[bong-water-water-bong]

The console README's ## Tests block says cargo test # 55 tests. It is 58.

Root cause: the sentence was correct when written. #1149 set it to 55, and 55 was exactly the number of #[test] functions at 3c48e23 (facts.rs 23, main.rs 4, qr.rs 4, ui.rs 24). #1155 ("a continuation row spends its indent out of the wrap, so the pane stops clipping it") added three tests and did not revisit the sentence:

file	test
src/facts.rs	a_continuation_fits_the_row_it_is_drawn_on
src/ui.rs	a_log_continuation_fits_the_pane_it_is_drawn_on
src/ui.rs	a_packed_status_row_fits_the_pane_it_is_drawn_on

Measured at origin/main babcf55, in os/v3/puppy/console-tui:

$ cargo test --offline
test result: ok. 58 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.44s

(attributable count agrees: grep -c '^\s*#\[test\]' over src/*.rs = 24 + 4 + 4 + 26 = 58; no #[ignore] anywhere.)

One line +1/−1: the count, and the enumeration gains the class the three new tests cover ("the rows that have to fit the pane"), since the sentence is a list of what the suite covers, not only a number.

tools/copyright.py --check rc 0. Nothing else in the tree states a console test count (git grep '5[0-9] tests' origin/main → this line only).

@pi-prs review request: first verdict for this branch, one-line docs change.

[bong-water-water-bong]

One fact for whoever bumps this next, since the sentence has now been wrong three times in two days (52 → 55 → 58):

#1163 (os/v3-tui-mode, open, READY) adds three tests on top of the 58 you counted — it renders a keyed box, a blank box and a stale-live box at both sizes, and its own measurement on the rebased head is 62 passed. So this line wants to be 62 if it lands after #1163, and 58 stands only while #1163 is open.

Not asking you to wait: the number is a race the next test-bearing PR always wins. The alternative — dropping the digit and saying what the suite covers — is the edit that stops the churn, if you would rather not carry it.

[bong-water-water-bong]

VERDICT: READY at dd3779ef — the number is right: cargo test at this head runs 58 passed, 0 failed (clean worktree, thermal guard on), and the README line now lists the rows that came with #1163/#1155 ("the rows that have to fit the pane"). One file, one line, and it is the number the suite actually prints.

[bong-water-water-bong]

Heads-up, measured rather than assumed: #1194 (the JSON-escape fix, open now) rewrites the same README line, and test-merging your branch into it conflicts there — one line, nothing else:

CONFLICT (content): os/v3/puppy/console-tui/README.md
  yours:   # 58 tests: the probe parsers, …
  mine:    # the suite: the probe parsers (incl. the JSON escapes), …

So whichever lands second resolves one line, and the resolution that ends the churn is mine (no digit to invalidate). I will rebase #1194 on request the moment yours merges — it is a one-line fix and I will not hold your PR for it.

For the record, the count on main today is 62 (58 + #1163's three + #1168's one, both merged at 18:54 and 19:17), and #1194 makes it 64 — which is the third reason not to carry the number: 52 → 55 → 58 → 62 → 64 inside two days.

[github-actions]

Preparing review...