Gate sc-tray on static analysis and a coverage ratchet #4

Merged

geusebio merged 3 commits from feature/static-analysis-and-coverage-gate into main

2026-08-15 13:04:34 +00:00

geusebio commented

2026-08-14 19:59:12 +00:00

Owner

Companion to scclient #4, which explains the shape. Same three files.

17% is the honest number here, and it is the reason for a ratchet rather than a target. Most of this package is menu callbacks the OS drives and systray mutation that needs a display — a line-coverage goal would be met by testing the wrong things. The capture suite is what actually exercises those paths, and the ratchet stops the 17% rotting downwards while that suite grows to replace it (#627).

One finding, and a fair one: the notification path hands a deployment name to osascript. It was already quoted into an AppleScript literal for exactly that reason — the pragma now says so on the line.

Clean at hack/lint.sh; go test, go vet and all four static cross-compile targets still green.

Companion to [scclient #4](https://git.grey.ooo/Grey.ooo/someones.computer_scclient/pulls/4), which explains the shape. Same three files. **17% is the honest number here, and it is the reason for a ratchet rather than a target.** Most of this package is menu callbacks the OS drives and systray mutation that needs a display — a line-coverage goal would be met by testing the wrong things. [The capture suite](hack/screenshot) is what actually exercises those paths, and the ratchet stops the 17% rotting downwards while that suite grows to replace it ([#627](https://git.grey.ooo/Grey.ooo/Someones.Computer/issues/627)). One finding, and a fair one: the notification path hands a deployment name to `osascript`. It was already quoted into an AppleScript literal for exactly that reason — the pragma now says so on the line. Clean at `hack/lint.sh`; `go test`, `go vet` and all four static cross-compile targets still green.

geusebio added 1 commit

2026-08-14 19:59:13 +00:00

Gate sc-tray on static analysis and a coverage ratchet

CI / build (pull_request) Failing after 50s

Details

2ba86c8b03

Matches scclient and sc: golangci-lint from a pinned image, and coverage as a
ratchet rather than a threshold.

The recorded floor is 17%, which is the honest number and the reason a
ratchet rather than a target. Most of this package is menu callbacks the OS
drives; a line-coverage goal would be met by testing the wrong things, and
the capture suite is what actually exercises them. The ratchet stops that 17%
rotting downwards while the capture suite grows to replace it.

One finding, and a fair one: the notification path hands a deployment name to
osascript. It was already quoted into an AppleScript literal for exactly that
reason, so the pragma says so on the line.

geusebio added 1 commit

2026-08-14 20:17:33 +00:00

Run the linter from PATH in CI, where a bind mount reads nothing

CI / build (pull_request) Successful in 5m27s

Details

e5fbd72d81

A CI job step and the docker daemon are in different filesystem namespaces,
so a `-v "$PWD":/app` in a step mounts an empty directory and the linter
passes having checked nothing. Same fix as scclient: prefer a PATH binary at
exactly the pinned version, keep the image as the laptop path.

geusebio added 1 commit

2026-08-14 20:46:37 +00:00

Note that coverage floors are Linux figures, since CI is what they gate

CI / build (pull_request) Successful in 4m5s

Details

7077d4960f

geusebio merged commit 027c153540 into main

2026-08-15 13:04:34 +00:00

geusebio referenced this pull request from a commit

2026-08-15 13:04:35 +00:00

Gate sc-tray on static analysis and a coverage ratchet (#4)