XSF Editor Team - 2020-06-18

I wonder if there’s any reason why we shouldn’t tie our pipeline to gitlab users, either.

I think that’s better than email addresses

we don’t have to tie ourselves to signed-off-by if we have a separate ACK process anyways

allows people to change email addresses without hassle, and it’s less blatant to have a list of gitlab users compared to a list of emailaddress/name pairs

so you can still contribute pseudonymously to some extent

so you’d do an MR against the ipr-list repo, add yourself as name to the list, the CI on the ipr-list would check that name you added == name of MR author

when an MR against xeps is made, the CI checks that the author name is part of the list

uuuuuuugh

we can’t do that from a pipeline

we have to go by webhooks, no way around that -.-

reason is that the pipeline on a MR from a forked repository is run within the forked context, not in the parent context, so it wouldn’t have our credentials

"reason is that the pipeline on a MR from a forked repository" yep. Also a reason why I often push my branches on the main repo and not the fork, just so I get the exact same CI setup

nothing to see here, move along: https://gitlab.com/xsf/test/-/merge_requests/1#note_364050284

(note: the thing was running in a temporary session on my server and is stopped now)