*: update to image-spec v1.0.0-rc5 #133
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
I will follow up with a |
cyphar
force-pushed
the
imagespec-v1.0.0-rc5
branch
2 times, most recently
from
c1eafdd to
d2ee3b9
Compare
cyphar
force-pushed
the
imagespec-v1.0.0-rc5
branch
4 times, most recently
from
f4d67a0 to
61fa801
Compare
Since PutBlobJSON is not actually an underlying property of the CAS, it makes more sense in casext to remove the amount of things that CAS implementations need to deal with. In addition it allows us to switch to deterministic JSON encoding in the future without changing the CAS implementations. In addition, move the corresponding PutBlob tests into casext so that the tests make more sense. The readonly and readwrite methods of testing the API guarantees is not as nice as it should be at the moment, and should be improved by breaking up the CAS "types" in the future (sort of like io.Reader and io.Writer). Signed-off-by: Aleksa Sarai <[email protected]>
This is equivalent to "path/filepath".SkipDir, but for descriptor walking. It is necessary in order to implement parts of the reference resolution code sanely (and also efficiently). Signed-off-by: Aleksa Sarai <[email protected]>
This is just a temporary implementation which will be replaced with the switch to v1.0.0-rc5. Note that this interface is *very* likely to change in the future due to the arbitrary nesting requirements of v1.0.0-rc5. In addition, move the tests for references to oci/casext (using the new interfaces). On paper this means that both the old and new interfaces should still pass all of the oci/casext tests. Signed-off-by: Aleksa Sarai <[email protected]>
cyphar
force-pushed
the
imagespec-v1.0.0-rc5
branch
from
61fa801 to
8277b2b
Compare
This requires some very significant changes to various portions of umoci. At this point, umoci supports all of the necessary bits for "standard" images (old-style structure with new layout). There are some leftover bits in mutate/ as well as in the UX that are blocking supporting the largest possible range of control over the structure of OCI images generated and mutated by this library. These will be fixed in future revisions before the next release. Signed-off-by: Aleksa Sarai <[email protected]>
cyphar
force-pushed
the
imagespec-v1.0.0-rc5
branch
from
8277b2b to
c29001a
Compare
Signed-off-by: Aleksa Sarai <[email protected]>
|
LGTM. There's plenty of improvements left that are necessary but this works for the moment. |
|
I just realised that the tests passed even though |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is a fairly large semantic and library change because of the semantic
change of tags no longer being a first-class property of the underlying
storage. This required a change to
oci/casand adding reference resolution tooci/casext(which is actually nicer on some level but conflicts are hard toresolve now).
Currently platform filtering is not implemented, and the current handling of
ambiguous tags is not really ironed out (errors are returned). This is
something that will need to be ironed out in the future especially if we want
to support creating multi-level trees of
Indexes.In addition, due to the way that our CI works we need to update to the latest
skopeobut we can't do this in the distribution because this change is stillbeing worked on. IMO it would be better if we first resolve #132 before merging
this so that we actually can test this in a more sane way going forward.
Signed-off-by: Aleksa Sarai [email protected]