[
https://issues.apache.org/jira/browse/LIBCLOUD-986?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16407536#comment-16407536
]
Quentin Pradet commented on LIBCLOUD-986:
-----------------------------------------
The first thing to do would be to run the tests locally:
* git clone https://github.com/apache/libcloud.git
* cd libcloud
* pip install -r requirements-test.txt
* cp libcloud/test/secrets.py-dist libcloud/test/secrets.py
* pytest
When you'll be working on the Azure Blobs tests, running `pytest -k
azure_blobs` will be faster.
The next step is to understand how the test fixtures work. Here's an example
for another driver:
https://github.com/apache/libcloud/pull/1157#issuecomment-355883772
Next we'll need to write a test that fails without your changes, and succeeds
with your changes. Try to come up with a way to do this.
If you're stuck, make sure to ask questions!
> List / Iterate container with prefix for Azure Blob service (This support
> exists in the other drivers already)
> --------------------------------------------------------------------------------------------------------------
>
> Key: LIBCLOUD-986
> URL: https://issues.apache.org/jira/browse/LIBCLOUD-986
> Project: Libcloud
> Issue Type: Improvement
> Components: Storage
> Reporter: Joshua Hawkinson
> Priority: Major
>
> While testing out the libcloud storage abstraction I've come across a curious
> omission in driver support. It seems that for many drivers there is
> additional support for the {{list_container_objects()}} and
> {{iterate_container_objects()}} to support a prefix (or {{ex_prefix)}}.
> This functionality is not present in the Azure driver for some reason. I've
> gone and checked the list_blobs API and it also supports this functionality
> as seen here:
> [https://docs.microsoft.com/en-us/rest/api/storageservices/list-blobs]
>
> I've attempted a quick and dirty patch that seems to work as well.
> {code}
> azure_blobs.py Line 384
> - def iterate_container_objects(self, container):
> + def iterate_container_objects(self, container, ex_prefix=None):
> """
> @inherits: :class:`StorageDriver.iterate_container_objects`
> """
> params = {'restype': 'container',
> 'comp': 'list',
> + 'prefix': ex_prefix,
> 'maxresults': RESPONSES_PER_REQUEST,
> 'include': 'metadata'}
> Line 419 (Add new function)
> def list_container_objects(self, container, ex_prefix=None):
> """
> Return a list of objects for the given container.
>
> :param container: Container instance.
> :type container: :class:`Container`
>
> :param ex_prefix: Only return objects starting with ex_prefix
> :type ex_prefix: ``str``
>
> :return: A list of Object instances.
> :rtype: ``list`` of :class:`Object`
> """
> return list(self.iterate_container_objects(container,
> ex_prefix=ex_prefix))
>
> {code}
>
> I've just worked around this at a higher level for now. There isn't any
> urgency on my end, but it would be nice if the abstraction was more
> consistent so I could remove the workaround code.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
