Tests: download a target with succinct_roles enabled

[MVrachev]

Related to issue #1909

Description of the changes being introduced by the pull request:

Add test downloading a target file when succonct_roles is used and as
such test the whole updater downloading workflow.

Signed-off-by: Martin Vrachev mvrachev@vmware.com

Please verify and check that the pull request fulfills the following
requirements:

• The code follows the Code Style Guidelines
• Tests have been added for the bug fix or new feature
• Docs have been added for the bug fix or new feature

[coveralls]

Pull Request Test Coverage Report for Build 2516412956

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 97.215%

---

Totals
Change from base Build 2515189215:	0.0%
Covered Lines:	553
Relevant Lines:	565

---

💛 - Coveralls

[jku]

So, I'm happy to merge this, it's better than what we have. However it feels like it's just the delegation graph test test_succinct_roles_graph_traversal with download_target() call added to the end...

Some potential ways to improve:

• avoid duplicating all delegation graph checks: those checks are complex and I feel like they harm the readability of this test. Maybe it makes sense to verify the delegated role (exp_calls) in the end but even that is debatable
• actually check that the target content is the expected one, stop using the same content for all targets
• it's very minimal WRT repository content: e.g. there is never more than one target file in a delegated targets metadata, or even the whole repository. I think we could get more real coverage if we setup a repository with succinct delegation to lots of generated targets (like maybe targets could be every two character combination as file content -- 26*26 files, and filename could be the same as file content), and then the subtests could be a dozen target downloads from that repo?

I'm marking approve, but let me know your opinion

[MVrachev]

I rewrote the whole test and used a different approach.
When I decided to follow your advice and generate more target files I realized that I don't need to use the decorator at all anymore.
That's why I instead used for loops which in this case seems like a reasonable choice.

If you don't like that approach I can rever the new commit.

[jku]

I like this, very nice! 1000 files sounds like a lot but if it's not slow I'm not complaining.

Left two comments but this is great as is 👍

[jku]

if it's not slow I'm not complaining.

looking at just results for this CI run, this seems to increase complete test suite runtime by ~50%

Please test this before writing 1000 files.

[jku]

Actually (since martin may be on a holiday) I can push a commit here myself that fixes the minor issues: will merge after that

[MVrachev]

Thank you for pushing the changes @jku!
I was indeed on a holiday.