Add a parse_gitignore_string so you don't need a real .gitignore file.

[tewalds]

No description provided.

[mherrmann]

Thank you for the PR. Why is it necessary? Can you add a test?

[tewalds]

Thank you, I added a test.

I'm using this in https://github.com/tewalds/replace , which is a simple find/replace tool, ie a worse ripgrep but that actually replaces as well. It's useful to ignore everything that isn't tracked as they're almost certainly generated files or otherwise not interesting, but it kept returning matches/failures from the .git directory, which is also not tracked and not interesting to search. I could add .git to all my .gitignore, but that's per repo and not really correct. An alternative approach would be for gitignore_parser to accept the file string instead of a filename (as the test does), and I could prepend the ".git/" line myself, but this was easier, and I figured you'd appreciate it too.

[mherrmann]

I see, thank you for explaining. I understand the need and appreciate the PR, but I feel that a "gitignore parser" should parse .gitignore files, and nothing else. I believe you have a point from a practical perspective, but I'm a purist at heart and can't bring myself to violate that.

[tewalds]

Well, then it'd be nice to have a way of parsing a string instead of a file without mocking open. Would you be open to making that change, or to a PR making that change? That is much more than a couple lines, so I figure you might have some opinions on how to approach it.

[tewalds]

I guess another option would be to make it optional, eg a ignore_git_dir, defaulted to `False.

[mherrmann]

Yes, I'm open to a PR that adds a function parse_gitignore_str, so you don't have to mock open.

[tewalds]

Ok, this is a much bigger change, both to the library and the tests, but hopefully you agree this is an improvement. This also fixes #6 along the way.

[tewalds]

Done.

[tewalds]

The CI seems to be failing due to python versions, so fixed in: #75

[mherrmann]

If you wanted to add a few lines to the README that explain the new function, that would be great; if not, I can do it as well.

[tewalds]

I added an example to the README as well.