Commit 98e86fd
committed
Fix CI dependency resolution tests to actually test target versions
The dependency resolution tests were using --frozen flag (for highest)
or no upgrade flag (for lowest-direct), which meant both tests just
used whatever was locked in uv.lock from the main branch rather than
actually resolving to the target versions.
This caused the CI to miss breaking changes in new dependency versions
(like pydantic 2.12) that were compatible with version constraints but
not yet in the lockfile.
Changes:
- Use --upgrade --resolution highest for the highest resolution test
- Use --upgrade --resolution lowest-direct for the lowest resolution test
(both need --upgrade to ignore lockfile and re-resolve)
- Simplify test execution to use --frozen --no-sync since
resolution already happened during uv sync
- Remove redundant UV_RESOLUTION env var (command-line flags suffice)
This ensures both tests actually resolve to their intended versions
and catch compatibility issues before they reach main.1 parent be73067 commit 98e86fd
File tree
1 file changed
+5
-7
lines changed- .github/workflows
1 file changed
+5
-7
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
38 | 38 | | |
39 | 39 | | |
40 | 40 | | |
41 | | - | |
| 41 | + | |
42 | 42 | | |
43 | | - | |
| 43 | + | |
44 | 44 | | |
45 | 45 | | |
46 | 46 | | |
| |||
57 | 57 | | |
58 | 58 | | |
59 | 59 | | |
60 | | - | |
61 | | - | |
62 | | - | |
63 | | - | |
64 | | - | |
| 60 | + | |
| 61 | + | |
| 62 | + | |
65 | 63 | | |
66 | 64 | | |
67 | 65 | | |
| |||
0 commit comments