Help! :)  Let's get this guy off the list :)


---------- Forwarded message ----------
From:  <support@afnic.fr>
Date: Mon, Dec 7, 2009 at 11:39 AM
Subject: <Ref682782> AFNIC / ACCUSE RECEPTION MAIL
To: fred@redhotpenguin.com


Bonjour,

Votre courriel "Fwd: svn commit: r888069 -
/perl/modperl/docs/trunk/src/about/pronounce.pod" a bien été reçu par
le support de l'AFNIC et porte le numéro 682782. Il fera l'objet d'un
traitement dès que possible.

Dans l'intervalle, nous vous invitons à consulter notre FAQ et les
actualités des opérations à l'AFNIC :

http://www.afnic.fr/doc/faq
http://www.afnic.fr/operations

Cordialement,
----------
Hello,

AFNIC support has received your message "Fwd: svn commit: r888069 -
/perl/modperl/docs/trunk/src/about/pronounce.pod". We will handle it
as soon as possible.

Regards.
>>>>>>>>>>>>>>>>>>>>>>>
Sujet : Fwd: svn commit: r888069 -
/perl/modperl/docs/trunk/src/about/pronounce.pod
N° dossier : 682782
Corps du mail : Just committed this to hopefully kill the warnings we
have been getting.

Thanks to Randy for the great suggestion.


---------- Forwarded message ----------
From:  <phred@apache.org>
Date: Mon, Dec 7, 2009 at 10:34 AM
Subject: svn commit: r888069 - /perl/modperl/docs/trunk/src/about/pronounce.pod
To: docs-cvs@perl.apache.org


Author: phred
Date: Mon Dec  7 18:34:32 2009
New Revision: 888069

URL: http://svn.apache.org/viewvc?rev=888069&view=rev
Log:
Remove blank line which may be causing "spurious '>'" error.

Modified:
   perl/modperl/docs/trunk/src/about/pronounce.pod

Modified: perl/modperl/docs/trunk/src/about/pronounce.pod
URL: http://svn.apache.org/viewvc/perl/modperl/docs/trunk/src/about/pronounce.pod?rev=888069&r1=888068&r2=888069&view=diff
==============================================================================
--- perl/modperl/docs/trunk/src/about/pronounce.pod (original)
+++ perl/modperl/docs/trunk/src/about/pronounce.pod Mon Dec  7 18:34:32 2009
@@ -79,7 +79,6 @@
        <td class="table-bg">
                <a href="pronounce/pereinar.ogg">Listen</a>
        </td>
-
 </tr>
 </table>




---------------------------------------------------------------------
To unsubscribe, e-mail: docs-cvs-unsubscribe@perl.apache.org
For additional commands, e-mail: docs-cvs-help@perl.apache.org

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org






NOUVEAU: Foire aux Questions (FAQ) dédiée aux Bureaux d'enregistrement AFNIC
http://operations.afnic.fr/fr/faq

--

Pensez-y : toute l'actualité des opérations de l'Afnic est sur
http://operations.afnic.fr/
Support Afnic
Tel: +33 1 39 30 83 00
Fax: +33 1 39 30 83 83
support@afnic.fr

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Just committed this to hopefully kill the warnings we have been getting.

Thanks to Randy for the great suggestion.


---------- Forwarded message ----------
From:  <phred@apache.org>
Date: Mon, Dec 7, 2009 at 10:34 AM
Subject: svn commit: r888069 - /perl/modperl/docs/trunk/src/about/pronounce.pod
To: docs-cvs@perl.apache.org


Author: phred
Date: Mon Dec  7 18:34:32 2009
New Revision: 888069

URL: http://svn.apache.org/viewvc?rev=888069&view=rev
Log:
Remove blank line which may be causing "spurious '>'" error.

Modified:
   perl/modperl/docs/trunk/src/about/pronounce.pod

Modified: perl/modperl/docs/trunk/src/about/pronounce.pod
URL: http://svn.apache.org/viewvc/perl/modperl/docs/trunk/src/about/pronounce.pod?rev=888069&r1=888068&r2=888069&view=diff
==============================================================================
--- perl/modperl/docs/trunk/src/about/pronounce.pod (original)
+++ perl/modperl/docs/trunk/src/about/pronounce.pod Mon Dec  7 18:34:32 2009
@@ -79,7 +79,6 @@
        <td class="table-bg">
                <a href="pronounce/pereinar.ogg">Listen</a>
        </td>
-
 </tr>
 </table>




---------------------------------------------------------------------
To unsubscribe, e-mail: docs-cvs-unsubscribe@perl.apache.org
For additional commands, e-mail: docs-cvs-help@perl.apache.org

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On 11/23/2009 11:02 PM, dorian taylor wrote:
> Hey guys, been a while.
>
> Was wondering about the behaviour of Apache2::Reload in modules that
> use warnings FATAL =>  'all'; it croaks with a Subroutine foo redefined
> error which of course is remedied by no warnings 'redefine' in the
> offending module. Is there a way to incorporate that into
> Apache2::Reload itself or is the scope too narrow?
>
> Regards,
>    
Set the unload method as so:

   $ModPerl::Util::DEFAULT_UNLOAD_METHOD = 'unload_package_xs';

and you will be good to go.  I am running with a prefork mpm and use
this to aid with thread, i.e., process, reuse issues:

<IfModule prefork.c>
     StartServers       1
     MinSpareServers    1
     MaxSpareServers    1
     ServerLimit        1
     MaxClients         1
     MaxRequestsPerChild  4000
</IfModule>

The above pragma will also eliminate 'Prototype mismatch' warnings.

Cheers,
-Ryan

OT, would this be better posted to modperl@perl.apache.org?

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hey guys, been a while.

Was wondering about the behaviour of Apache2::Reload in modules that
use warnings FATAL => 'all'; it croaks with a Subroutine foo redefined
error which of course is remedied by no warnings 'redefine' in the
offending module. Is there a way to incorporate that into
Apache2::Reload itself or is the scope too narrow?

Regards,

-- 
Dorian Taylor
http://doriantaylor.com/

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On 09-11-19 23:09 , Adam Prime wrote:
> Can someone change this ticket to resolved (or whatever is appropriate)
> 
> https://rt.cpan.org/Ticket/Display.html?id=38084
> 
> I can't.

Done. Thanks!

-- 
Philippe M. Chiasson     GPG: F9BFE0C2480E7680 1AE53631CB32A107 88C3A5A5
http://gozer.ectoplasm.org/       m/gozer\@(apache|cpan|ectoplasm)\.org/

Can someone change this ticket to resolved (or whatever is appropriate)

https://rt.cpan.org/Ticket/Display.html?id=38084

I can't.

Adam

aprime@apache.org wrote:
> Author: aprime
> Date: Fri Nov 20 04:05:37 2009
> New Revision: 882407
> 
> URL: http://svn.apache.org/viewvc?rev=882407&view=rev
> Log:
> Fixed issue with solaris version numbers being treated as decimal numbers, not
> dotted tuples.
> Fixes RT #38084.
> 
> Modified:
>     perl/Apache-SizeLimit/trunk/lib/Apache/SizeLimit/Core.pm
> 
> Modified: perl/Apache-SizeLimit/trunk/lib/Apache/SizeLimit/Core.pm
> URL: http://svn.apache.org/viewvc/perl/Apache-SizeLimit/trunk/lib/Apache/SizeLimit/Core.pm?rev=882407&r1=882406&r2=882407&view=diff
> ==============================================================================
> --- perl/Apache-SizeLimit/trunk/lib/Apache/SizeLimit/Core.pm (original)
> +++ perl/Apache-SizeLimit/trunk/lib/Apache/SizeLimit/Core.pm Fri Nov 20 04:05:37 2009
> @@ -131,7 +131,9 @@
>  }
>  
>  BEGIN {
> -    if ($Config{'osname'} eq 'solaris' && $Config{'osvers'} >= 2.6 ) {
> +    my ($major,$minor) = split(/\./, $Config{'osvers'});
> +    if ($Config{'osname'} eq 'solaris' && 
> +        (($major > 2) || ($major == 2 && $minor >= 6))) {
>          *_platform_check_size   = \&_solaris_2_6_size_check;
>          *_platform_getppid = \&_perl_getppid;
>      }
> 
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Fred Moyer wrote:
> On Mon, Nov 16, 2009 at 8:25 AM, Fred Moyer <fred@redhotpenguin.com> wrote:
>> On Mon, Nov 16, 2009 at 8:07 AM, Adam Prime <adam.prime@utoronto.ca> wrote:
>>> Any comments?
>> Looks pretty safe.  +1
> 
> That being said, I think this line:
> 
> +        ($major > 2 || ($major == 2 && $minor >= 6))) {
> 
> is better written as
> 
> (($major > 2) or (($major == 2) && ($minor >= 6))) {
> 
> I can't remember the style guide off the top of my head, I have it
> plugged into my mod_perl perltidy instance.
> 

Sure.

Adam

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Mon, Nov 16, 2009 at 8:25 AM, Fred Moyer <fred@redhotpenguin.com> wrote:
> On Mon, Nov 16, 2009 at 8:07 AM, Adam Prime <adam.prime@utoronto.ca> wrote:
>> Any comments?
>
> Looks pretty safe.  +1

That being said, I think this line:

+        ($major > 2 || ($major == 2 && $minor >= 6))) {

is better written as

(($major > 2) or (($major == 2) && ($minor >= 6))) {

I can't remember the style guide off the top of my head, I have it
plugged into my mod_perl perltidy instance.

>
>>
>> Petr Sumbera wrote:
>>>
>>> Fred Moyer wrote:
>>>>
>>>> On Wed, Nov 11, 2009 at 7:56 PM, Adam Prime <adam.prime@utoronto.ca>
>>>> wrote:
>>>>>
>>>>> Petr Sumbera wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> can someone please reevaluate and commit attached patch? I would
love
>>>>>> to
>>>>>> see
>>>>>> it in mod_perl 2.0.5 and 1.32 when they are released..
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Petr
>>>>>>
>>>>>> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff
>>>>>
>>>>> If no one objects in the next few day or so i'll commit this (finally)
>>>>
>>>> Can you post the patch inline to this email so everyone can review it?
>>>
>>> Resending (this time directly).
>>>
>>> Thanks,
>>>
>>> Petr
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
>>> For additional commands, e-mail: dev-help@perl.apache.org
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
>> For additional commands, e-mail: dev-help@perl.apache.org
>>
>>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Mon, Nov 16, 2009 at 8:07 AM, Adam Prime <adam.prime@utoronto.ca> wrote:
> Any comments?

Looks pretty safe.  +1

>
> Petr Sumbera wrote:
>>
>> Fred Moyer wrote:
>>>
>>> On Wed, Nov 11, 2009 at 7:56 PM, Adam Prime <adam.prime@utoronto.ca>
>>> wrote:
>>>>
>>>> Petr Sumbera wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> can someone please reevaluate and commit attached patch? I would love
>>>>> to
>>>>> see
>>>>> it in mod_perl 2.0.5 and 1.32 when they are released..
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Petr
>>>>>
>>>>> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff
>>>>
>>>> If no one objects in the next few day or so i'll commit this (finally)
>>>
>>> Can you post the patch inline to this email so everyone can review it?
>>
>> Resending (this time directly).
>>
>> Thanks,
>>
>> Petr
>>
>>
>> ------------------------------------------------------------------------
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
>> For additional commands, e-mail: dev-help@perl.apache.org
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Any comments?

Petr Sumbera wrote:
> Fred Moyer wrote:
>> On Wed, Nov 11, 2009 at 7:56 PM, Adam Prime <adam.prime@utoronto.ca> 
>> wrote:
>>> Petr Sumbera wrote:
>>>> Hi,
>>>>
>>>> can someone please reevaluate and commit attached patch? I would 
>>>> love to
>>>> see
>>>> it in mod_perl 2.0.5 and 1.32 when they are released..
>>>>
>>>> Thanks,
>>>>
>>>> Petr
>>>>
>>>> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff
>>> If no one objects in the next few day or so i'll commit this (finally)
>>
>> Can you post the patch inline to this email so everyone can review it?
> 
> Resending (this time directly).
> 
> Thanks,
> 
> Petr
> 
> 
> ------------------------------------------------------------------------
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Content-type: text/plain; CHARSET=US-ASCII; format=flowed
Content-transfer-encoding: 7BIT

Fred Moyer wrote:
> On Wed, Nov 11, 2009 at 7:56 PM, Adam Prime <adam.prime@utoronto.ca> wrote:
>> Petr Sumbera wrote:
>>> Hi,
>>>
>>> can someone please reevaluate and commit attached patch? I would love to
>>> see
>>> it in mod_perl 2.0.5 and 1.32 when they are released..
>>>
>>> Thanks,
>>>
>>> Petr
>>>
>>> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff
>> If no one objects in the next few day or so i'll commit this (finally)
> 
> Can you post the patch inline to this email so everyone can review it?

Resending (this time directly).

Thanks,

Petr

On Wed, Nov 11, 2009 at 7:56 PM, Adam Prime <adam.prime@utoronto.ca> wrote:
> Petr Sumbera wrote:
>>
>> Hi,
>>
>> can someone please reevaluate and commit attached patch? I would love to
>> see
>> it in mod_perl 2.0.5 and 1.32 when they are released..
>>
>> Thanks,
>>
>> Petr
>>
>> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff
>
> If no one objects in the next few day or so i'll commit this (finally)

Can you post the patch inline to this email so everyone can review it?

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Petr Sumbera wrote:
> Hi,
> 
> can someone please reevaluate and commit attached patch? I would love to see
> it in mod_perl 2.0.5 and 1.32 when they are released..
> 
> Thanks,
> 
> Petr
> 
> http://old.nabble.com/file/p26284847/solaris.diff solaris.diff 

If no one objects in the next few day or so i'll commit this (finally)

Adam

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hi,

can someone please reevaluate and commit attached patch? I would love to see
it in mod_perl 2.0.5 and 1.32 when they are released..

Thanks,

Petr

http://old.nabble.com/file/p26284847/solaris.diff solaris.diff 
-- 
View this message in context: http://old.nabble.com/Apache%3A%3ASizeLimit-%2B-solaris-Update-10-tp22367060p26284847.html
Sent from the mod_perl - Dev mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Petr Sumbera wrote:
> What is status of this issue? I don't see any related change in svn.
> 
> Thanks,
> 
> Petr

No one every replied to that last question, and nothing has happened 
since then by the looks of things.

Adam

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

What is status of this issue? I don't see any related change in svn.

Thanks,

Petr


Adam Prime wrote:
> 
> Fred Moyer wrote:
>> On Wed, Mar 18, 2009 at 7:14 PM, Adam Prime <adam.prime@utoronto.ca>
>> wrote:
>>> Fred suggested in resend these patches relative to the root of the
>>> distributions, so here they are.  SizeLimit.patch is against the version
>>> of Apache2::SizeLimit in mod_perl 2.0.4 and generated via git diff.
>>> svn.SizeLimit.patch is against the version of SizeLimit in svn, and was
>>> generated with svn diff.
>> 
>> This git diff doesn't match what I have in svn trunk for
>> Apache2/SizeLimit.pm, 
> 
> Thats because it's against the version in mod_perl 2.0.4, which is 
> nothing like the version that currently in svn.
> 
>> applied to your second part of the patch.  I'd like to see the
>> $major/$minor logic use a couple of more parentheses to spell out
>> exactly what it is doing without having to worry about whether the
>> conditional parser in my head is interpreting this conditional
>> correctly.
> 
> Do you want me to resubmit it like :
> 
>     (($major > 2) || ($major == 2 && $minor >= 6))
> 
> I'm guessing that's what you mean.
> 
> Adam
> 
> 
>> Index: lib/Apache/SizeLimit/Core.pm
>> ===================================================================
>> --- lib/Apache/SizeLimit/Core.pm        (revision 755867)
>> +++ lib/Apache/SizeLimit/Core.pm        (working copy)
>> @@ -131,7 +131,11 @@
>>  }
>> 
>>  BEGIN {
>> -    if ($Config{'osname'} eq 'solaris' && $Config{'osvers'} >= 2.6 )
{
>> +     my ($major,$minor) = split('.',$Config{'osvers'});
>> +
>> +     if ($Config{'osname'} eq 'solaris' &&
>> +         ($major > 2 || $major == 2 && $minor >= 6)) {
>> +
>>          *_platform_check_size   = \&_solaris_2_6_size_check;
>>          *_platform_getppid = \&_perl_getppid;
>> 
>> 
>> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Apache%3A%3ASizeLimit-%2B-solaris-Update-10-tp22367060p26008045.html
Sent from the mod_perl - Dev mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Fred Moyer wrote:
> I spoke too soon - kharma needed for this one also.
> 
> Pgollucci, do you have the keys to the list controls?

yes ping me monday directly so I don't forget.
-- 
------------------------------------------------------------------------
1024D/DB9B8C1C B90B FBC3 A3A1 C71A 8E70  3F8C 75B8 8FFB DB9B 8C1C
Philip M. Gollucci (pgollucci@p6m7g8.com) c: 703.336.9354
Consultant          - P6M7G8 Inc.                http://p6m7g8.net
Senior Sys Admin    - RideCharge, Inc.           http://ridecharge.com
ASF Member          - Apache Software Foundation http://apache.org
FreeBSD Committer   - FreeBSD Foundation         http://freebsd.org

Work like you don't need the money,
love like you'll never get hurt,
and dance like nobody's watching.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

I spoke too soon - kharma needed for this one also.

Pgollucci, do you have the keys to the list controls?


---------- Forwarded message ----------
From:  <support@afnic.fr>
Date: Fri, Sep 25, 2009 at 11:43 AM
Subject: <Ref611354> AFNIC / ACCUSE RECEPTION MAIL
To: fred@redhotpenguin.com


Bonjour,

Votre courriel "Fwd: Vacation reply" a bien été reçu par le support de
l'AFNIC et porte le numéro 611354. Il fera l'objet d'un traitement dès
que possible.

Dans l'intervalle, nous vous invitons à consulter notre FAQ et les
actualités des opérations à l'AFNIC :

http://www.afnic.fr/doc/faq
http://www.afnic.fr/operations

Cordialement,
----------
Hello,

AFNIC support has received your message "Fwd: Vacation reply". We will
handle it as soon as possible.

Regards.
>>>>>>>>>>>>>>>>>>>>>>>
Sujet : Fwd: Vacation reply
N° dossier : 611354
Corps du mail : Does anyone have the kharma to unsubscribe this guy?
Anyone posting
to the list gets one of this I think:


---------- Forwarded message ----------
From:  <granitegadde@hotmail.com>
Date: Fri, Sep 25, 2009 at 11:14 AM
Subject: Vacation reply
To: fred@redhotpenguin.com


Do not send me an email to this address granitegadde@hotmail.com. This
is an automated reply. I do not use this email
address any more regularly.
Srinivas Gadde.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org






Pensez-y : toute l'actualité des opérations de l'Afnic est sur
http://operations.afnic.fr/

--
Support Afnic
Tel: +33 1 39 30 83 00
Fax: +33 1 39 30 83 83
support@afnic.fr

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Does anyone have the kharma to unsubscribe this guy?  Anyone posting
to the list gets one of this I think:


---------- Forwarded message ----------
From:  <granitegadde@hotmail.com>
Date: Fri, Sep 25, 2009 at 11:14 AM
Subject: Vacation reply
To: fred@redhotpenguin.com


Do not send me an email to this address granitegadde@hotmail.com. This
is an automated reply. I do not use this email
address any more regularly.
Srinivas Gadde.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Does anyone have the kharma to remove this address from the user's list?


---------- Forwarded message ----------
From:  <granitegadde@hotmail.com>
Date: Wed, Sep 16, 2009 at 2:42 PM
Subject: Vacation reply
To: fred@redhotpenguin.com


Do not send me an email to this address granitegadde@hotmail.com. This
is an automated reply. I do not use this email
address any more regularly.
Srinivas Gadde.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Apache 2.2.13
Perl 5.10.1
Ubuntu Linux 2.6.24-19

Relevant Failures:

t/hooks/authen_basic.t ..................
1..4
# Running under perl version 5.010001 for linux
# Current time local: Tue Sep 15 16:45:32 2009
# Current time GMT:   Tue Sep 15 23:45:32 2009
# Using Test.pm version 1.25_02
# Using Apache/Test.pm version 1.31
ok 1
ok 2
ok 3
not ok 4
# Failed test 4 in t/hooks/authen_basic.t at line 26
Failed 1/4 subtests

t/hooks/authz.t .........................
1..4
# Running under perl version 5.010001 for linux
# Current time local: Tue Sep 15 16:45:33 2009
# Current time GMT:   Tue Sep 15 23:45:33 2009
# Using Test.pm version 1.25_02
# Using Apache/Test.pm version 1.31
ok 1
ok 2
ok 3
not ok 4
# Failed test 4 in t/hooks/authz.t at line 19
Failed 1/4 subtests

Error log and full test verbose output attached.

Best,

David

On Fri, Aug 14, 2009 at 06:19:29PM +0200, Jan Pazdziora wrote:
> 
> Having it in core mod_perl is definitelly also an option. 
> 
> That however will not bring the functionality to already-released 
> mod_perl 2 versions around the world. So my idea was to release this 
> module on CPAN and if it gets to mod_perl core eventually, even 
> better.

I have uploaded Apache2-RequestRec-Time-1.0.tar.gz to CPAN. It
provides request_duration_microseconds, request_duration, and
request_time_microseconds (I left out the _hires variants in the
end). I've sent announcement to modperl@perl.apache.org.

Yours,

-- 
Jan Pazdziora

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hello

I don't quite understand the interaction between Perl and sendmail.

If I send an email like this:
  open(MAIL,"|$mailprog -t -oi") or die "mail can't open ";
  print MAIL "To: $recipient\n";
.... other headers and body
close(MAIL);

will Perl have to wait for sendmail (the mail program) to return a value
after open() ? Are there any other delays here?

with thanks

Kropotkin


-- 
View this message in context: http://www.nabble.com/Sneding-mail-from-Perl-tp25342957p25342957.html
Sent from the mod_perl - Dev mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Fri, 28 Aug 2009 21:35:50 -0400
Philippe wrote:

> On 28/08/09 08:35 , Ryan Gies wrote:
> > This patch for Apache2::Reload does two things:
> > 
> >   A) ignores require-hooks which exist in %INC. (fix)
> 
> That's a very good fix indeed, and should be fixed by itself, yes.
> 
> >   B) reloads by file, not module name (fix?)
> 
> Good feature, good patch. I've got only one simple style nit with it.
> 
> Instead of:
> 
> +    foreach my $kvpair (@changed) {
> +        my $name = $ReloadByModuleName ? $kvpair->[0] : $kvpair->[1];
> +        require $name;
> 
> 
> I just find $kvpair and $kvpair->[0] somewhat hard to read, could you
> simply change it to something like:
> 
> +    foreach my $change (@changed) {
> +        my $module = $change->[0];
> +        my $file = $change->[1];
> +        my $name = $ReloadByModuleName ? $module: $file;
> +        require $name;

Agreed. I used this style presuming it's equally readable:

        my ($module, $file) = @$change;

> 
> Apart from that, it's a great patch!
> 
> Could you resend 2 patches (splitting A) and B) in different ones)
> with that small concern adressed ?
> 

No problem.  I created patch-B as a diff from patch-A, i.e., it's a 
cumulative patch.  I also bumped $VERSION once for each patch.  I
have only ran tests against the final v0.12 code.

I also introduced a new change in patch-B.  The debug output when an
entry is reloaded uses the format:

  Apache2::Reload: process %d reloading %s from %s\n

the last %s *was* populated with the file's full path.  It is *now*
populated with whatever value was passed to 'require'.  For example, if
you are using the old-style (ReloadByModuleName) method, it will look
like this:

  ... from Apache2/Trace.pm

and by default (new-style) it will look like:

  ... from /usr/local/src/lsn/perl/lib/Apache2/Trace.pm

Please let me know if you have any other changes, questions, or
comments! Cheers,
-Ryan

On 28/08/09 08:35 , Ryan Gies wrote:
> This patch for Apache2::Reload does two things:
> 
>   A) ignores require-hooks which exist in %INC. (fix)

That's a very good fix indeed, and should be fixed by itself, yes.

>   B) reloads by file, not module name (fix?)

Good feature, good patch. I've got only one simple style nit with it.

Instead of:

+    foreach my $kvpair (@changed) {
+        my $name = $ReloadByModuleName ? $kvpair->[0] : $kvpair->[1];
+        require $name;


I just find $kvpair and $kvpair->[0] somewhat hard to read, could you
simply change it to something like:

+    foreach my $change (@changed) {
+        my $module = $change->[0];
+        my $file = $change->[1];
+        my $name = $ReloadByModuleName ? $module: $file;
+        require $name;

Apart from that, it's a great patch!

Could you resend 2 patches (splitting A) and B) in different ones) with
that small concern adressed ?

-- 
Philippe M. Chiasson     GPG: F9BFE0C2480E7680 1AE53631CB32A107 88C3A5A5
http://gozer.ectoplasm.org/       m/gozer\@(apache|cpan|ectoplasm)\.org/

This patch for Apache2::Reload does two things:

  A) ignores require-hooks which exist in %INC. (fix)
  B) reloads by file, not module name (fix?)

I consider B) to be a fix because I cannot imagine why anyone would
want the previous functionality.  As such, the accompanying
configuration variable 'ReloadByModuleName' is used to enabled the old
behavior (as opposed to enabling the new behavior).

I was asked to include this patch inline in this email, however I know
the line-breaks will clobber the diff output, so I have attached it as
well.

Cheers,
-Ryan

--- Apache2/Reload.pm	2009-08-12 10:35:04.000000000 -0400
+++ Apache2/Reload.pm.v0.11	2009-08-28 08:28:25.000000000 -0400
@@ -20,7 +20,7 @@
 
 use mod_perl2;
 
-our $VERSION = '0.10';
+our $VERSION = '0.11';
 
 use Apache2::Const -compile => qw(OK);
 
@@ -86,6 +86,8 @@
 
     my $DEBUG = ref($o) && (lc($o->dir_config("ReloadDebug") || '') eq
'on'); 
+    my $ReloadByModuleName = ref($o) &&
(lc($o->dir_config("ReloadByModuleName") || '') eq 'on'); +
     my $TouchFile = ref($o) && $o->dir_config("ReloadTouchFile");
 
     my $ConstantRedefineWarnings = ref($o) && 
@@ -138,6 +140,9 @@
         my $file = $Apache2::Reload::INCS{$key};
 
         next unless defined $file;
+        # TODO when $file is a hook, call the hook like require does
to 
+        # determine the path (obtain a file handle). See %INC in
perlvar.
+        next if ref $file;
         next if @watch_dirs && !grep { $file =~ /^$_/ } @watch_dirs;
         warn "Apache2::Reload: Checking mtime of $key\n" if $DEBUG;
 
@@ -158,24 +163,27 @@
         }
 
         if ($mtime > $Stat{$file}) {
-            push @changed, $key;
+            push @changed, [$key, $file];
         }
         $Stat{$file} = $mtime;
     }
     
     #First, let's unload all changed modules
-    foreach my $module (@changed) {
-        my $package = module_to_package($module);
+    foreach my $kvpair (@changed) {
+        my $package = module_to_package($kvpair->[0]);
         ModPerl::Util::unload_package($package);
     }
-    
-    #Then, let's reload them all, so that module dependencies can
satisfy
-    #themselves in the correct order.
-    foreach my $module (@changed) {
-        my $package = module_to_package($module);
-        require $module;
-        warn("Apache2::Reload: process $$ reloading $package from
$module\n")
-            if $DEBUG;
+
+    #Then, let's reload each changed file, so that module dependencies
can 
+    #satisfy themselves in the correct order.
+    foreach my $kvpair (@changed) {
+        my $name = $ReloadByModuleName ? $kvpair->[0] : $kvpair->[1];
+        require $name;
+        if ($DEBUG) {
+          my $package = module_to_package($kvpair->[0]);
+          warn sprintf("Apache2::Reload: process %d reloading %s from
%s\n",
+            $$, $package, $kvpair->[1]);
+        }
     }
 
     return Apache2::Const::OK;
@@ -206,6 +214,7 @@
   PerlSetVar ReloadAll Off
   PerlSetVar ReloadModules "ModPerl::* Apache2::*"
   #PerlSetVar ReloadDebug On
+  #PerlSetVar ReloadByModuleName On
   
   # Reload a single module from within itself:
   package My::Apache2::Module;
@@ -226,16 +235,28 @@
 also do the check for modified modules, when a special touch-file has
 been modified.
 
-Note that C<Apache2::Reload> operates on the current context of
-C<@INC>.  Which means, when called as a C<Perl*Handler> it will not
-see C<@INC> paths added or removed by C<ModPerl::Registry> scripts, as
-the value of C<@INC> is saved on server startup and restored to that
-value after each request.  In other words, if you want
-C<Apache2::Reload> to work with modules that live in custom C<@INC>
-paths, you should modify C<@INC> when the server is started.  Besides,
-C<'use lib'> in the startup script, you can also set the C<PERL5LIB>
-variable in the httpd's environment to include any non-standard 'lib'
-directories that you choose.  For example, to accomplish that you can
+Require-hooks, i.e., entries in %INC which are references, are
ignored.  The +hook should modify %INC itself, adding the path to the
module file, for it to +be reloaded.
+
+C<Apache2::Reload> inspects and reloads the B<file> associated with a
given +module.  Changes to @INC are not recognized, as it is the file
which is +being re-required, not the module name.
+
+In version 0.10 and earlier the B<module name>, not the file, is
re-required. +Meaning it operated on the the current context of @INC.
If you still want this +behavior set this environment variable in
I<httpd.conf>: +
+  PerlSetVar ReloadByModuleName On
+
+This means, when called as a C<Perl*Handler>, C<Apache2::Reload> will
not see +C<@INC> paths added or removed by C<ModPerl::Registry>
scripts, as the value of +C<@INC> is saved on server startup and
restored to that value after each +request.  In other words, if you
want C<Apache2::Reload> to work with modules +that live in custom
C<@INC> paths, you should modify C<@INC> when the server is +started.
Besides, C<'use lib'> in the startup script, you can also set the
+C<PERL5LIB> variable in the httpd's environment to include any
non-standard +'lib' directories that you choose.  For example, to
accomplish that you can include a line: 
   PERL5LIB=/home/httpd/perl/extra; export PERL5LIB

On 24/08/09 15:54 , Fred Moyer wrote:
> On Mon, Aug 24, 2009 at 11:31 AM, Torsten
> Foertsch<torsten.foertsch@gmx.net> wrote:
>> On Mon 24 Aug 2009, Philippe M. Chiasson wrote:
>>>> Modified: perl/modperl/trunk/lib/Apache2/Build.pm
>>>> URL:
>>>> http://svn.apache.org/viewvc/perl/modperl/trunk/lib/Apache2/Build.p
>>>> m?rev=807116&r1=807115&r2=807116&view=diff
>>>> ===================================================================
>>>> =========== --- perl/modperl/trunk/lib/Apache2/Build.pm (original)
>>>> +++ perl/modperl/trunk/lib/Apache2/Build.pm Mon Aug 24 08:29:43
>>>> 2009 @@ -2068,6 +2068,7 @@
>>>>  }
>>>>
>>>>  sub inc {
>>>> +    local $_;
>>>>      my @includes = map { "-I$_" } @{ shift->includes };
>>>>      "@includes";
>>>>  }
>>>
>>> With the rest of the changes in this patch, is this particular change
>>> still necessary?
>>
>> Not really, but it's a safety measure. The "for ()" idiom is used in
>> many places. And I know these pieces of code that had bitten me now are
>> years old. Don't know why it happened now. So, I'd like to have it
>> there.
> 
> What about replacing the for () instances with 'for my'?  'local $_'
> may affect other parts of the code that haven't been looked at -
> that's my concern with that approach.

Well said, same concern here.

-- 
Philippe M. Chiasson     GPG: F9BFE0C2480E7680 1AE53631CB32A107 88C3A5A5
http://gozer.ectoplasm.org/       m/gozer\@(apache|cpan|ectoplasm)\.org/

On Mon, Aug 24, 2009 at 11:31 AM, Torsten
Foertsch<torsten.foertsch@gmx.net> wrote:
> On Mon 24 Aug 2009, Philippe M. Chiasson wrote:
>> > Modified: perl/modperl/trunk/lib/Apache2/Build.pm
>> > URL:
>> > http://svn.apache.org/viewvc/perl/modperl/trunk/lib/Apache2/Build.p
>> >m?rev=807116&r1=807115&r2=807116&view=diff
>> > ===================================================================
>> >=========== --- perl/modperl/trunk/lib/Apache2/Build.pm (original)
>> > +++ perl/modperl/trunk/lib/Apache2/Build.pm Mon Aug 24 08:29:43
>> > 2009 @@ -2068,6 +2068,7 @@
>> >  }
>> >
>> >  sub inc {
>> > +    local $_;
>> >      my @includes = map { "-I$_" } @{ shift->includes };
>> >      "@includes";
>> >  }
>>
>> With the rest of the changes in this patch, is this particular change
>> still necessary?
>
> Not really, but it's a safety measure. The "for ()" idiom is used in
> many places. And I know these pieces of code that had bitten me now are
> years old. Don't know why it happened now. So, I'd like to have it
> there.

What about replacing the for () instances with 'for my'?  'local $_'
may affect other parts of the code that haven't been looked at -
that's my concern with that approach.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Mon 24 Aug 2009, Philippe M. Chiasson wrote:
> > Modified: perl/modperl/trunk/lib/Apache2/Build.pm
> > URL:
> > http://svn.apache.org/viewvc/perl/modperl/trunk/lib/Apache2/Build.p
> >m?rev=807116&r1=807115&r2=807116&view=diff
> > ===================================================================
> >=========== --- perl/modperl/trunk/lib/Apache2/Build.pm (original)
> > +++ perl/modperl/trunk/lib/Apache2/Build.pm Mon Aug 24 08:29:43
> > 2009 @@ -2068,6 +2068,7 @@
> >  }
> >  
> >  sub inc {
> > +    local $_;
> >      my @includes = map { "-I$_" } @{ shift->includes };
> >      "@includes";
> >  }
>
> With the rest of the changes in this patch, is this particular change
> still necessary?

Not really, but it's a safety measure. The "for ()" idiom is used in 
many places. And I know these pieces of code that had bitten me now are 
years old. Don't know why it happened now. So, I'd like to have it 
there.

Torsten

-- 
Need professional mod_perl support?
Just hire me: torsten.foertsch@gmx.net

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On 24/08/09 04:29 , torsten@apache.org wrote:
> Author: torsten
> Date: Mon Aug 24 08:29:43 2009
> New Revision: 807116
> 
> URL: http://svn.apache.org/viewvc?rev=807116&view=rev
> Log:
> Fix a typo in ModPerl::BuildMM. Fix a compile time issue by introducing
> lexically scoped loop variables in ModPerl::MM::WriteMakefile().

Great, thanks for the fixes!

> Modified:
>     perl/modperl/trunk/Changes
>     perl/modperl/trunk/lib/Apache2/Build.pm
>     perl/modperl/trunk/lib/ModPerl/BuildMM.pm
>     perl/modperl/trunk/lib/ModPerl/MM.pm
> 
> Modified: perl/modperl/trunk/Changes
> URL: http://svn.apache.org/viewvc/perl/modperl/trunk/Changes?rev=807116&r1=807115&r2=807116&view=diff
> ==============================================================================
> --- perl/modperl/trunk/Changes (original)
> +++ perl/modperl/trunk/Changes Mon Aug 24 08:29:43 2009
> @@ -12,6 +12,9 @@
>  
>  =item 2.0.5-dev
>  
> +Fix a typo in ModPerl::BuildMM and introduce lexically scoped loop
> +variables in ModPerl::MM::WriteMakefile(). [Torsten Foertsch]

I'd reword that to be a bit more descriptive to the folks likely to be reading
this to determine if they should update. Something like:

"MODPERL_INC= now correctly supported as an argument to Makefile.PL"

The lexically scoped loop variables should have no impact at all to a potential
upgrader, so it doesn't really need to be mentionned.

>  Fix an XSS issue in Apache2::Status reported by Richard J. Brain
>  <richard@procheckup.com>. [Torsten Foertsch]
>  
> 
> Modified: perl/modperl/trunk/lib/Apache2/Build.pm
> URL: http://svn.apache.org/viewvc/perl/modperl/trunk/lib/Apache2/Build.pm?rev=807116&r1=807115&r2=807116&view=diff
> ==============================================================================
> --- perl/modperl/trunk/lib/Apache2/Build.pm (original)
> +++ perl/modperl/trunk/lib/Apache2/Build.pm Mon Aug 24 08:29:43 2009
> @@ -2068,6 +2068,7 @@
>  }
>  
>  sub inc {
> +    local $_;
>      my @includes = map { "-I$_" } @{ shift->includes };
>      "@includes";
>  }

With the rest of the changes in this patch, is this particular change still
necessary?

-- 
Philippe M. Chiasson     GPG: F9BFE0C2480E7680 1AE53631CB32A107 88C3A5A5
http://gozer.ectoplasm.org/       m/gozer\@(apache|cpan|ectoplasm)\.org/

On Sun, Aug 23, 2009 at 9:03 AM, Torsten
Foertsch<torsten.foertsch@gmx.net> wrote:
> Hi,
>
> recently I stumbled upon a curious thing. A freshly unpacked modperl
> aborted compilation after compiling in src/modules/perl. These are the
> symptoms:

....


+1 to this patch.  Has the added benefit of using a lexical loop
variable which is a better practice than 'for ... $_' in my humble
opinion.

> This patch introduces $o as loop variable instead of $_. For me the
> @default_opts loop was the culprit but I have changed all loops in
> WriteMakefile().
>
> Index: lib/ModPerl/MM.pm
> ===================================================================
> --- lib/ModPerl/MM.pm   (revision 806134)
> +++ lib/ModPerl/MM.pm   (working copy)
> @@ -132,22 +132,22 @@
>     my_import(__PACKAGE__);
>
>     # set top-level WriteMakefile's values if weren't set already
> -    for (@default_opts) {
> -        $args{$_} = get_def_opt($_) unless exists $args{$_}; # already defined
> +    for my $o (@default_opts) {
> +        $args{$o} = get_def_opt($o) unless exists $args{$o}; # already defined
>     }
>
>     # set dynamic_lib-level WriteMakefile's values if weren't set already
>     $args{dynamic_lib} ||= {};
>     my $dlib = $args{dynamic_lib};
> -    for (@default_dlib_opts) {
> -        $dlib->{$_} = get_def_opt($_) unless exists $dlib->{$_};
> +    for my $o (@default_dlib_opts) {
> +        $dlib->{$o} = get_def_opt($o) unless exists $dlib->{$o};
>     }
>
>     # set macro-level WriteMakefile's values if weren't set already
>     $args{macro} ||= {};
>     my $macro = $args{macro};
> -    for (@default_macro_opts) {
> -        $macro->{$_} = get_def_opt($_) unless exists $macro->{$_};
> +    for my $o (@default_macro_opts) {
> +        $macro->{$o} = get_def_opt($o) unless exists $macro->{$o};
>     }
>
>     ExtUtils::MakeMaker::WriteMakefile(%args);

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hi,

recently I stumbled upon a curious thing. A freshly unpacked modperl 
aborted compilation after compiling in src/modules/perl. These are the 
symptoms:

- during perl Makefile.PL:

...
WARNING: #define AP_SERVER_MAJORVERSION_NUMBER 2
 is not a known parameter.
Checking if your kit is complete...
Looks good
Writing Makefile for Apache2::Reload
!!! no default argument defined for argument: #define 
AP_SERVER_MAJORVERSION_NUMBER 2
 at ./Makefile.PL line 60
WARNING: #define AP_SERVER_MAJORVERSION_NUMBER 2
 is not a known parameter.
Checking if your kit is complete...
Looks good

- and then in make:

...
cp lib/Apache2/porting.pm blib/lib/Apache2/porting.pm
make[1]: Entering directory 
`/usr/src/packages/BUILD/mod_perl-2.0.5threading3/Apache-Reload'
Makefile:14: *** empty variable name.  Stop.
make[1]: Leaving directory 
`/usr/src/packages/BUILD/mod_perl-2.0.5threading3/Apache-Reload'
make: *** [subdirs] Error 2

A second call of "perl Makefile.PL; make" made it work normally.

I could track it down to $_ being used as loop variable and overwritten 
by a function called in that loop in ModPerl::MM.

There are obviously 2 ways to fix that. First, introduce a local loop 
variable and second, ensure $_ is not changed inside a function unless 
this is the purpose of the function.

What is the way to go? Or are both ways ok?

This patch introduces $o as loop variable instead of $_. For me the
@default_opts loop was the culprit but I have changed all loops in
WriteMakefile().

Index: lib/ModPerl/MM.pm
===================================================================
--- lib/ModPerl/MM.pm	(revision 806134)
+++ lib/ModPerl/MM.pm	(working copy)
@@ -132,22 +132,22 @@
     my_import(__PACKAGE__);
 
     # set top-level WriteMakefile's values if weren't set already
-    for (@default_opts) {
-        $args{$_} = get_def_opt($_) unless exists $args{$_}; # already defined
+    for my $o (@default_opts) {
+        $args{$o} = get_def_opt($o) unless exists $args{$o}; # already defined
     }
 
     # set dynamic_lib-level WriteMakefile's values if weren't set already
     $args{dynamic_lib} ||= {};
     my $dlib = $args{dynamic_lib};
-    for (@default_dlib_opts) {
-        $dlib->{$_} = get_def_opt($_) unless exists $dlib->{$_};
+    for my $o (@default_dlib_opts) {
+        $dlib->{$o} = get_def_opt($o) unless exists $dlib->{$o};
     }
 
     # set macro-level WriteMakefile's values if weren't set already
     $args{macro} ||= {};
     my $macro = $args{macro};
-    for (@default_macro_opts) {
-        $macro->{$_} = get_def_opt($_) unless exists $macro->{$_};
+    for my $o (@default_macro_opts) {
+        $macro->{$o} = get_def_opt($o) unless exists $macro->{$o};
     }
 
     ExtUtils::MakeMaker::WriteMakefile(%args);

And this one prevents inc() from changing $_:

Index: lib/Apache2/Build.pm
===================================================================
--- lib/Apache2/Build.pm	(revision 806134)
+++ lib/Apache2/Build.pm	(working copy)
@@ -2068,6 +2068,7 @@
 }
 
 sub inc {
+    local $_;
     my @includes = map { "-I$_" } @{ shift->includes };
     "@includes";
 }


The attached patch contains these 2 plus a fix for the typo mentioned
in my previous mail.

Can I apply it to trunk?

Torsten

Hi,

in lib/ModPerl/BuildMM.pm:WriteMakefile() I found these lines (80-83):

    my $inc;
    $inc = $args{INC} if $args{INC};
    $inc = " " . $build->inc;

Shouldn't the last one read instead:

    $inc .= " " . $build->inc;

Torsten

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Fri 21 Aug 2009, Torsten Foertsch wrote:
> I think the whole reconfig thing should be skipped
> with -run-tests.

How about the enclosed patch? I don't know enough about AT to judge. So, 
please comment.

Torsten

-- 
Need professional mod_perl support?
Just hire me: torsten.foertsch@gmx.net

Hi,

I thought I had found an issue in t/api/add_config.t but the problem is 
probably AT.

t/api/add_config.t creates a vhost. If I start httpd 
with "t/TEST -start-httpd" a new configuration is created in 
t/conf/apache_test_config.pm and the shows the correct vhost=>port 
mapping:

  'TestAPI::add_config' => {
                            'hostport' => 'localhost:8567',
                            ...

httpd runs and I can connect to localhost:8567 and run the test via 
curl.

Now, if I try to run the test via "t/TEST -verbose -run-tests 
t/api/add_config.t" it tries to connect to a completely different port:

# connecting to http://localhost:8607/TestAPI__add_config/

The point is, t/TEST overwrites t/conf/apache_test_config.pm or rather 
reconfigures the system. But now the default ports at 8529 upwards are 
in use. So, it assigns new ones. This must be prevented in 
the -run-tests case. I think the whole reconfig thing should be skipped 
with -run-tests.

Does anyone know where to lay hands on or do I have to figure it out 
myself?

Torsten

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hi all,

I was directed here after asking around regarding the following. I hope this
is the right place.

I'm trying to use Apache2::HookRun according to the example provided in
http://perl.apache.org/docs/2.0/api/Apache2/HookRun.html

Two problems so far:

   1. Apache2::RequestRec doesn't appear to provide a new() method, so the
   example fails.
   2. A subsequent "$r->location_merge($path);" seems, according to the
   comments, to be required or bad things will happen -- but $path isn't
   declared or defined, and location_merge() seems not to be a method for
   Apache2::RequestRec either.

Okay, so the example code seems a bit broken. Does anyone know how it should
be fixed/rewritten? Thanks.

(In the greater scheme of things, I'm trying to get mod_perl to dryrun the
authn/authz phases with config, handlers, .htaccess and all, and provide the
results "as a service". HookRun seemed like an ideal way to implement that,
but if there are better ways of getting there, I'd also appreciate any
pointers.)

Best,
Thomas

I'm going to be on IRC this weekend, #modperl on irc.perl.org.  Let's
get 2.0.5 under way!

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Thu, Aug 13, 2009 at 12:46:50PM -0700, Fred Moyer wrote:
> On Thu, Aug 13, 2009 at 12:21 PM, Jan Pazdziora<jpz-mlmp@adelton.com> wrote:
> > related to my previous post about getting to microsecond values
> > from APR/mod_perl, I propose to submit to CPAN a module
> > Apache2::RequestRec::Time which will add four functions (methods)
> > to Apache2::RequestRec: request_time_hires, request_duration,
> > request_duration_microseconds, and request_duration_hires.
> 
> I can't speak for the whole dev team, but I don't see any reason
> offhand why these methods couldn't be included in the core since they
> are RequestRec based.  This code is based off of the C api so it makes
> a bit more sense to me to have it in core than as an external module.

Having it in core mod_perl is definitelly also an option. 

That however will not bring the functionality to already-released 
mod_perl 2 versions around the world. So my idea was to release this 
module on CPAN and if it gets to mod_perl core eventually, even 
better.

-- 
Jan Pazdziora

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

On Thu, Aug 13, 2009 at 12:21 PM, Jan Pazdziora<jpz-mlmp@adelton.com> wrote:
> related to my previous post about getting to microsecond values
> from APR/mod_perl, I propose to submit to CPAN a module
> Apache2::RequestRec::Time which will add four functions (methods)
> to Apache2::RequestRec: request_time_hires, request_duration,
> request_duration_microseconds, and request_duration_hires.

I can't speak for the whole dev team, but I don't see any reason
offhand why these methods couldn't be included in the core since they
are RequestRec based.  This code is based off of the C api so it makes
a bit more sense to me to have it in core than as an external module.

>From my limited understanding of the mod_perl XS api this looks like
great start!

>
> My main goal is to be able to get value equivalent to custom
> log's %D (time taken to serve the request, in microseconds) from
> mod_perl, which request_duration_microseconds will do, and I wanted
> the set of methods to be a big more complete in case people wanted
> similar yet not exact functionality. Initially, I also planned to
> have request_time_microseconds but I hit integer overflows as epoch
> in microseconds is fairly big number.
>
> I'd appreciate any comments about suitability of such module and/or
> its name, purpose, or implementation (for code, see below) before
> I do the CPAN submission. Of course, the CPAN distribution will be
> complete with Makefile.PL and test.
>
> Thank you.
>
> For the reference, my planned XS code is:
>
> #include <mod_perl.h>
>
> typedef request_rec *Apache2__RequestRec;
>
> MODULE = Apache2::RequestRec::Time      PACKAGE = Apache2::RequestRec   PREFIX = mpxs_Apache2__RequestRec_
>
> double
> mpxs_Apache2__RequestRec_request_time_hires(r)
>                Apache2::RequestRec r
>        CODE:
>                RETVAL = (double)(r->request_time) / APR_USEC_PER_SEC;
>        OUTPUT:
>                RETVAL
>
> long
> mpxs_Apache2__RequestRec_request_duration(r)
>                Apache2::RequestRec r
>        CODE:
>                apr_time_t duration = apr_time_now() - r->request_time;
>                RETVAL = apr_time_sec(duration);
>        OUTPUT:
>                RETVAL
>
> double
> mpxs_Apache2__RequestRec_request_duration_microseconds(r)
>                Apache2::RequestRec r
>        CODE:
>                RETVAL = (double)(apr_time_now() - r->request_time);
>        OUTPUT:
>                RETVAL
>
> double
> mpxs_Apache2__RequestRec_request_duration_hires(r)
>                Apache2::RequestRec r
>        CODE:
>                RETVAL = (double)(apr_time_now() - r->request_time) / APR_USEC_PER_SEC;
>        OUTPUT:
>                RETVAL
>
> --
> Jan Pazdziora
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org

Hello,

related to my previous post about getting to microsecond values
from APR/mod_perl, I propose to submit to CPAN a module
Apache2::RequestRec::Time which will add four functions (methods)
to Apache2::RequestRec: request_time_hires, request_duration,
request_duration_microseconds, and request_duration_hires.

My main goal is to be able to get value equivalent to custom
log's %D (time taken to serve the request, in microseconds) from
mod_perl, which request_duration_microseconds will do, and I wanted
the set of methods to be a big more complete in case people wanted
similar yet not exact functionality. Initially, I also planned to
have request_time_microseconds but I hit integer overflows as epoch
in microseconds is fairly big number.

I'd appreciate any comments about suitability of such module and/or
its name, purpose, or implementation (for code, see below) before
I do the CPAN submission. Of course, the CPAN distribution will be
complete with Makefile.PL and test.

Thank you.

For the reference, my planned XS code is:

#include <mod_perl.h>

typedef request_rec *Apache2__RequestRec;

MODULE = Apache2::RequestRec::Time      PACKAGE = Apache2::RequestRec   PREFIX = mpxs_Apache2__RequestRec_

double
mpxs_Apache2__RequestRec_request_time_hires(r)
                Apache2::RequestRec r
        CODE:
                RETVAL = (double)(r->request_time) / APR_USEC_PER_SEC;
        OUTPUT:
                RETVAL

long
mpxs_Apache2__RequestRec_request_duration(r)
                Apache2::RequestRec r
        CODE:
                apr_time_t duration = apr_time_now() - r->request_time;
                RETVAL = apr_time_sec(duration);
        OUTPUT:
                RETVAL

double
mpxs_Apache2__RequestRec_request_duration_microseconds(r)
                Apache2::RequestRec r
        CODE:
                RETVAL = (double)(apr_time_now() - r->request_time);
        OUTPUT:
                RETVAL

double
mpxs_Apache2__RequestRec_request_duration_hires(r)
                Apache2::RequestRec r
        CODE:
                RETVAL = (double)(apr_time_now() - r->request_time) / APR_USEC_PER_SEC;
        OUTPUT:
                RETVAL

-- 
Jan Pazdziora

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org