[links-list] Re: gzipped files - final

Witold Filipczyk juandon at poczta.onet.pl
Wed May 1 11:37:52 PDT 2002

On Wed, May 01, 2002 at 07:53:24PM +0200, Petr Baudis wrote:
> Dear diary, on Wed, May 01, 2002 at 01:02:41PM CEST, I got a letter,
> where Witold Filipczyk <juandon at poczta.onet.pl> told me, that...
> > 
> > Compressed local files and compressed http it will be easier and clearer
> > to handle independently (without additional "meta-library").
> > I know it is difficult to understand what I mean.
> > In attachment there is an example.
> See below. The code duplication may rise dramatically with adding of the new
> encodings, and I just want to be able to add it on one place and feel safe.
> > Now gzipped files support is clearer.
> > Maybe bufsize_for_gzip should be increased (8192 will be better ?)
> > I don't know how to get size of decompressed file before decompression.
> > There is no need to test if file is gzipped or not because zlib
> > handle both.
> Yes, the new patch looks much better! :)
Which patch ? ;-)

> > I don't really know how it should be.
> > In today's version zlib is not enabled by default and configure.in surely
> > needs some tuning.
> Hmm, why it's not enabled by default?

First of all, not to prejudice :)

> > > This is something similiar. Well, no sense to comment on this, as we're going
> > > to move this away anyway, aren't we?
> > 
> > I don't think so.
> > Now support for gzip files is clear (only in file.c).
> And when we're going to add support for another encodings? From bzip2 to
> crypted files? ;)

I think it is a progress.
And now is better than was before.
Is there anybody interested in bzip2 support?

> I just want to do things properly, when we're already doing them. Otherwise
> only mess and code duplication will arise in the future..

Oh sorry, it seem to me, that E in elinks stands for experimental.

Unsubscribe: send email to links-list-request at linuxfromscratch.org
and put unsubscribe in the subject header of the message

More information about the links-list mailing list