Skip to content

fix(fedora): expose recorded reasons and reverse requirements - #224

Merged
PyRo1121 merged 1 commit into
PyRo1121/fix-fedora-size-clifrom
PyRo1121/fix-fedora-why-cli
Sep 5, 2026
Merged

fix(fedora): expose recorded reasons and reverse requirements#224
PyRo1121 merged 1 commit into
PyRo1121/fix-fedora-size-clifrom
PyRo1121/fix-fedora-why-cli

Conversation

@PyRo1121

@PyRo1121 PyRo1121 commented Sep 5, 2026

Copy link
Copy Markdown
Owner

Changes

why now displays the recorded DNF5 installation reason. why --reverse lists current direct requiring packages with native reasons and full identities. Neither mode claims historical causality, an installer identity, or removal safety.

Reuse the bounded native query adapter. Keep reason labels intact, reject malformed rows and unsafe operands, and reject missing or ambiguous selectors. Installed-fact queries ignore exclusion filters; transaction policy is unchanged. Reason-only mode does not fetch reverse requirements.

Verification

  • 31 focused DNF tests passed, including malformed records and reverse-query option injection.
  • Fedora library, binary, and integration-test Clippy passed with warnings denied. No-backend Clippy and the real no-backend why entry-point test passed.
  • The live integration test failed against the previous binary and passed against this candidate in the Fedora 44 KVM guest.
  • Current reverse requirements matched native DNF output. Live examples preserved Group, User, and External User reasons.
  • Installed-metadata exclusion tests and the size native-parity regression passed. Ambiguous local RPM versions were rejected; qualified selection and an empty reverse result worked.
  • The RPM inventory was unchanged across final checks. Formatting and diff checks passed. Cargo builds used two jobs.

Verified debug CLI SHA-256: 1995cd1970ff941f518d953acd86d47de4b6ba89188d96d744aae84c0f7f6c17.

Limits

No historical dependency-chain reconstruction, exhaustive CLI coverage, benchmark, or published-release repair is claimed. Other distro feature combinations and live multilib remain to be verified.

Manual review only. No subagents were used. The automated review tool abstained on this external checkout. Draft pending independent review. No host package changes, merges, releases, or deployments.

@PyRo1121
PyRo1121 marked this pull request as ready for review September 5, 2026 21:01
@PyRo1121
PyRo1121 merged commit dedf54b into PyRo1121/fix-fedora-size-cli Sep 5, 2026
1 check passed
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Sep 5, 2026

Copy link
Copy Markdown

Codex Review Summary

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

Review Status Commit Review trigger
📝 Code Review Completed 2026-09-05T21:02:02.819332Z ee517f2 Draft marked ready
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.

@cursor cursor Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This PR is already merged (dedf54b of ee517f2 into PyRo1121/fix-fedora-size-cli at 2026-09-05T21:01:15Z). I am not reopening, reverting, rebasing, approving, or stacking further work onto this closed PR.

No leftover merge conflict remains. I did not find a leftover non-breaking hole that I can land without changing author-scoped product limits or widening package-name validation.

I will not force these through:

  • Historical dependency-chain reconstruction, an installer identity, or removal-safety claims (explicitly out of scope in the PR)
  • Allowing : in package selectors so full NEVRA epochs can be typed (that would widen validate_package_name)
  • Exhaustive CLI coverage, live multilib, or published-release repair
  • Sibling #219's explicit breaking sha2 / p256 / p384 / toml_edit pin bump

A later author-owned change can take those; this run will not.

Open in Web View Automation 

Sent by Cursor Automation: Untitled

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant