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
When creating an image, if HEAD is used, then the repo2docker.ref label will contain that value which is not really useful. One must start a container and use git to know which commit was used to build the container.
The same goes if --ref is not used, then the ref will be None
Expected behaviour
When HEAD is used, the commit hash should be put it in repo2docker.ref.
Same goes if --ref is not used.
Actual behaviour
repo2docker.ref contains the literal HEAD when --ref HEAD is used. repo2docker.ref contains the literal None when --ref is not used.
How to reproduce
clone a repo2docker repository
run repo2docker --no-run --no-push --image-name label-test .
docker image inspect label-test -f '{{ index .Config.Labels "repo2docker.ref" }}'
After having done some testing and gone through the ContentProviders subclasses, I am wondering whether repo2dock.ref should simply contain the value returned by content_id ?
Bug description
When creating an image, if HEAD is used, then the
repo2docker.ref
label will contain that value which is not really useful. One must start a container and use git to know which commit was used to build the container.The same goes if
--ref
is not used, then the ref will beNone
Expected behaviour
When HEAD is used, the commit hash should be put it in
repo2docker.ref
.Same goes if
--ref
is not used.Actual behaviour
repo2docker.ref
contains the literalHEAD
when--ref HEAD
is used.repo2docker.ref
contains the literalNone
when--ref
is not used.How to reproduce
repo2docker --no-run --no-push --image-name label-test .
None
Your personal set up
The text was updated successfully, but these errors were encountered: