Fix: Correctly load built-in providers when using explicit 'provider' in ModelConfig

[siddikisahil47]

Description

Fixes a crash in _create_model_with_schema where built-in providers were not loaded when ModelConfig was initialized with an explicit provider argument.

Previously, providers.load_builtins_once() and providers.load_plugins_once() were only called inside the else block — meaning they were skipped when config.provider was set. This caused provider resolution to fail with an InferenceConfigError.

Root cause (in langextract/factory.py, _create_model_with_schema):

# BEFORE (buggy):
if config.provider:
    provider_class = router.resolve_provider(config.provider)  # builtins NOT loaded!
else:
    providers.load_builtins_once()   # only called here
    providers.load_plugins_once()
    provider_class = router.resolve(config.model_id)

Fix:

# AFTER (fixed):
providers.load_builtins_once()   # always load first
providers.load_plugins_once()

if config.provider:
    provider_class = router.resolve_provider(config.provider)
else:
    provider_class = router.resolve(config.model_id)

This aligns _create_model_with_schema with create_model, which already handles this correctly (lines 142-149).

Reproduction:

from langextract import factory
config = factory.ModelConfig(model_id='gpt-4o', provider='openai')
factory._create_model_with_schema(config)

Error before fix:

Traceback (most recent call last):
  File "langextract/factory.py", line 224, in _create_model_with_schema
    provider_class = router.resolve_provider(config.provider)
  File "langextract/providers/router.py", line 211, in resolve_provider
    raise exceptions.InferenceConfigError(
langextract.core.exceptions.InferenceConfigError: No provider found matching: 'openai'. Available providers can be listed with list_providers()

After fix: Provider resolves successfully ✅

Fixes #320

Bug fix

How Has This Been Tested?

Added a new unit test file tests/factory_provider_loading_test.py:

$ python -m unittest tests/factory_provider_loading_test.py
.
----------------------------------------------------------------------
Ran 1 test in 2.316s

OK

Test details:

• Clears the provider registry to simulate a fresh state.
• Creates a ModelConfig with explicit provider="openai".
• Calls _create_model_with_schema and asserts that we do not get a "No provider found matching" error.
• Accepts other errors (e.g., missing API key) as success — they prove the provider class was found and loaded.

Code style verified with ./autoformat.sh (all pre-commit hooks passed).

Checklist:

• I have read and acknowledged Google's Open Source Code of conduct.
• I have read the Contributing page, and I either signed the Google Individual CLA or am covered by my company's Corporate CLA.
• I have discussed my proposed solution with code owners in the linked issue(s) and we have agreed upon the general approach.
• I have made any needed documentation changes, or noted in the linked issue(s) that documentation elsewhere needs updating.
• I have added tests, or I have ensured existing tests cover the changes
• I have followed Google's Python Style Guide and ran pylint over the affected code.

[telackey]

I think this is essentially the same as #331 .

[github-actions]

⚠️ Branch Update Required

Your branch is 1 commits behind main. Please update your branch to ensure CI checks run with the latest code:

git fetch origin main
git merge origin/main
git push

Note: Enable "Allow edits by maintainers" to allow automatic updates.

[github-actions]

⚠️ Branch Update Required

Your branch is 1 commits behind main. Please update your branch to ensure CI checks run with the latest code:

git fetch origin main
git merge origin/main
git push

Note: Enable "Allow edits by maintainers" to allow automatic updates.