Fixes loss when serializing/deserializing large decimals #9180 (take 2)

[ybressler]

Change Summary

Serializing/deserializing large decimal values is lossy. These changes enable expected behavior - no loss in information with large decimals.

Note: This PR replaces work done by #9291

Help Needed

I was able to solve the deserialization part. But I'm having trouble identifying how/where to control for model export to json (serialization is still lossy). <-- that message was on my work in the previous PR. Presently, am not sure where the interface to control serialization/deserialization. Could someone point me in the direction?

Related issue number

fix #9180

Checklist

• [X ] The pull request title is a good summary of the changes - it will be used in the changelog
• Unit tests for the changes exist
• Tests pass on CI
• Documentation reflects the changes where applicable
• My PR is ready to review, please add a comment including the phrase "please review" to assign reviewers

[codspeed-hq]

CodSpeed Performance Report

Merging #9292 will not alter performance

_{Comparing ybressler:ybressler_feat/decoding-decimal-from-json-is-lossy-#9180_ii (b3f0d74) with main (77b0e1c)}

Summary

✅ 13 untouched benchmarks

[sydney-runkle]

Hi @ybressler,

Apologies for the delay here - got distracted with some work on FastUI. My initial thought would be that we'd want to investigate this further in the decimal validator and serializer in pydantic-core. You can find that logic here:

• https://github.com/pydantic/pydantic-core/blob/1cefeae57609a876842b29c3ccba9df14aeae9ad/src/validators/decimal.rs
• https://github.com/pydantic/pydantic-core/blob/1cefeae57609a876842b29c3ccba9df14aeae9ad/src/serializers/type_serializers/decimal.rs

I think @davidhewitt will have the best idea of the issue in core, but feel free to ping me with questions as well!