You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
/usr/bin/git -c protocol.version=2 fetch --no-tags --prune --progress --no-recurse-submodules --depth=1 origin +refs/heads/docs-remove-outdated-invitation*:refs/remotes/origin/docs-remove-outdated-invitation* +refs/tags/docs-remove-outdated-invitation*:refs/tags/docs-remove-outdated-invitation*
The process '/usr/bin/git' failed with exit code 1
I submitted this pull request from a fork of the W3C-owned repository. It's tough to say from the output alone (which, for those unfamiliar, is available by clicking the button labeled "0 / 2 checks OK"), but the continuous integration process may be incorrectly attempting to fetch the fork's branch as if it were present in the W3C-owned repository. The CI process may not be configured to support contributions issued from forks. In the past month, just two other pull requests have been submitted like that (gh-434 and gh-441), and they both exhibit the same error.
I discussed this briefly with @s3ththompson earlier today, forks make testing and contributions unnecessarily complicated so my vote is to have all PRs use a branch in this repo. Any technical or logistical objections? My reasons include:
With an existing clone of the repo, switching to a fork is more involved than just checking out a new branch.
If people have write or other non-default permissions under this repo, they don't apply to forks.
Only people who have been explicitly granted "write" access to the repository will be able to create branches there. That requirement will increase friction for new contributors.
@jugglinmike That's fair. So the policy should be, if someone already has write access, they should create a branch under this repo. Otherwise, a fork is their only option and shouldn't be forbidden. Members of the Bocoup and PAC teams should always be creating branches here.
More information about the failure:
I submitted this pull request from a fork of the W3C-owned repository. It's tough to say from the output alone (which, for those unfamiliar, is available by clicking the button labeled "0 / 2 checks OK"), but the continuous integration process may be incorrectly attempting to fetch the fork's branch as if it were present in the W3C-owned repository. The CI process may not be configured to support contributions issued from forks. In the past month, just two other pull requests have been submitted like that (gh-434 and gh-441), and they both exhibit the same error.
Originally posted by @jugglinmike in #433 (comment)
Is there something we can fix in how the GitHub Actions are set up to succeed for PRs from forks? (And not introduce new security holes)
The text was updated successfully, but these errors were encountered: