Signed-off-by: Adam Vandenberg <flangy@gmail.com>

This is needed for Ruby 1.9.x compatibility, but we still don't support
running Homebrew with anything other than a system provided 1.8.[67].

Fixes Homebrew/homebrew#6546

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Thus the url we use can change.

This only works provided the urls both contain git repos with the same blobs. But this should always be true.

Adjust mirror ohai slightly.

All Apache Formulae should be updated to use the closer.vgi script
to specify downloads rather than a random mirror that could be out
of date or compromised. Apache's closer.cgi does periodic health
checks.

The base URL for the mirror system is

  http://www.apache.org/dyn/closer.cgi?path=#{filepath}

e.g.:

  http://www.apache.org/dyn/closer.cgi?path=/couchdb/1.0.3/apache-couchdb-1.0.3.tar.gz

Note: The addition of the "Actually downloading..." message is sub-optimal
as the message should probably be emitted in _fetch() rather than fetch(),
but I didn't want to change the way Homebrew works today, so I'm leaving
this for mxcl & team to sort out or adopt :)

Currently, `brew doctor` complains about GREP_OPTIONS being set,
regardless of its value. The cmake formula, for which this check was
introduced, is more specific about what actually makes it fail.

GREP_OPTIONS='--color=auto' will not cause the build to fail, so don't
warn about this.

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

XCode 4.x only includes GCC 4.2. Therefore, having `brew doctor` report that
GCC 4.0 is missing only spreads fear, uncertainty and doubt over a situation
that is status quo.

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Fixes Homebrew/homebrew#6065. Fixes Homebrew/homebrew#6218.

Fixes Homebrew/homebrew#6065.

My pre-emptive fix that avoided calling Pathname.ensure_writable because I was not convinced it worked broke this function due to incorrect logic.

The lesson is, don’t write pre-emptive fixes. Wait until you've seen the bug first. All code has bugs in, so write less. I'm an idiot sometimes.

This line is getting long, and having these on a separate line
will help with some rebasing in some work-in-progress branches.

A spurious "--" argument was being passed to the installer script.

Also clean-up this external command in preparation
for becoming a built-in command.

Make which_versions available in Homebrew module so
that it can be used by other commands in the future.

This allows its use in Leopard, which has Ruby 1.8.6.

- Airfoil during the install of InstantOn changes /usr/local to
  root:wheel ownership

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

This is how we do it in general in Homebrew.

Fixes Homebrew/homebrew#2709.

By forcing dylibs to have an install_name id that is the HOMEBREW_PREFIX path, ie. the symlink’s path. Stuff that links to these dylibs will use this id and thus by immune to upgrades of underlying libraries.

Thus whatever keg is "current" ie. linked, will be the library that is used by the tool.

This fix is not retroactive. So there will still be breakage for existing installations of stuff.

The fix_install step in install is moved after the link step as the symlinking
is required to determine the eventual ids for each dylib.

Before we would always make the symlink itself have a basename the same as what it pointed to.

I don’t think this breaks any of the other usages as they were working by coincidence.