Re: canonicalize-lgpl bug

[Eric Blake]

Eric Blake <ebb9 <at> byu.net> writes:

> I noticed that glibc has applied some fixes to canonicalize.c that we were 
> missing; and sure enough, was able to enhance the testsuite to trip these 
> bugs.  They also applied some assert's proving that their optimization to 
avoid 
> strcpy are safe; I left the asserts out, but copied their optimization.  OK 
to 
> commit?

Hmm, on thinking about this more, there are probably existing Linux machines 
where glibc canonicalize_file_name is broken (fails to detect ENOTDIR 
on "file/"), and therefore we need to replace it.  Additionally, it would be 
nice to declare canonicalize_file_name in <stdlib.h>.  Also, now that POSIX 
requires realpath(name,NULL) to malloc() its results as an absolute path, it 
would be nice to have a realpath module that replaces broken versions (Solaris 
sometimes gives relative paths, and not all existing realpath implementations 
malloc).

-- 
Eric Blake