Chris Jerdonek

@Tinche Do you have a preference for what aiofiles module `os.path` functions should go in?

Maybe this is in part caused by the fact that colorama's `reset_all()` which it calls on exit only resets `orig_stdout` and not `orig_stderr`, which is where the color is happening:...

Thanks, @wiggin15. > The code you mentioned is only called at program exit. Yes, this is why I hedged and said "_in part_ caused by the fact..." My thinking was...

Note: when I said "permanently," I had in mind this issue-- https://github.com/pypa/pip/issues/6354 rather than the issue as reported in the original comment of the current issue.

> I didn't expect this and we may need to look into this a bit. Maybe the streams shouldn't be separated like this. My instinct is that people should be...

Just a quick note in response to the original post: > This is a little messy. Is there a better way to do this? I'm actually somewhat okay with how...

I meant to add that `fswatch` is also a descendant / grandchild process (a child of one of the two additional child processes), so I'm seeing four in all. Yes,...

I also noticed when changing "*" in the first line to "/*" (which you can do in Git to match the beginning of a path) that `docker-osx-dev` incorrectly interprets the...

> I believe our excludes follow rsync rules, not Docker rules The `docker-osx-dev` code manually reimplements its own line-by-line parsing and interpretation of .gitignore and .dockerignore files (see [here](https://github.com/brikis98/docker-osx-dev/blob/2f201735e24fee5db0de600db21da344da3d3163/src/docker-osx-dev#L956) and...

> Out of curiosity, does !foo/\* work? No, that syncs even fewer files: ``` 2015-12-20 22:08:23 [INFO] Performing initial sync of paths: /Users/chris/temp-sync 2015-12-20 22:08:23 [INFO] Syncing temp-sync/: cd+++++++ 2015-12-20...