On Fri, Sep 16, 2011 at 5:36 PM, Reuben Thomas <r...@sc3d.org> wrote: > On 16 September 2011 21:25, Hisham <hisham...@gmail.com> wrote: >> >> I'm using uname, but I'm "normalizing" the received values a bit, as >> some platforms tend to provide different results (for example, I spot >> `elseif proc:match("amd64") or proc:match("x86_64") then proc = >> "x86_64"` in the code). > > Why normalize? On a particular system is one not more likely to use > that system's uname -m names? I don't expect those names to be > portable, since if I'm using multiple OSes, I'll need to add the OS > name to the arch-specific path too.
IIRC, it's because I have things such as "pass these flags to GCC when running on this processor", and I don't care if the OS calls it "amd64" or "x86_64". -- -- Hisham http://hisham.hm/ - http://colorbleed.com.br/ ------------------------------------------------------------------------------ BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA http://p.sf.net/sfu/rim-devcon-copy2 _______________________________________________ Luarocks-developers mailing list Luarocks-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/luarocks-developers