From ad13b0bf5c384449eee932fa8e80a2ac2d1e89fa Mon Sep 17 00:00:00 2001 From: Matt Davis Date: Sat, 7 Nov 2026 16:41:47 -0500 Subject: [PATCH] Apply suggested fix to tests/performance/active-recording.md from Copilot Autofix Co-authored-by: Copilot Autofix powered by AI <223894421+github-code-quality[bot]@users.noreply.github.com> --- tests/performance/active-recording.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/performance/active-recording.md b/tests/performance/active-recording.md index 6aeecacd2..544cea45d 100644 --- a/tests/performance/active-recording.md +++ b/tests/performance/active-recording.md @@ -14,7 +14,7 @@ descriptors and cumulative native-process CPU ticks once per second. Source FFmpeg processes share the host; their CPU/memory are not included in the native process samples. Streams start sequentially before the sampling interval. -After manual stop, every catalogued file must pass ffprobe and full FFmpeg +After the harness automatically stops recordings at the configured duration, every catalogued file must pass ffprobe and full FFmpeg `-xerror` decoding, with footage present for every stream. Ordinary FFmpeg exit status alone is insufficient: it can report corrupt frames while exiting zero. The regression in `mediaLifecycle.api.spec.ts` uses one stream for ten seconds;