Antwort: Re: method name of listener

2007-03-06 Thread Kristian Marinkovic
definining a listener for my component looks sth like this (Tap standard):

component id=asyncOrder type=cross:AsyncListenerCaller
   binding name=listener value=listener:contactOrder /
   ...
/component

in my component i have a method returning a IActionListener:
public abstract IActionListener getListener();

now i want to generate a js function with the same name and i dont know how
to get it from IActionListener (at least i found no public methods:)). My
workaround is to add another component parameter that contains the
name of the listener as a literal.

... i hope this does not sound that stupid :)




   
 Jesse Kuhnert   
 [EMAIL PROTECTED] 
 omAn 
Tapestry users   
 06.03.2007 05:01   users@tapestry.apache.org
 Kopie 
   
  Bitte antwortenThema 
an  Re: method name of listener
 Tapestry users  
 [EMAIL PROTECTED] 
pache.org 
   
   
   




Isn't the name of a listener known by the very definition? I mean,
when ~don't~ you know the name of it?

On 2/19/07, Kristian Marinkovic [EMAIL PROTECTED] wrote:

 hi all,

 is there a way to obtain the method name of a listener?

 I need it to generate a javascript function with the same
 name. calling this js function will trigger an asynchronous XHR
 call (Tapestry.bind) that triggers the corresponding listener.
 The js method also accepts parameters that get send to the page.

 in case anyone else needs this very, very simple component
 i'll post it somewhere.

 g,
 kris


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



@Shell component, refresh and IE

2007-03-06 Thread Patrick Klein

Hello!

I'm using the @Shell component's refresh tag to set trigger the reload 
of a given page every 10 seconds. All works fine using Firefox. However, 
using IE all goes havoc as IE seems to drop at least parts of the reload 
url. The refresh part of the page looks like the following:


   meta http-equiv=Refresh content=10; 
URL=http://localhost:8080/myapp/ImportMatrix.page;jsessionid=C79B17BAFCE7166D80F92708F500D4B9; 
/


after the refresh i'm seeing the following url in IE:
  
   http://localhost:8080/myapp/ImportMatrix.page


completely disregarding the jsessionid, which brings me back to the 
login page.


Used Tapestry-version: 4.1.1, 4.1.2-SNAPSHOT (15.2.2007)

Could anyone give me a hint on how to fix this?

Regards,
   Patrick

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



AW: @Shell component, refresh and IE

2007-03-06 Thread Peter Schröder
hi patrick,

this might be because of the ; as delimiter for the sessionId wich is also 
used as delimiter for the values of the content-parameter. so i think that IE 
splits this entry to
10
URL=http://localhost:8080/myapp/ImportMatrix.page
jsessionid=C79B17BAFCE7166D80F92708F500D4B9

wich brings you to your start-page. 
instead of a meta-tag you might use javascript for page-refresh as a workaround.

kind regars

-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:20
An: Tapestry users
Betreff: @Shell component, refresh and IE

Hello!

I'm using the @Shell component's refresh tag to set trigger the reload 
of a given page every 10 seconds. All works fine using Firefox. However, 
using IE all goes havoc as IE seems to drop at least parts of the reload 
url. The refresh part of the page looks like the following:

meta http-equiv=Refresh content=10; 
URL=http://localhost:8080/myapp/ImportMatrix.page;jsessionid=C79B17BAFCE7166D80F92708F500D4B9;
 
/

after the refresh i'm seeing the following url in IE:
   
http://localhost:8080/myapp/ImportMatrix.page

completely disregarding the jsessionid, which brings me back to the 
login page.

Used Tapestry-version: 4.1.1, 4.1.2-SNAPSHOT (15.2.2007)

Could anyone give me a hint on how to fix this?

Regards,
Patrick

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: AW: @Shell component, refresh and IE

2007-03-06 Thread Patrick Klein

Hi, Peter,

somehow i allready thought it was something like that. Thanks for the 
answer :)

Switching to using cookies also does the trick.
Both solutions somehow are not really satisfying as the Shell component 
should be able to attend this...


Regards,
   Patrick

hi patrick,

this might be because of the ; as delimiter for the sessionId wich is also 
used as delimiter for the values of the content-parameter. so i think that IE splits this 
entry to
10
URL=http://localhost:8080/myapp/ImportMatrix.page
jsessionid=C79B17BAFCE7166D80F92708F500D4B9

wich brings you to your start-page. 
instead of a meta-tag you might use javascript for page-refresh as a workaround.


kind regars

-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:20

An: Tapestry users
Betreff: @Shell component, refresh and IE

Hello!

I'm using the @Shell component's refresh tag to set trigger the reload 
of a given page every 10 seconds. All works fine using Firefox. However, 
using IE all goes havoc as IE seems to drop at least parts of the reload 
url. The refresh part of the page looks like the following:


meta http-equiv=Refresh content=10; 
URL=http://localhost:8080/myapp/ImportMatrix.page;jsessionid=C79B17BAFCE7166D80F92708F500D4B9; 
/


after the refresh i'm seeing the following url in IE:
   
http://localhost:8080/myapp/ImportMatrix.page


completely disregarding the jsessionid, which brings me back to the 
login page.


Used Tapestry-version: 4.1.1, 4.1.2-SNAPSHOT (15.2.2007)

Could anyone give me a hint on how to fix this?

Regards,
Patrick

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


  



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Jesse--Temporary jfly public repository

2007-03-06 Thread Andrea Chiumenti

Hi Jesse,
I've set a temporary jfly repository (it's my host)
http://www.wingstech.it/jfly/nightly/

p.s.
do you know where can I put it when officially announced, currently the
project is hosted by sourceforge.

Ciao,
kiuma


Re: T5 binding problem

2007-03-06 Thread Jiri Mares

Hi Howard,

 2) OGNL isn't supported in T5.

OGNL won't be supported or not yet.

-- 
Jiří Mareš (mailto:[EMAIL PROTECTED])
ČSAD SVT Praha, s.r.o. (http://www.svt.cz)
Czech Republic

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Tapestry 4.1.2-SNAPSHOT

2007-03-06 Thread Paulo Ramos

Hello,

I am using Tapestry 4.1.2-SNAPSHOT and yesterday, when i deployed my 
aplication the server throw this error:
   java.lang.NoSuchMethodError: 
javassist.CtClass.toClass(Ljava/lang/ClassLoader;Ljava/security/ProtectionDomain;)Ljava/lang/Class;


Can anyone help me?

Thanks

Paulo Ramos



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



AW: Tapestry 4.1.2-SNAPSHOT

2007-03-06 Thread Peter Schröder
does your server contain the javassist lib in the expected version? 

-Ursprüngliche Nachricht-
Von: Paulo Ramos [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 12:49
An: users@tapestry.apache.org
Betreff: Tapestry 4.1.2-SNAPSHOT

Hello,

I am using Tapestry 4.1.2-SNAPSHOT and yesterday, when i deployed my 
aplication the server throw this error:
java.lang.NoSuchMethodError: 
javassist.CtClass.toClass(Ljava/lang/ClassLoader;Ljava/security/ProtectionDomain;)Ljava/lang/Class;

Can anyone help me?

Thanks

Paulo Ramos



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Friendly URLs in libraries

2007-03-06 Thread Stephane Decleire

Hi,

I've tried to implement friendly urls with the solution described at 
http://tapestry.apache.org/tapestry4/UsersGuide/friendly-urls.html
This works fine in my project but not in the pages included in libraries 
used by my project.

Any advice is welcome.

--
Stephane




Re: AW: Tapestry 4.1.2-SNAPSHOT

2007-03-06 Thread Paulo Ramos

I am using javassist 3.0

Peter Schröder escreveu:
does your server contain the javassist lib in the expected version? 


-Ursprüngliche Nachricht-
Von: Paulo Ramos [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 12:49

An: users@tapestry.apache.org
Betreff: Tapestry 4.1.2-SNAPSHOT

Hello,

I am using Tapestry 4.1.2-SNAPSHOT and yesterday, when i deployed my 
aplication the server throw this error:
java.lang.NoSuchMethodError: 
javassist.CtClass.toClass(Ljava/lang/ClassLoader;Ljava/security/ProtectionDomain;)Ljava/lang/Class;


Can anyone help me?

Thanks

Paulo Ramos



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




  




smime.p7s
Description: S/MIME Cryptographic Signature


Re: AW: AW: @Shell component, refresh and IE

2007-03-06 Thread Patrick Klein

Hello Peter,

thanks for the hint. The simplest solution seems to be extending 
RequestCycle and override the encodeURL method to replace the semicolon. 
The extension is already done but right now i don't know where configure 
it to make sure the custome RequestCycle is used instead of tapestrys 
RequestCycle. I guess it has to be added to the hivemind.xml or my 
application file, but could someone show me how? Thnx in advance.


Regards,
   Patrick

hi patrick,

you may also try to url-encode the tapestry-url so that ; will be %3B and IE 
wont get confused

kind regards 


-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:46

An: Tapestry users
Betreff: Re: AW: @Shell component, refresh and IE

Hi, Peter,

somehow i allready thought it was something like that. Thanks for the 
answer :)

Switching to using cookies also does the trick.
Both solutions somehow are not really satisfying as the Shell component 
should be able to attend this...


Regards,
Patrick
  

hi patrick,

this might be because of the ; as delimiter for the sessionId wich is also 
used as delimiter for the values of the content-parameter. so i think that IE splits this 
entry to
10
URL=http://localhost:8080/myapp/ImportMatrix.page
jsessionid=C79B17BAFCE7166D80F92708F500D4B9

wich brings you to your start-page. 
instead of a meta-tag you might use javascript for page-refresh as a workaround.


kind regars

-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:20

An: Tapestry users
Betreff: @Shell component, refresh and IE

Hello!

I'm using the @Shell component's refresh tag to set trigger the reload 
of a given page every 10 seconds. All works fine using Firefox. However, 
using IE all goes havoc as IE seems to drop at least parts of the reload 
url. The refresh part of the page looks like the following:


meta http-equiv=Refresh content=10; 
URL=http://localhost:8080/myapp/ImportMatrix.page;jsessionid=C79B17BAFCE7166D80F92708F500D4B9; 
/


after the refresh i'm seeing the following url in IE:
   
http://localhost:8080/myapp/ImportMatrix.page


completely disregarding the jsessionid, which brings me back to the 
login page.


Used Tapestry-version: 4.1.1, 4.1.2-SNAPSHOT (15.2.2007)

Could anyone give me a hint on how to fix this?

Regards,
Patrick

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


  




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


  



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: T5 binding problem

2007-03-06 Thread Howard Lewis Ship

It may never be; the design of T5 allows for easy property expressions
right in the binding, and more complex expression delegate out to Java
code easily (this is practical in T5 because of the automatic class
reloading).

A lot of the magic logic in BeanEditForm and Grid is based on some
APIs that OGNL will have trouble replicating ... in terms of
identifying a property is a type-safe manner, and gaining access to
annotations on the getter and/or setter.

On 3/6/07, Jiri Mares [EMAIL PROTECTED] wrote:


Hi Howard,

 2) OGNL isn't supported in T5.

OGNL won't be supported or not yet.

--
Jiří Mareš (mailto:[EMAIL PROTECTED])
ČSAD SVT Praha, s.r.o. (http://www.svt.cz)
Czech Republic

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com


Re: Problem with assignment of variables between pages

2007-03-06 Thread Patrick Klein

Hello Jesse,

the problem is a bit harder to track. All of my test cases (with tons of 
debug output) worked out as the should.


The problem as it seems is that under some circumstances one jsession is 
able to access the (persistent) variables of another jsessions instance 
of the same page (or maybe also a different page). The problem mostly 
occurs after some exceptions were thrown and maybe a hibernate session 
was left in a questionable state, but i'm not 100% sure if it really has 
to do with the hibernate session itself.


As soon as i find another (hopefully reproducable) appearance of this 
behavior i'll try to track it again.


Regards,
   Patrick


This may have been the case with a previous 4.1.2-SNAPSHOT version
that was doing some ~things~ with cglib and properties but I doubt
it...

If your properties aren't persistent there is no magic happening,
they just get set on the object.

Perhaps you are doing this in some kind of page loading method like
validate() or similar?

Even with persistent properties the object is first handed off to the
persistent store changes manager impl and then just set on the page
object, so there's no valid reason I can think of where this would
happen as you've described it.

I'd add some breakpoints to your code and run them in any of the
standard ide debuggers to find out definitively what is happening if
it were me. They leave 0 room for doubt in these situations.

On 2/27/07, Patrick Klein [EMAIL PROTECTED] wrote:

Hi!

cache is not disabled.

Regards,
Patrick
 Are you disabling the cache ?

 On 2/27/07, Patrick Klein [EMAIL PROTECTED] wrote:

 Hallo!

 I'm hoping I'm not completely OT with this problem here...

 We are using setters and getters to pass over objects between pages,
 like (in pseudo-java)

 page a:
 (PageB) page = (PageB) cycle.getPage(PageB);
 page.setVar(obj);
 [...]
 cycle.activate(page);

 page b:
 Obj var = getVar();
 if(var != null) {
 [...]
 }
 [...]
 public abstract Obj getVar();
 public abstract void setVar(Obj var);

 The problem which is driving us up the walls is that in some 
cases, the
 transition seems to get messed up, as a not-null object is handed 
over,

 but on the receiving side this object is null...
 An additional problem is that this behavior seems to be not valid for
 all http-sessions handled by tomcat, some work as they should.

 We are using Tap 4.1.2 (SNAPSHOT from 16.02.2007), Tomcat 5.5.17 (and
 5.5.20) and Hibernate 3.2.1ga with java 1.5.11SE.

 Did anyone here encounter a similar problem and if yes, give me a 
hint

 of where to look or how to fix it?

 Regards,
 Patrick Klein

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]





-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]








-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Autocompleter Questions

2007-03-06 Thread Chris Chiappone

It appears that tapestry isn't finding the location of dojo for some reason.

On 3/6/07, Jesse Kuhnert [EMAIL PROTECTED] wrote:

No idea, but a working example always helps me:

http://svn.apache.org/viewvc/tapestry/tapestry4/trunk/tapestry-examples/TimeTracker/src/

On 2/22/07, Chris Chiappone [EMAIL PROTECTED] wrote:
 I am trying to get the autocompleter component to work with tap 4.1.1
 I used to have tacos:Autocompleter working and wanted to replace it
 with the new default component.  I am not getting any exceptions but
 there are no values in my drop down.  Here is my page class and html

 public IAutocompleteModel getCompanyModel() {
 log.info(SETTING UP MODEL);
 ListCompany searchList = new ArrayListCompany();

 CollectionCompany comps = getCompanyDao().findAll();
 for(Company c : comps){
 log.info(Adding company:  + c.getId() + + 
c.getName());
 searchList.add(c);

 }

 return(new DefaultAutocompleteModel(searchList, id, 
name));

 }


 @Component(id=autoCompleter, type=Autocompleter, bindings={
 model=ognl:companyModel, value=ognl:TppName,
 focus=literal:false})
 public abstract Autocompleter getAutoCompleter();

 


 form jwcid=[EMAIL PROTECTED] 
listener=listener:searchForm
 pinput jwcid=autoCompleter / input 
jwcid=@Submit
 size=200px value=message:searchTpp class=searchSubmit //p
 /form

 --
 ~chris

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
~chris

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Tapestry 4.1.2 DirectLink updateComponents within a Dialog

2007-03-06 Thread Renat Zubairov

Upate of the components inside the Dialog is not working indeeed, I've
tried to debug it once, and it seems that dialog do not scan the
internal components when it searches for the updated client IDs.


On 06/03/07, Jesse Kuhnert [EMAIL PROTECTED] wrote:

Damnit...I don't know how I let something that obvious slip through.
I'll test this out tomorrow either way.

On 3/6/07, Leffel, Daniel [EMAIL PROTECTED] wrote:
 So simply updating the entire dialog worked perfectly. This suits my needs 
because the response is small, so I haven't debugged more. I'll check it out 
tomorrow and see if I can find out what's going on when trying to update 
components within the dialog.

 Thanks,

 Danny


 -Original Message-
 From: Jesse Kuhnert [mailto:[EMAIL PROTECTED]
 Sent: Mon 3/5/2007 9:05 PM
 To: Tapestry users
 Subject: Re: Tapestry 4.1.2 DirectLink updateComponents within a Dialog

 That's odd. I have some pretty complicated components being updated
 inside of a dialog, though admittedly I normally just specify the
 dialog as being the thing to update.

 I can take a look tomorrow to see if there's anything obvious I've
 missed...Did you have something more specific as an example? (assuming
 that it isn't obviously broken when I try it )

 There are some more debugging options available now as well, esp with
 intercepting the interactions between DojoAjaxResponseBuilder and
 client, maybe seeing what the server is doing will help you:

 http://tapestry.apache.org/tapestry4.1/ajax/debugging.html

 On 3/5/07, Leffel, Daniel [EMAIL PROTECTED] wrote:
  Am I doing something wrong? I can't get any components nested within a 
dialog to update using DirectLink. If I move the very same components outside of the 
dialog, everything updates correctly using async or json. It doesn't matter whether 
the dialog is hidden or not. Inside the dialog, nothing updates - there is nothing in 
the AJAX response. Is there a trick to making this work?
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 


 --
 Jesse Kuhnert
 Tapestry/Dojo team member/developer

 Open source based consulting work centered around
 dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]



 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]



--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Best regards,
Renat Zubairov

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Tapestry 5 screencasts for download ?

2007-03-06 Thread Renat Zubairov

I guess you can directly download them from the page by the source

For example

http://howardlewisship.com/screencasts/tapestry5_screencast5.mov

From

http://svn.apache.org/viewvc/tapestry/tapestry5/tapestry-project/trunk/src/site/resources/screencast_5.html?view=markup

On 06/03/07, Dwi Ardi Irawan [EMAIL PROTECTED] wrote:

Is there Tapestry 5 screencasts for download ?


dwi ardi irawan

http://dwiardiirawan.blogspot.com




--
Best regards,
Renat Zubairov

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Asset Service/JS file throws exception in Perl5Matcher

2007-03-06 Thread Chris Norris
We've found an obscure and hard to repeat bug in our app that is caused
when a page tries to load a javascript file.  The code that causes it is
located just inside the body tag:

 

body class=contentBody id=Body
script type=text/javascript
src=/assets/static/org/apache/tapestry/contrib/palette/PaletteFunctions
.js/script

 

 

The error the browser reports is a syntax error.  On a GET with the URL
https://www.domain.com/assets/static/org/apache/tapestry/contrib/palette
/PaletteFunctions.js, the file returned is actually a Tapestry Error
page, not a javascript file.  Digging through the logs revealed an error
throw when trying to retrieve the PaletteFunctions.js file:

 

java.lang.ArrayIndexOutOfBoundsException: 25

at
org.apache.oro.text.regex.Perl5Matcher.__interpret(Unknown Source)

at org.apache.oro.text.regex.Perl5Matcher.contains(Unknown
Source)

at org.apache.oro.text.regex.Perl5Matcher.contains(Unknown
Source)

at
org.apache.tapestry.util.RegexpMatcher.contains(RegexpMatcher.java:106)

at
org.apache.tapestry.asset.ResourceMatcherImpl.containsResource(ResourceM
atcherImpl.java:71)

at
$ResourceMatcher_11127c1ee2e.containsResource($ResourceMatcher_11127c1ee
2e.java)

at
org.apache.tapestry.asset.AssetService.service(AssetService.java:218)

at
$IEngineService_11127c1ed2e.service($IEngineService_11127c1ed2e.java)

 

I did not expect to see an ArrayIndexOutOfBoundsException there.  We put
a copy of RegexpMatcher on our classpath to get some more info,
specifically what the path and pattern strings were that it was using.
The path was /org/apache/tapestry/contrib/palette/PaletteFunctions.js
and the pattern was ^/org/apache/tapestry/.*.css.  I wrote a small
class to execute code almost identical to that of the RegexpMatcher:

 

String s_pattern = ^/org/apache/tapestry/.*.css;

String path =
/org/apache/tapestry/contrib/palette/PaletteFunctions.js;

Perl5Compiler compiler = new Perl5Compiler();

Pattern pattern = compiler.compile(s_pattern,
Perl5Compiler.SINGLELINE_MASK | Perl5Compiler.READ_ONLY_MASK);

Perl5Matcher matcher = new Perl5Matcher();

System.out.println(matcher.contains(path, pattern));

 

No exception is thrown.  We can't even get this error to show up
reliably.  It seems like closing your browser and restarting your
session makes it go away.  This is the kind of bug that drives me
insane.



Re: T5 binding problem

2007-03-06 Thread Howard Lewis Ship

Yep, that's the intention.

Not using OGNL falls into the blazing speed - no reflection goal.
Jesse's busy taking the reflection out of OGNL, so it may make a
comeback as an add-on.

On 3/6/07, Davor Hrg [EMAIL PROTECTED] wrote:

thnx,
I did add a syntethic property to do the task,
but thought ognl was supposed to work 

actualy I like the approach via property more, since then I have
autocomplete and all the stuff I'm used to while writing java code,
and auto reloading of classes makes it as fast to test, as if it was written
in the template :)

Davor Hrg

On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:

 Two problems here:

 1) Not well formed XML.  That  inside the test attribute isn't
 allow.  Use lt;
 2) OGNL isn't supported in T5.

 For anything more complicated than a property access, you do the work
 in Java code as a synthetic property.

 On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
  Hi,
  maybe this was reported and solved but here it goes ...
 
  I've tried the tapestry tutorial and it was very nice and easy.
  I've made few mistakes but error reporting is great so it was fixed
 quickly.
 
  the problem occured when I tried to add some of the proposed changes to
 the
  app..
 
  just to test the expressions I added:
  t:comp type=If test=ognl:index5
 
  but I get the error
  Failure parsing template context:WEB-INF/Guess.html: The value of
 attribute
  test associated with an element type t:comp must not contain the ''
  character.
 
  Am I doing it wrong, or it's just a bug to be fixed later
 
  Davor Hrg
 


 --
 Howard M. Lewis Ship
 TWD Consulting, Inc.
 Independent J2EE / Open-Source Java Consultant
 Creator and PMC Chair, Apache Tapestry
 Creator, Apache HiveMind

 Professional Tapestry training, mentoring, support
 and project work.  http://howardlewisship.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]






--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: Tapestry 4.1.2 DirectLink updateComponents within a Dialog

2007-03-06 Thread Andreas Andreou

Actually Dialog extends from AbstractWidget

It's my understanding that abstractWidgets are NOT containers of other
components, though
i don't know why they shouldn't be :)

Anyway, this led my to
http://tacos.svn.sourceforge.net/viewvc/tacos/tacos-4.1/trunk/tacos-core/src/java/net/sf/tacos/components/dojo/GenericWidget.java?view=markup
which has an isContainer property that allow component users to specify
this...

Perhaps we can combine the two and allow such behavior.




On 3/6/07, Renat Zubairov [EMAIL PROTECTED] wrote:


Upate of the components inside the Dialog is not working indeeed, I've
tried to debug it once, and it seems that dialog do not scan the
internal components when it searches for the updated client IDs.


On 06/03/07, Jesse Kuhnert [EMAIL PROTECTED] wrote:
 Damnit...I don't know how I let something that obvious slip through.
 I'll test this out tomorrow either way.

 On 3/6/07, Leffel, Daniel [EMAIL PROTECTED] wrote:
  So simply updating the entire dialog worked perfectly. This suits my
needs because the response is small, so I haven't debugged more. I'll check
it out tomorrow and see if I can find out what's going on when trying to
update components within the dialog.
 
  Thanks,
 
  Danny
 
 
  -Original Message-
  From: Jesse Kuhnert [mailto:[EMAIL PROTECTED]
  Sent: Mon 3/5/2007 9:05 PM
  To: Tapestry users
  Subject: Re: Tapestry 4.1.2 DirectLink updateComponents within a
Dialog
 
  That's odd. I have some pretty complicated components being updated
  inside of a dialog, though admittedly I normally just specify the
  dialog as being the thing to update.
 
  I can take a look tomorrow to see if there's anything obvious I've
  missed...Did you have something more specific as an example? (assuming
  that it isn't obviously broken when I try it )
 
  There are some more debugging options available now as well, esp with
  intercepting the interactions between DojoAjaxResponseBuilder and
  client, maybe seeing what the server is doing will help you:
 
  http://tapestry.apache.org/tapestry4.1/ajax/debugging.html
 
  On 3/5/07, Leffel, Daniel [EMAIL PROTECTED] wrote:
   Am I doing something wrong? I can't get any components nested within
a dialog to update using DirectLink. If I move the very same components
outside of the dialog, everything updates correctly using async or json. It
doesn't matter whether the dialog is hidden or not. Inside the dialog,
nothing updates - there is nothing in the AJAX response. Is there a trick to
making this work?
  
  
-
   To unsubscribe, e-mail: [EMAIL PROTECTED]
   For additional commands, e-mail: [EMAIL PROTECTED]
  
  
 
 
  --
  Jesse Kuhnert
  Tapestry/Dojo team member/developer
 
  Open source based consulting work centered around
  dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 


 --
 Jesse Kuhnert
 Tapestry/Dojo team member/developer

 Open source based consulting work centered around
 dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




--
Best regards,
Renat Zubairov

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Andreas Andreou - [EMAIL PROTECTED] - http://andyhot.di.uoa.gr
Tapestry / Tacos developer
Open Source / JEE Consulting


Re: T5 binding problem

2007-03-06 Thread Angelo Luis

No no no...

How can have a null handler now???

If a have a customer - address - street in a text component, for example 
custumer.address.street and i dont instanciate a address, a
NullPointerException is throw. With a ognl null handler, if the address is
null he creates one , like Webwork... i dont need to instanciate all object
graph to use in a page...

i will have this same problem in t5?

On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:


Two problems here:

1) Not well formed XML.  That  inside the test attribute isn't
allow.  Use lt;
2) OGNL isn't supported in T5.

For anything more complicated than a property access, you do the work
in Java code as a synthetic property.

On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
 Hi,
 maybe this was reported and solved but here it goes ...

 I've tried the tapestry tutorial and it was very nice and easy.
 I've made few mistakes but error reporting is great so it was fixed
quickly.

 the problem occured when I tried to add some of the proposed changes to
the
 app..

 just to test the expressions I added:
 t:comp type=If test=ognl:index5

 but I get the error
 Failure parsing template context:WEB-INF/Guess.html: The value of
attribute
 test associated with an element type t:comp must not contain the ''
 character.

 Am I doing it wrong, or it's just a bug to be fixed later

 Davor Hrg



--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




Re: T5 binding problem

2007-03-06 Thread Jesse Kuhnert

I don't think it's even a matter of choice right now to be honest.
There are still quite a few changes that need to be made before ognl
is suitable for use with T5. (such as the addition of parameter
accessors to handle annotated params, generics, IoC injection ,
removal of static references to make jvm class reloads work , etc)

The other projects want these kinds of things too (as well as Drew
telling me he specifically wanted to see the IoC stuff happen ), so
there's more than enough pressure to get it done.

Howard and I both know it's not ready ~yet~, but if/when it does get
to that point I'll ping him and let him know...

On 3/6/07, Angelo Luis [EMAIL PROTECTED] wrote:

No no no...

How can have a null handler now???

If a have a customer - address - street in a text component, for example 
custumer.address.street and i dont instanciate a address, a
NullPointerException is throw. With a ognl null handler, if the address is
null he creates one , like Webwork... i dont need to instanciate all object
graph to use in a page...

i will have this same problem in t5?

On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:

 Two problems here:

 1) Not well formed XML.  That  inside the test attribute isn't
 allow.  Use lt;
 2) OGNL isn't supported in T5.

 For anything more complicated than a property access, you do the work
 in Java code as a synthetic property.

 On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
  Hi,
  maybe this was reported and solved but here it goes ...
 
  I've tried the tapestry tutorial and it was very nice and easy.
  I've made few mistakes but error reporting is great so it was fixed
 quickly.
 
  the problem occured when I tried to add some of the proposed changes to
 the
  app..
 
  just to test the expressions I added:
  t:comp type=If test=ognl:index5
 
  but I get the error
  Failure parsing template context:WEB-INF/Guess.html: The value of
 attribute
  test associated with an element type t:comp must not contain the ''
  character.
 
  Am I doing it wrong, or it's just a bug to be fixed later
 
  Davor Hrg
 


 --
 Howard M. Lewis Ship
 TWD Consulting, Inc.
 Independent J2EE / Open-Source Java Consultant
 Creator and PMC Chair, Apache Tapestry
 Creator, Apache HiveMind

 Professional Tapestry training, mentoring, support
 and project work.  http://howardlewisship.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]






--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



[T5] ?. operator for null-safe graph traversal

2007-03-06 Thread Howard Lewis Ship

customer?.address?.street

I've borrowed some syntax from Groovy to handle this case; the ?.
operator returns null if the LHS is null, otherwise is traverses into
the property on the RHS.

On 3/6/07, Angelo Luis [EMAIL PROTECTED] wrote:

No no no...

How can have a null handler now???

If a have a customer - address - street in a text component, for example 
custumer.address.street and i dont instanciate a address, a
NullPointerException is throw. With a ognl null handler, if the address is
null he creates one , like Webwork... i dont need to instanciate all object
graph to use in a page...

i will have this same problem in t5?

On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:

 Two problems here:

 1) Not well formed XML.  That  inside the test attribute isn't
 allow.  Use lt;
 2) OGNL isn't supported in T5.

 For anything more complicated than a property access, you do the work
 in Java code as a synthetic property.

 On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
  Hi,
  maybe this was reported and solved but here it goes ...
 
  I've tried the tapestry tutorial and it was very nice and easy.
  I've made few mistakes but error reporting is great so it was fixed
 quickly.
 
  the problem occured when I tried to add some of the proposed changes to
 the
  app..
 
  just to test the expressions I added:
  t:comp type=If test=ognl:index5
 
  but I get the error
  Failure parsing template context:WEB-INF/Guess.html: The value of
 attribute
  test associated with an element type t:comp must not contain the ''
  character.
 
  Am I doing it wrong, or it's just a bug to be fixed later
 
  Davor Hrg
 


 --
 Howard M. Lewis Ship
 TWD Consulting, Inc.
 Independent J2EE / Open-Source Java Consultant
 Creator and PMC Chair, Apache Tapestry
 Creator, Apache HiveMind

 Professional Tapestry training, mentoring, support
 and project work.  http://howardlewisship.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]






--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: [T5] ?. operator for null-safe graph traversal

2007-03-06 Thread Jesse Kuhnert

Nice. I like that and the 0... that you've added. Why can't we
replicate you for work on OGNL ? =p

On 3/6/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:

customer?.address?.street

I've borrowed some syntax from Groovy to handle this case; the ?.
operator returns null if the LHS is null, otherwise is traverses into
the property on the RHS.

On 3/6/07, Angelo Luis [EMAIL PROTECTED] wrote:
 No no no...

 How can have a null handler now???

 If a have a customer - address - street in a text component, for example 
 custumer.address.street and i dont instanciate a address, a
 NullPointerException is throw. With a ognl null handler, if the address is
 null he creates one , like Webwork... i dont need to instanciate all object
 graph to use in a page...

 i will have this same problem in t5?

 On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:
 
  Two problems here:
 
  1) Not well formed XML.  That  inside the test attribute isn't
  allow.  Use lt;
  2) OGNL isn't supported in T5.
 
  For anything more complicated than a property access, you do the work
  in Java code as a synthetic property.
 
  On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
   Hi,
   maybe this was reported and solved but here it goes ...
  
   I've tried the tapestry tutorial and it was very nice and easy.
   I've made few mistakes but error reporting is great so it was fixed
  quickly.
  
   the problem occured when I tried to add some of the proposed changes to
  the
   app..
  
   just to test the expressions I added:
   t:comp type=If test=ognl:index5
  
   but I get the error
   Failure parsing template context:WEB-INF/Guess.html: The value of
  attribute
   test associated with an element type t:comp must not contain the ''
   character.
  
   Am I doing it wrong, or it's just a bug to be fixed later
  
   Davor Hrg
  
 
 
  --
  Howard M. Lewis Ship
  TWD Consulting, Inc.
  Independent J2EE / Open-Source Java Consultant
  Creator and PMC Chair, Apache Tapestry
  Creator, Apache HiveMind
 
  Professional Tapestry training, mentoring, support
  and project work.  http://howardlewisship.com
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 



--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: [T5] ?. operator for null-safe graph traversal

2007-03-06 Thread Howard Lewis Ship

It is definately a twisty path, before long I'll have reproduced all of OGNL.

It might be fun to support low..high syntax for properties rather
than constants, but that's getting to be as far as I want to take it.
Do the heavy lifting in Java (or Groovy) code.

On 3/6/07, Jesse Kuhnert [EMAIL PROTECTED] wrote:

Nice. I like that and the 0... that you've added. Why can't we
replicate you for work on OGNL ? =p

On 3/6/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:
 customer?.address?.street

 I've borrowed some syntax from Groovy to handle this case; the ?.
 operator returns null if the LHS is null, otherwise is traverses into
 the property on the RHS.

 On 3/6/07, Angelo Luis [EMAIL PROTECTED] wrote:
  No no no...
 
  How can have a null handler now???
 
  If a have a customer - address - street in a text component, for example 
  custumer.address.street and i dont instanciate a address, a
  NullPointerException is throw. With a ognl null handler, if the address is
  null he creates one , like Webwork... i dont need to instanciate all object
  graph to use in a page...
 
  i will have this same problem in t5?
 
  On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:
  
   Two problems here:
  
   1) Not well formed XML.  That  inside the test attribute isn't
   allow.  Use lt;
   2) OGNL isn't supported in T5.
  
   For anything more complicated than a property access, you do the work
   in Java code as a synthetic property.
  
   On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
Hi,
maybe this was reported and solved but here it goes ...
   
I've tried the tapestry tutorial and it was very nice and easy.
I've made few mistakes but error reporting is great so it was fixed
   quickly.
   
the problem occured when I tried to add some of the proposed changes to
   the
app..
   
just to test the expressions I added:
t:comp type=If test=ognl:index5
   
but I get the error
Failure parsing template context:WEB-INF/Guess.html: The value of
   attribute
test associated with an element type t:comp must not contain the ''
character.
   
Am I doing it wrong, or it's just a bug to be fixed later
   
Davor Hrg
   
  
  
   --
   Howard M. Lewis Ship
   TWD Consulting, Inc.
   Independent J2EE / Open-Source Java Consultant
   Creator and PMC Chair, Apache Tapestry
   Creator, Apache HiveMind
  
   Professional Tapestry training, mentoring, support
   and project work.  http://howardlewisship.com
  
   -
   To unsubscribe, e-mail: [EMAIL PROTECTED]
   For additional commands, e-mail: [EMAIL PROTECTED]
  
  
 


 --
 Howard M. Lewis Ship
 TWD Consulting, Inc.
 Independent J2EE / Open-Source Java Consultant
 Creator and PMC Chair, Apache Tapestry
 Creator, Apache HiveMind

 Professional Tapestry training, mentoring, support
 and project work.  http://howardlewisship.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: [T5] ?. operator for null-safe graph traversal

2007-03-06 Thread Jesse Kuhnert

Yeah...Thinking about it more now I realize that you don't really have
a choice. I'm still wondering how much mileage I'm liable to get out
of all of these OGNL changes if there is a better alternative..(ie if
groovy works just as well for property set/get operations ? ...)

I may send out an email to a few people to explore around first. There
are so many cool things to play with I don't want to waste my time if
there is an obvious alternative. (like maybe T5 itself, though I'm
even more wary of that right now - mostly because I think I'd be
getting in the way / distracting when some portions are still
forming...too many cooks can destroy a meal / er something... ;) )

On 3/6/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:

It is definately a twisty path, before long I'll have reproduced all of OGNL.

It might be fun to support low..high syntax for properties rather
than constants, but that's getting to be as far as I want to take it.
Do the heavy lifting in Java (or Groovy) code.

On 3/6/07, Jesse Kuhnert [EMAIL PROTECTED] wrote:
 Nice. I like that and the 0... that you've added. Why can't we
 replicate you for work on OGNL ? =p

 On 3/6/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:
  customer?.address?.street
 
  I've borrowed some syntax from Groovy to handle this case; the ?.
  operator returns null if the LHS is null, otherwise is traverses into
  the property on the RHS.
 
  On 3/6/07, Angelo Luis [EMAIL PROTECTED] wrote:
   No no no...
  
   How can have a null handler now???
  
   If a have a customer - address - street in a text component, for example 

   custumer.address.street and i dont instanciate a address, a
   NullPointerException is throw. With a ognl null handler, if the address is
   null he creates one , like Webwork... i dont need to instanciate all 
object
   graph to use in a page...
  
   i will have this same problem in t5?
  
   On 3/5/07, Howard Lewis Ship [EMAIL PROTECTED] wrote:
   
Two problems here:
   
1) Not well formed XML.  That  inside the test attribute isn't
allow.  Use lt;
2) OGNL isn't supported in T5.
   
For anything more complicated than a property access, you do the work
in Java code as a synthetic property.
   
On 3/5/07, Davor Hrg [EMAIL PROTECTED] wrote:
 Hi,
 maybe this was reported and solved but here it goes ...

 I've tried the tapestry tutorial and it was very nice and easy.
 I've made few mistakes but error reporting is great so it was fixed
quickly.

 the problem occured when I tried to add some of the proposed changes 
to
the
 app..

 just to test the expressions I added:
 t:comp type=If test=ognl:index5

 but I get the error
 Failure parsing template context:WEB-INF/Guess.html: The value of
attribute
 test associated with an element type t:comp must not contain the 
''
 character.

 Am I doing it wrong, or it's just a bug to be fixed later

 Davor Hrg

   
   
--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind
   
Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com
   
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
   
   
  
 
 
  --
  Howard M. Lewis Ship
  TWD Consulting, Inc.
  Independent J2EE / Open-Source Java Consultant
  Creator and PMC Chair, Apache Tapestry
  Creator, Apache HiveMind
 
  Professional Tapestry training, mentoring, support
  and project work.  http://howardlewisship.com
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 


 --
 Jesse Kuhnert
 Tapestry/Dojo team member/developer

 Open source based consulting work centered around
 dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




--
Howard M. Lewis Ship
TWD Consulting, Inc.
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Apache HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





--
Jesse Kuhnert
Tapestry/Dojo team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind. http://blog.opencomponentry.com

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional 

T5: Acegi support/integration anyone?

2007-03-06 Thread Stephan Schwab

Hi,

is anyone working on Acegi integration into Tapestry 5?

Stephan

--
Blog: http://www.stephan-schwab.com
-- 
View this message in context: 
http://www.nabble.com/T5%3A-Acegi-support-integration-anyone--tf3358378.html#a9341261
Sent from the Tapestry - User mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Tap 4.0.3 HttpSession and WebSession

2007-03-06 Thread Miguel Angel Hernández

Hi all,

Anybody knows how I can access the HttpSession from tapestry... I found
RequestCycle: getRequestContext():getSession()... but its
deprecated.

I need to invalidate the session in order to logout from JAAS realm, I've
tried WebSession():invalidate() but it doesn't work :(


Re: Tap 4.0.3 HttpSession and WebSession

2007-03-06 Thread Renat Zubairov

You can try to use Restart service. It should be pretty simple to call
a restart service via service link.
Otherwise you can have a look to the

http://tapestry.apache.org/tapestry4/tapestry/hivedocs/module/tapestry.globals.html

For example you can get a request injected to your Hivemind services.
Or via WebRequest you will get a WebSession

http://tapestry.apache.org/tapestry4/tapestry/apidocs/org/apache/tapestry/web/WebRequest.html#getSession(boolean)

There you will find exactly what you are looking for :)

http://tapestry.apache.org/tapestry4/tapestry/apidocs/org/apache/tapestry/web/WebSession.html#invalidate()



On 06/03/07, Miguel Angel Hernández [EMAIL PROTECTED] wrote:

Hi all,

Anybody knows how I can access the HttpSession from tapestry... I found
RequestCycle: getRequestContext():getSession()... but its
deprecated.

I need to invalidate the session in order to logout from JAAS realm, I've
tried WebSession():invalidate() but it doesn't work :(




--
Best regards,
Renat Zubairov


Re: Tap 4.0.3 HttpSession and WebSession

2007-03-06 Thread Miguel Angel Hernández

Renat Thanks a lot,

Unfortunately I've already tried those... but it's not working. I'm using
Geronimo as AppServer... maybe it's the configuration of my realm or
something...

I'll make you know if I find out what is it.

regards,

migs

On 3/6/07, Renat Zubairov [EMAIL PROTECTED] wrote:


You can try to use Restart service. It should be pretty simple to call
a restart service via service link.
Otherwise you can have a look to the


http://tapestry.apache.org/tapestry4/tapestry/hivedocs/module/tapestry.globals.html

For example you can get a request injected to your Hivemind services.
Or via WebRequest you will get a WebSession


http://tapestry.apache.org/tapestry4/tapestry/apidocs/org/apache/tapestry/web/WebRequest.html#getSession(boolean)

There you will find exactly what you are looking for :)


http://tapestry.apache.org/tapestry4/tapestry/apidocs/org/apache/tapestry/web/WebSession.html#invalidate()



On 06/03/07, Miguel Angel Hernández [EMAIL PROTECTED] wrote:
 Hi all,

 Anybody knows how I can access the HttpSession from tapestry... I found
 RequestCycle: getRequestContext():getSession()... but its
 deprecated.

 I need to invalidate the session in order to logout from JAAS realm,
I've
 tried WebSession():invalidate() but it doesn't work :(



--
Best regards,
Renat Zubairov



T5:Grid with Checkbox

2007-03-06 Thread Weisu

Hi, has anyone uses checkbox inside a grid, i just wonder how to do it in T5.
Thanks
-- 
View this message in context: 
http://www.nabble.com/T5%3AGrid-with-Checkbox-tf3360245.html#a9347153
Sent from the Tapestry - User mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: AW: AW: @Shell component, refresh and IE

2007-03-06 Thread Patrick Klein

Hello Peter,

i tried that now by using a custom Shell component which simply 
replaces ; with %3B in the url. Result in firefox _and_ IE :
  
   HTTP Status 404 - 
/myapp/ImportMatrix.page%3Bjsessionid=E903773F2AFE3C2F2AADB2EE2D1C0C3E


Would've been too easy if it had worked that way...

Does anyone have another idea for a workaround?

Regards,
   Patrick

hi patrick,

you may also try to url-encode the tapestry-url so that ; will be %3B and IE 
wont get confused

kind regards 


-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:46

An: Tapestry users
Betreff: Re: AW: @Shell component, refresh and IE

Hi, Peter,

somehow i allready thought it was something like that. Thanks for the 
answer :)

Switching to using cookies also does the trick.
Both solutions somehow are not really satisfying as the Shell component 
should be able to attend this...


Regards,
Patrick
  

hi patrick,

this might be because of the ; as delimiter for the sessionId wich is also 
used as delimiter for the values of the content-parameter. so i think that IE splits this 
entry to
10
URL=http://localhost:8080/myapp/ImportMatrix.page
jsessionid=C79B17BAFCE7166D80F92708F500D4B9

wich brings you to your start-page. 
instead of a meta-tag you might use javascript for page-refresh as a workaround.


kind regars

-Ursprüngliche Nachricht-
Von: Patrick Klein [mailto:[EMAIL PROTECTED] 
Gesendet: Dienstag, 6. März 2007 10:20

An: Tapestry users
Betreff: @Shell component, refresh and IE

Hello!

I'm using the @Shell component's refresh tag to set trigger the reload 
of a given page every 10 seconds. All works fine using Firefox. However, 
using IE all goes havoc as IE seems to drop at least parts of the reload 
url. The refresh part of the page looks like the following:


meta http-equiv=Refresh content=10; 
URL=http://localhost:8080/myapp/ImportMatrix.page;jsessionid=C79B17BAFCE7166D80F92708F500D4B9; 
/


after the refresh i'm seeing the following url in IE:
   
http://localhost:8080/myapp/ImportMatrix.page


completely disregarding the jsessionid, which brings me back to the 
login page.


Used Tapestry-version: 4.1.1, 4.1.2-SNAPSHOT (15.2.2007)

Could anyone give me a hint on how to fix this?

Regards,
Patrick

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


  




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


  



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]