Implement mapped collection create #898
Closed
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
This is an almost direct port of the gcs-cli command with no significant changes. No tests are included yet.
`storage_gateway_id` and `base_path` are now specified via options
rather than positionals. These have two advantages:
- order independent (too many positional arguments becomes confusing)
- we can specify defaults
Because we make a strong habit of avoiding required options, it is
desirable to have some default behaviors.
`--base-path=/` is simple, but `--storage-gateway` (note that it does
not end in `-id`) is more complex.
After discussion, we're going to have it default to "the only gateway
on the endpoint", and leave any more sophisticated behaviors as future
work. Future possibilities include
# selecting by type
--storage-gateway=posix
# selecting by name
--storage-gateway=mygateway
# selecting by any of the above with prefix mapping
--storage-gateway="type:posix"
--storage-gateway="name:mygateway"
Adjust option declarations, ensure they pass mypy and the click-type-test integration, and update the phrasing in many small ways to match the current expectations and usages of the CLI.
This primarily consists of a new set of tests which exercise the command exactly as written. One thing this revealed right away was that the CLI output from mapped collection creation was missing most of the useful fields to inspect. To remedy this, the command was updated with the "standard_collection_fields".
sirosen
force-pushed
the
implement-collection-create
branch
from
1d91c7c to
a891dd7
Compare
|
Superseded by #919 |
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Outdated suggestions cannot be applied.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 rebase of the prior work which I had on this, with some minimal cleanup to get things into a reasonable state. I'm opening it as a draft to improve visibility. It conflicts with the guest collection creation command and will need a rebase after that merges (so please ignore the conflicts).
I see a few things which need additional attention in here, and I especially note that there were no tests when I dusted this off. I think I had done some measure of manual testing which at this point is not very relevant.
Some areas of interest:
collection update, and it will share some with guest creation as well -- we should move those to common helpers where possible.policiessub-object is tricky. The current implementation has a small collection of mutually exclusive options which can be used. (These could be declared as mutex, but I believe mutex checking ofmultiple=Trueoptions requires some care.) I would like it better if we could declare these usages separately somehow, but I think the current solution is likely to be a good pragmatic solution for the near-to-medium term.--storage-gatewayrather than--storage-gateway-id. The expectation baked into this is that a typical endpoint will have only one or two gateways attached, and that typically when there are two (or more), there will only be one of each type. It is therefore a planned enhancement to support--storage-gateway posixas a method for selecting the only POSIX gateway on the Endpoint. I think this branch of work should be updated to either implement that logic OR switch to--storage-gateway-id. I'm not yet sure which.