Dry Run

Planned behaviour

When --dry-run ships, passing it to nolapse run will:

Run coverage collection as normal.
Compute the delta against the baseline.
Print the outcome (pass / warn / fail) and the delta to stdout.
Not write any files (no .nolapse_run_state update).
Always exit 0, regardless of whether the result would normally be a warn or fail.

This makes it safe to call in exploratory or debug contexts without affecting CI exit codes or state files.

# Planned — not yet supported
nolapse run --dry-run

Expected output (illustrative):

[dry-run] current coverage: 81.20%
[dry-run] baseline:         82.00%
[dry-run] delta:            -0.80pp
[dry-run] outcome:          WARN (would exit 0 normally, exit 1 in strict mode)
[dry-run] no files written.

Current workaround

Until --dry-run is available, you can simulate the behaviour by temporarily setting a very high --fail-threshold so that nolapse never exits non-zero:

nolapse run --warn-threshold 999 --fail-threshold 999

This will:

Run coverage collection normally.
Always produce a pass result regardless of the actual delta.
Still write .nolapse_run_state, so do not follow this with nolapse baseline update unless you intend to commit the result.

To observe what the outcome would have been under your normal thresholds, read the .nolapse_run_state file after the run:

cat .nolapse_run_state
# {"coverage": 81.20, "head_sha": "abc123..."}

Then compare manually to the baseline in .audit/coverage/baseline.md:

cat .audit/coverage/baseline.md
# coverage: 82.00%
# timestamp: 2026-01-01T10:00:00Z
# commit: def456...

The difference (81.20 - 82.00 = -0.80 pp) tells you what result your configured thresholds would produce.

Tracking the feature

The --dry-run flag is tracked as a P1 item on the nolapse roadmap. When it ships, this page will be updated with the full flag reference.