chore: sync 1.0-dev to main

[lkawka]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @lkawka, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request integrates a robust extension mechanism across the client and server components, enabling more flexible communication by allowing clients to declare supported extensions. It also improves the developer experience by enhancing the type generation script with versioning capabilities and provides server-side control over request payload sizes. Additionally, the handling of task history length has been made more precise.

Highlights

• Client-side Extension Support: Introduced the ability for clients to specify and pass 'extensions' and 'request_metadata' through various client methods (e.g., send_message, get_task, get_card) to the underlying transport layers. This includes adding an 'extensions' field to ClientConfig.
• Transport Layer Extension Handling: Modified gRPC, JSON-RPC, and REST transports to properly handle and propagate 'extensions'. This involves constructing gRPC metadata or setting the 'X-A2A-Extensions' HTTP header based on the provided extensions, ensuring they reach the server.
• Configurable Max Content Length for JSON-RPC Servers: Added a 'max_content_length' parameter to JSON-RPC server applications (FastAPI and Starlette) allowing for configuration of the maximum allowed size for incoming request payloads, preventing 'Payload too large' errors for oversized requests.
• Enhanced Type Generation Script: The 'generate_types.sh' script now supports specifying the A2A specification version (tag, branch, or commit SHA) via environment variables or command-line flags, and includes validation for remote URL accessibility.
• Refined Task History Management: Adjusted the 'apply_history_length' utility to correctly handle history length parameters, ensuring that a history_length of 1 returns the last message and other positive values are respected.

Ignored Files

• Ignored by pattern: .github/workflows/** (4)
  • .github/workflows/linter.yaml
  • .github/workflows/python-publish.yml
  • .github/workflows/unit-tests.yml
  • .github/workflows/update-a2a-types.yml
• Ignored by pattern: CHANGELOG.md (1)
  • CHANGELOG.md

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request syncs changes from the 1.0-dev branch to main, introducing significant enhancements. The most notable changes include the implementation of an extensions framework across the client stack, allowing for more flexible and extensible communication with agents. Additionally, the server-side applications now support configurable maximum content length for requests, improving robustness. The generate_types.sh script has also been greatly improved with versioning support and better validation.

My review has identified a couple of important issues:

• A potential bug in ClientFactory that could lead to unintended side effects by mutating the factory's configuration.
• A logic error in apply_history_length that incorrectly handles history_length=0.

I've also included a couple of minor suggestions for code style and a typo fix. Overall, these are great improvements to the SDK.

[gemini-code-assist]

The current implementation mutates the ClientFactory's configuration (self._config.extensions). This side effect can lead to unexpected behavior where extensions from one client creation leak into subsequent ones. To avoid this, a new configuration object should be created for each client, combining the factory's base extensions with any provided for the specific create call.

Note: You'll need to add import dataclasses at the top of the file for the suggestion to work.

Suggested change

	all_extensions = self._config.extensions.copy()
	if extensions:
	all_extensions.extend(extensions)
	self._config.extensions = all_extensions
	transport = self._registry[transport_protocol](
	card, transport_url, self._config, interceptors or []
	)
	return BaseClient(
	card, self._config, transport, all_consumers, interceptors or []
	card,
	self._config,
	transport,
	all_consumers,
	interceptors or [],
	)
	final_config = self._config
	if extensions:
	all_extensions = self._config.extensions + extensions
	final_config = dataclasses.replace(self._config, extensions=all_extensions)
	transport = self._registry[transport_protocol](
	card, transport_url, final_config, interceptors or []
	)
	return BaseClient(
	card,
	final_config,
	transport,
	all_consumers,
	interceptors or [],
	)

[gemini-code-assist]

This change introduces a bug. When history_length is 0, the condition history_length > 0 is false, and the original task with full history is returned. The specification for historyLength states that if the value is 0, no history should be returned. The previous implementation handled this correctly. I suggest a fix that correctly handles zero and negative values for history_length.

Suggested change

	if history_length is not None and history_length > 0 and task.history:
	# Limit history to the most recent N messages
	limited_history = (
	task.history[-history_length:] if history_length > 0 else []
	)
	limited_history = task.history[-history_length:]
	# Create a new task instance with limited history
	return task.model_copy(update={'history': limited_history})
	if history_length is not None and task.history:
	if history_length <= 0:
	limited_history = []
	else:
	limited_history = task.history[-history_length:]
	return task.model_copy(update={'history': limited_history})

[gemini-code-assist]

Using if extensions: is more idiomatic and avoids setting an empty header when extensions is an empty list. An empty list is a valid value for extensions, but it probably shouldn't result in an X-A2A-Extensions: header being sent.

Suggested change

	if extensions is not None:
	if extensions:

[gemini-code-assist]

Typo: lager should be larger.

Suggested change

	# Treat payloads lager than allowed as invalid request (-32600) before routing
	# Treat payloads larger than allowed as invalid request (-32600) before routing