2.02-dev release coming

[Joe Schaefer <jo...@sunstarsys.com>]

Based on the (ack'd) segfault in Apache::Cookie::new()
posted to the modperl list today, I think it'd be a 
good idea to put out another release as soon 
as we can.  I'd like to add a few more tests to
to make sure the fix I committed today is correct,
and then I'll post a 2.02-dev release candidate
for review.

If there's anything else folks want to see in
2.02-dev, now would be a good time to say so.

-- 
Joe Schaefer

Re: 2.02-dev release coming

[Stas Bekman <st...@stason.org>]

Randy Kobes wrote:
> ... The problems I had before about the APR module
> not calling apr_initialize() have now disappeared, perhaps
> due to some changes in mp2 or libapreq2, but more probably
> due to me understanding things better. What this boils down
> to is I now have a glue/perl/t/cgi.t test (essentially the
> same as the env/t/cgi.t test), which uses a
> glue/perl/t/cgi-bin/test_cgi.pl (essentially a perl version
> of env/test_cgi.c). test_cgi.pl, as well as needing
> libapreq2.(so|dll), needs mod_perl.so to resolve some
> symbols, 

Eventually, we will make a clean separation and there will be no dependency on 
mod_perl.so for external usages of APR and its subclasses. I did some changes 
to APR.pm BOOT quite recently, so that's why you no longer see problems with 
apr_initialize. I could be wrong.



__________________________________________________________________
Stas Bekman            JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/     mod_perl Guide ---> http://perl.apache.org
mailto:stas@stason.org http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org   http://ticketmaster.com

Re: 2.02-dev release coming

[Randy Kobes <ra...@theoryx5.uwinnipeg.ca>]

On Wed, 12 Nov 2003, Joe Schaefer wrote:

> Based on the (ack'd) segfault in Apache::Cookie::new()
> posted to the modperl list today, I think it'd be a
> good idea to put out another release as soon
> as we can.  I'd like to add a few more tests to
> to make sure the fix I committed today is correct,
> and then I'll post a 2.02-dev release candidate
> for review.
>
> If there's anything else folks want to see in
> 2.02-dev, now would be a good time to say so.

This is probably premature, and perhaps could wait for a
more leisurely examination, but I've been looking at the CGI
perl glue. The problems I had before about the APR module
not calling apr_initialize() have now disappeared, perhaps
due to some changes in mp2 or libapreq2, but more probably
due to me understanding things better. What this boils down
to is I now have a glue/perl/t/cgi.t test (essentially the
same as the env/t/cgi.t test), which uses a
glue/perl/t/cgi-bin/test_cgi.pl (essentially a perl version
of env/test_cgi.c). test_cgi.pl, as well as needing
libapreq2.(so|dll), needs mod_perl.so to resolve some
symbols, and creates the objects as
   my $p = APR::Pool->new();
   my $req = Apache::Request->new($p);
   ....
   my %cookies = Apache::Cookie->fetch($p);
Unfortunately, there seems to be a problem with large posted
data. But if one keeps the posts to under some maximum size,
all request and cookie tests pass.

-- 
best regards,
randy