The LDC snap package is currently without any automated testing beyond the fact that the package itself builds. It would be nice to improve on this to ensure that regressions do not sneak into updated package releases.
Broadly speaking, there are 2 things that we could consider doing:
tests during the build process itself (e.g. running any test suite available in the regular build scripts): this validates LDC itself on the target architecture, but doesn't really add much to the existing validation of LDC done during its own release process
The LDC snap package is currently without any automated testing beyond the fact that the package itself builds. It would be nice to improve on this to ensure that regressions do not sneak into updated package releases.
Broadly speaking, there are 2 things that we could consider doing:
tests during the build process itself (e.g. running any test suite available in the regular build scripts): this validates LDC itself on the target architecture, but doesn't really add much to the existing validation of LDC done during its own release process
automated tests once an updated package is released into the
edgechannel. Some ideas on how to do this are discussed in this snapcraft forum thread: https://forum.snapcraft.io/t/automated-testing-of-newly-built-snaps/881