search

diprism / fggs

Factor Graph Grammars in Python
MIT License
13 stars 3 forks source link

[WIP] Add CI configuration and test for building perpl. #174

Closed chihyang closed 11 months ago

chihyang commented 11 months ago

Support using "correct" tests from perpl to test sum product of fggs. Because of one bug in fail.ppl, once this is merged, the CI will always fail. This can be merged later when the bug is fixed.

ccshan commented 11 months ago

What do you think @davidweichiang? The plan is to use PERPL tests to exercise the FGG library.

davidweichiang commented 11 months ago

I think it sounds like a good idea. Can the CI be set up so that it is pinned to a particular version of PERPL so that a bug over there doesn't cause tests to fail here?

ccshan commented 11 months ago

I think it sounds like a good idea. Can the CI be set up so that it is pinned to a particular version of PERPL so that a bug over there doesn't cause tests to fail here?

Good idea -- maybe a tag in the perpl repository?

chihyang commented 11 months ago

I think it sounds like a good idea. Can the CI be set up so that it is pinned to a particular version of PERPL so that a bug over there doesn't cause tests to fail here?

Good idea -- maybe a tag in the perpl repository?

If there is a tag in PERPL, I can change the CI script to always use that version. For now I just use the master branch.

ccshan commented 11 months ago

If there is a tag in PERPL, I can change the CI script to always use that version. For now I just use the master branch.

Ok, I created a tag testing (while thinking of Debian) in the perpl repository.

Thanks for knowing how to use CI...

ccshan commented 11 months ago

If you merge 542ae59ab1f78b28bf1eee640a111a02abb107af then the new tests should pass?