Currently when you are utilizing the Sentry Ruby SDK if you do not specify a release name, it uses git-based automatic detection which uses a git rev-parse --short HEAD command.
This works nicely, however when creating a release using the Github Action Sentry Release the only option is to use the full SHA as it is all that Github exposes as a variable. This release is preferable to default to as it associates commits with the release and gives you a clearer picture of what is broken
For example you can wind up with two releases within your sentry project:
- one
f029dd0a822e8e99e19ee50f21a7 with github commits associated and files changed
- another
f029dd0 that has the actual errors being sent to it
Would it be possible to change the Sentry Ruby Gem to use the full SHA with git rev-parse HEAD to align with the Github Action?
FWIW this change has already been discussed and changed by the sentry-python SDK a few years back. getsentry/sentry-python#908
Currently when you are utilizing the Sentry Ruby SDK if you do not specify a release name, it uses git-based automatic detection which uses a
git rev-parse --short HEADcommand.This works nicely, however when creating a release using the Github Action Sentry Release the only option is to use the full SHA as it is all that Github exposes as a variable. This release is preferable to default to as it associates commits with the release and gives you a clearer picture of what is broken
For example you can wind up with two releases within your sentry project:
f029dd0a822e8e99e19ee50f21a7with github commits associated and files changedf029dd0that has the actual errors being sent to itWould it be possible to change the Sentry Ruby Gem to use the full SHA with
git rev-parse HEADto align with the Github Action?FWIW this change has already been discussed and changed by the sentry-python SDK a few years back. getsentry/sentry-python#908