Consider exclude_none in computed_field serialization

[hramezani]

Related issue: pydantic/pydantic#6549

Selected Reviewer: @adriangb

[codecov]

Codecov Report

Merging #768 (f7037b3) into main (bf0bce0) will decrease coverage by 0.01%.
The diff coverage is 100.00%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #768      +/-   ##
==========================================
- Coverage   93.66%   93.66%   -0.01%     
==========================================
  Files          99       99              
  Lines       14233    14247      +14     
  Branches       25       25              
==========================================
+ Hits        13331    13344      +13     
- Misses        896      897       +1     
  Partials        6        6              

Impacted Files	Coverage Δ
src/serializers/computed_fields.rs	98.30% <100.00%> (+0.04%)	⬆️

... and 7 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 bf0bce0...f7037b3. Read the comment docs.

[codspeed-hq]

CodSpeed Performance Report

Merging #768 will not alter performance

_{Comparing computed_field_serialization_exclude_none (f7037b3) with main (bf0bce0)}

Summary

✅ 126 untouched benchmarks

[hramezani]

please review

[samuelcolvin]

LGTM

[enricotagliani]

I can confirm that the model_dump_json(exclude_none=True) case is not resolved yet as of latest pydantic version (2.4.2).

This behaviour can easily be replicated with the following code:

class Model(BaseModel):
    x: int
    y: int

    @computed_field
    def sum(self) -> int:
        return self.x + self.y

    @computed_field
    def none(self) -> None:
        return None


m = Model(x=1, y=2)
print(
    m.model_dump(exclude_none=True),
    m.model_dump_json(exclude_none=True),
    sep='\n'
)

# {'x': 1, 'y': 2, 'sum': 3}
# {"x":1,"y":2,"sum":3,"none":null}

[davidhewitt]

@enricotagliani I can confirm this reproduces in 2.5.3 still, however I cannot reproduce on main, so I believe your case will be fixed on 2.6.0.

[majadesignz]

I can still reproduce it with 2.14.6. Is there any kind of workaround for that?