You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have lot of 'grandfathered' issues which we are fixing, below are the workflow we typically use
1. ./gradlew clean
2. infer --reactive -- ./gradlew --offline assembleDebug
3. Note down the issues and fix
4. infer --reactive --continue -- ./gradlew --offline assembleDebug
Above works fine, however step 4 seems to analyse the complete project even though only 1 file changed and it takes more time to analyse and report.
is there any flags or options which we can use to build only a file or two when using gradle, I'm aware its possible for normal java builds but how about gradle builds?
The text was updated successfully, but these errors were encountered:
Note that using --continue option, it will keep on growing the set of files to be analyzed. So in your particular case, since it follows a complete compilation, it will repeatedly analyze everything.
You might want to try without --continue, i.e. step 4 is the same as step 2, so only the files compiled in the last gradlew invocation will be analyzed.
Infer's gradle integration is very primitive and loses most of the incrementality of builds. @cristianoc's suggestion may help, but if you want fast and truly incremental analyses you may want to look into using okbuck and use infer's much better performing Buck integration. Another solution would be for someone to write a gradle plugin for infer.
We have lot of 'grandfathered' issues which we are fixing, below are the workflow we typically use
Above works fine, however step 4 seems to analyse the complete project even though only 1 file changed and it takes more time to analyse and report.
is there any flags or options which we can use to build only a file or two when using gradle, I'm aware its possible for normal java builds but how about gradle builds?
The text was updated successfully, but these errors were encountered: