Uclibc size

Erik Andersen andersen at codepoet.org
Thu Nov 10 18:02:52 UTC 2005


On Thu Nov 10, 2005 at 03:06:22PM +0000, Mike Frysinger wrote:
> On Thu, Nov 10, 2005 at 01:53:19PM +0200, Dmitry.Kasatkin at nokia.com wrote:
> > Can somebody tell me why there is so big difference and how it is
> > achieved?
> 
> for starters, we dont include every single locale glibc does

And we are very very sneaky in how we store the data.  By
carefully exploiting the enormous amount of redundancy in the
glibc files, Manuel was able to develop data representation that
focus on eliminating this redundancy.  He still intends to
implement another / final version of his uClibc locale work at
some point that completes all the interfaces and adds gettext
support, but his full time job has thus far been keeping him
busy.  And thus far, nobody else has gotten excited about doing
the hard work needed.

 -Erik

--
Erik B. Andersen             http://codepoet-consulting.com/
--This message was written using 73% post-consumer electrons--



More information about the uClibc mailing list