Puppet users - Apr 2014 - vcsrepo module: is my patch faulty or are the test failures unjustified?

Hi,
I have some vcsrepo resources that check out out a revision specified by 
SHA1 hash.
However, on every run 'git fetch' is run to verify that the correct 
revision is checked out.
I think that shouldn't be necessary, so I tried to patch vcsrepo to avoid 
such fetches in
that case:

https://github.com/mheistermann/puppetlabs-vcsrepo/commit/68cf48ec624dddfd1cdd034638131829fd2d46fb

After this change, several spec tests fail: 
https://gist.github.com/anonymous/11399436

You can find the source of the failing tests (current HEAD) here:
https://github.com/puppetlabs/puppetlabs-vcsrepo/blob/46588aa62d444c5f0a538cccec357a80cd191b93/spec/unit/puppet/provider/vcsrepo/git_spec.rb#L173

The test  "checking the revision property when its SHA1 is not different 
than the current
SHA should return the ref", which fits my setup, seems to fails because the
changed
code does not do some git invocations required by the test, which i believe 
to be superfluous.

If I understand it correctly (I'm new to ruby and rspec), the other failing 
tests hand the string
'currentsha' to the tested code both as resource revision as well as
'git
rev-parse HEAD'
output (L176, L178). I think that 'currentsha' would be the correct
method
return value, with no
git invocations except for 'git rev-parse HEAD'.

I am not sure if my code actually introduces wrong behavior (how?) or if 
the tests
failures are actually not justified and would be glad about your comments.


Thanks in advance!

Martin



-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to puppet-users+unsubscribe@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/puppet-users/72d53153-fba6-4ff9-9588-d3c134e97c68%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.