date:20090611


On 
http://code.google.com/intl/sv/webtoolkit/doc/1.6/DevGuideOrganizingProjects.html
we can read the following about static content, specifically in the
example an image called logo.gif

An image file available to the application code. You might load this
file programmatically using this URL: GWT.getModuleBaseURL() +
logo.gif.

I have a project called ArrivalJournal which I´m building in
Eclipse. I´m running Jboss 4.2.2 on localhost for web server.

Calling GWT.getModuleBaseURL() returns http://localhost:8080/
ArrivalJournal/arrivaljournal/

But if I put any content what so ever inside my war\arrivaljournal
direcory, it gets deleted when I compile my project. What to do?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Where to put static content?




On 11 juin, 10:02, Dalla dalla_man...@hotmail.com wrote:
 Onhttp://code.google.com/intl/sv/webtoolkit/doc/1.6/DevGuideOrganizingP...
 we can read the following about static content, specifically in the
 example an image called logo.gif

 An image file available to the application code. You might load this
 file programmatically using this URL: GWT.getModuleBaseURL() +
 logo.gif.

 I have a project called ArrivalJournal which I´m building in
 Eclipse. I´m running Jboss 4.2.2 on localhost for web server.

 Calling GWT.getModuleBaseURL() returns http://localhost:8080/
 ArrivalJournal/arrivaljournal/

 But if I put any content what so ever inside my war\arrivaljournal
 direcory, it gets deleted when I compile my project. What to do?

You have two places where to put your static content:
 - in the output folder (war); use GWT.getHostPageBaseURL() as a
prefix (or eventually no prefix at all, but I think using the host
page base URL is safer...)
 - in the public subpackage of your module (where the name public
can be overriden if you want); everything in there will be copied into
war/arrivaljournal, so you should use GWT.getModuleBaseURL() as a
prefix (in case you change the name of your module or its rename-to,
you won't have to change your code).

The advantage of public is that your resources are tied to your
module; and if its a reusable module (rather than a full-fledge
application for instance), you'll bring your resource with it with a
simple inherits (that's what the themes in GWT do)
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Question about SuggestBox.wrap




On 11 juin, 00:38, Ben benzhe...@gmail.com wrote:
 I have a text box insides a FlexTable and I am trying to use
 SuggestBox.wrap to wrap the text box into a SuggestBox. Code snippet
 is like this.

 .
 FlexTable ft = new FlexTable();
 TextBox tb = new TextBox();

 ft.setWidget(0,0,tb);

 MultiWordSuggestOracle oracle = new MultiWordSuggestOracle();

 suggest = SuggestBox.wrap(oracle, tb.getElement());
 .

 But it throws error in hosted mode like this:

 java.lang.AssertionError: A widget that has an existing parent widget
 may not be added to the detach list
         at com.google.gwt.user.client.ui.RootPanel.detachOnWindowClose
 (RootPanel.java:122)
         at com.google.gwt.user.client.ui.SuggestBox.wrap(SuggestBox.java:
 267)...

 But it does not have this problem when you compile it and run it in
 browsers. I look into the source code of SuggestBox, it throws error
 in this line.

  266     suggestBox.onAttach();
  267     RootPanel.detachOnWindowClose(suggestBox);

 What is the problem? Is there a way to solve it?

Do not use wrap(), use the ctor instead.

FlexTable ft = new FlexTable();
TextBox tb = new TextBox();
MultiWordSuggestOracle oracle = new MultiWordSuggestOracle();
suggest = new SuggestBox(oracle, tb);

ft.setWidget(0,0,suggest);

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Is it possible to call the GWT Service outside the entry point?

2009-06-11 Thread Jonathan Kushner

As long as you're obiding by the rules of GWT client-side handling, you can
call anywhere you'd like within the client package.
It looks like your experience a rendering versus service calling effect. I
would review the documentations on best practices for creating a GWT
Application with RPC's. Something in the logic isn't accurate, but your
question over RPC calling is fine as long as it remains within the client
domain.

Good luck.


On Wed, Jun 10, 2009 at 10:08 PM, retha pasalli re...@wacana-inti.comwrote:


 Hi everybody...

 I want to know whether I can call the AsyncCallback outside the entry
 point. I have tried to create a class named UserServiceCall to call
 all service to retrieve and insert/edit/delete User data. The class
 contain the code to call the service (AsyncCallback object with
 onFailure() and onSuccess process). Then I keep the result object in
 the attribute of that class. In entry point, I create an object of
 UserServiceCall class and call the service caller method and get the
 object by the attribute getter method of UserServiceCall class. But
 when I run the project, it shows me nothing in the html page. Can you
 help with this problem?

 Thx before...

 Regards,
 Retha

 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Server initialization

2009-06-11 Thread Steve


where you define your servlet in web.xml add load-on-startup1/load-
on-startup

servlet
servlet-nameMyAppServiceImpl/servlet-name
servlet-classmy.package.app.MyAppServiceImpl/servlet-class
load-on-startup1/load-on-startup
/servlet

On Jun 10, 6:29 pm, Jamie jamiesharbor-sou...@yahoo.com wrote:
 You might try using the servlet configuration 'load-on-startup'
 setting in your web.xml servlet definition.

 You could also define your own base class and derive your servlets
 from that so you at least have the code in one place...

 Jamie.
 ---
 Search for analog and digital television broadcast antennas in your
 area:http://www.antennamap.com/
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: ImageBundle reuse images




On 10 juin, 23:44, mrpantsuit kevin.peter.w...@gmail.com wrote:
 This might be a stupid question, but can I use the same Image created
 from an ImageBundle in multiple places on my page?  I guess the more
 general question is can I use the same Image in multiple places in my
 page.  I assume so.

Image is a widget, so you can't reuse the same Image object (it'll be
automatically removed from where it was before being added).
But you can of course reuse the AbstractImagePrototype and call
getImage() each time you need the Image.

...or you could use getHTML() and reuse it at will with setHTML on the
appropriate widgets (not being an Image widget, you won't have event
handlers though)
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Server initialization

2009-06-11 Thread Jason Morris


Personally I would say use a ServletContextListener if the structures you are 
creating will be 
shared by several of your Servlets. That way the init is finished before any of 
your Servlets are 
created.

Load-on-startup always feels like a bit of a hack to me ;)

Just my 2c

Steve wrote:
 where you define your servlet in web.xml add load-on-startup1/load-
 on-startup
 
 servlet
 servlet-nameMyAppServiceImpl/servlet-name
 servlet-classmy.package.app.MyAppServiceImpl/servlet-class
 load-on-startup1/load-on-startup
 /servlet
 
 On Jun 10, 6:29 pm, Jamie jamiesharbor-sou...@yahoo.com wrote:
 You might try using the servlet configuration 'load-on-startup'
 setting in your web.xml servlet definition.

 You could also define your own base class and derive your servlets
 from that so you at least have the code in one place...

 Jamie.
 ---
 Search for analog and digital television broadcast antennas in your
 area:http://www.antennamap.com/
  
 

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: GWT Support for other browsers




On 10 juin, 19:24, Jeff Chimene jchim...@gmail.com wrote:
 If you can't get the browser to send an alternate ident string, you
 might try hacking the *.nocache.js to handle the Galio browser ident
 string and instead reply with either the IE or Firefox branch to the
 client. Compile your GWT code as PRETTY to get code that's easier to
 understand.

It's far easier to override the property-provider
name=user.agent: copy it from UserAgent.gwt.xml (within gwt-
user.jar) into your own *.gwt.xml and tweak it so that Galio is
identified as another supported browser.

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Where to put static content?


Oh... guess you just create a folder rather than a package for the
resources...

On 11 Juni, 11:23, Dalla dalla_man...@hotmail.com wrote:
 So if I get you right, I should have

 com/example/app/arrivaljournal/client,  Sub-package containing all the
 client-side source code.
 com/example/app/arrivaljournal/server, Sub-package containing all the
 server-side source code
 com/example/app/arrivaljournal/public, Static resources

 But in Eclipse, when I try to create com/example/app/arrivaljournal/
 public
 I get : 'public' is not a vaild java identifier

 Is that just an Eclipse bug then?

 On 11 Juni, 10:44, Thomas Broyer t.bro...@gmail.com wrote:



  On 11 juin, 10:02, Dalla dalla_man...@hotmail.com wrote:

   Onhttp://code.google.com/intl/sv/webtoolkit/doc/1.6/DevGuideOrganizingP...
   we can read the following about static content, specifically in the
   example an image called logo.gif

   An image file available to the application code. You might load this
   file programmatically using this URL: GWT.getModuleBaseURL() +
   logo.gif.

   I have a project called ArrivalJournal which I´m building in
   Eclipse. I´m running Jboss 4.2.2 on localhost for web server.

   Calling GWT.getModuleBaseURL() returns http://localhost:8080/
   ArrivalJournal/arrivaljournal/

   But if I put any content what so ever inside my war\arrivaljournal
   direcory, it gets deleted when I compile my project. What to do?

  You have two places where to put your static content:
   - in the output folder (war); use GWT.getHostPageBaseURL() as a
  prefix (or eventually no prefix at all, but I think using the host
  page base URL is safer...)
   - in the public subpackage of your module (where the name public
  can be overriden if you want); everything in there will be copied into
  war/arrivaljournal, so you should use GWT.getModuleBaseURL() as a
  prefix (in case you change the name of your module or its rename-to,
  you won't have to change your code).

  The advantage of public is that your resources are tied to your
  module; and if its a reusable module (rather than a full-fledge
  application for instance), you'll bring your resource with it with a
  simple inherits (that's what the themes in GWT do)- Dölj citerad text -

  - Visa citerad text -- Dölj citerad text -

 - Visa citerad text -
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Possible to insert a widget into the DOM?

2009-06-11 Thread gscholt


Hi,

Widgets are added to widgets, elements are added to elements.
If you want to put a widget into the dom, I'd use a RootPanel, since
it will take care of the unloading that is needed:

p = RootPanel.get(currentElement.getId());
p.add(someWidget);

if currentElement doesn't have an Id, you could use
  currentElement.setId(DOM.createUniqueId());
to set one.

Note that these are Elements, not Nodes; not every node is an element,
and not every node is allowed to have children (which is probably why
the appendChild fails, but without more information then an
exception it is hard to be sure).

Gert

On Jun 10, 11:51 pm, peterk peter.ke...@gmail.com wrote:
 Hey Jason,

 Thanks for your reply!

 I'm afraid I'm still running into problems here. First, even just
 trying to insert an element. I have this code:

 SpanElement test_element = Document.get().createSpanElement();
 test_element.setInnerText(Hello there);
 currentNode.appendChild(test_element);

 currentNode is of type Node. This compiles OK, but at run time I get
 an exception on the appendChild call.

 How can I add or append an element to a node like this?

 Secondly..I can't seem to find any wrap method in the Widget class?

 Any further insight would be greatly appreciated..thanks so much!

 On Jun 10, 8:41 pm, Jason Essington jason.essing...@gmail.com wrote:

  Probably the safest thing to do would be to insert an element, then  
  use the widget's .wrap() method to promote that element to a widget.

  there are other ways, but they are prone to leaking memory.

  -jason

  On Jun 10, 2009, at 1:36 PM, peterk wrote:

   Hey all,

   I'm walking the DOM tree and would like to append a Widget to certain
   nodes where certain conditions are met, but I'm not having much
   success.

   appendChild expects a Node as its parameter, so I can't just append a
   type of Widget. It will accept Widget.getElement() as a parameter, but
   this leads to runtime errors.

   Is it possible to mix Widgets with the DOM like this or is the only
   way to add Widgets to a page via the RootPanel.get() ..? If so is
   there a way to add Widgets to the page based on a desired position in
   the DOM?

   I'd like to use a custom widget I've made, having everything wrapped
   neatly in a class, and not have to revert to lower level stuff..

   Thanks for any help :)
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Possible to insert a widget into the DOM?

2009-06-11 Thread peterk


Hey Gert,

Thanks for the reply! I realised my problem with appending an
Element..I was trying to append it to a TextNode which I think isn't
allowed. So I've sorted that, I append to the text node's Parent now.

On adding Widgets to a certain point in the DOM..this seems to work
for me:

I create the Widget, and add it to RootPanel. Then, I get its element
and append that to the appropriate Node. This moves the widget to the
correct place in the DOM, and at first glance I get the behaviours I
would expect (e.g. the Widget's clicklisteners work etc.).

Is it safe to do things this way or are there other 'gotchas' I should
be aware of, when manipulating the position of a Widget's element in
the page?

Thanks again for all your help!

On Jun 11, 10:48 am, gscholt gsch...@gmail.com wrote:
 Hi,

 Widgets are added to widgets, elements are added to elements.
 If you want to put a widget into the dom, I'd use a RootPanel, since
 it will take care of the unloading that is needed:

 p = RootPanel.get(currentElement.getId());
 p.add(someWidget);

 if currentElement doesn't have an Id, you could use
   currentElement.setId(DOM.createUniqueId());
 to set one.

 Note that these are Elements, not Nodes; not every node is an element,
 and not every node is allowed to have children (which is probably why
 the appendChild fails, but without more information then an
 exception it is hard to be sure).

 Gert

 On Jun 10, 11:51 pm, peterk peter.ke...@gmail.com wrote:



  Hey Jason,

  Thanks for your reply!

  I'm afraid I'm still running into problems here. First, even just
  trying to insert an element. I have this code:

  SpanElement test_element = Document.get().createSpanElement();
  test_element.setInnerText(Hello there);
  currentNode.appendChild(test_element);

  currentNode is of type Node. This compiles OK, but at run time I get
  an exception on the appendChild call.

  How can I add or append an element to a node like this?

  Secondly..I can't seem to find any wrap method in the Widget class?

  Any further insight would be greatly appreciated..thanks so much!

  On Jun 10, 8:41 pm, Jason Essington jason.essing...@gmail.com wrote:

   Probably the safest thing to do would be to insert an element, then  
   use the widget's .wrap() method to promote that element to a widget.

   there are other ways, but they are prone to leaking memory.

   -jason

   On Jun 10, 2009, at 1:36 PM, peterk wrote:

Hey all,

I'm walking the DOM tree and would like to append a Widget to certain
nodes where certain conditions are met, but I'm not having much
success.

appendChild expects a Node as its parameter, so I can't just append a
type of Widget. It will accept Widget.getElement() as a parameter, but
this leads to runtime errors.

Is it possible to mix Widgets with the DOM like this or is the only
way to add Widgets to a page via the RootPanel.get() ..? If so is
there a way to add Widgets to the page based on a desired position in
the DOM?

I'd like to use a custom widget I've made, having everything wrapped
neatly in a class, and not have to revert to lower level stuff..

Thanks for any help :)- Hide quoted text -

 - Show quoted text -
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Is it possible to call the GWT Service outside the entry point?

2009-06-11 Thread retha pasalli


Hi Mr. Kushner,

Thx for your reply. I've tried to call the service from a class
extending Composite class. And it works! But I still can't find why it
doesn't work when I call it in class extending nothing. But that's not
an urgent problem for me for a while. Hehehe..

Once more thx...

Regards,
Retha
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Server initialization

2009-06-11 Thread Keith Whittingham

Yup. That does the ticket!

Here's the code if anyone's interested...

The listener...

package com.inexas.test.server;

import javax.servlet.*;

public class AppStarter implements ServletContextListener {

public void contextInitialized(ServletContextEvent event) {
ServletContext sc = event.getServletContext();
System.out.println(sc.getServerInfo())  ;
}

public void contextDestroyed(ServletContextEvent event) {
System.out.println(Bye!);
}
}

In the web.xml file...

web-app

listener
listener-class
com.inexas.test.server.AppStarter
/listener-class
/listener
...




On Jun 11, 2009, at 11:21 AM, Jason Morris wrote:

 Personally I would say use a ServletContextListener if the  
 structures you are creating will be
 shared by several of your Servlets. That way the init is finished  
 before any of your Servlets are
 created.

 Load-on-startup always feels like a bit of a hack to me ;)

 Just my 2c


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: LoginSecurityFAQ and sessionID/tokens

2009-06-11 Thread Paul Robinson


If you store the session ID in a cookie so that user logins can persist
beyond browser refreshes (as suggested in the FAQ), then the session ID
will end up in the header anyway.

eags wrote:
 I am implementing user logins and authentication using the model
 presented in the login security FAQ.  In particular I plan on manually
 maintaining a table of {sessionID,User,timeout} values for each active
 session and not using the normal servlet session functionality.

 So, my question is, where do I get the ID that is returned to the
 client?  I know that I can get one from the servlet session using
 HttpServletRequest.getSession().getid() but it seems like I could just
 use any randomly generated key right?  And maybe I if face should not
 use that technique because that sessionID is also in the header where
 it can be easily snooped right?  So, what is a good technique for
 generating the sessionID?  To avoid duplicates I would just check the
 sessionID table before returning the sessionID to the client and if it
 is already in use I just call generateSessionID() again.  So my
 question is what should getSessionID() look like?

 I realize the recommended approach in the LoginSecurityFAQ is
 controversial and I've already read all that debate so I'm not really
 interested in more of that.  I just need specific help regarding these
 questions assuming I am doing what is recommended in the FAQ.

 Thanks in advance for any help.
 

   

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

This is pretty interesting...

2009-06-11 Thread chhatrapal . sharma


   
Hi !
Your friend chhatrapal.sharma from  , just invited you to  his/her  online 
photo albums and journals at Zorpia.com.



So what is Zorpia?
It is an online community that allows you to upload unlimited amount of photos, 
write journals and make friends. We also have a variety of skins in store for 
you so that you can customize your homepage freely.

Join now for free! Please click the following link to join Zorpia:
http://signup.zorpia.com/signup?invitation_key=200906af6eeda93b3c9f562eb817a2dbreferral=chhatrapalsharmatt=14hour=2009061107sub=13

This message was delivered with the chhatrapal.sharma's initiation.

If you wish to discontinue receiving invitations from us, please click the 
following link:
http://signup.zorpia.com/email/optout/Google-Web-Toolkit@googlegroups.com


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Help with EventHandlers

2009-06-11 Thread Kelo


I'm glad it works fine on your project.

Saludos, Marcelo.


On Jun 10, 6:43 pm, Iván López lopez.i...@gmail.com wrote:
    Thank you very much!. It works ok and now, with my code working I
 think I understand a little bit better how handlers works!.

    Muchas gracias compañero.

    Saludos, Iván.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Event handler for model ?


No clues anyone?

On 10 Juni, 14:40, Dalla dalla_man...@hotmail.com wrote:
 Hi

 I´m having trouble understanding how to implement an event handler for
 my custom objects with the new event handler system. Let´s pretend I
 have a very simple model class like this one:

 public class ArrivalData {
         private Date date;
         private boolean processing = false;

         public Date getDate() {
                 return date;
         }

         public void setDate(Date date) {
                 this.date = date;
         }

 }

 When setDate is called, I want to fire an event that my view
 components can listen on, and act on when called for.
 How would you go about doing that? Are there any existing handler
 interfaces that can be used?
 I found this tutorial (see link below), but it seems pretty
 complicated for what I´m trying to 
 acheive.http://www.itsolut.com/chrismusings/2009/04/28/business-events-with-g...

 What´s best practice here?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: no createDocumentFragment() in com.google.gwt.dom.client.Document

2009-06-11 Thread andrej



I'm running off trunk right now. There's no
Document.createDocumentFragment() method either. Just wondering why
that is.

Could someone comment on issue 2955? --
http://code.google.com/p/google-web-toolkit/issues/detail?id=2955
The issue was created almost a year ago, but it's still in NEW status.

(I just tried out OOPHM and it rocks! Thank you gwt team!)
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: This is pretty interesting...

2009-06-11 Thread Ian Bambury

If your friend (an invisible friend?) wants me on their list, why don't they
ask me themselves?
Preferably not via the GWT group.

Ian

http://examples.roughian.com


2009/6/11 chhatrapal.sharma chhatrapal.sha...@gmail.com

  Hi ,
 I started using this and love it. If you use the link below you'll
 automatically be on my friend's list.

 http://signup.zorpia.com/signup?invitation_key=200906af6eeda93b3c9f562eb817a2dbreferral=chhatrapalsharmatt=14hour=2009061107sub=13
 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Dialog Box Move Listener kind of thing

2009-06-11 Thread abhiram


Hi,

  Is there any way by which I could know if my Dialog Box has moved? I
mean at run time thru some listener or something? I am trying to find
out the position of the Dialog Box when it has moved for which I need
to know when it moved!!!

regards,
abhiram
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

gwt compiler... parameterized types with enums causing problems?

2009-06-11 Thread Ian Smith


The following class appears to be legal java (eclipse accepts it ok)
but it causes an internal error in the gwt compiler (java -
javascript).  Has anyone found a workaround to allow code like this to
run?

If you try to instantiate this class with
public enum MyFormField {
  FIELD1, OTHER_FIELD;
};

and then this declaration:

public class MyFormErrorSet extends ErrorSetMyFormField {
};

the compiler blows up with an internal error Thoughts?

thanks
ian smith




package whatever.whatever;

import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;

public abstract class ErrorSetE extends EnumE {
protected ListE errors = new ArrayListE();

public boolean hasNoErrors() {
return errors.isEmpty();
}

public E errorAtRow(int rowTarget) {
IteratorE iter = errors.iterator();
while (iter.hasNext()) {
E f = iter.next();
if (getRowForEnumItem(f) == rowTarget) {
return f;
}
}
return null;
}

public void add(E f) {
if (errorAtRow(getRowForEnumItem(f)) != null) {
return;
}
errors.add(f);
}

public int size() {
return errors.size();
}

public E get(int element) {
return errors.get(element);
}

public abstract int getRowForEnumItem(E item);
}

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Gears Automatically generated Google Gears manifest and rpc files

2009-06-11 Thread Tibo


Hello,

I am currently developping an app using both GWT and GAE. I'd like
this app to use Gears mostly to used offline and also to improve
performance.

Taking first steps with gwt-gears API I found quite convenient to have
an automatically generated manifest for files needed to be downloaded.

However I am using RPC and the Gears manifest generation includes the
gwt.rpc files to the list of files to download. Unless I missed
something, these files are useless to Gears localserver and over all
the default configuration of App Engine doesn't allow to serve gwt.rpc
files which results in a HTTP 404 error when trying to download them.

I would like to know what solution could be found.

I tried to let the rpc files being served which allowed Gears to
dowload all the files listed in the manifest but I guess this would be
quite unsecure to let this that way.

Could we change the folder where the rpc files are generated ?

I guess it is possible to make the manifest file manually but it seems
to me it goes against a certain spirit of convenience GWT brought to
me.

Note that even with all files donwloaded I still have trouble in
having my local server being able to serve. But this will be for
later.

Thanks for your interest. :)

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Gears Automatically generated Google Gears manifest and rpc files

2009-06-11 Thread Eric Ayers


My thought on this is that we should modify the manifest generator to
filter out these files, and also add a config option to allow users to
specify files to exclude in case something else like this comes up -
maybe through (yet another) annotation on the Gadget class?

On Thu, Jun 11, 2009 at 5:19 AM, Tibotpou...@gmail.com wrote:

 Hello,

 I am currently developping an app using both GWT and GAE. I'd like
 this app to use Gears mostly to used offline and also to improve
 performance.

 Taking first steps with gwt-gears API I found quite convenient to have
 an automatically generated manifest for files needed to be downloaded.

 However I am using RPC and the Gears manifest generation includes the
 gwt.rpc files to the list of files to download. Unless I missed
 something, these files are useless to Gears localserver and over all
 the default configuration of App Engine doesn't allow to serve gwt.rpc
 files which results in a HTTP 404 error when trying to download them.

 I would like to know what solution could be found.

 I tried to let the rpc files being served which allowed Gears to
 dowload all the files listed in the manifest but I guess this would be
 quite unsecure to let this that way.

 Could we change the folder where the rpc files are generated ?

 I guess it is possible to make the manifest file manually but it seems
 to me it goes against a certain spirit of convenience GWT brought to
 me.

 Note that even with all files donwloaded I still have trouble in
 having my local server being able to serve. But this will be for
 later.

 Thanks for your interest. :)

 




-- 
Eric Z. Ayers - GWT Team - Atlanta, GA USA
http://code.google.com/webtoolkit/

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Unit testing GWT and App Engine in Eclipse

2009-06-11 Thread Johan Lundberg


Thanks Jason.
I thought it might be possible to do the round trip testing with app
engine but this does not seem to be the case. Thanks for the code
sample.
/johan

On Jun 10, 3:56 pm, Jason Parekh jasonpar...@gmail.com wrote:
 Hi Johan,
 I don't think there are any frameworks that can let you test a round-trip
 between the two.  Instead, you'll likely want to test each side individually
 (seehttp://code.google.com/appengine/docs/java/howto/unittesting.htmlfor
 App Engine junit testing).

 If you're using the Eclipse plugin, I have a sample project that tests an
 App Engine project, 
 seehttp://www.wuala.com/jasonparekh/Public/google/beingTested.zip/.

 Hope this helps,
 jason



 On Wed, Jun 10, 2009 at 8:10 AM, Johan Lundberg joe...@gmail.com wrote:

  Have been trying to make a unit test for my GWT and App Engine app but
  there is something I do wrong and I cannot find any documentation
  about this combination.

 http://code.google.com/webtoolkit/tutorials/1.6/appengine.htmlshows
  the relation between gwt and appengine but it doesn't explain how to
  set up the unit tests.

 http://code.google.com/webtoolkit/doc/1.6/DevGuideTesting.htmlis a
  great guide I have been using to make write my first tests, but it
  doesn't explain how I should invoke the local app engine server when
  making a unit test in hosted mode for my gwt app.

  Any suggestions or pointers to guides?
  /johan
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Gears Automatically generated Google Gears manifest and rpc files

2009-06-11 Thread BobV


On Thu, Jun 11, 2009 at 9:43 AM, Eric Ayerszun...@google.com wrote:
 My thought on this is that we should modify the manifest generator to
 filter out these files, and also add a config option to allow users to
 specify files to exclude in case something else like this comes up -
 maybe through (yet another) annotation on the Gadget class?

The gears linker should go ahead and exclude *.gwt.rpc* from being
included in the manifest.  Looking at upcoming changes for GWT 2.0,
such as runAsync, the manifest linker is going to need to get smarter
about generating multiple manifests, one for each permutation.

-- 
Bob Vawter
Google Web Toolkit Team

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Gears Automatically generated Google Gears manifest and rpc files

2009-06-11 Thread Thibault Pouget

It would be great :) !
I was saying at the end of my previous message that I still could get my
local server to serve and I think I found why.

I am also using GXT which needs its image folder to be in the war one. And
looking at the generated manifest, it seems that files outside the
projectname folder are not included in the list.

So I am trying to do my own manifest by hand and to download all needed
files including specific css and images.

And I was thinking that some convenient methods to manually build the
manifest would be awesome. For example, directly including an entire
folder... I am going to try and write these. But for now I 'd like to get my
local serve do what he is intended to. :)

By the way, thanks for the quick reply.

2009/6/11 Eric Ayers zun...@google.com


 My thought on this is that we should modify the manifest generator to
 filter out these files, and also add a config option to allow users to
 specify files to exclude in case something else like this comes up -
 maybe through (yet another) annotation on the Gadget class?

 On Thu, Jun 11, 2009 at 5:19 AM, Tibotpou...@gmail.com wrote:
 
  Hello,
 
  I am currently developping an app using both GWT and GAE. I'd like
  this app to use Gears mostly to used offline and also to improve
  performance.
 
  Taking first steps with gwt-gears API I found quite convenient to have
  an automatically generated manifest for files needed to be downloaded.
 
  However I am using RPC and the Gears manifest generation includes the
  gwt.rpc files to the list of files to download. Unless I missed
  something, these files are useless to Gears localserver and over all
  the default configuration of App Engine doesn't allow to serve gwt.rpc
  files which results in a HTTP 404 error when trying to download them.
 
  I would like to know what solution could be found.
 
  I tried to let the rpc files being served which allowed Gears to
  dowload all the files listed in the manifest but I guess this would be
  quite unsecure to let this that way.
 
  Could we change the folder where the rpc files are generated ?
 
  I guess it is possible to make the manifest file manually but it seems
  to me it goes against a certain spirit of convenience GWT brought to
  me.
 
  Note that even with all files donwloaded I still have trouble in
  having my local server being able to serve. But this will be for
  later.
 
  Thanks for your interest. :)
 
  
 



 --
 Eric Z. Ayers - GWT Team - Atlanta, GA USA
 http://code.google.com/webtoolkit/

 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Gears Automatically generated Google Gears manifest and rpc files

2009-06-11 Thread Eric Ayers


I created an issue for the filtering problem:

  http://code.google.com/p/gwt-google-apis/issues/detail?id=280

Feel free to come over to the gwt-google-apis group and discuss a
proposal for other changes to the gadget API that might help with the
app you are building.

-Eric.

On Thu, Jun 11, 2009 at 9:56 AM, Thibault Pougettpou...@gmail.com wrote:
 It would be great :) !
 I was saying at the end of my previous message that I still could get my
 local server to serve and I think I found why.
 I am also using GXT which needs its image folder to be in the war one. And
 looking at the generated manifest, it seems that files outside the
 projectname folder are not included in the list.
 So I am trying to do my own manifest by hand and to download all needed
 files including specific css and images.
 And I was thinking that some convenient methods to manually build the
 manifest would be awesome. For example, directly including an entire
 folder... I am going to try and write these. But for now I 'd like to get my
 local serve do what he is intended to. :)
 By the way, thanks for the quick reply.

 2009/6/11 Eric Ayers zun...@google.com

 My thought on this is that we should modify the manifest generator to
 filter out these files, and also add a config option to allow users to
 specify files to exclude in case something else like this comes up -
 maybe through (yet another) annotation on the Gadget class?

 On Thu, Jun 11, 2009 at 5:19 AM, Tibotpou...@gmail.com wrote:
 
  Hello,
 
  I am currently developping an app using both GWT and GAE. I'd like
  this app to use Gears mostly to used offline and also to improve
  performance.
 
  Taking first steps with gwt-gears API I found quite convenient to have
  an automatically generated manifest for files needed to be downloaded.
 
  However I am using RPC and the Gears manifest generation includes the
  gwt.rpc files to the list of files to download. Unless I missed
  something, these files are useless to Gears localserver and over all
  the default configuration of App Engine doesn't allow to serve gwt.rpc
  files which results in a HTTP 404 error when trying to download them.
 
  I would like to know what solution could be found.
 
  I tried to let the rpc files being served which allowed Gears to
  dowload all the files listed in the manifest but I guess this would be
  quite unsecure to let this that way.
 
  Could we change the folder where the rpc files are generated ?
 
  I guess it is possible to make the manifest file manually but it seems
  to me it goes against a certain spirit of convenience GWT brought to
  me.
 
  Note that even with all files donwloaded I still have trouble in
  having my local server being able to serve. But this will be for
  later.
 
  Thanks for your interest. :)
 
  
 



 --
 Eric Z. Ayers - GWT Team - Atlanta, GA USA
 http://code.google.com/webtoolkit/




 




-- 
Eric Z. Ayers - GWT Team - Atlanta, GA USA
http://code.google.com/webtoolkit/

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

eval() with ActiveXObject

2009-06-11 Thread JF


Hi,

i have a string that contain a javascript function. Within that
function, there is a ActiveXObject(Microsoft.XMLDOM); object.

When i pass that string to the eval function, i get an error
automation server can't create object. I am using IE 6.

Anyone have an idea ?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Return type of AsyncCallback methods

2009-06-11 Thread Damon Lundin


You can also have your async method return RequestBuilder which will
not dispatch the RPC call, but rather returns the RequestBuilder
already set up and ready to go.  You can then make changes to it
before making the call such as setting a timeout.  Then you dispatch
the RPC call which will return you the Request object you ask for
above.

RequestBuilder builder = service.myMethod(new AsyncCallback
() {...});
builder.setTimeoutMillis(REQUEST_TIMEOUT);
Request request = builder.send();
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: ImageBundle reuse images

2009-06-11 Thread Jason Essington


Images widgets at their core contain a DOM element and you can't have  
an element in two places at once.

However the beauty of ImageBundle is that it caches that actual image  
so you don't incur additional bandwidth when you create multiple image  
widgets using the same image bundle.

-jason

On Jun 10, 2009, at 3:44 PM, mrpantsuit wrote:


 This might be a stupid question, but can I use the same Image created
 from an ImageBundle in multiple places on my page?  I guess the more
 general question is can I use the same Image in multiple places in my
 page.  I assume so.
 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: ImageBundle reuse images

2009-06-11 Thread Kevin Wong


Awesome.  Thanks guys.

On Jun 11, 11:19 am, Jason Essington jason.essing...@gmail.com
wrote:
 Images widgets at their core contain a DOM element and you can't have  
 an element in two places at once.

 However the beauty of ImageBundle is that it caches that actual image  
 so you don't incur additional bandwidth when you create multiple image  
 widgets using the same image bundle.

 -jason

 On Jun 10, 2009, at 3:44 PM, mrpantsuit wrote:





  This might be a stupid question, but can I use the same Image created
  from an ImageBundle in multiple places on my page?  I guess the more
  general question is can I use the same Image in multiple places in my
  page.  I assume so.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Possible to insert a widget into the DOM?

2009-06-11 Thread Jason Essington


You need to make sure that your node is an element.

also .wrap() is not found in widget itself but rather in its  
subclasses (Label for instance)

-jason

On Jun 10, 2009, at 3:51 PM, peterk wrote:


 Hey Jason,

 Thanks for your reply!

 I'm afraid I'm still running into problems here. First, even just
 trying to insert an element. I have this code:

 SpanElement test_element = Document.get().createSpanElement();
 test_element.setInnerText(Hello there);
 currentNode.appendChild(test_element);

 currentNode is of type Node. This compiles OK, but at run time I get
 an exception on the appendChild call.

 How can I add or append an element to a node like this?

 Secondly..I can't seem to find any wrap method in the Widget class?

 Any further insight would be greatly appreciated..thanks so much!

 On Jun 10, 8:41 pm, Jason Essington jason.essing...@gmail.com wrote:
 Probably the safest thing to do would be to insert an element, then
 use the widget's .wrap() method to promote that element to a  
 widget.

 there are other ways, but they are prone to leaking memory.

 -jason

 On Jun 10, 2009, at 1:36 PM, peterk wrote:



 Hey all,

 I'm walking the DOM tree and would like to append a Widget to  
 certain
 nodes where certain conditions are met, but I'm not having much
 success.

 appendChild expects a Node as its parameter, so I can't just  
 append a
 type of Widget. It will accept Widget.getElement() as a parameter,  
 but
 this leads to runtime errors.

 Is it possible to mix Widgets with the DOM like this or is the only
 way to add Widgets to a page via the RootPanel.get() ..? If so is
 there a way to add Widgets to the page based on a desired position  
 in
 the DOM?

 I'd like to use a custom widget I've made, having everything wrapped
 neatly in a class, and not have to revert to lower level stuff..

 Thanks for any help :)
 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Is it possible to call the GWT Service outside the entry point?

2009-06-11 Thread Jason Essington


perhaps this little discussion would be of some help:
http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/faca1575f306ba0f/3be719c021aa19bd

-jason
On Jun 11, 2009, at 4:05 AM, retha pasalli wrote:


 Hi Mr. Kushner,

 Thx for your reply. I've tried to call the service from a class
 extending Composite class. And it works! But I still can't find why it
 doesn't work when I call it in class extending nothing. But that's not
 an urgent problem for me for a while. Hehehe..

 Once more thx...

 Regards,
 Retha
 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: GWT Support for other browsers

2009-06-11 Thread Jeff Chimene


On Thu, Jun 11, 2009 at 2:21 AM, Thomas Broyert.bro...@gmail.com wrote:



 On 10 juin, 19:24, Jeff Chimene jchim...@gmail.com wrote:
 If you can't get the browser to send an alternate ident string, you
 might try hacking the *.nocache.js to handle the Galio browser ident
 string and instead reply with either the IE or Firefox branch to the
 client. Compile your GWT code as PRETTY to get code that's easier to
 understand.

 It's far easier to override the property-provider
 name=user.agent: copy it from UserAgent.gwt.xml (within gwt-
 user.jar) into your own *.gwt.xml and tweak it so that Galio is
 identified as another supported browser.

Nice solution. Does that mean that they will have to test all
supoorted browsers for the best match?

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Question about SuggestBox.wrap

2009-06-11 Thread Ben


It seems that is the only one to avoid it.

Thanks!

On Jun 11, 4:50 am, Thomas Broyer t.bro...@gmail.com wrote:
 On 11 juin, 00:38, Ben benzhe...@gmail.com wrote:





  I have a text box insides a FlexTable and I am trying to use
  SuggestBox.wrap to wrap the text box into a SuggestBox. Code snippet
  is like this.

  .
  FlexTable ft = new FlexTable();
  TextBox tb = new TextBox();

  ft.setWidget(0,0,tb);

  MultiWordSuggestOracle oracle = new MultiWordSuggestOracle();

  suggest = SuggestBox.wrap(oracle, tb.getElement());
  .

  But it throws error in hosted mode like this:

  java.lang.AssertionError: A widget that has an existing parent widget
  may not be added to the detach list
          at com.google.gwt.user.client.ui.RootPanel.detachOnWindowClose
  (RootPanel.java:122)
          at com.google.gwt.user.client.ui.SuggestBox.wrap(SuggestBox.java:
  267)...

  But it does not have this problem when you compile it and run it in
  browsers. I look into the source code of SuggestBox, it throws error
  in this line.

   266     suggestBox.onAttach();
   267     RootPanel.detachOnWindowClose(suggestBox);

  What is the problem? Is there a way to solve it?

 Do not use wrap(), use the ctor instead.

 FlexTable ft = new FlexTable();
 TextBox tb = new TextBox();
 MultiWordSuggestOracle oracle = new MultiWordSuggestOracle();
 suggest = new SuggestBox(oracle, tb);

 ft.setWidget(0,0,suggest);
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Event handler for model ?


Could you please post some examples will?
Or maybe send me some code? mnenchevs example was good, but more
examples never hurt :-)

On 11 Juni, 14:08, Will wil...@gmail.com wrote:
 That's what I use as well.  I created an Abstract Class with
 HashmapEventHander,HandlerRegistration for removing a handler
 (because HandlerManager.removeHandler(..) is depricated) and use that
 for all my UI components that create custom events.  I then create sub-
 Classes of EventHandler for each logical group of events but that's
 just to help keep the project readable.  Doing it once for the base UI
 class didn't seem overweight but we'll see how it performs once I get
 the project into a perf test.

 On Jun 10, 8:40 am, Dalla dalla_man...@hotmail.com wrote:



  Hi

  I´m having trouble understanding how to implement an event handler for
  my custom objects with the new event handler system. Let´s pretend I
  have a very simple model class like this one:

  public class ArrivalData {
          private Date date;
          private boolean processing = false;

          public Date getDate() {
                  return date;
          }

          public void setDate(Date date) {
                  this.date = date;
          }

  }

  When setDate is called, I want to fire an event that my view
  components can listen on, and act on when called for.
  How would you go about doing that? Are there any existing handler
  interfaces that can be used?
  I found this tutorial (see link below), but it seems pretty
  complicated for what I´m trying to 
  acheive.http://www.itsolut.com/chrismusings/2009/04/28/business-events-with-g...

  What´s best practice here?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Specify where Tomcat Lite is Installed

2009-06-11 Thread Sledged


I'm using GWT 1.6.4, and when I run my JUnit test cases, the GWTShell
installs Tomcat Lite in the tomcat directory under my project root
directory.  How can I specify where Tomcat Lite is installed?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

problem when

2009-06-11 Thread net


Hi all, i'm newbie in techno thingsbut i'm now developing a UI
with gwt ext. I develop on Windows XP.
I'd like to call web services. i have a .wsdl file, compile it using
cxf (wsdl2java) by maven to generate java sources from wsdl. I wrote a
method to call this service in order to get a list of technical
services like this:

@Transactional
public class SampleServiceImpl implements SampleService {

public ListTechnicalServiceTO createServiceDescription()throws
Exception {
ServiceManagerImplService ssDr = new ServiceManagerImplService(); //
wsdl service generated
ServiceManager sm = ssDr.getServiceManagerImplPort();
ListTechnicalServiceTO sdragon = sm.getAllServices();
return sdragon;
}

in my presentation layer, i have GWTSampleServiceAsync,
GWTSampleServiceImple... to link with the above service. once i
compiled this layer, everything seems ok.
but, when i click a button that invokes this service, there's problem
of Class not found: javax/xml/ws/Service

Then, i add a dependency of this class (jaxws.jar) into my pom.xml of
presentation layerI compiled once again...there's still not
work...and gave me another error which is: javax/xml/ws cannot be
serialized.
So, now I dont know what to do.

Anyone has ideas how to solve such problem?

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Large Data Set problems

2009-06-11 Thread Joe Larson


I am having a performance problem.  I am trying to build a page that
allows the user to select from a fairly large list of objects (up to
20,000).  We have filtering capabilities to trim the list so the user
doesn't have to scroll through all of them.  It's a reasonable UI.

Unfortunately, I only get decent performance on Safari.  IE6 is
exceptionally miserable.  It's okay when the list has 100 items, but
it crashes as we exceed 300.  So I've been experimenting, and I
haven't found a way around it.

So I created a sample app.  All I did was populate an
ArrayListString with 20,000 fairly short strings that I read out of
the HTML file.  Under Safari 4 on a G5 Mac Pro, this is almost
instant.  Under IE6 on a similar age Windows box, it's taking 4
minutes.  Firefox on the G5 runs out of memory.

Is it possible to store large amounts of data in my GWT applications?
Or do I have to store fairly small amounts of data and go back and
forth to the server a lot?  If it's the latter, I may as well just go
back to what we used to do (WebObjects).

I'm including my little sample program.  Maybe there's something
obvious that I can do?

-Joe

--- GWTMinimum.java ---
package com.missionmode.gwtminimum.client;

import java.util.*;

import com.google.gwt.core.client.*;
import com.google.gwt.user.client.*;
import com.google.gwt.user.client.ui.*;

/**
 * Entry point classes define codeonModuleLoad()/code.
 */
public class GWTMinimum implements EntryPoint {
public FlexTablepanel;
public Button   loadBtn;
public Button   displayBtn;
public ListBox  listBox;
public Labellabel;
public ArrayListStringusernames = new ArrayListString();
public String[] parts;

/**
 * This is the entry point method.
 */
public void onModuleLoad() {
panel = new FlexTable();
RootPanel.get(web2).add(panel);

loadBtn = new Button(Click to start load);
panel.setWidget(0, 0, loadBtn);

loadBtn.addClickListener(new ClickListener() {
public void onClick(Widget sender) {
loadUsers();
}
});
}

/**
 * We've loaded the user table.  Set up so they
 * can press another button to display the data.
 */
public void setupForDisplay() {
displayBtn = new Button(Click to Display);
panel.setWidget(0, 1, displayBtn);
displayBtn.addClickListener(new ClickListener() {
public void onClick(Widget sender) {
showUsers();
}
});
}

/**
 * Load the users from the HTML page.
 */
public void loadUsers() {
if (parts == null) {
String str = stringFromDocument(users);
parts = str.split(===);
loadBtn.setText(Click me again);
}
else {
for (String name : parts) {
name = name.trim();
if (name.length()  0) {
usernames.add(name);
}
}
showCount();
setupForDisplay();
}
}

/**
 * Just show how many users were actually loaded.
 */
public void showCount() {
label = new Label(Total user count:  + usernames.size());
panel.setWidget(1, 0, label);
}

/**
 * Show the loaded users.
 */
public void showUsers() {
try {
listBox = new ListBox();
listBox.setVisibleItemCount(20);
panel.setWidget(2, 0, listBox);
for (String str : usernames) {
listBox.addItem(str);
}
}
catch (Exception e) {
panel.setText(3, 0, Error:  + e.getMessage());
}
}

/**
 * Return a string from the document.
 */
public String stringFromDocument(String elementName) {
String retVal = ;
Element params = DOM.getElementById(elementName);
if (params != null) {
retVal = DOM.getInnerText(params);
}
return retVal;
}
}

--- Foo.html ---
!DOCTYPE HTML PUBLIC -//W3C//DTD HTML 4.01 Transitional//EN
html
head
meta http-equiv=content-type content=text/html;
charset=UTF-8
titleGWTMinimum/title
script type=text/javascript

Background image shows up in hosted mode but not in web mode

2009-06-11 Thread Manolis Platakis


Hello,

I have created the following css style to add a background image to
the html page of my application.

.bg {
  background: url(images/bg.gif) 50% 50% no-repeat;
}

and in the .java I have:

RootPanel.get().addStyleName(bg);

In hosted mode I can see the background image but in web mode (after
compiling), although the application is up to date, the background
image does not show up.

Any ideas on why this might be happening?

Thanks in advance,
Manolis

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Using Filters

2009-06-11 Thread Jamie


I have implemented something similar using a filter.

In the doFilter, you can cast the ServletRequest parameter to an
HttpServletRequest.
You will then have access to cookies and the container's session.

For example,

HttpServletRequest hRequest = (HttpServletRequest)request;
Cookie[] cookies = hRequest.getCookies();
hRequest.getSession().setAttribute(someattr, object);
(MyObject)myObject = hRequest.getSession().getAttribute(blap);

You could store your sessionid in a cookie, or in the container
session.

This can provide a pretty decent auth mechanism, I think.

1. My filter takes an initialization parameter, named requiredrole.
If set, then the user must have that role or else the request is
denied or redirected or whatever.
2. If you use a cookie to send the sessionid, you can get it back
using hRequest.getCookies().
3. You can avoid sending your own sessionid to the client altogether,
and instead, store it in the container's session if you wish, using
hRequest.getSession().setAttribute().
(that is dependent on the container's session cookie coming from the
browser.)

You can stop there can continue on your way...
The following is some detail on how I have used this to create my own
login filter.


As for the user logged in, I have implemented a class that implements
java.security.Principal,
This must implement getName().  To that I added a set of roles.

public class MyPrincipal implements Principal
{
String name;
SetString roles = new HashSetString();
etc...
public String getName() { return name; }
public boolean isUserInRole(String role) { return roles.contains
(role); }
}


Additionally, I have a class that extends HttpServletRequestWrapper.

private class UserPrincipalRequestWrapper extends
HttpServletRequestWrapper
{
  MyPrincipal p;
  public UserPrincipalRequestWrapper(HttpServletRequest arg0,
MyPrincipal p)
  {
 super(arg0);
 this.p = p;
  }
  public Principal getUserPrincipal()
  {
 return p;
  }
  public boolean isUserInRole(String role)
  {
return p.isUserInRole(role);
  }


So!  How does this all work?
- A request comes in to the filter.
- The filter looks in the container session for a MyPrincipal object
- If not found, then redirect to the login page
- If found, then
  - if requiresRole is set, verify the MyPrincipal object has that
role.
  - create a UserPrincipalRequestWrapper using that MyPrincipal object
  - call chain.doFilter(wrapper, response);

The login page is therefore required to assign a MyPrincipal object to
the session using
hRequest.getSession().setAttribute(principal, new MyPrincipal(...));


Inside the Servlet method, I can then call the standard auth methods,
for example:

  // Get the request inside GWT servlet
  HttpServletRequest request = getThreadLocalRequest();
  request.getUserPrincipal();
  request.isUserInRole(admin);


Finally, in my web.xml, I initialize multiple filters, with
requiredRole set to different things; I then apply that to different
servlets.
EG., I have a servlet for admin functions, and another for user
functions.

For bonus points, you can implement a persistent cookie (random
number?) that will reload the MyPrincipal into the container session,
if the container session does not already have one.  This is to
support 'log in for two weeks' feature of many popular email sites for
example.

Well there you go, more detail than you ever needed to know...

Jamie.
---
Search for analog and digital television broadcast antennas in your
area:
http://www.antennamap.com/

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Large Data Set problems

2009-06-11 Thread Tom Hjellming


Hi Joe,

The trick is to not even try to load that much data into the browser.  
We use the SmartGWT widget library which has very good support for 
virtual pagination of lists and comboboxes.  By default, it only loads 
the first 75 rows and then as the user scrolls down, calls are made back 
to the server to retrieve the desired data - which is then shown to the 
user.  There is an occasional delay when the user sees blank space which 
is then filled in when the data comes in.  But overall, it is a great 
solution -we're viewing tables with thousands of rows and seeing better 
performance than our old Swing-based application.

Tom


Joe Larson wrote:
 I am having a performance problem.  I am trying to build a page that
 allows the user to select from a fairly large list of objects (up to
 20,000).  We have filtering capabilities to trim the list so the user
 doesn't have to scroll through all of them.  It's a reasonable UI.

 Unfortunately, I only get decent performance on Safari.  IE6 is
 exceptionally miserable.  It's okay when the list has 100 items, but
 it crashes as we exceed 300.  So I've been experimenting, and I
 haven't found a way around it.

 So I created a sample app.  All I did was populate an
 ArrayListString with 20,000 fairly short strings that I read out of
 the HTML file.  Under Safari 4 on a G5 Mac Pro, this is almost
 instant.  Under IE6 on a similar age Windows box, it's taking 4
 minutes.  Firefox on the G5 runs out of memory.

 Is it possible to store large amounts of data in my GWT applications?
 Or do I have to store fairly small amounts of data and go back and
 forth to the server a lot?  If it's the latter, I may as well just go
 back to what we used to do (WebObjects).

 I'm including my little sample program.  Maybe there's something
 obvious that I can do?

 -Joe

 --- GWTMinimum.java ---
 package com.missionmode.gwtminimum.client;

 import java.util.*;

 import com.google.gwt.core.client.*;
 import com.google.gwt.user.client.*;
 import com.google.gwt.user.client.ui.*;

 /**
  * Entry point classes define codeonModuleLoad()/code.
  */
 public class GWTMinimum implements EntryPoint {
   public FlexTablepanel;
   public Button   loadBtn;
   public Button   displayBtn;
   public ListBox  listBox;
   public Labellabel;
   public ArrayListStringusernames = new ArrayListString();
   public String[] parts;

   /**
* This is the entry point method.
*/
   public void onModuleLoad() {
   panel = new FlexTable();
   RootPanel.get(web2).add(panel);

   loadBtn = new Button(Click to start load);
   panel.setWidget(0, 0, loadBtn);

   loadBtn.addClickListener(new ClickListener() {
   public void onClick(Widget sender) {
   loadUsers();
   }
   });
   }

   /**
* We've loaded the user table.  Set up so they
* can press another button to display the data.
*/
   public void setupForDisplay() {
   displayBtn = new Button(Click to Display);
   panel.setWidget(0, 1, displayBtn);
   displayBtn.addClickListener(new ClickListener() {
   public void onClick(Widget sender) {
   showUsers();
   }
   });
   }

   /**
* Load the users from the HTML page.
*/
   public void loadUsers() {
   if (parts == null) {
   String str = stringFromDocument(users);
   parts = str.split(===);
   loadBtn.setText(Click me again);
   }
   else {
   for (String name : parts) {
   name = name.trim();
   if (name.length()  0) {
   usernames.add(name);
   }
   }
   showCount();
   setupForDisplay();
   }
   }

   /**
* Just show how many users were actually loaded.
*/
   public void showCount() {
   label = new Label(Total user count:  + usernames.size());
   panel.setWidget(1, 0, label);
   }

   /**
* Show the loaded users.
*/
   public void showUsers() {
   try {
   listBox = new ListBox();
   listBox.setVisibleItemCount(20);
   panel.setWidget(2, 0, listBox);
   for (String str : usernames) {
   listBox.addItem(str);
   }
   }
   catch (Exception e) {

Re: HELLO MODERATOR

Hi eags,
Sorry for the delay in moderation. Unfortunately I'm the primary moderator
for the forum, and so posting in a different timezone does sometimes add a
bit more delay than I'd like to a post. Perhaps I should try a polyphasic
sleep schedule :-)

Regarding moderated posts - you should be cleared for posting now that your
first post has gone through. I only check the first post for new users
registered to the forum for spam. If that clears, then you become free to
post in the future without moderation. That said, I'm afraid there isn't
much I can do about changing the timestamp on the previously moderated
message. I'm guessing that this post referencing the previous post should
get enough attention on it so that other group members will post up.

Sorry again for the delay,
-Sumit Chandel

On Tue, Jun 9, 2009 at 11:03 AM, eags eagsala...@gmail.com wrote:


 Hi.  I posted a comment last night starting a new thread called Very
 basic LoginSecurityFAQ and GWT-RPC questions.  I guess because of the
 moderation it wasn't actually posted until about a half an hour ago
 but the timestamp on it still is for last night which means that it is
 buried in last nights conversations and no one is seeing it.

 So I have two questions:

 1. Should I repost?  Or is it possible for you to reset the timestamp
 so it gets seen?
 2. How do I become a regular user and not get moderated anymore?  Is
 it a certain number of posts?  Or is it a certain average rating on my
 posts?  Thanks.


 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Passing bag of Serializable objects to RPC

2009-06-11 Thread Daniel Jue


Hi, I've read some past emails on this subject, but I want to know if
anyone has a best practice for this.

Lets say I have MyService(HashMapString,Serializable reportBag) .
I've read it's advisable to narrow down the Serializable to a type
that has a smaller set of implementations, such as using GXT2's
serializable BeanModel or making up a marker like SerializaleDTO (or
for that matter, the old IsSerializable)  I'd be willing to do this,
except I'd like to put ArrayListBeanModel in my
HashMapString,Serializable reportBag

When I start stuffing things in my bag from widgets like selections
from grids, etc, these my come back as ArrayListBeanModel.

This means if I want to do

reportBag.put(StuffFromWidget1,myGrid.getSelection());
it won't work unless myGrid.getSelection() is null.

I tried wrapping ArrayListBeanModel in a new implementation of
BeanModel, but it still did not work (same error, even after
cleaning).

I guess what I'm looking for is a way to sent an object with an
arbitrary number and combination of beans and lists of beans, where a
bean is a Serializable object.  I'd prefer to use some sort of map, so
I can look up the objects in my service impl.

i.e.
reportBag = new HashMapString, Serializable();
reportBag.put(SomeText,JustText);
reportBag.put(SomeBean,imaSerializableBean);
reportBag.put(ListOBeans,thatListOBeansFromAWidget);
reportBag.put(OneMoreBean,forGoodMeasureBean);
etc etc

then

MyServiceAsync.RPC.runReport(reportBag, callback);

Is this possible?


I know topics on serialization get hashed and rehashed here and I sure
don't want to bother people with noob questions.  I must have read 40
threads on it today!  Now I'm off to google code search to try and
find something in the wild...

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Please help! GWT out of the box examples stopped running correctly in hosted mode under Windows Vista

Hi sbarr10,
Here is the previous thread where you posted about this issue:

http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/54e1ffa04907b569

http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/54e1ffa04907b569I
replied back on there recommending that you check if the Windows Firewall or
new security mechanism is potentially blocking the hosted mode process. If
that's the case, you should be getting a security popup prompting you to
grant access to the process / allow it to connect to the network, unless it
has been set to always allow processes to run (although I'm not quite sure
if there exists a setting to do that). Feel free to get back on the previous
thread if you're still having problems getting hosted mode working.

Hope that helps,
-Sumit Chandel

On Sun, Jun 7, 2009 at 9:36 PM, sbar...@earthlink.net sbar...@earthlink.net
 wrote:


 I cannot find my previous post on this topic.

  I am new to Vista and for weeks have been happily coding away in GWT
 1.6.4 using hosted mode.   I strictly use DOS shells (no Eclipse).

 The issue is not specific to just my code, it affects the GWT sample
 applications out of the box too.

 When I type:

 %ant_home%\bin\ant clean -The relevant directories were cleaned.
 %ant_home%\bin\ant build -Yes, new files are being generated.
 %ant_home%\bin\ant hosted-  I get two blank white screens and nothing
 happens.

 If I then bang the Compile/Browse button and wait a minute, the sample
 app will eventually launch - in a THIRD window.  This did not used to
 happen!

 I looked at the Windows Update history and noticed that an IE8 install
 had failed, so I went and completed that.

 What is it that I am not understanding about Vista?!?




 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Rebind for Android/Webkit?

2009-06-11 Thread Matt Bishop


I think you have to specify an for each if in your replace-with.
Try:

replace-with
class=com.hs.moose.common.client.widget.NextPreviousButtonCSS
when-type-is
class=com.hs.moose.common.client.widget.NextPreviousButton/
any
when-property-is name=user.agent value=safari/
when-property-is name=WebKit value=yes/
/any
/replace-with


On Jun 10, 6:18 am, Evan Ruff evan.r...@gmail.com wrote:
 Hey guys,

 I'm trying to add some rebind parameters for an Android implementation
 of my application and I just can't seem to get it to work right. In my
 Application.gwt.xml I'm doing this:

 define-property name=WebKit values=yes,no/
             property-provider name=WebKit![CDATA[ return
 navigator.userAgent.indexOf('WebKit')  -1 ? 'yes' : 'no'; ]]
         /property-provider

     replace-with
 class=com.hs.moose.common.client.widget.NextPreviousButtonCSS
         when-type-is
 class=com.hs.moose.common.client.widget.NextPreviousButton/
         when-property-is name=user.agent value=safari/
         when-property-is name=WebKit value=yes/
     /replace-with

         !-- Rebind the entry point if Gears is not installed --
         replace-with class=com.hs.moose.app.client.Application
                 when-type-is class=com.hs.moose.app.client.HasGearsEntry /
                 when-property-is name=gears.installed value=false /
         /replace-with

 But $WebKit never gets set, either yes or no. I was wondering if
 anyone could help me out on this sort of thing (It's my first go at
 it.) Also, is there a way to do Android, iPhone and Chrome seperately,
 or am I only limited to Webkit?

 Thanks!

 Evan
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Mouse Wheel Handler on a TextBox

Hi D L H,
I'm guessing the browser you were testing on when you tried this out was
Firefox 3? There is a known (and fixed) issue for this reported on the Issue
Tracker. See Issue #2902 (link below). The fix for the issue has been
committed all will be included in the next release.

Issue #2902:
http://code.google.com/p/google-web-toolkit/issues/detail?id=2902

If you're working from the GWT trunk, you can pick up and patch the issue
with r5398.

Hope that helps,
-Sumit Chandel

On Mon, Jun 8, 2009 at 7:09 AM, D L H thed2...@gmail.com wrote:


 hello.

 i'm trying to make a TextBox that will change value when someone
 scrolls the mouse wheel over it. specifically, my goal is for the
 number in the text box to increase when i scroll up, and decrease when
 i scroll down. however, i'm having trouble figuring out the
 MouseWheelHandler. i simplified my code to just change the value to
 UP or DOWN, but it just doesn't work. it compiles though. i also
 tried it with event.preventDefault(), but that didn't seem to have any
 effect.

 private TextBox valueField = new TextBox();
 ...
 ...
 valueField.addMouseWheelHandler(new MouseWheelHandler() {
   public void onMouseWheel(MouseWheelEvent event) {
  //event.preventDefault();
  if(event.isNorth()) {
 valueField.setText(UP);
  } else {
 valueField.setText(DOWN);
  }
   }
 });
 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: 404 error: gwt server can not be found

2009-06-11 Thread Paul Hargreaves



Hi

I had a similar problem a few days ago. It was down to an entry in my web.xml 
file. Perhaps if you post your xml servlet-mapping entry, someone may be able 
to help you with it. I do not see why it should make a difference, I know it 
does not in standard browsers, but should there be an extra forward slash in 
/notamsearchpreference//searchService ? I am a novice and may be getting it 
all wrong. Sorry if I am.

Paul


- Original Message 
From: Linda linda.ctr.c...@faa.gov
To: Google Web Toolkit Google-Web-Toolkit@googlegroups.com
Sent: Thursday, 11 June, 2009 17:10:26
Subject: 404 error: gwt server can not be found


I upgraded my project from gwt-1.5 to gwt-1.6 and updated my web.xml
file. Now I am running a Error 404 problem after I get a warning :
No file found for: /notamsearchpreference//searchService. I checked
my WEB-INF/class/myClient folder, the searchService.class was there.

here is the sample code how do I create the server:

public static SearchServiceAsync searchService = (SearchServiceAsync)
GWT
        .create(SearchService.class);
    static
    {
        ServiceDefTarget endpoint = (ServiceDefTarget) searchService;
        String moduleRelativeURL = GWT.getModuleBaseURL()
            + /searchService;
        endpoint.setServiceEntryPoint(moduleRelativeURL);
    }

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: GWT + hosted mode + jboss

Hi mmenchev,
You should still be able to use hosted mode with -noserver through the
plugin, although there are some steps you need to take first to get
everything setup for hosted mode debugging on your server. Check out the FAQ
link below.

FAQ How do I use my own server instead of GWT's built-in Jetty instance:
http://code.google.com/webtoolkit/doc/1.6/FAQ_DebuggingAndCompiling.html#How_do_I_use_my_own_server_in_hosted_mode_instead_of_GWT's

http://code.google.com/webtoolkit/doc/1.6/FAQ_DebuggingAndCompiling.html#How_do_I_use_my_own_server_in_hosted_mode_instead_of_GWT'sOnce
you've followed the steps outlined there, you should be able to launch
hosted mode with the plugin and work with your own server for hosted mode
debugging.

Hope that helps,
-Sumit Chandel

On Sat, Jun 6, 2009 at 7:49 AM, mnenchev nenchev.mari...@gmail.com wrote:

Hi, all,

I dig the web, but i could not find how to run hosted mode with
external jboss.

Here is what i do:

1) Create web project with google plugin for eclipse, that creates me
hello world project with gwt.

2) test it within the internal server and it works.

3) Add new server runtime jboss 4.2.

4) Change the project facets - runtimes check jboss.

5) run configurations - set the port 8080 and uncheck the use
default server checkbox

6) Start the jboss from eclipse(view servers)

6) run the application with the google plugin

When the hosted browser is displayed it says that it could not find
the resource on the server.

So my question is how to deploy on the server, if i have to do it
every time when i change something this host mode is useless with
external server. I also added -noserver, but the application is not
deployed on the jboss so the effect is the same. How to manage this. I
look all over the net but i could not find any complete example how to
do this, every one just says use -noserver, but that does not solve
the problem.

Pls help me i am new with gwt.
Regards.

Re: gwt 1.6 Embedded Jetty log

Hi soncio69,
Try setting the -logLevel DEBUG argument in your hosted mode launch
configuration. It should tell you a lot more about the exception that was
thrown in the hosted mode console. Simply select the exception in the upper
pane of the console window and the exception stack trace should be displayed
in the console window output in the lower pane.

Hope that helps,
-Sumit Chandel

On Mon, Jun 8, 2009 at 2:45 AM, sonci...@gmail.com sonci...@gmail.comwrote:


 I'm trying to upgrade my project from GWT 1.5 to 1.6
 Now everything sems to be ok on my client side but  i'm having a
 problem running my serviceImpl on the server side.
 My AsyncCallback on failure mathod get executed with the following
 message: The call failed on the server; see server log for details
 I'm new to Jetty, where can i find server log?
 The same application on GWT 1.5 and embedded tomcat runs without
 problems.






 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

GWT architecture MVP/EventBus (mentioned at Google I/O)

2009-06-11 Thread Benju


I did not attend Google I/O but as soon as the video Google I/O 2009
- Best Practices for Architecting GWT App (http://www.youtube.com/
watch?v=PDuhR18-EdM) was posted I reviewed it and was a bit confused
by the idea of an EventBus.

From what I can tell the idea is that UI widgets requiring data from
the server are able to fire off requests for some form of data like
void getTransactionsForAccount(Account acct) then at some unknown
later time (ms to seconds ussually) when a response comes in from the
RPC call the eventbus is what actually directly recives the data and
then it is dispatched


Client UI: Hey call some RPC method with these parameters, my
AsyncCallback is this EventBus thing

...some unknown time passes while the server does magic...

Client Event Bus: A response came in of type X/Y/Z I should fire an
event to all interested parties

Client UI: According to this event I just recieved some of my UI code
needs to change


A few things are still very hazy for me...


1- What is missing here is when would the Client UI typically
subscribe/unsubscribe from the event bus.  If this were a desktop
application I would simply use weak reference so I would not have to
unsubscribe my UI manually to prevent a memory leak.

2- Is there some EventBus code I should be using that already exists
in the GWT SDK?  Is this the same code that is used for handling
widget events like clicking a button?

3- Would you typically have one eventbus code for everything?

4- How does this tie in with the MVP pattern?

If anybody else watched the video I would be curious what they have to
say.  Also Ray Ryan mentioned he was going to be creating a new demo
application for GWT developers to see this sort of thing in action.
That would be awesome, does anybody have any thoughts on this as well?

Thanks in advance for your comments.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Can`t use HttpServlet in hosted mode debugger after migrating to GWT 1.6

2009-06-11 Thread mabead


Thanks for your feedback.

The WEB-INFO (instead of WEB-INF) was a typo.

The issue is resolved.  Here is an explanation of why it failed.
First, the request that I wanted to send to my HttpServlet looked like
the following:

String url = GWT.getModuleBaseURL() + RESTProxy/system;
RequestBuilder requestBuilder = new RequestBuilder
(RequestBuilder.GET, url);
requestBuilder.sendRequest(, new RequestCallback()

On the other hand, my web.xml looked like the following:

servlet
  servlet-nameRESTProxyServlet/servlet-name
  servlet-classcom.foo.bar.server.RESTProxy/servlet-class
   /servlet

   servlet-mapping
  servlet-nameRESTProxyServlet/servlet-name
  url-pattern/foobar/RESTProxy/url-pattern
   /servlet-mapping

As you can see, there is no wildcard in my url-pattern.  I had to
modify my url-pattern to the following

  url-pattern/foobar/RESTProxy/*/url-pattern

After this change, everything goes fine.

Regards,
Maxime
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: GWT app with Spring Security caching/refresh issue

2009-06-11 Thread tim.clymer

http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/9e41d90e33fd294e/668f9b62079b8f0e?lnk=gstq=blank+until+refresh#668f9b62079b8f0e

This thread also seems to have run into the issue. Seems to be a
tomcat issue. Odd though that I haven't been able to find more
instances of it.

Has anyone else had this problem and fixed it with a method other than
renaming the entry point from .html to .jsp?

Tim

On Jun 11, 2:27 pm, tim.clymer tim.cly...@gmail.com wrote:
I've been having an odd caching/refresh issue with my application. I
have my GWT (1.6.4) app secured with Spring Security (2.0.4). For the
most part, this works perfectly. I have a login page which redirects
to my entry point page once authenticated, and any attempt to access
my entry point page without being authenticated redirects to the login
page. However sometimes when the redirect is performed on login I get
a blank entry point page and the browser shows that loading is
complete. When I look at the page source, the entry point HTML is
there along with the reference to the generated nocache js. However,
my app doesn't load and the page title doesn't appear in the browser
tab, just the URL.

I'm wondering if this is caching related, but this is just my best
guess. If I refresh the page, the app always pops right up. It's
happening across all browsers. Any ideas?

What I'm using:
GWT 1.6.4
SmartGWT 1.1
Tomcat 6

Thanks,
Tim
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

invoking a GWT service from a J2SE app

2009-06-11 Thread ajperez4


Hi all,

has somebody tried to invoke a GWT service from a J2SE app?

When trying to invoke the service as the same way in a GWT client app
I get the following:

Exception in thread AWT-EventQueue-0
java.lang.UnsupportedOperationException: ERROR: GWT.create() is only
usable in client code!  It cannot be called, for example, from server
code.  If you are running a unit test, check that your test case
extends GWTTestCase and that GWT.create() is not called from within an
initializer or constructor.
at com.google.gwt.core.client.GWT.create(GWT.java:91)

thanks!

Alfredo
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Eclipse Classic 3.5

2009-06-11 Thread Joakim


Any news on the progress of the 3.5 support? Can we expect a working
plugin in time for the final release of Galileo in two weeks?


On 28 Maj, 20:37, Miguel Méndez mmen...@google.com wrote:
 The plugin code is not open sourced at this time, but we do plan to open
 source it.





 On Tue, May 26, 2009 at 7:50 PM, acabler acab...@gmail.com wrote:

  This will be really nice to have.  Is the plugin code available in svn
  yet?  I would like to check it out so I can contribute patches for
  issues like these.

  thanks,
  adam

  On May 23, 10:40 am, Alex Rudnick a...@google.com wrote:
   Hey LiR,

   The plugin doesn't supportEclipse3.5yet. We're aware of the
   incompatibility, and it'll be fixed in an upcoming release -- likely
   soon before Galileo gets officially released.

   Thanks!

   On Fri, May 22, 2009 at 11:46 PM, LiR kirill...@gmail.com wrote:

   EclipseClassic3.5is available but google plugin (for 3.4 version)
doesn`t want to install in new version.

I have a Install Details message in top of window:

The operation cannot be completed.  See the details.

And details block:

Cannot complete the install because one or more required items could
not be found.
Software being installed:
com.google.gdt.eclipse.suite.e34.feature.feature.group
1.0.1.v200905131143
Missing requirement:
com.google.gdt.eclipse.suite.e34.feature.feature.group
1.0.1.v200905131143 requires 'org.eclipse.platform.feature.group
[3.4.0,3.5.0)' but it could not be found

   --
   Alex Rudnick
   swe, gwt, atl

 --
 Miguel

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Name a function that will not get changed when compliling?

2009-06-11 Thread Shedokan


Hello,
How can I name a function like start() and it will not change while
compliling so that other javascript can call it?

thanks.

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Mysql access dienied in GWT hosted mode

2009-06-11 Thread Bonor


never mind

On 10 jun, 17:35, Bonor bono...@gmail.com wrote:
 When doing a RPC call my Service get's an Access Denied error when
 trying to connect with mysql.

 java.security.AccessControlException: access denied
 (java.lang.RuntimePermission modifyThreadGroup)

 Where do i grant access?

 Suggestions?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Name a function that will not get changed when compliling?

2009-06-11 Thread Ian Petersen


Search for GWT Exporter and use that.

On Thu, Jun 11, 2009 at 10:58 AM, Shedokanshedok...@gmail.com wrote:

 Hello,
 How can I name a function like start() and it will not change while
 compliling so that other javascript can call it?

 thanks.

 


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Name a function that will not get changed when compliling?

2009-06-11 Thread Alex Rudnick


Hey Shedokan,

The approach I've used for that is to attach a GWT function to the
window object ($wnd).

There's an example in the docs over here:
http://code.google.com/webtoolkit/doc/1.6/DevGuideCodingBasics.html#DevGuideJavaScriptNativeInterface

Check under the subsection Calling a Java Method from Handwritten JavaScript.

Like Ian said, you could also use GWT Exporter.

Hope this helps!

On Thu, Jun 11, 2009 at 1:58 PM, Shedokanshedok...@gmail.com wrote:

 Hello,
 How can I name a function like start() and it will not change while
 compliling so that other javascript can call it?

 thanks.

-- 
Alex Rudnick
swe, gwt, atl

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: wow, total newbie question

Hi Steve,
Typically, you wouldn't want to define multiple entrypoints but instead use
a single entrypoint, and potentially divide your project into multiple
modules. More replies inlined to your post below:

Should this be implemented as 4 separate entry points? Should it be
one entry point and one page with all interfaces on it, showing and
hiding them as appropriate? If so won't this get bulky?

Using a single EntryPoint class, you could define four main panels that
represent each of the four pages you described above. You could then add
these panels to a TabPanel, for example, where each tab corresponds to one
of the four panels, and clicking on the tab opens up that panel. There are
many other options for how you can set this up, but the TabPanel example is
one that is hopefully easiest to visualize and understand. From the
functionality you described so far, designing the application this way would
probably not be too bulky, in terms of both size of your code base and
initial download size of your application for your user.

However, as your code base grows and you add features, you might want to
start separating your code into different modules to make clearer
distinctions between them and also make modules reusable in other components
of your application. This is when you can consider splitting your code into
multiple modules, where the module containing your entry point class
inherits from whichever other modules it needs to use. You can read more
about creating more modules for your application by checking out the doc
linked below.

See Dividing code into multiple modules section:
http://code.google.com/webtoolkit/doc/1.6/DevGuideOrganizingProjects.html#DevGuideModuleXml

For the question about whether this will become bulky for the user, the
answer is, it will. Which is why we're working on the new code splitting
runAsync() feature in GWT 2.0. Using the runAsync() feature, you can control
you can split up your code into individual parts and control when they get
loaded. This way, you might have the login and add new member panel coded
to load as soon as the user visits your application, but split out the edit
existing user and reminder email panels downloaded as separate fragments
in the background to reduce the initial download of your application. Until
GWT 2.0 releases and you can use the feature in a stable release, you can
either work from GWT trunk if you want to use the feature now (probably not
the best thing to do since you're just getting started with GWT), or design
your application following the principle of on-demand widget creation (link
below). Your user will still pay for the initial download size, but at least
you will cut down on rendering time and memory usage at a given moment. You
should also be able to revisit areas of your code where you lazily loaded
your components and replace these by runAsync() calls when GWT 2.0 lands.

On-demand widget creation:
http://googlewebtoolkit.blogspot.com/2008/11/improving-performance-with-on-demand.html

Are the HTML pages I add after the main entrypoint always pure html?
or can I add GWT goodness to them?

In general, you won't be adding HTML pages to an GWT application through the
entrypoint. Once you've coded your GWT application and run the GWT compiler
over the code base, it will generate a number of JavaScript and HTML files,
one of which is the JS bootstrap file. Your main HTML page (the page your
users land on when they visit your site) will include a script tag
referencing this JS bootstrap file and load your GWT application in the
page. You could instead add another HTML page into your GWT module through a
Frame widget, but that would only be done in very specific situations.

You can add a GWT component to an existing part of an application by
defining an element in your host HTML page that your GWT component can bind
to. For example:

index.html:

div id='gwtComponent'/div

MyEntryPoint.java:

public class MyEntryPoint implements EntryPoint {

public void onModuleLoad() {
VerticalPanel mainPanel = new VerticalPanel();

// Add components to the main panel, and other stuff...

RootPanel.get(gwtComponent).add(mainPanel);
}
}

Hope that helps,
-Sumit Chandel

In general it would be great to have some guidance as to entry points,
multiple entry points and how they should be used. I've read all the
documentation I can find and haven't seen this explained for someone
at a basic level as me.

Thanks in Advance
Steve

Preventing JavaScript Injection cient/server side solutions

2009-06-11 Thread tamsler


I am trying to figure out what the best way is to handle JavaScript
injection cases. Since any client side input validation handling
doesn't truly prevent one from injecting JS such as using tools like
Firebug to re-post RPC calls etc.

I am wondering if anybody has attempted to intercept JS injection on
the server side by scanning RPC calls . I could imagine using a
servlet filter to do this or or some other way.

Any ideas/feeback is greatly appreciated.

Best,
-- Thomas Amsler
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Preventing JavaScript Injection cient/server side solutions

2009-06-11 Thread Jeff Chimene


On 06/11/2009 04:18 PM, tamsler wrote:
 I am trying to figure out what the best way is to handle JavaScript
 injection cases. Since any client side input validation handling
 doesn't truly prevent one from injecting JS such as using tools like
 Firebug to re-post RPC calls etc.

 I am wondering if anybody has attempted to intercept JS injection on
 the server side by scanning RPC calls . I could imagine using a
 servlet filter to do this or or some other way.

 Any ideas/feeback is greatly appreciated.


It's a good question, but it's not really GWT related.
You're talking about server-side code. The  JS code generated by GWT 
executes in the browser.
I may be completely missing your point, but perhaps these articles may 
be apropos:
http://code.google.com/webtoolkit/articles/using_gwt_for_json_mashups.html
and
http://code.google.com/webtoolkit/articles/put_your_gwt_app_on_facebook.html


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Couldn't run the Getting Started example with JSON

2009-06-11 Thread Dagvadorj Galbadrakh


Anyway I got it working.

On Jun 8, 12:19 am, Dagvadorj Galbadrakh dagvad...@gmail.com wrote:
 Hello everybody, I am totally new to GWT. I Couldn't run the Getting
 Started example with JSON implementation at 
 address,http://code.google.com/webtoolkit/tutorials/1.6/JSON.html.

 The problem is that the code doesn't identify id field with getId
 method of GroupData class.

 Does anybody has a working full codes left from your walk through? I'd
 be appreciated with any help. Thanks.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Browser memory usage grows

2009-06-11 Thread Craig Spry

Hello All,

We've built an application using GWT 1.5.4 that calls a json server and the
memory usage for the browser that it is running in keeps going up, I've
tested this in IE, Firefox and Chrome with the same result.  This
application is a widget based application, we have a map widget(open street
maps) and some list widgets but it doesn't seem to matter which widgets that
we use the memory seems to grow.  I've tried using each widget individually
with the same result.  The json that is returned is big, Firebug doesn't
give me a value.  Our application polls for new information.

I've also tried to use this tool:
http://blogs.msdn.com/gpde/pages/javascript-memory-leak-detector.aspx
But it didn't detect any leak in javascript.

Unfortunatley this is an internal application so I can't point you to our
application.  If it would be helpful I'll try and knock up a test case.

Has anyone else had this problem?
If so what was the solution?

Thanks,

Craig Spry

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Is it possible to call the GWT Service outside the entry point?

2009-06-11 Thread retha pasalli


 perhaps this little discussion would be of some 
 help:http://groups.google.com/group/Google-Web-Toolkit/browse_thread/threa...

 -jason


Great illustration!! Thx jason..
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: Preventing JavaScript Injection cient/server side solutions

2009-06-11 Thread Shawn Brown


 It's a good question, but it's not really GWT related.

Sure it is.

 You're talking about server-side code. The  JS code generated by GWT
 executes in the browser.

RPC calls execute where?

http://code.google.com/webtoolkit/articles/security_for_gwt_applications.html

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

Re: App engine data types

2009-06-11 Thread bcottam


okay, so I dug into this a bit and figured out how to port existing
code to GWT compliant standards.  I also included
_CustomFieldSerializers.  I am currently only using Key and Text in my
test code, but I ported  Key, Text, Blob, ShortBlob, Link and User
however I haven't tested much beyond the Key and Text classes.

here is a link to the jars and some notes about them:
www.resmarksystems.com/code

I hope it is helpful  :)
-bryce

On May 14, 11:22 am, bcottam bcot...@gmail.com wrote:
 yeah, I actually posted there 
 first:http://groups.google.com/group/google-appengine-java/browse_thread/th...
 and they suggested that I cross-post here  :)

 I actually started using the persisteddataclasses and sending them
 across the gwt rpc channel, and that was fine until the datanucleus
 JDO started putting fields on my classes that were instances of Object
 [] (which don't serialize all that well).  That, and I know there is
 going to be a performance hit with sending a lot ofdataover the
 channel.  So, I am thinking that doing a lite DTO just for transport
 would possibly be my best option.

 I will check out Ray Cromwells posts, mainly I guess what I am looking
 for at this point is a good reference for using the super-source tag
 and how to make custom wrappers for non GWT classes (in the same
 manner that the java.util api is wrapped with com.google.emul etc.
 Any pointers in that direction would be great.

 thanks!

 On May 13, 7:59 pm, Alyxandor a.revolution.ultra.b...@gmail.com
 wrote:

  I'm really not sure, but I'd recommend checking out theAppengine
  google group 
  first...http://groups.google.com/group/google-appengine-java?hl=en

  There's over a hundred posts a day there, so I'm sure someone there
  with a little more expertise than yours truly will be of more help.

  Out of curiosity, have you already tried sending your persistentdata
  classes down the wire?  Cos  There's a lot of discussion about it,
  and some workarounds {basically, the only ways I could do it was with
  Ray Cromwell's various hacks, or a drop-in called Gilead}.  You might
  find that using the same beans to be nigh impossible without making
  them non-detachable {read: you can send an object to the client,
  modify it and send it back to the server to save without copying the
  object, or manually copying thedatafrom every field}.

  ...Check the other group, like I said...  I'm no expert.

  Of course, Ray Cromwell IS, so if you can find his Google-Groups
  profile, try clicking on some of the threads he posts on...  You'll be
  glad you did!
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Moving PagingScrollTable Friends to Trunk

2009-06-11 Thread jay


I'd definitely like to see the paging ability as an extension (read
that as you will: interface implementation, derived class, plug-in
behavior, whatever) to the scrolling behavior. I don't really like how
the incubator has two totally separate, unrelated classes.

I wonder if some performance enhancements could be made by having a
TableModel interface which is just read-only, and then only if you
need editing would you use a MutableTableModel. I'm just wondering
here...I really don't know. Though I'm sure it would have other
implications, like a change to the ColumnDefinition (e.g., making it
read-only, and having a MutableColumnDefinition?)

jay


On Jun 10, 9:55 am, Isaac Truett itru...@gmail.com wrote:
 Re: Bruce's point about expectations and features vs. performance.

 Has there been (or should we start) discussion for the public record
 of different facets/features of tables, their impact on performance,
 and their possible class structure? What I'm thinking of specifically
 is bulk rendering vs. dynamic model-backed tables, scrolling, paging,
 and all the combinatorial permutations thereof. Will supported
 features be part of a single class hierarchy as they are in Incubator
 (i.e., PagingScrollTable extends ScrollTable; paging code is bound to
 scrolling code, although scrolling can be disabled it would still add
 code to the compiled app)? Or will they be pluggable features into a
 base Table class (like MonthSelectors for DatePickers)? Or separate
 implementations of a common Table interface?

 - Isaac

 On Wed, Jun 10, 2009 at 11:27 AM, Bruce Johnson br...@google.com wrote:
  I'll be the bad guy and try to lower expectations: whatever we end up doing,
  it has to be fast. We've seen some *horrible* usability problems with fancy
  tables -- even at a small number of rows -- and so don't be surprised if we
  have to pare back features and reduce API flexibility to ensure that a few
  key use cases are sufficiently high performance.

  -- Bruce

  On Tue, Jun 9, 2009 at 10:12 AM, John LaBanca jlaba...@google.com wrote:

  We'll definitely keep these things in mind when moving stuff over to GWT
  trunk.  We've also found a lot of general usability problems, such as the
  fact the the table doesn't layout naturally, which means apps require 
  active
  layout.  During the transfer, we'll refactor quite a few things to make 
  them
  more usable.  Specifically, we'd like to provide a version that allows you
  to bulk renderer the header and footer into the same table element,
  eliminated the three separate tables and fixed layout.  You would lose the
  scrolling feature, but you would not have to use active layout.

  When we start moving stuff into trunk or while its in my branch (as in
  right now), thats a good time to point out specific problems or requests.
  Its much harder to change the API after we make an official release.

  Thanks,
  John LaBanca
  jlaba...@google.com

  On Tue, Jun 9, 2009 at 5:01 AM, David david.no...@gmail.com wrote:

  Jay,

  We are experiencing the same ideas here. We store column ordering and
  widths on the server but we have no way of getting events in the UI to
  know when changes have been complete.

  wouldn't it be nice that the dnd was included as well, I could really
  use the DND of columns! Was it hard to implement ? We did not yet
  bother to investigate since we have to focus on getting functionality
  complete first.

  David

  On Tue, Jun 9, 2009 at 10:00 AM, jayjay.gin...@gmail.com wrote:

   As I see that this has begun (yeah), I'd like to throw out a few
   requests:

    * Please, please, please -- ensure that this is as extensible as
   possible. Here's just one example--I've integrated the gwt-dnd library
   to allow drag-n-drop re-ordering of columns. There are a couple of
   funny corner cases, though, because I have no way of knowing when a
   column resize has completed. Obviously, if you're resizing the column,
   you're not interested in dragging it to a new location. I strongly
   encourage you to think three, four, five times about making a method
   private or package protected. Liberal use of JavaDoc with strongly
   worded warnings to those of us who need to customize the widgets. I
   know this cuts down on your ability to make under-the-cover changes
   from release to release, but it makes it so that folks like me don't
   have to resort to things like JSNI trickery or copying the entire
   class or set of classes into our own code base.

    * As a direct follow up to #1, fire some more events. For example,
   fire an event when a column resize starts and when it ends.

    * Flexibility is great, but often I'm just interested in the simple
   cases...simple. My example here is the multiple-row header stuff. It's
   GREAT! I LOVE it! (And better yet, our customers have been screaming
   for this!) But, I don't always need/want it. And, it can make things
   more complex. One idea would be to overload methods like

[gwt-contrib] Re: Changing JsArrayT extends JavaScriptObject to JsArrayT

2009-06-11 Thread Joel Webber

+1 Ray. Now here's the really tricky question. Is there any way we can take
advantage of Javascript's for (x in y) { ... } syntax (and should we,
given its spotty performance)? My intuition tells me that the only way we
could use it would be through a callback, because there's nothing like .NET
generators/yield in Javascript.
On Wed, Jun 10, 2009 at 7:26 PM, Ray Cromwell cromwell...@gmail.com wrote:

 My take on this is that there is many places where I'd like to avoid
 JRE collections, but the basic JsArray is too much of a downgrade. I
 don't mind changing it to T if it doesn't effect performance cause
 then I could subclass it, but as an example of the stuff I would like
 in a 'FastArrayList' that is not a collections derivative:

 1) add() instead of just set(length(), item) (e.g. push)
 2) addAll(anotherFastArrayList) (e.g. concat)
 3) splice
 4) toArray() (webmode is reinterpret cast op)
 5) remove
 6) shift/unshift (useful for queue when combined with pop)

 I use the following pattern all over GQuery to avoid JRE collections
 but preserve for-each

 for(Foo f : fooList.elements()) { ... }

 where fooList is my own JsArrayT class, and elements() returns T[]
 via reinterpret cast in webmode.

 -Ray



 On Thu, Jun 11, 2009 at 2:43 AM, Lex Spoonsp...@google.com wrote:
 
  Bah, mis-send.  What I was typing was:
 
 
  I though the point was to get rid of JRE collections?  Anyway, the
  collection in question is used as a queue.  I would hate to see its
  performance get worse when there'
 
  ...when there's a known, straightforward alternative, and when that
  alternative provides a class people have been wanting for separate
  purposes.
 
 
  Lex
 
   
 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Optimize if statements to conditionals and boolean ops if possible

2009-06-11 Thread Joel Webber

FWIW:
http://code.google.com/p/google-web-toolkit/wiki/BooleanIfExpressionOptimization

http://code.google.com/p/google-web-toolkit/wiki/IfToTernaryExpressionOptimization

Thanks again for banging on these. I've been meaning to do it for ages...

On Wed, Jun 10, 2009 at 9:11 PM, mmastrac.altern...@gmail.com wrote:


 Reviewers: scottb,

 Description:
 As suggested by scottb, we can optimize certain if statements to their
 boolean equivalents, ie:

 if (a()) {b()} else {c()}   - a()?b():c()

 if (a()) {b()}  -  a()b()

 if (a()) {} else {b()}  - a()||b()



 Please review this at http://gwt-code-reviews.appspot.com/36801

 Affected files:
   dev/core/src/com/google/gwt/dev/js/JsStaticEval.java
   dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java



 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: ant improvements, round 1

2009-06-11 Thread Joel Webber

w00t indeed. This just saved me serious time this morning already.

On Wed, Jun 10, 2009 at 6:35 PM, Scott Blum sco...@google.com wrote:

 w00t!!


 On Wed, Jun 10, 2009 at 5:47 PM, Freeland Abbott fabb...@google.comwrote:

 As of r5537, my no-change ant build takes 1:55 instead of 19:43, and
 there's still some easy work to do, albeit with obviously diminishing
 returns

 Most of that difference is due to a rather annoying timestamp
 consideration with directory entries in jars; my patch introduces a new Ant
 task, LatestTimeJar, to resolve it.

 The issue is---was---that in general, we jar both
 .../src/com/google/gwt/.../Foo.java and also
 build/out/.../com/google/gwt/.../Foo.class.  The jar file will have one
 directory entry for com/, the existence of which is actually important to
 GWT as Scott pointed out in the first-round review comments.  But the two
 directories have different touch dates, and we archived the first-named,
 which was usually from .../src/..., with an old date by svn.  The second
 build would therefore notice that the *second* instance of com/ was
 newer than the archived com/, and therefore jar it again.  (Because we did
 updates, the entry would have been new after that second cycle.  In some
 cases, notably the servet API classes in alldeps.jar, we had up to four such
 duplicates, though.)  Worse, everything downstream of that error also had to
 be redone... including the samples.






 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] [google-web-toolkit commit] r5538 - Fixes typo that confused __MODULE_NAME with MODULE_FUNC__, leading to


Author: j...@google.com
Date: Thu Jun 11 06:24:20 2009
New Revision: 5538

Modified:
trunk/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js

Log:
Fixes typo that confused __MODULE_NAME__ with __MODULE_FUNC__, leading to
startup failures in pre-1.6-hosted-mode.
Review: http://gwt-code-reviews.appspot.com/37802 (TBR)


Modified: trunk/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js
==
--- trunk/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js  
(original)
+++ trunk/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js Thu Jun 
 
11 06:24:20 2009
@@ -347,7 +347,7 @@
//  onScriptLoad()).
var win = scriptFrame.contentWindow;
if (isHostedMode()) {
-win.name = '__MODULE_NAME__';
+win.name = '__MODULE_FUNC__';
}

// Set this *before* calling doc.write(), because the linux  
hosted-mode

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Changing JsArrayT extends JavaScriptObject to JsArrayT

2009-06-11 Thread John Tamplin

On Thu, Jun 11, 2009 at 7:51 AM, Joel Webber j...@google.com wrote:

 +1 Ray. Now here's the really tricky question. Is there any way we can take
 advantage of Javascript's for (x in y) { ... } syntax (and should we,
 given its spotty performance)? My intuition tells me that the only way we
 could use it would be through a callback, because there's nothing like .NET
 generators/yield in Javascript.


First problem is you have to have some knowledge of the key domain in order
to use the keys as raw JS attributes.  Ie, basically they need to be
integers or strings that don't conflict with particular reserved words like
__proto (or you prefix all the keys, like what is done in FastStringMap).

I think assuming the keys are safe is ok for specific use cases, but not ok
for a general purpose data structure (so I agree with prefixing keys in
FastStringMap while avoiding them in CurrencyList for example).

Regarding the for(x in y) construct, it seems conceivable the compiler could
recognize the code generated by the normal Java-JS compile of for(x :
y.elements()) and rewrite it to the native JS code where appropriate as a
peephole optimization.

-- 
John A. Tamplin
Software Engineer (GWT), Google

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: ant improvements, round 1

2009-06-11 Thread Freeland Abbott

Well, if I've saved serious time by 10:30am, I'm happy indeed.
I've got another depends-on-your-hardware-but-I-saw-4min-saving (for
work-to-do rebuild of samples, so no gain if you use buildonly) out to scott
already, though it's small enough that anyone who wants to review at
http://gwt-code-reviews.appspot.com/36802/show can help Scott do real work
instead of ant file review.



2009/6/11 Joel Webber j...@google.com

 w00t indeed. This just saved me serious time this morning already.


 On Wed, Jun 10, 2009 at 6:35 PM, Scott Blum sco...@google.com wrote:

 w00t!!


 On Wed, Jun 10, 2009 at 5:47 PM, Freeland Abbott fabb...@google.comwrote:

 As of r5537, my no-change ant build takes 1:55 instead of 19:43, and
 there's still some easy work to do, albeit with obviously diminishing
 returns

 Most of that difference is due to a rather annoying timestamp
 consideration with directory entries in jars; my patch introduces a new Ant
 task, LatestTimeJar, to resolve it.

 The issue is---was---that in general, we jar both
 .../src/com/google/gwt/.../Foo.java and also
 build/out/.../com/google/gwt/.../Foo.class.  The jar file will have one
 directory entry for com/, the existence of which is actually important to
 GWT as Scott pointed out in the first-round review comments.  But the two
 directories have different touch dates, and we archived the first-named,
 which was usually from .../src/..., with an old date by svn.  The second
 build would therefore notice that the *second* instance of com/ was
 newer than the archived com/, and therefore jar it again.  (Because we did
 updates, the entry would have been new after that second cycle.  In some
 cases, notably the servet API classes in alldeps.jar, we had up to four such
 duplicates, though.)  Worse, everything downstream of that error also had to
 be redone... including the samples.









 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Changing JsArrayT extends JavaScriptObject to JsArrayT




On 11 juin, 16:11, John Tamplin j...@google.com wrote:
 On Thu, Jun 11, 2009 at 7:51 AM, Joel Webber j...@google.com wrote:
  +1 Ray. Now here's the really tricky question. Is there any way we can take
  advantage of Javascript's for (x in y) { ... } syntax (and should we,
  given its spotty performance)? My intuition tells me that the only way we
  could use it would be through a callback, because there's nothing like .NET
  generators/yield in Javascript.


I'm afraid I don't understand the relationship between JsArray/
JsArrayList and for (x in y) { ... }

https://developer.mozilla.org/en/Core_JavaScript_1.5_Reference/Statements/for...in
Although it may be tempting to use this as a way to iterate over an
Array, this is a bad idea. The for...in statement iterates over user-
defined properties in addition to the array elements, so if you modify
the array's non-integer or non-positive properties (e.g. by adding a
foo property to it or even by adding a method or property to
Array.prototype), the for...in statement will return the name of your
user-defined properties in addition to the numeric indexes. Also,
because order of iteration is arbitrary, iterating over an array may
not visit elements in numeric order. Thus it is better to use a
traditional for loop with a numeric index when iterating over arrays.
Similar arguments might be used against even using for...in at all (at
least without propertyIsEnumerable() or hasOwnProperty() checks),
since it will also iterate over Object.prototype (which, though
usually discouraged, can, as in the case of Array.prototype, be
usefully extended by the user where are no namespacing concerns caused
by inclusion of other libraries which might not perform the above
checks on such iterations and where they are aware of the effect such
extension will have on their own use of iterators such as
for...in).

And Nicholas C. Zakas recently confirmed that for (var i=0, l =
arr.length; i  l; i++) { ... } is the fastest way of iterating
through an array's values: 
http://google-code-updates.blogspot.com/2009/06/nicholas-c-zakas-speed-up-your.html
(starts at slide #40)
(and this construct is the one generated by GWT from a Java for (T
i : arr) { ... }, thus also with Ray's proposed for (T i :
jsarr.elements()) { ... }; but beware if the jsarr can be modified
--in length-- during the loop!)

See also the toArray()s in
http://code.google.com/p/gwt-in-the-air/source/browse/trunk/src/net/ltgt/gwt/jscollections/client/JsArrays.java

Those methods (toArray(), elements()) are for people who really know
what they do, so I'm not sure they so should be included in the core
GWT... (the toJavaArray() from some Utils class in GALGWT always
copies the array, which is slower but always safe)

 Regarding the for(x in y) construct, it seems conceivable the compiler could
 recognize the code generated by the normal Java-JS compile of for(x :
 y.elements()) and rewrite it to the native JS code where appropriate as a
 peephole optimization.

What the compiler could eventually do is optimize the l=arr.length
out of the for-loop when used several times in a row. E.g.

void f(T[] arr)
{
   this.sb = new StringBuilder();
   for (T i : arr) { this.sb.append(i.toString()); }
   this.count = 0;
   for (T i : arr) { this.count++; }
}

to be compiled into (pseudo):
function f(arr)
{
   var l = arr.length;
   this.sb = new StringBuilder();
   for (var i = 0; i  l; i++) { this.sb.append(arr[i].toString()); }
   this.count = 0;
   for (var i = 0; i  l; i++) { this.count++; }
}

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Changing JsArrayT extends JavaScriptObject to JsArrayT

2009-06-11 Thread Stefan Haustein

On Thu, Jun 11, 2009 at 3:42 PM, Thomas Broyer t.bro...@gmail.com wrote:


 On 11 juin, 16:11, John Tamplin j...@google.com wrote:
  On Thu, Jun 11, 2009 at 7:51 AM, Joel Webber j...@google.com wrote:
   +1 Ray. Now here's the really tricky question. Is there any way we can
 take
   advantage of Javascript's for (x in y) { ... } syntax (and should we,
   given its spotty performance)? My intuition tells me that the only way
 we
   could use it would be through a callback, because there's nothing like
 .NET
   generators/yield in Javascript.


 I'm afraid I don't understand the relationship between JsArray/
 JsArrayList and for (x in y) { ... }


Javascript arrays are sparse. There are use case (such as protocol buffers
with extension fields) where this is relevant and for(x in y) may be faster.
I agree we should avoid this in the general case, though.

https://developer.mozilla.org/en/Core_JavaScript_1.5_Reference/Statements/for...in
 Although it may be tempting to use this as a way to iterate over an
 Array, this is a bad idea. The for...in statement iterates over user-
 defined properties in addition to the array elements, so if you modify
 the array's non-integer or non-positive properties (e.g. by adding a
 foo property to it or even by adding a method or property to
 Array.prototype), the for...in statement will return the name of your
 user-defined properties in addition to the numeric indexes. Also,
 because order of iteration is arbitrary, iterating over an array may
 not visit elements in numeric order. Thus it is better to use a
 traditional for loop with a numeric index when iterating over arrays.
 Similar arguments might be used against even using for...in at all (at
 least without propertyIsEnumerable() or hasOwnProperty() checks),
 since it will also iterate over Object.prototype (which, though
 usually discouraged, can, as in the case of Array.prototype, be
 usefully extended by the user where are no namespacing concerns caused
 by inclusion of other libraries which might not perform the above
 checks on such iterations and where they are aware of the effect such
 extension will have on their own use of iterators such as
 for...in).

 And Nicholas C. Zakas recently confirmed that for (var i=0, l =
 arr.length; i  l; i++) { ... } is the fastest way of iterating
 through an array's values:
 http://google-code-updates.blogspot.com/2009/06/nicholas-c-zakas-speed-up-your.html
 (starts at slide #40)
 (and this construct is the one generated by GWT from a Java for (T
 i : arr) { ... }, thus also with Ray's proposed for (T i :
 jsarr.elements()) { ... }; but beware if the jsarr can be modified
 --in length-- during the loop!)

 See also the toArray()s in

 http://code.google.com/p/gwt-in-the-air/source/browse/trunk/src/net/ltgt/gwt/jscollections/client/JsArrays.java

 Those methods (toArray(), elements()) are for people who really know
 what they do, so I'm not sure they so should be included in the core
 GWT... (the toJavaArray() from some Utils class in GALGWT always
 copies the array, which is slower but always safe)

  Regarding the for(x in y) construct, it seems conceivable the compiler
 could
  recognize the code generated by the normal Java-JS compile of for(x :
  y.elements()) and rewrite it to the native JS code where appropriate as a
  peephole optimization.

 What the compiler could eventually do is optimize the l=arr.length
 out of the for-loop when used several times in a row. E.g.

 void f(T[] arr)
 {
   this.sb = new StringBuilder();
   for (T i : arr) { this.sb.append(i.toString()); }
   this.count = 0;
   for (T i : arr) { this.count++; }
 }

 to be compiled into (pseudo):
 function f(arr)
 {
   var l = arr.length;
   this.sb = new StringBuilder();
   for (var i = 0; i  l; i++) { this.sb.append(arr[i].toString()); }
   this.count = 0;
   for (var i = 0; i  l; i++) { this.count++; }
 }

 



-- 
Stefan Haustein
Google UK Limited

Registered Office: Belgrave House, 76 Buckingham Palace Road, London SW1W
9TQ; Registered in England Number: 3977902

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: ant improvements, round 1

2009-06-11 Thread Rajeev Dayal

Nice job Freeland! You're an ant-master!

On Thu, Jun 11, 2009 at 10:40 AM, Freeland Abbott fabb...@google.comwrote:

 Well, if I've saved serious time by 10:30am, I'm happy indeed.
 I've got another depends-on-your-hardware-but-I-saw-4min-saving (for
 work-to-do rebuild of samples, so no gain if you use buildonly) out to scott
 already, though it's small enough that anyone who wants to review at
 http://gwt-code-reviews.appspot.com/36802/show can help Scott do real work
 instead of ant file review.



 2009/6/11 Joel Webber j...@google.com

 w00t indeed. This just saved me serious time this morning already.


 On Wed, Jun 10, 2009 at 6:35 PM, Scott Blum sco...@google.com wrote:

 w00t!!


 On Wed, Jun 10, 2009 at 5:47 PM, Freeland Abbott fabb...@google.comwrote:

 As of r5537, my no-change ant build takes 1:55 instead of 19:43, and
 there's still some easy work to do, albeit with obviously diminishing
 returns

 Most of that difference is due to a rather annoying timestamp
 consideration with directory entries in jars; my patch introduces a new Ant
 task, LatestTimeJar, to resolve it.

 The issue is---was---that in general, we jar both
 .../src/com/google/gwt/.../Foo.java and also
 build/out/.../com/google/gwt/.../Foo.class.  The jar file will have one
 directory entry for com/, the existence of which is actually important to
 GWT as Scott pointed out in the first-round review comments.  But the two
 directories have different touch dates, and we archived the first-named,
 which was usually from .../src/..., with an old date by svn.  The second
 build would therefore notice that the *second* instance of com/ was
 newer than the archived com/, and therefore jar it again.  (Because we 
 did
 updates, the entry would have been new after that second cycle.  In some
 cases, notably the servet API classes in alldeps.jar, we had up to four 
 such
 duplicates, though.)  Worse, everything downstream of that error also had 
 to
 be redone... including the samples.












 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Changing JsArrayT extends JavaScriptObject to JsArrayT

2009-06-11 Thread John Tamplin

On Thu, Jun 11, 2009 at 10:42 AM, Thomas Broyer t.bro...@gmail.com wrote:

 On 11 juin, 16:11, John Tamplin j...@google.com wrote:
  On Thu, Jun 11, 2009 at 7:51 AM, Joel Webber j...@google.com wrote:
   +1 Ray. Now here's the really tricky question. Is there any way we can
 take
   advantage of Javascript's for (x in y) { ... } syntax (and should we,
   given its spotty performance)? My intuition tells me that the only way
 we
   could use it would be through a callback, because there's nothing like
 .NET
   generators/yield in Javascript.


 I'm afraid I don't understand the relationship between JsArray/
 JsArrayList and for (x in y) { ... }


I was thinking more of a Map implementation to iterate over the key set.

-- 
John A. Tamplin
Software Engineer (GWT), Google

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Optimize if statements to conditionals and boolean ops if possible

2009-06-11 Thread Matt Mastracci

No problem...  I haven't handled the common super-expression case (if  
(test) { return a; } else {return b; } - return test ? a : b), but  
that one should be pretty easy to tackle for a limited set of common  
super-expressions.

On 11-Jun-09, at 5:53 AM, Joel Webber wrote:

 FWIW:
   
 http://code.google.com/p/google-web-toolkit/wiki/BooleanIfExpressionOptimization
   
 http://code.google.com/p/google-web-toolkit/wiki/IfToTernaryExpressionOptimization

 Thanks again for banging on these. I've been meaning to do it for  
 ages...

 On Wed, Jun 10, 2009 at 9:11 PM, mmastrac.altern...@gmail.com wrote:

 Reviewers: scottb,

 Description:
 As suggested by scottb, we can optimize certain if statements to their
 boolean equivalents, ie:

 if (a()) {b()} else {c()}   - a()?b():c()

 if (a()) {b()}  -  a()b()

 if (a()) {} else {b()}  - a()||b()



 Please review this at http://gwt-code-reviews.appspot.com/36801

 Affected files:
   dev/core/src/com/google/gwt/dev/js/JsStaticEval.java
   dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java






 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: ant improvements, round 1

2009-06-11 Thread Freeland Abbott

Thanks... I think.

On Thu, Jun 11, 2009 at 11:53 AM, Rajeev Dayal rda...@google.com wrote:

 Nice job Freeland! You're an ant-master!


 On Thu, Jun 11, 2009 at 10:40 AM, Freeland Abbott fabb...@google.comwrote:

 Well, if I've saved serious time by 10:30am, I'm happy indeed.
  I've got another depends-on-your-hardware-but-I-saw-4min-saving (for
 work-to-do rebuild of samples, so no gain if you use buildonly) out to scott
 already, though it's small enough that anyone who wants to review at
 http://gwt-code-reviews.appspot.com/36802/show can help Scott do real
 work instead of ant file review.



 2009/6/11 Joel Webber j...@google.com

 w00t indeed. This just saved me serious time this morning already.


 On Wed, Jun 10, 2009 at 6:35 PM, Scott Blum sco...@google.com wrote:

 w00t!!


 On Wed, Jun 10, 2009 at 5:47 PM, Freeland Abbott fabb...@google.comwrote:

 As of r5537, my no-change ant build takes 1:55 instead of 19:43, and
 there's still some easy work to do, albeit with obviously diminishing
 returns

 Most of that difference is due to a rather annoying timestamp
 consideration with directory entries in jars; my patch introduces a new 
 Ant
 task, LatestTimeJar, to resolve it.

 The issue is---was---that in general, we jar both
 .../src/com/google/gwt/.../Foo.java and also
 build/out/.../com/google/gwt/.../Foo.class.  The jar file will have one
 directory entry for com/, the existence of which is actually important 
 to
 GWT as Scott pointed out in the first-round review comments.  But the two
 directories have different touch dates, and we archived the first-named,
 which was usually from .../src/..., with an old date by svn.  The second
 build would therefore notice that the *second* instance of com/ was
 newer than the archived com/, and therefore jar it again.  (Because we 
 did
 updates, the entry would have been new after that second cycle.  In some
 cases, notably the servet API classes in alldeps.jar, we had up to four 
 such
 duplicates, though.)  Worse, everything downstream of that error also had 
 to
 be redone... including the samples.















 


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] reworking ant targets to allow single-platform use

2009-06-11 Thread fabbott


Reviewers: scottb,

Description:
Shaves 10min off a dist-one build vs dist (aka old build).

Please review this at http://gwt-code-reviews.appspot.com/37803

Affected files:
   build.xml



--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: initial load sequence for runAsync's

2009-06-11 Thread spoon


Thanks, Bob!  I have several things to change; my plans are in the
line-by-line comments.


http://gwt-code-reviews.appspot.com/33848/diff/1/4
File
dev/core/src/com/google/gwt/core/ext/soyc/impl/SplitPointRecorder.java
(right):

http://gwt-code-reviews.appspot.com/33848/diff/1/4#newcode107
Line 107: MapInteger, String names = new HashMapInteger, String();
I'll sort them.  It will work to have recordSplitPoints() loop from
1..size instead of looping over the map's key set.

http://gwt-code-reviews.appspot.com/33848/diff/1/5
File dev/core/src/com/google/gwt/core/linker/SoycReportLinker.java
(right):

http://gwt-code-reviews.appspot.com/33848/diff/1/5#newcode43
Line 43: if (soycFiles.getDepFile() != null) {
It's a new requirement.  A StandardCompilationAnalysis is now always
added even when -soyc is turned off, because it's both useful and fast
to generate.  When -soyc is off, the SCA will include the split point
map but not the other two files.

http://gwt-code-reviews.appspot.com/33848/diff/1/10
File dev/core/src/com/google/gwt/dev/jjs/JavaToJavaScriptCompiler.java
(right):

http://gwt-code-reviews.appspot.com/33848/diff/1/10#newcode327
Line 327:
Will do.

http://gwt-code-reviews.appspot.com/33848/diff/1/10#newcode363
Line 363: System.out.println(Completed SOYC phase in 
Will do.  I missed that one.

http://gwt-code-reviews.appspot.com/33848/diff/1/11
File dev/core/src/com/google/gwt/dev/jjs/ast/JProgram.java (right):

http://gwt-code-reviews.appspot.com/33848/diff/1/11#newcode231
Line 231: private MapInteger, RunAsyncReplacement runAsyncReplacements
= new HashMapInteger, RunAsyncReplacement();
It could be d.u.c.Maps.create().  The setter might not be called, in
which case an empty map leads to simpler code.

http://gwt-code-reviews.appspot.com/33848/diff/1/11#newcode1051
Line 1051: splitPointInitialSequence.addAll(list);
It should only be called once.  I'll add an assert that the existing
sequence is empty.

http://gwt-code-reviews.appspot.com/33848/diff/1/6
File dev/core/src/com/google/gwt/dev/jjs/impl/CodeSplitter.java (right):

http://gwt-code-reviews.appspot.com/33848/diff/1/6#newcode340
Line 340: * TODO(spoon) accept a labeled runAsync call, once runAsyncs
can be labeled
Some kind of annotation at the runAsync call site.  The details are TBD.

http://gwt-code-reviews.appspot.com/33848/diff/1/6#newcode349
Line 349: + PROP_INITIAL_SEQUENCE + :  + refString);
Yes!

http://gwt-code-reviews.appspot.com/33848/diff/1/6#newcode380
Line 380: Method includes multiple runAsync calls, so it's ambiguous
which one is meant: 
On 2009/06/10 23:08:17, bobv wrote:
 Split the string literal to improve formatting.

Sure.


 This happens after pruning, right?

I'll pull up the code and double check, but it shouldn't be called after
optimization starts.  It should run pre-optimization, so that the jsni
references can actually be looked up.  This method converts the JSNI
reference to an integer, which will be stable for the whole compile.

http://gwt-code-reviews.appspot.com/33848/diff/1/6#newcode451
Line 451: MapJMethod, ListInteger revmap = new
java.util.HashMapJMethod, ListInteger();
I'll change it to use GWT's hash map.

http://gwt-code-reviews.appspot.com/33848/diff/1/8
File dev/core/src/com/google/gwt/dev/jjs/impl/ReplaceRunAsyncs.java
(right):

http://gwt-code-reviews.appspot.com/33848/diff/1/8#newcode36
Line 36: *  by calls to a fragment loader.
Will delete.

http://gwt-code-reviews.appspot.com/33848

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: configurable -localWorkers for samples

2009-06-11 Thread scottb


LGTM, we've been needing this for a while. :)

http://gwt-code-reviews.appspot.com/36802

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: reworking ant targets to allow single-platform use

2009-06-11 Thread galgwt . reviews



http://gwt-code-reviews.appspot.com/37803/diff/1/2
File build.xml (right):

http://gwt-code-reviews.appspot.com/37803/diff/1/2#newcode21
Line 21: target name=dist-one depends=buildonly, tools, samples,
doc description=Make this platform's distributions
Now would be a good opportunity to add comments to document what these
different targets are intended to do.  I like to have a summary of the
'user oriented' targets at the top of the ant file.  I think that 'one'
refers to build a distribution for just one platform, but someone else
might think it means building one permutation of the samples or
something like that.

http://gwt-code-reviews.appspot.com/37803/diff/1/2#newcode72
Line 72: antcall target=buildtools
Maybe a macrodef would help shrink the amount of code after getting rid
of the -do construct?

macrodef name=call-checkstyle
   attribute name=target /
   sequential
 antcall targ...@{target}
   param name=target value=checkstyle /
 /antcall
   /sequential
/macrodef

   call-checkstyle target=buildtools /
   call-checkstyle target=dev /
   ...

(I wish I could find something like a for loop)

http://gwt-code-reviews.appspot.com/37803/diff/1/2#newcode93
Line 93: antcall target=buildtools
same here - consider a macrodef.

http://gwt-code-reviews.appspot.com/37803

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: reworking ant targets to allow single-platform use

2009-06-11 Thread fabbott



http://gwt-code-reviews.appspot.com/37803/diff/1/2
File build.xml (right):

http://gwt-code-reviews.appspot.com/37803/diff/1/2#newcode21
Line 21: target name=dist-one depends=buildonly, tools, samples,
doc description=Make this platform's distributions
On 2009/06/11 17:41:56, zundel wrote:
 Now would be a good opportunity to add comments to document what these
different
 targets are intended to do.  I like to have a summary of the 'user
oriented'
 targets at the top of the ant file.  I think that 'one' refers to
build a
 distribution for just one platform, but someone else might think it
means
 building one permutation of the samples or something like that.

I was actually planning to resort them (separately, and after this
approval, rather than introducing gratuitous diffs) so that ant -p
would have the sorting you describe, yes.  (Currently it's at least
mostly alphabetical, which is readily computable but not particularly
useful...)

And the description is intended to suggest one - this platform,
though I can obviously rework that wording.  I'd rather have a good
description than rely on comments (which, after all, might not be read,
aren't displayed by ant -p, etc.)

http://gwt-code-reviews.appspot.com/37803/diff/1/2#newcode72
Line 72: antcall target=buildtools
On 2009/06/11 17:41:56, zundel wrote:
 Maybe a macrodef would help shrink the amount of code after getting
rid of the
 -do construct?

 macrodef name=call-checkstyle
attribute name=target /
sequential
  antcall targ...@{target}
param name=target value=checkstyle /
  /antcall
/sequential
 /macrodef

call-checkstyle target=buildtools /
call-checkstyle target=dev /
...

 (I wish I could find something like a for loop)

There's a foreach in ant-contrib, but it invokes a target as its body,
which I've always found really annoying and counter-clarifying.  But
yes, we can make a macro (and parameterize both the subproject, what
you're calling target there, and the target within it, so the macro can
be reused below).

http://gwt-code-reviews.appspot.com/37803

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] [google-web-toolkit commit] r5539 - Merge trunk 5538 to the 6/2 snapshot branch.


Author: j...@google.com
Date: Thu Jun 11 11:36:32 2009
New Revision: 5539

Modified:
branches/snapshot-2009.06.02-r5498/branch-info.txt
 
branches/snapshot-2009.06.02-r5498/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js

Log:
Merge trunk 5538 to the 6/2 snapshot branch.


r5538 | j...@google.com | 2009-06-11 09:24:20 -0400 (Thu, 11 Jun 2009) | 4  
lines

Fixes typo that confused __MODULE_NAME__ with __MODULE_FUNC__, leading to
startup failures in pre-1.6-hosted-mode.
Review: http://gwt-code-reviews.appspot.com/37802 (TBR)


svn merge -c5538 https://google-web-toolkit.googlecode.com/svn/trunk .


Modified: branches/snapshot-2009.06.02-r5498/branch-info.txt
==
--- branches/snapshot-2009.06.02-r5498/branch-info.txt  (original)
+++ branches/snapshot-2009.06.02-r5498/branch-info.txt  Thu Jun 11 11:36:32  
2009
@@ -24,3 +24,5 @@
 svn merge -c5530 https://google-web-toolkit.googlecode.com/svn/trunk .
  /trunk 5532 was merged into this branch (yet more fixes from 5523)
 svn merge -c5532 https://google-web-toolkit.googlecode.com/svn/trunk .
+/trunk 5538 was merged into this branch (final(?) fix from 5523)
+   svn merge -c5538 https://google-web-toolkit.googlecode.com/svn/trunk .

Modified:  
branches/snapshot-2009.06.02-r5498/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js
==
---  
branches/snapshot-2009.06.02-r5498/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js
 
(original)
+++  
branches/snapshot-2009.06.02-r5498/dev/core/src/com/google/gwt/core/linker/IFrameTemplate.js
 
Thu Jun 11 11:36:32 2009
@@ -347,7 +347,7 @@
//  onScriptLoad()).
var win = scriptFrame.contentWindow;
if (isHostedMode()) {
-win.name = '__MODULE_NAME__';
+win.name = '__MODULE_FUNC__';
}

// Set this *before* calling doc.write(), because the linux  
hosted-mode

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: initial load sequence for runAsync's

2009-06-11 Thread spoon


How about now?  In addition to the requested changes, I changed two
other things:

1. The SOYC timings use PerfLogger now, rather than inlined timing code
2. There is a check for the same split point being specified multiple
times in the initial sequence



http://gwt-code-reviews.appspot.com/33848

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] [google-web-toolkit commit] r5540 - JsStaticEval now optimizes certain if statements to conditional operators.


Author: sco...@google.com
Date: Thu Jun 11 12:08:45 2009
New Revision: 5540

Modified:
trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java
trunk/dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java

Log:
JsStaticEval now optimizes certain if statements to conditional operators.

if (a()) {b()} else {c()}   - a()?b():c()

if (a()) {b()}  -  a()b()

if (a()) {} else {b()}  - a()||b()

Patch by: mmastrac
Suggested by: me
Review by: me

Modified: trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java
==
--- trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java  (original)
+++ trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java  Thu Jun 11  
12:08:45 2009
@@ -352,32 +352,50 @@
} else {
  boolean thenIsEmpty = isEmpty(thenStmt);
  boolean elseIsEmpty = isEmpty(elseStmt);
-
+JsExpression thenExpr = extractExpression(thenStmt);
+JsExpression elseExpr = extractExpression(elseStmt);
+
  if (thenIsEmpty  elseIsEmpty) {
+  // Convert if (a()) {} = a().
ctx.replaceMe(expr.makeStmt());
+} else if (thenExpr != null  elseExpr != null) {
+  // Convert if (a()) {b()} else {c()} = a()?b():c().
+  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
+  Replaced if statement with conditional);
+
+  JsConditional cond = new JsConditional(sourceInfo, x.getIfExpr(),
+  thenExpr, elseExpr);
+  ctx.replaceMe(accept(cond.makeStmt()));
+} else if (thenIsEmpty  elseExpr != null) {
+  // Convert if (a()) {} else {b()} = a()||b().
+  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
+  Replaced if statement with ||);
+
+  JsBinaryOperation op = new JsBinaryOperation(sourceInfo,
+  JsBinaryOperator.OR, x.getIfExpr(), elseExpr);
+  ctx.replaceMe(accept(op.makeStmt()));
  } else if (thenIsEmpty  !elseIsEmpty) {
-  /*
-   * If the then block is blank, but the else statement has  
statements,
-   * invert the test
-   */
+  // Convert if (a()) {} else {stuff} = if (!a()) {stuff}.
sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
Simplified if with empty then statement);

JsUnaryOperation negatedOperation = new  
JsPrefixOperation(sourceInfo,
JsUnaryOperator.NOT, x.getIfExpr());
JsIf newIf = new JsIf(sourceInfo, negatedOperation, elseStmt,  
null);
-
ctx.replaceMe(accept(newIf));
+} else if (elseIsEmpty  thenExpr != null) {
+  // Convert if (a()) {b()} = a()b().
+  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
+  Replaced if statement with );
+
+  JsBinaryOperation op = new JsBinaryOperation(sourceInfo,
+  JsBinaryOperator.AND, x.getIfExpr(), thenExpr);
+  ctx.replaceMe(accept(op.makeStmt()));
  } else if (elseIsEmpty  elseStmt != null) {
-  /*
-   * If the else statement is present but has no effective  
statements,
-   * prune it
-   */
+  // Convert if (a()) {b()} else {} = if (a()) {b()}.
sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
Pruned empty else statement);
-
JsIf newIf = new JsIf(sourceInfo, x.getIfExpr(), thenStmt, null);
-
ctx.replaceMe(accept(newIf));
  }
}
@@ -583,6 +601,26 @@

public static boolean exec(JsProgram program) {
  return (new JsStaticEval(program)).execImpl();
+  }
+
+  /**
+   * Attempts to extract a single expression from a given statement and  
returns
+   * it. If no such expression exists, returns codenull/code.
+   */
+  protected static JsExpression extractExpression(JsStatement stmt) {
+if (stmt == null) {
+  return null;
+}
+
+if (stmt instanceof JsExprStmt) {
+  return ((JsExprStmt) stmt).getExpression();
+}
+
+if (stmt instanceof JsBlock  ((JsBlock) stmt).getStatements().size()  
== 1) {
+  return extractExpression(((JsBlock) stmt).getStatements().get(0));
+}
+
+return null;
}

protected static boolean isEmpty(JsStatement stmt) {

Modified: trunk/dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java
==
--- trunk/dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java  
(original)
+++ trunk/dev/core/test/com/google/gwt/dev/js/JsStaticEvalTest.java Thu Jun 
 
11 12:08:45 2009
@@ -33,8 +33,13 @@
  assertEquals(a();, optimize(if (a()) { }));
}

+  public void testIfWithEmptyThenAndElseExpression() throws Exception {
+assertEquals(a()||b();, optimize(if (a()) { } else { b(); }));
+  }
+
public void testIfWithEmptyThenAndElse() throws Exception {
-

[gwt-contrib] -draftCompile optimizes less

2009-06-11 Thread spoon


Reviewers: scottb,

Description:
In trunk, -draftCompile does one full outer loop of the optimization
loop.  This patch pares that one loop down much further.

A one-permutation draft compile of Showcase now takes about 25 seconds,
as compared to 27 seconds before.  Additionally, the actual optimization
step is reduced from about 2.5 seconds to about 500 milliseconds.

Most of the remaining optimization time is in MakeCallsStatic, and that
would be possible to pare down further.  The only calls that really need
to be static are those to methods in a few key classes including String
and Object. I coded that up, but backed it out of this patch because it
looks too complicated for the speedup involved.

At this point the bottleneck looks like the front end: running the JDT
type check and converting the results to GWT internal ASTs.  Here are
some key parts of a perf log on my machine:

ModuleDef.normalize 764ms
CompilationState.compile 8837ms
Precompile 9759ms
   Build AST 2877ms
draft optimize 549ms

The entire compile takes 24424ms.



Please review this at http://gwt-code-reviews.appspot.com/38801

Affected files:
   dev/core/src/com/google/gwt/dev/jjs/JavaToJavaScriptCompiler.java


Index: dev/core/src/com/google/gwt/dev/jjs/JavaToJavaScriptCompiler.java
===
--- dev/core/src/com/google/gwt/dev/jjs/JavaToJavaScriptCompiler.java
(revision 5538)
+++ dev/core/src/com/google/gwt/dev/jjs/JavaToJavaScriptCompiler.java
(working copy)
@@ -366,11 +366,11 @@

  toReturn.getArtifacts().add(
  new StandardCompilationAnalysis(dependencies, stories,  
splitPoints));
-
+
  System.out.println(Completed SOYC phase in 
  + (System.currentTimeMillis() - soycStart) +  ms);
}
-
+
System.out.println(Permutation took 
+ (System.currentTimeMillis() - permStart) +  ms);
return toReturn;
@@ -542,6 +542,10 @@
  }
}

+  /**
+   * Perform the minimal amount of optimization to make sure the compile
+   * succeeds.
+   */
protected static void draftOptimize(JProgram jprogram) {
  /*
   * Record the beginning of optimizations; this turns on certain checks  
that
@@ -550,14 +554,17 @@
   */
  jprogram.beginOptimizations();

-optimizeLoop(jprogram, false);
+PerfLogger.start(draft optimize);

-/*
- * Ensure that references to dead clinits are removed. Otherwise, the
- * application won't run reliably.
- */
-jprogram.typeOracle.recomputeAfterOptimizations();
-DeadCodeElimination.exec(jprogram);
+PerfLogger.start(Finalizer);
+Finalizer.exec(jprogram);
+PerfLogger.end();
+
+PerfLogger.start(MakeCallsStatic);
+MakeCallsStatic.exec(jprogram);
+PerfLogger.end();
+
+PerfLogger.end();
}

protected static void optimize(JJSOptions options, JProgram jprogram)



--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Optimize if statements to conditionals and boolean ops if possible

2009-06-11 Thread scottb


LGTM.  Committed at r5540 with minor tweaks.

http://gwt-code-reviews.appspot.com/36801

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Optimize if statements to conditionals and boolean ops if possible

2009-06-11 Thread Scott Blum

Matt,
Could you review these two follow-on patches?  The second depends on the
first.  This is mostly non-functional cleanup to make things a little
simpler to read. The only substantive change is this:

Presently, if we have a CanBooleanEval that fails to resolve, we do nothing
else.  My second patch allows the failure case to fall through and check for
more simplifications.

Thanks,
Scott

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---



cleanup2.patch
Description: Binary data


cleanup1.patch
Description: Binary data

[gwt-contrib] Re: -draftCompile optimizes less

2009-06-11 Thread scottb


LGTM.

By the way, instant hosted mode should reduce CompilationState.compile
drastically, so help is on the way!


http://gwt-code-reviews.appspot.com/38801

--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] Re: Optimize if statements to conditionals and boolean ops if possible

2009-06-11 Thread Matt Mastracci


Both LGTM.  Good catch, thanks.  :)

On 11-Jun-09, at 1:15 PM, Scott Blum wrote:

 Matt,

 Could you review these two follow-on patches?  The second depends on  
 the first.  This is mostly non-functional cleanup to make things a  
 little simpler to read. The only substantive change is this:

 Presently, if we have a CanBooleanEval that fails to resolve, we do  
 nothing else.  My second patch allows the failure case to fall  
 through and check for more simplifications.


--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---

[gwt-contrib] [google-web-toolkit commit] r5541 - Simple refactoring of SourceInfo generation in JsStaticEval.


Author: sco...@google.com
Date: Thu Jun 11 13:01:10 2009
New Revision: 5541

Modified:
trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java

Log:
Simple refactoring of SourceInfo generation in JsStaticEval.

Review by: mmastrac

Modified: trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java
==
--- trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java  (original)
+++ trunk/dev/core/src/com/google/gwt/dev/js/JsStaticEval.java  Thu Jun 11  
13:01:10 2009
@@ -15,6 +15,7 @@
   */
  package com.google.gwt.dev.js;

+import com.google.gwt.dev.jjs.HasSourceInfo;
  import com.google.gwt.dev.jjs.SourceInfo;
  import com.google.gwt.dev.js.ast.CanBooleanEval;
  import com.google.gwt.dev.js.ast.JsBinaryOperation;
@@ -222,17 +223,14 @@
if (condExpr instanceof CanBooleanEval) {
  CanBooleanEval condEval = (CanBooleanEval) condExpr;
  if (condEval.isBooleanTrue()) {
-  // e.g. (true() ? then : else) - true()  then
-  JsBinaryOperation binOp = new JsBinaryOperation(
-  x.getSourceInfo().makeChild(StaticEvalVisitor.class,
-  Simplified always-true condition),  
JsBinaryOperator.AND,
+  JsBinaryOperation binOp = new JsBinaryOperation(makeSourceInfo(x,
+  Simplified always-true condition), JsBinaryOperator.AND,
condExpr, thenExpr);
ctx.replaceMe(accept(binOp));
  } else if (condEval.isBooleanFalse()) {
// e.g. (false() ? then : else) - false() || else
-  JsBinaryOperation binOp = new JsBinaryOperation(
-  x.getSourceInfo().makeChild(StaticEvalVisitor.class,
-  Simplified always-false condition),  
JsBinaryOperator.OR,
+  JsBinaryOperation binOp = new JsBinaryOperation(makeSourceInfo(x,
+  Simplified always-false condition), JsBinaryOperator.OR,
condExpr, elseExpr);
ctx.replaceMe(accept(binOp));
  }
@@ -256,8 +254,8 @@
FindBreakContinueStatementsVisitor visitor = new  
FindBreakContinueStatementsVisitor();
visitor.accept(x.getBody());
if (!visitor.hasBreakContinueStatements()) {
-JsBlock block = new JsBlock(x.getSourceInfo().makeChild(
-StaticEvalVisitor.class, Simplified always-false  
condition));
+JsBlock block = new JsBlock(makeSourceInfo(x,
+Simplified always-false condition));
  block.getStatements().add(x.getBody());
  block.getStatements().add(expr.makeStmt());
  ctx.replaceMe(accept(block));
@@ -290,8 +288,8 @@

  // If false, replace with initializers and condition.
  if (cond.isBooleanFalse()) {
-  JsBlock block = new JsBlock(x.getSourceInfo().makeChild(
-  StaticEvalVisitor.class, Simplified always-false  
condition));
+  JsBlock block = new JsBlock(makeSourceInfo(x,
+  Simplified always-false condition));
if (x.getInitExpr() != null) {
  block.getStatements().add(x.getInitExpr().makeStmt());
}
@@ -318,21 +316,19 @@
JsExpression expr = x.getIfExpr();
JsStatement thenStmt = x.getThenStmt();
JsStatement elseStmt = x.getElseStmt();
-  SourceInfo sourceInfo;
if (expr instanceof CanBooleanEval) {
+SourceInfo sourceInfo;
  CanBooleanEval cond = (CanBooleanEval) expr;
  JsStatement onlyStmtToExecute;
  JsStatement removed;
  if (cond.isBooleanTrue()) {
onlyStmtToExecute = thenStmt;
removed = elseStmt;
-  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
-  Simplified always-true condition);
+  sourceInfo = makeSourceInfo(x, Simplified always-true  
condition);
  } else if (cond.isBooleanFalse()) {
onlyStmtToExecute = elseStmt;
removed = thenStmt;
-  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
-  Simplified always-false condition);
+  sourceInfo = makeSourceInfo(x, Simplified always-false  
condition);
  } else {
return;
  }
@@ -360,42 +356,36 @@
ctx.replaceMe(expr.makeStmt());
  } else if (thenExpr != null  elseExpr != null) {
// Convert if (a()) {b()} else {c()} = a()?b():c().
-  sourceInfo = x.getSourceInfo().makeChild(StaticEvalVisitor.class,
-  Replaced if statement with conditional);
-
-  JsConditional cond = new JsConditional(sourceInfo, x.getIfExpr(),
+  JsConditional cond = new JsConditional(makeSourceInfo(x,
+  Replaced if statement with conditional), x.getIfExpr(),
thenExpr, elseExpr);
ctx.replaceMe(accept(cond.makeStmt()));
  } else if (thenIsEmpty  elseExpr != null) {
// Convert if (a()) {} else

[gwt-contrib] [google-web-toolkit commit] r5542 - Refactor JsStaticEval to allow additional simplification when static eval of the conditio...