Re: use of large java math library from javascript

2015-12-18 Thread Andreas Plesch

On Friday, December 18, 2015 at 10:37:59 AM UTC-5, Alberto Mancini wrote:
>
> Hello,
> well, in my opinion "porting large existing java libraries" is one of the 
> strengths of GWT and in my experience it works really well in this.
>
> Were to find informations. 
>
> Well, JSInterop is the way for the part  "use the exposed, public classes 
> from Javascript."
>
> https://docs.google.com/document/d/10fmlEYIHcyead_4R1S5wKGs1t2I7Fnp_PaNaa7XTEk0/edit#
>  
> and the talk of Julien Dramaix at GWTcon 
> https://drive.google.com/file/d/0BwVGJUurq6uVR1Y5anV5TC1SQkk/edit
>

Thanks, this is very instructive, in particular the talk. I would have to 
do quite a bit of annotating, it looks like. Is there a WebIDL approach 
using an external file as well ?
 

> For the part of transpiling the library you need  more or less to 
> transform the library in a GWT module, i think  you need
> to follow
>
> http://www.gwtproject.org/doc/latest/DevGuideOrganizingProjects.html , 
> where you will find essentially what a module is and what  is 
> intended to (in case you need) and 
> http://www.gwtproject.org/doc/latest/RefJreEmulation.html that covers the 
> emulated jre. 
>
 
Yes, I just discovered this. Hopefully, the emu would cover most what is 
used.
 

> I'm biased about that but a few years ago we wrote a post on this subject 
> http://jooink.blogspot.it/2012/10/gwt-augmented-reality-howto-step-0.html
> old but still not completely useless.  
>

This looks similar, and talks a bit about (unused) java.io . Should be very 
helpful.

Thanks !
 



-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at https://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: use of large java math library from javascript

2015-12-18 Thread Andreas Plesch
Replying to myself to collect some info:

Seems to be a use case for jsinterop: 
https://docs.google.com/document/d/10fmlEYIHcyead_4R1S5wKGs1t2I7Fnp_PaNaa7XTEk0/edit#

relevant posting: 
https://groups.google.com/forum/#!searchin/google-web-toolkit/library/google-web-toolkit/kfHB1R-USOk/pJMbEIFJdhkJ

javaemul for java.util perhaps not sufficiently complete. teavm looks more 
complete.

SRM uses java.io in one case to read a static data file. Would likely need 
to be avoided somehow.



On Friday, December 18, 2015 at 9:48:30 AM UTC-5, Andreas Plesch wrote:
>
> Hello,
>
> I came across GWT when I was looking for a way to use a large java math 
> library on the client with javascript. GWT may be targeted as building web 
> apps from the ground up using java rather than porting large existing java 
> libraries but from what I see there have been such ports nonetheless.
>
> The library is called SRM and makes a lot of classes and methods available 
> which can be used to compute sophisticated and well defined 3d coordinate 
> transformations.
>
> Here is a typical example of how it would be used in a small application.
>
> import SRM.*;
>
> public class CdToCcConv
> {
> public static void main (String args[])
> {
> System.out.println("*** Sample program using SRM Java API to 
> convert a 3D coordinate");
> System.out.println("*** from a Celestiodetic SRF to a 
> Celestiocentric SRF.");
>
> // Declare reference variables for the CD_3D and CC_3D SRFs
> SRF_Celestiodetic CdSrf = null;
> SRF_Celestiocentric CcSrf = null;
>
> try
> {
> // Create a Celestiodetic SRF with WGS 1984 and Identity 
> transformation
> CdSrf = new SRF_Celestiodetic(SRM_ORM_Code.ORMCOD_WGS_1984,
>   
> SRM_RT_Code.RTCOD_WGS_1984_IDENTITY);
>
> // Create a Celestiocentric SRF with WGS 1984 and Identity 
> transformation
> CcSrf = new SRF_Celestiocentric(SRM_ORM_Code.ORMCOD_WGS_1984,
> 
> SRM_RT_Code.RTCOD_WGS_1984_IDENTITY);
>
> // Create a 3D Celestiodetic coordinate with
> // longitude   => 10.0 degrees (note: this input 
> parameter is converted to radians)
> // latitude=> 20.0 degrees (note: this input 
> parameter is converted to radians)
> // ellipsoidal height => 100.0 meters
> Coord3D_Celestiodetic CdCoord =
> 
> (Coord3D_Celestiodetic)CdSrf.createCoordinate3D(Math.toRadians(10.0),
> 
> Math.toRadians(20.0),
> 100.0);
>
> // Instantiate a 3D Celestiocentric coordinate
> // This is an alternative method for instantiate a 3D 
> coordinate
> Coord3D_Celestiocentric CcCoord = new 
> Coord3D_Celestiocentric(CcSrf);
>
> // print out the SRF parameter values and the coordinate 
> component values
> System.out.println("CdSrf parameter =>  \n" + CdSrf);
> System.out.println("CcSrf parameter =>  \n" + CcSrf);
> System.out.println("CdCoord components (source) => \n" + 
> CdCoord);
>
> // convert the 3D Celestiodetic coordinate to 3D 
> Celestiocentric coordinate
> SRM_Coordinate_Valid_Region_Code valRegion =
> CcSrf.changeCoordinateSRF(CdCoord, CcCoord);
>
> // print out the values of the resulting 3D Celestiocentric 
> coordinate
> System.out.println("CcCoord components (destination) => \n" + 
> CcCoord + " is " + valRegion);
> }
> catch (SrmException ex)
> {
> // catch SrmException and print out the error code and text.
> System.out.println("Exception caught=> " + ex.what() + ", " + 
> ex);
> }
> }
> }
>
> What I would like is use the exposed, public classes from Javascript. I 
> would envision javascript objects (classes) with the same name and methods 
> as in java.
>
> I have a compiled .jar archive and also the source code. It has about 60k 
> lines of code. It only imports java.util.* . There is no UI, just a well 
> defined API.
>
> Would it be feasible to use GWT to transpile to javascript ? If yes, how 
> would I go about it ? I do not want to develop a new web app but use the 
> library in an existing code base. I could not really find a similar example 
> but perhaps there is one ?
>
> I can provide 

use of large java math library from javascript

2015-12-18 Thread Andreas Plesch
Hello,

I came across GWT when I was looking for a way to use a large java math 
library on the client with javascript. GWT may be targeted as building web 
apps from the ground up using java rather than porting large existing java 
libraries but from what I see there have been such ports nonetheless.

The library is called SRM and makes a lot of classes and methods available 
which can be used to compute sophisticated and well defined 3d coordinate 
transformations.

Here is a typical example of how it would be used in a small application.

import SRM.*;

public class CdToCcConv
{
public static void main (String args[])
{
System.out.println("*** Sample program using SRM Java API to 
convert a 3D coordinate");
System.out.println("*** from a Celestiodetic SRF to a 
Celestiocentric SRF.");

// Declare reference variables for the CD_3D and CC_3D SRFs
SRF_Celestiodetic CdSrf = null;
SRF_Celestiocentric CcSrf = null;

try
{
// Create a Celestiodetic SRF with WGS 1984 and Identity 
transformation
CdSrf = new SRF_Celestiodetic(SRM_ORM_Code.ORMCOD_WGS_1984,
  
SRM_RT_Code.RTCOD_WGS_1984_IDENTITY);

// Create a Celestiocentric SRF with WGS 1984 and Identity 
transformation
CcSrf = new SRF_Celestiocentric(SRM_ORM_Code.ORMCOD_WGS_1984,

SRM_RT_Code.RTCOD_WGS_1984_IDENTITY);

// Create a 3D Celestiodetic coordinate with
// longitude   => 10.0 degrees (note: this input 
parameter is converted to radians)
// latitude=> 20.0 degrees (note: this input 
parameter is converted to radians)
// ellipsoidal height => 100.0 meters
Coord3D_Celestiodetic CdCoord =

(Coord3D_Celestiodetic)CdSrf.createCoordinate3D(Math.toRadians(10.0),

Math.toRadians(20.0),
100.0);

// Instantiate a 3D Celestiocentric coordinate
// This is an alternative method for instantiate a 3D coordinate
Coord3D_Celestiocentric CcCoord = new 
Coord3D_Celestiocentric(CcSrf);

// print out the SRF parameter values and the coordinate 
component values
System.out.println("CdSrf parameter =>  \n" + CdSrf);
System.out.println("CcSrf parameter =>  \n" + CcSrf);
System.out.println("CdCoord components (source) => \n" + 
CdCoord);

// convert the 3D Celestiodetic coordinate to 3D 
Celestiocentric coordinate
SRM_Coordinate_Valid_Region_Code valRegion =
CcSrf.changeCoordinateSRF(CdCoord, CcCoord);

// print out the values of the resulting 3D Celestiocentric 
coordinate
System.out.println("CcCoord components (destination) => \n" + 
CcCoord + " is " + valRegion);
}
catch (SrmException ex)
{
// catch SrmException and print out the error code and text.
System.out.println("Exception caught=> " + ex.what() + ", " + 
ex);
}
}
}

What I would like is use the exposed, public classes from Javascript. I 
would envision javascript objects (classes) with the same name and methods 
as in java.

I have a compiled .jar archive and also the source code. It has about 60k 
lines of code. It only imports java.util.* . There is no UI, just a well 
defined API.

Would it be feasible to use GWT to transpile to javascript ? If yes, how 
would I go about it ? I do not want to develop a new web app but use the 
library in an existing code base. I could not really find a similar example 
but perhaps there is one ?

I can provide more background if directed to what I should look for. Any 
tip welcome,

Andreas

-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at https://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: GWT 2.8.0 release notes - not official

2015-12-08 Thread Andreas Ofner
Hi,

Am Dienstag, 8. Dezember 2015 04:00:19 UTC+1 schrieb Juan Pablo Gardella:
>
> I've created a google doc  with full GWT 2.8.0 
> release notes.
>

Great job!!! 

-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: Will GWT stay ALIVE

2015-12-07 Thread Andreas Ofner
Hi,

Am Montag, 7. Dezember 2015 09:50:42 UTC+1 schrieb mohammed al-hammouri:
>
> I am confused about the future of GWT 
> Thanks .. and please provide me will the resources that I can use to be up 
> to date with GWT .
>

Release Notes 2.8.0 Beta1:
http://www.gwtproject.org/release-notes.html#Release_Notes_2_8_0_BETA1

JsInterop Spec:
https://docs.google.com/document/d/10fmlEYIHcyead_4R1S5wKGs1t2I7Fnp_PaNaa7XTEk0/edit#

JsInterop @ GWTCon 2015

Keynote:
https://www.youtube.com/watch?v=8pGG1WjG6_w

JsInterop Deep Dive
https://drive.google.com/file/d/0BwVGJUurq6uVR1Y5anV5TC1SQkk/edit

News / Conference Videos & Slides
http://gwtdaily.com/
http://gwtcreate.com/
http://www.gwtcon.org/

Videos from GWT Contributor Meet-up Event, May 26-27. 2015:
GWT 2.8 and Beyond & Polymer & Modernizing GWT Apps ...
https://www.youtube.com/playlist?list=PL1yReUCGwGvrqscLu1EAyYRPrr0ceEHLE

Cheers,
Andreas

-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: (Server-side) AutoBeans and thread-safety

2015-03-04 Thread Andreas Kohn


On Tuesday, February 24, 2015 at 6:56:50 PM UTC+1, Andreas Kohn wrote:

> I understand that autobean's are actually parsed lazily, so for me this 
> looks like a concurrency issue somewhere inside the autobeans framework 
> itself. I didn't see any notes about sharing autobeans between threads in 
> the documentation, so:
> 1. Should it be possible to share them, or is additional synchronization 
> in the application code needed? 
> 2. Instead of synchronizing all accesses, would it be enough to force a 
> full parse run of the autobean before making it visible to other threads?
> 3. Is this problem only affecting server-side code, or would similar 
> issues also be possible in client-side code? (I'm suspecting: "It depends 
> on the browser's implementation of JS?")
>
> Regarding the second question: I did not manage to reproduce the issue by 
> using this additional code:
> AutoBeanUtils.getAutoBean(delegate).accept(new AutoBeanVisitor() { /* 
> Nothing */ });
>
>
>
Small update here: I was able to get my test to break again even after 
using the #accept() idea, by using a Map property in the bean, and 
iterating over its entry set. So my conclusion here: using autobeans on the 
server side without external synchronization and careful coding is 
dangerous. 

I've now worked around this issue by copying the contents of the bean into 
a separate class that implements the bean interface as well. In my case I 
do not need the ability to modify the contents, so I didn't bother working 
out how I'd get my copy serialized into JSON again.

Regards,
--
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


(Server-side) AutoBeans and thread-safety

2015-02-24 Thread Andreas Kohn
Hi,

I hope this is the right forum to also ask questions about the Autobeans 
framework, if not please do redirect me!

We're using autobeans both in client code and server code, and so far 
things work nicely except in one situation: sometimes parsing autobeans 
simply fails with a stack trace similar to this:
java.lang.NullPointerException
at com.google.web.bindery.autobean.vm.impl.ProxyAutoBean.makeProxy(
ProxyAutoBean.java:105)
at com.google.web.bindery.autobean.vm.impl.ProxyAutoBean.createShim(
ProxyAutoBean.java:393)
at com.google.web.bindery.autobean.vm.impl.ProxyAutoBean.as(ProxyAutoBean.
java:222)
at com.google.web.bindery.autobean.vm.impl.ShimHandler.maybeWrap(ShimHandler
.java:113)
at com.google.web.bindery.autobean.vm.impl.ShimHandler.invoke(ShimHandler.
java:91)
at com.sun.proxy.$Proxy368.get(Unknown Source)
at com.google.web.bindery.autobean.shared.impl.SplittableSimpleMap$1$1$1.<
init>(SplittableSimpleMap.java:90)
at com.google.web.bindery.autobean.shared.impl.SplittableSimpleMap$1$1.next(
SplittableSimpleMap.java:86)
at com.google.web.bindery.autobean.shared.impl.SplittableSimpleMap$1$1.next(
SplittableSimpleMap.java:76)
at sun.reflect.GeneratedMethodAccessor89.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(
DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:483)
at com.google.web.bindery.autobean.vm.impl.ShimHandler.invoke(ShimHandler.
java:85)
at com.sun.proxy.$Proxy347.next(Unknown Source)
at XXX.function(XXX.java:...)

When dumping the contents of the bean at that point everything looks 
proper, and it can be parsed just fine in an isolated unit test. After some 
more investigation I managed to reproduce this and other similar 
exceptions: it seems to happen when multiple threads access a recently 
created autobean proxy.

I understand that autobean's are actually parsed lazily, so for me this 
looks like a concurrency issue somewhere inside the autobeans framework 
itself. I didn't see any notes about sharing autobeans between threads in 
the documentation, so:
1. Should it be possible to share them, or is additional synchronization in 
the application code needed? 
2. Instead of synchronizing all accesses, would it be enough to force a 
full parse run of the autobean before making it visible to other threads?
3. Is this problem only affecting server-side code, or would similar issues 
also be possible in client-side code? (I'm suspecting: "It depends on the 
browser's implementation of JS?")

Regarding the second question: I did not manage to reproduce the issue by 
using this additional code:
AutoBeanUtils.getAutoBean(delegate).accept(new AutoBeanVisitor() { /* 
Nothing */ });

Regards,
--
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: GWT 2.7/GwtMockito: History support

2015-02-24 Thread Andreas Kohn
Hi,

Thanks for your answer. I'll look into putting a layer between GWT (at 
least the parts that use static helpers) and our application code then. For 
now the ugliness works, and I guess I'll be able to use it also in similar 
other situations.

Regards,
--
Andreas

On Tuesday, February 17, 2015 at 11:19:54 AM UTC+1, Jens wrote:
>
>
>  is there any specific reason not to make HistoryImpl accessible?
>>
>
>  Its good practice for a library to hide implementation details because 
> you do not want developers to accidentally or knowingly depend on these 
> details. I am pretty sure other features of GwtMockito look equally "ugly", 
> because GwtMockito itself is a huge hack to circumvent GWT.create() and 
> JSNI in unit tests.
>
> Often you can also just refactor your app a bit so it does not directly 
> call such GWT methods. You can wrap them in helper classes like 
> HistoryTokenEncoder that you can swap out during testing or wrap them using 
> a helper method that can be overwritten in tests (anonymously or by sub 
> classing)
>
> -- J.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


GWT 2.7/GwtMockito: History support

2015-02-17 Thread Andreas Kohn
Hi,

some of our code uses the GWT History class (specifically the 
#encodeHistoryToken() method), and while trying to write tests for that 
code I found that GwtMockito doesn't provide a specific fake for History 
(fair enough), but unfortunately also that providing one is quite hard: 
History delegates the encoding to an instance of HistoryImpl, but that one 
is private, and so the straight-forward way of creating and registering a 
fake doesn't work.

My work-around for this is to use reflection to make the class accessible 
enough, and that works reasonably well. But, the work-around is ugly, so I 
was wondering: is there any specific reason not to make HistoryImpl 
accessible?

Regards,
--
Andreas

PS: If it helps anyone in the same position, here's the code I'm using. 
This uses easymock for various reasons, using mockito should work as well:
// Prepare the private HistoryImpl class for mocking
Class historyImplClass = Class.forName(
"com.google.gwt.user.client.History$HistoryImpl");
Method encodeHistoryTokenMethod = historyImplClass.getMethod(
"encodeHistoryToken", String.class);
encodeHistoryTokenMethod.setAccessible(true);

// Create the mock itself
Object historyImpl = createNiceMock(historyImplClass);

// Mock #encodeHistoryToken()
encodeHistoryTokenMethod.invoke(historyImpl, isA(String.class));
expectLastCall().andAnswer(new IAnswer() {
@Override
public String answer() throws Throwable {
String value = (String) getCurrentArguments()[0];
if (value == null) {
throw new NullPointerException();
} else if (value.isEmpty()) {
return "";
}

try {
URI uri = new URI(null, null, value);
// URI#getRawFragment() is pretty close, but it does 
unfortunately retain characters >= U+0080,
// which we do not want.
// Instead, rely on #toASCIIString(), and cut away the '#' in 
the beginning.
// See also HistoryImpl#encodeHistoryToken()
String encoded = uri.toASCIIString();
assert encoded.charAt(0) == '#' : "Sanity";
return encoded.substring(1).replace("#", "%23");
} catch (URISyntaxException e) {
throw new IllegalArgumentException("Cannot use '" + value + "' 
as fragment", e);
}
}
}).anyTimes();
replay(historyImpl);

// Finally, tell GwtMockito about it.
GwtMockito.useProviderForType(historyImplClass, new FakeProvider() {
@Override
public Object getFake(Class type) {
return historyImpl;
}
});



-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: GWT selects no permutation due to unexpected locales from inherited module

2014-03-26 Thread Andreas Kohn
Hi Jens,

Thanks for the idea, I tried it but it did not lead to a change. Looking a 
bit further into how the  and  are handled 
that actually makes sense: set-property modifies only the *allowed* values 
of the property, but the code generating the values 'locale' is checked 
again looks at the *possible* values, which are controlled by 
extend-property.

Ultimately I now went with a custom generator, which looks like this:

import java.util.SortedSet;

import com.google.gwt.core.ext.TreeLogger;
import com.google.gwt.core.ext.UnableToCompleteException;
import com.google.gwt.core.ext.linker.ConfigurationProperty;
import com.google.gwt.i18n.linker.LocalePropertyProviderGenerator;

public class RestrictedLocalePropertyProviderGenerator extends 
LocalePropertyProviderGenerator {

@Override
public String generate(TreeLogger logger, SortedSetpossibleValues
, String fallback,
SortedSet configProperties) throws 
UnableToCompleteException {

String original = super.generate(logger, possibleValues, 
fallback,configProperties
);

// original is now "{ ... code returning locale }", so
// wrap that into a function, invoke it, and filter the result.
return "{" +
"var allowedLocales = {'en':1, 'fr':1 };" +
"var locale = function() " + original + "();" +
"while (locale && !(locale in allowedLocales)) {" +
"  var lastIndex = locale.lastIndexOf('_');" +
"  if (lastIndex < 0) {" +
"locale = null;" +
"break;" +
"  }" +
"  locale = locale.substring(0, lastIndex);" +
"}" +
"return locale || '" + fallback + "'" + 
"}";
}
}


Not pretty, but avoids most of the mentioned issues at the cost of 
duplicating my list of allowed values (which change very rarely), and 
ignoring a warning from the GWT compiler to not include gwt-dev as 
dependency.

Regards,
--
Andreas

On Wednesday, March 26, 2014 1:13:51 PM UTC+1, Jens wrote:
>
> Try in your module:
>
> 
> 
> 
>
> Yes I really mean duplicating that line. See: 
> https://code.google.com/p/google-web-toolkit/issues/detail?id=5769
>
> -- J.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


GWT selects no permutation due to unexpected locales from inherited module

2014-03-26 Thread Andreas Kohn
Hi,

we're using GWT 2.6.0 and recently added the gwt-cal module. Since then 
some users reported that instead of the application they only saw an empty 
page. We debugged the issue, but now are at a loss how to "properly" fix it.

Our own .gwt.xml looks (abbreviated) like this:



 
 
 

 
 
 
 
 

 
 
 
 
 


The goal is to have permutations for our two supported locales: en and fr. 

The gwt-cal modules support quite a few more locales, and looks (again 
trimmed) like this:













What happens now is that compiling our module produces permutations only 
for 'en' and 'fr', but the calculation of the 'locale' property value does 
also accept 'ar' and all other locales. As a result the permutation 
switching code does not select an existing permutation if a user for some 
reason uses one of those locales, and those users observe a completely 
empty page, rather than being forced into the 'en' locale.

It looks like we have the following options here:
1. modify the gwt-cal module and remove the  parts
2. modify GWT's PropertiesUtil#generatePropertyProvider() to produce the 
list of values that are allowed from SelectionProperty#getAllowedValues() 
rather than SelectionProperty#getPossibleValues()
3. create a new PropertyProvider for locale that does an explicit check of 
the detected locale

The first option would be very localized, and we would get breakage again 
when another module starts extending the 'locale' property values. 
Additionally the first options make updating a lot harder. I've looked 
briefly in the third approach, but it seems that I would have to copy/paste 
most code from LocalePropertyProviderGenerator, which seems also a bit 
sub-optimal :)

Am I missing an approach here, or could anyone see any other option?

Best Regards,
--
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: Super Dev Mode 2.6 : NoClassDefFoundError

2014-03-17 Thread Andreas Horst
If kind of everything fails you could clone
https://github.com/gwt-maven-plugin/gwt-maven-plugin to locally install and
use version 2.6.0-1-SNAPSHOT of the plugin in order to get going. It really
works quite smooth, though I had to skip tests because of a breaking API
change in one of the contributing plugins.

Give it a try. I did this myself in order to try out SDM.

2014-03-17 18:04 GMT+01:00 Philippe Pithon :

> I tried your solution but same problem...
>
>
>
> Le lundi 17 mars 2014 16:33:09 UTC+1, Thomas Broyer a écrit :
>
>>
>>
>> On Monday, March 17, 2014 4:19:21 PM UTC+1, Philippe Pithon wrote:
>>>
>>> Hello,
>>> As dev mode no longer works with firefox 27, I try to use Super Dev mode
>>> but
>>> Problem when I use command maven : gwt:run-codeserver
>>>
>>> NoClassDefFoundError ! Ideas ???
>>>
>>>
>>> [INFO] Caused by: java.lang.NoClassDefFoundError:
>>> fr.sigal.solrclientsmart.client.GreetingServiceAsync
>>>
>>
>> You might be running into a weird bug we have in gwt-maven-plugin 2.6.0
>> (fixed in master, will be in 2.6.1). Try using "mvn process-classes
>> gwt:run-codeserver".
>>
>  --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.


Re: JSONSplittable#setSize() fails if new size is bigger than existing size

2014-02-07 Thread Andreas Kohn
Hi,

I filed https://code.google.com/p/google-web-toolkit/issues/detail?id=8563for
it, with a patch at
https://gwt-review.googlesource.com/#/c/6330/

Regards,
--
Andreas


On Wed, Feb 5, 2014 at 2:33 PM, Thomas Broyer  wrote:

> Splittable is supposed to represent any JavaScript value (except
> *undefined*, which is seen the same as *null*), and Array.length in JS is
> settable to any value (note that in some (edge?) cases, the length property
> is not set to the exact value assigned to it), extending the array with
> *undefined* items.
> So I'd say it's a bug, and setSize should behave as you describe (in
> JsonSplittable; in JsoSplittable it just sets the length property of the
> underlying JS array)
>
> On Wednesday, February 5, 2014 12:14:17 PM UTC+1, Andreas Kohn wrote:
>>
>> Hi,
>>
>> I was working today with autobeans in a JSON format that is partially
>> fixed, and partially open. The fixed parts we handle with regular autobean
>> get/set methods, the open parts we handle with a single get/set pair that
>> produces a Map.
>>
>> One of the splittables I needed was an array of strings, and probably
>> because of premature optimization I used #setSize() before filling the
>> array:
>> 
>> List values = (coming from the caller)
>>
>> Splittable valuesSplittable = StringQuoter.createIndexed();
>> valuesSplittable.setSize(values.size());
>>   /// here.
>> for (int i = 0; i < values.size(); i++) {
>> Splittable valueSplittable = StringQuoter.create(values.
>> get(i));
>> valueSplittable.assign(valuesSplittable, i);
>> }
>> 
>>
>> This leads to an exception:
>> java.lang.reflect.UndeclaredThrowableException
>> at com.sun.proxy.$Proxy18.setValues(Unknown Source)
>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>> at sun.reflect.NativeMethodAccessorImpl.invoke(
>> NativeMethodAccessorImpl.java:57)
>> at sun.reflect.DelegatingMethodAccessorImpl.invoke(
>> DelegatingMethodAccessorImpl.java:43)
>> at java.lang.reflect.Method.invoke(Method.java:606)
>> at com.google.web.bindery.autobean.vm.impl.ShimHandler.
>> invoke(ShimHandler.java:85)
>> at com.sun.proxy.$Proxy18.setValues(Unknown Source)
>> at com.company.ClassTest.testSetValues(Unknown Source)
>> ...
>> Caused by: java.lang.reflect.InvocationTargetException
>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>> at sun.reflect.NativeMethodAccessorImpl.invoke(
>> NativeMethodAccessorImpl.java:57)
>> at sun.reflect.DelegatingMethodAccessorImpl.invoke(
>> DelegatingMethodAccessorImpl.java:43)
>> at java.lang.reflect.Method.invoke(Method.java:606)
>> at com.google.web.bindery.autobean.vm.impl.BeanMethod$5.
>> invoke(BeanMethod.java:155)
>> at com.google.web.bindery.autobean.vm.impl.SimpleBeanHandler.invoke(
>> SimpleBeanHandler.java:43)
>> ... 32 more
>> Caused by: java.lang.RuntimeException: org.json.JSONException:
>> JSONArray[0] not found.
>> at com.google.web.bindery.autobean.vm.impl.JsonSplittable.setSize(
>> JsonSplittable.java:278)
>> at com.company.Class.setValues(Unknown Source)
>> ... 38 more
>> Caused by: org.json.JSONException: JSONArray[0] not found.
>> at org.json.JSONArray.get(JSONArray.java:234)
>> at com.google.web.bindery.autobean.vm.impl.JsonSplittable.setSize(
>> JsonSplittable.java:276)
>> ... 39 more
>>
>> It seems that #setSize() works fine when reducing the size of the array,
>> but not when trying to increase it: #setSize() assumes that all indices
>> from 0..newSize-1 are valid.
>>
>> 1. Is this intentional? Or should the #setSize() method only copy to
>> Min(newSize, oldSize), and fill up the array with 'null' if newSize >
>> oldSize, so that the assumption #size() returns the size set with
>> #setSize() is retained?
>> 2. Should a note be added to the javadoc to better specify the #setSize()
>> method behavior to avoid such issues?
>>
>> Regards,
>> Andreas
>>
>  --
> You received this message because you are subscribed to a topic in the
> Google Groups "Google Web Toolkit" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/YE5HRc3x-H8/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-

JSONSplittable#setSize() fails if new size is bigger than existing size

2014-02-05 Thread Andreas Kohn
Hi,

I was working today with autobeans in a JSON format that is partially 
fixed, and partially open. The fixed parts we handle with regular autobean 
get/set methods, the open parts we handle with a single get/set pair that 
produces a Map.

One of the splittables I needed was an array of strings, and probably 
because of premature optimization I used #setSize() before filling the 
array:

List values = (coming from the caller)

Splittable valuesSplittable = StringQuoter.createIndexed();

valuesSplittable.setSize(values.size());
 
/// here.
for (int i = 0; i < values.size(); i++) {
Splittable valueSplittable = StringQuoter.create(values.get(i));
valueSplittable.assign(valuesSplittable, i);
}


This leads to an exception: 
java.lang.reflect.UndeclaredThrowableException
at com.sun.proxy.$Proxy18.setValues(Unknown Source)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at 
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at 
com.google.web.bindery.autobean.vm.impl.ShimHandler.invoke(ShimHandler.java:85)
at com.sun.proxy.$Proxy18.setValues(Unknown Source)
at com.company.ClassTest.testSetValues(Unknown Source)
...
Caused by: java.lang.reflect.InvocationTargetException
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at 
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at 
com.google.web.bindery.autobean.vm.impl.BeanMethod$5.invoke(BeanMethod.java:155)
at 
com.google.web.bindery.autobean.vm.impl.SimpleBeanHandler.invoke(SimpleBeanHandler.java:43)
... 32 more
Caused by: java.lang.RuntimeException: org.json.JSONException: JSONArray[0] 
not found.
at 
com.google.web.bindery.autobean.vm.impl.JsonSplittable.setSize(JsonSplittable.java:278)
at com.company.Class.setValues(Unknown Source)
... 38 more
Caused by: org.json.JSONException: JSONArray[0] not found.
at org.json.JSONArray.get(JSONArray.java:234)
at 
com.google.web.bindery.autobean.vm.impl.JsonSplittable.setSize(JsonSplittable.java:276)
... 39 more

It seems that #setSize() works fine when reducing the size of the array, 
but not when trying to increase it: #setSize() assumes that all indices 
from 0..newSize-1 are valid. 

1. Is this intentional? Or should the #setSize() method only copy to 
Min(newSize, oldSize), and fill up the array with 'null' if newSize > 
oldSize, so that the assumption #size() returns the size set with 
#setSize() is retained?
2. Should a note be added to the javadoc to better specify the #setSize() 
method behavior to avoid such issues?

Regards,
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


Re: StorageMap.StorageEntryIterator#remove() - Issues with order?

2014-01-23 Thread Andreas Kohn
Hi,

there we go:
https://code.google.com/p/google-web-toolkit/issues/detail?id=8544

--
Andreas


On Thu, Jan 16, 2014 at 7:14 PM, Jens  wrote:

> Should I open an issue for this?
>>
>
> Sure. I don't think the current implementation will change in the near
> future (as it seems to work in all browsers nowadays) but at least that
> problem will be tracked, can be found by others if it becomes relevant and
> can be used to discuss alternative implementations.
>
> -- J.
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "Google Web Toolkit" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/b6SLntTyTsU/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


Re: StorageMap.StorageEntryIterator#remove() - Issues with order?

2014-01-16 Thread Andreas Kohn
Should I open an issue for this?

On the question of using the other order guarantee, I guess this depends on
how reliably that is actually implemented in the browsers :)

--
Andreas


On Mon, Jan 6, 2014 at 1:16 PM, Jens  wrote:

> Looks like you are right and the iterator does not account for such a
> behavior in case of remove(). If remove() can really result in a different
> key ordering then iteration can fail to work properly. Adding an element to
> the underlying collection while iterating is not an issue as the Iterator
> JavaDoc/Contract already defines that iteration behavior is undefined if
> the underlying collection is modified by other means than Iterator.remove().
>
> I think GWT has been lucky here so far and browsers do not re-order
> existing storage entries if you remove a key/value pair. Also the key(n)
> doc says:
>
> The supported property names on a 
> Storage<http://dev.w3.org/html5/webstorage/#storage-0>object are the keys of 
> each key/value pair currently present in the list
> associated with the object, *in the order that the keys were last added
> to the storage area*.
>
> I am wondering if this can be useful for an alternative iterator
> implementation as this sentence defines an ordering which should not change
> when removing items.
>
>
> -- J.
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "Google Web Toolkit" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/b6SLntTyTsU/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


StorageMap.StorageEntryIterator#remove() - Issues with order?

2014-01-06 Thread Andreas Kohn
Hi,

while reading through the GWT StorageMap class to learn more about using 
HTML 5 local storage I noticed that the StorageEntryIterator#remove() 
method seems to assume a stable order of the keys in the local storage, 
i.e. that removing a key via the iterator does not change the order of the 
iteration, and the next unvisited key has simply the same index as the 
now-removed key.

According to http://dev.w3.org/html5/webstorage/#dom-storage-key this 
doesn't seem to be a good assumption: the order is explicitly allowed to 
change when the number of keys change, 
> The key(n) method must return the name of the nth key in the list. The 
order of keys is user-agent defined, but must be consistent within an 
object so long as the number of keys doesn't change. (Thus, 
adding<http://dev.w3.org/html5/webstorage/#dom-storage-setitem>or 
removing <http://dev.w3.org/html5/webstorage/#dom-storage-removeitem> a key 
may change the order of the keys, but merely changing the value of an 
existing key must not.) If n is greater than or equal to the number of 
key/value pairs in the object, then this method must return null.

Has this behavior ever been observed in the wild, or am I missing something 
here?

--
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


GwtMockito and GWT 2.6 Timers

2014-01-03 Thread Andreas Kohn
advise would be greatly appreciated :)

Best Regards,
--
Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


Re: http://zeroturnaround.com/rebellabs/the-2014-decision-makers-guide-to-java-web-frameworks/3/

2013-11-06 Thread Andreas Horst
I'd say it's even more than "a bit silly" because as one can see Vaadin,
which is based on GWT, is not that insecure. It seems to me like the author
just entirely omitted the fact that - of course - one can apply all
security mechanisms current containers provide on the server side with GWT
as well! Just look at what is written in the "evaluation" of Vaadin: "...by
declaring security constraints in the deployment descriptor...". AFAIK,
this has nothing to do with the applied (front end) framework.



2013/11/6 salk31 

>
> http://zeroturnaround.com/rebellabs/the-2014-decision-makers-guide-to-java-web-frameworks/3/
>
> Apparently GWT is insecure because it uses JavaScript. Am I reading this
> wrong or is it a bit silly?
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-web-toolkit+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.


Re: java arrays in JSNI

2013-05-24 Thread Andreas Berger
Like explained in http://stackoverflow.com/a/6130473/1666829 you can call:

public static final native void trigger(String eventName, JsArray 
params)/*-{ 
$wnd.mylib.trigger.apply(this, [eventName].concat(params)) ;
}-*/; 

Pay attention to the second parameter which is a JsArray. You can convert 
it in e seperate method.

Am Dienstag, 15. Mai 2012 18:54:03 UTC+2 schrieb Sebastián Gurin:
>
> Hi all. I'm trying to create a java API for a JavaScript function that can 
> accept variable number of parameters, for ex. trigger("eventName", param1, 
> param2, ...). 
>
> In java the desired java method signature, do not need to use java 
> varargs, could be simple something like this: 
>
> public static final native void trigger(String eventName, Object[] 
> params)/*-{ 
> ///$wnd.mylib.trigger(eventName, ) 
> }-*/; 
>
>
> The thing is that the params array elements can be of any and mixed types. 
> Could be POJOS or JavaScriptObject or java.lang.Integer. 
>
> My problem with this is that I don't know how to create a JsArray or a 
> native JavaScript array and put all the elements of parameter 
> object[]params. 
>
> I tried doing this in java using a JsArray and JsArrayMixed but elements 
> must be casted to JavaScriptObject, and when doing so I get a 
> "java.lang.Object cannot cast to com...JavaScriptObject" exception. This 
> happens when trying to push an element into the jsarray: it need to be 
> casted for java compiler: jsArray1.push((JavaScriptObject) myJavaArray[i]); 
>
> I also tried doing it in jsni javascript code, but I don't know how to 
> access java arrays for obtaining its elements. The documentation says that 
> from javascript java arrays are "opaque" objects. It also says : 
>
> "Although Java arrays are not directly usable in JavaScript, there are 
> some helper classes that efficiently achieve a similar effect: 
> JsArray,. These classes are wrappers around a native JavaScript array." 
>
> But I think in my case, if I need to access a java array with java objects 
> these classes have not use... ? 
>
> I think I'm missing something here... IMHO it is a common javascript 
> technic defining functions with variable parameters ,and it sould be 
> available to work with these from java using GWT JSNI. - 
>
> Any suggestion is mmost appreciated. Thanks in advance. 
>
> -- 
> Sebastian Gurin > 
>

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




Re: [POLL RESULTS] Maven project layout, what to "standardize"?

2012-11-19 Thread Andreas Horst
That's true. But I do not have to edit both Java files and GWT module
descriptors at the same time very often. More often that's the case for
UIBinder/CSS and Java files.

I have multi module projects with several hierarchy levels either but still
prefer the *src/main/resources* option(s).

2012/11/19 Joseph Lust 

> In multi module projects, it is quite nice to have the Java sources and
> *.gwt.xml files in a single package in src/main/java. Spreading it out into
> the resource folder would make such projects a bit more difficult to
> maintain/navigate.
>
> Sincerely,
> Joseph
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/google-web-toolkit/-/sxkhAHV--CkJ.
>
> 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.
>

-- 
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: Video from io 2012 : "The History and Future of Google Web Toolkit"

2012-07-11 Thread Andreas
Now available at http://www.youtube.com/watch?v=VOf27ez_Hvg


-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-web-toolkit/-/s63um8IqQEQJ.
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.



Requestfactory: How to send a HTTP status code to the client

2012-01-04 Thread Andreas
Hi,

what is the recommended way to inform the client about a 404 Not Found or a 
401 Unauthorized?

Greetings Andreas

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-web-toolkit/-/jmBZjbdDFdsJ.
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: Don't compile for the MAC and IE* when developing.

2011-11-29 Thread Andreas Horst
Use a working module. That's a very easy way to do this.

2011/11/29 skippy 

> When developing locally, I usually only test with one browser.  Before
> moving to regression testing we test multiple browsers just for the
> fun of it.
>
> My question is, is there a way to easily turn off the compiling for
> the MAC and Fire Fox when not testing in that space?
>
> I have implemented the localWorker compiler option, but our
> application is starting to grow and compile times are on the rise.
>
> 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.
>
>

-- 
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: What do you think?

2011-09-28 Thread Andreas Horst
+1

One of the worst (formatted) articles I've

2011/9/28 Kees de Kooter 

> I think the writer of this article should first read up on the manual of
> the enter key.
>
>  --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/google-web-toolkit/-/52W0q3coY30J.
>
> 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.
>

-- 
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: Why is my Inherits causing all the inherited module to run its onModuleLoad? :?

2011-09-21 Thread Andreas Horst
That's a problem I had as well, when it comes to creating some custom
widgets in a reusable module and at the same time being able to test them in
the same module.

I'd do the following:

- remove the  from your main module
- extend your main module with a "dev-"module which contains the
 for testing purpose
- use that "dev-"module in development and testing
- use the main module for inheriting in other projects

That way you can keep your entry point for development but distribute your
module without it. You could for example exclude that "dev-"module from
packaging or simply name it appropriately so people do not inherit the wrong
one.

Google for "Working Module". It's a good way to keep development related
configuration in a working module apart from the main module.

2011/9/21 darkflame 

> Ah, that makes a lot more sense. I have it working now, cheers.
>
> It seems previously I was simply within the same package, thus no
> inherits needed.
>
> As I am building up quite a nice selection of widgets to reuse (which
> at some point I'll probably want to share with others too), do I
> simply make a new gwt.xml without an entry point (and excluding
> anything else not needed)?
>
> Or should I bundle them some other way?
> My preferance would be to keep an entry point of some sort as it does
> make testing the widgets a lot easier rather then needing a seperate
> project to test/develope them from.
>
> Thanks,
> Thomas
>
>
> On Sep 21, 5:20 pm, Andreas Horst 
> wrote:
> > I do not have the URL at hand but AFAIK each EntryPoint gets executed
> > (in arbitrary order) by default. This is helpful if you have a module
> which
> > for example needs to (ensure) inject some specific styles (gwt-dnd for
> > example does this in its EntryPoint). So that is a -1 on Juan Pablo's
> reply:
> > also reusable modules might need to do stuff at start of the application.
> >
> > Now if your inherited module is in fact an application itself and it
> builds
> > up its UI in its EntryPoint your problem actually is that that module is
> not
> > designed to be reused. You might want to separate the API components of
> that
> > inherited module into an API module which you can inherit instead of the
> > full blown application module.
> >
> > 2011/9/21 Juan Pablo Gardella 
> >
> >
> >
> > > If is a reusable artifact, you don't need an entry point.
> >
> > > 2011/9/21 darkflame 
> >
> > >> I was simply trying to reuse code from one project in another.
> >
> > >> I've done this many times before, by simply adding the project to the
> > >> build path in Eclipse.
> > >> This normally has worked fine - I can use widgets and code from the
> > >> other projects just fine. It compiles fine.
> > >> Interestingly, I have never had to previously add any inherits to the
> > >> gwt.xml file for this to work. These projects compile just fine
> > >> without any reference in ther xml (and still do). I assumed eclipse or
> > >> the gwt compiler sorted this out itself.
> >
> > >> Trying it with a new project though, I now keep getting the "did you
> > >> forget to inherit the require module" error when I reference code
> > >> elsewhere.
> > >> Ok, I thought, this time I'll just add the module to the xml;
> >
> > >> 
> > >> 
> > >>  
> > >>  
> >
> > >>  
> >
> > >>  
> > >>  
> > >>  
> > >>  
> > >>  
> > >>  
> > >>  
> >
> > >>  
> >
> > >>  
> > >>  
> >
> > >>  
> > >>  
> > >>  
> >
> > >> 
> >
> > >> Note the "" is me
> > >> inheriting of the project I want to reuse code from.
> >
> > >> This removes the error and it compiles.but now the MyApplication
> > >> projects onModuleLoad actualy runs!
> > >> For some reason its not just referencing the code, but triggering the
> > >> whole module :?
> >
> > >> Whats going on?
> >
> > >> --
> > >> 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..

Re: Why is my Inherits causing all the inherited module to run its onModuleLoad? :?

2011-09-21 Thread Andreas Horst
I do not have the URL at hand but AFAIK each EntryPoint gets executed
(in arbitrary order) by default. This is helpful if you have a module which
for example needs to (ensure) inject some specific styles (gwt-dnd for
example does this in its EntryPoint). So that is a -1 on Juan Pablo's reply:
also reusable modules might need to do stuff at start of the application.

Now if your inherited module is in fact an application itself and it builds
up its UI in its EntryPoint your problem actually is that that module is not
designed to be reused. You might want to separate the API components of that
inherited module into an API module which you can inherit instead of the
full blown application module.

2011/9/21 Juan Pablo Gardella 

> If is a reusable artifact, you don't need an entry point.
>
> 2011/9/21 darkflame 
>
>> I was simply trying to reuse code from one project in another.
>>
>> I've done this many times before, by simply adding the project to the
>> build path in Eclipse.
>> This normally has worked fine - I can use widgets and code from the
>> other projects just fine. It compiles fine.
>> Interestingly, I have never had to previously add any inherits to the
>> gwt.xml file for this to work. These projects compile just fine
>> without any reference in ther xml (and still do). I assumed eclipse or
>> the gwt compiler sorted this out itself.
>>
>> Trying it with a new project though, I now keep getting the "did you
>> forget to inherit the require module" error when I reference code
>> elsewhere.
>> Ok, I thought, this time I'll just add the module to the xml;
>>
>> 
>> 
>>  
>>  
>>
>>  
>>
>>  
>>  
>>  
>>  
>>  
>>  
>>  
>>
>>  
>>
>>  
>>  
>>
>>
>>  
>>  
>>  
>>
>> 
>>
>>
>> Note the "" is me
>> inheriting of the project I want to reuse code from.
>>
>> This removes the error and it compiles.but now the MyApplication
>> projects onModuleLoad actualy runs!
>> For some reason its not just referencing the code, but triggering the
>> whole module :?
>>
>> Whats going on?
>>
>> --
>> 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.
>>
>>
>  --
> 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.
>

-- 
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: GWTTestCase run by surefire 2.9

2011-07-28 Thread Andreas Horst
Thanks Nicolas

I will try it using the gwt-maven-plugin test mojo.

Let me add another question please:

Is anybody using this particular setup for building GWT module jars and
successfully using the gwt-maven-plugin test mojo for obtaining coverage?
Actually the reason for using surefire for GWTTestCase was for obtaining
coverage for both pojo Unit and GWT tests to display in Sonar. We used emma
which seemed to be the only one working with GWT tests at that moment.

Andreas

2011/7/28 nicolas de loof 

> Sure, compile mojo is only one of the available mojos, you can use only
> gwt:test to run tests, and gwt:resources to package the xml and .java files
> with your gwt module
>
>
> 2011/7/28 Andreas Horst 
>
>> Hi Nicolas
>>
>> I'm not using the gwt-maven-plugin for this project since the project does
>> not produce GWT compiler output. It's a reusable GWT-based API and only
>> builds jars to use in concrete GWT projects which of course then use the
>> gwt-maven-plugin.
>>
>> I hence only need "normal" compilation and test execution besides the use
>> of some GWTTestCases. Would you suggest using the gwt-maven-plugin for this
>> even though the GWT compiler itself will not be used?
>>
>> Thanks
>>
>> Andreas
>>
>>
>> 2011/7/28 nicolas de loof 
>>
>>> Why don't you use gwt:test goal for that ? It has been designed to mimic
>>> surefire but don't requires such tricky configuration
>>>
>>> 2011/7/28 Andreas Horst 
>>>
>>>> Forgot the stack trace:
>>>>
>>>> com.google.gwt.core.ext.UnableToCompleteException: (see previous log
>>>> entries)
>>>>  at
>>>> com.google.gwt.dev.cfg.ModuleDefLoader.nestedLoad(ModuleDefLoader.java:262)
>>>>  at
>>>> com.google.gwt.dev.cfg.ModuleDefLoader$2.load(ModuleDefLoader.java:210)
>>>> at
>>>> com.google.gwt.dev.cfg.ModuleDefLoader.doLoadModule(ModuleDefLoader.java:308)
>>>>  at
>>>> com.google.gwt.dev.cfg.ModuleDefLoader.createSyntheticModule(ModuleDefLoader.java:102)
>>>> at
>>>> com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl2(CompileStrategy.java:165)
>>>>  at
>>>> com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl(CompileStrategy.java:112)
>>>> at
>>>> com.google.gwt.junit.SimpleCompileStrategy.maybeCompileModule(SimpleCompileStrategy.java:36)
>>>>  at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1340)
>>>> at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1309)
>>>>  at com.google.gwt.junit.JUnitShell.runTest(JUnitShell.java:650)
>>>> at com.google.gwt.junit.client.GWTTestCase.runTest(GWTTestCase.java:441)
>>>>  at junit.framework.TestCase.runBare(TestCase.java:127)
>>>> at junit.framework.TestResult$1.protect(TestResult.java:106)
>>>>  at junit.framework.TestResult.runProtected(TestResult.java:124)
>>>> at junit.framework.TestResult.run(TestResult.java:109)
>>>>  at junit.framework.TestCase.run(TestCase.java:118)
>>>> at com.google.gwt.junit.client.GWTTestCase.run(GWTTestCase.java:296)
>>>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>>>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>>>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>>>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>>>  at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>>>> at
>>>> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>>>>  at
>>>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>>>> at java.lang.reflect.Method.invoke(Method.java:597)
>>>>  at
>>>> org.apache.maven.surefire.junit.JUnitTestSet.execute(JUnitTestSet.java:98)
>>>> at
>>>> org.apache.maven.surefire.junit.JUnit3Provider.executeTestSet(JUnit3Provider.java:117)
>>>>  at
>>>> org.apache.maven.surefire.junit.JUnit3Provider.invoke(JUnit3Provider.java:94)
>>>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>>>>  at
>>>> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>>>> at
>>>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>>>>  at java.lang.reflect.Me

Re: GWTTestCase run by surefire 2.9

2011-07-28 Thread Andreas Horst
Hi Nicolas

I'm not using the gwt-maven-plugin for this project since the project does
not produce GWT compiler output. It's a reusable GWT-based API and only
builds jars to use in concrete GWT projects which of course then use the
gwt-maven-plugin.

I hence only need "normal" compilation and test execution besides the use of
some GWTTestCases. Would you suggest using the gwt-maven-plugin for this
even though the GWT compiler itself will not be used?

Thanks

Andreas

2011/7/28 nicolas de loof 

> Why don't you use gwt:test goal for that ? It has been designed to mimic
> surefire but don't requires such tricky configuration
>
> 2011/7/28 Andreas Horst 
>
>> Forgot the stack trace:
>>
>> com.google.gwt.core.ext.UnableToCompleteException: (see previous log
>> entries)
>>  at
>> com.google.gwt.dev.cfg.ModuleDefLoader.nestedLoad(ModuleDefLoader.java:262)
>>  at
>> com.google.gwt.dev.cfg.ModuleDefLoader$2.load(ModuleDefLoader.java:210)
>> at
>> com.google.gwt.dev.cfg.ModuleDefLoader.doLoadModule(ModuleDefLoader.java:308)
>>  at
>> com.google.gwt.dev.cfg.ModuleDefLoader.createSyntheticModule(ModuleDefLoader.java:102)
>> at
>> com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl2(CompileStrategy.java:165)
>>  at
>> com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl(CompileStrategy.java:112)
>> at
>> com.google.gwt.junit.SimpleCompileStrategy.maybeCompileModule(SimpleCompileStrategy.java:36)
>>  at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1340)
>> at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1309)
>>  at com.google.gwt.junit.JUnitShell.runTest(JUnitShell.java:650)
>> at com.google.gwt.junit.client.GWTTestCase.runTest(GWTTestCase.java:441)
>>  at junit.framework.TestCase.runBare(TestCase.java:127)
>> at junit.framework.TestResult$1.protect(TestResult.java:106)
>>  at junit.framework.TestResult.runProtected(TestResult.java:124)
>> at junit.framework.TestResult.run(TestResult.java:109)
>>  at junit.framework.TestCase.run(TestCase.java:118)
>> at com.google.gwt.junit.client.GWTTestCase.run(GWTTestCase.java:296)
>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>  at junit.framework.TestSuite.runTest(TestSuite.java:208)
>> at junit.framework.TestSuite.run(TestSuite.java:203)
>>  at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>> at
>> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>>  at
>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>> at java.lang.reflect.Method.invoke(Method.java:597)
>>  at
>> org.apache.maven.surefire.junit.JUnitTestSet.execute(JUnitTestSet.java:98)
>> at
>> org.apache.maven.surefire.junit.JUnit3Provider.executeTestSet(JUnit3Provider.java:117)
>>  at
>> org.apache.maven.surefire.junit.JUnit3Provider.invoke(JUnit3Provider.java:94)
>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>>  at
>> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
>> at
>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>>  at java.lang.reflect.Method.invoke(Method.java:597)
>> at
>> org.apache.maven.surefire.util.ReflectionUtils.invokeMethodWithArray(ReflectionUtils.java:164)
>>  at
>> org.apache.maven.surefire.booter.ProviderFactory$ProviderProxy.invoke(ProviderFactory.java:110)
>> at
>> org.apache.maven.surefire.booter.SurefireStarter.invokeProvider(SurefireStarter.java:172)
>>  at
>> org.apache.maven.surefire.booter.SurefireStarter.runSuitesInProcessWhenForked(SurefireStarter.java:104)
>> at
>> org.apache.maven.surefire.booter.ForkedBooter.main(ForkedBooter.java:70)
>>
>> 2011/7/28 Andreas Horst 
>>
>>> Hi All
>>>
>>> Is anybody successfully running GWTTestCase (2.3) using the surefire
>>> plugin version 2.9? If so could you share your surefire configuration?
>>>
>>>
>>> I just recently switched to latest version of surefire for staying up to
>>> date and now have problems running my GWTTestCases with it. My surefire
>>> configuration is derived from 
>>> here<http://stackoverflow.com/questions/2737173/error-when-running-a-gwttestcase-using-maven-gwt-plugin>
>>> :
>>>
>>> 
>>>
>>>
>>>
>>>
>>>  false
>>>
>>&g

Re: GWTTestCase run by surefire 2.9

2011-07-28 Thread Andreas Horst
Forgot the stack trace:

com.google.gwt.core.ext.UnableToCompleteException: (see previous log
entries)
at
com.google.gwt.dev.cfg.ModuleDefLoader.nestedLoad(ModuleDefLoader.java:262)
at com.google.gwt.dev.cfg.ModuleDefLoader$2.load(ModuleDefLoader.java:210)
at
com.google.gwt.dev.cfg.ModuleDefLoader.doLoadModule(ModuleDefLoader.java:308)
at
com.google.gwt.dev.cfg.ModuleDefLoader.createSyntheticModule(ModuleDefLoader.java:102)
at
com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl2(CompileStrategy.java:165)
at
com.google.gwt.junit.CompileStrategy.maybeCompileModuleImpl(CompileStrategy.java:112)
at
com.google.gwt.junit.SimpleCompileStrategy.maybeCompileModule(SimpleCompileStrategy.java:36)
at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1340)
at com.google.gwt.junit.JUnitShell.runTestImpl(JUnitShell.java:1309)
at com.google.gwt.junit.JUnitShell.runTest(JUnitShell.java:650)
at com.google.gwt.junit.client.GWTTestCase.runTest(GWTTestCase.java:441)
at junit.framework.TestCase.runBare(TestCase.java:127)
at junit.framework.TestResult$1.protect(TestResult.java:106)
at junit.framework.TestResult.runProtected(TestResult.java:124)
at junit.framework.TestResult.run(TestResult.java:109)
at junit.framework.TestCase.run(TestCase.java:118)
at com.google.gwt.junit.client.GWTTestCase.run(GWTTestCase.java:296)
at junit.framework.TestSuite.runTest(TestSuite.java:208)
at junit.framework.TestSuite.run(TestSuite.java:203)
at junit.framework.TestSuite.runTest(TestSuite.java:208)
at junit.framework.TestSuite.run(TestSuite.java:203)
at junit.framework.TestSuite.runTest(TestSuite.java:208)
at junit.framework.TestSuite.run(TestSuite.java:203)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at
org.apache.maven.surefire.junit.JUnitTestSet.execute(JUnitTestSet.java:98)
at
org.apache.maven.surefire.junit.JUnit3Provider.executeTestSet(JUnit3Provider.java:117)
at
org.apache.maven.surefire.junit.JUnit3Provider.invoke(JUnit3Provider.java:94)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at
org.apache.maven.surefire.util.ReflectionUtils.invokeMethodWithArray(ReflectionUtils.java:164)
at
org.apache.maven.surefire.booter.ProviderFactory$ProviderProxy.invoke(ProviderFactory.java:110)
at
org.apache.maven.surefire.booter.SurefireStarter.invokeProvider(SurefireStarter.java:172)
at
org.apache.maven.surefire.booter.SurefireStarter.runSuitesInProcessWhenForked(SurefireStarter.java:104)
at org.apache.maven.surefire.booter.ForkedBooter.main(ForkedBooter.java:70)

2011/7/28 Andreas Horst 

> Hi All
>
> Is anybody successfully running GWTTestCase (2.3) using the surefire plugin
> version 2.9? If so could you share your surefire configuration?
>
>
> I just recently switched to latest version of surefire for staying up to
> date and now have problems running my GWTTestCases with it. My surefire
> configuration is derived from 
> here<http://stackoverflow.com/questions/2737173/error-when-running-a-gwttestcase-using-maven-gwt-plugin>
> :
>
> 
>
>  false
>
>  
>
>
> ${basedir}/src/main/java
>
>
> ${basedir}/src/test/java
>
>  
>
>   
>
>
> and used to work fine with older surefire versions. Switching back to older
> versions also works but now I wonder what breaks it exactly.
>
> The error message is:
>
> Loading inherited module 'de.my.module.Module'
>[ERROR] Unable to find 'de/my/module/Module.gwt.xml' on your classpath;
> could be a typo, or maybe you forgot to include a classpath entry for
> source?
>
> Which actually seems to be pretty nonsense since
> a) it works with older versions
> b) the module is definitely on the classpath (see surefire configuration:
> .gwt.xml is in src/main/java; and a))
>
> Basic setup: Eclipse EE Indigo on Win7(64) with m2e 1.0.0.20110607, Maven
> 3.0.3, Sun JDK 1.6.0_24
>
> Cheers
>
> Andreas
>

-- 
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.



GWTTestCase run by surefire 2.9

2011-07-28 Thread Andreas Horst
Hi All

Is anybody successfully running GWTTestCase (2.3) using the surefire plugin
version 2.9? If so could you share your surefire configuration?


I just recently switched to latest version of surefire for staying up to
date and now have problems running my GWTTestCases with it. My surefire
configuration is derived from
here<http://stackoverflow.com/questions/2737173/error-when-running-a-gwttestcase-using-maven-gwt-plugin>
:


 false
 
   
${basedir}/src/main/java
   
${basedir}/src/test/java
 
  


and used to work fine with older surefire versions. Switching back to older
versions also works but now I wonder what breaks it exactly.

The error message is:

Loading inherited module 'de.my.module.Module'
   [ERROR] Unable to find 'de/my/module/Module.gwt.xml' on your classpath;
could be a typo, or maybe you forgot to include a classpath entry for
source?

Which actually seems to be pretty nonsense since
a) it works with older versions
b) the module is definitely on the classpath (see surefire configuration:
.gwt.xml is in src/main/java; and a))

Basic setup: Eclipse EE Indigo on Win7(64) with m2e 1.0.0.20110607, Maven
3.0.3, Sun JDK 1.6.0_24

Cheers

Andreas

-- 
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.



Prevent data url for images in css

2011-06-10 Thread Andreas
Is there a way to have background images in css files that are not
data urls? We have a bunch of small images that are shown only in
special case, so there is no need to load them on startup.

-- 
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 there a way to workaround GWT compiler/serializer/linker issue?

2011-03-16 Thread Andreas Horst
Do you use that abstract base class as a RPC parameter? If so, *of
course*GWT has to look for and try to compile all sub classes and
implementations
*since* they might get used in a RPC and hence need to be serializable and
compilable.

To avoid this you can either:
- refactor your RPC to use only the concrete types

*or* if you rely on using abstract classes or interfaces in RPC:
- you might want to configure your sub classes' packages to be part of a
module's client packages, so that the GWT compiler finds them (your "No
source found..." error indicates that this is not the case).

I personnaly think your problem is not the behaviour of the GWT compiler but
your project setup regarding separation into modules. Hope this can help
you.

2011/3/16 KD 

> Lets say I have package : com.mycom.model
>
> It has following classes
>
> com.mycom.model.PlatformMessage (an interface)
>
> com.mycom.model.AbstractMessage (an abstract class that implements
> PlatformMessage)
>
> com.mycom.model.QueryMessage (a concrete implementation of
> AbstractMessage)
>
> Now there is another package: com.mycom.app
>
> It has the following class...
>
> com.mycom.app.StreamMessage (a concrete implementation of
> AbstractMessage)
>
> ONLY com.mycom.model a GWT module.
>
> Now when I compile com.mycom.model, why does it need to go and look
> for all implementations of that abstract class AbstractMessage in
> OTHER packages (as I later find with many No Source found..."
> errors)?
>
> It seems RPC serializer is trying to serialize/compile all classes
> that IMPLEMENT an abstract class in my gwt module.
>
> Is there way to workaround this?
>
> --
> 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.
>
>

-- 
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.



Where to place external libs?

2011-03-08 Thread Andreas Wederbrand
Hi!

I'm running 2.2 on ubuntu and I'm including external libraries (log4j)
within my maven2 file but they don't get copied anywhere and when running
devMode I get noSuchClassException.

I've tried to put them in war/WEB-INF/lib but doing so gives me a 404 on my
normally working html start page.

I'm sure this is easy but I can't figure it out.

Thanks.

-- 
Read my blog @ http://www.raven.nu/blog

-- 
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: Upadate Data of CellList

2011-02-28 Thread Andreas
I hold the list internally and add the incoming data to it. So it
works when I call updateRow(0, wholeList). But I dont won't the
browser force to render the whole list. Lets say initially the list
has 50 items, I dont want to render 500 items after 9 loads, but only
to render 50 for every load.

This is the doc for m.google.gwt.view.client.HasData.setRowData:

Set a values associated with the rows in the visible range.

This method does not replace all rows in the display; it replaces the
row values starting at the specified start index through the length of
the the specified values.

If I understand this right, it should to what I want, only update the
values for specific rows, but it also delete all others.

On Feb 28, 3:30 pm, Jeff Schwartz  wrote:
> How are you maintaining the reference to the list of data objects you want
> to render?
>
> On Mon, Feb 28, 2011 at 9:26 AM, Greg Dougherty
> wrote:
>
>
>
>
>
>
>
>
>
> > Well, I haven't looked at the Showcase example, but I'd bet good money
> > that those null pointers are the result of your actions, not anything
> > in GWT.
>
> > Are you disposing of the data you got from the server after your call
> > to updateRow?  Remember, YOU are providing the data to the renderer.
> > That means you have to keep a copy of the data around for the renderer
> > whenever it wants it.
>
> > Greg
>
> > On Feb 25, 9:59 am, Andreas  wrote:
> > > > If you're only adding to the end of the list, then keep track of how
> > > > many items you've already added, and call updateRowData (prevNumAdded
> > > > - 1, newData);
>
> > > Unfortunately this works only on initial filling. The second time I
> > > call this I've got nullPointers in my cell.render method for the first
> > > items added. This is another thing I dont understand: why is the
> > > render method call for my whole list and not only for the one that
> > > I've added in the updateRowData method and why the first 10 are null.
>
> > > Also I dont understand which method forces the list to re render?
>
> > > > If you're changing things / adding things in the middle, you can do
> > > > all sorts of complicated things to figure out where the changes are,
> > > > and only give your CellList the changes, but you're probably better
> > > > off simply replacing everything and letting CellList worry about it.
>
> > > Draw the whole list is not that good cause you have a short delay
> > > where the list disappears. And there must be  a way to get it work, as
> > > shown in the showcase example.
>
> > --
> > 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.
>
> --
> *Jeff 
> Schwartz*http://jefftschwartz.appspot.com/http://www.linkedin.com/in/jefftschwartz
> follow me on twitter: @jefftschwartz

-- 
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: Upadate Data of CellList

2011-02-25 Thread Andreas

> If you're only adding to the end of the list, then keep track of how
> many items you've already added, and call updateRowData (prevNumAdded
> - 1, newData);
>

Unfortunately this works only on initial filling. The second time I
call this I've got nullPointers in my cell.render method for the first
items added. This is another thing I dont understand: why is the
render method call for my whole list and not only for the one that
I've added in the updateRowData method and why the first 10 are null.

Also I dont understand which method forces the list to re render?

> If you're changing things / adding things in the middle, you can do
> all sorts of complicated things to figure out where the changes are,
> and only give your CellList the changes, but you're probably better
> off simply replacing everything and letting CellList worry about it.

Draw the whole list is not that good cause you have a short delay
where the list disappears. And there must be  a way to get it work, as
shown in the showcase example.



-- 
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.



Upadate Data of CellList

2011-02-24 Thread Andreas
I have a celllist where I want add data loading from server and the
list should show all data. I use a AnsyncProvider and when I
succecfully get my data from the server I want to add the data to the
celllist. The only way I got this work is to hold a list of the data
I've got from the  server, where I add incoming data and then calling
updateRowData with the list of all incoming data:
List list;
onSuccess(Response response){
list.addAll(response.getItems());
  list.addAll(response);
  display.setVisibleRange(0, list.size());
  asyncDataProvider.updateRowData(0, list);
}

But this force the CellList to redraw the the list with the whole
dataset not only the new one. Like in the showcase example of cellList
(http://gwt.google.com/samples/Showcase/Showcase.html#!CwCellList).

So my question how can I show all data and only the new one will be
added to the view.

-- 
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.



@def in @if for Conditional CSS

2011-02-23 Thread Andreas
Is there a way to get @def statments to get work in @if statments.
Whenever I try this, the @def last @if statment wins.For example:

@def backgroundColor #e00;

@if  platform p1{
@def backgroundColor #F1C1BD;
}
@if platform p2{
   @def backgroundColor #FDDCC5;
}
@if platform p3{
@def backgroundColor #F7C2DB;
}

backgroundColor is always #F7C2DB. The only way I've got it work is to
have explicit css rules in the if statement, which isn't very elegant:

@if platform p1 {
  body {
background: #F1C1BD;
  }
}

@if platform p2 {
  body {
background: #FDDCC5;
  }
}

@if platform p3 {
  body {
background: #F7C2DB;
  }
}

The documentation says it should work, but isn't very helpfully btw:

What does it mean to have an @def or @eval in an @if block? Easy to
make this work for property-based @if statements; would have to
generate pretty gnarly runtime code to handle the expression-based @if
statement. Could have block-level scoping; but this seems like a
dubious use-case

-- 
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 is the best place to put client-side global constants ?

2011-01-12 Thread Andreas Horst
Well if you change the constant's visibility from private to public you can
access it from everywhere and hence only need to declare it once, that is in
one class or interface. Just make sure it's inside the module's client
package.

Besides this there is no really "good" place I can think of or it depends on
your overall architecture. For example if you wish to inherit constants via
a common interface or abstract base class. A simple solution would be to
place it (maybe along with other constants) in an interface or an abstract
utility (private constructor) class and access it statically, for example
"MyTokenizerConstants.SEPARATOR".

Note that you can also use default and protected visibility if you want
finer control, I think only private visibility is not that useful for shared
constants.

2011/1/12 zixzigma 

> where is the best place to put client-side global constants ?
>
> for example for Place tokenizing, I need to define a SEPARATOR
>
> private static final SEPARATOR = "!";
>
> and this appears in many packages,
> what is the correct way to define client-side constants once, in one
> place,
> to avoid duplication and scattering ?
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: Wave

2011-01-09 Thread Andreas Horst
2011/1/9 Ian Bambury 

> Polite is good, obviously. Asking for help is also, of course, OK. That's
> what the forum is about. The problem (I think) Andreas has is people making
> wildly general requests (how do I make something like Wave, tell me the
> answer) rather than trying to do it for themselves, hitting a block of some
> kind, and then asking for specific help with regard to how to approach a
> particular coding or design problem within GWT.
>
>
You read my mind! +1


> Anyone who has been here for a while recognises thinly veiled 'tell me the
> answer to my homework' posts - not that I'm saying the OP was doing that,
> but it *was* a _very_ general question rather than 'I get this error when I
> do something, what do I do to fix it?'
>
> You don't get much more generalised that 'Tell me how to recreate Wave'.
> Asked by someone who doesn't seem to realise that the 'i' in GUI actually
> stands for 'interface'.
>
> Ian
>
> On 9 January 2011 21:06, Harpal Grover  wrote:
>
>> On Sun, Jan 9, 2011 at 3:17 PM, nino ekambi 
>> wrote:
>> > Well i can not help you guys , but wanted to give my two cent  to Horst.
>> > I  disagree with  you saying that  people should not  asking about
>>  things
>> > they can not  figure out themselves.
>> > This is a community  and  i believe the main goal is to share knowledge.
>> > There are a lot people here in the group that spend their time studying
>> > GWT(Thomas Broyer and so on... )  and  give those informations for free.
>> > What would it  be  if they just say "Sorry i cant help, go and study the
>> > source code, the solution is there".
>> > If you hav  the infos and dont want to give them away that s your choice
>> and
>> >  i m ok with that. But it s alright to ask questions dough
>> > Regards,
>> > Alain
>>
>> I agree with you Alain. I believe we should try to be a bit more
>> polite when answering people's questions or trying to encourage
>> beginners to trace code a little bit before asking questions. Not
>> everyone is at the same level.
>>
>>
>>
>>
>>
>>
>>
>> --
>> Harpal Grover
>> President
>> Harpal Grover Consulting Inc
>>
>> --
>> 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-tool...@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.
>>
>>
>  --
> 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-tool...@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.
>

-- 
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-tool...@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: Wave

2011-01-09 Thread Andreas Horst
Hi,

2011/1/9 Harpal Grover 

> On Sun, Jan 9, 2011 at 3:17 PM, nino ekambi 
> wrote:
> > Well i can not help you guys , but wanted to give my two cent  to Horst.
> > I  disagree with  you saying that  people should not  asking about
>  things
> > they can not  figure out themselves.
>

I never said or meant to say that!


> > This is a community  and  i believe the main goal is to share knowledge.
> > There are a lot people here in the group that spend their time studying
> > GWT(Thomas Broyer and so on... )  and  give those informations for free.
> > What would it  be  if they just say "Sorry i cant help, go and study the
> > source code, the solution is there".
>

Sorry again, I just did not say that.


> > If you hav  the infos and dont want to give them away that s your choice
> and
> >  i m ok with that. But it s alright to ask questions dough
> > Regards,
> > Alain
>
> I agree with you Alain. I believe we should try to be a bit more
> polite when answering people's questions or trying to encourage
> beginners to trace code a little bit before asking questions. Not
> everyone is at the same level.
>
>
>
What motivated me to reply were the facts that the really simple answer has
already been posted by Dan and that this discussion now turns into a "how to
checkout&read foreign code" manual. In my opinion this is not really GWT
specific since as I posted Wave is a project of its own. One could kindly
ask there about UI development and specifics. Come on, just try and hit
Google with Wave, read and click and you'll immediately find the project and
even the source brows-able online; guys you don't even have to check it out.
What else besides Dan's reply do you need?

Now in what other reasonable manner could one answer politely and
constructive here?
- by showing people how to read and copy code? Well yes if non developers
ask but since this is GWT I assume members to have at least a bit knowledge
of software development.
- by manually copy/pasting code for the OP? Honestly please tell me no!
Especially since the targeted code does not belong to this group.
- ?


>
>
>
>
> --
> Harpal Grover
> President
> Harpal Grover Consulting Inc
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: Wave

2011-01-09 Thread Andreas Horst
To post my two cents: I clearly dislike the "Hey what a nice UI, where can I
get its widgets?"- attitude. That's why I will explicitly NOT give any
advice on how to scavenge the great Wave code base in order to rip out some
peaces of UI. I feel sorry for you if you are not able or willing to play
around with built-in GWT widgets and CSS to achieve a certain look and feel.
After all this is what the rest of us does every day.

In case you are not really interested in the project and only cloned the
repository in order to "get" some code please stop!

2011/1/9 Sudhakar 

> Guys,
>
> I have taken a clone of Wave protocol in my ubuntu.But i dont know how
> to start.
>
> Please help.
>
> Regards,
>
> Sudhakar
>
>
> On Sun, 2011-01-09 at 21:17 +0530, Deepak Singh wrote:
> >
> >
> > Hi All,
> > I also want to have UI like this. Any suggestion how to start this
> > way..
> >
> >
> >
> > On Sun, Jan 9, 2011 at 2:32 PM,  wrote:
> > Hi guys,
> >
> > Could you please help me creating a interface/GUI like wave.
> >
> > Regards,
> >
> > Sudhakar
> > Empower your Business with BlackBerry® and Mobile Solutions
> > from Etisalat
> >
> > --
> > 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-tool...@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.
> >
> >
> >
> >
> > --
> > 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-tool...@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.
>
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: Modular rpc blues

2010-12-23 Thread Andreas Horst
Good job, Patrick!

We're still using 1.3-SNAPSHOT ourself so I even wouldn't have been able to
solve the problem. I guess I'll think it over more than twice before
switching to newer versions. Anyway I'll leave a message as soon as I try it
myself with 2.1.0-1.

Until then maybe someone else could join who is already using 2.1.0-1.

Sorry I couldn't help.

2010/12/23 Coelho 

>  I checked the content and I have the servlet tag in a the module
> descriptor
> ( Check the contents of the .jar file. Does it contain the
> module descriptor? Does the module descriptor contain  tags for the
> RPCs? )
>
> I wonder if it is not a matter of gwt-maven-plugin version
> because the all sample compiles once and the version was 1.3-SNAPSHOT , if
> my memory is not failing
>
> then I had compilation problems ( not with thi project I think )
> so I cleared my maven local repository
> then from this moment on It complained about 1.3-SNAPSHOT that could not be
> found
> so I changed to 2.1.0-1
>
> this morning I reversed back to 1.3-SNAPSHOT : it compiles again ! an runs
>
> why not with 2.1.0-1 ?
>
> Thanks for reading
> Patrick
>
>  - Original Message -
> *From:* Andreas Horst 
> *To:* google-web-toolkit@googlegroups.com
> *Sent:* Wednesday, December 22, 2010 6:55 PM
> *Subject:* Re: Modular rpc blues
>
> Don't worry.
>
> I know this is not a trivial but nonetheless vital aspect of efficient
> development of reusable GWT modules. I also think that these topics are not
> well documented, at least at the time I found myself struggling with GWT
> module inheritance (+ Maven).
>
> 2010/12/22 Coelho 
>
>>  I'll Check and let you know
>> sorry for the trouble
>>
>> Patrick
>>
>>  - Original Message -
>> *From:* Andreas Horst 
>> *To:* google-web-toolkit@googlegroups.com
>>   *Sent:* Wednesday, December 22, 2010 6:41 PM
>> *Subject:* Re: Modular rpc blues
>>
>> Alright, now I get it!
>>
>> To bundle RPC functionality in a reusable (.jar) GWT module you only need
>> to declare any RPC servlet in the module's module descriptor. That's all.
>> "gwt:mergewebxml" will create appropriate servlet mappings in the target
>> web.xml. It's really that easy.
>>
>> Make sure the .jar really  contains the module descriptor (*.gwt.xml) and
>> that the RPC servlets are actually declared in it. If you try it with a
>> module that used to be an application and only declared its RPC servlets
>> directly in its web.xml you won't be able to use them via "gwt:mergewebxml"
>> AFAIK.
>>
>> Check the contents of the .jar file. Does it contain the module
>> descriptor? Does the module descriptor contain  tags for the RPCs?
>>
>> 2010/12/22 Metronome / Basic 
>>
>>>  In fact I simply want to be able to use a jar containing GWT-RPC code in
>>> any webapp
>>>
>>> As I had no succes with my code
>>>
>>> I tried to rely on "maven-googlewebtoolkit2-sample"
>>> that is a HelloWorld RPC example with code in different modules.
>>>
>>> here is the parent pom
>>>
>>> I did'nt change the code , I just modified the poms
>>>
>>>
>>>  - Original Message -
>>> *From:* Andreas Horst 
>>> *To:* google-web-toolkit@googlegroups.com
>>>   *Sent:* Wednesday, December 22, 2010 1:52 PM
>>> *Subject:* Re: Modular rpc blues
>>>
>>> I just took a look at your configuration files.
>>>
>>> Actually I can't tell much from just them especially since I don't have
>>> the parent POM. Anyway I'm a bit confused about what you are trying to do.
>>>
>>> Are you only trying to compile the sample project or are you trying to
>>> reuse something of it? What exactly do you mean with "project war rpc and
>>> server" or "war server"? A project that packages to a .war file (like GWT
>>> applications) or a .war file you are trying to include via module
>>> inheritance? I'm sorry but it's really not clear to me.
>>>
>>> 2010/12/22 Metronome / Basic 
>>>
>>>>  Hello
>>>> Thanks for trying to help me !
>>>>
>>>> One of my tries is : compile the "maven-googlewebtoolkit2-sample"
>>>> using gwt-maven-plugin
>>>> it was a three modules project war rpc and server
>>>> I reduced it to war server , by merging rpc and server
>>>>
>>>> I join 

Re: Modular rpc blues

2010-12-22 Thread Andreas Horst
Don't worry.

I know this is not a trivial but nonetheless vital aspect of efficient
development of reusable GWT modules. I also think that these topics are not
well documented, at least at the time I found myself struggling with GWT
module inheritance (+ Maven).

2010/12/22 Coelho 

>  I'll Check and let you know
> sorry for the trouble
>
> Patrick
>
> - Original Message -
> *From:* Andreas Horst 
> *To:* google-web-toolkit@googlegroups.com
> *Sent:* Wednesday, December 22, 2010 6:41 PM
> *Subject:* Re: Modular rpc blues
>
> Alright, now I get it!
>
> To bundle RPC functionality in a reusable (.jar) GWT module you only need
> to declare any RPC servlet in the module's module descriptor. That's all.
> "gwt:mergewebxml" will create appropriate servlet mappings in the target
> web.xml. It's really that easy.
>
> Make sure the .jar really  contains the module descriptor (*.gwt.xml) and
> that the RPC servlets are actually declared in it. If you try it with a
> module that used to be an application and only declared its RPC servlets
> directly in its web.xml you won't be able to use them via "gwt:mergewebxml"
> AFAIK.
>
> Check the contents of the .jar file. Does it contain the module descriptor?
> Does the module descriptor contain  tags for the RPCs?
>
> 2010/12/22 Metronome / Basic 
>
>>  In fact I simply want to be able to use a jar containing GWT-RPC code in
>> any webapp
>>
>> As I had no succes with my code
>>
>> I tried to rely on "maven-googlewebtoolkit2-sample"
>> that is a HelloWorld RPC example with code in different modules.
>>
>> here is the parent pom
>>
>> I did'nt change the code , I just modified the poms
>>
>>
>>  - Original Message -
>> *From:* Andreas Horst 
>> *To:* google-web-toolkit@googlegroups.com
>>   *Sent:* Wednesday, December 22, 2010 1:52 PM
>> *Subject:* Re: Modular rpc blues
>>
>> I just took a look at your configuration files.
>>
>> Actually I can't tell much from just them especially since I don't have
>> the parent POM. Anyway I'm a bit confused about what you are trying to do.
>>
>> Are you only trying to compile the sample project or are you trying to
>> reuse something of it? What exactly do you mean with "project war rpc and
>> server" or "war server"? A project that packages to a .war file (like GWT
>> applications) or a .war file you are trying to include via module
>> inheritance? I'm sorry but it's really not clear to me.
>>
>> 2010/12/22 Metronome / Basic 
>>
>>>  Hello
>>> Thanks for trying to help me !
>>>
>>> One of my tries is : compile the "maven-googlewebtoolkit2-sample"
>>> using gwt-maven-plugin
>>> it was a three modules project war rpc and server
>>> I reduced it to war server , by merging rpc and server
>>>
>>> I join the pom and web.xml of the war
>>>
>>>
>>> mvn package report
>>>
>>> ERROR] Failed to execute goal
>>> org.codehaus.mojo:gwt-maven-plugin:2.1.0-1:mergewebxml (default) on
>>> project
>>> maven-googlewebtoolkit2-sample-war: Unable to merge web.xml:
>>> NullPointerException -> [Help 1]
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>   - Original Message -
>>> *From:* Andreas Horst 
>>> *To:* google-web-toolkit@googlegroups.com
>>> *Sent:* Wednesday, December 22, 2010 12:26 AM
>>> *Subject:* Re: Modular rpc blues
>>>
>>> Another question just coming to my mind:
>>>
>>> Where in the inherited module are you declaring the RPC servlet?
>>>
>>> If you declare it in the inherited module's web.xml then make sure the
>>> Maven GWT plugin parameter "webXml" properly points to it (Not sure how one
>>> would do this - never did it myself - especially since  inherited modules
>>> usually come as a .jar. If so does yours include the web.xml?).
>>>
>>> If you declare it in the inherited module's module descriptor (a..k.a.
>>> *.gwt.xml) all should be fine. Hence I assume you do not declare it
>>> there? Look 
>>> here<http://www.gwtapps.com/doc/html/com.google.gwt..doc.DeveloperGuide.Fundamentals.Modules.ModuleXml.html>
>>>  for
>>> details about the module descriptor, note the  tag.
>>>
>>> My 2cents: Use the second option.
>>>
>>> Why? Because obviously your inherit

Re: Modular rpc blues

2010-12-22 Thread Andreas Horst
Alright, now I get it!

To bundle RPC functionality in a reusable (.jar) GWT module you only need to
declare any RPC servlet in the module's module descriptor. That's all.
"gwt:mergewebxml" will create appropriate servlet mappings in the target
web.xml. It's really that easy.

Make sure the .jar really  contains the module descriptor (*.gwt.xml) and
that the RPC servlets are actually declared in it. If you try it with a
module that used to be an application and only declared its RPC servlets
directly in its web.xml you won't be able to use them via "gwt:mergewebxml"
AFAIK.

Check the contents of the .jar file. Does it contain the module descriptor?
Does the module descriptor contain  tags for the RPCs?

2010/12/22 Metronome / Basic 

>  In fact I simply want to be able to use a jar containing GWT-RPC code in
> any webapp
>
> As I had no succes with my code
>
> I tried to rely on "maven-googlewebtoolkit2-sample"
> that is a HelloWorld RPC example with code in different modules.
>
> here is the parent pom
>
> I did'nt change the code , I just modified the poms
>
>
> - Original Message -
> *From:* Andreas Horst 
> *To:* google-web-toolkit@googlegroups.com
> *Sent:* Wednesday, December 22, 2010 1:52 PM
> *Subject:* Re: Modular rpc blues
>
> I just took a look at your configuration files.
>
> Actually I can't tell much from just them especially since I don't have the
> parent POM. Anyway I'm a bit confused about what you are trying to do.
>
> Are you only trying to compile the sample project or are you trying to
> reuse something of it? What exactly do you mean with "project war rpc and
> server" or "war server"? A project that packages to a .war file (like GWT
> applications) or a .war file you are trying to include via module
> inheritance? I'm sorry but it's really not clear to me.
>
> 2010/12/22 Metronome / Basic 
>
>>  Hello
>> Thanks for trying to help me !
>>
>> One of my tries is : compile the "maven-googlewebtoolkit2-sample"
>> using gwt-maven-plugin
>> it was a three modules project war rpc and server
>> I reduced it to war server , by merging rpc and server
>>
>> I join the pom and web.xml of the war
>>
>>
>> mvn package report
>>
>> ERROR] Failed to execute goal
>> org.codehaus.mojo:gwt-maven-plugin:2.1.0-1:mergewebxml (default) on
>> project
>> maven-googlewebtoolkit2-sample-war: Unable to merge web.xml:
>> NullPointerException -> [Help 1]
>>
>>
>>
>>
>>
>>
>>
>>   - Original Message -
>> *From:* Andreas Horst 
>> *To:* google-web-toolkit@googlegroups.com
>> *Sent:* Wednesday, December 22, 2010 12:26 AM
>> *Subject:* Re: Modular rpc blues
>>
>> Another question just coming to my mind:
>>
>> Where in the inherited module are you declaring the RPC servlet?
>>
>> If you declare it in the inherited module's web.xml then make sure the
>> Maven GWT plugin parameter "webXml" properly points to it (Not sure how one
>> would do this - never did it myself - especially since  inherited modules
>> usually come as a .jar. If so does yours include the web.xml?).
>>
>> If you declare it in the inherited module's module descriptor (a..k.a.
>> *.gwt.xml) all should be fine. Hence I assume you do not declare it
>> there? Look 
>> here<http://www.gwtapps.com/doc/html/com.google.gwt..doc.DeveloperGuide.Fundamentals.Modules.ModuleXml.html>
>>  for
>> details about the module descriptor, note the  tag.
>>
>> My 2cents: Use the second option.
>>
>> Why? Because obviously your inherited module realizes functionality that
>> is to be reused. It is hence some sort of library and not (only?) an
>> application or even a .war packed web application. All our inherited modules
>> are _library_ modules, they don't get deployed on an _application_ server on
>> their own. Now if one of those features functionality through RPC (we have
>> some of those) we thankfully use the above mentioned  tag in the
>> module descriptor and let the Maven GWT plugin do its job. IMHO on the one
>> hand a web.xml does not belong into a common not "runnable" module and on
>> the other one a (.war packed) application is not best suitable for
>> inheriting functionality.
>>
>> Regards
>>
>> 2010/12/21 Andreas Horst 
>>
>>>
>>>
>>> 2010/12/21 Thomas Broyer 
>>>
>>>
>>>>
>>>> On Tuesday, December 21, 2010 9:51:38 AM UTC+1, coelho wr

Re: Modular rpc blues

2010-12-22 Thread Andreas Horst
I just took a look at your configuration files.

Actually I can't tell much from just them especially since I don't have the
parent POM. Anyway I'm a bit confused about what you are trying to do.

Are you only trying to compile the sample project or are you trying to reuse
something of it? What exactly do you mean with "project war rpc and server"
or "war server"? A project that packages to a .war file (like GWT
applications) or a .war file you are trying to include via module
inheritance? I'm sorry but it's really not clear to me.

2010/12/22 Metronome / Basic 

>  Hello
> Thanks for trying to help me !
>
> One of my tries is : compile the "maven-googlewebtoolkit2-sample"
> using gwt-maven-plugin
> it was a three modules project war rpc and server
> I reduced it to war server , by merging rpc and server
>
> I join the pom and web.xml of the war
>
>
> mvn package report
>
> ERROR] Failed to execute goal
> org.codehaus.mojo:gwt-maven-plugin:2.1.0-1:mergewebxml (default) on project
>
> maven-googlewebtoolkit2-sample-war: Unable to merge web.xml:
> NullPointerException -> [Help 1]
>
>
>
>
>
>
>
> - Original Message -
> *From:* Andreas Horst 
> *To:* google-web-toolkit@googlegroups.com
> *Sent:* Wednesday, December 22, 2010 12:26 AM
> *Subject:* Re: Modular rpc blues
>
> Another question just coming to my mind:
>
> Where in the inherited module are you declaring the RPC servlet?
>
> If you declare it in the inherited module's web.xml then make sure the
> Maven GWT plugin parameter "webXml" properly points to it (Not sure how one
> would do this - never did it myself - especially since  inherited modules
> usually come as a .jar. If so does yours include the web.xml?).
>
> If you declare it in the inherited module's module descriptor (a.k.a.
> *.gwt.xml) all should be fine. Hence I assume you do not declare it
> there? Look 
> here<http://www.gwtapps.com/doc/html/com.google.gwt.doc.DeveloperGuide.Fundamentals.Modules.ModuleXml.html>
>  for
> details about the module descriptor, note the  tag.
>
> My 2cents: Use the second option.
>
> Why? Because obviously your inherited module realizes functionality that is
> to be reused. It is hence some sort of library and not (only?) an
> application or even a .war packed web application. All our inherited modules
> are _library_ modules, they don't get deployed on an _application_ server on
> their own. Now if one of those features functionality through RPC (we have
> some of those) we thankfully use the above mentioned  tag in the
> module descriptor and let the Maven GWT plugin do its job. IMHO on the one
> hand a web.xml does not belong into a common not "runnable" module and on
> the other one a (.war packed) application is not best suitable for
> inheriting functionality.
>
> Regards
>
> 2010/12/21 Andreas Horst 
>
>>
>>
>> 2010/12/21 Thomas Broyer 
>>
>>
>>>
>>> On Tuesday, December 21, 2010 9:51:38 AM UTC+1, coelho wrote:
>>>>
>>>>  Hello
>>>>
>>>> What seems to me great in GWT is that it's easy to build
>>>> client code and server code that can communicate through GWT-RPC.
>>>>
>>>> What 's great too is that you can write modules
>>>> and your webapp can use those modules.
>>>>
>>>> Then why is it so complicated ( is it possible ? ) to have a module with
>>>> GWT-RPC code
>>>> ( implementation and interfaces )
>>>> that could be used in a webapp
>>>>
>>>
>>> And by "so complicated" you mean adding half a dozen lines to your
>>> web.xml file, right?
>>>
>>>
>>
>> Actually the goal gwt:mergewebxml is really ALL you need (believe me, we
>> use it just like that for exactly what you are trying to). Please clarify
>> what you mean with "web.xml refers to external module". I assume either your
>> web.xml gets or already is troubled or your POM is not configured properly.
>>
>>
>>>   I tried many things in eclipse
>>>> I tried many things with maven
>>>> ( gwt-maven-plugin : goal mergewebxml ) fails when web.xml refers to
>>>> external module
>>>>
>>>> still no success !
>>>>
>>>
>>> Have a look at the cargo maven plugin (I haven't tried it though)
>>>
>>>   I wondered if there is such a project already done
>>>>
>>>> Is there somewhere a jar , ready

Re: Modular rpc blues

2010-12-21 Thread Andreas Horst
Another question just coming to my mind:

Where in the inherited module are you declaring the RPC servlet?

If you declare it in the inherited module's web.xml then make sure the Maven
GWT plugin parameter "webXml" properly points to it (Not sure how one would
do this - never did it myself - especially since  inherited modules usually
come as a .jar. If so does yours include the web.xml?).

If you declare it in the inherited module's module descriptor (a.k.a.
*.gwt.xml) all should be fine. Hence I assume you do not declare it
there? Look 
here<http://www.gwtapps.com/doc/html/com.google.gwt.doc.DeveloperGuide.Fundamentals.Modules.ModuleXml.html>
for
details about the module descriptor, note the  tag.

My 2cents: Use the second option.

Why? Because obviously your inherited module realizes functionality that is
to be reused. It is hence some sort of library and not (only?) an
application or even a .war packed web application. All our inherited modules
are _library_ modules, they don't get deployed on an _application_ server on
their own. Now if one of those features functionality through RPC (we have
some of those) we thankfully use the above mentioned  tag in the
module descriptor and let the Maven GWT plugin do its job. IMHO on the one
hand a web.xml does not belong into a common not "runnable" module and on
the other one a (.war packed) application is not best suitable for
inheriting functionality.

Regards

2010/12/21 Andreas Horst 

>
>
> 2010/12/21 Thomas Broyer 
>
>
>>
>> On Tuesday, December 21, 2010 9:51:38 AM UTC+1, coelho wrote:
>>>
>>>  Hello
>>>
>>> What seems to me great in GWT is that it's easy to build
>>> client code and server code that can communicate through GWT-RPC.
>>>
>>> What 's great too is that you can write modules
>>> and your webapp can use those modules.
>>>
>>> Then why is it so complicated ( is it possible ? ) to have a module with
>>> GWT-RPC code
>>> ( implementation and interfaces )
>>> that could be used in a webapp
>>>
>>
>> And by "so complicated" you mean adding half a dozen lines to your web.xml
>> file, right?
>>
>>
>
> Actually the goal gwt:mergewebxml is really ALL you need (believe me, we
> use it just like that for exactly what you are trying to). Please clarify
> what you mean with "web.xml refers to external module". I assume either your
> web.xml gets or already is troubled or your POM is not configured properly.
>
>
>>  I tried many things in eclipse
>>> I tried many things with maven
>>> ( gwt-maven-plugin : goal mergewebxml ) fails when web.xml refers to
>>> external module
>>>
>>> still no success !
>>>
>>
>> Have a look at the cargo maven plugin (I haven't tried it though)
>>
>>  I wondered if there is such a project already done
>>>
>>> Is there somewhere a jar , ready made , with GWT-RPC included that I
>>> could use as a reference ?
>>>
>>> or is hopeless ?
>>>
>>
>> I believe that's what web-fragments in Servlets 3.0 are meant to solve:
>>
>> http://java.sun.com/developer/technicalArticles/JavaEE/JavaEE6Overview_Part2.html#webfrags
>>
>>
>> --
>> 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-tool...@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.
>>
>


-- 
Andreas Horst
Schwicheldtstraße 23, 38704 Liebenburg
Tel. +49 (0)170 4162251, mailto:horst.andrea...@googlemail.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-tool...@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: Modular rpc blues

2010-12-21 Thread Andreas Horst
2010/12/21 Thomas Broyer 

>
>
> On Tuesday, December 21, 2010 9:51:38 AM UTC+1, coelho wrote:
>>
>>  Hello
>>
>> What seems to me great in GWT is that it's easy to build
>> client code and server code that can communicate through GWT-RPC.
>>
>> What 's great too is that you can write modules
>> and your webapp can use those modules.
>>
>> Then why is it so complicated ( is it possible ? ) to have a module with
>> GWT-RPC code
>> ( implementation and interfaces )
>> that could be used in a webapp
>>
>
> And by "so complicated" you mean adding half a dozen lines to your web.xml
> file, right?
>
>

Actually the goal gwt:mergewebxml is really ALL you need (believe me, we use
it just like that for exactly what you are trying to). Please clarify what
you mean with "web.xml refers to external module". I assume either your
web.xml gets or already is troubled or your POM is not configured properly.


> I tried many things in eclipse
>> I tried many things with maven
>> ( gwt-maven-plugin : goal mergewebxml ) fails when web.xml refers to
>> external module
>>
>> still no success !
>>
>
> Have a look at the cargo maven plugin (I haven't tried it though)
>
> I wondered if there is such a project already done
>>
>> Is there somewhere a jar , ready made , with GWT-RPC included that I could
>> use as a reference ?
>>
>> or is hopeless ?
>>
>
> I believe that's what web-fragments in Servlets 3.0 are meant to solve:
>
> http://java.sun.com/developer/technicalArticles/JavaEE/JavaEE6Overview_Part2.html#webfrags
>
>
> --
> 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-tool...@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.
>

-- 
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-tool...@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: execute same code for different widget event handle

2010-12-16 Thread Andreas Horst
The value of 
UIHandler<http://google-web-toolkit.googlecode.com/svn/javadoc/2.0/com/google/gwt/uibinder/client/UiHandler.html>
is
of type String[]. Just pass your arguments as an array. It's as "peace of
cake" as reading the docs ;-)

@UiHandler({"streetName", "streetNumber"})

Regards,

Andreas

2010/12/16 Jeff Schwartz 

> Have both handlers call the same routine
>
> On Dec 16, 2010 7:45 AM, "pieceovcake"  wrote:
>
> this works great
>
> @UiHandler("streetName")
> void handleStreetNameKeyPress(KeyPressEvent  e) {
> // code
> }
>
> but i want to execute the same handleStreetNameKeyPress triggered from
> another field on the page
> something like this
>
> @UiHandler("streetNumber")
> @UiHandler("streetName")
> void handleStreetNameKeyPress(KeyPressEvent  e) {
> // code
> }
>
> how do i write this?
>
>
>
>
>
>  --
> 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-tool...@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.
>
>  --
> 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-tool...@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.
>

-- 
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-tool...@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: Announcing Chrome Developer Plugin support for Mac and Linux

2010-12-10 Thread Andreas Horst
Thanks for the link!

Just starred http://code.google.com/p/chromium/issues/detail?id=5751.

Regards,

Andreas

2010/12/10 Chris Conroy 

> http://code.google.com/p/google-web-toolkit/issues/detail?id=4493#c24
>
> On Fri, Dec 10, 2010 at 11:38 AM, Andreas Horst <
> horst.andrea...@googlemail.com> wrote:
>
>> There was a 
>> post<http://code.google.com/p/google-web-toolkit/issues/detail?id=4325#c99> 
>> about
>> the performance of the plugin for Chrome. Obviously this is not related to
>> the plugin itself (see post for details) so I guess we will have to live
>> with it and/or use a different browser than Chrome.
>>
>> Maybe a plugin specialist could clear things up for us especially since
>> this is not a bug in the plugin?
>>
>> Regards,
>>
>> Andreas
>>
>>
>> 2010/12/10 koma 
>>
>> Same here, not usable; switched back to FF.
>>>
>>> On Dec 10, 12:19 pm, Vagner Araujo  wrote:
>>> > Hello Friends,
>>> >
>>> > gwt plugin work in my google-chrome 8.0.552.215. But in development
>>> mode
>>> > It's very very very slow.
>>> >
>>> > I'm using Slackware Linux Current 32bits with xfce4, eclipse Helios and
>>> 2GB
>>> > ram.
>>> >
>>> > Thanks,
>>> > --
>>> > *
>>> > Vagner Araujo
>>> >
>>> > O PLANETA É O MEU PAÍS, E A CIÊNCIA É A MINHA RELIGIÃO ! INDO AO
>>> INFINITO E
>>> > ALÉM... !!
>>> > *
>>>
>>> --
>>> 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.
>>>
>>>   --
>> 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-tool...@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.
>>
>
>  --
> 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-tool...@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.
>

-- 
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-tool...@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: Announcing Chrome Developer Plugin support for Mac and Linux

2010-12-10 Thread Andreas Horst
There was a 
post<http://code.google.com/p/google-web-toolkit/issues/detail?id=4325#c99>
about
the performance of the plugin for Chrome. Obviously this is not related to
the plugin itself (see post for details) so I guess we will have to live
with it and/or use a different browser than Chrome.

Maybe a plugin specialist could clear things up for us especially since this
is not a bug in the plugin?

Regards,

Andreas


2010/12/10 koma 

> Same here, not usable; switched back to FF.
>
> On Dec 10, 12:19 pm, Vagner Araujo  wrote:
> > Hello Friends,
> >
> > gwt plugin work in my google-chrome 8.0.552.215. But in development mode
> > It's very very very slow.
> >
> > I'm using Slackware Linux Current 32bits with xfce4, eclipse Helios and
> 2GB
> > ram.
> >
> > Thanks,
> > --
> > *
> > Vagner Araujo
> >
> > O PLANETA É O MEU PAÍS, E A CIÊNCIA É A MINHA RELIGIÃO ! INDO AO INFINITO
> E
> > ALÉM... !!
> > *
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: Announcing Chrome Developer Plugin support for Mac and Linux

2010-12-09 Thread Andreas Horst
Your SYSTEM's default browser is used. AFAIK this has nothing to do with
Eclipse and/or GWT.

Regards,

Andreas

2010/12/9 Kenneth Jacker 

> Thanks for the new Google Chrome GWT plugin for Linux.  Of course, I
> want to use it!
>
> I have looked (almost, apparently) everywhere, but cannot find a way
> to change the GWT devmode "default browser" from Firefox to Chrome.
>
> I've changed Eclipe's "General/Web Browser" preference to Chrome, but
> that didn't work.  Firefox keeps running when I hit the "Launch
> Default Browser" button in the devmode window.  So, I am forced to
> "Copy to Clipboard", run Chrome, and paste the URL ... :(
>
> How can we change the default browser?
>
> I appreciate your suggestions/comments!
>
> -Kenneth
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: RPC in separate module

2010-12-08 Thread Andreas Horst
Well, yes. But that's not what I meant. These are the servlet declarations
in the DEPLOYMENT descriptor (aka web.xml).

What I meant is you could declare the servlet in the MODULE descriptor (aka
mymodule.gwt.xml) of your inherited module.

A servlet declaration in a module descriptor looks like this:


However what you still need to do is get an appropriate servlet mapping in
the deployment descriptor of the inheriting module. There are several
options:
- with Maven: use goal "gwt:mergewebxml"
- with Ant/GWT compiler: I don't know, ask the group!
- manually edit the deployment descriptor like you just proposed, however
make sure the url-pattern matches the
RemoteServiceRelativePath<http://google-web-toolkit.googlecode.com/svn/javadoc/1.6/com/google/gwt/user/client/rpc/RemoteServiceRelativePath.html>
and
probably the "rename-to" directive (not sure about this last point though
especially in case both your inherited and the inheriting module declare
(different) directives)

Hope I did not create more confusion about this.

Regards,

Andreas

2010/12/8 Metronome / Basic 

> Thanks for your help , I think that it must be something like that
> so I tried
>
> 
> userServlet
>
> metro.app.tmupack.server.MyModServiceImpl
>
>
> 
>
>
> 
>
> userServlet
>
> /metro.app.tmupack/modervice
>
> 
>
> where metro.app.tmupack is from the main module but , ie does't work
>
> I guess that the url of the module have to be addes somehow ?
>
> Any Idea
>
> Thanks
>
> Patrick
>
>
> - Original Message - From: "Brian Reilly"  >
> To: 
> Sent: Tuesday, December 07, 2010 8:15 PM
>
> Subject: Re: RPC in separate module
>
>
> You do have to declare the servlet in the web.xml of the main module.
> As such, you have to treat it like it's in the main module. I suspect
> the "metro.module.rpcpack" part of the url-pattern is the problem. The
> URL pattern is determined by the main module name, not the inherited
> module name or the package containing the RPC code.
>
> -Brian
>
> On Tue, Dec 7, 2010 at 12:17 PM, Metronome / Basic
>  wrote:
>
>> It certainly is but I cannot find it
>>
>> I tried
>> 
>>
>> userServlet
>>
>>
>> metro.module.rpcpack.server.MyModServiceImpl
>>
>> 
>>
>>
>> 
>>
>> userServlet
>>
>> /metro.module.rpcpack/modervice
>>
>> 
>>
>> in the web.xml of the main module
>> metro.module.rpcpack is the package name of the module with the rpc
>>
>> should the declaration of the servlet be in the main module ?
>> what is the path
>>
>> I cannot sort it out
>>
>> Thanks
>>
>> Patrick
>>
>> - Original Message - From: "Didier Durand" <
>> durand.did...@gmail.com>
>> To: "Google Web Toolkit" 
>> Sent: Tuesday, December 07, 2010 1:08 PM
>> Subject: Re: RPC in separate module
>>
>>
>> Hi,
>>
>> Everything you need is detailed here:
>> http://code.google.com/webtoolkit/doc/1.6/DevGuideOrganizingProjects.html
>>
>> On Dec 7, 11:06 am, coelho  wrote:
>>
>>>
>>> Hello
>>>
>>> I'm trying to bild a small application using RPC to acces SQL data
>>> ( jdbc )
>>>
>>> It works no problem so far
>>>
>>> I'd like this application to use modules , so that the code could be
>>> reused in others
>>>
>>> so I'd like to put the SQL parts in a separate module
>>>
>>> and the question is :
>>>
>>> How do You declare a servlet that is not in the main module
>>> ( application)
>>>
>>> or am I missing something ? ( probably)
>>>
>>> Thanks for reading
>>>
>>> Patrick
>>>
>>
>> --
>> 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-tool...@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.
>>
>>
>>
>>
>> -
>> Aucun virus trouvé dans ce message.
>> Analyse effectuée par AVG - www.avg.fr
>> Version: 10.0.1170 / Base de données virale: 426/3301 - Date: 06/12/2010
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Google Web Toolkit&

Re: RPC in separate module

2010-12-07 Thread Andreas Horst
Hi,

if you are using Maven you do not need to define a servlet mapping in any
web.xml manually in case of inheriting modules. The GWT Maven plugin is
capable of merging servlet mappings declared in a super-module's gwt.xml
into the inheriting module's web.xml for you; see the goal
"gwt:mergewebxml".

You must use the  tag in the module descriptor xml and NOT in the
web.xml to declare servlets to pass to inheriting modules. Maybe someone
else can tell you how the merge functionality may be used without Maven by
the GWT compiler itself?

Regards,

Andreas

2010/12/7 Brian Reilly 

> You do have to declare the servlet in the web.xml of the main module.
> As such, you have to treat it like it's in the main module. I suspect
> the "metro.module.rpcpack" part of the url-pattern is the problem. The
> URL pattern is determined by the main module name, not the inherited
> module name or the package containing the RPC code.
>
> -Brian
>
> On Tue, Dec 7, 2010 at 12:17 PM, Metronome / Basic
>  wrote:
> > It certainly is but I cannot find it
> >
> > I tried
> > 
> >
> > userServlet
> >
> >
> metro.module.rpcpack.server.MyModServiceImpl
> >
> > 
> >
> >
> > 
> >
> > userServlet
> >
> > /metro.module.rpcpack/modervice
> >
> > 
> >
> > in the web.xml of the main module
> > metro.module.rpcpack  is the package name of the module with the rpc
> >
> > should the declaration of the servlet be in the main module ?
> > what is the path
> >
> > I cannot sort it out
> >
> > Thanks
> >
> > Patrick
> >
> > - Original Message - From: "Didier Durand" <
> durand.did...@gmail.com>
> > To: "Google Web Toolkit" 
> > Sent: Tuesday, December 07, 2010 1:08 PM
> > Subject: Re: RPC in separate module
> >
> >
> > Hi,
> >
> > Everything you need is detailed here:
> >
> http://code.google.com/webtoolkit/doc/1.6/DevGuideOrganizingProjects.html
> >
> > On Dec 7, 11:06 am, coelho  wrote:
> >>
> >> Hello
> >>
> >> I'm trying to bild a small application using RPC to acces SQL data
> >> ( jdbc )
> >>
> >> It works no problem so far
> >>
> >> I'd like this application to use modules , so that the code could be
> >> reused in others
> >>
> >> so I'd like to put the SQL parts in a separate module
> >>
> >> and the question is :
> >>
> >> How do You declare a servlet that is not in the main module
> >> ( application)
> >>
> >> or am I missing something ? ( probably)
> >>
> >> Thanks for reading
> >>
> >> Patrick
> >
> > --
> > 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.
> >
> >
> >
> >
> > -
> > Aucun virus trouvé dans ce message.
> > Analyse effectuée par AVG - www.avg.fr
> > Version: 10.0.1170 / Base de données virale: 426/3301 - Date: 06/12/2010
> >
> >
> > --
> > 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.
> >
> >
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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 client-side thread or asynchronous call ?

2010-11-18 Thread andreas
If you really want to stick with the client side and want to have an
async-like experience you could also use a Timer to trigger the
search. Schedule the Timer as soon as the user begins to enter his
search and cancel + re-schedule it on every key hit. This way the
blocking search will only start as soon as the user has stopped typing
for a specific amount of time. You will have to try and check what is
a reasonable schedule time. This should work I think.

On 18 Nov., 15:09, MickeyR  wrote:
> Is there anything similar to using java threads that I can use in myclient 
> sidecode ?
>
> In my application a user types a search term into a Text box. For
> every character typed in my code goes and searches in a dataset for
> results containing the search string. These results are then returned
> and displayed in a List. But the problem is that because my search +
> display code is so slow, it makes the typing in very sluggish for the
> user (i.e. they can’t type the next character in until the search +
> display code has returned).
>
> Ideally I’d like to run the search+display code in a separatethread–
> so that the user can continue to type in the Text box. But I
> understand that threads are not supported inGWTclient-sidecoding ?
> Or is it possible to have the search + display in an 
> onlyclient-sideasynchronouscall?
>
> Any help, much appreciated,
> M.

-- 
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-tool...@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.



Using REST with GWT

2010-11-05 Thread Andreas
We're start a project using gwt 2.1. What is the best way to
communicate with a rest interface. Especially how to batch multiple
requests and how to cache response on the client using webstorage or
other client side technics.

-- 
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-tool...@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.



RPC and session timeout

2010-10-26 Thread andreas
Hi everyone,

what is a reasonable setup/realization for enabling "restoring" RPCs
after a session timeout, server restart/disconnect and re-login?

Take the following scenario:
- GWT application on standard Tomcat
- container managed authentication via login.jsp (simply redirects to
GWT html upon authentication)

Now if a timeout occurs or the application server is restarted or the
client was disconnect long enough for the session to expire and the
application then issues a RPC it will fail, since the session is lost
and a new login is required. A copy/paste, "hello world" login.jsp
would simply redirect to the starting point GWT application html. This
causes the application to be simply started again (restarted) and all
previous internal state like in memory cache is lost. So are the RPCs
issued as the session was already expired.

Related posts in this group:
http://groups.google.com/group/google-web-toolkit/browse_thread/thread/91f74a1b6293b89b/ae90dee11b49408e
http://groups.google.com/group/google-web-toolkit/browse_thread/thread/7d6b8dd5175e0e47/d999a1dd0e50287b

We already realized the heart beat pattern (periodical ping) to
prevent the session from timing out however such cases like a server
restart or client workstation temporarily losing network connection
still form scenarios where RPCs and hence data might get lost.

Regards

-- 
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-tool...@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: Compile with different color themes

2010-10-25 Thread Andreas
I want to change the color during the compiling. But I dont understand
how to create a generator for the different themes.

On Oct 22, 4:33 pm, Stefan Bachert  wrote:
> Hi Andreas,
>
> Your question does not make clear how you want to change thecolor
> values.
> Surely you can change the value incssand recompile, but this is
> probably not what you want
>
> Stefan Bacherthttp://gwtworld.de
>
> On 21 Okt., 09:48, Andreas Köberle  wrote:
>
>
>
>
>
>
>
> > Is there a way to get the application compiled with different colors
> > themes? I want to have acsswhere I only write:
>
> > h1 {color: color1} and the compiler generates different versions
> > where "color1" is replaced by the right hexcolorvalue.

-- 
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-tool...@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.



Compile with different color themes

2010-10-21 Thread Andreas Köberle
Is there a way to get the application compiled with different colors
themes? I want to have a css where I only write:

h1 {color : color1} and the compiler generates different versions
where "color1" is replaced by the right hex color value.

-- 
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-tool...@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.



History broken after using ClosingHandler

2010-10-21 Thread andreas
Hey,

we encountered a strange problem when using a ClosingHandler(GWT
2.0.3) to display the browser's prompt dialog before leaving our
application. We use this as a temporary workaround before having our
history support wired up completely so that users do not lose data by
accidently leaving the application.

The problem was encountered in Chrome. I just submitted a bug for
Chromium here: http://code.google.com/p/chromium/issues/detail?id=60074.

However we noticed that it also did not work on IE(8) so I decided to
post it here as well just in case it is also a problem with the
generated JavaScript. It works in FF (3.x and 4).

Now the problem:
After clicking back and "Stay on this Page" the forward button is
enabled and contains the history entries previously shown for the
backwards history. However the browser remained on the current page.
It seems like it stays on the page but internally manipulates the
navigation history as if it would have navigated back, though it did
not.

Has anyone encountered this before? I'm not sure how much JavaScript
is involved in the functionality (though I doubt it's a lot) but if
necessary I could raise an issue for GWT as well.

Regards,

Andreas

-- 
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-tool...@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: show scrollPanel only when necesseary

2010-08-15 Thread andreas
Let me ask you something first:

Why do you need a ScrollPanel if its content's dimension is set to
"100%"? If I get it right you only want to have vertical scrolling?

Setting width to 100% actually should work fine. BUT: you have to
consider that the vertical scroll bar will consume horizontal space.
So even if the scroll panel's content's width is equal to the scroll
panel's width as soon as the vertical scroll bar appears the
horizontal one will as well...

If you want to have only vertical scrolling your scroll panel's
content width should be 100% minus the width of the vertical scroll
bar. You could hook up a window resize listener and calculate the
width, but I bet there are better solutions. Maybe someone else more
experienced in GWT layout will answer too...

Regards,

Andreas

On 15 Aug., 17:16, asianCoolz  wrote:
> my layout is like below
>
> rootlayoutpanel < -- scrollPanel <--- htmlPanel
>
> but my x and y always show scrollPanel even though my htmlpanel is
> smaller size
>
> i already set   scrollPanel.setAlwaysShowScrollBars(false);
>
> i think this is because i set  width 100% for htmlpanel,  if i set
> width 960px, then it showing properly without x horizontal bar. how to
> properly fix this?

-- 
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-tool...@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: Building two war-files with Maven using different web.xml

2010-08-06 Thread andreas
I don't think you can do that in one build.

But you can use profiles to have two different builds. Of course you
will have to build twice, once for each profile but with that you'll
have the different builds. I use profiles to run hosted_mode target
with fake service servlet implementations and web.xml with appropriate
mapping and no security constraints. The normal build will use the
deploy web.xml with real servlet mapped and security constraints.

For me this works pretty good and you will not have to alter files
(especially pom.xml) for this or that and hence get messy conflicts
with your version system.

Regards,

Andreas

On 6 Aug., 19:02, "dane.molotok"  wrote:
> I've thought about doing that also, in order to to have a war where
> my .gwt.xml inherits the Debug module, and one that does not, but I'm
> beginning to think it's not even buying me much to find out how to do
> it. I'm assuming you have a similar reason for wanting to do this?
>
> On Aug 6, 3:57 am, Stephan T  wrote:
>
>
>
> > I'm using the GWT-plugin for Maven to build my application. I want to
> > build two war-files where the only difference is that I want to use a
> > different web.xml.
>
> > How do I achieve that?
>
> > Here's my pom.xml:
> > 
> > http://maven.apache.org/POM/4.0.0"; 
> > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance";
> >         
> > xsi:schemaLocation="http://maven.apache.org/POM/4.0.0http://maven.apache.org/maven-v4_0_0...";>
> >         4.0.0
>
> >         Web Application
> >         my.company
> >         web
> >         ${web-version}
> >         war
>
> >         
> >                 my.company
> >                 build
> >                 ${version}
> >                 ../build/pom.xml
> >         
>
> >         
> >                 war/WEB-INF/classes
> >                 
> >                         
> >                                 org.codehaus.mojo
> >                                 gwt-maven-plugin
> >                                 
> >                                         
> >                                                 
> >                                                         compile
> >                                                         
> > 
> >                                                         test
> >                                                 
> >                                         
> >                                 
> >                         
> >                         
> >                                 org.apache.maven.plugins
> >                                 maven-war-plugin
> >                                 
> >                                         
> > war
> >                                         war/WEB-INF/web.xml
> >                                 
> >                         
> >                 
> >         
>
> >         
> >                 
> >                         my.company
> >                         env-configuration
> >                         ${env-configuration-version}
> >                         jar
> >                         provided
> >                 
> >                 
> >                         my.company
> >                         common
> >                         ${common-version}
> >                         jar
> >                 
> >                 
> >                         com.google.gwt
> >                         gwt-servlet
> >                 
> >                 
> >                         com.google.gwt
> >                         gwt-user
> >                 
> >                 
> >                         com.google.gwt
> >                         gwt-incubator
> >                 
> >                 
> >                         org.springframework.security
> >                         spring-security-taglibs
> >                 
> >                 
> >                         org.springframework.security
> >                         spring-security-config
> >                 
> >                 
> >                         org.springframework.security
> >                         spring-security-web
> >                 
> >                 
> >                         jstl
> >                         jstl
> >                 
> >                 
> >                         taglibs
> >                         standard
> >                 
> >         
>
> > 

-- 
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-tool...@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: Official tutorial of how to create GWT module?

2010-08-05 Thread andreas
Not sure about an official tutorial...

BUT, just have a look at other GWT modules (look into the GWT jars or
for example gwt-dnd, whatever...):

Modules include:
- the source (IMPORTANT: not only class files, the java files!)
- the module descriptor (*.gwt.xml)
- any static resources you want to include in the modules public path

Simply package that all in a jar and your good to go, that's it.

Regards,

Andreas

On 5 Aug., 09:09, hezjing  wrote:
> Hmmm ... I have developed some GWT applications, and I'm thinking to move
> the common functionalities into modules (or libraries).
> What is the steps to create these common modules, and then shared by the GWT
> applications?
>
>
>
>
>
> On Thu, Aug 5, 2010 at 2:13 PM, rudolf michael  wrote:
> >http://code.google.com/webtoolkit/doc/1.6/tutorial/create.html
>
> > <http://code.google.com/webtoolkit/doc/1.6/tutorial/create.html>if you
> > have the GWT Eclipse plugin installed then all you you need to do is New
> > Google application project, add the project name and package and hit the
> > finish button as simple as that.
>
> > best regards,
> > Rudolf Michael
>
> > On Thu, Aug 5, 2010 at 8:34 AM, hezjing  wrote:
>
> >> Hi
>
> >> I might be missing something, but I couldn't find a tutorial of how to
> >> create a GWT module from the GWT documentation.
>
> >> Is there any tutorial or article from GWT Team?
>
> >> --
>
> >> Hez
>
> >> --
> >> 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-tool...@googlegroups.com.
> >> To unsubscribe from this group, send email to
> >> google-web-toolkit+unsubscr...@googlegroups.com >>  cr...@googlegroups.com>
> >> .
> >> For more options, visit this group at
> >>http://groups.google.com/group/google-web-toolkit?hl=en.
>
> >  --
> > 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-tool...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > google-web-toolkit+unsubscr...@googlegroups.com > cr...@googlegroups.com>
> > .
> > For more options, visit this group at
> >http://groups.google.com/group/google-web-toolkit?hl=en.
>
> --
>
> Hez

-- 
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-tool...@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: DockPanelLayout Displays nothing at all

2010-08-01 Thread andreas
Hey Jose,

I'm not really sure but I remember there were a few messages around
here regarding the *LayoutPanel widgets. Try adding your UI to
RootLayoutPanel instead of RootPanel.

Regards,

Andreas

On 2 Aug., 02:33, Jose Luis Estrella Campaña 
wrote:
> Hello everyone !
>
> I've been developing this application using UIBinders and I'm very
> frustrated because when I use DockPanelLayout as my Layout Panel I
> can't get it to display anything at all. I realize I must be missing
> something. but I don't know what it is...
>
> Here are my code Snippets:
>
> public class GroupsFrontEnd implements EntryPoint {
>         public void onModuleLoad() {
>                 RootPanel.get().add(new Application());
>         }}
>
> --- 
> --- 
> -
> public class Application extends Composite {
>         private static ApplicationUiBinder uiBinder = GWT
>                         .create(ApplicationUiBinder.class);
>         interface ApplicationUiBinder extends UiBinder {
>         }
>         @UiField
>         VerticalPanel content;
>         public Application() {
>                 initWidget(uiBinder.createAndBindUi(this));
>                 this.addWidget(new LoginWidget());
>         }
>         public void addWidget(Widget widget) {
>                 content.clear();
>                 content.add(widget);
>         }}
>
> --- 
> --- 
> -
> public class LoginWidget extends Composite {
>
>         private static LoginWidgetUiBinder uiBinder = GWT
>                         .create(LoginWidgetUiBinder.class);
>
>         interface LoginWidgetUiBinder extends UiBinder {
>         }
>
>         @UiField
>         Label loginLabel;
>         @UiField
>         TextBox loginValueBox;
>         @UiField
>         Label passwordLabel;
>         @UiField
>         TextBox passwordValueBox;
>
>         public LoginWidget() {
>                 initWidget(uiBinder.createAndBindUi(this));
>                 loginLabel.setText("login:");
>                 passwordLabel.setText("password:");
>         }}
>
> --- 
> --- 
> -
> http://dl.google.com/gwt/DTD/xhtml.ent";>
>          xmlns:g="urn:import:com.google.gwt.user.client.ui">
>         
>                 .form {
>                         float: right;
>                         width: 400px;
>                         height: 180px;
>                 }
>         
>
>         
>                 
>                         
>                                 
>                                         This is where I want to display some 
> Text Content
>                                 
>                         
>                 
>                 
>                         
>                                 
>                                         
>                                                 
>                                                          ui:field="loginLabel" />
>                                                 
>                                                 
>                                                          ui:field="loginValueBox" />
>                                                 
>                                         
>                                         
>                                                 
>                                                          ui:field="passwordLabel" />
>                                                 
>                                                 
>                                                          ui:field="passwordValueBox" />
>                                                 
>                                         
>                                 
>                         
>                 
>         
> 
> --- 
> --- 
> -
>
> I hope somebody could tell me What is wrong. I think it may be
> something really obvious, But I just don't see it, please help... the
> browser is blank when I run this, but throws No error.
>
> Thanks in advance,
>
> Best Regards,
>
> Jose.

-- 
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-tool...@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: Dev plugin for firefox 3.7

2010-07-21 Thread andreas
I'm having the same problem ... just built development plugin from
trunk but FF 4.0 still displays its "Plugin required..." stuff.

Working on Ubuntu with actually only newest versions of browsers
(Chrome dev channel and FF 3.6 & 4.0). At least for FF 3.6 a working
development plugin exists. Hope this will change soon...

Does anyone have a working setup for this combination?:
- Ubuntu
- GWT Development Mode
- Browser supporting Websockets

What about other OS?

I had it running with FF 4.0 and development plugin build from trunk
like two weeks ago but apparently it does not work anymore.

Regards,

Andreas

On 7 Jul., 23:15, 01fetischist <01fetisch...@googlemail.com> wrote:
> Doesn't work withFirefox4.0 beta1 andFirefox4.0 beta2pre
> (nightlybuild).
> Tested with gecko1.9.3 plugin-sdk from google and with self build
> gecko2.0b2pre sdk.
> I still get the "Development Mode requires the Google Web Toolkit
> Developer Plugin" page.
>
> Any hints?

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-07-21 Thread andreas
Hey Magnus,

to be honest I was very glad to read this post:

http://googleenterprise.blogspot.com/2010/01/modern-browsers-for-modern-applications.html

Besides, what exactly is looking ugly?

Me personally, I would not spend much time in fixing it for IE6 if it
can not be fixed by slightly adjusting the UI setup in such a way that
it still works correctly on modern browsers.

Regards,

Andreas

On 21 Jul., 19:00, Magnus  wrote:
> Hi Andreas,
>
> may I come back to your code? I have build my chess game on your idea
> using SimplePanels for the cells. It always worked great, but today I
> found that it looks ugly under IE6. What would you do in this case? Is
> IE6 so old that you can ignore it?
>
> Thanks
> Magnus
>
> On Jun 28, 12:24 pm, andreas  wrote:
>
>
>
> > Here's code I'd use to create a simplistic chess board:
>
> > // the chess board with no spaces
> > Grid cb = new Grid(10, 10);
> > cb.setCellPadding(0);
> > cb.setCellSpacing(0);
> > cb.setBorderWidth(0);
>
> > // assembles the board by inserting colored panels
> > for (int i = 1; i < 9; i++) {
> >     // panels of the top row
> >     SimplePanel pHTop = new SimplePanel();
> >     pHTop.setPixelSize(40, 20);
> >     pHTop.getElement().getStyle().setBackgroundColor("red");
>
> >     // panels of the bottom row
> >     SimplePanel pHBottom = new SimplePanel();
> >     pHBottom.setPixelSize(40, 20);
> >     pHBottom.getElement().getStyle().setBackgroundColor("red");
>
> >     // panels of the left column
> >     SimplePanel pVLeft = new SimplePanel();
> >     pVLeft.setPixelSize(20, 40);
> >     pVLeft.getElement().getStyle().setBackgroundColor("green");
>
> >     // panels of the right column
> >     SimplePanel pVRight = new SimplePanel();
> >     pVRight.setPixelSize(20, 40);
> >     pVRight.getElement().getStyle().setBackgroundColor("green");
>
> >     // insert the border cells
> >     cb.setWidget(0, i, pHTop);
> >     cb.setWidget(9, i, pHBottom);
> >     cb.setWidget(i, 0, pVLeft);
> >     cb.setWidget(i, 9, pVRight);
>
> >     for (int j = 1; j < 9; j++) {
> >         // the inner chess board panels
> >         SimplePanel cP = new SimplePanel();
> >         cP.setPixelSize(40, 40);
> >         // switches between black and white
> >         if (j % 2 == 0) {
> >             cP.getElement().getStyle().setBackgroundColor(
> >                             i % 2 == 0 ? "black" : "white");
> >         } else {
> >             cP.getElement().getStyle().setBackgroundColor(
> >                             i % 2 == 0 ? "white" : "black");
> >         }
> >         cb.setWidget(i, j, cP);
> >     }
>
> > }
>
> > // there it is
> > RootPanel.get().add(cb, 1, 1);
>
> > Programmatic styles are of course not as good as using stylesheets,
> > consider this just a demo.
>
> > On 28 Jun., 11:51, andreas  wrote:
>
> > > Where exactly are the vertical spaces? From what I see, there are no
> > > spaces between the cells of the top and bottom row and the spaces
> > > between the cells in the left and right column are of the same color
> > > as the image background, so I assume there are actually also no
> > > spaces, correct me on this one?
>
> > > For better debug you could also assign a border and background color
> > > to the Grid. If none of these colors will be visible you can be sure
> > > that there are no spaces or anything left.
>
> > > For the inner cells I can not say if there are any spaces.
>
> > > Also I assume your style "pnl-r" adds the red border to the Grid? If
> > > there were any spaces left caused by the border width you would see a
> > > red Grid.
>
> > > I think you got what you wanted...
>
> > > BTW: I think you do not need to set the cell dimensions manually; Grid
> > > will automatically adjust cell dimensions so that the cell widgets
> > > fit, in other words a columns width for example will be adjusted so
> > > that the cell widget with the biggest width is completely visible;
> > > same goes for rows and heights and so on
>
> > > BTW2: the two for-blocks in init() can be realized in one single for-
> > > block since they iterate over exactly the same interval (0-9)
>
> > > On 28 Jun., 11:27, Magnus  wrote:
>
> > > > Here is the screenshot:
>
> > > >http://yfrog.com/j7chessboardj
>
> > > > Magnus

-- 
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-tool...@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 client as Maven module in multi-module project

2010-07-16 Thread andreas
Good to hear...

I think the output .class files have nothing to do with what you see
in the browser. What you see in action there is JavaScript. What
Hosted Mode does (afaik) is take the Java code, not the .class files,
and sort of just in time compiles them into JavaScript, but I'm not an
expert on this. That's why you can use this cool refresh feature and I
tell you, we love it, too!

Andreas

On 16 Jul., 20:27, David Vree  wrote:
> Andreas -- wanted to let you know it works perfectly...including
> debugging.  However, I am totally mystified about how it is that I can
> change a file in eclipse and hit refresh in firefox and see it work!
> How can this be?
>
> When I make a change to a java file in Eclipse, it compiles to
> myproject/build/classes -- how then is that the hosted mode, running
> off an unchanged build project war directory somewhere else "knows"
> about this -- and dynamically incorporates it at runtime.  It works,
> but it seems like magic!
>
> thx,
> Dave
>
> On Jul 16, 10:03 am, David Vree  wrote:
>
>
>
> > Andreas -- thank you very much for the HUGE help...been struggling
> > with all this for almost 4 days now!
>
> > On Jul 16, 4:51 am, andreas  wrote:
>
> > > Hi David,
>
> > > I did not try using the latest GPE with the changes/fixes you
> > > mentioned. I remember that configuration of the "war" directory was
> > > introduced, however like you say it wasn't enough and I still was not
> > > able to get it run. I wouldn't want to use anything in src/ as an
> > > output directory as well (nor will our SVN do) and I don't know how to
> > > make Eclipse/GPE properly create/fill WEB-INF/lib... for us it's
> > > simply maven who does it all.
>
> > > Launching the maven build configuration takes the same time as
> > > launching Hosted Mode via GPE from what I know. Anyway the launch time
> > > is not that important to me because the maven build mentioned above
> > > supports code change reflections upon refreshing the browser, so
> > > regarding client side code changes you can use one launch for several
> > > edit cycles. Even server side code changes can be reflected upon
> > > restarting the Hosted Mode server without relaunching the build. Only
> > > if you change/add static resources you have to terminate the build and
> > > launch it again so that the resource plugin does its job.
>
> > > Actually I did not make a lot of use of the Eclipse debugger. However
> > > your question made me curious and a quick search gave me this:
>
> > >http://claudiushauptmann.com/tutorial-gwt-maven-and-eclipse-with-m2ec...
>
> > > After 10 minutes of reading and applying I was able to use Debug
> > > perspective and halt our GWT application with breakpoints. It works
> > > just like debugging a normal Java application... pretty cool.
>
> > > Regards,
>
> > > Andreas
>
> > > On 16 Jul., 09:15, David Vree  wrote:
>
> > > > Thanks Andreas -- it makes a lot of sense to me.  A couple of
> > > > thoughts:
>
> > > > In the 1.3.3 version of GPE they fixed a few thing:
> > > >    1) For what its worth, GPE now has the ability to point to any
> > > > directory for the "WAR"...I have it pointing to /src/main/webapp.
> > > >    2) You can disable GPE from complaining when the SDK isn't in the
> > > > class path.  However, you still need to have the SDK lying around
> > > > somewhere.
>
> > > > These things help, but is not enough I think -- because it also
> > > > requires Eclipse to use "src/main/webapp/WEB-INF/classes" as the
> > > > output directory for all sources in the project.  This seems odd to
> > > > me.  Also I don't understand how GPE or Eclipse is to put stuff into /
> > > > WEB-INF/lib...do you?
>
> > > > On last question if you don't mind:  Can you comment on your edit/
> > > > build/debug cycle?  How long does it take for the "clean compile
> > > > war:exploded gwt:run" to execute so you can being another cycle and
> > > > how do you use the Eclipse debugger while running the web app?
>
> > > > On Jul 16, 1:57 am, andreas  wrote:
>
> > > > > Hey David,
>
> > > > > I was trying to get our newly introduced maven based build cycle to
> > > > > work with GPE a few months ago. At that time GPE had some issues
> > > > > regarding flexibility in configuration to work with maven

Re: GWT client as Maven module in multi-module project

2010-07-16 Thread andreas
Hi David,

I did not try using the latest GPE with the changes/fixes you
mentioned. I remember that configuration of the "war" directory was
introduced, however like you say it wasn't enough and I still was not
able to get it run. I wouldn't want to use anything in src/ as an
output directory as well (nor will our SVN do) and I don't know how to
make Eclipse/GPE properly create/fill WEB-INF/lib... for us it's
simply maven who does it all.

Launching the maven build configuration takes the same time as
launching Hosted Mode via GPE from what I know. Anyway the launch time
is not that important to me because the maven build mentioned above
supports code change reflections upon refreshing the browser, so
regarding client side code changes you can use one launch for several
edit cycles. Even server side code changes can be reflected upon
restarting the Hosted Mode server without relaunching the build. Only
if you change/add static resources you have to terminate the build and
launch it again so that the resource plugin does its job.

Actually I did not make a lot of use of the Eclipse debugger. However
your question made me curious and a quick search gave me this:

http://claudiushauptmann.com/tutorial-gwt-maven-and-eclipse-with-m2eclipse.html

After 10 minutes of reading and applying I was able to use Debug
perspective and halt our GWT application with breakpoints. It works
just like debugging a normal Java application... pretty cool.

Regards,

Andreas

On 16 Jul., 09:15, David Vree  wrote:
> Thanks Andreas -- it makes a lot of sense to me.  A couple of
> thoughts:
>
> In the 1.3.3 version of GPE they fixed a few thing:
>    1) For what its worth, GPE now has the ability to point to any
> directory for the "WAR"...I have it pointing to /src/main/webapp.
>    2) You can disable GPE from complaining when the SDK isn't in the
> class path.  However, you still need to have the SDK lying around
> somewhere.
>
> These things help, but is not enough I think -- because it also
> requires Eclipse to use "src/main/webapp/WEB-INF/classes" as the
> output directory for all sources in the project.  This seems odd to
> me.  Also I don't understand how GPE or Eclipse is to put stuff into /
> WEB-INF/lib...do you?
>
> On last question if you don't mind:  Can you comment on your edit/
> build/debug cycle?  How long does it take for the "clean compile
> war:exploded gwt:run" to execute so you can being another cycle and
> how do you use the Eclipse debugger while running the web app?
>
> On Jul 16, 1:57 am, andreas  wrote:
>
>
>
> > Hey David,
>
> > I was trying to get our newly introduced maven based build cycle to
> > work with GPE a few months ago. At that time GPE had some issues
> > regarding flexibility in configuration to work with maven-gwt project
> > layout (in particular no "war/" directory, which GPE was expecting).
>
> > After all our solution was to drop usage of GPE and use only maven
> > together with maven gwt plugin and maven war plugin. We run our
> > application in Hosted Mode using a maven build configuration. We
> > configured our project to host static resources in src/main/webapp and
> > src/main/resources and use the project build directory for the hosted
> > web application. With this configuration which is basically default
> > maven(-gwt) project layout we run Hosted Mode with these goals in
> > maven build configuration: clean compile war:exploded gwt:run. That's
> > all.
>
> > There is no need to mix maven goals with launching of GPE or similar.
> > As I said in the beginning I was trying to get GPE run with maven but
> > that was actually because I did not know what maven and maven gwt
> > plugin can allready do for you. You might want to give this a try...
>
> > Andreas
>
> > On 16 Jul., 06:29, David Vree  wrote:
>
> > > Thanks -- makes sense, although I hate the idea of having to do a "mvn
> > > package" everytime before running in host mode.  If I punt on the
> > > maven directory structure and go with the "war" directory can I
> > > shorten my edit-debug cycle?  What do most maven users do?
>
> > > On Jul 15, 6:25 pm, Daniel  wrote:
>
> > > > The GWT Maven plugin deviates from the standard Maven directory
> > > > structure by default, to accommodate the Google Plugin for Eclipse's
> > > > default directory structure. If you want to use the standard Maven
> > > > directory layout (with the static resources for your War file in src/
> > > > main/webapp instead of the war directory) with the Google Plugin for
> > > > Eclipse, there are some things yo

Re: GWT client as Maven module in multi-module project

2010-07-15 Thread andreas
Hey David,

I was trying to get our newly introduced maven based build cycle to
work with GPE a few months ago. At that time GPE had some issues
regarding flexibility in configuration to work with maven-gwt project
layout (in particular no "war/" directory, which GPE was expecting).

After all our solution was to drop usage of GPE and use only maven
together with maven gwt plugin and maven war plugin. We run our
application in Hosted Mode using a maven build configuration. We
configured our project to host static resources in src/main/webapp and
src/main/resources and use the project build directory for the hosted
web application. With this configuration which is basically default
maven(-gwt) project layout we run Hosted Mode with these goals in
maven build configuration: clean compile war:exploded gwt:run. That's
all.

There is no need to mix maven goals with launching of GPE or similar.
As I said in the beginning I was trying to get GPE run with maven but
that was actually because I did not know what maven and maven gwt
plugin can allready do for you. You might want to give this a try...

Andreas

On 16 Jul., 06:29, David Vree  wrote:
> Thanks -- makes sense, although I hate the idea of having to do a "mvn
> package" everytime before running in host mode.  If I punt on the
> maven directory structure and go with the "war" directory can I
> shorten my edit-debug cycle?  What do most maven users do?
>
> On Jul 15, 6:25 pm, Daniel  wrote:
>
>
>
> > The GWT Maven plugin deviates from the standard Maven directory
> > structure by default, to accommodate the Google Plugin for Eclipse's
> > default directory structure. If you want to use the standard Maven
> > directory layout (with the static resources for your War file in src/
> > main/webapp instead of the war directory) with the Google Plugin for
> > Eclipse, there are some things you need to make sure of.
>
> > 1. In the GWT Maven Plugin , add $
> > {project.build.directory}/${project.build.finalName}.
> > That will cause the plugin to use target/my-example-project-1.0.0-
> > SNAPSHOT (or whatever your project is called) instead of the war
> > directory.
> > 2. Configure the Google Plugin for Eclipse to use src/main/webapp
> > instead of war
> > 3. Before you can run the project in hosted mode, you'll need to run
> > mvn package, to copy your static resources from your GWT public
> > package and src/main/webapp to your hosted mode directory. You'll only
> > need to do this the first time.
>
> > On Jul 16, 5:17 am, David Vree  wrote:
>
> > > The documentation is very complex, but ultimately it provided the
> > > answer.  I needed to configure the maven-war-plugin to filter (e.g.
> > > copy) the files from my webapp directory to the war directory.  I
> > > accomplished this via the following snippet in my module level POM:
>
> > >                         
> > >                                 
> > > org.apache.maven.plugins
> > >                                 maven-war-plugin
> > >                                 
> > >                                         
> > > war
> > >                                 
> > >                         
>
> > > Thanks for the help.  The debugging stop points don't work, but I'll
> > > start a new thread on that
>
> > > On Jul 15, 11:19 am, SalvadorDiaz  wrote:
>
> > > > Hi,
>
> > > > You might want to take a look at the GWT maven plugin documentation
> > > > (there are lots of useful tips):
>
> > > >http://mojo.codehaus.org/gwt-maven-plugin/
>
> > > > Hope that helps,
>
> > > > Salvador
>
> > > > On 15 juil, 03:35, David Vree  wrote:
>
> > > > > Manually moving index.html to the WEB-INF directory solver the 404
> > > > > problem. But there is still something wrong.
>
> > > > > The pop-up window I get with the regular GWT application doesn't pop-
> > > > > up in my application.  And the debugging stop point I added for
> > > > > onModuleLoad doesn't catch.
>
> > > > > On Jul 14, 6:40 pm, Thomas Broyer  wrote:
>
> > > > > > On 14 juil, 19:25, David Vree  wrote:
>
> > > > > > > Total GWT newbie here trying to get the first app up and running. 
> > > > > > >  I'm
> > > > > > > using Eclipse 3.5.2 and have created and run the sample app you 
> > > > > > > get
> > > > > > > with New->Google->Web Application.  The app a

Re: This is getting beyond a joke

2010-07-15 Thread andreas
Thanks David!

I'll spread the news in our team...

Andreas

On 15 Jul., 04:18, David Chandler  wrote:
> Hang tight, folks, we're on it. As has been noted in this thread, GWT
> itself is not built with maven, so it's not a 100% automated process
> to push to maven central. But we're working hard to get gwt-user-2.0.4
> and gwt-dev-2.0.4 uploaded by Friday and to make deployment to maven
> central a part of future releases.
>
> David Chandler
> GWT Developer Relations
> Atlanta, GA USA
>
> On Jul 14, 9:15 pm, David Vree  wrote:
>
>
>
> > Thanks for that.
>
> > On Jul 14, 6:37 pm, monkeyboy  wrote:
>
> > > Nice. From 23 to 37 stars in less than a day.
>
> > > On Jul 14, 9:09 am, monkeyboy  wrote:
>
> > > > On Wed, Jul 14, 2010 at 7:22 AM, Kasper Hansen 
> > > > wrote:
>
> > > > > I would also wish that Google would takeMavenmore seriously. I can
> > > > > only recommend all that agree to star the issues regardingMavenin
> > > > > the bug tracker.
>
> > > > Here it is (the issue 
> > > > regardingMaven):http://code.google.com/p/google-web-toolkit/issues/detail?id=4673
>
> > > > There are only 23 stars so far.- 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-tool...@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: Client side mapping between model objects

2010-07-13 Thread andreas
Maybe I was not clear about it:

1) We do not "map anything together" from the model at client side and
the best is: we do not need to; our (GWT) RPC returns one instance of
our domain model containing all references among elements like the way
we designed them, it's just that easy... the referencing between the
elements is persisted and loaded by hibernate, just like the elements
are

2) We don't do that... each model object is represented by its own
view object

Regarding your "few points":

1) Why is it different? It's simply model based...

Andreas

On 13 Jul., 20:44, munna kaka  wrote:
> Thanks A.
> 2 more selfish questions
> 1) In your case after simply loading various model objects via RPC do you
> again map them togather ?
> 2) If one of model object is used (read-only) by two different MVP widgets
> than how do you share that model object.
>
> Few points
> 1) I understand that your proj domain is different from anybody else
> 2) I do have coarse grained GWT RPC calls but I need to make fine grained
> calls now
>
> On Tue, Jul 13, 2010 at 12:11 PM, andreas 
> wrote:
>
>
>
> > Are you sure you mean 'client side mapping'?
>
> > What I understand from it is you have a domain model with classes
> > (customer, order, item) which somehow reference each other.
>
> > How do you retrieve instances of these classes from your application
> > backend? Are instances of these classes referencing each other at all,
> > for example before being persisted or after being retrieved from a
> > database?
>
> > If you use hibernate you can persist the associations between domain
> > model classes as well (see hibernate documentation) and of course load
> > it again as well. Now if you break up these associations and retrieve
> > each classes instances separately I'd suggest your associations are
> > lost, but why would you do something like that?
>
> > Get your model instance out of your database and pass it to the client
> > at once. For example request one CUSTOMER which references some ORDERs
> > which reference some ITEMs. GWT RPC will not break up these
> > references.
>
> > If you have a very large database things are getting different of
> > course because you do not want to load the whole database because all
> > entries references cover the whole database. Then you'll want to
> > consider other fetching strategies.
>
> > We are working with (quite) complex domain models containing
> > hierarchies and references as well. In our case we will not load the
> > whole database by accident and simply loading via RPC works great.
>
> > Hope this help...
>
> > Andreas
>
> > On 13 Jul., 19:55, mk  wrote:
> > > How do you maintain mapping between Model objects at client browser?
>
> > > Say for example, over the course of user conversation, there were
> > > three DIFFERENT ajax calls to load CUSTOMER, ORDERS and ITEMS.
>
> > > Now do you manually map CUSTOMER to ORDER and ORDER to ITEM in client
> > > to maintain mapping between Model objects..
> > > ( or do you store CUSTOMER, ORDER, ITEM  separatly with no mapping or
> > > there is a framework like hibernate mappings but for browser which
> > > maps model objects with configuration)
>
> > > 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-tool...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > google-web-toolkit+unsubscr...@googlegroups.com > cr...@googlegroups.com>
> > .
> > For more options, visit this group at
> >http://groups.google.com/group/google-web-toolkit?hl=en.

-- 
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-tool...@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: Client side mapping between model objects

2010-07-13 Thread andreas
Are you sure you mean 'client side mapping'?

What I understand from it is you have a domain model with classes
(customer, order, item) which somehow reference each other.

How do you retrieve instances of these classes from your application
backend? Are instances of these classes referencing each other at all,
for example before being persisted or after being retrieved from a
database?

If you use hibernate you can persist the associations between domain
model classes as well (see hibernate documentation) and of course load
it again as well. Now if you break up these associations and retrieve
each classes instances separately I'd suggest your associations are
lost, but why would you do something like that?

Get your model instance out of your database and pass it to the client
at once. For example request one CUSTOMER which references some ORDERs
which reference some ITEMs. GWT RPC will not break up these
references.

If you have a very large database things are getting different of
course because you do not want to load the whole database because all
entries references cover the whole database. Then you'll want to
consider other fetching strategies.

We are working with (quite) complex domain models containing
hierarchies and references as well. In our case we will not load the
whole database by accident and simply loading via RPC works great.

Hope this help...

Andreas

On 13 Jul., 19:55, mk  wrote:
> How do you maintain mapping between Model objects at client browser?
>
> Say for example, over the course of user conversation, there were
> three DIFFERENT ajax calls to load CUSTOMER, ORDERS and ITEMS.
>
> Now do you manually map CUSTOMER to ORDER and ORDER to ITEM in client
> to maintain mapping between Model objects..
> ( or do you store CUSTOMER, ORDER, ITEM  separatly with no mapping or
> there is a framework like hibernate mappings but for browser which
> maps model objects with configuration)
>
> 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-tool...@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: StringBuffer vs StringBuilder in GWT

2010-07-13 Thread Andreas Karlsson
Manolo,

but if I understood you correctly you said there was a client side
performance penalty. If it's only in compile time I think it's easier
to let devs use stringbuilder in JS mode as well to be consistent.

/Andreas

On Tue, Jul 13, 2010 at 10:59 AM, Manuel Carrasco Moñino
 wrote:
> Yeah, Gwt should optimize them in compiler time, but at the end the
> optimization will produce arithmetic operations with javascript String
> implementation, and this optimization will penalize the time spent to
> compile.
> So why use StringBuffer instead of String unless this code was shared
> in both sides (js/jre).
>
> -Manolo
>
> On Tue, Jul 13, 2010 at 10:44 AM, Andreas Karlsson  wrote:
>> Is this really true? Shouldn't GWT optimize any overhead away and make
>> it similar to using String directly?
>>
>> /Andreas
>>
>> On Tue, Jul 13, 2010 at 10:41 AM, Manuel Carrasco Moñino
>>  wrote:
>>> I think the use of either will penalize the performance in client
>>> side, String should be faster.
>>>
>>> - Manolo
>>>
>>>
>>>
>>> On Tue, Jul 13, 2010 at 10:21 AM, guandalino  wrote:
>>>> Hi, GWT provides JRE emulation for both StringBuffer and
>>>> StringBuilder. The Java API says that in single threaded environments
>>>> the preferred choice is to use StringBuilder as it is faster. I also
>>>> remember to have read that browsers way to work is single threaded.
>>>>
>>>> So I'm wondering why and when one should use StringBuffer at all. Can
>>>> you clarify?
>>>>
>>>> Thank you.
>>>>
>>>> --
>>>> 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-tool...@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.
>>>>
>>>>
>>>
>>> --
>>> 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-tool...@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.
>>>
>>>
>>
>> --
>> 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-tool...@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.
>>
>>
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: StringBuffer vs StringBuilder in GWT

2010-07-13 Thread Andreas Karlsson
Is this really true? Shouldn't GWT optimize any overhead away and make
it similar to using String directly?

/Andreas

On Tue, Jul 13, 2010 at 10:41 AM, Manuel Carrasco Moñino
 wrote:
> I think the use of either will penalize the performance in client
> side, String should be faster.
>
> - Manolo
>
>
>
> On Tue, Jul 13, 2010 at 10:21 AM, guandalino  wrote:
>> Hi, GWT provides JRE emulation for both StringBuffer and
>> StringBuilder. The Java API says that in single threaded environments
>> the preferred choice is to use StringBuilder as it is faster. I also
>> remember to have read that browsers way to work is single threaded.
>>
>> So I'm wondering why and when one should use StringBuffer at all. Can
>> you clarify?
>>
>> Thank you.
>>
>> --
>> 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-tool...@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.
>>
>>
>
> --
> 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-tool...@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.
>
>

-- 
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-tool...@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: Serialization exception

2010-07-08 Thread andreas
Another idea from looking over your exception again:

Since you are loading an object with references to other objects you
also would want to make sure that hibernate really loads the
references. I think by default it uses lazy fetching and hence
initializes the references with proxy objects. These proxy objects
again are not serializable by GWT (by itself). You can deactivate lazy
fetching in the hibernate mapping file:



This makes hibernate really load and instantiate a Category object and
not a proxy.

Andreas

On 8 Jul., 12:13, andreas  wrote:
> I had a similar problem once.
>
> You could sysou the rtnAryList.class to make sure it really is an
> instance of ArrayList and not a collection of hibernate. Although you
> instantiate an ArrayList the parsed call to list() may return
> something different. If it is one of hibernates collections (which it
> uses for lazy fetching and so on afaik) you could "manually" add all
> entries in an instance of ArrayList.
>
> The hibernate collections are not serializable by GWT by default. I
> think there is also at least one project out there for integrating
> hibernate to GWT which also avoids this problem, you might want to
> give it a quick search.
>
> Andreas
>
> On 8 Jul., 11:56, Ho Jimmy  wrote:
>
>
>
> > Hi,
>
> > I keep getting the serialization exception and causing the rpc failed. Can
> > someone tell me what's going on?
>
> >  com.google.gwt.user.client.rpc.SerializationException: Type
> > 'com.jobscout.frontpage.client.Category$$EnhancerByCGLIB$$424f9bbb' was not
> > included in the set of types which can be serialized by this
> > SerializationPolicy or its Class object could not be loaded. For security
> > purposes, this type will not be serialized.: instance =
> > com.jobscout.frontpage.client.categ...@9cdbbc
>
> > I use hibernate to extract the records, put the result in the ArrayList. To
> > test the serialization, I dropped the hibernate part and built the ArrayList
> > of the type in the Impl and it works. So, I am not sure how the hibernate
> > return the arraylist causes the serialization exception.
>
> > The following are extracted from my code that is relevent.
>
> > public class Job implements Serializable {
> >  private long oid;
> >  private long acctOid;
> >  private String subject;
> >  private String description;
> >  private Category category;            //Serializable
> >  private SubCat1 subcat1;            //Serializable
> >  private Mesgtype mesgtype;            //Serializable
>
> > This part is the testing and it works. I create the Arraylist instead of
> > using hibernate to extract records and return an Arraylist.
>
> >  public ArrayList getJobAryLstTest(){
> >   ArrayList jobarylst = new ArrayList();
> >   Category catparttime = new Category(0,"Parttime");
> >   SubCat1 painter = new SubCat1(0, catparttime,"Partime Painter");
> >   Mesgtype postjobmesgtype = new Mesgtype(0, "Post Job");
> >   Job job1 = new Job(0,"subject1","descruotuib1",catparttime,
> > painter,postjobmesgtype);
> >   Job job2 = new Job(0,"subject2","descruotuib2",catparttime,
> > painter,postjobmesgtype);
> >   jobarylst.add(job1);
> >   jobarylst.add(job2);
> >   return jobarylst;
>
> > This part call the hibernate dao and cause serialization exception
> >  public ArrayList getAllJobAryLst() {
> >   JobDAO jobdao=new JobDAO();
> >   ArrayList rtnJobList;
> >   rtnJobList = jobdao.getAllJob();
> >   return rtnJobList;
> >  }
>
> > This part is the DAO
> >  public ArrayList getAllJob()
> >  {
> >   ArrayList rtnAryList=new ArrayList();
> >   try
> >   {
> >    if(!session.isOpen())
> >    {
> >     session = HibernateUtil.getSessionFactory().openSession();
> >    }
> >    session.beginTransaction();
> >    String hql="from Job job";
> >    rtnAryList = (ArrayList)session.createQuery(hql).list();
> >    return rtnAryList;
> >   }
> >   catch(Exception e)
> >   {
> >    System.out.println(e.toString());
> >    return null;
> >   }
> >  }
>
> > Thanks
> > Jimmy

-- 
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-tool...@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: Serialization exception

2010-07-08 Thread andreas
I had a similar problem once.

You could sysou the rtnAryList.class to make sure it really is an
instance of ArrayList and not a collection of hibernate. Although you
instantiate an ArrayList the parsed call to list() may return
something different. If it is one of hibernates collections (which it
uses for lazy fetching and so on afaik) you could "manually" add all
entries in an instance of ArrayList.

The hibernate collections are not serializable by GWT by default. I
think there is also at least one project out there for integrating
hibernate to GWT which also avoids this problem, you might want to
give it a quick search.

Andreas

On 8 Jul., 11:56, Ho Jimmy  wrote:
> Hi,
>
> I keep getting the serialization exception and causing the rpc failed. Can
> someone tell me what's going on?
>
>  com.google.gwt.user.client.rpc.SerializationException: Type
> 'com.jobscout.frontpage.client.Category$$EnhancerByCGLIB$$424f9bbb' was not
> included in the set of types which can be serialized by this
> SerializationPolicy or its Class object could not be loaded. For security
> purposes, this type will not be serialized.: instance =
> com.jobscout.frontpage.client.categ...@9cdbbc
>
> I use hibernate to extract the records, put the result in the ArrayList. To
> test the serialization, I dropped the hibernate part and built the ArrayList
> of the type in the Impl and it works. So, I am not sure how the hibernate
> return the arraylist causes the serialization exception.
>
> The following are extracted from my code that is relevent.
>
> public class Job implements Serializable {
>  private long oid;
>  private long acctOid;
>  private String subject;
>  private String description;
>  private Category category;            //Serializable
>  private SubCat1 subcat1;            //Serializable
>  private Mesgtype mesgtype;            //Serializable
>
> This part is the testing and it works. I create the Arraylist instead of
> using hibernate to extract records and return an Arraylist.
>
>  public ArrayList getJobAryLstTest(){
>   ArrayList jobarylst = new ArrayList();
>   Category catparttime = new Category(0,"Parttime");
>   SubCat1 painter = new SubCat1(0, catparttime,"Partime Painter");
>   Mesgtype postjobmesgtype = new Mesgtype(0, "Post Job");
>   Job job1 = new Job(0,"subject1","descruotuib1",catparttime,
> painter,postjobmesgtype);
>   Job job2 = new Job(0,"subject2","descruotuib2",catparttime,
> painter,postjobmesgtype);
>   jobarylst.add(job1);
>   jobarylst.add(job2);
>   return jobarylst;
>
> This part call the hibernate dao and cause serialization exception
>  public ArrayList getAllJobAryLst() {
>   JobDAO jobdao=new JobDAO();
>   ArrayList rtnJobList;
>   rtnJobList = jobdao.getAllJob();
>   return rtnJobList;
>  }
>
> This part is the DAO
>  public ArrayList getAllJob()
>  {
>   ArrayList rtnAryList=new ArrayList();
>   try
>   {
>    if(!session.isOpen())
>    {
>     session = HibernateUtil.getSessionFactory().openSession();
>    }
>    session.beginTransaction();
>    String hql="from Job job";
>    rtnAryList = (ArrayList)session.createQuery(hql).list();
>    return rtnAryList;
>   }
>   catch(Exception e)
>   {
>    System.out.println(e.toString());
>    return null;
>   }
>  }
>
> Thanks
> Jimmy

-- 
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-tool...@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: the button and its infoLabel problem...

2010-07-07 Thread andreas
Hey Shelley,

a bit of further information is missing but may it be that you issue
an RPC between infoLabel.setText("Beginning"); and
infoLabel.setText("Completed");?

If so you simply forgot the async nature of GWT RPC. Your code in
onClick(...) will not sort of stop or wait until the RPC finishes and
continue afterwards. That's why infoLabel.setText("Completed");
will be immediately executed after infoLabel.setText("Beginning");
and you're simply "way to slow" to see it.

What you could do instead is call infoLabel.setText("Completed");
in your onSuccess(...) in case I'm right with RPC or in the callback
of whatever request you do for your search operation.

Andreas

On 7 Jul., 10:22, Shelley  wrote:
> hello all:
>    i came across a strange problem which seems quite simple:
>
>    i have a button and a label on a panel, the button has been
> registered a listener:
>
>   Button searchButton = new Button( "Search" );
>         searchButton.addClickHandler( new ClickHandler()
>         {
>             @Override
>             public void onClick( ClickEvent event )
>             {
>                 infoLabel.setText("Beginning");
>                 //do a search operation here which will take more than
> 5 seconds...
>                 infoLabel.setText("Completed...");
>             }
>         } );
>
> that's all, but i never see the "Beginning“ on the label, but only
> see the "completed...", seem the label will not refresh it's text
> until onClick is  finished? how can i achieved the function that
> display "beginning..." first and "Completed..." when the operation
> completed?
>
> thanks in advance.
>
> -Shelley

-- 
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-tool...@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: onSuccess() return value other than void?

2010-07-07 Thread andreas
http://lmgtfy.com/?q=gwt+module+xml+source

http://www.gwtapps.com/doc/html/com.google.gwt.doc.DeveloperGuide.Fundamentals.Modules.ModuleXml.html
http://www.gwtapps.com/doc/html/com.google.gwt.doc.DeveloperGuide.Fundamentals.Modules.html

It's a subpackage of the package your module is located in (your
*.gwt.xml).

Andreas

On 7 Jul., 11:44, day_trader  wrote:
> Does not appear to be working even with the ,
> assuming I'm giving the correct path. Is the path of the form of a
> directory or package?
>
> On Jul 7, 9:51 am, andreas  wrote:
>
>
>
> > The inherits tag is for inheriting GWT modules not for inheriting Java
> > files or similar (afaik).
>
> > You add packages to the GWT translatable sources via source tag:
>
> > 
>
> > This should be enough assuming your GWT module is in 'com.myApp'.
>
> > Andreas
>
> > On 7 Jul., 10:44, day_trader  wrote:
>
> > > It could possibly be because I am using Enunciate to develop my
> > > application with. Here is my complete error:
>
> > > [INFO]
> > > 
> > > [ERROR] BUILD ERROR
> > > [INFO]
> > > 
> > > [INFO] Problem assembling the enunciate app.
>
> > > Embedded error: Class not found: TreeModel
>
> > > I've just tried putting TreeModel into my gwt.xml file :  > > name='com.myApp.client.models.TreeModel'/> however this doesn't appear
> > > to work either. I have posted this to the Enunciate Mailing List as
> > > well, so hopefully one of those guys can enlighten me if it's a
> > > probably with Enunciate and not GWT/GXT.
>
> > > Malcolm
>
> > > On Jul 7, 9:28 am, andreas  wrote:
>
> > > > Sorry, it's war/WEB-INF/classes/.
>
> > > > On 7 Jul., 10:26, andreas  wrote:
>
> > > > > You do not have to move something out of 'client' packages to have it
> > > > > on your server. This is definitely wrong!
>
> > > > > You do have to move classes you want to have on the client or both
> > > > > client and server into the client packages (or declare those packages
> > > > > in your *.gwt.xml files).
>
> > > > > Since you want your class to be passed from the client to the server
> > > > > this class definitely needs to be in your 'client' package. I don't
> > > > > know why you get an error on compiling, maybe you can give more
> > > > > informations on this one. GWT compiles all Java files (also those in
> > > > > 'client') to .class files in your war/WEB-INF/src/ folder so all
> > > > > classes will be accessible on the server.
>
> > > > > Andreas
>
> > > > > On 7 Jul., 10:02, day_trader  wrote:
>
> > > > > > This is where I am rather confused. As when i move it to
> > > > > > a package which isn't named in the .gwt.xml file I receive
> > > > > > an error of "cannot be resolved to a type". Which is fair enough,
> > > > > > as I assumed everything in the .gwt.xml was putting the classes
> > > > > > in 'view' of everything in the application? Or perhaps I'm very
> > > > > > muddled up!
>
> > > > > > Sorry for the continued annoyance! I very much apologise.
>
> > > > > > Malcolm
>
> > > > > > On Jul 7, 5:11 am, Sunny  wrote:
>
> > > > > > > Hi day_trader
>
> > > > > > > As far as i can think is that you are trying to pass an object 
> > > > > > > with
> > > > > > > the data of the class..
>
> > > > > > > now to make your class go to the server side all you need to do 
> > > > > > > is to
> > > > > > > move it to a package that is in the package that is *not* 
> > > > > > > mentioned in
> > > > > > > your .gwt.xml file as the classes in these
> > > > > > > packages,which are mentioned in the gwt.xml file, get "ajaxified" 
> > > > > > > and
> > > > > > > sent on to the client.
>
> > > > > > > Hope this helps.
>
> > > > > > > Sunny.
>
> > > > > > > On Jul 6, 5:04 pm, day_trader  wrote:
>
> > > > > > > > Tha

Re: onSuccess() return value other than void?

2010-07-07 Thread andreas
The inherits tag is for inheriting GWT modules not for inheriting Java
files or similar (afaik).

You add packages to the GWT translatable sources via source tag:



This should be enough assuming your GWT module is in 'com.myApp'.

Andreas

On 7 Jul., 10:44, day_trader  wrote:
> It could possibly be because I am using Enunciate to develop my
> application with. Here is my complete error:
>
> [INFO]
> 
> [ERROR] BUILD ERROR
> [INFO]
> 
> [INFO] Problem assembling the enunciate app.
>
> Embedded error: Class not found: TreeModel
>
> I've just tried putting TreeModel into my gwt.xml file :  name='com.myApp.client.models.TreeModel'/> however this doesn't appear
> to work either. I have posted this to the Enunciate Mailing List as
> well, so hopefully one of those guys can enlighten me if it's a
> probably with Enunciate and not GWT/GXT.
>
> Malcolm
>
> On Jul 7, 9:28 am, andreas  wrote:
>
>
>
> > Sorry, it's war/WEB-INF/classes/.
>
> > On 7 Jul., 10:26, andreas  wrote:
>
> > > You do not have to move something out of 'client' packages to have it
> > > on your server. This is definitely wrong!
>
> > > You do have to move classes you want to have on the client or both
> > > client and server into the client packages (or declare those packages
> > > in your *.gwt.xml files).
>
> > > Since you want your class to be passed from the client to the server
> > > this class definitely needs to be in your 'client' package. I don't
> > > know why you get an error on compiling, maybe you can give more
> > > informations on this one. GWT compiles all Java files (also those in
> > > 'client') to .class files in your war/WEB-INF/src/ folder so all
> > > classes will be accessible on the server.
>
> > > Andreas
>
> > > On 7 Jul., 10:02, day_trader  wrote:
>
> > > > This is where I am rather confused. As when i move it to
> > > > a package which isn't named in the .gwt.xml file I receive
> > > > an error of "cannot be resolved to a type". Which is fair enough,
> > > > as I assumed everything in the .gwt.xml was putting the classes
> > > > in 'view' of everything in the application? Or perhaps I'm very
> > > > muddled up!
>
> > > > Sorry for the continued annoyance! I very much apologise.
>
> > > > Malcolm
>
> > > > On Jul 7, 5:11 am, Sunny  wrote:
>
> > > > > Hi day_trader
>
> > > > > As far as i can think is that you are trying to pass an object with
> > > > > the data of the class..
>
> > > > > now to make your class go to the server side all you need to do is to
> > > > > move it to a package that is in the package that is *not* mentioned in
> > > > > your .gwt.xml file as the classes in these
> > > > > packages,which are mentioned in the gwt.xml file, get "ajaxified" and
> > > > > sent on to the client.
>
> > > > > Hope this helps.
>
> > > > > Sunny.
>
> > > > > On Jul 6, 5:04 pm, day_trader  wrote:
>
> > > > > > Thanks for the replies thus far.
>
> > > > > > That all makes sense. I did have strong doubts about the likelihood 
> > > > > > of
> > > > > > what I wanted but as I am not an expert in the field I thought it 
> > > > > > was
> > > > > > better to make sure.
>
> > > > > > Actually, my problems would be solved quite quickly if I got around 
> > > > > > my
> > > > > > initial problem that led to this posting. Basically, there is a 
> > > > > > class
> > > > > > in my client side, under myApp.client.models.SomeModel that I need 
> > > > > > to
> > > > > > pass to the server. The problem is that the server cannot see this
> > > > > > class as I get an error when I try to compile it. My question is: 
> > > > > > How
> > > > > > do I check if the class is on my SERVER classpath?
>
> > > > > > 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-tool...@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: onSuccess() return value other than void?

2010-07-07 Thread andreas
Sorry, it's war/WEB-INF/classes/.

On 7 Jul., 10:26, andreas  wrote:
> You do not have to move something out of 'client' packages to have it
> on your server. This is definitely wrong!
>
> You do have to move classes you want to have on the client or both
> client and server into the client packages (or declare those packages
> in your *.gwt.xml files).
>
> Since you want your class to be passed from the client to the server
> this class definitely needs to be in your 'client' package. I don't
> know why you get an error on compiling, maybe you can give more
> informations on this one. GWT compiles all Java files (also those in
> 'client') to .class files in your war/WEB-INF/src/ folder so all
> classes will be accessible on the server.
>
> Andreas
>
> On 7 Jul., 10:02, day_trader  wrote:
>
>
>
> > This is where I am rather confused. As when i move it to
> > a package which isn't named in the .gwt.xml file I receive
> > an error of "cannot be resolved to a type". Which is fair enough,
> > as I assumed everything in the .gwt.xml was putting the classes
> > in 'view' of everything in the application? Or perhaps I'm very
> > muddled up!
>
> > Sorry for the continued annoyance! I very much apologise.
>
> > Malcolm
>
> > On Jul 7, 5:11 am, Sunny  wrote:
>
> > > Hi day_trader
>
> > > As far as i can think is that you are trying to pass an object with
> > > the data of the class..
>
> > > now to make your class go to the server side all you need to do is to
> > > move it to a package that is in the package that is *not* mentioned in
> > > your .gwt.xml file as the classes in these
> > > packages,which are mentioned in the gwt.xml file, get "ajaxified" and
> > > sent on to the client.
>
> > > Hope this helps.
>
> > > Sunny.
>
> > > On Jul 6, 5:04 pm, day_trader  wrote:
>
> > > > Thanks for the replies thus far.
>
> > > > That all makes sense. I did have strong doubts about the likelihood of
> > > > what I wanted but as I am not an expert in the field I thought it was
> > > > better to make sure.
>
> > > > Actually, my problems would be solved quite quickly if I got around my
> > > > initial problem that led to this posting. Basically, there is a class
> > > > in my client side, under myApp.client.models.SomeModel that I need to
> > > > pass to the server. The problem is that the server cannot see this
> > > > class as I get an error when I try to compile it. My question is: How
> > > > do I check if the class is on my SERVER classpath?
>
> > > > 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-tool...@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: onSuccess() return value other than void?

2010-07-07 Thread andreas
You do not have to move something out of 'client' packages to have it
on your server. This is definitely wrong!

You do have to move classes you want to have on the client or both
client and server into the client packages (or declare those packages
in your *.gwt.xml files).

Since you want your class to be passed from the client to the server
this class definitely needs to be in your 'client' package. I don't
know why you get an error on compiling, maybe you can give more
informations on this one. GWT compiles all Java files (also those in
'client') to .class files in your war/WEB-INF/src/ folder so all
classes will be accessible on the server.

Andreas

On 7 Jul., 10:02, day_trader  wrote:
> This is where I am rather confused. As when i move it to
> a package which isn't named in the .gwt.xml file I receive
> an error of "cannot be resolved to a type". Which is fair enough,
> as I assumed everything in the .gwt.xml was putting the classes
> in 'view' of everything in the application? Or perhaps I'm very
> muddled up!
>
> Sorry for the continued annoyance! I very much apologise.
>
> Malcolm
>
> On Jul 7, 5:11 am, Sunny  wrote:
>
>
>
> > Hi day_trader
>
> > As far as i can think is that you are trying to pass an object with
> > the data of the class..
>
> > now to make your class go to the server side all you need to do is to
> > move it to a package that is in the package that is *not* mentioned in
> > your .gwt.xml file as the classes in these
> > packages,which are mentioned in the gwt.xml file, get "ajaxified" and
> > sent on to the client.
>
> > Hope this helps.
>
> > Sunny.
>
> > On Jul 6, 5:04 pm, day_trader  wrote:
>
> > > Thanks for the replies thus far.
>
> > > That all makes sense. I did have strong doubts about the likelihood of
> > > what I wanted but as I am not an expert in the field I thought it was
> > > better to make sure.
>
> > > Actually, my problems would be solved quite quickly if I got around my
> > > initial problem that led to this posting. Basically, there is a class
> > > in my client side, under myApp.client.models.SomeModel that I need to
> > > pass to the server. The problem is that the server cannot see this
> > > class as I get an error when I try to compile it. My question is: How
> > > do I check if the class is on my SERVER classpath?
>
> > > 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-tool...@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: RPC - which exception types may be used???

2010-07-06 Thread andreas
1) I'm not sure if you can use SQLException since its source code is
probably not known by GWT (it's not in an inherited module) so you
better catch it and throw your own exception just like you suggested

2) I'm not 100% sure but I don't think the async service definition
can throw exceptions; it just sort of wraps your service as an async
version where the return type is wrapped in a async callback, the true
service logic (return type, parameters and exception) is in the
service and its implementation; also the exception thrown by the
service method is accessible in the onFailure() method if the async
callback

Try removing the throws in your async service, it should work.

Andreas

On 6 Jul., 18:29, Magnus  wrote:
> Hi,
>
> another point is: If I declare my own exception type SystemService,
> and if I let the method throw this exception in all three
> implementation files (...Service, ...ServiceAsync, ...ServiceImpl),
> then I get an "uncaught exception" compiler error at the service
> invocation point.
>
> But this is not the case in the GreetingService example...
>
> What do I have to do that I can invoke the service callback without
> try/catch?
>
> Thanks
> Magnus

-- 
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-tool...@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: RPC - which exception types may be used???

2010-07-05 Thread andreas
>From what I know you can use any type of exception, but:

- GWT needs to know the source (file) of the exception (it's either in
an inherited module or in your 'client' code or what ever package you
declared in your *.gwt.xml)
- it needs to be serializable (implement Serializable and do not
forget empty constructor)

That's it I think.

Andreas

On 5 Jul., 18:55, Magnus  wrote:
> Hi,
>
> I am trying to implement an RPC call that loads a record from a
> database:
>
>  User loadUser (String nickname) throws Exception;
>
> I tried different Exception types:
>
> SQLException: results in mystic errors (not serializable)
> Exception: results in error "uncaught exception" at the service call
> IllegalArgumentException: works, but is not that what I want
>
> Question 1:
> What Exception types may be used for RPC?
>
> Question 2:
> Where (which source files) do I have to declare the exception?
> Service
> ServiceAsync
> ServiceImpl
>
> Thanks
> Magnus

-- 
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-tool...@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: gradient background over multiple panels?

2010-07-04 Thread andreas
You could assign that gradient background to the superior panel in
which the other ones are inserted and make them transparent.

Andreas

On 4 Jul., 15:21, Magnus  wrote:
> Hi,
>
> I have multiple different panels stacked vertically I assigned a
> gradient background to each of them. But this does not look good,
> because the gradient begins again at each panel. Can I assign a common
> background image over multiple panels?
>
> Magnus

-- 
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-tool...@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: How to show a load screen during rpc call?

2010-07-04 Thread andreas
OK, I have not used RpcRequestBuilder yet and only used RequestBuilder
to alter the target of the request.

How exactly would you use RpcRequestBuilder to hide the panel? And
also from what I understood RpcRequestBuilder has nothing to do with
actually sending the request so displaying a panel on creation of the
request is kind of a short moment too early.

I think I'd stick to a combination of a subclass of RequestBuilder
where in send() the panel is shown and an implicitly assigned custom
abstract AsyncCallBack implementation where onSuccess()/onFailure
methods hide the panel and delegate the onSuccess()/onFailure() to a
subclass.

This of course depends on two things:
1) usage of return type in async service interface
2) possibility to use a subclass of RequestBuilder in async service
interface (not sure about that but I might give it a try)

Andreas

On 3 Jul., 23:27, Thomas Broyer  wrote:
> On 3 juil, 21:45, andreas  wrote:
>
> > This sounds really great Thomas, haven't tried that yet.
>
> > Can this be done by subclassing RequestBuilder? And can I actually
> > really use my subclass of RequestBuilder as return type in
> > RemoteService methods? Would make me love the ability to use this
> > "return" feature of GWT RPC even more!
>
> It's nothing to do with the return type of methods in your Async
> interface. I'm talking about RpcRequestBuilder, not 
> RequestBuilder:http://google-web-toolkit.googlecode.com/svn/javadoc/2.0/com/google/g...
>
> (and yes, it allows you to use your subclass of RequestBuilder; but
> IMO you'd rather show the popup in RpcRequestBuilder#doFinish (or
> maybe even earlier, such as in doCreate()), and wrap the
> RequestCallback in doSetCallback; but if you already have a
> RequestBuilder-subclass, then just return it from doCreate)

-- 
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-tool...@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: How to show a load screen during rpc call?

2010-07-03 Thread andreas
This sounds really great Thomas, haven't tried that yet.

Can this be done by subclassing RequestBuilder? And can I actually
really use my subclass of RequestBuilder as return type in
RemoteService methods? Would make me love the ability to use this
"return" feature of GWT RPC even more!

On 3 Jul., 02:23, Thomas Broyer  wrote:
> On 2 juil, 18:22, andreas  wrote:
>
> > You could show your load screen (for example a popup panel or even
> > just a label) right after issuing the RPC and hide it in onSuccess()
> > and onFailure().
>
> And if you want the same panel to be shown in most cases, you could
> bake it into a RpcRequestBuilder that you then inject inside you
> RemoteService/Async, and that's it, nothing special to be done at the
> "call site" of your service's method.

-- 
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-tool...@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: How to show a load screen during rpc call?

2010-07-02 Thread andreas
You could show your load screen (for example a popup panel or even
just a label) right after issuing the RPC and hide it in onSuccess()
and onFailure().

On 2 Jul., 18:02, crojay78  wrote:
> Hi,
>
> I need to load a few things from my server after a user gives input.
> Can somebody give an example how I can show a load screen as long as
> the rpc is not finished? I structured the app in mvp style ... any
> example would help me
>
> Best regards
>
> 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-tool...@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: onSuccess() return value other than void?

2010-07-02 Thread andreas
Welcome to the async world! ;-)

I had the same problem. Since the code of an async callback is not
executed in the order of the statements in your myMethod() you can not
directly return the results from onSuccess() in myMethod(). And
myMethod() is not able to "wait" for onSuccess().

You could for example do the things you want to do with
ArrayList in onSuccess() instead of where you call myMethod()
of if you want to keep the processing logic for ArrayList
where myMethod() is called use a method to pass ArrayList from
onSuccess().

You can think of it as separating the logic in two methods:
1) first one does initial stuff and then requests something via RPC
(in your example via myMethod() but with return type void)
2) second one is called by onSuccess() passing the results of the RPC
and continues with the requested data where first method "ends"

Hope it helps,

Andreas

On 2 Jul., 14:50, day_trader  wrote:
> At present, an AsyncCallback contains a 'public void onSuccess()'
> method. This is posing significant problems for me at the moment.
>
> I have a method myMethod() being called which has a return value type
> of ArrayList. MyMethod contains this AsynCallback which is
> used to query a database on the server side of the code. I need to,
> either somehow make the method WAIT for the AsyncCallback's
> onSuccess() method to return which is the thing that is giving me the
> ArrayList to return to the code which called myMethod, or
> change the return type of the AsyncCallback to ArrayList and
> let this be the 'return' of myMethod.
>
> Is this possible? Am I very confused? Could someone please advise?

-- 
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-tool...@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: why do I sometimes get a javascript error of "uncaught exception: java.lang.NullPointerException"?

2010-06-30 Thread andreas
You could try using an UncaughtExceptionHandler. Assign your
implementation in onModuleLoad() by

GWT.setUncaughtExceptionHandler(...);

Then use a DeferredCommand for calling your actual entry point method:

DeferredCommand.addCommand(new Command() {
public void execute() {
yourEntryPointMethod();
}
});

You can use the UncaughtExceptionHandler to print stack trace for
example.

See 
http://google-web-toolkit.googlecode.com/svn/javadoc/1.6/com/google/gwt/core/client/GWT.UncaughtExceptionHandler.html
and 
http://google-web-toolkit.googlecode.com/svn/javadoc/1.6/com/google/gwt/core/client/GWT.html

On 30 Jun., 22:03, Shedokan  wrote:
> Why there are some times while doing the same actions in my app I get
> this javascript error:
> uncaught exception: java.lang.NullPointerException
>
> I have no idea where it is coming from since there is no line number
> and have no idea what is causing it.
>
> how can I trace it and fix it?
> and why does it show up in some times and in other times it doesn't?
>
> 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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-28 Thread andreas
I think the difference was that you never set the size of the Image
instances.

You know the resolution of the images you use as background. Try
calling setPixelSize(...) on each Image instance with the image
resolution values. This way you can make sure the Image has the
correct size, thus the superior cell and finally the overall Grid as
well.

I develop applications with more than 100 Panel instances (not to
mention all other Widgets and custom Widgets) and there is no
noticeable resource problem. I think GWT does a pretty good job there.

In my example I never call setPixelSize(...) on the Grid and it
apparently is not of height 0. Why should it be?

Additionally as you saw in your code, calling it led to strange
dimension problems/overlapping, ...

On 28 Jun., 18:28, Magnus  wrote:
> BTW: I must set the grids pixel size, because it will be of height 0.
> The parent is a DockLayoutPanel...
>
> But it's ok, since I know the correct size...
>
> Magnus

-- 
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-tool...@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 Popup is not centered

2010-06-28 Thread andreas
Did you try calling center() after the constructor, for example right
after where you instantiate the PopupPanel without the center() inside
the constructor?

I'm not that deep into GWT core but maybe it's related to the fact
that elements get their real/actual dimensions (and these are needed
to calculate the center position) only after being completely attached
to the DOM. Maybe this is not the case from within the constructor.

You can observe this behavior with a Panel. Set its pixel size but
don't add it to the RootPanel or any other Panel. Calling
getOffsetHeight/Width will not return the expected values until the
Panel is attached.

On 28 Jun., 15:32, RPB  wrote:
> I see the exact same behaviour in my application, and would appreciate
> any ideas with this issue.
> The popup is not centered for the first time only, and afterwards
> popup.center() works fine. I am using GWT 2.0.3. Here is some sample
> code:
>
> public class LoginForm extends PopupPanel {
>         public LoginForm(){
>                 this.setStyleName("bwPopupPanel");
>                 this.setGlassEnabled(true);
>                 this.add(new LoginFields());
>                 this.center();
>         }
>
> }
>
> //..Different class..
>         private void showLoginPopUp() {
>                 @SuppressWarnings("unused")
>                 LoginForm loginPanel = new LoginForm();
>         }
>
> Any thoughts?
>
> On Jun 27, 9:06 pm, "Ricardo.M"  wrote:
>
>
>
> > Hi,
>
> > I use gwtpopupto show some messages, but it isnotdisplayed in thecenterof 
> > the display event if i callpopup.center(). Actually it isnotcentered
> > only the first time, if i close it and open it again every thing is
> > ok,
> > butnotthe first time. How to fix that?
>
> > I tried using  .setPopupPosition(350, 250);  but thepopupisnot
> > centered only the first time.
>
> > Regards.

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-28 Thread andreas
Yes I see.

Maybe the image file dimensions and the dimensions of the Image
instances are "conflicting".

When I get it right the dimensions of the Image instances are never
set. Instead they are resized based on the dimension of the Grid
itself. Now if the Image instances display an image with different
dimensions I can not tell what this results in.

Look at my code, I never call setSize(...) or setPixelSize(...) on the
Grid. It will automatically resize based on the dimensions of its
cells.

Try setting the pixel size of the Image instances (rather than Grid)
so that the (raw) image is perfectly fitted (resolution of the png or
whatever).

In my example the SimplePanel dimensions are simply 40x40 and for the
border cells 40x20/20x40 ... the overall size results in the row and
column sum of all cells dimensions AND is never set manually.

On 28 Jun., 14:48, Magnus  wrote:
> Hello,
>
> I made another screenshot with red borders around the 
> images:http://yfrog.com/cachessboardj
>
> As you can see there is vedrtical space below each image.
>
> I just found out that this is not the case in IE.
>
> Thanks
> Magnus
>
> On 28 Jun., 11:51, andreas  wrote:
>
>
>
> > Where exactly are the vertical spaces? From what I see, there are no
> > spaces between the cells of the top and bottom row and the spaces
> > between the cells in the left and right column are of the same color
> > as the image background, so I assume there are actually also no
> > spaces, correct me on this one?
>
> > For better debug you could also assign a border and background color
> > to the Grid. If none of these colors will be visible you can be sure
> > that there are no spaces or anything left.
>
> > For the inner cells I can not say if there are any spaces.
>
> > Also I assume your style "pnl-r" adds the red border to the Grid? If
> > there were any spaces left caused by the border width you would see a
> > red Grid.
>
> > I think you got what you wanted...
>
> > BTW: I think you do not need to set the cell dimensions manually; Grid
> > will automatically adjust cell dimensions so that the cell widgets
> > fit, in other words a columns width for example will be adjusted so
> > that the cell widget with the biggest width is completely visible;
> > same goes for rows and heights and so on
>
> > BTW2: the two for-blocks in init() can be realized in one single for-
> > block since they iterate over exactly the same interval (0-9)
>
> > On 28 Jun., 11:27, Magnus  wrote:
>
> > > Here is the screenshot:
>
> > >http://yfrog.com/j7chessboardj
>
> > > Magnus- Zitierten Text ausblenden -
>
> > - Zitierten Text anzeigen -

-- 
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-tool...@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: Unable to find 'Student.gwt.xml' on your classpath; could be a typo, or maybe you forgot to include a classpath entry for source?

2010-06-28 Thread andreas
Did you remove all inherits tags in all other modules referencing
Student.gwt.xml?

It may be a module still referencing it...

On 28 Jun., 12:26, Aditya <007aditya.b...@gmail.com> wrote:
> hello,
>
>       Its been now long time i m using GWT , I am facing some problem
> with GWT modules.
>
> I was having several modules in my application but over the period I
> updated my application and just wanted to remove some unused stuff
> from the application.
>
> So I deleted the module Student and then i tried running application
> in hosted mode it worked fine but I got error on the console saying
>
> [ERROR] Unable to find 'Student.gwt.xml' on your classpath; could be a
> typo, or maybe you forgot to include a classpath entry for source?
>
> I am not able to figure out why I am getting this error  as that is
> module is not present and I have deleted all the files which were
> associated with that module so no chance of having any reference.
>
> Still i m getting that error message and just because of that I can't
> run my application in debug mode so i cant hit the client side break
> points as by passing extra URL param application refuses to run.
>
> Any one has encountered this problem before if yes then please let me
> know.
>
> thank you
>
> --
> Aditya

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-28 Thread andreas
Here's code I'd use to create a simplistic chess board:

// the chess board with no spaces
Grid cb = new Grid(10, 10);
cb.setCellPadding(0);
cb.setCellSpacing(0);
cb.setBorderWidth(0);

// assembles the board by inserting colored panels
for (int i = 1; i < 9; i++) {
// panels of the top row
SimplePanel pHTop = new SimplePanel();
pHTop.setPixelSize(40, 20);
pHTop.getElement().getStyle().setBackgroundColor("red");

// panels of the bottom row
SimplePanel pHBottom = new SimplePanel();
pHBottom.setPixelSize(40, 20);
pHBottom.getElement().getStyle().setBackgroundColor("red");

// panels of the left column
SimplePanel pVLeft = new SimplePanel();
pVLeft.setPixelSize(20, 40);
pVLeft.getElement().getStyle().setBackgroundColor("green");

// panels of the right column
SimplePanel pVRight = new SimplePanel();
pVRight.setPixelSize(20, 40);
pVRight.getElement().getStyle().setBackgroundColor("green");

// insert the border cells
cb.setWidget(0, i, pHTop);
cb.setWidget(9, i, pHBottom);
cb.setWidget(i, 0, pVLeft);
cb.setWidget(i, 9, pVRight);

for (int j = 1; j < 9; j++) {
// the inner chess board panels
SimplePanel cP = new SimplePanel();
cP.setPixelSize(40, 40);
// switches between black and white
if (j % 2 == 0) {
cP.getElement().getStyle().setBackgroundColor(
i % 2 == 0 ? "black" : "white");
} else {
cP.getElement().getStyle().setBackgroundColor(
i % 2 == 0 ? "white" : "black");
}
cb.setWidget(i, j, cP);
}
}

// there it is
RootPanel.get().add(cb, 1, 1);

Programmatic styles are of course not as good as using stylesheets,
consider this just a demo.

On 28 Jun., 11:51, andreas  wrote:
> Where exactly are the vertical spaces? From what I see, there are no
> spaces between the cells of the top and bottom row and the spaces
> between the cells in the left and right column are of the same color
> as the image background, so I assume there are actually also no
> spaces, correct me on this one?
>
> For better debug you could also assign a border and background color
> to the Grid. If none of these colors will be visible you can be sure
> that there are no spaces or anything left.
>
> For the inner cells I can not say if there are any spaces.
>
> Also I assume your style "pnl-r" adds the red border to the Grid? If
> there were any spaces left caused by the border width you would see a
> red Grid.
>
> I think you got what you wanted...
>
> BTW: I think you do not need to set the cell dimensions manually; Grid
> will automatically adjust cell dimensions so that the cell widgets
> fit, in other words a columns width for example will be adjusted so
> that the cell widget with the biggest width is completely visible;
> same goes for rows and heights and so on
>
> BTW2: the two for-blocks in init() can be realized in one single for-
> block since they iterate over exactly the same interval (0-9)
>
> On 28 Jun., 11:27, Magnus  wrote:
>
>
>
> > Here is the screenshot:
>
> >http://yfrog.com/j7chessboardj
>
> > Magnus

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-28 Thread andreas
Where exactly are the vertical spaces? From what I see, there are no
spaces between the cells of the top and bottom row and the spaces
between the cells in the left and right column are of the same color
as the image background, so I assume there are actually also no
spaces, correct me on this one?

For better debug you could also assign a border and background color
to the Grid. If none of these colors will be visible you can be sure
that there are no spaces or anything left.

For the inner cells I can not say if there are any spaces.

Also I assume your style "pnl-r" adds the red border to the Grid? If
there were any spaces left caused by the border width you would see a
red Grid.

I think you got what you wanted...

BTW: I think you do not need to set the cell dimensions manually; Grid
will automatically adjust cell dimensions so that the cell widgets
fit, in other words a columns width for example will be adjusted so
that the cell widget with the biggest width is completely visible;
same goes for rows and heights and so on

BTW2: the two for-blocks in init() can be realized in one single for-
block since they iterate over exactly the same interval (0-9)

On 28 Jun., 11:27, Magnus  wrote:
> Here is the screenshot:
>
> http://yfrog.com/j7chessboardj
>
> Magnus

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-28 Thread andreas
Try setting the border width on the Grid instance to 0:

http://google-web-toolkit.googlecode.com/svn/javadoc/1.5/com/google/gwt/user/client/ui/HTMLTable.html#setBorderWidth(int)

- Andreas

On 28 Jun., 09:43, Magnus  wrote:
> Hi Andreas,
>
> thank you!
>
> I found out something:
>
> I thought I could get a solution by inserting images of the correct
> pixel size into each cell.
>
> So I created transparent images with the exact pixel sizes:
> 50x50 for the inner cells, 50x10 for the horizontal borders, 10x50 vor
> the vertical borders.
>
> As a result, I found out that there is always some vertical space
> between the images, one or some pixels. I also found out, that the
> pixel size of the whole grid has changed after inserting the images,
> so that it is not a square anymore.
>
> I inserted some visible pixels into the transparent images, and I
> verified that there is vertical space between the images.
>
> (I set cellspacing and cellpadding to 0)
>
> How can I get the grid to show its cells without any space?
>
> Thanks
> Magnus
>
> On 27 Jun., 12:59, andreas  wrote:
>
>
>
> > Hey,
>
> > maybe this can help you:
>
> >http://groups.google.com/group/google-web-toolkit/msg/f59a0c87d0cf300e
>
> > On 27 Jun., 07:23, Magnus  wrote:
>
> > > Hi,
>
> > > I need a Grid with the following simple requirements:
>
> > > - 10 * 10 cells
>
> > > - inner 8*8 cells (form a chess board):
> > >   rows 1-8; cols 1-8: pixel size 50 * 50
>
> > > - outer cells (form the annotations, e. g. "A", "1"):
> > >   rows 0 and 9: pixel height: 10
> > >   cols 0 and 9pixel width: 10
>
> > > The total pixel size of the grid must therefore be 8*50 + 2*10 = 420
> > > (width and height).
>
> > > I cannot get this realized with a grid. Although I set the pixel sizes
> > > for every cell, it's never correct. The annotation cells are always of
> > > the same size as the other cells.
>
> > > Below is my code.
>
> > > What can I do?
>
> > > Thanks
> > > Magnus
>
> > > private void init ()
> > >  {
> > >   setPixelSize (420,420);
>
> > >   grd.setPixelSize (420,420);
> > >   grd.setBorderWidth (0);
> > >   grd.setCellPadding (0);
> > >   grd.setCellSpacing (0);
>
> > >   CellFormatter cf = grd.getCellFormatter ();
>
> > >   // set the inner cells
>
> > >   for (int y = 1;y < 9;y++)
> > >    for (int x = 1;x < 9;x++)
> > >    {
> > >     String s;
>
> > >     if (((x + y) & 1) != 0)
> > >      s = "blackCell";
> > >     else
> > >      s = "whiteCell";
>
> > >     cf.addStyleName (y,x,s); // just the cells color, no sizes here
> > >     cf.setWidth (y,x,"50px");
> > >     cf.setHeight (y,x,"50px");
> > >    }
>
> > >   // set the outer cells
>
> > >   for (int x = 0;x < 10;x++)
> > >   {
> > >    cf.setHeight (x,0,"10px");
> > >    cf.setHeight (x,9,"10px");
> > >   }
>
> > >   for (int y = 0;y < 10;y++)
> > >   {
> > >    cf.setWidth (0,y,"10px");
> > >    cf.setWidth (9,y,"10px");
> > >   }
> > >  }- Zitierten Text ausblenden -
>
> > - Zitierten Text anzeigen -

-- 
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-tool...@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: how to control the pixel sizes of a Grid?

2010-06-27 Thread andreas
Hey,

maybe this can help you:

http://groups.google.com/group/google-web-toolkit/msg/f59a0c87d0cf300e

On 27 Jun., 07:23, Magnus  wrote:
> Hi,
>
> I need a Grid with the following simple requirements:
>
> - 10 * 10 cells
>
> - inner 8*8 cells (form a chess board):
>   rows 1-8; cols 1-8: pixel size 50 * 50
>
> - outer cells (form the annotations, e. g. "A", "1"):
>   rows 0 and 9: pixel height: 10
>   cols 0 and 9pixel width: 10
>
> The total pixel size of the grid must therefore be 8*50 + 2*10 = 420
> (width and height).
>
> I cannot get this realized with a grid. Although I set the pixel sizes
> for every cell, it's never correct. The annotation cells are always of
> the same size as the other cells.
>
> Below is my code.
>
> What can I do?
>
> Thanks
> Magnus
>
> private void init ()
>  {
>   setPixelSize (420,420);
>
>   grd.setPixelSize (420,420);
>   grd.setBorderWidth (0);
>   grd.setCellPadding (0);
>   grd.setCellSpacing (0);
>
>   CellFormatter cf = grd.getCellFormatter ();
>
>   // set the inner cells
>
>   for (int y = 1;y < 9;y++)
>    for (int x = 1;x < 9;x++)
>    {
>     String s;
>
>     if (((x + y) & 1) != 0)
>      s = "blackCell";
>     else
>      s = "whiteCell";
>
>     cf.addStyleName (y,x,s); // just the cells color, no sizes here
>     cf.setWidth (y,x,"50px");
>     cf.setHeight (y,x,"50px");
>    }
>
>   // set the outer cells
>
>   for (int x = 0;x < 10;x++)
>   {
>    cf.setHeight (x,0,"10px");
>    cf.setHeight (x,9,"10px");
>   }
>
>   for (int y = 0;y < 10;y++)
>   {
>    cf.setWidth (0,y,"10px");
>    cf.setWidth (9,y,"10px");
>   }
>  }

-- 
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-tool...@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: Button look and feel

2010-06-26 Thread andreas
Ok, that's everything but "overwriting". It's simply assigning a
different style.

And by the way: why subclass? You can simply call setStylePrimaryName
on an instance of Button, no need for a subclass to just change the
style.

On 27 Jun., 00:58, Jim Douglas  wrote:
> Apparently I created a cascade of confusion here.  When I said
> "override the default gwt-Button style", I simply meant subclass it
> and add something like this to the constructor:
>
>         setStylePrimaryName("my-custom-button");
>
> On Jun 26, 4:14 am, andreas  wrote:
>
>
>
> > Are you sure it is possible to overwrite the default gwt-Button style?
>
> > I tried it once and found out that it did not work. This was the case
> > because the default GWT styles are somehow otherwise injected than
> > manually added styles. The default style is loaded via style element
> > in the module xml.
>
> > I found a link where it says that styles injected via module xml can
> > not be overwritten by hand. I'll have a quick look and paste it
> > here...
>
> > There it 
> > is:http://groups.google.com/group/Google-Web-Toolkit/browse_thread/threa...
>
> > See Thomas Broyers answer. You can not overwrite default GWT styles,
> > unless using important rules.
>
> > You should stick to writing your own CSS style and assigning it via
> > setStyleName(...).
>
> > Greetings,
>
> > Andreas
>
> > On 25 Jun., 23:14, Jim Douglas  wrote:
>
> > > Chris --
>
> > > PushButton is a styled DIV, but Button is a standard HTML button
> > > object.  You want to use Button, and override the default gwt-Button
> > > style.
>
> > >http://google-web-toolkit.googlecode.com/svn/javadoc/2.1/com/google/g...
>
> > > On Jun 25, 3:50 am, Takapa  wrote:
>
> > > > Hi,
>
> > > > I have been using GWT for around 18 months via Smart GWT but have
> > > > decided to create a pure GWT project but having a few hiccups.
>
> > > > I'm having a seemingly trivial problem with look and feel at the
> > > > moment. When I create a button it is coloured grey to white with a
> > > > gradient fill. I just want a regular HTML button that matches my OS
> > > > standard buttons. Can anyone tell me how to specify this either
> > > > programatically or via CSS.
>
> > > > Thanks,
>
> > > > Chris

-- 
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-tool...@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: Button look and feel

2010-06-26 Thread andreas
So if one uses a module which defines styles and injects them using
 one can override these styles in ones own ?

ModuleA has StylesA injected via ;
MyModule inherits ModuleA and can override StylesA in MStyleSheet
injected via 

Is this the case?



On 26 Jun., 17:23, Ian Bambury  wrote:
> You can just add your style sheets after the GWT one in the *.gwt.xml file
>
>     
>
> relative to the /war/ directory
>
> These are added after the HTML is processed and therefore overwrite them
>
> On 26 June 2010 15:54, andreas  wrote:
>
>
>
>
>
> > On 26 Jun., 15:13, Thomas Broyer  wrote:
> > > On 26 juin, 13:14, andreas  wrote:
>
> > > > Are you sure it is possible to overwrite the default gwt-Button style?
>
> > > Yes.
>
> > Well, actually it is, but not by only adding .gwt-Button style in my
> > custom stylesheet and adding different values than in the standard.css
> > file where the original css values are in. As you said those
> > overwriting values will be overwritten by those in the standard.css
> > since these are injected, right? So by only "overwriting" it is not.
> > You have to do the resource trick and manually copy/add standard.css.
> > In this setup you can literally overwrite standard.css settings. Did I
> > get you right there?
>
> > > > See Thomas Broyers answer. You can not overwrite default GWT styles,
> > > > unless using important rules.
>
> > > Please follow the links from my message there, this is NOT what I
> > > said.
> >http://groups.google.com/group/google-web-toolkit/msg/a741cc9a600109c7
>
> > You're right sorry about that. This "important rule"-thing was an
> > advice from someone else.
>
> > --
> > 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-tool...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > google-web-toolkit+unsubscr...@googlegroups.com > cr...@googlegroups.com>
> > .
> > For more options, visit this group at
> >http://groups.google.com/group/google-web-toolkit?hl=en.

-- 
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-tool...@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: Button look and feel

2010-06-26 Thread andreas


On 26 Jun., 15:13, Thomas Broyer  wrote:
> On 26 juin, 13:14, andreas  wrote:
>
> > Are you sure it is possible to overwrite the default gwt-Button style?
>
> Yes.

Well, actually it is, but not by only adding .gwt-Button style in my
custom stylesheet and adding different values than in the standard.css
file where the original css values are in. As you said those
overwriting values will be overwritten by those in the standard.css
since these are injected, right? So by only "overwriting" it is not.
You have to do the resource trick and manually copy/add standard.css.
In this setup you can literally overwrite standard.css settings. Did I
get you right there?

>
> > See Thomas Broyers answer. You can not overwrite default GWT styles,
> > unless using important rules.
>
> Please follow the links from my message there, this is NOT what I
> said.http://groups.google.com/group/google-web-toolkit/msg/a741cc9a600109c7

You're right sorry about that. This "important rule"-thing was an
advice from someone else.

-- 
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-tool...@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.



  1   2   >