Commit 50cd900
committed
ci: filter usermod matrix to only build changed usermod directories
Instead of building every usermod with a library.json on each PR, use
git diff to identify which usermods/ subdirectories were actually touched
and intersect that with the known-good library.json list.
This reduces CI time significantly for PRs that only modify one or two
usermods (previously every PR triggered ~40 usermods × 4 chipsets).
Also removes the unnecessary PlatformIO install from get_usermod_envs
(the step only uses shell/jq, not pio) and adds fetch-depth: 0 to
ensure the base branch is available for the diff.1 parent e68bd2a commit 50cd900
1 file changed
Lines changed: 20 additions & 8 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
20 | 20 | | |
21 | 21 | | |
22 | 22 | | |
23 | | - | |
24 | 23 | | |
25 | | - | |
26 | | - | |
27 | | - | |
28 | | - | |
29 | | - | |
| 24 | + | |
| 25 | + | |
30 | 26 | | |
31 | 27 | | |
32 | | - | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
| 33 | + | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
| 39 | + | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
| 43 | + | |
33 | 44 | | |
34 | 45 | | |
35 | 46 | | |
36 | 47 | | |
37 | 48 | | |
38 | 49 | | |
39 | | - | |
| 50 | + | |
| 51 | + | |
40 | 52 | | |
41 | 53 | | |
42 | 54 | | |
| |||
0 commit comments