You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cocoon.apache.org by Antonio Gallardo <ag...@agsoftware.dnsalias.com> on 2003/01/20 07:34:04 UTC

Re: DO NOT REPLY [Bug 16248] New: - Reloading a modified page causes NPE

Hi!

I can confirm that the new cache implementation is a little buggy. I was
the last 5 days trying to make run my application (that is already tested)
in the lastest CVS.

I am another error with the new cache interface. I dont know where is the
problem. Throw an exception about some UnnamedSelector:

ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
> Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
> Attempted to retrieve component with null hint.

Carsten said me about trying to find the component. But I dont know where
the problem is. The sitemap with redirection (line 2) is:

1  <map:match pattern="">
2    <map:redirect-to uri="aut-inicio"/>
3  </map:match>
4
5  <map:match pattern="aut-inicio">
6    <map:act type="auth-loggedIn">
7      <map:parameter name="handler" value="myhandler"/>
8        <map:redirect-to uri="bienvenidos"/>
9    </map:act>
10   <map:generate src="docs/aut-inicio.xsp" type="serverpages"/>
11   <map:transform src="stylesheets/tohtml.xsl"/>
12   <map:transform type="encodeURL"/>
13   <map:serialize/>
14 </map:match>

After redirection it execute the action auth-loggedIn (line 6) and the it
crash with:

ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
> Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
> Attempted to retrieve component with null hint.

I think that the problem can be related to the XSP generator (line 10) .
But this XSP page is just a simple form for logging. I am going really
crazy with this.

I am searching in the 3 books of Cocoon trying to figure out how works
internally but I think I need a debuger to do this.

Please tell me if there is a chance to run Cocoon under Debug mode using
clipse.org or another tool. I am just thinking in the Sun environment
"Forte". Maybe it can help?!

Currently I am setting some logging using logkit but this is a time
consuming approach.

Best Regards,

Antonio Gallardo


Note: Here is the docs/aut-inicio.xsp file. As you can see it is very
simple. This is translated from spanish to english to post on the list.

***************** The XSP Page ****************
<xsp:page xmlns:xsp="http://apache.org/xsp">
> <document menu="false">
>   <permiso>publico</permiso>
>   <header>
>     <title>Price Exchange System</title>
>   </header>
>   <t1 title="User Authentication">
>     <loginform>
>       <url resource="bienvenidos">my-login</url>
>       <field name="username" type="text" size="24"
>         label="User" help="Write the username"/>
>       <field name="password" type="password" size="10"
>         label="Password" help="Write the password"/>
>     </loginform>
>   </t1>
>   <note>Please write your username and password</note>
> </document>
> </xsp:page>

***********************************************



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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Sylvain Wallez <sy...@anyware-tech.com>.
Jeff Turner wrote:

>On Mon, Jan 20, 2003 at 12:34:04AM -0600, Antonio Gallardo wrote:
>  
>
>>Hi!
>>
>>I can confirm that the new cache implementation is a little buggy. I was
>>the last 5 days trying to make run my application (that is already tested)
>>in the lastest CVS.
>>
>>I am another error with the new cache interface. I dont know where is the
>>problem. Throw an exception about some UnnamedSelector:
>>
>>ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
>>    
>>
>>>Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
>>>Attempted to retrieve component with null hint.
>>>      
>>>
>
>Oh.. I was getting those errors too :)  I found it was because my sitemap used
>'map:pipelines' inside map:components, and the syntax recently changed to
>'map:pipes':
>
><map:components>
>  ...
>  <map:pipes default="caching">
>     <map:pipe name="caching" src="org.apache.cocoon.components.pipeline.impl.CachingProcessingPipeline"/>
>     <map:pipe name="noncaching" src="org.apache.cocoon.components.pipeline.impl.NonCachingProcessingPipeline"/>
>  </map:pipes>
></map:components>
>
>Erm.  Shouldn't the sitemap version be incremented for large
>backwards-incompatible changes like this?  Users are NOT going to be amused
>when all their sitemaps break with such an unhelpful message..
>  
>

This change is in the 2.1, which is still "pre-alpha", even if it the 
code is largely stable. Version numbers change will occur when entering 
the beta stage.

Now you're right that the error message doesn't really help in 
diagnosing the problem. I'll take a look at it.

Sylvain

-- 
Sylvain Wallez                                  Anyware Technologies
http://www.apache.org/~sylvain           http://www.anyware-tech.com
{ XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects }



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


RE: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Carsten Ziegeler dijo:
> Antonio Gallardo wrote:
>> Jeff Turner dijo:
>>
>> >Glad I could help.. looks like you were first penguin into the water,
>> and
>> got >severely mauled by the waiting killer whales :/
>>
>> Yeah! His name is Carsten, lol!
>>
>> Antonio Gallardo.
>>
>> PS: Carsten, please dont take it bad. It is just a joke.in the good
>> sense. ;-)
>>
>
> :) (No problem, actually, I'm glad if at least someone mentions my name
> anywhere).

lol

Antonio
>
> Carsten
>
>
> --------------------------------------------------------------------- To
> unsubscribe, e-mail: cocoon-dev-unsubscribe@xml.apache.org
> For additional commands, email: cocoon-dev-help@xml.apache.org




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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Sylvain Wallez <sy...@anyware-tech.com>.
Stefano Mazzocchi wrote:

> Carsten Ziegeler wrote:
>
>>
>> :) (No problem, actually, I'm glad if at least someone mentions my name
>> anywhere).
>
>
> what about: Carsten is a dork!
>
> ahahah
>
> just kidding :)
>
> Sorry couldn't stop myself submitting this one :)))


I did not knew that word. Thanks for helping improving our vocabulary ;-P

Sylvain

-- 
Sylvain Wallez                                  Anyware Technologies
http://www.apache.org/~sylvain           http://www.anyware-tech.com
{ XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects }



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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Stefano Mazzocchi <st...@apache.org>.
Carsten Ziegeler wrote:
> Antonio Gallardo wrote:
> 
>>Jeff Turner dijo:
>>
>>
>>>Glad I could help.. looks like you were first penguin into the water, and
>>
>>got >severely mauled by the waiting killer whales :/
>>
>>Yeah! His name is Carsten, lol!
>>
>>Antonio Gallardo.
>>
>>PS: Carsten, please dont take it bad. It is just a joke.in the good sense.
>>;-)
>>
> 
> 
> :) (No problem, actually, I'm glad if at least someone mentions my name
> anywhere).

what about: Carsten is a dork!

ahahah

just kidding :)

Sorry couldn't stop myself submitting this one :)))

-- 
Stefano Mazzocchi                               <st...@apache.org>
--------------------------------------------------------------------



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


RE: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Carsten Ziegeler <cz...@s-und-n.de>.
Antonio Gallardo wrote:
> Jeff Turner dijo:
>
> >Glad I could help.. looks like you were first penguin into the water, and
> got >severely mauled by the waiting killer whales :/
>
> Yeah! His name is Carsten, lol!
>
> Antonio Gallardo.
>
> PS: Carsten, please dont take it bad. It is just a joke.in the good sense.
> ;-)
>

:) (No problem, actually, I'm glad if at least someone mentions my name
anywhere).

Carsten


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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Jeff Turner dijo:

>Glad I could help.. looks like you were first penguin into the water, and
got >severely mauled by the waiting killer whales :/

Yeah! His name is Carsten, lol!

Antonio Gallardo.

PS: Carsten, please dont take it bad. It is just a joke.in the good sense.
;-)




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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Jeff Turner <je...@apache.org>.
On Mon, Jan 20, 2003 at 02:14:23AM -0600, Antonio Gallardo wrote:
> Hi Jeff:
> 
> You are true. It solves my problem. Many thanks again !
> 
> It works fine!

Glad I could help.. looks like you were first penguin into the water, and
got severely mauled by the waiting killer whales :/  I'd class this as a
bug, not user error.

--Jeff

> Antonio Gallardo
> 

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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Hi Jeff:

You are true. It solves my problem. Many thanks again !

It works fine!

Antonio Gallardo




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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Thanks, Jeff.

You are a genious!

Many thanks. I lost 5 DAYS with this issue. I never thinked that this was
the PROBLEM!

I was checking the authentication framework, then the XSPGenerator, Java
config, Tomcat, etc....

Damn, I am so stupid!

I will try your tip and send back the comments to the list. By the way I
think all the docs must be rewritten to SHOW this LITTLE change. ;-)

Antonio Gallardo

Jeff Turner dijo:
> On Mon, Jan 20, 2003 at 12:34:04AM -0600, Antonio Gallardo wrote:
>> Hi!
>>
>> I can confirm that the new cache implementation is a little buggy. I
>> was the last 5 days trying to make run my application (that is already
>> tested) in the lastest CVS.
>>
>> I am another error with the new cache interface. I dont know where is
>> the problem. Throw an exception about some UnnamedSelector:
>>
>> ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
>> > Thread-7/ExcaliburComponentSelector: UnnamedSelector:
>> ComponentSelector Attempted to retrieve component with null hint.
>
> Oh.. I was getting those errors too :)  I found it was because my
> sitemap used 'map:pipelines' inside map:components, and the syntax
> recently changed to 'map:pipes':
>
> <map:components>
>   ...
>   <map:pipes default="caching">
>      <map:pipe name="caching"
> src="org.apache.cocoon.components.pipeline.impl.CachingProcessingPipeline"/>
> <map:pipe name="noncaching"
> src="org.apache.cocoon.components.pipeline.impl.NonCachingProcessingPipeline"/>
>   </map:pipes>
> </map:components>
>
> Erm.  Shouldn't the sitemap version be incremented for large
> backwards-incompatible changes like this?  Users are NOT going to be
> amused when all their sitemaps break with such an unhelpful message..
>
>
> --Jeff
>
> <snip/>
>>
>> Best Regards,
>>
>> Antonio Gallardo
>
> --------------------------------------------------------------------- To
> unsubscribe, e-mail: cocoon-dev-unsubscribe@xml.apache.org
> For additional commands, email: cocoon-dev-help@xml.apache.org




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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Stefano Mazzocchi <st...@apache.org>.
Jeff Turner wrote:
> On Mon, Jan 20, 2003 at 05:27:29PM +0000, Stefano Mazzocchi wrote:
> 
>>Jeff Turner wrote:
> 
> ...
> 
>>>Users are NOT going to be amused
>>>when all their sitemaps break with such an unhelpful message..
>>
>>Read the WARNING.txt file on the xml-cocoon2 CVS module.
>>
>>Once we exit alpha stage, those things won't change, but now it's hard 
>>enough to come up with a serious design without having to back-support 
>>anything that gets tried out.
>>
>>The reason why we don't release anything right now is *exactly* to 
>>prevent those issues and protect our users.
>>
>>Besides, we had a vote on that change that passed and you didn't vote -1 
>>on it. I think it's very unfair to complain on it afterwards.
> 
> 
> I have no problem with the change, but think it deserves a sitemap
> version increment.  Sylvain's mail explained that one will be made at
> some stage before release, so now I'm happy.

Cool :)

-- 
Stefano Mazzocchi                               <st...@apache.org>
--------------------------------------------------------------------



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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Jeff Turner dijo:
Sylvain's mail explained that one will be made at
> some stage before release, so now I'm happy.


lol

Antonio



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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Jeff Turner <je...@apache.org>.
On Mon, Jan 20, 2003 at 05:27:29PM +0000, Stefano Mazzocchi wrote:
> Jeff Turner wrote:
...
> >Users are NOT going to be amused
> >when all their sitemaps break with such an unhelpful message..
> 
> Read the WARNING.txt file on the xml-cocoon2 CVS module.
> 
> Once we exit alpha stage, those things won't change, but now it's hard 
> enough to come up with a serious design without having to back-support 
> anything that gets tried out.
> 
> The reason why we don't release anything right now is *exactly* to 
> prevent those issues and protect our users.
> 
> Besides, we had a vote on that change that passed and you didn't vote -1 
> on it. I think it's very unfair to complain on it afterwards.

I have no problem with the change, but think it deserves a sitemap
version increment.  Sylvain's mail explained that one will be made at
some stage before release, so now I'm happy.

--Jeff

> -- 
> Stefano Mazzocchi                               <st...@apache.org>
> --------------------------------------------------------------------
> 

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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Antonio Gallardo <ag...@agsoftware.dnsalias.com>.
Stefano Mazzocchi dijo:
> Jeff Turner wrote:
>> On Mon, Jan 20, 2003 at 12:34:04AM -0600, Antonio Gallardo wrote:
>>
>>>Hi!
>>>
>>>I can confirm that the new cache implementation is a little buggy. I
>>> was the last 5 days trying to make run my application (that is already
>>> tested) in the lastest CVS.
>>>
>>>I am another error with the new cache interface. I dont know where is
>>> the problem. Throw an exception about some UnnamedSelector:
>>>
>>>ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
>>>
>>>>Thread-7/ExcaliburComponentSelector: UnnamedSelector:
>>>> ComponentSelector Attempted to retrieve component with null hint.
>>
>>
>> Oh.. I was getting those errors too :)  I found it was because my
>> sitemap used 'map:pipelines' inside map:components, and the syntax
>> recently changed to 'map:pipes':
>>
>> <map:components>
>>   ...
>>   <map:pipes default="caching">
>>      <map:pipe name="caching"
>> src="org.apache.cocoon.components.pipeline.impl.CachingProcessingPipeline"/>
>> <map:pipe name="noncaching"
>> src="org.apache.cocoon.components.pipeline.impl.NonCachingProcessingPipeline"/>
>>   </map:pipes>
>> </map:components>
>>
>> Erm.  Shouldn't the sitemap version be incremented for large
>> backwards-incompatible changes like this?
>
> no, because these things were never released anyway.
>
>> Users are NOT going to be amused
>> when all their sitemaps break with such an unhelpful message..
>
> Read the WARNING.txt file on the xml-cocoon2 CVS module.
>
> Once we exit alpha stage, those things won't change, but now it's hard
> enough to come up with a serious design without having to back-support
> anything that gets tried out.
>
> The reason why we don't release anything right now is *exactly* to
> prevent those issues and protect our users.
>
> Besides, we had a vote on that change that passed and you didn't vote -1
>  on it. I think it's very unfair to complain on it afterwards.
>
> --
> Stefano Mazzocchi                               <st...@apache.org>
> --------------------------------------------------------------------

Hi Stefano:

Sorry about the problem. It was my fault. But anyway, I think we need to
put a little better error description than:

ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
Attempted to retrieve component with null hint.

Please take the advise in the good sense. And lets continue this good job
called Cocoon.

I will continue being a alpha-beta to gamma tester of Cocoon. I enjoy to
see the new features and trying it.

By the wat the new cache system + XSLTC is amazing faster! I am impresed
of it speed with XSP.

Best Regards,

Antonio Gallardo




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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Stefano Mazzocchi <st...@apache.org>.
Jeff Turner wrote:
> On Mon, Jan 20, 2003 at 12:34:04AM -0600, Antonio Gallardo wrote:
> 
>>Hi!
>>
>>I can confirm that the new cache implementation is a little buggy. I was
>>the last 5 days trying to make run my application (that is already tested)
>>in the lastest CVS.
>>
>>I am another error with the new cache interface. I dont know where is the
>>problem. Throw an exception about some UnnamedSelector:
>>
>>ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
>>
>>>Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
>>>Attempted to retrieve component with null hint.
> 
> 
> Oh.. I was getting those errors too :)  I found it was because my sitemap used
> 'map:pipelines' inside map:components, and the syntax recently changed to
> 'map:pipes':
> 
> <map:components>
>   ...
>   <map:pipes default="caching">
>      <map:pipe name="caching" src="org.apache.cocoon.components.pipeline.impl.CachingProcessingPipeline"/>
>      <map:pipe name="noncaching" src="org.apache.cocoon.components.pipeline.impl.NonCachingProcessingPipeline"/>
>   </map:pipes>
> </map:components>
> 
> Erm.  Shouldn't the sitemap version be incremented for large
> backwards-incompatible changes like this?

no, because these things were never released anyway.

> Users are NOT going to be amused
> when all their sitemaps break with such an unhelpful message..

Read the WARNING.txt file on the xml-cocoon2 CVS module.

Once we exit alpha stage, those things won't change, but now it's hard 
enough to come up with a serious design without having to back-support 
anything that gets tried out.

The reason why we don't release anything right now is *exactly* to 
prevent those issues and protect our users.

Besides, we had a vote on that change that passed and you didn't vote -1 
on it. I think it's very unfair to complain on it afterwards.

-- 
Stefano Mazzocchi                               <st...@apache.org>
--------------------------------------------------------------------



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


Re: [Bug 16248] New: - Reloading a modified page causes NPE

Posted by Jeff Turner <je...@apache.org>.
On Mon, Jan 20, 2003 at 12:34:04AM -0600, Antonio Gallardo wrote:
> Hi!
> 
> I can confirm that the new cache implementation is a little buggy. I was
> the last 5 days trying to make run my application (that is already tested)
> in the lastest CVS.
> 
> I am another error with the new cache interface. I dont know where is the
> problem. Throw an exception about some UnnamedSelector:
> 
> ERROR   (2003-01-14) 17:12.17:171   [core.manager] (/myapp/aut-inicio)
> > Thread-7/ExcaliburComponentSelector: UnnamedSelector: ComponentSelector
> > Attempted to retrieve component with null hint.

Oh.. I was getting those errors too :)  I found it was because my sitemap used
'map:pipelines' inside map:components, and the syntax recently changed to
'map:pipes':

<map:components>
  ...
  <map:pipes default="caching">
     <map:pipe name="caching" src="org.apache.cocoon.components.pipeline.impl.CachingProcessingPipeline"/>
     <map:pipe name="noncaching" src="org.apache.cocoon.components.pipeline.impl.NonCachingProcessingPipeline"/>
  </map:pipes>
</map:components>

Erm.  Shouldn't the sitemap version be incremented for large
backwards-incompatible changes like this?  Users are NOT going to be amused
when all their sitemaps break with such an unhelpful message..


--Jeff

<snip/>
>
> Best Regards,
> 
> Antonio Gallardo

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