[x15sr71]

In raising this pull request, I confirm the following (please check boxes):

• I have read and understood the contributors guide.
• I have checked that another pull request for this purpose does not exist.
• I have considered, and confirmed that this submission will be valuable to others.
• I accept that this submission may not be used, and the pull request closed at the will of the maintainer.
• I give this submission freely, and claim no ownership to its content.

My familiarity with the project is as follows (check one):

• I have never used the project.
• I have used the project briefly.
• I have used the project extensively, but have not contributed previously.
• I am an active contributor to the project.

---

Summary

Add regression test coverage for ATSC XMLTV output generation (--xmltv=1) using the ATSC broadcast sample I uploaded to the platform.

This test exercises:

• ATSC PSI/SI table parsing (VCT, ETT, EIT)
• XMLTV format generation from broadcast EPG data
• Exit code validation (expected_rc=10) with OutputType.null

Technical Details

Sample Selection

Sample 187 (ch29FullTS.ts) – ATSC broadcast sample for XMLTV testing with --xmltv=1 --out=null:

• SHA256: 6b10f8aef895469ec9cb89ecd0c2463eb2442052c32da9d81c6a6f5cec9137b8
• Size: 300.52 MB | Duration: 2m 10s
• ATSC PSI/SI Tables:
  • Virtual Channel Table (VCT) - channels 5.1-5.5 (KPIX-TV, StartTV, etc.)
  • Extended Text Table (ETT) - program titles/descriptions
  • Event Information Table (EIT) - multi-day EPG scheduling
• Captions: EIA-608/708 across 5 video PIDs

I reviewed the TS samples currently available on the platform and did not find any that contain the ATSC PSI/SI tables (VCT, EIT, ETT) needed to exercise the ATSC XMLTV generation paths. For this reason, I recently added an ATSC broadcast sample to provide regression coverage.

Database Deployment Strategy

Following the pattern from the WebVTT PR (#993):

Environment	Sample Used	Mechanism	Purpose
Production	Sample 187	Alembic migration	Real ATSC XMLTV validation
Fresh install	Samples 1-2	install/sample_db.py	Bootstrap without FK errors
CI Tests	Samples 1-2	tests/base.py fixtures	Fast smoke tests

Why no test fixture changes:

• Sample 187 is a 300MB production upload (not in test fixtures)
• Adding it would cause foreign key violations in test DB
• Migration-only registration follows the existing pattern for uploaded samples

Test Characteristics

• OutputType.null: Exit-code-only validation (no golden file comparison)
• expected_rc = 10: Correct ATSC XMLTV processing return code
• ignore = false: Standard validation (no special error handling)
• Category: "General" - Uses existing category for organization
• Idempotent: Migration checks for existing test before inserting (safe to re-run)

Migration Details

• Revision: eb7303e132c0
• Revises: c8f3a2b1d4e5 (latest migration on master)
• Creates: Regression test + category linkage + output metadata
• Downgrade: Clean removal in correct FK order

Related Work

• [FIX]: Restore XMLTV generation for ATSC EIT/VCT streams and correct EIT bounds checks ccextractor#1773 — ATSC XMLTV implementation
• [FIX]: Add ATSC VCT virtual channel numbers and call signs to XMLTV output ccextractor#1837 — ATSC table parsing

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[cfsmp3]

• Migration-head collision: revises head c8f3a2b1d4e5 — but so does #1071. Second to merge → two alembic heads → flask db upgrade breaks.
• Contradictory output row: correct='', ignore=0 will perpetually diff, yet the PR claims "exit-code-only" validation. Pick one (set ignore=1, or store a real hash).

[x15sr71]

Thanks for the review and suggesting the fix — fixed the output-row contradiction and found a MySQL-only blocker, fixed it too:

• Output row (fixed): set ignore=1 → exit-code-only, matching the PR's intent. Checked against CCExtractorTester: with ignore=1 it reports an exit-code-only result and never compares output, so there's no perpetual diff.
• ignore reserved word (new — blocker): the raw INSERT listed ignore unquoted, which passes on SQLite (CI) but fails with ERROR 1064 on MySQL/prod. Backticked it.

Head collision: If #1071 merges first, I'm happy to rebase this PR onto it. Just let me know your preferred order.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[canihavesomecoffee]

Can you rebase this (once more)? :)

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[x15sr71]

I've updated the branch with the latest master. Since this PR was opened, master got a couple of new database migrations, so I pointed this PR's migration at the newest one (e7c9b4a2d1f0). Without that change, the database update step in the deploy would have failed, because it would have found two "latest" migrations.

Before pushing, I checked that:

• the production database is on that latest migration, sample 187 is there and doesn't have this test yet, and the "General" category exists
• the migration now lines up cleanly after the newest one
• on the same MySQL version as production, it adds the test correctly, does nothing if run twice, and removes everything cleanly when rolled back
• running --xmltv=1 --out=null on sample 187 with current master exits with 10, which is what the test expects
• all tests pass on my fork's CI (810 tests, Python 3.12–3.15)

Thanks for taking a look!