Philippe M. Chiasson wrote:
Also, one last thing of importance I guess, is anybody seeing test
failures with current CVS of any platform ?
Fine on all my FreeBSD systems:
FreeBSD 4.8-STABLE perl 5.6.1 OK
FreeBSD 4.8-RELEASE perl 5.6.1 OK
FreeBSD 4.1-RELEASE perl 5.6.1 OK
FreeBSD 4.1-RELEASE p
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
what happens if the key is used in a numeric context? won't this
scheme blow?
Sure would, if someone's expecting a key to have
a normal numeric value. I won't try to pursue
the point further, I'm just trying to leave you
with
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
> what happens if the key is used in a numeric context? won't this
> scheme blow?
Sure would, if someone's expecting a key to have
a normal numeric value. I won't try to pursue
the point further, I'm just trying to leave you
with an idea of how cr
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
I'd hate to take the joy of adding these wonderful features away from
Philippe, who's now busy wrestling with mod_perl 1.28 release, but
once done will certainly love to do those. Right Philippe?
Curiosity got the better of me. Here'
Stas Bekman wrote:
[EMAIL PROTECTED] wrote:
randyk 2003/06/05 20:45:53
select undef, undef, undef, 0.25
-while -e $file && -s _ == 10 || $t++ == 20;
+while -e $file && -s _ == SIZE || $t++ == TRIES;
Looking at this code, it's totally bogus, bad copy-n-paste from
Stas Bekman <[EMAIL PROTECTED]> writes:
> I'd hate to take the joy of adding these wonderful features away from
> Philippe, who's now busy wrestling with mod_perl 1.28 release, but
> once done will certainly love to do those. Right Philippe?
Curiosity got the better of me. Here's a patch that s
[EMAIL PROTECTED] wrote:
randyk 2003/06/05 20:45:53
select undef, undef, undef, 0.25
-while -e $file && -s _ == 10 || $t++ == 20;
+while -e $file && -s _ == SIZE || $t++ == TRIES;
Looking at this code, it's totally bogus, bad copy-n-paste from cleanup.t on
my part.
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
so we can't simply alias FETCH to apr_table_get, the following sort
of works (but breaks some other things):
static MP_INLINE
const char *mpxs_APR__Table_FETCH(pTHX_ SV *tsv, const char *key)
{
apr_table_t *t = mp_xs_sv2_APR__
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
> so we can't simply alias FETCH to apr_table_get, the following sort
> of works (but breaks some other things):
>
> static MP_INLINE
> const char *mpxs_APR__Table_FETCH(pTHX_ SV *tsv, const char *key)
> {
> apr_table_t *t = mp_xs_sv2_APR__Table
Randy Kobes wrote:
my $target_dir = catfile $vars->{documentroot}, "util";
+my $perl = catfile $Config{bin},
+(Apache::TestConfig::WIN32 ? 'perl.exe' : 'perl');
+
In that case, we do know the path to perl, it's stored in Apache::Build:
require Apache::Build;
my $build = A
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
Hmm, what about storing the pointer to apreq_value_t as val
AFAICT, that helps nada.
true, again because of newSVpv
in the table and subclass get() to do what you want?
Jus
On Thu, 5 Jun 2003, Stas Bekman wrote:
> Randy Kobes wrote:
[ .. ]
> > Yes, it does - thanks. Here's a diff that enables all the
> > apache/subprocess tests to pass on Win32 (I haven't tested
> > it on Unix):
> >
> > ==
> > Index: subprocess.
On Thu, 2003-06-05 at 09:42, Stas Bekman wrote:
> Philippe M. Chiasson wrote:
> > On Wed, 2003-06-04 at 13:38, Stas Bekman wrote:
> >
> >>blead-perl w/o ithreads fails modules/module.t with this error:
> >>
> >>[Wed Jun 4 15:34:41 2003] [error] Can't locate
> >>auto/Apache/Module/top_module.al i
[EMAIL PROTECTED] wrote:
gozer 2003/06/05 00:09:20
Modified:.Makefile.PL
t/docs/auth .htaccess
Log:
s/Apache::(OK|DECLINED)/Apache::Constants::$1/g
argh, now that I looked in eagle: ch06, it uses:
PerlAuthenHandler Apache::OK
And it has this footnote:
*Apa
Stas Bekman <[EMAIL PROTECTED]> writes:
> Joe Schaefer wrote:
> > Stas Bekman <[EMAIL PROTECTED]> writes:
> >
> >> Hmm, what about storing the pointer to apreq_value_t as val
> > AFAICT, that helps nada.
>
> true, again because of newSVpv
>
> >>in the table and subclass get() to do what you want
Randy Kobes wrote:
On Wed, 4 Jun 2003, Stas Bekman wrote:
Randy Kobes wrote:
Hi,
On Win32, the apache/subprocess tests fail on Win32.
This is due to the following: currently, the tests do
something like
@argv = qw(potentially something);
$command = catfile $target_dir, "some_script.pl";
On Wed, 4 Jun 2003, Stas Bekman wrote:
> Randy Kobes wrote:
> > Hi,
> >On Win32, the apache/subprocess tests fail on Win32.
> > This is due to the following: currently, the tests do
> > something like
> >@argv = qw(potentially something);
> >$command = catfile $target_dir, "some_script
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
Hmm, what about storing the pointer to apreq_value_t as val
AFAICT, that helps nada.
true, again because of newSVpv
in the table and subclass get() to do what you want?
Just subclassing get() might work! At the moment
I'm concerne
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
Thanks for the drawing. Now I don't understand how do you get to the
apreq_value_t pointer from table_entry->val. Do you just subtract the
size of the apreq_value_t struct and based on the knowledge that the
memory allocation is c
Stas Bekman <[EMAIL PROTECTED]> writes:
> Hmm, what about storing the pointer to apreq_value_t as val
AFAICT, that helps nada.
> in the table and subclass get() to do what you want?
Just subclassing get() might work! At the moment
I'm concerned about how each() is implemented; does
each make
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
> Thanks for the drawing. Now I don't understand how do you get to the
> apreq_value_t pointer from table_entry->val. Do you just subtract the
> size of the apreq_value_t struct and based on the knowledge that the
> memory allocation is continues, you
Hmm, what about storing the pointer to apreq_value_t as val in the table and
subclass get() to do what you want?
__
Stas BekmanJAm_pH --> Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide ---> http://
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
I still don't see how that's useful here. There's nothing wrong with the
data inside apr_table. It's the c2perl
translation (newSVpv) that's causing the lossage.
and I don't get where the lossage is :(
OK- I see we're having a
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
> > I still don't see how that's useful here. There's nothing wrong with the
> > data inside apr_table. It's the c2perl
> > translation (newSVpv) that's causing the lossage.
>
> and I don't get where the lossage is :(
OK- I see we're having a disc
Fwd
--- Begin Message ---
Stas Bekman <[EMAIL PROTECTED]> writes:
> Joe Schaefer wrote:
> > Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
> >>What if you use mpxs_apr_table_do to fixup the elements on the first
> >> FETCH? Would that be a bad approach?
> > Not sure what you have in mind.
>
> yo
Index: xs/APR/Table/APR__Table.h
===
RCS file: /home/cvspublic/modperl-2.0/xs/APR/Table/APR__Table.h,v
retrieving revision 1.9
diff -u -r1.9 APR__Table.h
--- xs/APR/Table/APR__Table.h 24 Aug 2002 17:15:58 - 1.9
+++ xs/APR/Table/APR
Stas Bekman wrote:
Cool, now if you add:
require Apache::TestClient;
at the beginning of sub lwp_call {
it should have loaded it on line 30 of TestRequest.pm
waiting for server to start: ..[Thu Jun 05 12:25:48 2003] [info] 22
Apache:: modules loaded
[Thu Jun 05 12:25:48 2003] [info] 5 APR
Carl Brewer wrote:
Stas Bekman wrote:
what happens if you replace:
my $r = (\&{$name})->(@_);
with:
require Carp;
my $r = Apache::TestClient::GET->(@_);
Carp::confess("oops") unless defined $r;
Undefined subroutine &Apache::TestClient::GET called at
/data/src/modperl-2.0/Apache
Stas Bekman wrote:
what happens if you replace:
my $r = (\&{$name})->(@_);
with:
require Carp;
my $r = Apache::TestClient::GET->(@_);
Carp::confess("oops") unless defined $r;
waiting for server to start: ok (waited 4 secs)
server localhost:8529 started
server localhost:8530 lis
Carl Brewer wrote:
Can't call method "method" on an undefined value at
/data/src/modperl-2.0/Apache-Test/lib/Apache/TestRequest.pm line 411.
*** Error code 61
Stop.
make: stopped in /data/src/modperl-2.0
steel1: {10}
Do I need to send you something else as well? That doesn't look any
more usefu
Philippe M. Chiasson wrote:
I've narrowed down a strange bug I reported a while back.
Seems that using constant for Perl* directives doesn't quite work for
perl 5.6.1 and 5.6.x
PerlModule Apache::Constants
SetHandler perl-script
PerlHandler Apache::OK
I don't think it's supposed to work. It
Stas Bekman wrote:
I thought that the reason is that you don't have LWP installed. So I was
trying to reproduce it, but it still worked. Can you try again with this
patch. I want to know whom to blame.
Patch applied. Here's the output from another make test :
steel1: {9} make test
cd "src/m
Philippe M. Chiasson wrote:
Only 1 issue still not quite cleared yet. Stas suggested bundling
Apache-Test with this release of mp1. I don't see how that could be a
bad thing, since getting Apache-Test out there and adopted asap is a
good thing. Any strong feelings against it? Otherwise I'd just bu
Philippe M. Chiasson wrote:
On Wed, 2003-06-04 at 13:38, Stas Bekman wrote:
blead-perl w/o ithreads fails modules/module.t with this error:
[Wed Jun 4 15:34:41 2003] [error] Can't locate
auto/Apache/Module/top_module.al in @INC (@INC contains:
/home/stas/apache.org/modperl/blib/lib /home/stas/
Carl Brewer wrote:
-8<-- Start Bug Report 8<--
1. Problem Description:
Seeing this at make test time from today's CVS snapshot of mp2.
I haven't tried this with older snapshots or releases. The
box is running NetBSD 1.6.1 i386. :
steel1# make test
cd "src/mo
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
[...]
Still, once user tries to modify the data, what happens to meta-data?
When COW is performed, does it copy the whole thing or just whatever
was pointed by SvCUR?
Doesn't
We get more and more bug reports with obscure failures when the real reason is
that wrong mp2 libraries get loaded. e.g., older builds, incompatible builds.
I thought it'd be only us developers, hit by this issue, but apparently the
problem is relevant for normal users as well. Hence, I'm thinki
-8<-- Start Bug Report 8<--
1. Problem Description:
Seeing this at make test time from today's CVS snapshot of mp2.
I haven't tried this with older snapshots or releases. The
box is running NetBSD 1.6.1 i386. :
steel1# make test
cd "src/modules/perl" && make
Philippe M. Chiasson wrote:
On Wed, 2003-06-04 at 14:55, Issac Goldstand wrote:
After a few fruitless days of fiddling with this, I find myself getting
nowhere... Can anyone really explain how Apache::ReadConfig and/or direct
me to the source code (for handling the namespace, not the code for
se
Stas Bekman <[EMAIL PROTECTED]> writes:
> Joe Schaefer wrote:
> > Stas Bekman <[EMAIL PROTECTED]> writes:
> > [...]
> >
> >>Still, once user tries to modify the data, what happens to meta-data?
> >>When COW is performed, does it copy the whole thing or just whatever
> >>was pointed by SvCUR?
> > D
40 matches
Mail list logo
