$commits
$commits : int
What one verify pass found.
Counted rather than described: a pass over a year of history touches millions of frames, so the report holds totals plus the findings, and the findings are already clamped by Finding itself.
A pass that read everything and found nothing is the only clean result. Anything else names what it could not read, which is what makes "the backups work" a measured claim rather than an assumption.
__construct(int $commits = 0, int $trees = 0, int $frames = 0, int $segments = 0, int $bytes = 0, list<\Drupal\strata\Health\Finding> $findings = [], float $seconds = 0.0, bool $complete = true) : mixed
Constructs a report.
| int | $commits | Commits walked. |
| int | $trees | Tree nodes read. |
| int | $frames | Frames fetched and decoded. |
| int | $segments | Segment manifests read. |
| int | $bytes | Decoded bytes verified. |
| list<\Drupal\strata\Health\Finding> | $findings | What went wrong, in the order it was found. |
| float | $seconds | How long the pass took. |
| bool | $complete | FALSE when the pass stopped at a limit rather than reaching the root of history, so a clean result covers only what it walked. |