[astral-automations-bot]

uv workspace metadata omits installed packages outside the selected resolution. For example, a package can remain installed after it is removed from the dependency graph; this is especially important for astral-sh/ruff#28192.

This PR adds observed distributions in environment.packages, including their installed versions, metadata paths, and editable status. The inventory includes installed packages outside the locked resolution.

Installed distributions are matched to selected non-virtual packages by name (i.e. they're matched even when their versions differ). Matched entries reuse the selected resolution node ID and report the actual installed version separately. Only unmanaged entries get generated IDs based on their installed name, version, and metadata path. If multiple installations match the same resolution node, uv keeps the first record and warns.

The affected fields, expressed as TypeScript types:

type NodeId = string; // Opaque identifier.

type InstalledPackage = {
  name: string;
  version: string; // Actual installed version.
  path: string; // Absolute path to installed metadata.
  editable: boolean;
};

type Metadata = {
  // ... (other metadata fields omitted)
  environment?: {
    // Existing root and python fields omitted.
    packages: Record<NodeId, InstalledPackage>; // New; present even when empty.
  };

  // Existing. Module name -> owners; omitted when empty.
  // package_id indexes environment.packages;
  // matched IDs also index resolution.
  module_owners?: Record<string, { package_id: NodeId }[]>;
};

Unmanaged packages are entries in environment.packages whose IDs are absent from resolution. Use module_owners to find the modules they provide. Unmatched packages without discovered modules appear only in environment.packages.

[codspeed]

Merging this PR will degrade performance by 0.37%

⚠️ 8 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• discover_workspace_from_all_members_with_excludes
• discover_workspace_from_all_members
• install_wheel_many_files
• resolve_warm_jupyter
• run_python_version_synthetic_workspace
• resolve_warm_airflow
• unpack_sdist_many_files
• resolve_warm_jupyter_universal

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 2 improved benchmarks
❌ 1 regressed benchmark
✅ 51 untouched benchmarks
⏩ 54 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌ ⚠️	Simulation	discover_workspace_from_all_members_with_excludes	140 ms	165.5 ms	-15.41%
⚡ ⚠️	Simulation	discover_workspace_from_all_members	156 ms	142.2 ms	+9.68%
⚡	Simulation	from_hex/uppercase[48]	337 µs	316.1 µs	+6.6%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing zsol/workspace-metadata-installed-packages (37b0674) with main (f61b2fc)}

Footnotes

1. 54 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-14T11:00:51.776278Z	735ac8f	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[astral-sh-bot]

uv test inventory changes

This PR changes the tests when compared with the main base revision.

• Added tests: 4
• Removed tests: 1
• Changed suites: 1

uv::workspace: +4 / -1

Added:

• uv::workspace::workspace_metadata::workspace_metadata_installed_packages_with_duplicate_names
• uv::workspace::workspace_metadata::workspace_metadata_installed_packages_without_modules
• uv::workspace::workspace_metadata::workspace_metadata_module_owners_after_dependency_changes
• uv::workspace::workspace_metadata::workspace_metadata_module_owners_include_stale_virtual_package

Removed:

• uv::workspace::workspace_metadata::workspace_metadata_module_owners_ignore_stale_virtual_package

[zanieb]

I feel like this should be modules with owners as a child array for each item for future expansion?

[zsol]

Yeah we can make that happen separately.

[charliermarsh]

(Separately because it's already this way on main?)

[zsol]

Yes, and ty depends on the current structure. I'll want to make a change to ty first to handle the new format before changing uv