Mailing List Archive: Catalyst: Users

Anybody who fancies some LWP poking ...

Index | Next | Previous | View Threaded

dbix-class at trout

May 4, 2008, 9:02 AM

Post #1 of 21 (605 views)
Permalink

Anybody who fancies some LWP poking ...

http://use.perl.org/~acme/journal/36321

--
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/

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 4, 2008, 12:30 PM

Post #2 of 21 (594 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

yeah, I started setting up cat on a new system yesterday and hit this :-(

The salient line seems to be :
"HTTP::Message content not bytes at lib/Test/WWW/Mechanize/Catalyst.pm
line 88"

I will spend an hour or two on it now, as I hate force installing, but
am starting almost from scratch and don't have much intuition yet as
what this is about. Any pointers welcome ...

D

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 4, 2008, 2:21 PM

Post #3 of 21 (593 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

OK, I walked through the code and found out roughly what is going on.
Hard to say where the real source of the bug is (could be at least two
modules) but I have a workaround which I am using (basically just to
remove the "croak" in HTTP::Message, a bit dirty but probably
harmless).

http://rt.cpan.org/Public/Bug/Display.html?id=34802

I forwarded this to Leon also.

HTH

D

On Sun, May 4, 2008 at 9:30 PM, Daniel McBrearty
<danielmcbrearty[at]gmail.com> wrote:
> yeah, I started setting up cat on a new system yesterday and hit this :-(
>
> The salient line seems to be :
> "HTTP::Message content not bytes at lib/Test/WWW/Mechanize/Catalyst.pm
> line 88"
>
> I will spend an hour or two on it now, as I hate force installing, but
> am starting almost from scratch and don't have much intuition yet as
> what this is about. Any pointers welcome ...
>
> D
>

--
Daniel McBrearty
email : danielmcbrearty at gmail.com
http://www.engoi.com
http://danmcb.vox.com
http://danmcb.blogger.com
find me on linkedin and facebook
BTW : 0873928131

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

apv at sedition

May 4, 2008, 4:00 PM

Post #4 of 21 (593 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

There was a recent SoPW on PerlMonks about this. Seems HTTP::Message is
doing the right thing.

http://perlmonks.org/?node_id=683833

-Ashley

On May 4, 2008, at 2:21 PM, Daniel McBrearty wrote:

> OK, I walked through the code and found out roughly what is going on.
> Hard to say where the real source of the bug is (could be at least two
> modules) but I have a workaround which I am using (basically just to
> remove the "croak" in HTTP::Message, a bit dirty but probably
> harmless).
>
> http://rt.cpan.org/Public/Bug/Display.html?id=34802
>
> I forwarded this to Leon also.
>
> HTH
>
> D
>
> On Sun, May 4, 2008 at 9:30 PM, Daniel McBrearty
> <danielmcbrearty[at]gmail.com> wrote:
>> yeah, I started setting up cat on a new system yesterday and hit
>> this :-(
>>
>> The salient line seems to be :
>> "HTTP::Message content not bytes at lib/Test/WWW/Mechanize/
>> Catalyst.pm
>> line 88"
>>
>> I will spend an hour or two on it now, as I hate force installing,
>> but
>> am starting almost from scratch and don't have much intuition yet as
>> what this is about. Any pointers welcome ...
>>
>> D
>>
>
>
>
> --
> Daniel McBrearty
> email : danielmcbrearty at gmail.com
> http://www.engoi.com
> http://danmcb.vox.com
> http://danmcb.blogger.com
> find me on linkedin and facebook
> BTW : 0873928131
>
> _______________________________________________
> List: Catalyst[at]lists.scsys.co.uk
> Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
> Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
> Dev site: http://dev.catalyst.perl.org/

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 5, 2008, 2:18 AM

Post #5 of 21 (583 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

tht's reasonable.

what is not so reasonable is suddenly deciding that you are going to
*croak* on something which you didn't even test for with previous
releases.

A warning would have been fine, at least for half a year or so.

D

On Mon, May 5, 2008 at 1:00 AM, Ashley <apv[at]sedition.com> wrote:
> There was a recent SoPW on PerlMonks about this. Seems HTTP::Message is
> doing the right thing.
>
> http://perlmonks.org/?node_id=683833
>
> -Ashley
>
> On May 4, 2008, at 2:21 PM, Daniel McBrearty wrote:
>
>> OK, I walked through the code and found out roughly what is going on.
>> Hard to say where the real source of the bug is (could be at least two
>> modules) but I have a workaround which I am using (basically just to
>> remove the "croak" in HTTP::Message, a bit dirty but probably
>> harmless).
>>
>> http://rt.cpan.org/Public/Bug/Display.html?id=34802
>>
>> I forwarded this to Leon also.
>>
>> HTH
>>
>> D
>>
>> On Sun, May 4, 2008 at 9:30 PM, Daniel McBrearty
>> <danielmcbrearty[at]gmail.com> wrote:
>>>
>>> yeah, I started setting up cat on a new system yesterday and hit this :-(
>>>
>>> The salient line seems to be :
>>> "HTTP::Message content not bytes at lib/Test/WWW/Mechanize/Catalyst.pm
>>> line 88"
>>>
>>> I will spend an hour or two on it now, as I hate force installing, but
>>> am starting almost from scratch and don't have much intuition yet as
>>> what this is about. Any pointers welcome ...
>>>
>>> D
>>>
>>
>>
>>
>> --
>> Daniel McBrearty
>> email : danielmcbrearty at gmail.com
>> http://www.engoi.com
>> http://danmcb.vox.com
>> http://danmcb.blogger.com
>> find me on linkedin and facebook
>> BTW : 0873928131
>>
>> _______________________________________________
>> List: Catalyst[at]lists.scsys.co.uk
>> Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
>> Searchable archive:
>> http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
>> Dev site: http://dev.catalyst.perl.org/
>
>
> _______________________________________________
> List: Catalyst[at]lists.scsys.co.uk
> Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
> Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
> Dev site: http://dev.catalyst.perl.org/
>

--
Daniel McBrearty
email : danielmcbrearty at gmail.com
http://www.engoi.com
http://danmcb.vox.com
http://danmcb.blogger.com
find me on linkedin and facebook
BTW : 0873928131

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

jon at jrock

May 5, 2008, 1:46 PM

Post #6 of 21 (576 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

* On Mon, May 05 2008, Daniel McBrearty wrote:
> tht's reasonable.
>
> what is not so reasonable is suddenly deciding that you are going to
> *croak* on something which you didn't even test for with previous
> releases.
>
> A warning would have been fine, at least for half a year or so.

In that case we would be having this exact discussion 6 months from
now. Breaking broken code is always good. +1 for HTTP::Message.

If you don't want new versions of modules, then *don't upgrade them*.

Regards,
Jonathan Rockway

--
print just => another => perl => hacker => if $,=$"

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 5, 2008, 2:12 PM

Post #7 of 21 (576 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

"If you don't want new versions of modules, then *don't upgrade them*."

and when you (or a total newcomer to the language/framework) do a
*new* install? and the "latest greatest" is broken right out of the
box? looks great, hey?

The code may have been broken - but not so broken that it couldn't
work (acceptably) well in the last N releases. But, hey, all of a
sudden, it must be changed *now*.

Dogmatism.

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

zzbbyy at gmail

May 5, 2008, 11:04 PM

Post #8 of 21 (570 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

On Mon, May 5, 2008 at 11:12 PM, Daniel McBrearty
<danielmcbrearty[at]gmail.com> wrote:
> "If you don't want new versions of modules, then *don't upgrade them*."
>
> and when you (or a total newcomer to the language/framework) do a
> *new* install? and the "latest greatest" is broken right out of the
> box? looks great, hey?
>
> The code may have been broken - but not so broken that it couldn't
> work (acceptably) well in the last N releases. But, hey, all of a
> sudden, it must be changed *now*.

There is a proposal from Michael Schwern for a Debian-like releases
(unstable, test, stable and old-stable)
http://www.perlfoundation.org/perl5/index.cgi?cpan_stability_project .
I just thought this might be interesting to you.

--
Zbigniew Lukasiak
http://brudnopis.blogspot.com/

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

dbix-class at trout

May 6, 2008, 8:59 AM

Post #9 of 21 (558 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

On Mon, May 05, 2008 at 03:46:09PM -0500, Jonathan Rockway wrote:
> * On Mon, May 05 2008, Daniel McBrearty wrote:
> > tht's reasonable.
> >
> > what is not so reasonable is suddenly deciding that you are going to
> > *croak* on something which you didn't even test for with previous
> > releases.
> >
> > A warning would have been fine, at least for half a year or so.
>
> In that case we would be having this exact discussion 6 months from
> now. Breaking broken code is always good. +1 for HTTP::Message.

Not when you're that far down the dependency chain.

Anyway, this thread was started to try and help get the fallout fixed,
not for you to posture. Either write code, shut up, or start a separate
thread so I can safely killfile it.

--
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/

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 7, 2008, 1:51 PM

Post #10 of 21 (543 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

Not sure who that's pointed at Matt, but if you mean me, sorry for that.

In all honesty, if I could've worked out *what* needed fixing and
where, I would have done so. What I did was at least try to indicate
to people where the error was coming from and why, and what they might
do temporarily in order to get a workaround. In the absence of any
actual understanding of *why* the changes which triggered the problem
were made, trying to actually implement a fix didn't seem a very good
idea.

At any rate, the suggestion of some kind of stable/testing rating for
module releases sounds like a pretty good idea.

>
> Not when you're that far down the dependency chain.
>
> Anyway, this thread was started to try and help get the fallout fixed,
> not for you to posture. Either write code, shut up, or start a separate
> thread so I can safely killfile it.
>

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

dbix-class at trout

May 9, 2008, 1:34 PM

Post #11 of 21 (520 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

On Wed, May 07, 2008 at 10:51:37PM +0200, Daniel McBrearty wrote:
> Not sure who that's pointed at Matt, but if you mean me, sorry for that.

No, I meant John and his soapbox.

> In all honesty, if I could've worked out *what* needed fixing and
> where, I would have done so. What I did was at least try to indicate
> to people where the error was coming from and why, and what they might
> do temporarily in order to get a workaround. In the absence of any
> actual understanding of *why* the changes which triggered the problem
> were made, trying to actually implement a fix didn't seem a very good
> idea.

I think there was some discussion over the same error in the POE HTTP
client over on perlmonks that included a "this is the line of code that
is wrong". I can't find the link offhand though.

--
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/

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

danielmcbrearty at gmail

May 10, 2008, 5:26 AM

Post #12 of 21 (509 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

> I think there was some discussion over the same error in the POE HTTP
> client over on perlmonks that included a "this is the line of code that
> is wrong". I can't find the link offhand though.

well, I see really three places to look at this:

1. in HTTP::Message - IMO the check on the content is fine, but there
is really no need to die on failure, at least not until the rest of
the codebase gets fixed. Hence my short term fix of just removing
that. Bah.

2. in Test::WWW:Mechanize::Catalyst , you get:

# For some reason Test::WWW::Mechanize uses $response->content everywhere
# instead of $response->decoded_content;
$response->content( $response->decoded_content );

which is the line that causes problems up the chain (I think that the
content check that happens in HTTP::Message->new() is not an issue).

So here, the question is, *why*? Is the right fix for T:W:M to change
to using decoded_content and remove this line? Or just remove this
line anyway? Or is there another issue?

I'd like Leon's opinion on this. Forwarding to him again.

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

acme at astray

May 11, 2008, 11:10 AM

Post #13 of 21 (497 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

2008/5/10 Daniel McBrearty <danielmcbrearty[at]gmail.com>:

> I'd like Leon's opinion on this. Forwarding to him again.

I understand many bits of it but have given up trying to get it
working. I would love a patch which passes tests on both old and new
lib-www-perls.

Leon

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

domm at cpan

May 15, 2008, 4:35 AM

Post #14 of 21 (435 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

Hi!

On Sun, May 11, 2008 at 07:10:27PM +0100, Leon Brocard wrote:
> 2008/5/10 Daniel McBrearty <danielmcbrearty[at]gmail.com>:
>
> > I'd like Leon's opinion on this. Forwarding to him again.
>
> I understand many bits of it but have given up trying to get it
> working. I would love a patch which passes tests on both old and new
> lib-www-perls.

Today we where hit by this, and I dug into the code...

The problems seems to lie in WWW::Mechanize and
Test::WWW::Mechanize::Catalyst.

I was able to (sort of) fix it with the attached two patches.

T:W:M:C tests work after those patches (as do our tests...), but
WWW::Mechanize spews some "Parsing of undecoded UTF-8 will give garbage
when decoding entitie" warnings. And I'm not in the mood for utf8
debugging ...

I have to say that I did not analyse the whole problem, and in fact we
just downgrade to libwww-perl-5.808. But if these findings help someone
with deeper knowledge to really solve the problem, I'd be delighted!

--
#!/usr/bin/perl http://domm.plix.at
for(ref bless{},just'another'perl'hacker){s-:+-$"-g&&print$_.$/}

Attachments:

WWW-Mechanize-1.34.patch (1.50 KB)

Test-WWW-Mechanize-Catalyst-0.42.patch (0.60 KB)

domm at cpan

May 15, 2008, 4:35 AM

Post #15 of 21 (436 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

Attachments:

WWW-Mechanize-1.34.patch (1.50 KB)

Test-WWW-Mechanize-Catalyst-0.42.patch (0.60 KB)

jgoulah at gmail

May 15, 2008, 7:20 AM

Post #16 of 21 (435 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

On Thu, May 15, 2008 at 7:35 AM, Thomas Klausner <domm[at]cpan.org> wrote:
> Hi!
>
> On Sun, May 11, 2008 at 07:10:27PM +0100, Leon Brocard wrote:
>> 2008/5/10 Daniel McBrearty <danielmcbrearty[at]gmail.com>:
>>
>> > I'd like Leon's opinion on this. Forwarding to him again.
>>
>> I understand many bits of it but have given up trying to get it
>> working. I would love a patch which passes tests on both old and new
>> lib-www-perls.
>
> Today we where hit by this, and I dug into the code...
>
> The problems seems to lie in WWW::Mechanize and
> Test::WWW::Mechanize::Catalyst.
>
> I was able to (sort of) fix it with the attached two patches.
>
> T:W:M:C tests work after those patches (as do our tests...), but
> WWW::Mechanize spews some "Parsing of undecoded UTF-8 will give garbage
> when decoding entitie" warnings. And I'm not in the mood for utf8
> debugging ...
>
> I have to say that I did not analyse the whole problem, and in fact we
> just downgrade to libwww-perl-5.808. But if these findings help someone
> with deeper knowledge to really solve the problem, I'd be delighted!
>

Since these fix tests, will these modules get patched and released
with this applied?

Thanks
John

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

jgoulah at gmail

May 15, 2008, 7:20 AM

Post #17 of 21 (435 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

domm at cpan

May 16, 2008, 2:54 AM

Post #18 of 21 (421 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

Hi!

On Thu, May 15, 2008 at 10:20:23AM -0400, John Goulah wrote:

> Since these fix tests, will these modules get patched and released
> with this applied?

Hm, actually I only notified Leon about the patches. I will also submit
the patches directly to Andy (for WWW::Mechanize)

--
#!/usr/bin/perl http://domm.plix.at
for(ref bless{},just'another'perl'hacker){s-:+-$"-g&&print$_.$/}

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

domm at cpan

May 16, 2008, 2:54 AM

Post #19 of 21 (421 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

acme at astray

May 16, 2008, 5:59 PM

Post #20 of 21 (402 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

2008/5/15 John Goulah <jgoulah[at]gmail.com>:

> Since these fix tests, will these modules get patched and released
> with this applied?

Thanks for the investigation Thomas, but I'd like to hold off until
someone actually fixes the real problems and quietens the warnings
using understanding.

For now just downgrade LWP.

Regards, Leon

_______________________________________________
List: Catalyst[at]lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst[at]lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/

acme at astray

May 16, 2008, 5:59 PM

Post #21 of 21 (402 views)
Permalink

Re: Anybody who fancies some LWP poking ... [In reply to]

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact lists@gossamer-threads.com