No offense but the use of --with-framework-name option is totally wrong in the
Formula.
As noted in the Mac build [notes][1] --with-framework-name is *not* to pass
the SDK path, but to rename it (e.g. "AwesomePython.framework" instead of just
"Python.framework"). To build it as a Mac OS X framework you need to use the
--enable-framework flag instead.
The same with the other option -- to build Python universally. The flag
--enable-universalsdk is missing.
[1]: http://svn.python.org/projects/python/branches/release26-maint/Mac/README
Beanstalk is a simple, fast workqueue service. Its interface is generic,
but was originally designed for reducing the latency of page
views in high-volume web applications by running
time-consuming tasks asynchronously.
Signed-off-by: Max Howell <max@methylblue.com>
I changed it to beanstalk.rb as that's how they refer to it in general.
Merged the clj_repl and clj commands into one clj command, which either
invokes the REPL, or runs a script (similar to how the python command works).
Moved the script to DATA, instead of inline
Also fixes etc config.
The call to inreplace had an incorrect path. I tried correcting it
and leaving it as a post-make-install step, but it still did not work.
Moving the inreplace call did the trick.
Signed-off-by: Max Howell <max@methylblue.com>
The formula was committed as liboggz.rb, but the homepage is consistent in
calling it Oggz so I renamed the formula.
ClosesHomebrew/homebrew#85. ClosesHomebrew/homebrew#86.
Removing .la files is proving problematic in places. But oh well. Homebrew is
all about optimizing init?
Also only set ARCHFLAGS on non-Core Duo machines... though it seemed to build
fine even with it, it doenst make much sense to me... these changes are kind
of up for debate but it "works for me" this way...
ClosesHomebrew/homebrew#75
It's a lost cause trying to get 1.0beta3 compiling under Snow Leopard,
and the trunk already compiles 64-bit clean, so we might as well
use that.
Note that the DATA patch makes a few changes that are then rewritten
by inreplace. Why not patch clean and not run inreplace later? Because
I'm going to submit the patch back upstream, which should hopefully
allow us to drop it from the formula and only use inreplace to select
our build environment (native arch and framework for either 10.5 or 10.6)
