Improve provider plugin documentation with plugin creation script (google#144)

* Improve provider plugin documentation with plugin creation script

- Add comprehensive 7-step checklist for provider creation
- Create automated plugin generator script (create_provider_plugin.py)
- Fix sample plugin with proper schema support and apply_schema method
- Update documentation to simplify testing and add community engagement

* Add links to provider plugin generator script in README files

Author: aksg87

examples/custom_provider_plugin/README.md

@@ -4,6 +4,12 @@ This example demonstrates how to create a custom provider plugin that extends La
 
 **Note**: This is an example included in the LangExtract repository for reference. It is not part of the LangExtract package and won't be installed when you `pip install langextract`.
 
+**Automated Creation**: Instead of manually copying this example, use the [provider plugin generator script](../../scripts/create_provider_plugin.py):
+```bash
+python scripts/create_provider_plugin.py MyProvider --with-schema
+```
+This will create a complete plugin structure with all boilerplate code ready for customization.
+
 ## Structure
 
 ```
@@ -133,13 +139,76 @@ result = lx.extract(
 # )
 ```
 
-## Creating Your Own Provider
+## Creating Your Own Provider - Step by Step
+
+### 1. Copy and Rename
+```bash
+# Copy this example directory
+cp -r examples/custom_provider_plugin/ ~/langextract-myprovider/
+
+# Rename the package directory
+cd ~/langextract-myprovider/
+mv langextract_provider_example langextract_myprovider
+```
+
+### 2. Update Package Configuration
+Edit `pyproject.toml`:
+- Change `name = "langextract-myprovider"`
+- Update description and author information
+- Change entry point: `myprovider = "langextract_myprovider:MyProvider"`
+
+### 3. Modify Provider Implementation
+Edit `provider.py`:
+- Change class name from `CustomGeminiProvider` to `MyProvider`
+- Update `@register()` patterns to match your model IDs
+- Replace Gemini API calls with your backend
+- Add any provider-specific parameters
+
+### 4. Add Schema Support (Optional)
+Edit `schema.py`:
+- Rename to `MyProviderSchema`
+- Customize `from_examples()` for your extraction format
+- Update `to_provider_config()` for your API requirements
+- Set `supports_strict_mode` based on your capabilities
+
+### 5. Install and Test
+```bash
+# Install in development mode
+pip install -e .
+
+# Test your provider
+python -c "
+import langextract as lx
+lx.providers.load_plugins_once()
+print('Provider registered:', any('myprovider' in str(e) for e in lx.providers.registry.list_entries()))
+"
+```
+
+### 6. Write Tests
+- Test that your provider loads and handles basic inference
+- Verify schema support works (if implemented)
+- Test error handling for your specific API
+
+### 7. Publish to PyPI and Share with Community
+```bash
+# Build package
+python -m build
+
+# Upload to PyPI
+twine upload dist/*
+```
+
+**Share with the community:**
+- Open an issue on [LangExtract GitHub](https://github.com/google/langextract/issues) to announce your provider and get feedback
+- Consider submitting a PR to add your provider to the community providers list (coming soon)
+
+## Common Pitfalls to Avoid
 
-1. Copy this example as a starting point
-2. Update the provider class name and registration pattern
-3. Replace the Gemini implementation with your own backend
-4. Update package name in `pyproject.toml`
-5. Install and test your plugin
+1. **Forgetting to trigger plugin loading** - Plugins load lazily, use `load_plugins_once()` in tests
+2. **Pattern conflicts** - Avoid patterns that conflict with built-in providers
+3. **Missing dependencies** - List all requirements in `pyproject.toml`
+4. **Schema mismatches** - Test schema generation with real examples
+5. **Not handling None schema** - Provider must clear schema when `apply_schema(None)` is called (see provider.py for implementation)
 
 ## License
 

examples/custom_provider_plugin/langextract_provider_example/provider.py

@@ -68,6 +68,8 @@ def __init__(
       temperature: Sampling temperature.
       **kwargs: Additional parameters.
     """
+    super().__init__()
+
     # TODO: Replace with your own client initialization
     try:
       from google import genai  # pylint: disable=import-outside-toplevel
@@ -97,8 +99,6 @@ def __init__(
 
     self._client = genai.Client(api_key=self.api_key)
 
-    super().__init__()
-
   @classmethod
   def get_schema_class(cls) -> type[lx.schema.BaseSchema] | None:
     """Return our custom schema class.
@@ -111,6 +111,30 @@ def get_schema_class(cls) -> type[lx.schema.BaseSchema] | None:
     """
     return custom_schema.CustomProviderSchema
 
+  def apply_schema(self, schema_instance: lx.schema.BaseSchema | None) -> None:
+    """Apply or clear schema configuration.
+
+    This method is called by LangExtract to dynamically apply schema
+    constraints after the provider is instantiated. It's important to
+    handle both the application of a new schema and clearing (None).
+
+    Args:
+      schema_instance: The schema to apply, or None to clear existing schema.
+    """
+    super().apply_schema(schema_instance)
+
+    if schema_instance:
+      # Apply the new schema configuration
+      config = schema_instance.to_provider_config()
+      self.response_schema = config.get('response_schema')
+      self.enable_structured_output = config.get(
+          'enable_structured_output', False
+      )
+    else:
+      # Clear the schema configuration
+      self.response_schema = None
+      self.enable_structured_output = False
+
   def infer(
       self, batch_prompts: Sequence[str], **kwargs: Any
   ) -> Iterator[Sequence[lx.inference.ScoredOutput]]: