Closed christianwbsn closed 2 years ago
This looks good, but could the testing also be triggered when someone submitted a PR (not a draft, per se)?
I'm just asking, since I think trying this functionality alone is already ok for a mainstream submission. I think if someone wants to run a test on a different subset, this workflow still need to be enhanced (I just look up at your example here, where the test returns an error due to the subset requirement)
This looks good, but could the testing also be triggered when someone submitted a PR (not a draft, per se)?
I'm just asking, since I think trying this functionality alone is already ok for a mainstream submission. I think if someone wants to run a test on a different subset, this workflow still need to be enhanced (I just look up at your example here, where the test returns an error due to the subset requirement)
Ah, for that to work, I think we need to standardize how we write titles in PR / use branches (people are not really consistent with this). When someone opens a PR, we can't precisely know which data loader needs to be tested, hence we still need to type the dataset name manually. Dataset with subset gives another reason that we still need to trigger it manually rather than automatically trigger it whenever someone submits a PR. Open for new idea for this issue, I think it could be great if we can do that automatically
And for subset, I've already updated it! https://github.com/christianwbsn/nusa-crowd/actions/runs/3035492504
PoC here: https://github.com/christianwbsn/nusa-crowd/pull/4
For easier testing, let's![Screen Shot 2022-09-09 at 16 43 15](https://user-images.githubusercontent.com/26519918/189321736-9d02263d-aeba-49f3-af07-0d73829423f1.png)
abuseutilize GitHub actions computing power, and memory How to use: