Perl 5 LWP Design

Roy T. Fielding (fielding@avron.ICS.UCI.EDU)
Sun, 05 Mar 1995 05:33:07 -0800


> One thought that strikes me is why we don't just provide a perl
> interface to the CERN libwww code.  Does anybody know the internals of
> libwww well enough to comment on how difficult this would be?

Very difficult in terms of understanding.  The CERN library is massively
mucked-up with C macros.  You would also lose the ability to interactively
debug the library internals.  When I originally wrote libwww-perl, the CERN
code was also a bit buggy and out-of-date.

OTOH, Henrik has improved things greatly in v3 and it will always have more
people supporting it than lwp does.  I think Dan Connolly once wrote a
perl to CERN libwww interface, but that was over a year ago and was never
complete.


......Roy Fielding   ICS Grad Student, University of California, Irvine  USA
                                     <fielding@ics.uci.edu>
                     <URL:http://www.ics.uci.edu/dir/grad/Software/fielding>