diff options
| author | Craig Jennings <c@cjennings.net> | 2026-05-14 07:51:12 -0500 |
|---|---|---|
| committer | Craig Jennings <c@cjennings.net> | 2026-05-14 07:51:12 -0500 |
| commit | 9dc45a753b74b818054f861e2999adb012fa46b6 (patch) | |
| tree | 09c35e25de0f4c391ee1d634d55731a4202a7f01 /scripts | |
| parent | e98d9b02a167fa3db19dc7f459b698df3c619b9a (diff) | |
| download | dotemacs-9dc45a753b74b818054f861e2999adb012fa46b6.tar.gz dotemacs-9dc45a753b74b818054f861e2999adb012fa46b6.zip | |
refactor(system-commands): use string interactive spec so undercover instruments
`cj/system-cmd' had `(interactive (list (read-shell-command "System
command: ")))' -- the destructured-list interactive spec. Undercover.el
relies on edebug instrumentation, which doesn't see past that form, so
the entire function body registered as 0 hits under coverage even
though tests call the function directly.
Switch to the equivalent string spec `(interactive "sSystem command: ")'.
Same UX (prompt, history, single-string result), but the body now
instruments correctly and coverage moves from 34/49 to 50/51.
Add one more test that captures the `run-at-time' lambda in
`cj/system-cmd-restart-emacs' and invokes it directly so the inner
`call-process-shell-command' branch registers, taking coverage to 51/51.
Diffstat (limited to 'scripts')
0 files changed, 0 insertions, 0 deletions
