Good first issue. Two small things on lines users read, both about a line
of text rather than about behaviour.
1. Eighteen stray spaces in the tune header
$ launchbound tune reduce-flip --backend model --cc 8.6
reduce-flip — ESTIMATED tuning (analytical model, cc 8.6); the gate is full, the timings are NOT measurements:
crates/launchbound-cli/src/main.rs:423:
"{} — ESTIMATED tuning (analytical model, cc {cc}); the gate is full, the timings are NOT measurements:",
The indentation of a wrapped literal ended up inside the string — a \ line-continuation lost in a reformat. It is the first line tune prints, and it is also a sentence carrying a caveat that matters (the timings are NOT measurements), so it should not look like a rendering fault.
While it is being touched, the line is 130 characters and wraps awkwardly in an 80-column terminal. Two lines would read better:
reduce-flip — ESTIMATED tuning (analytical model, cc 8.6)
the gate is full; the timings are NOT measurements
2. configuration(s)
crates/launchbound-tui/src/app.rs:137:
"{} REFUSED configuration(s) measured FASTER than the chosen one — view 3",
which renders, in the golden frame, as:
1 REFUSED configuration(s) measured FASTER than the chosen one — view 3
1 configuration(s) is the case that reads worst, and it is the common one — that banner appears when the count is small.
Taking both singular and plural rather than appending an s keeps the verb right too, which matters in this sentence: 1 configuration measured faster, 2 configurations measured faster happens to work, but the next such line may not.
mossaic did exactly this sweep and the helper is four lines:
pub fn plural<'a, N: PartialEq + From<u8>>(n: N, one: &'a str, many: &'a str) -> &'a str {
if n == N::from(1) { one } else { many }
}
Worth grepping for others while you are there — these were the two I found by running the tool, not by searching.
Done when
tune's header has no run of spaces in it, and the TUI banner agrees with its own count.
Getting started
CONTRIBUTING.md has the setup and the DCO sign-off (git commit -s). AI help is welcome; AI attribution is not. The TUI's golden frames live in crates/launchbound-tui/tests/golden/ and are re-blessed with LAUNCHBOUND_BLESS=1 — check tests/tui.rs for the exact variable before regenerating, and review the diff like code.
Good first issue. Two small things on lines users read, both about a line
of text rather than about behaviour.
1. Eighteen stray spaces in the tune header
crates/launchbound-cli/src/main.rs:423:The indentation of a wrapped literal ended up inside the string — a
\line-continuation lost in a reformat. It is the first linetuneprints, and it is also a sentence carrying a caveat that matters (the timings are NOT measurements), so it should not look like a rendering fault.While it is being touched, the line is 130 characters and wraps awkwardly in an 80-column terminal. Two lines would read better:
2.
configuration(s)crates/launchbound-tui/src/app.rs:137:which renders, in the golden frame, as:
1 configuration(s)is the case that reads worst, and it is the common one — that banner appears when the count is small.Taking both singular and plural rather than appending an
skeeps the verb right too, which matters in this sentence: 1 configuration measured faster, 2 configurations measured faster happens to work, but the next such line may not.mossaic did exactly this sweep and the helper is four lines:
Worth grepping for others while you are there — these were the two I found by running the tool, not by searching.
Done when
tune's header has no run of spaces in it, and the TUI banner agrees with its own count.Getting started
CONTRIBUTING.mdhas the setup and the DCO sign-off (git commit -s). AI help is welcome; AI attribution is not. The TUI's golden frames live incrates/launchbound-tui/tests/golden/and are re-blessed withLAUNCHBOUND_BLESS=1— checktests/tui.rsfor the exact variable before regenerating, and review the diff like code.