You are viewing a plain text version of this content. The canonical link for it is here.
Posted to apreq-dev@httpd.apache.org by Joe Schaefer <jo...@sunstarsys.com> on 2004/08/05 18:32:39 UTC

plans for 2.04-dev

I think we've got enough new stuff in cvs
to start talking about our goals for a 2.04-dev 
release.  The only thing I plan to do before
then is flesh out Table.pod and Error.pod,
so if there are any issues/problems with
the current exception scheme (or table api)
we should flesh it out now, preferably in the 
doc tests!

Any other ideas about what we want to include in 
2.04-dev? (Remember our ultimate goal is to 
release a stable 2.05 around November's ApacheCON).

-- 
Joe Schaefer


Re: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
On Thu, 2004-08-26 at 18:22, Stas Bekman wrote:

> No, there are all not required modules. Module::Build is used by 
> Apache::Test, if you are using Apache::TestMB (which is not used by 
> mod_perl). Data::Flow is used only by developers of mod_perl 2.0. You 
> should look for the Makefile.PL requirements, not any use()/require() 
> statements you can see in the distribution.

Thanks.

-- 
Bojan


Re: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
Quoting Stas Bekman <st...@stason.org>:

> No, there are all not required modules. Module::Build is used by
> Apache::Test, if you are using Apache::TestMB (which is not used by
> mod_perl). Data::Flow is used only by developers of mod_perl 2.0. You
> should look for the Makefile.PL requirements, not any use()/require()
> statements you can see in the distribution.

Once again, thanks for letting me know about this. I have found how Red Hat do
this in mod_perl - they use a little script called filter-requires.sh that
hoses the bogus requirements out of perl.req file. I've added all of these to
the list and now mod_perl RPM installs cleanly. Whether it works or not will
have to be determined by someone that actually does know Perl and mod_perl.

In any event, I'll upload the new mod_perl RPMS to the same location once I get
to the box that is allowed to do that. Shouldn't take more than a day.

--
Bojan

Re: plans for 2.04-dev

Posted by Stas Bekman <st...@stason.org>.
Bojan Smojver wrote:
> Quoting Bojan Smojver <bo...@rexursive.com>:
> 
> 
>>Once I install that, I'm back to building libapreq2 RPMs. We'll see how that
>>goes...
> 
> 
> A question in relation to this for all Red Hat/Fedora/Perl folk out there...
> Once I've built the module (mod_perl), I attempted to install, which came out
> with missing dependencies, namely perl(Carp::Heavy), perl(Data::Flow) and
> perl(Module::Build). I can see Carp::Heavy inside Perl RPM and cpan script
> claims it's part of Perl (which seems to make sense). The other two I fetched
> from CPAN and they installed OK, but the dependency problem is still there. I'm
> guessing the problem is actually FC2 RPM and it's "provides" list, right? Or is
> there something more sinister at work?

No, there are all not required modules. Module::Build is used by 
Apache::Test, if you are using Apache::TestMB (which is not used by 
mod_perl). Data::Flow is used only by developers of mod_perl 2.0. You 
should look for the Makefile.PL requirements, not any use()/require() 
statements you can see in the distribution.


-- 
__________________________________________________________________
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: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
Quoting Bojan Smojver <bo...@rexursive.com>:

> Once I install that, I'm back to building libapreq2 RPMs. We'll see how that
> goes...

A question in relation to this for all Red Hat/Fedora/Perl folk out there...
Once I've built the module (mod_perl), I attempted to install, which came out
with missing dependencies, namely perl(Carp::Heavy), perl(Data::Flow) and
perl(Module::Build). I can see Carp::Heavy inside Perl RPM and cpan script
claims it's part of Perl (which seems to make sense). The other two I fetched
from CPAN and they installed OK, but the dependency problem is still there. I'm
guessing the problem is actually FC2 RPM and it's "provides" list, right? Or is
there something more sinister at work?

I have forced the install of mod_perl and mod_perl-devel, simply to complete
build of libapreq2 RPMs. This worked and I'll put them up as soon as I get to
the system  that allows me to do that. Will let everyone know what the URLs
are.

--
Bojan

Re: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
Just building mod_perl 1.99_16 RPM on FC2. It's easier than I thought, thanks to
Red Hat's excellent spec file and, of course, the module itself.

Once I install that, I'm back to building libapreq2 RPMs. We'll see how that
goes...

Quoting Bojan Smojver <bo...@rexursive.com>:

> Sorry, been swamped recently. Will have a look at it when I get a
> minute...
>
> On Sat, 2004-08-21 at 05:31, Joe Schaefer wrote:
> > Bojan Smojver <bo...@rexursive.com> writes:
> >
> > [...]
> >
> > > Thanks for this piece of info. At present, the CVS fails to build as an
> > > RPM, I'm guessing due to mod_perl issues (I'm not a Perl guy, so you'll
> > > have to excuse my ignorance).
> >
> > mp2 1.99_15 was released today, can you build the rpm now?
> --
> Bojan
>
>


--
Bojan

Re: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
Sorry, been swamped recently. Will have a look at it when I get a
minute...

On Sat, 2004-08-21 at 05:31, Joe Schaefer wrote:
> Bojan Smojver <bo...@rexursive.com> writes:
> 
> [...]
> 
> > Thanks for this piece of info. At present, the CVS fails to build as an
> > RPM, I'm guessing due to mod_perl issues (I'm not a Perl guy, so you'll
> > have to excuse my ignorance).
> 
> mp2 1.99_15 was released today, can you build the rpm now?
-- 
Bojan


Re: plans for 2.04-dev

Posted by Joe Schaefer <jo...@sunstarsys.com>.
Bojan Smojver <bo...@rexursive.com> writes:

[...]

> Thanks for this piece of info. At present, the CVS fails to build as an
> RPM, I'm guessing due to mod_perl issues (I'm not a Perl guy, so you'll
> have to excuse my ignorance).

mp2 1.99_15 was released today, can you build the rpm now?

-- 
Joe Schaefer


Re: plans for 2.04-dev

Posted by Bojan Smojver <bo...@rexursive.com>.
On Fri, 2004-08-06 at 02:32, Joe Schaefer wrote:

> Any other ideas about what we want to include in 
> 2.04-dev? (Remember our ultimate goal is to 
> release a stable 2.05 around November's ApacheCON).

Thanks for this piece of info. At present, the CVS fails to build as an
RPM, I'm guessing due to mod_perl issues (I'm not a Perl guy, so you'll
have to excuse my ignorance).

-- 
Bojan


Re: plans for 2.04-dev

Posted by Stas Bekman <st...@stason.org>.
Joe Schaefer wrote:
[...]
> One major issue we need to discuss concerns our mp2 1.99_15
> (current-cvs) dependency.  I don't know if mp2 is planning
> to roll another release anytime soon.  If they are planning
> to roll one this month, I don't mind waiting until then.
> If not, I'd like to start the release candidate process for
> 2.04-dev sometime this week.

I hope there will be _15 in a week or two. It has been a while since _14 
release, but there were lots of API tweaks, which went back and forth. 
There are a few bugs that need to be addressed before we can release _15.


-- 
__________________________________________________________________
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: plans for 2.04-dev

Posted by Randy Kobes <ra...@theoryx5.uwinnipeg.ca>.
On Sun, 8 Aug 2004, Joe Schaefer wrote:

> Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:
>
> > On Thu, 5 Aug 2004, Joe Schaefer wrote:
> >
> > > Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:
> > >
> > > [...]
> > >
> > > > - there's something amiss with $cookie->secure(); it
> > > > always seems to be set to 0;
> > >
> > > Just on Win32, or more generally?
> >
> > Sorry - just on Win32 there was a problem.

Actually, that seems to have beenn cleared up in the
lates cvs version.

> >
> > > > I'm working on those, but will be away for a couple of weeks
> > > > - if you'd like to put out a release before these are
> > > > resolved, we could just arrange to have these tests skipped
> > > > on Win32 (in particular, the tests involving POSIX::close(0)
> > > > on Win32 now cause a crash).
> > >
> > > I recently recoded apreq_params around the new
> > > req->body_status flag, so I'm pretty sure we can
> > > just set it to APR_SUCCESS and do away with
> > > POSIX::close(0) in the doc tests.
> >
> > I've seen you've done that now - I'll take a look
> > tonight on Win32. Thanks.
>
> Randy, any news on these?

Unfortunately, I'm away from my Win32 machine for
almost the next 2 weeks. But just before leaving I tried
the current cvs - it hung on docs/Request.t,
but passed all tests prior to that. I didn't have
time to look into it though - I will when I get back.

-- 
best regards,
randy


Re: plans for 2.04-dev

Posted by Joe Schaefer <jo...@sunstarsys.com>.
Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:

> On Thu, 5 Aug 2004, Joe Schaefer wrote:
> 
> > Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:
> >
> > [...]
> >
> > > - there's something amiss with $cookie->secure(); it
> > > always seems to be set to 0;
> >
> > Just on Win32, or more generally?
> 
> Sorry - just on Win32 there was a problem.
> 
> > > I'm working on those, but will be away for a couple of weeks
> > > - if you'd like to put out a release before these are
> > > resolved, we could just arrange to have these tests skipped
> > > on Win32 (in particular, the tests involving POSIX::close(0)
> > > on Win32 now cause a crash).
> >
> > I recently recoded apreq_params around the new
> > req->body_status flag, so I'm pretty sure we can
> > just set it to APR_SUCCESS and do away with
> > POSIX::close(0) in the doc tests.
> 
> I've seen you've done that now - I'll take a look
> tonight on Win32. Thanks.

Randy, any news on these?

Also, I just rolled current-cvs and put it at 

  http://cvs.apache.org/~joes/libapreq2-2.04-dev.tar.gz

One major issue we need to discuss concerns our mp2 1.99_15
(current-cvs) dependency.  I don't know if mp2 is planning
to roll another release anytime soon.  If they are planning
to roll one this month, I don't mind waiting until then.
If not, I'd like to start the release candidate process for
2.04-dev sometime this week.

-- 
Joe Schaefer


Re: plans for 2.04-dev

Posted by Randy Kobes <ra...@theoryx5.uwinnipeg.ca>.
On Thu, 5 Aug 2004, Joe Schaefer wrote:

> Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:
>
> [...]
>
> > - there's something amiss with $cookie->secure(); it
> > always seems to be set to 0;
>
> Just on Win32, or more generally?

Sorry - just on Win32 there was a problem.

> > I'm working on those, but will be away for a couple of weeks
> > - if you'd like to put out a release before these are
> > resolved, we could just arrange to have these tests skipped
> > on Win32 (in particular, the tests involving POSIX::close(0)
> > on Win32 now cause a crash).
>
> I recently recoded apreq_params around the new
> req->body_status flag, so I'm pretty sure we can
> just set it to APR_SUCCESS and do away with
> POSIX::close(0) in the doc tests.

I've seen you've done that now - I'll take a look
tonight on Win32. Thanks.

-- 
best regards,
randy

Re: plans for 2.04-dev

Posted by Joe Schaefer <jo...@sunstarsys.com>.
Randy Kobes <ra...@theoryx5.uwinnipeg.ca> writes:

[...]

> - there's something amiss with $cookie->secure(); it
> always seems to be set to 0;

Just on Win32, or more generally?

> I'm working on those, but will be away for a couple of weeks
> - if you'd like to put out a release before these are
> resolved, we could just arrange to have these tests skipped
> on Win32 (in particular, the tests involving POSIX::close(0)
> on Win32 now cause a crash).

I recently recoded apreq_params around the new
req->body_status flag, so I'm pretty sure we can
just set it to APR_SUCCESS and do away with 
POSIX::close(0) in the doc tests.

-- 
Joe Schaefer


Re: plans for 2.04-dev

Posted by Randy Kobes <ra...@theoryx5.uwinnipeg.ca>.
On Thu, 5 Aug 2004, Joe Schaefer wrote:

>
> I think we've got enough new stuff in cvs
> to start talking about our goals for a 2.04-dev
> release.  The only thing I plan to do before
> then is flesh out Table.pod and Error.pod,
> so if there are any issues/problems with
> the current exception scheme (or table api)
> we should flesh it out now, preferably in the
> doc tests!

There's a couple of Win32-specific things in the perl glue
that I haven't tracked down yet, but would be good to
resolve before a release:
- as you suspected, Win32 doesn't like the use of POSIX
in the docs/*.t tests (specifically, POSIX::close(0));
- there's something amiss with $cookie->secure(); it
always seems to be set to 0;
I'm working on those, but will be away for a couple of weeks
- if you'd like to put out a release before these are
resolved, we could just arrange to have these tests skipped
on Win32 (in particular, the tests involving POSIX::close(0)
on Win32 now cause a crash).

-- 
best regards,
randy