Fix quick-check skip logic for ignore-times parity#2064
Conversation
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
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".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| return false; | ||
| } | ||
|
|
||
| if modify_window.is_zero() { | ||
| return files_content_equal(source_path, destination_path).unwrap_or(false); | ||
| } | ||
|
|
||
| true |
There was a problem hiding this comment.
Regress quick‑check accuracy when modify_window is zero
After removing the content comparison for zero modify_window, should_skip_copy now returns true whenever size and timestamp match, even if file contents differ. This means a local copy will leave stale data when a file is edited without changing its mtime (a case previously covered by should_skip_copy_detects_content_changes_with_identical_timestamps). The earlier files_content_equal check avoided this false skip; without it the copy logic no longer verifies the data and breaks the existing test’s expectation.
Useful? React with 👍 / 👎.
…#3933) Document the four wire-compatible zsync-inspired matching speedups (bithash prefilter #2059-2062, seq-match extend-run #2064-2066, matched-block pruning #2068-2070, compact-keys layout #2072-2073) with CHANGELOG snippets, a release notes block, an anti-spec forbidding wire-feature advertisement, and a placeholder for benchmark numbers pending #2081 and #2082.
…#3933) Document the four wire-compatible zsync-inspired matching speedups (bithash prefilter #2059-2062, seq-match extend-run #2064-2066, matched-block pruning #2068-2070, compact-keys layout #2072-2073) with CHANGELOG snippets, a release notes block, an anti-spec forbidding wire-feature advertisement, and a placeholder for benchmark numbers pending #2081 and #2082.
Summary
Testing
Codex Task