Skip to content

Bound browser memory for large Source File downloads #36

Description

@sam2tom

Outcome

Bound browser memory use for large Source File downloads or provide an explicit safe browser-side limit and fallback.

Current behavior

Core can expose Source Files up to 512 MiB. The current Client materializes the response into a complete Uint8Array, and Web then creates a Blob, which can cause a large transient memory peak even though small-file lifecycle behavior is correct.

Acceptance criteria

  • Choose and document a browser-safe strategy: streaming to an approved sink where available, or an explicit lower Web download threshold with a clear fallback.
  • Keep response status, content length, content type, and filename validation fail-closed.
  • Never place bearer credentials in a URL, browser history, or persistent storage.
  • Do not auto-retry an interrupted download.
  • Add focused tests for the selected threshold/streaming behavior and cleanup after success, cancellation, and failure.

Non-goals

  • Changing Core's Source File size contract.
  • Persisting execution-principal credentials in the browser.
  • Treating a partial download as success.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions