Expose wrapped iterator in SerializationIterator

[yorickvP]

Change Summary

We're currently working around pydantic#8907 by parsing the pointer out of the repr(). To make this workaround slightly better, we'd like to expose the wrapped iterator in SerializationIterator via .iterator.

Related issue number

• Related to: https://github.com/pydantic/pydantic/issues/8907

Checklist

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

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 89.54%. Comparing base (ab503cb) to head (6e3ed61). Report is 211 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1399      +/-   ##
==========================================
- Coverage   90.21%   89.54%   -0.68%     
==========================================
  Files         106      109       +3     
  Lines       16339    17348    +1009     
  Branches       36       41       +5     
==========================================
+ Hits        14740    15534     +794     
- Misses       1592     1794     +202     
- Partials        7       20      +13     

Files with missing lines	Coverage Δ
src/serializers/type_serializers/generator.rs	81.39% <ø> (-13.46%)	:arrow_down:

... and 39 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 39a6b10...6e3ed61. Read the comment docs.

[yorickvP]

Test failure looks unrelated.

[codspeed-hq[bot]]

CodSpeed Performance Report

Merging #1399 will not alter performance

_{Comparing yorickvP:yorickvp/expose-sit-iterator (6e3ed61) with main (39a6b10)}

Summary

✅ 155 untouched benchmarks

[davidhewitt]

I'm not convinced this is a good idea; it seems like it's part of the same old pattern that SerializationIterator is lazy validation that this has become a problem.

Why I don't think this is a good idea: if you call next(my_model.field.iterator) you'll be consuming a value from the underlying iterator without ever validating it. I don't think we want that.

I'd prefer we actually fixed the root problem rather than exposing a backdoor to make workarounds easier.

[sydney-runkle]

Agreed with @davidhewitt here. We're planning a rework of Iterable stuff in v2.11, so going to close this for now.

[yorickvP]

Here's the workaround for anyone else running into this issue in the meantime:

from ctypes import *
import re

def unwrap_SerializationIterator(x: SerializationIterator) -> Any:
    return cast(int(re.findall(r'0x[0-9A-F]+', repr(x), re.I)[0], 16), py_object).value