Commit 72694e6
1 file changed
+0
-1
lines changedLines changed: 0 additions & 1 deletion
Original file line number | Diff line number | Diff line change | |
---|---|---|---|
| |||
228 | 228 |
| |
229 | 229 |
| |
230 | 230 |
| |
231 |
| - | |
232 | 231 |
| |
233 | 232 |
| |
234 | 233 |
| |
|
Original file line number | Diff line number | Diff line change | |
---|---|---|---|
| |||
228 | 228 |
| |
229 | 229 |
| |
230 | 230 |
| |
231 |
| - | |
232 | 231 |
| |
233 | 232 |
| |
234 | 233 |
| |
|
3 commit comments
scordio commentedon Jun 9, 2025
Hi @fmbenhassine, is this a temporary change, or would you like to prevent ErrorProne checks from triggering build failures in the long term as well?
Once #4864 is introduced, NullAway should ideally cause build failures to spot incorrectly annotated APIs.
Taking inspiration from Spring Data, ErrorProne could be set as
-XepDisableAllChecks -Xep:NullAway:ERROR
, which should ignore all ErrorProne checks except the ones triggered by NullAway.Should I proceed this way in my PR?
fmbenhassine commentedon Jun 9, 2025
I prefer failing the build only for errors (mainly compilation errors and test failures), but not warnings. Warnings should be reported, but should not result in a build failure (they are warnings after all).
For the jSpecify PR, you can proceed as you started. Ideally, we should be able to specify which options to include/exclude at the CLI when running the plugin. We can discuss this further when merging the PR.
scordio commentedon Jun 9, 2025
Then I propose removing all the exclusions introduced in #4807 as well, to keep the POM cleaner. I'll do that in my PR.