Ideally we could do a get request and fill in the issue form at GitHub. Presumably though, we'd only do this if the user typed brew report or somesuch.

Someone should get that implemented!

It's a great feature but it seems to conflict with our new system() replacement. Clearly the bug is still in our system() but what exactly it is is not clear and this bug sucks.

See Homebrew/homebrew#124

As I understand it (I tested too), signal propogation is handled by the parent Ruby process. However this was working mostly anyway. So I don't fully understand what is going on. However this seems to not hang in Process.wait where it was before for one test case.

It just seems to behave strangely with SIGINT. Eg. SIGINT causes tar to exit, but the SIGINT is ignored by our process. This is not the case when used with curl.

We need a test for this.

Closes Homebrew/homebrew#116

Thus install.rb gets the path to the instantiating brew process rather than whichever is first in the PATH. Not to mention Homebrew doesn't *have* to be in the PATH.

Because it's a global.h constant and that file can be included by other projects.

It's a constant and Dir.getwd can change.

The rationale here is that the --verbose mode had a bug where it didn't escape its parameters properly. Which caused ocassionally cryptic issues.

The problem was that git checkout -b doesn't work the second time. This simple solution works by never creating a local branch.

More sensible, and fixes brew

We trap the INT in the install process now so the newline race condition is avoided. Much neater and more polished now.