Google Web Toolkit: February 2013

Thursday, February 28, 2013

Re: adding custom widget to FormPanel

Em quinta-feira, 28 de fevereiro de 2013 16h43min55s UTC-3, seven.reeds escreveu:

Saw a post in a different forum that suggested putting a "Hidden" widget in the custom one that would receive the actual custom "name" and "value". It suggested that the form would find the Hidden widget and sent that properly. Makes sense to me.

My test this morning says, No, that does not work.

Remember that all GWT can do is compile to HTML, and an HTML form will submit only the values of form elements such as input and hidden.

--
P.

--
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+unsubscribe@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: GWT dev mode and GIN and dependency injection in general

I believe you're right. If you "mess" with DI stuff (for example add a new parameter to a constructor annotated with @Inject) the code will break when you reload. You have to rebuild the project at that point. I believe this is due to gin using generators which are not triggered on reload.

You should update your GIN version ;-) With GIN 2.0 you can simply reload the page and nothing will break when you have changed dependencies in constructors, etc. or you have changed bindings in your GinModule.

-- J.

Re: GWT dev mode and GIN and dependency injection in general

On Thursday, February 28, 2013 11:21:48 PM UTC+1, alucard wrote:

On Thursday, February 28, 2013 6:43:11 PM UTC+1, Marc wrote:
Hello!

I just experimented somewhat with Gin and realized that it (v2.0) seems to break the incremental compilation feature of GWT's development mode, ie. where I just have to save the .java source file and the gwt dev mode server automatically incorporates the changes.

Can someone confirm this or is it my own fault?

I believe you're right. If you "mess" with DI stuff (for example add a new parameter to a constructor annotated with @Inject) the code will break when you reload. You have to rebuild the project at that point. I believe this is due to gin using generators which are not triggered on reload.

While this was true in previous versions, I believe it no longer is starting with GIN 2.0.

How to detect when user DnD something into the app and the page unloads?

I'd like to disable DnD in my GWT app unless I specifically allow it (which I do in places). Is there a way to detect and optionally prevent drops globally (prevent page unloads in this case)?

My use case is that we do support DnD in certain ways but if they drop wrong it causes the entire page to unload and then when they select the back browser button...the app will come back but they have to login again and reinitialize the app.

I thought of just adding a global Window.ClosingHandler() handler but that doesn't work because...there are lots of valid reasons the page is unloading and our security filters have already processed the logout...so that's too late...I think I need to detect that they did a drop, possibly by accident, and allow them to cancel the drop.

-Dave

Re: GWT dev mode and GIN and dependency injection in general

On Thursday, February 28, 2013 6:43:11 PM UTC+1, Marc wrote:

Hello!

I just experimented somewhat with Gin and realized that it (v2.0) seems to break the incremental compilation feature of GWT's development mode, ie. where I just have to save the .java source file and the gwt dev mode server automatically incorporates the changes.

Can someone confirm this or is it my own fault?

I believe you're right. If you "mess" with DI stuff (for example add a new parameter to a constructor annotated with @Inject) the code will break when you reload. You have to rebuild the project at that point. I believe this is due to gin using generators which are not triggered on reload.

In general: what DI frameworks are there for GWT? I'd like to have something akin to Spring's auto-wiring based on interfaces+class path scanning, but, of course, *before* runtime, ie. at compile time to avoid performance loss in production. I'm even thinking about an annotation processor that generates the GinModule bindery code, however, using something that already exists and is properly maintained is always an advantage ;-).

I believe gin is the best DI for GWT out there.

Best regards,
Mark

Re: IE 8 issues with GWT 2.5

Our app runs with SSL and IE8 and any GWT version just fine.

Make sure that IE 8 does not automatically activate compatibility mode for intranet sites. Also check your IE 8 security settings, because you have said that it works when you add your site to the list of trusted sites.

-- J.

GWT generator problem

Hello,

I am using GWT 2.5.0 with Sencha GXT 3.0.1. I want to use com.google.gwt.core.ext.Generator to generate dynamic code for my application. I created a Web Application Project in eclipse and follow the tutorial http://francisshanahan.com/index.php/2010/a-simple-gwt-generator-example/. The example of the tutorial worked correctly. When I add in file ---.gwt.xml of my project the inherits <inherits name='com.sencha.gxt.ui.GXT'/> to use GXT in my project and compile project the generator does not work. It invoked the method PaintScreen in class ExampleScreen instead of PaintScreen defined by generator. However, in debug mode everything works fine. Only when compiling (Compile GWT) is swapping the classes

How to solve this situation?
thanks

Re: adding custom widget to FormPanel

Saw a post in a different forum that suggested putting a "Hidden" widget in the custom one that would receive the actual custom "name" and "value". It suggested that the form would find the Hidden widget and sent that properly. Makes sense to me.

My test this morning says, No, that does not work.

Re: New widgets: Best way to implement them?

[1] http://code.google.com/p/google-web-toolkit/wiki/CellBackedWIdgets
[2] http://www.sencha.com/products/gxt/whats-new-in-sencha-gxt-3/
[3] https://groups.google.com/d/topic/google-web-toolkit-contributors/g8WPRxkdqPA/discussion
[4] https://groups.google.com/d/topic/google-web-toolkit/neeh5YxKm-I/discussion
[5] http://code.google.com/p/collide/source/browse/#git%2Fjava%2Fcom%2Fgoogle%2Fcollide%2Fclient%2Fui

Thanks for that cross links :-)

-- J.

Re: New widgets: Best way to implement them?

Interesting topic.
Can't really comment on those things but I guess the CellWidgets are a good starting point.
I also came across a good stackoverflow reply which I think might also be useful: http://stackoverflow.com/a/7481137/356594

Yeah I have also came across this link and it looks pretty similar to the appearance pattern. Downside is that when your widget only implements IsWidget then its not fully UiBinder compatible as <g:Spinner addStyleNames=""/> does not work anymore. You probably also want to define your own IsWidget interface that extends GWT's IsWidget so you can add all methods that GWT's Widget class provides. Otherwise you can't call the Widget setter methods from UiBinder (setHeight(), etc.).

-- J.

Re: GWT dev mode and GIN and dependency injection in general

I just experimented somewhat with Gin and realized that it (v2.0) seems to break the incremental compilation feature of GWT's development mode, ie. where I just have to save the .java source file and the gwt dev mode server automatically incorporates the changes.

Not exactly sure what you mean with incremental compilation but when I reload my app everything works like expected and file changes are picked up by DevMode (Gin changes are also picked up on the fly). Just realized that I have never tried to change a file and see if changes are picked up by DevMode when I just navigate back and forth in the app so the modified classes need to be re-instantiated.

Can someone confirm this or is it my own fault?

In general: what DI frameworks are there for GWT? I'd like to have something akin to Spring's auto-wiring based on interfaces+class path scanning, but, of course, *before* runtime, ie. at compile time to avoid performance loss in production.

For GWT client code your DI options are pretty much

- google-gin (based on google-guice)

- sheath (based on Dagger from Square). But sheath is only a proof of concept. Not sure how well it currently works.

- I think Errai from JBoss also provides DI functionalities.

-- J.

GWT Developer plugin

I have been trying to install the GWT Developer plugin for the latest Chrome browser but keep getting

'This application is not supported on this computer. Installation has been disabled'

Does anyone Please know the reason for this?

I have just upgraded to Windows 8 Pro from Vista and the plugin worked fine before with Chrome, as it is now with

IE9 and FireFox.

GWT dev mode and GIN and dependency injection in general

Hello!

I just experimented somewhat with Gin and realized that it (v2.0) seems to break the incremental compilation feature of GWT's development mode, ie. where I just have to save the .java source file and the gwt dev mode server automatically incorporates the changes.

Can someone confirm this or is it my own fault?

In general: what DI frameworks are there for GWT? I'd like to have something akin to Spring's auto-wiring based on interfaces+class path scanning, but, of course, *before* runtime, ie. at compile time to avoid performance loss in production. I'm even thinking about an annotation processor that generates the GinModule bindery code, however, using something that already exists and is properly maintained is always an advantage ;-).

Best regards,
Mark

CssResource RTLsupport, textBox issue

Currently when injecting a style using a ClientBundle interface like this TasklistResourceBundle.INSTANCE.style().ensureInjected();

If we set the locale ?locale=ar_iq to arabic everything is moved to the RHS which is fine, however when typing into textBoxes the cursor is positioned on the left not the right.

The css we are injecting

.tasklistwidget {

width: 100%;

height: 100%;

}

I have found a workaround, but struggling to understand how this works.

i have created another CSS file

.tasklistwidget {

float:right;

direction: rtl;

width: 100%;

height: 100%;

}

and inject this instead. The text box now functions as expected and all elements are on the RHS.

Here is our extent of ClientBundle interface

public interface TasklistResourceBundle extends ClientBundle

{

public static final TasklistResourceBundle INSTANCE = GWT.create(TasklistResourceBundle.class);

/* note the source path is relative to the location of this file! */

@Source("css/tasklist-widget.css")

@CssResource.NotStrict

CssResource style();

@Source("css/tasklist-widgetRTL.css")

@CssResource.NotStrict

CssResource styleRTL();

}

Brainstorming desired - easier styling/css approaches

Hey all,
I'm curious if anyone already has some good, applied blogs or posts around the styling/css approaches with GWT applications. In particular, when going through some of these usecases and maybe from a UX designer point of view (I'm a developer, so do not know what tools (dreamweaver?) a UX/styling person may be more comfortable using, particularly with pure GWT apps):

1) Creating a new widget, how to plan for styling that would be applied to different applications using your widget.

2) Creating an application quickly, but style afterwards (or prepare for re-styling/re-branding changes).

3) Creating a 'stock' CSS template to apply to many GWT applications.

4) Given a specific application, is there an easier way to review 80% of the changes without going to each screen/widget at a time?

4-example) I was trying to brainstorm if there was an easier way to 'create a single web page with fragments of each display element in-context relevant to a specific application' --> a single page with table headers (if used in app), if table headers are always embedded in the same panel or different panels (if used in app), buttons, pulldowns, textfields with random values (or default values if you use a grey-text approach), etc. That way, instead of having different people review each screen/page of an entire app, if simply create a single page (or only a couple of pages) that covers 80% of what portions of the app would look like for easier review of changes.

thoughts?
-Darren

Re: RequestFactory issue with overloaded setters in domain type

FYI writing my own ServiceLayerDecorator with the following worked like a charm so extra thanks for the suggestion

    private static Method getBeanMethod( BeanMethod methodType, Class<?> domainType, String property, Class<?> propertyType ) {
        for ( Method m : domainType.getMethods() ) {
            if ( methodType.matches( m ) && property.equals( methodType.inferName( m ) )
                    && m.getParameterTypes()[0].isAssignableFrom( propertyType ) ) {
                m.setAccessible( true );
                return m;
            }
        }
        return null;
    }

    @Override
    public Method getSetter( Class<?> domainType, String property ) {
        Class<?> propertyType = getTop().getGetter( domainType, property ).getReturnType();
        Method setter = getBeanMethod( BeanMethod.SET, domainType, property, propertyType );
        if ( setter == null ) {
            setter = getBeanMethod( BeanMethod.SET_BUILDER, domainType, property, propertyType );
        }
        return setter;
    }

On 28 February 2013 13:44, James Horsley <james.horsley@gmail.com> wrote:

Makes sense. Thanks!

On 28 February 2013 13:27, Thomas Broyer <t.broyer@gmail.com> wrote:

On Thursday, February 28, 2013 1:50:03 PM UTC+1, James Horsley wrote:

I'll take a stab at writing a ServiceLayerDecorator as that should get me unblocked without writing DTO's and if the correct setter is called the NPE's won't be an issue. Thanks to you both for the suggestions!

Thomas, is this issue likely to end up as PatchesWelcome or WontFix? If the former then I'll look at getting a patch together.

I you send a patch, I'll look at it; the outcome will depend on what changes the patch has to make.

--
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+unsubscribe@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.

IE 8 issues with GWT 2.5

We recently started using GWT 2.5 for our GWT applications. Prior to a couple of months ago, we used GWT 2.4. After deploying the GWT 2.5 applications, we noticed that the applications no longer work with IE 8. When the user navigates to the application via an index.html page, the page is accessed successfully but the GWT java script is not loaded. We do not see the login dialog as we would expect. With Chrome and IE 9, we do not have issues.

We are using SSL.

We noticed that if we access the application using IE 8 and we add the site as a trusted site in IE 8, it works ok.

Also, we noticed that if we access the application using IE 8 and use the IP address of the machine versus the domain name, it works ok.

We've looked at the problem in Fiddler and it looks like with IE there is a problem in the SSL handshake. Things hang after about 2 steps in the handshake, whereas in the good case there are several more messages back and forth before receiving the dialog page.

For the user agent string in the gwt module, we have:

<set-property name="gxt.user.agent" value="ie8,ie9,chrome"/>

In our index.html file, we have:

<!doctype html>

One thing worth noting is that we also started using UIBinder when we moved to GWT 2.5. I'm not sure if that makes any difference or not.

Thanks for any help anyone can provide.

Re: RequestFactory issue with overloaded setters in domain type

Makes sense. Thanks!

On 28 February 2013 13:27, Thomas Broyer <t.broyer@gmail.com> wrote:

On Thursday, February 28, 2013 1:50:03 PM UTC+1, James Horsley wrote:

I'll take a stab at writing a ServiceLayerDecorator as that should get me unblocked without writing DTO's and if the correct setter is called the NPE's won't be an issue. Thanks to you both for the suggestions!

Thomas, is this issue likely to end up as PatchesWelcome or WontFix? If the former then I'll look at getting a patch together.

I you send a patch, I'll look at it; the outcome will depend on what changes the patch has to make.

--
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+unsubscribe@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: Unable to open GWT designer in

Encountered this today win7x64, cleared out the \.metadata\.plugins *gwt folders in my workspace, reloaded eclipse and the designer worked. It must have been holding on to old settings.

On Wednesday, 26 January 2011 05:14:54 UTC, Rao Venu wrote:

Hi

I am trying to open an entry point module file in the "design" view as
per
http://code.google.com/webtoolkit/tools/gwtdesigner/quick_start.html

However, I am getting the following message
"64-bit Windows is not supported for this version of GWT. Please use
32-bit Eclipse and a 32-bit Java VM."

Is there a workaround to this problem? Apart from installing on a
32bit environment?

Thanks
-Venu

My environment : Windows 7 Ultimate 64 bit with Springsource ToolSuite
2.5.1, GWT 2.1

Re: RequestFactory issue with overloaded setters in domain type

On Thursday, February 28, 2013 1:50:03 PM UTC+1, James Horsley wrote:

I'll take a stab at writing a ServiceLayerDecorator as that should get me unblocked without writing DTO's and if the correct setter is called the NPE's won't be an issue. Thanks to you both for the suggestions!

Thomas, is this issue likely to end up as PatchesWelcome or WontFix? If the former then I'll look at getting a patch together.

I you send a patch, I'll look at it; the outcome will depend on what changes the patch has to make.

Re: send event (click) to an external html page

ARRRGGG !

Thank you all for your search and help.

I don't know how to workaround my problem.

So, I'll search (alone, again, ;-)
If I find anything, I'll tell you

Best,

Laurent

Le mercredi 27 février 2013 17:59:06 UTC+1, Milan Cvejic a écrit :

Thanks Thomas for clarification of messaging api. I believed that it is working only on same domain and that it is build just to standardize communication.

Best,
Milan

On Wed, Feb 27, 2013 at 3:24 PM, Thomas Broyer <t.br...@gmail.com> wrote:

On Wednesday, February 27, 2013 3:07:48 PM UTC+1, Milan Cvejic wrote:
On the other hand, if you are loading iframe from the same domain, and have full control of iframe content you can use window.postMessage api for child - parent communication.

If it's from the same origin, you don't need postMessage. postMessage was specifically made for cross-origin communications.

See http://dev.w3.org/html5/postmsg/#web-messaging and http://www.whatwg.org/specs/web-apps/current-work/multipage/web-messaging.html#web-messaging for the spec, and http://html5demos.com/postmessage2 for a demo.

You will need to implement some JSNI methods in order to accomplish this in GWT.

or use Elemental.

I doubt it'll solve Laurent's use case though: he was talking about loading a third-party site, and never mentioned that site was "friendly" to the "calling GWT app" (he used Google as an example)

--
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/xyIjJvWyRD4/unsubscribe?hl=en.
To unsubscribe from this group and all its topics, send an email to google-web-toolkit+unsubscribe@googlegroups.com.
To post to this group, send email to google-we...@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.

--
LiquidBrain!

Re: RequestFactory issue with overloaded setters in domain type

I'll take a stab at writing a ServiceLayerDecorator as that should get me unblocked without writing DTO's and if the correct setter is called the NPE's won't be an issue. Thanks to you both for the suggestions!

Thomas, is this issue likely to end up as PatchesWelcome or WontFix? If the former then I'll look at getting a patch together.

On 28 February 2013 11:19, Thomas Broyer <t.broyer@gmail.com> wrote:

On Thursday, February 28, 2013 11:18:22 AM UTC+1, Ignacio Baca Moreno-Torres wrote:

If you don't want to have a delegator DTO, you may try AOP. For example, with guice you can match the classes you have the NPE problem, and return null before de NPE.

or in this specific case, use a ServiceLayerDecorator whose getSetter returns the appropriate setter for the given known class and property name (or possibly do the right thing in setProperty, without the need to go down to the java.lang.reflect level)

--
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+unsubscribe@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: How to align DockLayoutPanel:east content floated to the right?

Especially I'm looking for this:
           <g:east size='10'>
                    <g:FlowPanel styleName="{style.right}">
                          <g:ToggleButton text="toggleMe" width="60px" />
                    </g:FlowPanel>
                </g:east>

This would still align the togglebutton to the left edge of the <g:east> panel. Why?

How to align DockLayoutPanel:east content floated to the right?

How can align the content of the east panel docked to the right edge?
Using the following would just align everything in it's panel to their left edge:

<g:DockLayoutPanel unit="EM">
<g:west size="10"><g:FlowPanel styleName="{style.left}" >
<g:east size="10"><g:FlowPanel styleName="{style.right}" >
<g:center />

.left { float: left; }
.right { float: right; }

Re: RequestFactory issue with overloaded setters in domain type

On Thursday, February 28, 2013 11:18:22 AM UTC+1, Ignacio Baca Moreno-Torres wrote:

If you don't want to have a delegator DTO, you may try AOP. For example, with guice you can match the classes you have the NPE problem, and return null before de NPE.

or in this specific case, use a ServiceLayerDecorator whose getSetter returns the appropriate setter for the given known class and property name (or possibly do the right thing in setProperty, without the need to go down to the java.lang.reflect level)

Re: PopupPanel.hide() steals focus from TextBox?

Tried this but does not work for me. Only what helps is disabling the animation...

2013/2/27 <rjfleck@gmail.com>

I had a similar problem with a DialogBox
setModal(false) fixed it.

--
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+unsubscribe@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: RequestFactory issue with overloaded setters in domain type

If you don't want to have a delegator DTO, you may try AOP. For example, with guice you can match the classes you have the NPE problem, and return null before de NPE.

On Feb 27, 2013, at 6:30 PM, James Horsley <james.horsley@gmail.com> wrote:

Issue filed https://code.google.com/p/google-web-toolkit/issues/detail?id=8039

On 27 February 2013 17:04, James Horsley <james.horsley@gmail.com> wrote:

Thanks for the quick and detailed reply.

Given a big win using RequestFactory is having it work well with existing server side data models it'd seem like something we'd want to fix, relative to other priorities of course. In this specifc case I don't own the domain class as it's from a 3rd party and it's been fantastic with RF to just use a proxy interface to pass the data across without another DTO. I can extend it in this specific case to override the behaviour of setUnit(SomeUnitEnum) but that's not always possible and I'm still at the mercy of the order the methods are iterated in as to what type I have to use in my Proxy interface, this throws it into the want to fix case for me (not surprising perhaps given I'm the one asking the question).

I'll file an issue for it and am fully behind the fact that if I think it's important that I should get a patch out for it so will try to do that. The geTop().getGetter(domainClass, property) suggestion you made sounds like a quick and mildly dirty way through; although I may look at the first fallback based solution you mentioned.

Cheers,
James

On 27 February 2013 16:05, Thomas Broyer <t.broyer@gmail.com> wrote:

On Wednesday, February 27, 2013 4:05:30 PM UTC+1, James Horsley wrote:

I'm running into an issue where RequestFactory isn't calling the correct setter on the domain object in DevMode and I've not tried it compiled yet. Looking at the code it appears that ReflectiveServiceLayer.getBeanMethod is just picking the first setter it finds based on SET/SET_BUILDER, are only verifying that the method is a setter, and the method name.

I was going to submit an issue for it and see about getting a patch together but thought I'd ask here first to see if I'm missing something.

A more concrete example that I'm running into is that the domain class has something like the following:

DomainClass.java
void setUnit(String unit) { this.unit = unit; }
void setUnit(SomeUnitEnum unit) { this.unit = unit.toString(); }

ProxyInterface.java
void setUnit(String unit);

String getUnit();

when unit is null, the server side RF code is calling the enum version which is causing an NPE.

Anything I'm missing or should just file the issue?

Fixing this would require breaking changes in ServiceLayerDecorator though (getSetter would need a new expectedType parameter, similar to setProperty) so I wonder how best we could provide backwards compatibility (ideally, getSetter(Class,String,Class) would fallback to getSetter(Class,String) (to ensure backwards compatibility with custom ServiceLayerDecorators) but we'd then like ReflectiveServiceLayer#getSetter(Class,String) to answer based on the Class originally passed to getSetter(Class,String,Class). Alternatively, ReflectiveServiceLayer#getSetter could call getTop().getGetter(domainClass, property) and then use the getter's return type to disambiguate overloaded setters. That would then make it impossible to have proxies with setter-only properties (see https://code.google.com/p/google-web-toolkit/issues/detail?id=5760, there's much to be done before this can be supported, so the server-side implementation might be revisited at that point, i.e. not a showstopper)

The question is: is this something we really *want* to fix? And honestly, I don't have a problem rejecting overloaded setters (see also the comment on https://code.google.com/p/google-web-toolkit/issues/detail?id=6587: if it weren't for protobuf where the idiom is common, having "overloaded" hasXxx and getXxx getters would have been declared illegal).

Feel free to file an issue, but be warned that I'll categorize it as PatchesWelcome.

--
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+unsubscribe@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.

--
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+unsubscribe@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.

Wednesday, February 27, 2013

Re: Problem use GWT 2.5.0 Super Dev Mode

See https://groups.google.com/group/google-web-toolkit/browse_thread/thread/efc1e5e3b7853cac/0d745bfee422f8eb

2013/2/27 emurmur <emurphy@conceptuamath.com>

Here is a blog post I wrote on getting the code server running with the Elemental library. You might find stuff in it that is useful.

http://lumpofcode.blogspot.com/2012/11/configuring-gwt-codeserver-and.html

On Wednesday, February 27, 2013 1:09:59 AM UTC-8, Dominic Warzok wrote:
Hey I try to use Super Dev Mode in my web app.

But I always get the message
"Can't find any GWT Modules on this page. code server"

I follow these Steps to get Super Dev Mode running in eclipse:

1. add <add-linker name="xsiframe" />

<set-configuration-property name="devModeRedirectEnabled" value="true" />

to my webapp.xml

2. build a new Java configuration "Run as Java App"

3. Start up code server Output:
A lot of compiling packages and this:

Source Maps Enabled
Compile of permutations succeeded

Linking into C:\Users\warzok\AppData\Local\Temp\gwt-codeserver-2196663031883485286.tmp\com.SchumannGmbH.cam.web.Web_CAM\compile-1\war\web_cam; Writing extras to C:\Users\warzok\AppData\Local\Temp\gwt-codeserver-2196663031883485286.tmp\com.SchumannGmbH.cam.web.Web_CAM\compile-1\extras\web_cam

Link succeeded
Compilation succeeded -- 50,274s

Compile completed in 50873 ms
SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
SLF4J: Defaulting to no-operation (NOP) logger implementation

SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

The code server is ready.

Next, visit: http://localhost:9876/

4. Run normal dev mode with these programm arguments:

-war "${workspace_loc:Web-CAM/war}" -remoteUI "${gwt_remote_ui_server_port}:${unique_id}" -startupUrl Web_CAM.html -logLevel INFO -codeServerPort 4321 -port 8888 com.SchumannGmbH.cam.web.Web_CAM

Maybe here is the error ?!

5. Go to http://localhost/myWebApp.html

6. Click Dev Mode On

If I set up a Testproject this will work but not with my bigger webapp,

Any sugesstions

Thanks in advance .

--
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+unsubscribe@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.

Nick Cannon, Catalaya , and 6 others have Tweets for you

Nick Cannon @NickCannon

Exceptional work @TheBestManager! celeb.bz/Wghx8v

26 Feb

Guns for Greatness

Micahel Blue Williams, the head of Family Tree Entertainment, proposes a gun buyback program that would reward New Yorkers who trade in their guns with mentorship and Beyonce concert tickets for The M …