davidhewitt · 2024-06-25T11:15:33Z

Can we ignore this case for now and simply add a note in the documentation?

Yes, we can xfail this case when running on PyPy. If you're willing to also report the case on the PyPy GitHub, it would help them be aware they need to fix :)

Another question regarding test-pydantic-integration. How do I handle this mutual dependency between pydantic and pydantic-core so that this test can pass?

In this case it won't pass until we update pydantic main for these changes, so once we merge this PR we should make a minor release and integrate them into the main repository.

davidhewitt

Thanks, this is looking great! There's a few edge cases to discuss wrt config, constraints, and unions. Beyond that, this implementation looks rock-solid 🪨

src/errors/types.rs

davidhewitt · 2024-06-25T11:22:16Z

src/input/input_python.rs

+    fn validate_complex<'a>(&'a self, _py: Python<'py>) -> ValResult<ValidationMatch<EitherComplex<'py>>> {
+        if let Ok(complex) = self.downcast::<PyComplex>() {
+            return Ok(ValidationMatch::exact(EitherComplex::Py(complex.to_owned())));
+        } else if let Ok(s) = self.downcast::<PyString>() {
+            return Ok(ValidationMatch::exact(EitherComplex::Py(string_to_complex(s, self)?)));
+        } else if self.is_exact_instance_of::<PyFloat>() {
+            return Ok(ValidationMatch::exact(EitherComplex::Complex([
+                self.extract::<f64>().unwrap(),
+                0.0,
+            ])));
+        } else if self.is_exact_instance_of::<PyInt>() {
+            return Ok(ValidationMatch::exact(EitherComplex::Complex([
+                self.extract::<i64>().unwrap() as f64,
+                0.0,
+            ])));
+        }
+        Err(ValError::new(ErrorTypeDefaults::ComplexType, self))
+    }


We should have a think here about what behaviour we expect in strict mode. Probably we only want to accept complex objects in strict. Do we want to accept integers / floats in "strict" mode? In lax, we can accept everything for sure.

Also, similarly the ValidationMatch::exact calls should probably be downgraded to either strict or lax - these will set the level of priority in e.g. complex | int union. I think most would agree that complex | int would expect an integer value to not be cast to a complex. We might want some according tests for some of these union cases.

strict mode is strict in terms of typing, right? I was mostly thinking from the perspective of mathematics when I added float and int. I suppose then we should only allow complex objects in strict mode? Then ValidationMatch::exact is the same as strict. This sounds reasonable to me, but I'm not exactly sure about the how different exact and strict usually are.

One more question about strict mode: how does it apply to other input types? Is it like when the input is in JSON, we only accept strings in strict mode?

I suppose then we should only allow complex objects in strict mode?

I think so, yes (note that in JSON it's less straightforward as you note). In JSON I think we only accept strings in strict mode, yes.

Then ValidationMatch::exact is the same as strict.

IIRC in other cases I made subclasses strict, if complex objects allow subclassing. I'd check what I did for ints or floats.

davidhewitt · 2024-06-25T11:26:25Z

src/validators/complex.rs

+}
+
+#[derive(Debug)]
+pub struct ComplexValidator {}


I think we probably need a strict mode flag here at the very least.

What about constraints like Le, or MultipleOf? They might get painful, so I'm ok to defer those to later as long as they fail gracefully.

Do we want to support constraints for complex numbers?

I thought about constraints initially when I checked other validators to get some ideas of how to implement this complex validator, but I don't think constraints are going to be useful. For example, Le and relevant constraints don't really make sense as complex numbers don't have a default ordering. Users can apply additional filtering ad hoc based on the ordering they define, but I don't think we can add that kind of constraints to the validator.

I am happy for us to not support constraints for now.

tests/serializers/test_complex.py

changhc · 2024-07-07T08:53:57Z

Hi @davidhewitt, how should we proceed with this? Do we wait for input from the team or should we manage to make a decision here ourselves?

sydney-runkle · 2024-08-02T14:08:20Z

Hey @changhc,

Sorry we've dropped the ball on feedback here. Will chat with the team early next week so that we can move this forward and include it in our v2.9 release 🚀 !

changhc · 2024-08-04T23:18:33Z

Thanks @sydney-runkle! I'm making some changes to address David's comments. They will be ready soon!

changhc · 2024-08-06T21:04:13Z

Hi @davidhewitt, I've implemented the strict mode for complex numbers. As we discussed, when strict mode is on,

only complex objects are accepted for python input.
only complex strings are accepted for JSON input.

Because of the strict mode, I tweaked the messages for validation errors a little bit to make errors less confusing to users. Specifically for python input,

when strict mode is on, tell users that only python complex objects are accepted.
when strict mode is off and the input value is an invalid complex string, give the generic message explaining all acceptable input values instead of telling users to correct the string since at this point we are not sure if users actually want to place a complex string or just give a value of an incorrect type.

For other input types, the error is rather simple as complex strings are the only acceptable input values when strict mode is on.

I also added some test cases for dictionaries involving complex keys just to make sure things work as expected.

davidhewitt

This generally looks great to me, thanks.

The only change I'd like to see would be to remove the custom py_strict error type and use an instance-of error instead.

davidhewitt · 2024-08-08T08:30:04Z

src/input/input_python.rs

+            return Ok(ValidationMatch::strict(EitherComplex::Py(complex.to_owned())));
+        }
+        if strict {
+            return Err(ValError::new(ErrorTypeDefaults::ComplexTypePyStrict, self));


I think rather than a new type here, we can use isinstance error:

Suggested change

return Err(ValError::new(ErrorTypeDefaults::ComplexTypePyStrict, self));

return Err(ValError::new(

ErrorType::IsInstanceOf {

class: PyComplex::type_object(py).qualname()

.and_then(|name| name.extract())

.unwrap_or_else(|_| "complex".to_owned()),

context: None,

},

self,

))

Updated. The error message now looks like:

pydantic_core._pydantic_core.ValidationError: 1 validation error for complex Input should be an instance of complex [type=is_instance_of, input_value='foobar', input_type=str] For further information visit https://errors.pydantic.dev/2.8/v/is_instance_of

sydney-runkle · 2024-08-13T16:50:54Z

@davidhewitt, are we ready to move this across the line?

davidhewitt · 2024-08-13T16:54:15Z

Yep, LGTM! Thanks @changhc

changhc · 2024-08-13T22:04:12Z

Thanks! I'll update pydantic/pydantic#9654 once this PR is merged and the next minor release includes these changes.

sydney-runkle · 2024-08-14T01:51:51Z

Great! I'll get this merged tomorrow 🚀

sydney-runkle · 2024-08-15T14:36:24Z

Great work @changhc, going to go ahead and merge this - the failing integration tests should soon be fixed with your branch.

Do we need to make any more updates to pydantic/pydantic#9654 other than supporting a new version of pydantic-core with this change?

This PR contains the following updates: | Package | Change | Age | Adoption | Passing | Confidence | |---|---|---|---|---|---| | [pydantic](https://redirect.github.com/pydantic/pydantic) ([changelog](https://docs.pydantic.dev/latest/changelog/)) | `2.8.2` -> `2.9.0` | [![age](https://developer.mend.io/api/mc/badges/age/pypi/pydantic/2.9.0?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![adoption](https://developer.mend.io/api/mc/badges/adoption/pypi/pydantic/2.9.0?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![passing](https://developer.mend.io/api/mc/badges/compatibility/pypi/pydantic/2.8.2/2.9.0?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![confidence](https://developer.mend.io/api/mc/badges/confidence/pypi/pydantic/2.8.2/2.9.0?slim=true)](https://docs.renovatebot.com/merge-confidence/) | --- ### Release Notes <details> <summary>pydantic/pydantic (pydantic)</summary> ### [`v2.9.0`](https://redirect.github.com/pydantic/pydantic/blob/HEAD/HISTORY.md#v290-2024-09-05) [Compare Source](https://redirect.github.com/pydantic/pydantic/compare/v2.8.2...v2.9.0) [GitHub release](https://redirect.github.com/pydantic/pydantic/releases/tag/v2.9.0) The code released in v2.9.0 is practically identical to that of v2.9.0b2. ##### What's Changed ##### Packaging - Bump `ruff` to `v0.5.0` and `pyright` to `v1.1.369` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9801](https://redirect.github.com/pydantic/pydantic/pull/9801) - Bump `pydantic-extra-types` to `v2.9.0` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9832](https://redirect.github.com/pydantic/pydantic/pull/9832) - Support compatibility with `pdm v2.18.1` by [@Viicos](https://redirect.github.com/Viicos) in [#10138](https://redirect.github.com/pydantic/pydantic/pull/10138) - Bump `v1` version stub to `v1.10.18` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10214](https://redirect.github.com/pydantic/pydantic/pull/10214) - Bump `pydantic-core` to `v2.23.2` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10311](https://redirect.github.com/pydantic/pydantic/pull/10311) ##### New Features - Add support for `ZoneInfo` by [@Youssefares](https://redirect.github.com/Youssefares) in [#9896](https://redirect.github.com/pydantic/pydantic/pull/9896) - Add `Config.val_json_bytes` by [@josh-newman](https://redirect.github.com/josh-newman) in [#9770](https://redirect.github.com/pydantic/pydantic/pull/9770) - Add DSN for Snowflake by [@aditkumar72](https://redirect.github.com/aditkumar72) in [#10128](https://redirect.github.com/pydantic/pydantic/pull/10128) - Support `complex` number by [@changhc](https://redirect.github.com/changhc) in [#9654](https://redirect.github.com/pydantic/pydantic/pull/9654) - Add support for `annotated_types.Not` by [@aditkumar72](https://redirect.github.com/aditkumar72) in [#10210](https://redirect.github.com/pydantic/pydantic/pull/10210) - Allow `WithJsonSchema` to inject `$ref`s w/ `http` or `https` links by [@dAIsySHEng1](https://redirect.github.com/dAIsySHEng1) in [#9863](https://redirect.github.com/pydantic/pydantic/pull/9863) - Allow validators to customize validation JSON schema by [@Viicos](https://redirect.github.com/Viicos) in [#10094](https://redirect.github.com/pydantic/pydantic/pull/10094) - Support parametrized `PathLike` types by [@nix010](https://redirect.github.com/nix010) in [#9764](https://redirect.github.com/pydantic/pydantic/pull/9764) - Add tagged union serializer that attempts to use `str` or `callable` discriminators to select the correct serializer by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in in [pydantic/pydantic-core#1397](https://redirect.github.com/pydantic/pydantic-core/pull/1397) ##### Changes - Breaking Change: Merge `dict` type `json_schema_extra` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9792](https://redirect.github.com/pydantic/pydantic/pull/9792) - For more info (how to replicate old behavior) on this change, see [here](https://docs.pydantic.dev/dev/concepts/json_schema/#merging-json_schema_extra) - Refactor annotation injection for known (often generic) types by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9979](https://redirect.github.com/pydantic/pydantic/pull/9979) - Move annotation compatibility errors to validation phase by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9999](https://redirect.github.com/pydantic/pydantic/pull/9999) - Improve runtime errors for string constraints like `pattern` for incompatible types by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10158](https://redirect.github.com/pydantic/pydantic/pull/10158) - Remove `'allOf'` JSON schema workarounds by [@dpeachey](https://redirect.github.com/dpeachey) in [#10029](https://redirect.github.com/pydantic/pydantic/pull/10029) - Remove `typed_dict_cls` data from `CoreMetadata` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10180](https://redirect.github.com/pydantic/pydantic/pull/10180) - Deprecate passing a dict to the `Examples` class by [@Viicos](https://redirect.github.com/Viicos) in [#10181](https://redirect.github.com/pydantic/pydantic/pull/10181) - Remove `initial_metadata` from internal metadata construct by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10194](https://redirect.github.com/pydantic/pydantic/pull/10194) - Use `re.Pattern.search` instead of `re.Pattern.match` for consistency with `rust` behavior by [@tinez](https://redirect.github.com/tinez) in [pydantic/pydantic-core#1368](https://redirect.github.com/pydantic/pydantic-core/pull/1368) - Show value of wrongly typed data in `pydantic-core` serialization warning by [@BoxyUwU](https://redirect.github.com/BoxyUwU) in [pydantic/pydantic-core#1377](https://redirect.github.com/pydantic/pydantic-core/pull/1377) - Breaking Change: in `pydantic-core`, change `metadata` type hint in core schemas from `Any` -> `Dict[str, Any] | None` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [pydantic/pydantic-core#1411](https://redirect.github.com/pydantic/pydantic-core/pull/1411) - Raise helpful warning when `self` isn't returned from model validator by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10255](https://redirect.github.com/pydantic/pydantic/pull/10255) ##### Performance - Initial start at improving import times for modules, using caching primarily by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10009](https://redirect.github.com/pydantic/pydantic/pull/10009) - Using cached internal import for `BaseModel` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10013](https://redirect.github.com/pydantic/pydantic/pull/10013) - Simplify internal generics logic - remove generator overhead by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10059](https://redirect.github.com/pydantic/pydantic/pull/10059) - Remove default module globals from types namespace by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10123](https://redirect.github.com/pydantic/pydantic/pull/10123) - Performance boost: skip caching parent namespaces in most cases by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10113](https://redirect.github.com/pydantic/pydantic/pull/10113) - Update ns stack with already copied ns by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10267](https://redirect.github.com/pydantic/pydantic/pull/10267) ##### Minor Internal Improvements - ⚡️ Speed up `multiple_of_validator()` by 31% in `pydantic/_internal/_validators.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9839](https://redirect.github.com/pydantic/pydantic/pull/9839) - ⚡️ Speed up `ModelPrivateAttr.__set_name__()` by 18% in `pydantic/fields.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9841](https://redirect.github.com/pydantic/pydantic/pull/9841) - ⚡️ Speed up `dataclass()` by 7% in `pydantic/dataclasses.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9843](https://redirect.github.com/pydantic/pydantic/pull/9843) - ⚡️ Speed up function `_field_name_for_signature` by 37% in `pydantic/_internal/_signature.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9951](https://redirect.github.com/pydantic/pydantic/pull/9951) - ⚡️ Speed up method `GenerateSchema._unpack_refs_defs` by 26% in `pydantic/_internal/_generate_schema.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9949](https://redirect.github.com/pydantic/pydantic/pull/9949) - ⚡️ Speed up function `apply_each_item_validators` by 100% in `pydantic/_internal/_generate_schema.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9950](https://redirect.github.com/pydantic/pydantic/pull/9950) - ⚡️ Speed up method `ConfigWrapper.core_config` by 28% in `pydantic/_internal/_config.py` by [@misrasaurabh1](https://redirect.github.com/misrasaurabh1) in [#9953](https://redirect.github.com/pydantic/pydantic/pull/9953) ##### Fixes - Respect `use_enum_values` on `Literal` types by [@kwint](https://redirect.github.com/kwint) in [#9787](https://redirect.github.com/pydantic/pydantic/pull/9787) - Prevent type error for exotic `BaseModel/RootModel` inheritance by [@dmontagu](https://redirect.github.com/dmontagu) in [#9913](https://redirect.github.com/pydantic/pydantic/pull/9913) - Fix typing issue with field_validator-decorated methods by [@dmontagu](https://redirect.github.com/dmontagu) in [#9914](https://redirect.github.com/pydantic/pydantic/pull/9914) - Replace `str` type annotation with `Any` in validator factories in documentation on validators by [@maximilianfellhuber](https://redirect.github.com/maximilianfellhuber) in [#9885](https://redirect.github.com/pydantic/pydantic/pull/9885) - Fix `ComputedFieldInfo.wrapped_property` pointer when a property setter is assigned by [@tlambert03](https://redirect.github.com/tlambert03) in [#9892](https://redirect.github.com/pydantic/pydantic/pull/9892) - Fix recursive typing of `main.IncEnx` by [@tlambert03](https://redirect.github.com/tlambert03) in [#9924](https://redirect.github.com/pydantic/pydantic/pull/9924) - Allow usage of `type[Annotated[...]]` by [@Viicos](https://redirect.github.com/Viicos) in [#9932](https://redirect.github.com/pydantic/pydantic/pull/9932) - `mypy` plugin: handle frozen fields on a per-field basis by [@dmontagu](https://redirect.github.com/dmontagu) in [#9935](https://redirect.github.com/pydantic/pydantic/pull/9935) - Fix typo in `invalid-annotated-type` error code by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9948](https://redirect.github.com/pydantic/pydantic/pull/9948) - Simplify schema generation for `uuid`, `url`, and `ip` types by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9975](https://redirect.github.com/pydantic/pydantic/pull/9975) - Move `date` schemas to `_generate_schema.py` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9976](https://redirect.github.com/pydantic/pydantic/pull/9976) - Move `decimal.Decimal` validation to `_generate_schema.py` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9977](https://redirect.github.com/pydantic/pydantic/pull/9977) - Simplify IP address schema in `_std_types_schema.py` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9959](https://redirect.github.com/pydantic/pydantic/pull/9959) - Fix type annotations for some potentially generic `GenerateSchema.match_type` options by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#9961](https://redirect.github.com/pydantic/pydantic/pull/9961) - Add class name to "has conflict" warnings by [@msabramo](https://redirect.github.com/msabramo) in [#9964](https://redirect.github.com/pydantic/pydantic/pull/9964) - Fix `dataclass` ignoring `default_factory` passed in Annotated by [@kc0506](https://redirect.github.com/kc0506) in [#9971](https://redirect.github.com/pydantic/pydantic/pull/9971) - Fix `Sequence` ignoring `discriminator` by [@kc0506](https://redirect.github.com/kc0506) in [#9980](https://redirect.github.com/pydantic/pydantic/pull/9980) - Fix typing for `IPvAnyAddress` and `IPvAnyInterface` by [@haoyun](https://redirect.github.com/haoyun) in [#9990](https://redirect.github.com/pydantic/pydantic/pull/9990) - Fix false positives on v1 models in `mypy` plugin for `from_orm` check requiring from_attributes=True config by [@radekwlsk](https://redirect.github.com/radekwlsk) in [#9938](https://redirect.github.com/pydantic/pydantic/pull/9938) - Apply `strict=True` to `__init__` in `mypy` plugin by [@kc0506](https://redirect.github.com/kc0506) in [#9998](https://redirect.github.com/pydantic/pydantic/pull/9998) - Refactor application of `deque` annotations by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10018](https://redirect.github.com/pydantic/pydantic/pull/10018) - Raise a better user error when failing to evaluate a forward reference by [@Viicos](https://redirect.github.com/Viicos) in [#10030](https://redirect.github.com/pydantic/pydantic/pull/10030) - Fix evaluation of `__pydantic_extra__` annotation in specific circumstances by [@Viicos](https://redirect.github.com/Viicos) in [#10070](https://redirect.github.com/pydantic/pydantic/pull/10070) - Fix `frozen` enforcement for `dataclasses` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10066](https://redirect.github.com/pydantic/pydantic/pull/10066) - Remove logic to handle unused `__get_pydantic_core_schema__` signature by [@Viicos](https://redirect.github.com/Viicos) in [#10075](https://redirect.github.com/pydantic/pydantic/pull/10075) - Use `is_annotated` consistently by [@Viicos](https://redirect.github.com/Viicos) in [#10095](https://redirect.github.com/pydantic/pydantic/pull/10095) - Fix `PydanticDeprecatedSince26` typo by [@kc0506](https://redirect.github.com/kc0506) in [#10101](https://redirect.github.com/pydantic/pydantic/pull/10101) - Improve `pyright` tests, refactor model decorators signatures by [@Viicos](https://redirect.github.com/Viicos) in [#10092](https://redirect.github.com/pydantic/pydantic/pull/10092) - Fix `ip` serialization logic by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10112](https://redirect.github.com/pydantic/pydantic/pull/10112) - Warn when frozen defined twice for `dataclasses` by [@mochi22](https://redirect.github.com/mochi22) in [#10082](https://redirect.github.com/pydantic/pydantic/pull/10082) - Do not compute JSON Schema default when plain serializers are used with `when_used` set to `'json-unless-none'` and the default value is `None` by [@Viicos](https://redirect.github.com/Viicos) in [#10121](https://redirect.github.com/pydantic/pydantic/pull/10121) - Fix `ImportString` special cases by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10137](https://redirect.github.com/pydantic/pydantic/pull/10137) - Blacklist default globals to support exotic user code with `__` prefixed annotations by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10136](https://redirect.github.com/pydantic/pydantic/pull/10136) - Handle `nullable` schemas with `serialization` schema available during JSON Schema generation by [@Viicos](https://redirect.github.com/Viicos) in [#10132](https://redirect.github.com/pydantic/pydantic/pull/10132) - Reorganize `BaseModel` annotations by [@kc0506](https://redirect.github.com/kc0506) in [#10110](https://redirect.github.com/pydantic/pydantic/pull/10110) - Fix core schema simplification when serialization schemas are involved in specific scenarios by [@Viicos](https://redirect.github.com/Viicos) in [#10155](https://redirect.github.com/pydantic/pydantic/pull/10155) - Add support for stringified annotations when using `PrivateAttr` with `Annotated` by [@Viicos](https://redirect.github.com/Viicos) in [#10157](https://redirect.github.com/pydantic/pydantic/pull/10157) - Fix JSON Schema `number` type for literal and enum schemas by [@Viicos](https://redirect.github.com/Viicos) in [#10172](https://redirect.github.com/pydantic/pydantic/pull/10172) - Fix JSON Schema generation of fields with plain validators in serialization mode by [@Viicos](https://redirect.github.com/Viicos) in [#10167](https://redirect.github.com/pydantic/pydantic/pull/10167) - Fix invalid JSON Schemas being generated for functions in certain scenarios by [@Viicos](https://redirect.github.com/Viicos) in [#10188](https://redirect.github.com/pydantic/pydantic/pull/10188) - Make sure generated JSON Schemas are valid in tests by [@Viicos](https://redirect.github.com/Viicos) in [#10182](https://redirect.github.com/pydantic/pydantic/pull/10182) - Fix key error with custom serializer by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10200](https://redirect.github.com/pydantic/pydantic/pull/10200) - Add 'wss' for allowed schemes in NatsDsn by [@swelborn](https://redirect.github.com/swelborn) in [#10224](https://redirect.github.com/pydantic/pydantic/pull/10224) - Fix `Mapping` and `MutableMapping` annotations to use mapping schema instead of dict schema by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10020](https://redirect.github.com/pydantic/pydantic/pull/10020) - Fix JSON Schema generation for constrained dates by [@Viicos](https://redirect.github.com/Viicos) in [#10185](https://redirect.github.com/pydantic/pydantic/pull/10185) - Fix discriminated union bug regression when using enums by [@kfreezen](https://redirect.github.com/kfreezen) in [pydantic/pydantic-core#1286](https://redirect.github.com/pydantic/pydantic-core/pull/1286) - Fix `field_serializer` with computed field when using `*` by [@nix010](https://redirect.github.com/nix010) in [pydantic/pydantic-core#1349](https://redirect.github.com/pydantic/pydantic-core/pull/1349) - Try each option in `Union` serializer before inference by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [pydantic/pydantic-core#1398](https://redirect.github.com/pydantic/pydantic-core/pull/1398) - Fix `float` serialization behavior in `strict` mode by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [pydantic/pydantic-core#1400](https://redirect.github.com/pydantic/pydantic-core/pull/1400) - Introduce `exactness` into Decimal validation logic to improve union validation behavior by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in in [pydantic/pydantic-core#1405](https://redirect.github.com/pydantic/pydantic-core/pull/1405) - Fix new warnings assertions to use `pytest.warns()` by [@mgorny](https://redirect.github.com/mgorny) in [#10241](https://redirect.github.com/pydantic/pydantic/pull/10241) - Fix a crash when cleaning the namespace in `ModelMetaclass` by [@Viicos](https://redirect.github.com/Viicos) in [#10242](https://redirect.github.com/pydantic/pydantic/pull/10242) - Fix parent namespace issue with model rebuilds by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10257](https://redirect.github.com/pydantic/pydantic/pull/10257) - Remove defaults filter for namespace by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10261](https://redirect.github.com/pydantic/pydantic/pull/10261) - Use identity instead of equality after validating model in `__init__` by [@Viicos](https://redirect.github.com/Viicos) in [#10264](https://redirect.github.com/pydantic/pydantic/pull/10264) - Support `BigInt` serialization for `int` subclasses by [@kxx317](https://redirect.github.com/kxx317) in [pydantic/pydantic-core#1417](https://redirect.github.com/pydantic/pydantic-core/pull/1417) - Support signature for wrap validators without `info` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10277](https://redirect.github.com/pydantic/pydantic/pull/10277) - Ensure `__pydantic_complete__` is set when rebuilding `dataclasses` by [@Viicos](https://redirect.github.com/Viicos) in [#10291](https://redirect.github.com/pydantic/pydantic/pull/10291) - Respect `schema_generator` config value in `TypeAdapter` by [@sydney-runkle](https://redirect.github.com/sydney-runkle) in [#10300](https://redirect.github.com/pydantic/pydantic/pull/10300) ##### New Contributors ##### `pydantic` - [@kwint](https://redirect.github.com/kwint) made their first contribution in [#9787](https://redirect.github.com/pydantic/pydantic/pull/9787) - [@seekinginfiniteloop](https://redirect.github.com/seekinginfiniteloop) made their first contribution in [#9822](https://redirect.github.com/pydantic/pydantic/pull/9822) - [@a-alexander](https://redirect.github.com/a-alexander) made their first contribution in [#9848](https://redirect.github.com/pydantic/pydantic/pull/9848) - [@maximilianfellhuber](https://redirect.github.com/maximilianfellhuber) made their first contribution in [#9885](https://redirect.github.com/pydantic/pydantic/pull/9885) - [@karmaBonfire](https://redirect.github.com/karmaBonfire) made their first contribution in [#9945](https://redirect.github.com/pydantic/pydantic/pull/9945) - [@s-rigaud](https://redirect.github.com/s-rigaud) made their first contribution in [#9958](https://redirect.github.com/pydantic/pydantic/pull/9958) - [@msabramo](https://redirect.github.com/msabramo) made their first contribution in [#9964](https://redirect.github.com/pydantic/pydantic/pull/9964) - [@DimaCybr](https://redirect.github.com/DimaCybr) made their first contribution in [#9972](https://redirect.github.com/pydantic/pydantic/pull/9972) - [@kc0506](https://redirect.github.com/kc0506) made their first contribution in [#9971](https://redirect.github.com/pydantic/pydantic/pull/9971) - [@haoyun](https://redirect.github.com/haoyun) made their first contribution in [#9990](https://redirect.github.com/pydantic/pydantic/pull/9990) - [@radekwlsk](https://redirect.github.com/radekwlsk) made their first contribution in [#9938](https://redirect.github.com/pydantic/pydantic/pull/9938) - [@dpeachey](https://redirect.github.com/dpeachey) made their first contribution in [#10029](https://redirect.github.com/pydantic/pydantic/pull/10029) - [@BoxyUwU](https://redirect.github.com/BoxyUwU) made their first contribution in [#10085](https://redirect.github.com/pydantic/pydantic/pull/10085) - [@mochi22](https://redirect.github.com/mochi22) made their first contribution in [#10082](https://redirect.github.com/pydantic/pydantic/pull/10082) - [@aditkumar72](https://redirect.github.com/aditkumar72) made their first contribution in [#10128](https://redirect.github.com/pydantic/pydantic/pull/10128) - [@changhc](https://redirect.github.com/changhc) made their first contribution in [#9654](https://redirect.github.com/pydantic/pydantic/pull/9654) - [@insumanth](https://redirect.github.com/insumanth) made their first contribution in [#10229](https://redirect.github.com/pydantic/pydantic/pull/10229) - [@AdolfoVillalobos](https://redirect.github.com/AdolfoVillalobos) made their first contribution in [#10240](https://redirect.github.com/pydantic/pydantic/pull/10240) - [@bllchmbrs](https://redirect.github.com/bllchmbrs) made their first contribution in [#10270](https://redirect.github.com/pydantic/pydantic/pull/10270) ##### `pydantic-core` - [@kfreezen](https://redirect.github.com/kfreezen) made their first contribution in [pydantic/pydantic-core#1286](https://redirect.github.com/pydantic/pydantic-core/pull/1286) - [@tinez](https://redirect.github.com/tinez) made their first contribution in [pydantic/pydantic-core#1368](https://redirect.github.com/pydantic/pydantic-core/pull/1368) - [@fft001](https://redirect.github.com/fft001) made their first contribution in [pydantic/pydantic-core#1362](https://redirect.github.com/pydantic/pydantic-core/pull/1362) - [@nix010](https://redirect.github.com/nix010) made their first contribution in [pydantic/pydantic-core#1349](https://redirect.github.com/pydantic/pydantic-core/pull/1349) - [@BoxyUwU](https://redirect.github.com/BoxyUwU) made their first contribution in [pydantic/pydantic-core#1379](https://redirect.github.com/pydantic/pydantic-core/pull/1379) - [@candleindark](https://redirect.github.com/candleindark) made their first contribution in [pydantic/pydantic-core#1404](https://redirect.github.com/pydantic/pydantic-core/pull/1404) - [@changhc](https://redirect.github.com/changhc) made their first contribution in [pydantic/pydantic-core#1331](https://redirect.github.com/pydantic/pydantic-core/pull/1331) </details> --- ### Configuration 📅 **Schedule**: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Enabled. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about this update again. --- - [ ] If you want to rebase/retry this PR, check this box --- This PR was generated by [Mend Renovate](https://mend.io/renovate/). View the [repository job log](https://developer.mend.io/github/spiraldb/ziggy-pydust).  Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

changhc added 3 commits June 12, 2024 22:06

implement complex

441188e

Merge branch 'main' into implement-complex

a60b6a0

fix tests

ca12c2b

handle real numbers

1e7dbd9

sydney-runkle reviewed Jun 13, 2024

View reviewed changes

changhc added 7 commits June 22, 2024 22:16

handle complex strings

419ec8d

add test cases

930519e

fix tests

8529a72

fix typo

c68a975

add error type

c3bc5c7

fix tests

e589080

fix tests

90b0946

davidhewitt reviewed Jun 25, 2024

View reviewed changes

changhc added 2 commits June 25, 2024 22:14

xfail test for pypy

1ea4bd9

fix format

7da4284

sydney-runkle mentioned this pull request Jul 30, 2024

pydantic weekly project plan (Sydney Runkle) - July 29th, 2024 pydantic/pydantic#9994

Closed

15 tasks

changhc added 2 commits August 5, 2024 00:30

strict mode

7eb84c8

update tests

ae774c4

changhc added 3 commits August 6, 2024 21:14

Merge branch 'main' into implement-complex

ee119b0

fix tests

9d373ef

update tests

cc53b24

fix tests

6c8c222

changhc marked this pull request as ready for review August 6, 2024 21:12

davidhewitt reviewed Aug 8, 2024

View reviewed changes

use isinstance error

8919589

sydney-runkle self-assigned this Aug 15, 2024

sydney-runkle added the enhancement New feature or request label Aug 15, 2024

sydney-runkle merged commit 3d8295e into pydantic:main Aug 15, 2024
28 of 29 checks passed

vasudev-gm mentioned this pull request Aug 19, 2024

Bump pydantic-core from 2.20.1 to 2.22.0 vasudev-gm/magika_demo#26

Closed

vasudev-gm mentioned this pull request Sep 2, 2024

Bump pydantic-core from 2.20.1 to 2.23.1 vasudev-gm/magika_demo#29

Closed

attevaltojarvi mentioned this pull request Sep 20, 2024

Discriminated Union with callable discriminator and Django models is invalidly serialized pydantic/pydantic#10437

Open

1 task

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support complex numbers #1331

Support complex numbers #1331

changhc commented Jun 12, 2024 •

edited

Loading

codecov bot commented Jun 12, 2024 •

edited

Loading

codspeed-hq bot commented Jun 12, 2024 •

edited

Loading

sydney-runkle left a comment

changhc commented Jun 13, 2024

davidhewitt commented Jun 18, 2024

changhc commented Jun 18, 2024 via email •

edited

Loading

sydney-runkle commented Jun 18, 2024

sydney-runkle commented Jun 20, 2024

changhc commented Jun 22, 2024

davidhewitt commented Jun 25, 2024

davidhewitt left a comment

davidhewitt Jun 25, 2024

changhc Jun 25, 2024

davidhewitt Jul 25, 2024

davidhewitt Jun 25, 2024

changhc Jun 25, 2024

davidhewitt Jul 25, 2024

changhc commented Jul 7, 2024

sydney-runkle commented Aug 2, 2024

changhc commented Aug 4, 2024

changhc commented Aug 6, 2024 •

edited

Loading

davidhewitt left a comment

davidhewitt Aug 8, 2024

changhc Aug 8, 2024

sydney-runkle commented Aug 13, 2024

davidhewitt commented Aug 13, 2024

changhc commented Aug 13, 2024 •

edited

Loading

sydney-runkle commented Aug 14, 2024

sydney-runkle commented Aug 15, 2024

-            return Err(ValError::new(ErrorTypeDefaults::ComplexTypePyStrict, self));
+            return Err(ValError::new(
+                ErrorType::IsInstanceOf {
+                    class: PyComplex::type_object(py).qualname()
+                        .and_then(|name| name.extract())
+                        .unwrap_or_else(|_| "complex".to_owned()),
+                    context: None,
+                },
+                self,
+            ))

Support complex numbers #1331

Support complex numbers #1331

Conversation

changhc commented Jun 12, 2024 • edited Loading

Change Summary

Related issue number

Checklist

codecov bot commented Jun 12, 2024 • edited Loading

Codecov Report

codspeed-hq bot commented Jun 12, 2024 • edited Loading

Merging #1331 will not alter performance

Summary

sydney-runkle left a comment

Choose a reason for hiding this comment

changhc commented Jun 13, 2024

davidhewitt commented Jun 18, 2024

changhc commented Jun 18, 2024 via email • edited Loading

sydney-runkle commented Jun 18, 2024

sydney-runkle commented Jun 20, 2024

changhc commented Jun 22, 2024

davidhewitt commented Jun 25, 2024

davidhewitt left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

changhc commented Jul 7, 2024

sydney-runkle commented Aug 2, 2024

changhc commented Aug 4, 2024

changhc commented Aug 6, 2024 • edited Loading

davidhewitt left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sydney-runkle commented Aug 13, 2024

davidhewitt commented Aug 13, 2024

changhc commented Aug 13, 2024 • edited Loading

sydney-runkle commented Aug 14, 2024

sydney-runkle commented Aug 15, 2024

changhc commented Jun 12, 2024 •

edited

Loading

codecov bot commented Jun 12, 2024 •

edited

Loading

codspeed-hq bot commented Jun 12, 2024 •

edited

Loading

changhc commented Jun 18, 2024 via email •

edited

Loading

changhc commented Aug 6, 2024 •

edited

Loading

changhc commented Aug 13, 2024 •

edited

Loading