Create new empty gh-pages branch if missing

[manics]

This makes it easier to publish a new chart. If you think this feature is worthwhile I'll add some tests.

[consideRatio]

This feature makes sense to me. The implementation seems robust enough and safe enough (safe as in no/low risk of overriding some content).

[consideRatio]

Did you mean to include the push parameter to the publish_pages function? It doesn't seem to have a purpose for this change.

I think the test seemed a bit too detailed, where the key thing to test would be that we have the new git switch --orphan gh-pages call rather than re-verify all other parts. For example, if someone like myself would change a call with ["helm", "repo", "index", "--url", "https://example.org/chartpress"] to ["helm", "repo", "index", "--url=https://example.org/chartpress"] suddenly this test named test_publish_pages_firsttime would error.

[manics]

As far as I can tell all the publish_pages tests are mocked, which means we aren't fully testing it. For example the existing test_publish_pages test wouldn't succeed, since it uses test-chart as the chartname when the chart directory under tests/test_helm_chart is testchart. If you remove the mock:

$> helm package test-chart --dependency-update --destination /tmp/tmpdevp15ck/
Error: stat test-chart: no such file or directory
FAILED

I added the push arg because all the preceding commands are local, which means they can be run for real. If you think it's better I could remove the use of CheckCallWrapper (i.e. don't verify any commands at all), and instead focus on verifying the output files?

[consideRatio]

👍 on adding the push parameter to publish_pages as it helps you run tests!

If you think it's better I could remove the use of CheckCallWrapper (i.e. don't verify any commands at all), and instead focus on verifying the output files?

Yes I think that would be better as this otherwise locks in how chartpress should behave in sharp detail beyond the scope of testing that we manage to handle the situation when a gh-pages branch isn't available etc as indicated by test_publish_pages_firsttime.

If you disagree I suggest we go for a merge straight away though. If we run into a test failure on this if something is changed down the line that breaks this test, we can address it then as well.