In particular, Predictive Test Selection selects tests based on the Git changes in a build (among other data).
You can start Recording test results with the Launchable CLI with the
--no-build option. This is a good strategy for getting data flowing to Launchable quickly.
However, if you plan to use Predictive Test Selection, you should then come back and start recording builds.
To record a build, run
launchable record build before you create a build in your CI script:
launchable record build --name <BUILD NAME> --source src=<PATH TO SOURCE>
--nameoption, you assign a unique identifier to this build. You will use this value later when you record test results. See Choosing a value for <BUILD NAME> for tips on choosing this value.
--sourceoption points to the local copy of the Git repository (or repositories) used to produce this build, such as
Make sure to point to a full clone of each repository, not a partial clone.
GitHub Actions users: if you use
actions/checkoutto check out the current repo, set
You can view your recorded builds on the Builds page of the Launchable dashboard at app.launchableinc.com.
After you record a build, you can record test results against it by updating the
launchable record tests invocation(s) you created after following Recording test results with the Launchable CLI:
--build <BUILD NAME> where
<BUILD NAME> is the value you used for
launchable record build.
# beforelaunchable record tests --no-build <TOOL> <PATH TO REPORTS># afterlaunchable record tests --build <BUILD NAME> <TOOL> <PATH TO REPORTS>
If done correctly, you'll see the corresponding build name for each test session on the Test Sessions page.