Mailing List Archive: Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02)

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02)

Jul 16, 2008, 3:39 PM

Post #1 of 9 (5277 views)

Brian Cassidy wrote:
> Hey All,
>
> We're putting out one final development release before we go full bore
> with 5.7100.
>
> NB: It is IMPERATIVE that you TEST this release against YOUR code. Quoth
> mst: "If you do not test now any bugs we ship with are your fault!" --

Broke my application. First post too big, scaling it back. I have

package App::Munchies::Controller::Root;

use strict;
use warnings;
use base qw(CatalystX::Usul::Controller);
use NEXT;

__PACKAGE__->config( namespace => q() );

sub lang : Chained('/') PathPart('') CaptureArgs(1) {
}

sub about : Chained('lang') PathPrefix Args(0) {
}

Which produced this debug loading output under 5.7014

.-------------------------------------+-----------------------------
| Path Spec | Private

+-------------------------------------+-----------------------------
| /*/about | /lang (1)

| | => /about

and under 5.7099_02

.-------------------------------------+--------------------------.
| Path Spec | Private

+-------------------------------------+-------------------------
| /* | /lang (1)

| | => /about

Will continue to dig deeper

--

Regards

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

Jul 16, 2008, 3:57 PM

Post #2 of 9 (5115 views)

Peter Flanigan wrote:
> Brian Cassidy wrote:
>> Hey All,
>>
>> We're putting out one final development release before we go full bore
>> with 5.7100.
>>
>> NB: It is IMPERATIVE that you TEST this release against YOUR code. Quoth
>> mst: "If you do not test now any bugs we ship with are your fault!" --
>
> sub about : Chained('lang') PathPrefix Args(0) {
> }
>

If I replace PathPrefix in the above with PathPart('about') the debug
loading output returns to normal. I note the the change log has

5.7099_02 2008-07-16 19:10:00
- Added PathPrefix attribute

I have been using the PathPrefix attribute in 5.7014

Still digging

--

Regards

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

ash_cpan at firemirror

Jul 16, 2008, 3:57 PM

Post #3 of 9 (5131 views)

On 16 Jul 2008, at 23:39, Peter Flanigan wrote:

> Brian Cassidy wrote:
>> Hey All,
>> We're putting out one final development release before we go full
>> bore
>> with 5.7100.
>> NB: It is IMPERATIVE that you TEST this release against YOUR code.
>> Quoth
>> mst: "If you do not test now any bugs we ship with are your fault!"
>> --
>
> Broke my application. First post too big, scaling it back. I have
>
> package App::Munchies::Controller::Root;
>
> use strict;
> use warnings;
> use base qw(CatalystX::Usul::Controller);
> use NEXT;
>
> __PACKAGE__->config( namespace => q() );
>
> sub lang : Chained('/') PathPart('') CaptureArgs(1) {
> }
>
> sub about : Chained('lang') PathPrefix Args(0) {
> }
>
> Which produced this debug loading output under 5.7014
>
> .-------------------------------------+-----------------------------
> | Path Spec | Private
> +-------------------------------------+-----------------------------
> | /*/about | /lang (1)
> | | => /about
>
> and under 5.7099_02
>
> .-------------------------------------+--------------------------.
> | Path Spec | Private
> +-------------------------------------+-------------------------
> | /* | /lang (1)
> | | => /about
>
>
> Will continue to dig deeper
>
> --
>
> Regards
>
>

Might this be because PathPrefix was only recently re-added as an
attribute that did anything? I remember something being mentioned
about it on #catalyst-dev irc

-ash

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

Jul 16, 2008, 4:21 PM

Post #4 of 9 (5106 views)

Peter Flanigan wrote:
> Peter Flanigan wrote:
>> Brian Cassidy wrote:
>>> Hey All,
>>>
>>> We're putting out one final development release before we go full bore
>>> with 5.7100.
>>>
>>> NB: It is IMPERATIVE that you TEST this release against YOUR code. Quoth
>>> mst: "If you do not test now any bugs we ship with are your fault!" --
>>
>> sub about : Chained('lang') PathPrefix Args(0) {
>> }
>>

Removed the 48 occurrences of PathPrefix from my application and problem
goes away

--

Regards

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

brian.cassidy at gmail

Jul 16, 2008, 4:43 PM

Post #5 of 9 (5128 views)

On Wed, Jul 16, 2008 at 7:39 PM, Peter Flanigan <pjf@roxsoft.co.uk> wrote:
> __PACKAGE__->config( namespace => q() );
>
> sub lang : Chained('/') PathPart('') CaptureArgs(1) {
> }
>
> sub about : Chained('lang') PathPrefix Args(0) {
> }
>
> Which produced this debug loading output under 5.7014
>
> .-------------------------------------+-----------------------------
> | Path Spec | Private
> +-------------------------------------+-----------------------------
> | /*/about | /lang (1)
> | | => /about
>
> and under 5.7099_02
>
> .-------------------------------------+--------------------------.
> | Path Spec | Private
> +-------------------------------------+-------------------------
> | /* | /lang (1)
> | | => /about

I think there's a logical explanation here, as ash stated, you were
using PathPrefix before it existed. :)

In a little more detail:

sub about : Chained('lang') PathPrefix Args(0) {

With a working PathPrefix, that would translate to, in this particular
case, PathPart('') -- hence "/*".

Without a working PathPrefix, it's a no-op (hrmm, do invalid
attributes not throw errors?), and the default PathPart setting is the
sub name, hence "/*/about"

HTH,

-Brian

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

Jul 16, 2008, 5:01 PM

Post #6 of 9 (5123 views)

Brian Cassidy wrote:
> I think there's a logical explanation here, as ash stated, you were
> using PathPrefix before it existed. :)

Since I'm not smart enough to have invented PathPrefix as a no-op I must
have copied from a post to the list or someone else's published code.
It implies that there is at least one more Catalyst application out
there waiting to break

--

Regards

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

brian.cassidy at gmail

Jul 16, 2008, 5:10 PM

Post #7 of 9 (5112 views)

On Wed, Jul 16, 2008 at 9:01 PM, Peter Flanigan <pjf@roxsoft.co.uk> wrote:
> Since I'm not smart enough to have invented PathPrefix as a no-op I must
> have copied from a post to the list or someone else's published code. It
> implies that there is at least one more Catalyst application out there
> waiting to break

Did you have PathPrefix in a base controller? What did it look like?

FWIW, I've posted about PathPrefix before, but the code for it hasn't
changed -- besides it's only 1 line of code. :)

-Brian

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

Jul 16, 2008, 5:27 PM

Post #8 of 9 (5123 views)

Brian Cassidy wrote:
> Did you have PathPrefix in a base controller? What did it look like?

No, there was no code it was just an attribute, which I now know was
unnecessary. My application would have worked under 5.7014 without it.

--

Regards

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Re: [IMPORTANT] TEST Catalyst::Runtime 5.7100 RC1 (aka 5.7099_02) [ In reply to ]

dbix-class at trout

Jul 19, 2008, 6:49 PM

Post #9 of 9 (5079 views)

On Wed, Jul 16, 2008 at 11:57:46PM +0100, Peter Flanigan wrote:
> Peter Flanigan wrote:
> >Brian Cassidy wrote:
> >>Hey All,
> >>
> >>We're putting out one final development release before we go full bore
> >>with 5.7100.
> >>
> >>NB: It is IMPERATIVE that you TEST this release against YOUR code. Quoth
> >>mst: "If you do not test now any bugs we ship with are your fault!" --
> >
> >sub about : Chained('lang') PathPrefix Args(0) {
> >}
> >
>
> If I replace PathPrefix in the above with PathPart('about') the debug
> loading output returns to normal. I note the the change log has
>
> 5.7099_02 2008-07-16 19:10:00
> - Added PathPrefix attribute
>
> I have been using the PathPrefix attribute in 5.7014

Then you must have your own implementation of it that behaves differently.

--
Matt S Trout Need help with your Catalyst or DBIx::Class project?
Technical Director http://www.shadowcat.co.uk/catalyst/
Shadowcat Systems Ltd. Want a managed development or deployment platform?
http://chainsawblues.vox.com/ http://www.shadowcat.co.uk/servers/

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev@lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev