[Web][Experimental] Add support for cross-origin storage caching

[akaashrp]

1. Add support for cross-origin storage caching
2. Bump version to 0.24.0-dev2

[gemini-code-assist]

Summary of Changes

Hello, 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 introduces experimental support for cross-origin storage caching, enhancing the web runtime's ability to store and retrieve data efficiently across different origins. This feature aims to improve performance by leveraging a new browser API for shared storage. The existing caching infrastructure has been refactored to accommodate this new cache type, providing a unified interface for managing various storage backends and improving flexibility in cache configuration.

Highlights

• Cross-Origin Storage Support: Introduced a new CrossOriginStorage class and associated interfaces to manage cross-origin data caching, leveraging an experimental browser API.
• New Cache Type Integration: Implemented ArtifactCrossOriginStorageCache to integrate the new cross-origin storage mechanism into the existing artifact caching system, expanding the available cache types.
• Refactored Cache Access: Refactored hasTensorInCache and deleteTensorCache functions to support a new TensorCacheAccessOptions object and a createArtifactCache factory function, centralizing and simplifying cache type selection logic.
• Version Bump: Updated the project version in package.json from 0.24.0-dev1 to 0.24.0-dev2.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• web/package.json
  • Updated the project version to 0.24.0-dev2.
• web/src/artifact_cache.ts
  • Defined new types and interfaces for CrossOriginHashDescriptor, CrossOriginStorageHandle, CrossOriginStorageWritable, and CrossOriginStorageAPI.
  • Implemented the CrossOriginStorage class to interact with the experimental navigator.crossOriginStorage API, including methods for matching, putting, and deleting cached items, and resolving hash descriptors.
  • Added ArtifactCacheType enum to include 'cross-origin' as a cache option.
  • Introduced TensorCacheAccessOptions interface for configuring cache access.
  • Created ArtifactCrossOriginStorageCache class, implementing ArtifactCacheTemplate using the CrossOriginStorage backend.
  • Developed createArtifactCache factory function to abstract cache instance creation based on ArtifactCacheType.
  • Refactored hasTensorInCache and deleteTensorCache to accept TensorCacheAccessOptions and utilize the createArtifactCache factory, simplifying cache type selection logic.
  • Updated JSDoc comments for hasTensorInCache and deleteTensorCache to reflect the new cross-origin cache type.
• web/src/index.ts
  • Exported new types (ArtifactCacheType, TensorCacheAccessOptions, FetchTensorCacheOptions) and classes (ArtifactCrossOriginStorageCache, createArtifactCache) from artifact_cache.ts and runtime.ts.
• web/src/runtime.ts
  • Updated imports to use TensorCacheAccessOptions and createArtifactCache.
  • Defined FetchTensorCacheOptions interface, extending TensorCacheAccessOptions.
  • Refactored the fetchTensorCache method in the Instance class to support the new FetchTensorCacheOptions object and handle legacy positional arguments, ensuring compatibility while integrating the new cross-origin cache.

Activity

• No specific activity (comments, reviews, progress updates) has been recorded for this pull request yet.

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 adds support for a new experimental cross-origin storage caching mechanism, introducing ArtifactCrossOriginStorageCache and refactoring public API functions like hasTensorInCache, deleteTensorCache, and fetchTensorCache for a more extensible options-based configuration. However, the current implementation of the hash cache uses a globally accessible Map on globalThis, which can lead to the leakage of sensitive information such as private repository tokens and content hashes to other scripts on the same origin. While the overall API design is improved, there is a suggestion to simplify some of the new logic in fetchTensorCache.

[akaashrp]

cc @tqchen for review