Upgrade to pydantic2

[BenGalewsky]

Background

Pydantic V2 is a ground-up rewrite that offers many new features, performance improvements, and some breaking changes compared to Pydantic V1. The main reasons for the changes include:

• Significant performance improvements: Based on benchmarks, Pydantic v2 is between 4x and 50x faster than Pydantic v1.
• Improved reliability and consistency: Pydantic v2 offers a strict mode that no longer magically coerces data, clear rules on required vs nullable fields, and more flexible validation options.
• New features: Pydantic V2 introduces many new features, such as support for autocomplete in VS Code via dataclass_transform.
• Breaking changes: While the transition to Pydantic V2 requires some breaking changes, the changes are not so numerous or game-changing that they cannot be understood by reading the documentation. The migration guide and new API docs provide detailed information on the changes between V1 and V2.
• Maintainability: The aggressive removals between major versions are likely better for maintainability, considering Pydantic's complexity and the fact that it was basically a rewrite

[codecov]

Codecov Report

Attention: Patch coverage is 97.18310% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 73.88%. Comparing base (5bc25f1) to head (9604c29).

Files	Patch %	Lines
servicex/databinder_models.py	93.93%	2 Missing ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##           3.0_develop     #368      +/-   ##
===============================================
+ Coverage        73.78%   73.88%   +0.10%     
===============================================
  Files               42       42              
  Lines             2300     2305       +5     
===============================================
+ Hits              1697     1703       +6     
+ Misses             603      602       -1     

Flag	Coverage Δ
unittests	73.88% <97.18%> (+0.10%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[matthewfeickert]

Resolves #363

[kyungeonchoi]

Working nicely!

[matthewfeickert]

Once this is merged in, can a 3.0.0-alpha.16 release please be made as well? I can use that to update the AB-dev container images.

[kyungeonchoi]

@matthewfeickert - if you are referring to the ServiceX client version in the ATLAS idap repo that's still in the another branch (3.0_develop_uprootraw). It includes some refactoring so we are planning to merge after having a discussion during the coming ServiceX chat :)

[matthewfeickert]

if you are referring to the ServiceX client version in the ATLAS idap repo

More broadly than the repo but iris-hep/idap-200gbps-atlas#60 in general with the servicex client getting added to the install list in https://github.com/usatlas/analysisbase-dask.

that's still in the another branch (3.0_develop_uprootraw). It includes some refactoring so we are planning to merge after having a discussion during the coming ServiceX chat

I guess I'm confused about how development is done here, but that branch is a separate feature request,

3.0_develop...3.0_develop_uprootraw

,so why isn't that branch just rebased off of 3.0_develop and updated to the new pydantic after this is merged in?

[matthewfeickert]

Thanks @BenGalewsky, for the great PR and for merging this in!

@kyungeonchoi I'm sorry earlier for writing a note hastily and without proper clarity. I didn't mean to suggest that a alpha release be cut from the 3.0_develop_uprootraw branch (until 3.0_develop is merged into main with the 3.0 release I assume that all alpha and rcs will be cut from the 3.0_develop branch). I had misunderstood

It includes some refactoring so we are planning to merge after having a discussion during the coming ServiceX chat

to mean that you were waiting on 3.0_develop_uprootraw to merge it into this PR's branch. But I understand that I was mistaken now, and if I attended the serviceX meetings I probably wouldn't be.

I can't promise that I'll be at the Friday 2024-05-03 meeting, but if the 3.0_develop_uprootraw branch isn't ready to PR tomorrow would it still be possible to make another alpha release now that this PR is in?