Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Audit existing fixtures against quality criteria (litmus-sk2k)

StatusDone · archived
TypeTask
Prioritynormal
ParentFixtures iteration — more realistic, better color showcase (litmus-49jz)

Review the 7 existing fixtures against the quality criteria:

  1. Color variety (≥4 distinct ANSI colors)
  2. Instant recognition
  3. Fits 80x24
  4. Deterministic
  5. Self-contained

Fix any that don’t meet criteria. Known issues:

  • htop fallback may not be deterministic (real top output varies)
  • Some fixtures rely on tool color defaults which may vary by system
  • shell-prompt and python-repl are simulated (printf/echo) — check if they still make sense as fixtures or should be replaced with real tool output

This can run in parallel with new fixture work.

Audit Results

FixtureColorRecognition80x24DeterministicSelf-ContainedAction
git-diffPASSPASSPASSPASSPASSNone
git-logPASSPASSPASSPASSPASSNone
ls-colorPASSPASSPASSFIXEDPASSSet explicit LS_COLORS
cargo-buildPASSPASSPASSPASSPASSNone
shell-promptPASSPASSPASSPASSPASSSimulated (OK)
python-replPASSPASSPASSPASSPASSSimulated (OK)
htopPASSPASSPASSFIXEDPASSAlways use scripted output

Summary of Changes

  • htop: Removed non-deterministic top -b -n 1 path that produced varying process data. Now always uses scripted htop-like display with fixed processes and ANSI colors.
  • ls-color: Added explicit LS_COLORS export to ensure consistent file type coloring across systems.
  • shell-prompt, python-repl: Confirmed simulated approach is appropriate for reproducibility.
  • 5 fixtures (git-diff, git-log, cargo-build, shell-prompt, python-repl) passed all criteria with no changes needed.