Conversation
.cirrus.yml
Outdated
| # We try to clone htslib using the same branch name and owner as this | ||
| # samtools branch. If it exists, it's likely the user is making a | ||
| # joint samtools+htslib PR and wants both checked in unison. | ||
| # Failing that we use samtools/htslib:devel |
There was a problem hiding this comment.
| # Failing that we use samtools/htslib:devel | |
| # Failing that we use samtools/htslib:develop |
There was a problem hiding this comment.
Thanks.
Also I was hoping it'd work more like AppVeyor than Travis, but maybe I need a .cirrus dir with an updated script. Will explore. This PR is using samtools/htslib develop branch, not the cirrus-ci branch. For me in my repos it "does the right thing" here. Will explore more so expect a flurry of commits ;)
There was a problem hiding this comment.
Seems not possible. Oh well, we had the same problem with Travis. Env set or this PR:
CIRRUS_BASE_BRANCH=develop
CIRRUS_BRANCH=pull/1335
CIRRUS_REPO_FULL_NAME=samtools/samtools
CIRRUS_REPO_NAME=samtools
CIRRUS_REPO_OWNER=samtools
Lacking jkbonfield/samtools or my branch of cirrus-ci branch means it has no means to find the matching htslib branch. In this case it's irrelevant, but often not (eg the two fastq branches). Shame. Only choice I could think of for fixing that, if we care, would be grokking the commit messages.
No description provided.