fix(crossseed): compare Site and Sum fields for anime releases

[s0up4200]

Anime releases use bracket notation for fansub groups and checksums. The rls library parses these into Site and Sum fields respectively, not Group.

Previously, releases from different fansub groups could incorrectly match as cross-seed candidates because the matching logic only checked Group, which was empty for both.

Now releasesMatch() compares Site and Sum fields using the same logic as Group: only reject if both releases have values and they differ.

Summary by CodeRabbit

• Bug Fixes

  • Stricter matching: site identity is now enforced and checksums are required to match when present, reducing incorrect cross-seed matches across trackers.

• Tests

  • Added comprehensive unit tests for anime release matching, covering site comparisons, checksum scenarios, case-insensitivity, and real-world tracker examples.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

Added normalized Site and Sum pre-checks to releasesMatch so mismatched non-empty site or checksum fields cause an early false. Added unit tests covering site and checksum matching, including case-insensitivity and integrated real-world examples.

Changes

Cohort / File(s)	Summary
Matching logic enhancement internal/services/crossseed/matching.go	Added normalized Site and Sum pre-checks in releasesMatch: if source.Site (or source.Sum) is non-empty and differs from candidate.Site (or candidate.Sum) after normalization — or if candidate value is empty — return false early.
Test coverage internal/services/crossseed/matching_anime_test.go	New unit tests (table-driven and integrated examples) validating site and checksum (Sum) matching behavior, including case-insensitivity and real parsed release comparisons (e.g., SubsPlease vs AnoZu).

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

• Review normalization function behavior and ensure consistent case/whitespace handling.
• Check edge cases: empty vs non-empty fields and when candidate.Site or candidate.Sum is missing.
• Verify tests cover negative and positive scenarios and real-world data paths.

Possibly related PRs

• feat(crossseed): add Unicode normalization for title and file matching #742 — Modified normalization/title logic used by releasesMatch; touches the same matching pathway.
• bugfix(cross-seed): large amounts of 0day false positives #678 — Added other field pre-checks (Artist/date) to releasesMatch, similar tightening of matching logic.
• fix(crossseed): normalize punctuation in title matching #718 — Adjusted title normalization in releasesMatch; related to matching heuristics.

Suggested labels

bugfix

Poem

🐰 I nibble bytes and hop through sums,
I sniff the Site where each seed comes from.
When fields align, I quietly hum,
When they don't — I bounce and say “not one!”
Hooray for tidy matches, down the burrow we run.

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	You can run @coderabbitai generate docstrings to improve docstring coverage.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'fix(crossseed): compare Site and Sum fields for anime releases' directly and specifically describes the main change: adding Site and Sum field comparisons for anime release matching.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch fix/crossseed-anime-matching

---

📜 Recent review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 36c42b8 and c9606a3.

📒 Files selected for processing (2)

• internal/services/crossseed/matching.go (1 hunks)
• internal/services/crossseed/matching_anime_test.go (1 hunks)

🚧 Files skipped from review as they are similar to previous changes (1)

• internal/services/crossseed/matching_anime_test.go

🧰 Additional context used

🧠 Learnings (2)

📚 Learning: 2025-11-28T20:32:30.126Z

Learnt from: s0up4200
Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go:209-212
Timestamp: 2025-11-28T20:32:30.126Z
Learning: Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go
Learning: The cross-seed recheck-resume worker intentionally runs for the process lifetime and keys pending entries by hash only. This is acceptable under the current constraint that background seeded-search runs operate on a single instance at a time; graceful shutdown and instanceID|hash keying are deferred by design.

Applied to files:

• internal/services/crossseed/matching.go

📚 Learning: 2025-11-21T21:11:50.633Z

Learnt from: s0up4200
Repo: autobrr/qui PR: 625
File: internal/qbittorrent/sync_manager.go:1112-1123
Timestamp: 2025-11-21T21:11:50.633Z
Learning: In internal/qbittorrent/sync_manager.go, the GetCachedFilesBatch interface documentation (around line 39-40) should specify "uppercase hex" instead of "lowercase hex" to match the actual normalization practice used throughout the codebase (e.g., normalizeHash in internal/services/crossseed/service.go uses strings.ToUpper, and hash filtering uses uppercase).

Applied to files:

• internal/services/crossseed/matching.go

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Run tests

🔇 Additional comments (2)

internal/services/crossseed/matching.go (2)

208-217: Site matching logic is correct and follows Group pattern.

The asymmetric pattern (source-enforced) is appropriate here. When the source has a Site (anime fansub group like [SubsPlease]), the candidate must have the same Site. When the source has no Site (non-anime release), the candidate's Site is ignored. This correctly prevents different fansub groups from cross-seeding while allowing flexibility when Site is not specified.

Note: The past review comment flagged asymmetry concerns, but this implementation correctly follows the same pattern as Group matching (lines 194-206), which is the intended design per the PR description.

---

219-227: Sum matching logic is correct and consistent with Site/Group patterns.

The asymmetric checksum matching is appropriate: when the source has a CRC32 checksum, the candidate must have an identical checksum; when the source lacks a checksum, the candidate's checksum is ignored. This strikes a good balance between strictness (preventing mismatches when checksums are available) and flexibility (allowing releases without checksums to participate in cross-seeding).

The normalizer correctly handles case-insensitive comparison for hex checksums, consistent with the codebase's uppercase hex normalization practices.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

internal/services/crossseed/matching_anime_test.go (1)

12-117: Good table-driven coverage for Site/Sum (including case-insensitivity + missing-field behavior).
Optional: add t.Parallel() inside each t.Run (and/or at the top of each test) since s.releasesMatch should be read-only here.

Also applies to: 119-232

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between dc83102 and 91c54cd.

📒 Files selected for processing (2)

• internal/services/crossseed/matching.go (1 hunks)
• internal/services/crossseed/matching_anime_test.go (1 hunks)

🧰 Additional context used

🧠 Learnings (2)

📚 Learning: 2025-11-28T20:32:30.126Z

Learnt from: s0up4200
Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go:209-212
Timestamp: 2025-11-28T20:32:30.126Z
Learning: Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go
Learning: The cross-seed recheck-resume worker intentionally runs for the process lifetime and keys pending entries by hash only. This is acceptable under the current constraint that background seeded-search runs operate on a single instance at a time; graceful shutdown and instanceID|hash keying are deferred by design.

Applied to files:

• internal/services/crossseed/matching_anime_test.go
• internal/services/crossseed/matching.go

📚 Learning: 2025-11-28T22:21:20.730Z

Learnt from: s0up4200
Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go:2415-2457
Timestamp: 2025-11-28T22:21:20.730Z
Learning: Repo: autobrr/qui PR: 641
File: internal/services/crossseed/service.go
Learning: The determineSavePath function intentionally includes a contentLayout string parameter for future/content-layout branching and API consistency. Its presence is by design even if unused in the current body; do not flag as an issue in reviews.

Applied to files:

• internal/services/crossseed/matching_anime_test.go

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Run tests

🔇 Additional comments (1)

internal/services/crossseed/matching_anime_test.go (1)

234-271: Nice “real-world” regression test (SubsPlease vs AnoZu, plus no-checksum cross-tracker case).
This directly guards the bug described in the PR and complements the table tests well.