cli: setup gitignore when working from git directory at init.

[msis]

• Have you followed the guidelines for contributing?
• Have you signed the CLA?
• If this is a new feature. Have you discussed the design on the forum?
• Have you successfully run ./runtests.sh static?
• Have you successfully run ./runtests.sh unit?

---

This adds predefined .gitignore if .git directory exists.

[kyrofa]

Thanks @msis! I've tested this successfully and I like it, although I suggest a few code tweaks before I can wholeheartedly give it a +1.

[kyrofa]

The use of os.path.join is unnecessary, here. In fact, I think this entire variable is unnecessary, no?

[kyrofa]

Rather than passing, perhaps this would be easier to read:

# Don't mess with an existing .gitignore
if not os.path.exists('.gitignore'):
    with open('.gitignore', mode='w') as f:
        f.write(_TEMPLATE_GITIGNORE)

[kyrofa]

Because I apparently don't know how to use the review tools, I gave it a wholehearted +1 anyway. That's okay 😄 .

[msis]

Thanks @kyrofa !
I updated the code to follow your guidelines.
PS: Do I need to register somewhere for that integration test to succeed?

[kyrofa]

@msis no, you're good. Something else is broken in that check, so I invoke @elopio.

[kalikiana]

Neat :-D Love at first sight!

[sergiusens]

Looks good, mind if I move stuff around in the PR a bit? Just use of dedent and driving the ignore creation from the cli

[elopio]

There's a typo here: Verify the .gitignore was NOT created.
I would just remove this comment, the assertion is clear enough, I think.

[elopio]

This is great, thanks @msis!
There is one typo in one of the comments, so I'll use this opportunity to be nitpicky and the three # Verify ... comments. It seems to me that the name of the test + the nice assertions you have there are very clear and readable.

Also, as Sergio mentioned, please use dedent. It's a very recent rule in our code style book: https://github.com/snapcore/snapcraft/blob/master/CODE_STYLE.md#multiline-strings

[elopio]

Sorry @msis, I delayed my review and now this pull request has a conflict. Can you please fix it?

[sergiusens]

thanks for the contribution. I think we will need some more smarts in this to make it bullet proof and remove any potential error prone behavior.