Add malformed tests check

[jtratner]

Currently, if you have a single quote in your test command, it'll break the bioconda build because of shell escaping. Not sure if the upstream will fix this (or if a fix is really worth it, given that you can always put your test in a test file), but the error is confusing to the user. This PR adds a single_quote_in_tests check to the linter.

Aside - I haven't been able to get the test suite to run yet (not sure how to let anaconda for python3 and python2 exist). But I'm hoping Travis will run 'em for me :)

[bgruening]

Cool thanks @jtratner. xref: #163
I think the fix might be easier if we can get this to quote the parameters correctly: https://github.com/bioconda/bioconda-utils/blob/master/bioconda_utils/pkg_test.py#L134

[jtratner]

@bgruening

I think the fix might be easier if we can get this to quote the parameters correctly: https://github.com/bioconda/bioconda-utils/blob/master/bioconda_utils/pkg_test.py#L134

yes, but also shell escaping gets so complicated that it makes my head hurt :-/ (so I picked the dumb way ;P)

[bgruening]

That is perfectly fine for the time being! Thanks!

[jtratner]

Test failures are unrelated... especially the ENCRYPTED_KEY_VAR for docs - wdyt @bgruening ?

[eseiler]

Could the ENCRYPTED_KEY_VAR error be related to #167 ?

The current failure is due to running on a fork, which doesn't have the encrypted var available when cloning the repo over ssh. Should be a quick fix to use https when not on the main repo, I will fix this tomorrow.

Cloning via SSH was fixed after this PR, so you may need just to clone again.

[epruesse]

@bgruening @jtratner What is the status of this? Do we still need this?

[bgruening]

Looks still good to me. Could we have a rebase?

[epruesse]

@bgruening - merged and conflicts resolved. It looks like it might be overzealous, but it is KISS.