dict_kwargs restricted to classes with an unresolved **kwargs

[mauvilsa]

What does this PR do?

dict_kwargs in a subclass spec exists for the keyword arguments that the parameter resolvers can't determine. Resolved parameters are meant to be given through init_args. Until now dict_kwargs accepted any key for any class, so a key that the class doesn't accept was either silently ignored or only showed up as a TypeError at instantiation. Now such a key fails during parsing, unless the class has a **kwargs that the resolvers could not resolve.

Keys that are resolved parameters are still accepted in dict_kwargs and get moved to init_args. This keeps backward compatibility, and it means configs keep working when an improvement to the resolvers turns a previously unresolved parameter into a resolved one.

The parameter resolvers now track whether a **kwargs was left unresolved. They represent it with a VAR_KEYWORD parameter named **, which can't collide with a real name. Cases that count as unresolved: **kwargs not used in the body, forwarded to something that has its own unresolved **kwargs, used in an unsupported way (unsupported assign, super call or call target, given as a keyword parameter), and the assumptions and stubs resolver fallbacks. Unpack[TypedDict] counts as resolved.

This is a behavior change: configs that put a key the class doesn't accept in dict_kwargs used to parse, and now fail. The changelog entry is under Changed for v5.0.0, and the "Unresolved parameters" section of the documentation describes the new behavior.

Before submitting

• Did you read the contributing guideline?
• If you used a coding agent, did you fully understand and validate all generated code and ensure it follows the contributing guidelines?
• Did you update the documentation? (readme and public docstrings)
• Did you write unit tests such that there is 100% coverage on related code? (required for bug fixes and new features)
• Did you verify that new and existing tests pass locally?
• [n/a] If this is a bug fix, did you verify that the tests fail without the code fix?
• Did you make sure that all changes preserve backward compatibility?
• Did you update the CHANGELOG including a pull request link? (not for typos, docs, test updates, or minor internal changes/refactors)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (5e494f1) to head (a2340f4).
⚠️ Report is 1 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff            @@
##              main      #981   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           27        27           
  Lines         9078      9114   +36     
=========================================
+ Hits          9078      9114   +36     

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

[greptile-apps]

Confidence Score: 4/5

The implementation appears safe to merge after correcting the non-blocking changelog wording.

Findings

1.  Malformed changelog sentence ▶

Summary

This PR tracks unresolved **kwargs through parameter resolution and uses that information to restrict subclass dict_kwargs during parsing, help generation, and JSON Schema completion.

• Records whether each generated class parser accepts arbitrary keyword names.
• Moves resolved dict_kwargs parameters into init_args while rejecting unknown keys for closed signatures.
• Preserves permissive behavior for genuinely unresolved forwarding and supported model extras.
• Expands resolver, subclass, Pydantic, stub, and completion-schema coverage.

Diagram

%%{init: {'theme': 'neutral'}}%%
flowchart TD
  A[Resolve class signature] --> B{Unresolved **kwargs?}
  B -->|Yes| C[Record accepted kwargs as True]
  B -->|No| D[Record resolved names and aliases]
  C --> E[Parse subclass specification]
  D --> E
  E --> F[Move resolved dict_kwargs keys to init_args]
  F --> G{Unexpected keys remain?}
  G -->|No| H[Parse init_args and retain valid unresolved kwargs]
  G -->|Yes, arbitrary kwargs accepted| H
  G -->|Yes, closed signature| I[Raise parsing error]

Loading

_{Reviews (1) · Last reviewed commit: "dict_kwargs restricted to classes with a..."}

[sonarqubecloud]

Quality Gate passed

Issues
2 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud