Re: Custom Post Linker and Super Dev Mode

[Geoffrey Wiseman]

On Tuesday, January 28, 2020 at 4:48:32 AM UTC-5, Thomas Broyer wrote:
>
> Those methods are to be called from the client code, not the linker itself.
>

Ok -- makes sense. If I modify the client code that makes those requests to 
use getModuleBaseForStaticFiles(), it looks like it's now trying to load 
them from the code server port instead -- assume that's expected?  If I log 
the output of both methods, I can see the ports are different.

Unfortunately, the new base URL doesn't seem to have helped me. Now I see 
"[WARN] ignored get request: /path/to/file" for the URL and if I make a 
curl request, I can see that it's coming back with a 200 and content length 
0.

So  it's different, but not totally working either?

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit/8f3c1799-877d-4779-b4f7-abbb1f5ca9a1%40googlegroups.com.

Re: Custom Post Linker and Super Dev Mode

[Geoffrey Wiseman]

On Wednesday, January 15, 2020 at 3:49:34 AM UTC-5, Thomas Broyer wrote:
>
> IIRC, with Super Dev Mode, public artifacts aren't generated in the 
> launcher dir; you need to build your path relative to 
> GWT.getModuleBaseForStaticFiles() 
> 
>  
> rather than GWT.getModuleBase(). HTH
>

Cautionary preamble: I wasn't involved in the creation of this linker, and 
I haven't done much with GWT linkers, so it is entirely possible that I'm 
missing something here. Having said all that:

The linker that I've seen break down in SuperDevMode just calls 
`emitString` to create some text resources that are then requested later by 
the browser. Those resources were visible to dev mode in the past, and 
they're visible in the built GWT application, but in SuperDevMode, they 
404. Near as I can tell, the linker isn't using `getModuleBase` or 
`getModuleBaseForStaticFiles` at all.

But with this response I was able to find another thread:
https://www.mail-archive.com/google-web-toolkit@googlegroups.com/msg105012.html

Sounds like you were saying there that it may not be possible to fix this 
for SuperDevMode? It definitely *does* work in the final built WAR file, 
it's only SuperDevMode where this seems to break down. So if I create 
artifacts using `emitString` in a custom linker, there's no way to have 
these be accessible in SuperDevMode?



-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit/c81f5a6e-b776-4b76-ba54-34e5e4a7f98b%40googlegroups.com.

Re: Custom Post Linker and Super Dev Mode

[Geoffrey Wiseman]

I know this is a while back -- did you ever make any headway on this? I've 
seen this on a project and it wasn't a pressing concern, but if anyone has 
a fix for this, I'd be happy to hear it.

On Tuesday, February 20, 2018 at 11:55:59 AM UTC-5, MCMicS wrote:
>
> Hi,
> I have written an own linker to generate file.
> If I compile the App it works and the legacy DevMode works also fine.
>
> But If I use SuperDevMode the file is not generated.
>
> The reason ist that the SuperDevListener has following method:
>   @Override
>   public void writeCompilerOutput(StandardLinkerContext linkerStack, 
> ArtifactSet artifacts,
>   ModuleDef module, boolean isRelink) throws UnableToCompleteException 
> {
> // The code server will do this.
>   }
>
> The linker is called but no artifact will be written.
>
> How can I reach to generate a file (index.html, readme, ...) during 
> compile step
>
> Regards
> Michael
>
>

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit/0f68e354-18c1-494c-baad-d7eb58d76e4b%40googlegroups.com.

Re: [gwt-contrib] Re: Maven-ization Status

[Geoffrey Wiseman]

On Sunday, September 29, 2013 8:51:47 PM UTC-4, Matthew Dempsky wrote:

 Anyway, I'm of the same opinion here as Thomas: I want to make it easy for 
 developers to use GWT in their projects and to contribute to GWT itself.  I 
 supported switching to Maven as a means to this end, but I'm no fan of it 
 beyond that, and I don't think any other core GWT contributors are at this 
 point.


 To summarize: if anyone still feels strongly that GWT should use Maven, 
 those individuals need to roll up their sleeves and work on making it 
 happen.


I neither love or hate Maven. It's one of the best supported build tools in 
Java, and I end up using it for that reason as much as anything else. It 
simplifies some things and makes other things more complex. If one of its 
alternatives got really popular, I would probably try it. In the end, I 
probably prefer Maven to Ant, mainly because Maven projects /tend/ to be a 
little more likely to work out of the box without configuration on an 
individual developer's machine, but that's more to do with how people use 
Ant than with Ant itself. 

That said, I agree with a bunch of the comments here that getting GWT to 
the point where someone can contribute to it without days of setup is key; 
I once tried to contribute a patch to something, but after a few hours of 
not getting the project fully set up, I gave up and went back to what I was 
actually working on. If you can reduce the barrier to entry so that it's 
not hard to contribute, even if that means installing Gradle or Buck, I 
think the problem is solved. If you moved to Maven and still had immense 
setup hurdles, the problem still isn't solved, AFAICS.

So, I'd just be happy if that barrier to entry could be reduced -- reduced 
setup, increased modularity, simpler out-of-the-box configuration, etc. 
That'd make it easier for people like me to be more involved.

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

Re: GWT App -- Firefox 20 regression?

[Geoffrey Wiseman]

On 2013-04-09, at 8:20 AM, Jens jens.nehlme...@gmail.com wrote:

 
 Probably not related but hit a weird issue with FF20 too:
 * undefined in a variable that the Java source (compiled to JS) says should 
 be initialised.
 * works fine in all other browsers.
 * works in FF if firebug or the built in debugger turned on...
 
 Sounds like a JIT Compiler issue especially because it goes away when a JS 
 debugger is active which often (always?) disables JIT compilation in 
 browsers. You can try disabling the JIT compiler in about:config.

Out of curiosity, I tried this with my issue, and no dice. I disabled all four 
of the flags containing 'jit' in about:config. Refreshing, quitting didn't 
solve the problem, so I did another --safe-mode reset in case it helped clear 
some sort of jit-compiler cache, and went back through the process I use to 
trigger my problem (load up dev mode, go through a sequence in my app) and 
voila, the problem resurfaced as always.

So -- if there is a JIT bug, it either doesn't cause the problem I'm 
experiencing, or turning off those four flags is insufficient to stop it from 
recurring.

  - Geoffrey
--
Geoffrey Wiseman
geoffrey.wise...@gmail.com
http://www.geoffreywiseman.ca

-- 
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: GWT App -- Firefox 20 regression?

[Geoffrey Wiseman]

On 2013-04-09, at 10:31 AM, Jens jens.nehlme...@gmail.com wrote:

 I think you have to set javascript.options.ion.content to false to disable 
 IonMonkey (the name of the JIT Compiler in FF) in FF20.

It briefly seemed like this might have had an impact (I failed to reproduce the 
same error state with the Ion Monkey flags set to false), but then I couldn't 
reproduce it after a reset with the ion monkey flags turned on, and discovered 
that /other/ selects are now similarly broken, and that that's true with the 
Ion Monkey flags set to false as well. So at this point, all that I'm learning 
is that reproducing the error state is less reliable than it had seemed before, 
although I still seem quite able to reproduce, but with less consistency as to 
the when and where. ;)

I'm going back to getting work done in FF19; I'll keep an eye out for FF20.0.1 
or more signs that the cause might have been found.

  - Geoffrey
--
Geoffrey Wiseman
geoffrey.wise...@gmail.com
http://www.geoffreywiseman.ca

-- 
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: GWT App -- Firefox 20 regression?

[Geoffrey Wiseman]

On Friday, April 5, 2013 4:49:20 PM UTC-4, Geoffrey Wiseman wrote:

 I hit an issue this afternoon where I have an unexplained issue with a 
 select where I can't select any of the options in it, and it seems like the 
 handler isn't even being called (or isn't triggering the debugger in 
 Eclipse, anyway), in either dev mode or after deployment in regular browser 
 usage (including in FF safe mode, which disables all the plugins). This is 
 true in FF20, but not in FF19. 

 It's only one select out of many in our application, so there are clearly 
 complicating circumstances, it's not just select controls don't work in 
 FF20/GWT, but it's odd to me that it works in FF19 and doesn't work in 
 FF20. Since I don't have any idea of the cause yet, I'm not sounding 
 alarms, but I thought I'd put a placeholder in here in case anyone else is 
 experiencing a FF19/FF20 regression and is farther along in their analysis 
 than I am.


I'm having a hard time pinning this down; I can make it go away by 
resetting my firefox profile, but it comes back. I haven't reproduced it on 
another machine, but since I don't have a known sequence of steps that I 
use to trigger the problem, I'd probably have to use another machine as a 
primary machine for a while to see if that helps.

I can trigger this with a simple jsfiddle test:
http://jsfiddle.net/codiform/LgvRb/8/

Unfortunately, this test works fine on a clean install of Firefox, or after 
resetting the profile, so if you're not already experiencing the problem 
that I'm experiencing, you probably won't experience it by visiting that 
page.

I've posted to the Mozilla questions site to see if anyone there is 
encountering it:
https://support.mozilla.org/en-US/questions/955922#answer-425122

Anyway -- I don't have any real answers to what's going on, and I don't 
feel like I'm making a lot of headway on figuring out the cause, so I'm 
probably going to just go back to FF19 for now, but I thought I'd post an 
update in case someone else experiences this.

-- 
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.

GWT App -- Firefox 20 regression?

[Geoffrey Wiseman]

I hit an issue this afternoon where I have an unexplained issue with a 
select where I can't select any of the options in it, and it seems like the 
handler isn't even being called (or isn't triggering the debugger in 
Eclipse, anyway), in either dev mode or after deployment in regular browser 
usage (including in FF safe mode, which disables all the plugins). This is 
true in FF20, but not in FF19. 

It's only one select out of many in our application, so there are clearly 
complicating circumstances, it's not just select controls don't work in 
FF20/GWT, but it's odd to me that it works in FF19 and doesn't work in 
FF20. Since I don't have any idea of the cause yet, I'm not sounding 
alarms, but I thought I'd put a placeholder in here in case anyone else is 
experiencing a FF19/FF20 regression and is farther along in their analysis 
than I am.


-- 
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: GWT Multi-Module Server Classpath Problem

[Geoffrey Wiseman]

On Friday, 10 August, 2012 at 10:34 AM, Thomas Broyer wrote:

 Are the JARs exported by the common Eclipse project?

They aren't -- both M2E or eclipse:eclipse handles transitive dependencies by 
adding the JARs to the class path of the 'child' project, so you don't need the 
'parent' to export, AFAICS.

Over the weekend, I discovered that the child project was showing up as a WST 
project; I haven't done much with WST so I don't know what its builders and 
natures do, and the Deployment Assembly seemed like it might be getting in the 
way. When I took those natures and builders out of my .project so that it more 
closely matched the single-project version I'd been using previously the SLF4J 
errors went away, but only because my problems got more severe. Now a launch 
configuration (even a completely new one) starts up without errors or warnings, 
but then 404s on all requests, including for the project's simple HTML page, 
e.g.:
[WARN] 404 - GET /Project.html?gwt.codesvr=127.0.0.1:9997 (127.0.0.1) 1395 bytes


I still don't feel like I have much of a sense of why this would be happening 
and in what way this project is different from other GWT projects that I can 
still launch from Eclipse, which is frustrating. The multi-project nature of 
this doesn't seem to explain all my problems (although perhaps it does).

The same project still launches successfully from a mvn gwt:run, so it's not 
entirely a lost cause, but I'd prefer the Google Plugin for Eclipse to work 
with it too.

  - Geoffrey-- 
Geoffrey Wiseman
http://www.geoffreywiseman.ca/

 
 -- 
 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/-/xZ4Ve-ZuQdAJ.
 To post to this group, send email to google-web-toolkit@googlegroups.com 
 (mailto:google-web-toolkit@googlegroups.com).
 To unsubscribe from this group, send email to 
 google-web-toolkit+unsubscr...@googlegroups.com 
 (mailto: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.

GWT Multi-Module Server Classpath Problem

[Geoffrey Wiseman]

I have a 2yo GWT project that's getting a little big for a single module, 
and we're bringing in a new project and we'd like to share some code.

Accordingly, I've split the project into a 'common' codebase and a 
project-specific codebase. That split isn't perfect yet, but I can't seem 
to get Eclipse to launch the project. If I launch the project through mvn 
gwt:run, it launches successfully, but if I launch it through Eclipse (Run 
As  Web Application) the server part fails to start up when a filter can't 
find the SLF4J LoggerFactory.

If I do some introspection on the classpath from the servlet, it seems like 
all of the JARs in the project aren't on the classpath -- only the classes 
and dependent projects are there.

The launch configuration classpath seems ok, and the command-line startup 
seems to work. It doesn't seem to matter if I use m2e (with its Maven 
Dependencies container) or mvn eclipse:eclipse, which adds the JARs to 
the project .classpath directly. I'm running low on ideas.

So:

   1. Anyone experience this or something like it and know what I should 
   look for?
   2. How does the server get its classpath -- from the launch 
   configuration classpath or something different?
   3. Is anyone using multiple eclipse projects with the Google Eclipse 
   Plugin successfully? (from other threads, it sounded like there were people 
   who had it working, but ...)

Any other thoughts or suggestions welcome.

-- 
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/-/ty-qiodo-4wJ.
To post to this group, send email to google-web-toolkit@googlegroups.com.
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.

GWT_CONTAINER will not be exported or published

[Geoffrey Wiseman]

When I use M2E and GWT together in Eclipse these days, I get this warning:
Classpath entry com.google.gwt.eclipse.core.GWT_CONTAINER will not be 
exported or published. Runtime ClassNotFoundExceptions may result.

It's not a big deal, but I'd rather get rid of it -- any suggestions?

  - Geoffrey

-- 
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/-/SUMvdp218okJ.
To post to this group, send email to google-web-toolkit@googlegroups.com.
To unsubscribe from this group, send email to 
google-web-toolkit+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.

Re: GWT_CONTAINER will not be exported or published

[Geoffrey Wiseman]

On Wednesday, 25 April, 2012 at 12:17 PM, Thomas Broyer wrote:
 When using M2E, the Google Plugin for Eclipse should automatically pick the 
 GWT SDK from the Maven dependencies, and as a result there shouldn't be a 
 GWT_CONTAINER entry in the build path.
  
 So, in your case, I'd try removing the GWT_CONTAINER entry from the build 
 path, and/or possibly use Maven → Update Project Configuration…
I'd done the latter, but I hadn't considered the idea of removing the 
container, so I hadn't tried that manually. Worked like a charm, and makes 
sense to boot. Thanks.

  - Geoffrey
--  
Geoffrey Wiseman
http://www.geoffreywiseman.ca/

-- 
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.

Exporting Instance Methods to Hand-Written Javascript

[Geoffrey Wiseman]

I have a piece of GWT code that I wanted to invoke from outside handwritten 
JavaScript. There's an example shown here under Calling a Java method from 
Handwritten JavaScript:
https://developers.google.com/web-toolkit/doc/latest/DevGuideCodingBasicsJSNI

That example uses a static method; I can make that work, but my first 
instinct was that i'd rather use an instance, so using the syntax shown 
under JSNI invocations, I tried:

/* package */ native void exportJavascriptMethods( PreviewPane x ) /*-{
   $wnd.imageSelected =
  $entry(this
@ca.cpp.spike.gwtapplet.client.PreviewPane::imageSelected
(Ljava/lang/String;Ljava/lang/String;));
  }-*/;

That doesn't work. Eclipse (and the GWT compiler) complain:
JavaScript parsing: Missing ) after argument list PreviewPane.java

If I replace 'this' with 'x', same problem. If I take 'this' out and leave 
it binding to the static method:

/* package */ native void exportJavascriptMethods( PreviewPane x ) /*-{
   $wnd.imageSelected =
  $entry(@ca.cpp.spike.gwtapplet.client.PreviewPane::imageSelected
(Ljava/lang/String;Ljava/lang/String;));
  }-*/;


All is well in the world.

So -- is that just not an option, exporting an instance method like that?

-- 
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/-/mC9EADg67lcJ.
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: Aspect Security / Sonatype Study - GWT Vunlerabilities

[Geoffrey Wiseman]

On Thursday, March 29, 2012 4:23:17 PM UTC-4, Joseph Lust wrote:

 They appear to be companies using antiquated software and GWT being called 
 out is a bit of sensationalist cry by the authors. For example, they place 
 in their chart GWT at the top, not GWT 1.6/7. That is to say that not 
 all GWT applications are vulnerable, just the really old, rot in place 
 ones. They also call out SpringMVC 2.5.6, while we're rocking on 3.0.10 
 these days.


Yeah, I wish I had more information on how they derived those numbers and 
what they mean. Do those numbers for GWT only include those versions that 
had reported vulnerabilities, and which ones? Being able to trace their 
chart back to specific details would be useful, because without that I'm 
not sure how much weight to put on their results.
 

 The gaping omission of the article is that most such *Global 500* firms 
 software development is for *internal components*. If at my office and 
 most others, we don't see an internal meeting scheduling app written in GWT 
 1.6 to be a serious issue. However, client/external facing applications are 
 a whole different can of beans which have many rounds of reviews before 
 release and continuing audits. I'd estimate only 5% of our applications are 
 externally visible, and the real number is likely lower than that.


Yes, that's certainly something to consider; although ideally you wouldn't 
be using components with known vulnerabilities internally, the risk of 
doing so is somewhat lower.
 

 The real take away message is that Maven needs an audit feature to check 
 your POM for known vulnerabilities, say at compile time.


Although it's often the technology management that wants the audit 
features, not the individual developer -- that's why the repository-level 
oversight has some appeal for some organizations, I think.

  - Geoffrey

-- 
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/-/NFXatA0yUpoJ.
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.

Aspect Security / Sonatype Study - GWT Vunlerabilities

[Geoffrey Wiseman]

This study by Aspect Security and Sonatype is making the rounds, and 
implies that GWT is the most-downloaded component in Maven central with 
security vulnerabilities:
http://www.sonatype.com/Products/Sonatype-Insight/Why-Insight/Mitigate-Security-Risks/Security-Brief

I've asked, but I'm curious which GWT vulnerabilities they might be 
including here.

-- 
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/-/j74s381Q2TUJ.
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: Aspect Security / Sonatype Study - GWT Vunlerabilities

[Geoffrey Wiseman]

On Wednesday, March 28, 2012 11:10:44 AM UTC-4, Geoffrey Wiseman wrote:

 This study by Aspect Security and Sonatype is making the rounds, and 
 implies that GWT is the most-downloaded component in Maven central with 
 security vulnerabilities:

 http://www.sonatype.com/Products/Sonatype-Insight/Why-Insight/Mitigate-Security-Risks/Security-Brief

 I've asked, but I'm curious which GWT vulnerabilities they might be 
 including here.


The one that comes up the most in searches for me is this (relatively 
ancient) GWT 1.5/1.6-era RSS/XSS vulnerability:
http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2010-4322
http://code.google.com/p/google-web-toolkit/issues/detail?id=3637

If they're using this one, I'm curious if their download stats only include 
affected versions.

-- 
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/-/eHOwW9yCMOQJ.
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: Re : Re: UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

On May 12, 4:13 pm, Thomas Broyer t.bro...@gmail.com wrote:
 It's not HTML vs. Widget, it's innerHTML vs. DOM's
 document.createElement/appendChild. So in GWT, as soon as you start making
 reusable components, you're breaking the innerHTML layer and have to
 introduce a Widget.

From what I've seen in most of the modern browsers, the difference
between innerHTML and DOM manipulation isn't vast (e.g.
http://andrew.hedges.name/experiments/innerhtml/).  I have to admit,
the presentations that have talked about performance differences
between using widgets and emphasizing html in GWT haven't been
incredibly clear on the reasons why -- other than talking about the
level at which events are handled.

Given that I don't have a performance problem that I'm trying to
solve, I'm happy to chalk this up to premature optimization.

 The point of HTML vs. Widget is: don't use FlexTable/Grid/FlowPanel for
 layout and Button/TextBox/etc. if you can simply use an HTMLPanel with HTML
 inside (and event delegation for event handling)

Sure. From the limited descriptions I'd seen, I imagined that a
reusable component containing HTML might offer the same benefit over a
reusable component containing GWT widgets. That might not be true.

-- 
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: UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

On May 12, 3:19 pm, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 On May 12, 1:22 pm, Gal Dolber gal.dol...@gmail.com wrote:
  In real its ok what you are doing(in theory Widget should be used only when
  you need events), but the truth is that gwt's panels don't have any support
  for UiObjects so you need to use Widget.

 So even HTMLPanel can't easily absorb HTML once you bundle it up as a
 GWT UiBuilder component.  Ah well.

Coming back to this point -- the UiBinder documentation shows bundling
HTML as a UiObject (http://code.google.com/webtoolkit/doc/latest/
DevGuideUiBinder.html#Hello_World) and the Google Plugin for Eclipse
has an option to allow you to do this, but there doesn't seem to be
much you can do with it having done so: all panels from RootPanel on
down require you to insert a widget.

So why would you create a UiObject component using UiBinder?

-- 
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: UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

On May 13, 11:12 am, Thomas Broyer t.bro...@gmail.com wrote:
 ...when you need an UIObject ;-)

Thanks! Helpful food for thought.

It definitely sounds like I was trying to make an optimization I don't
need based on advice I didn't fully understand. ;)

-- 
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.

UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

Just to give a little background, I've been doing GWT for a while, but
I'm pretty new to UiBinder.  I'm starting to fold it in to some of the
GWT work I'm doing, and I certainly like some aspects of it. I'm
building a page with a lot of repeating display elements that don't
require any events (a simple dashboard with some stats), so I started
thinking that I'd use HTML elements instead of widgets, and then I ran
into some troubles on how to do that with UiBinder.

I have a component representing the dashboard page (more specifically,
the 'body' part of the page leaving out the template and navigation),
and within that a bunch of stat widgets. I originally laid it out as a
FlowPanel, with stat widgets, e.g.:

!DOCTYPE ui:UiBinder SYSTEM http://dl.google.com/gwt/DTD/xhtml.ent;
ui:UiBinder xmlns:ui='urn:ui:com.google.gwt.uibinder'
xmlns:g='urn:import:com.google.gwt.user.client.ui'
xmlns:w=urn:import:myproject.client.widget
g:FlowPanel
g:LabelGeneral/g:Label
w:Stat heading=apps in-force ui:field=appsInForce /
/g:FlowPanel
/ui:UiBinder

Once I started the Stat component, I realized I didn't need it to use
any widgets, that HTML would be ok, so I developed it that way:

public class Stat extends UIObject {

private static StatUiBinder uiBinder =
GWT.create( StatUiBinder.class );

@UiField
SpanElement heading;

@UiField
SpanElement value;

interface StatUiBinder extends UiBinderElement, Stat {
}

public Stat() {
setElement( uiBinder.createAndBindUi( this ) );
}

public void setHeading( String heading ) {
this.heading.setInnerText( heading );
}

public void setValue( String value ) {
this.value.setInnerText( value );
}

}

!DOCTYPE ui:UiBinder SYSTEM http://dl.google.com/gwt/DTD/xhtml.ent;
ui:UiBinder xmlns:ui=urn:ui:com.google.gwt.uibinder
ui:style
.statHeading {
}
.statValue {
}
/ui:style
div
span class={style.statHeading} ui:field=heading/span
span class={style.statValue} ui:field=value/span
/div
/ui:UiBinder

GWT then complained that I was calling add(Widget) with a UIObject,
which made sense -- I can't put HTML elements into a flow panel. So I
switched the container to an HTMLPanel, switched the label for a Div:

!DOCTYPE ui:UiBinder SYSTEM http://dl.google.com/gwt/DTD/xhtml.ent;
ui:UiBinder xmlns:ui='urn:ui:com.google.gwt.uibinder'
xmlns:g='urn:import:com.google.gwt.user.client.ui'
xmlns:w=urn:import:myproject.client.widget
g:HTMLPanel
divGeneral/div
w:Stat heading=apps in-force ui:field=appsInForce /
/g:HTMLPanel
/ui:UiBinder

Now I'm getting this message:
[ERROR] [myproject] - Line 23: The method
addAndReplaceElement(Widget, Element) in the type HTMLPanel is not
applicable for the arguments (Stat, Element)

I'm confused as to why it's trying to treat the Stat as a Widget
(which it isn't), so I'm missing something about UiBinder or HTMLPanel
here. I did a few google searches, didn't see anything obvious.

Help appreciated.

-- 
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: UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

On May 12, 1:07 pm, Thomas Broyer t.bro...@gmail.com wrote:
 What I don't understand is why Stat is not a Widget if you want to use it
 like a Widget.

[nod] I've changed it to being a widget in the meantime.

Basically, I need a reusable piece of display content, but it doesn't
need any events, so it seemed like using HTML would be 'lighter' than
using widgets. Certainly some of the best practice presentations
called out that using too many widgets where you could use HTML was
likely to pose performance problems in the long run.

I think for this page, a widget won't pose any real problems, so I'll
just switch to doing that, but if I were doing something obviously
wrong, then I'd rather know what that is rather than just ignore it
and do something different.

  - Geoffrey

-- 
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: UiBinder and child HTML 'widgets'

[Geoffrey Wiseman]

On May 12, 1:22 pm, Gal Dolber gal.dol...@gmail.com wrote:
 In real its ok what you are doing(in theory Widget should be used only when
 you need events), but the truth is that gwt's panels don't have any support
 for UiObjects so you need to use Widget.

So even HTMLPanel can't easily absorb HTML once you bundle it up as a
GWT UiBuilder component.  Ah well.

  - Geoffrey

-- 
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.

View-Presenter Interaction Patterns in Google Web Toolkit (GWT)

[Geoffrey Wiseman]

If you're using Model-View-Presenter with GWT (or you're thinking
about it), there are a number of different patterns you can use to co-
ordinate between your view and the corresponding presenter. Ray Ryan's
Best Practices for Architecting your GWT App at I/O 2009 showed off
one pattern, Daniel Danilatos refined that for his GWT Testing Best
Practices presentation during I/O 2010, and a third pattern was used
in the Large-Scale Application Development and MVP article in the
GWT wiki.  Using these and my own experiences with these patterns,
I've taken a small example, demonstrated how to apply each of these
patterns to that example, and written up some of the advantages and
disadvantages.

http://blog.codiform.com/2011/03/view-presenter-interaction-patterns-in.html

These patterns are by and large available already in the sources
above, but by bringing them together and talking about some of the
pros and cos, I hope to save some people some of the sorts of
experimentation that many of us will have already gone through to
decide which one of these patterns works best for us.

-- 
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: View-Presenter Interaction Patterns in Google Web Toolkit (GWT)

[Geoffrey Wiseman]

On Mar 1, 3:14 pm, Brian Reilly brian.irei...@gmail.com wrote:
 It looks to me like Guit leans toward the Abstracted Control View
 approach, as indicated by the @ViewField annotations on the HasText
 fields.

That would have been my reading too, although it seems as if the
abstracted controls are injected into the presenter (proxies?), and
with the handler methods, there's touches of the 'View Delegate'
pattern as well, I agree.

It's not an exact match for any of the patterns I've described, but
I'd say there's enough in common to draw some strong parallels.  I'm
actually quite curious to dig further into some of these MVP
frameworks for GWT now and get a better sense of how they match up.

  - Geoffrey

-- 
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.

Building Testable Applications in Google Web Toolkit (GWT)

[Geoffrey Wiseman]

I've been polishing my thoughts on testing and GWT for a while, but
I've finally posted them to the Codiform blog:
http://blog.codiform.com/2011/01/building-testable-applications-with.html

Those of you who've been using GWT for a while and come from a
developer-test-automation (TDD or otherwise) background may not find
that much new here, but I'm hopeful that it will help someone starting
out with GWT to find a style for building testable applications with
GWT more rapidly.

  - Geoffrey

-- 
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: Debug problem since GWT 2.1.0, gwt-maven-plugin 2.1.0, Maven 3.0

[Geoffrey Wiseman]

On Dec 2, 3:53 pm, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 I'm guessing this is caused by this filed 
 issue?http://code.google.com/p/google-web-toolkit/issues/detail?id=4853

Repeating what I said there on that issue:

Near as I can tell, these INDEX.LIST are in gwt-servlet.jar and gwt-
user.jar from GWT 1.6 through to GWT 2.1.0.  With that in mind, I
don't think they're the only cause of the current issues with
debugging in Eclipse, although they may be partially responsible.  I
suspect there's also a recent regression with the Google Plugin for
Eclipse.

See files below:

v.2.0:
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.0.0/gwt-servlet-2.0.0.jar/!/META-INF/INDEX.LIST

v2.0.1:
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.0.1/gwt-servlet-2.0.1.jar/!/META-INF/INDEX.LIST

v2.0.2:
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.0.2/gwt-servlet-2.0.2.jar/!/META-INF/INDEX.LIST

v2.0.3:
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.0.3/gwt-servlet-2.0.3.jar/!/META-INF/INDEX.LIST

v2.0.4:
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.0.4/gwt-servlet-2.0.4.jar/!/META-INF/INDEX.LIST

v2.1.0
https://repository.sonatype.org/service/local/repositories/central-proxy/archive/com/google/gwt/gwt-servlet/2.1.0/gwt-servlet-2.1.0.jar/!/META-INF/INDEX.LIST

-- 
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: NullPointerException in PassthroughParser

[Geoffrey Wiseman]

On Dec 7, 11:32 pm, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 Unfortunately, I have no idea how it gets in that state, since I'm
 unable to reproduce it in a simple GWT Test Case, nor find a hole in
 the GWT code that clearly identifies how a null could get in there.
 setText() seems well-protected and setValue() clearly invokes
 setText().

 I don't have the more recent stack trace copied down at the moment,
 but I may try and regress our codebase to the point at which either
 one of these errors was evident so that I can try and reproduce it.
 I'm very curious now, much more so than when I thought it was just a
 simple bug that we had a workaround for.

Ok, so I feel better.  It's essentially our bug (although you could
argue that it's a bug in the subclassability of TextBox, perhaps).

Essentially, we have a subclass of TextBox that we use for displaying
guide text when the text box is empty.  If the guide text is applied
and getText() is called, it returns null rather than .  It seems
like changing that to  would be sufficient to make the bug go away
and probably won't introduce any weird side effects.

That said, it seems like either PassthroughParser should check for
null, or TextBox should be slightly more clear about the contract
expected for getText() for subclassors.  But that's pretty minor, and
I'll leave that to your judgement.  For now, I'm just happy to know
what the problem is and that it's not strictly a GWT bug.

Thanks for gently steering me in the right direction.

-- 
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.

NullPointerException in PassthroughParser

[Geoffrey Wiseman]

I'm seeing some NPEs in PassthroughParser when using TextBox,
particularly if you touch getValue(), setValue().  getValueOrThrow()
seems to pass a null right up to PassthroughParser.parse(), which
calls toString() without checking.  It happend once, we switched to
setText(), problem went away.  Happened a second time, seemed worth
digging a little deeper.

I can probably create a simple test case if that's helpful.  I'm not
seeing any filed bug reports or discussions, which surprises me.

I believe this is a regression in 2.1.0 that we didn't see in 2.0.4,
although my memory's a little fuzzy on that point.

I'm inclined to file it as a bug, but I thought I'd check here to see
if (a) anyone else has encountered his (b) it's expected, and I'm
doing it wrong (c) there's a discussion or filed bug that I missed.

Thanks,

  - Geoffrey

-- 
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: Debug problem since GWT 2.1.0, gwt-maven-plugin 2.1.0, Maven 3.0

[Geoffrey Wiseman]

I'm guessing this is caused by this filed issue?
http://code.google.com/p/google-web-toolkit/issues/detail?id=4853

-- 
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: Safari 5.0 with GWT 1.7.x?

[Geoffrey Wiseman]

On Jun 8, 10:47 am, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 Anyone tried Safari 5.0 with GWT 1.7.x to see if it solves the crash
 issue?
 (http://code.google.com/p/google-web-toolkit/issues/detail?id=4220)

Got bored of waiting.  Tried it myself.  Seems to work so far;
pleasing.

-- 
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.

Safari 5.0 with GWT 1.7.x?

[Geoffrey Wiseman]

Anyone tried Safari 5.0 with GWT 1.7.x to see if it solves the crash
issue?
(http://code.google.com/p/google-web-toolkit/issues/detail?id=4220)

-- 
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.

Invalid Memory Access - GWT 1.7.1 and OS X 10.6.3?

[Geoffrey Wiseman]

Anyone tried this combination?  I took down some software updates this
morning including 10.6.3, and voila, Invalid memory access when I
try to launch hosted mode:
http://emberapp.com/geoffrey.wiseman/images/gwt-hosted-mode-in-10-6-3

Either way, I'd like to know what's happening, but I'd also be happy
to hear from anyone else using this combination so that I know if
that's the problem.

I'm using:
Mac OS X 10.6.3
Eclipse 3.5 (20100218-1602)
Google Plugin for Eclipse 3.5 1.3.2.v201003242055
Google Web Toolkit 1.7.1v200909221731

-- 
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: Invalid Memory Access - GWT 1.7.1 and OS X 10.6.3?

[Geoffrey Wiseman]

On Apr 2, 11:41 am, kozura koz...@gmail.com wrote:
 Wouldn't happen to be this one, I ran into it to due to Safari
 upgrading past 4.0.3:

 http://code.google.com/p/google-web-toolkit/issues/detail?id=4220

 There's a workaround in comment 22 that worked.

Aw, hell.

I explicitly didn't take Safari 4.0.4 or Safari 4.0.5.  After this
surfaced, I checked software update to ensure that 'Safari 4.0.5' was
still visible, and it was, so I was sure I didn't take the upgrade.
Unfortunately, About Safari shows 4.0.4, so it looks like the OS X
10.6.3 may have moved me up to 4.0.4 anyway.

That's probably it then.

  - Geoffrey

-- 
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.

Safari 4.0.5 GWT 1.x?

[Geoffrey Wiseman]

Anyone tried Safari 4.0.5 with GWT 1.x Hosted Mode?  The latest
comment on the bug implies the issue hasn't been resolved, but thought
I'd check to see if there were more data points with people here:
http://code.google.com/p/google-web-toolkit/issues/detail?id=4220

-- 
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.

Making a Custom GWT Theme

[Geoffrey Wiseman]

IE uses a significantly larger font for font-size: small than most
of the other browsers.  I've been able to override the theme defaults
for much of the text in the GWT UI, but for some reason, IE7 doesn't
seem to want me to fix the font-size: small on the SELECT.  No
matter what I do in the CSS, the font stays larger, and the IE Dev
Toolbar informs me that it's using font-size:small from chrome.css.

As a result, I'm curious to try making a custom theme and see if that
makes controlling the CSS for IE7 slightly more tractable.  Has anyone
built a custom theme and documented the process?  I did some google
searches and found a few references to the idea, but nothing
comprehensive.

  - Geoffrey

-- 
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: Download, IE7 Download Blocker

[Geoffrey Wiseman]

On Feb 18, 10:15 pm, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 I have a GWT application that uses GWT-RPC to send data back and forth
 to the server.  I've got a button that downloads a PDF based on that
 data from the server.  PDF Generation works fine, and the download
 worked fine using Window.open().

 However, as it stood, the download would not include any data that had
 been changed on-screen; but if I add a GWT-RPC call before the
 download, then IE7 blocks the download because it decides that the
 download is not the direct result of user action (I guess because the
 Window.open happens after the async 'save my data' call).

I came to understand that IE7 would allow me to open the file (Content-
Disposition: inline) even in situations that it wouldn't let me
download it (Content-Disposition: attachment).  Although I was hoping
for the download, opening a PDF worked well enough to suffice for the
time being.

-- 
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.

Download, IE7 Download Blocker

[Geoffrey Wiseman]

I have a GWT application that uses GWT-RPC to send data back and forth
to the server.  I've got a button that downloads a PDF based on that
data from the server.  PDF Generation works fine, and the download
worked fine using Window.open().

However, as it stood, the download would not include any data that had
been changed on-screen; but if I add a GWT-RPC call before the
download, then IE7 blocks the download because it decides that the
download is not the direct result of user action (I guess because the
Window.open happens after the async 'save my data' call).

I've tried some other suggestions in the group's archive for iframes
and the like, but all of these are also blocked by the download
blocker.  Has anyone run into this and found a solution that they're
happy with?  Sadly, IE7 is the target browser for this client, so it's
a bit of an issue for me.

-- 
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.

Bug in Browsers that's Not in Hosted Mode

[Geoffrey Wiseman]

I'm just encountering my first piece of code that works fine in hosted
mode (GWT 1.7.1) but totally fails on every browser I try it in.  I
guess that means I'm going to have to debug the Javascript that I
didn't write, since I can't debug in hosted mode.  If someone's been
through this already and wants to pass on any tips, I'm all ears.  Are
there common reasons why this might happen?  Known Issues?

Just looking for general feedback, although I'm happy to return with
more information once I've tracked down the problem.

-- 
You received this message because you are subscribed to the Google Groups 
Google Web Toolkit group.
To post to this group, send email to google-web-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: Bug in Browsers that's Not in Hosted Mode

[Geoffrey Wiseman]

On Feb 8, 5:04 pm, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 Just looking for general feedback, although I'm happy to return with
 more information once I've tracked down the problem.

Ah, this is less surprising than I anticipated; I think it boils down
to a regular expression issue, which is definitely a source of known
surprises between hosted mode and browsers.

-- 
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: Installing GWT Eclipse plugin, version 1.1.0

[Geoffrey Wiseman]

On Dec 12, 1:08 am, Geoffrey Wiseman geoffrey.wise...@gmail.com
wrote:
 On Dec 11, 2:14 pm, Rajeev Dayal rda...@google.com wrote:

  We did not opt for this approach for simplicity's sake. Since we were
  planning on maintaining backwards compatibility with previous SDK versions,
  there was no good reason to provide older versions of the plugin for
  installation.

 That makes sense for the plugin; could you not have the update site
 list the current plugin and current-and-previous SDKs, allowing people
 the option of installing GPE 1.2 with SDK 1.7.1 after a few checkbox
 clicks?

   - Geoffrey

Is this worth filing as a desired enhancement somewhere?

--

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: Installing GWT Eclipse plugin, version 1.1.0

[Geoffrey Wiseman]

On Dec 11, 2:14 pm, Rajeev Dayal rda...@google.com wrote:
 We did not opt for this approach for simplicity's sake. Since we were
 planning on maintaining backwards compatibility with previous SDK versions,
 there was no good reason to provide older versions of the plugin for
 installation.

That makes sense for the plugin; could you not have the update site
list the current plugin and current-and-previous SDKs, allowing people
the option of installing GPE 1.2 with SDK 1.7.1 after a few checkbox
clicks?

  - Geoffrey

--

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.

Hibernate, DTOs and Moo

[Geoffrey Wiseman]

If you're using Hibernate with GWT, there's a solid writeup on your
available choices:
http://code.google.com/webtoolkit/articles/using_gwt_with_hibernate.html

If you're not entirely happy with any of those choices, do feel free
to take a quick peek at Moo:
http://github.com/geoffreywiseman/Moo

It's an approach I've used before, although this is the first time
I've put it into an open-source version that I could share.  It's
still pretty early days, so I haven't built significant documentation
other than the test cases and the README, but just wanted to share in
case it suited someone else better than Dozer or Gilead.  Feedback
welcome.


--~--~-~--~~~---~--~~
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: SerializationException is lost

[Geoffrey Wiseman]

On Sep 12, 7:39 am, Sripathi Krishnan sripathi.krish...@gmail.com
wrote:
 No - GWT doesn't propagate that exception/message to the client.

Is this a bug, tracked somewhere, or a deliberate decision that
doesn't currently make sense to me?  :)
--~--~-~--~~~---~--~~
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: SerializationException is lost

[Geoffrey Wiseman]

On Oct 8, 6:56 pm, Sripathi Krishnan sripathi.krish...@gmail.com
wrote:
 Deliberate decision ..

 Few reasons that I know of --

I buy the not transmitting the errors to the client part; what I don't
understand is why the server log doesn't seem to show anything useful
in this case, at least, where it's come up for me.
--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---

History in Hosted Mode on OS X

[Geoffrey Wiseman]

I've been doing some development of the project instrastructure to
support the GWT history mechanism, and in hosted mode on OS X as
launched from the latest 'Google Eclipse Plugin', the history tokens
don't show up (as in the URL doesn't change).  On friday they were
showing up some of the time, and today not at all.  I haven't been
able to ascertain any change that would have resulted in this
difference.

Further, when I compile the whole thing, run it in Jetty, and access
it from either Safari 4 or Firefox, the history tokens work fine.  So
it seems like there may be some quirk in supporting history in hosted
mode under OS X, but that's really all I've found at this point.  A
quick look around implied that this wasn't a known issue, but if I
missed something lemme know.

Alternately, if you've got history working in hosted mode under OS X,
I'd be interested to hear it.  I'm running Eclipse 3.5, Leopard
(10.5.X), Java 5.

  - Geoffrey
--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---

Hosted Mode - Keyboard Shortcuts?

[Geoffrey Wiseman]

Are there keyboard shortcuts in the hosted mode browser?  In
particular, for 'Refresh'?  ⌘R doesn't do the trick in OS X.  I
haven't had a lot of luck searching for GWT hosted mode browser
keyboard shortcuts in general.  There's nothing obvious in the menus.

--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---