Re: [WARN] Lost communication with remote process ?

[Ed Bras]

I decreased the Max heap size to 1756MB and the builds seem to go fine
currently.

On 24 January 2017 at 19:56, Ed Bras <post2edb...@gmail.com> wrote:

> Thanks Thomas,
> It probably has to do with OutOfMemory problems, but I thought I had it
> kind of solved.
> After the upgrade to 2.8 I had to increase the max memory from 1.5 to 2G,
> else I keep getting out of memory problems :(
> Currently I have, on a quad core:
>
> -Xms1G -Xmx2G -Xss1024k -Dgwt.persistentunitcache=false extraJvmArgs>
>
> 3
>
> However, I think the 2G might be too much too allocate on the server. I
> will play around with the settings, thanks.
>
> - Ed
>
> On 24 January 2017 at 18:02, Thomas Broyer <t.bro...@gmail.com> wrote:
>
>> A quick look at the code reveals that after the exception is logged
>> <https://github.com/gwtproject/gwt/blob/2.8.0/dev/core/src/com/google/gwt/dev/ExternalPermutationWorkerFactory.java#L154-L162>,
>> a TransientWorkerException
>> <https://github.com/gwtproject/gwt/blob/2.8.0/dev/core/src/com/google/gwt/dev/TransientWorkerException.java>
>> is thrown, and a TransientWorkerException means that compilation on a given
>> worker has failed and should be retried in another worker.
>> This explains why it works in the end.
>>
>> Now as to why the external worker seemingly closed the connection,
>> possibly an OufOfMemory.
>> You can enable debug logging to get the stdout and stderr of the forked
>> JVM.
>> Then try playing with the gwt.jjs.javaArgs system properties to tune the
>> forked JVM (memory if it's a memory issue).
>> Or you could possibly simply use fewer workers.
>>
>>
>> On Tuesday, January 24, 2017 at 4:01:47 PM UTC+1, Ed wrote:
>>>
>>> Anybody any idea what to do with the following error:
>>> I haven't seen this before, I see it since GWT 2.8. It occurs sometimes
>>> between permutation 15-16 and sometimes between 11-12 like the exception
>>> below.
>>> However, it ends with success, strange, and the application does work.
>>>
>>> I receive it when it's been build by Jenkins.
>>> Any idea what it is and how to solve it ?
>>>
>>> - Ed
>>>
>>> ---
>>>
>>> [INFO] Compiling permutation 10...
>>> [INFO]  Compiling
>>> [INFO] Compiling permutation 11...
>>> [INFO]   [WARN] Lost communication with remote process
>>> [INFO] java.io.EOFException
>>> [INFO]  at 
>>> java.io.ObjectInputStream$BlockDataInputStream.peekByte(ObjectInputStream.java:2601)
>>> [INFO]  at 
>>> java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1319)
>>> [INFO]  at 
>>> java.io.ObjectInputStream.readObject(ObjectInputStream.java:371)
>>> [INFO]  at 
>>> com.google.gwt.dev.ExternalPermutationWorkerFactory$ExternalPermutationWorker.compile(ExternalPermutationWorkerFactory.java:154)
>>> [INFO]  at 
>>> com.google.gwt.dev.PermutationWorkerFactory$Manager$WorkerThread.run(PermutationWorkerFactory.java:74)
>>> [INFO]  at java.lang.Thread.run(Thread.java:745)
>>> [INFO]   Compiling permutation 12...
>>> [INFO]  Compiling
>>> [INFO] Compiling permutation 13...
>>> [INFO]   Compiling permutation 14...
>>> [INFO]  Compiling
>>> [INFO] Compiling permutation 15...
>>> [INFO]   Compiling permutation 16...
>>> [INFO]  Compiling
>>> [INFO] Compiling permutation 17...
>>> [INFO]   Compiling permutation 18...
>>> [INFO]  Compiling
>>> [INFO] Compiling permutation 10...
>>> [INFO]Compile of permutations succeeded
>>> [INFO]Compilation succeeded -- 866.139s
>>> [INFO] Linking into /home/develop/gwt/work/prod/total/plus
>>> [INFO]Link succeeded
>>> [INFO]Linking succeeded -- 14.454s
>>>
>>>
>>>
>>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "GWT Users" group.
>> To unsubscribe from this topic, visit https://groups.google.com/d/to
>> pic/google-web-toolkit/zIYAvNMnGtU/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to
>> google-web-toolkit+unsubscr...@googlegroups.com.
>> To post to this group, send email to google-web-toolkit@googlegroups.com.
>> Visit this group at https://groups.google.com/group/google-web-toolkit.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

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

Re: [WARN] Lost communication with remote process ?

[Ed Bras]

Thanks Thomas,
It probably has to do with OutOfMemory problems, but I thought I had it
kind of solved.
After the upgrade to 2.8 I had to increase the max memory from 1.5 to 2G,
else I keep getting out of memory problems :(
Currently I have, on a quad core:

-Xms1G -Xmx2G -Xss1024k -Dgwt.persistentunitcache=false

3

However, I think the 2G might be too much too allocate on the server. I
will play around with the settings, thanks.

- Ed

On 24 January 2017 at 18:02, Thomas Broyer  wrote:

> A quick look at the code reveals that after the exception is logged
> ,
> a TransientWorkerException
> 
> is thrown, and a TransientWorkerException means that compilation on a given
> worker has failed and should be retried in another worker.
> This explains why it works in the end.
>
> Now as to why the external worker seemingly closed the connection,
> possibly an OufOfMemory.
> You can enable debug logging to get the stdout and stderr of the forked
> JVM.
> Then try playing with the gwt.jjs.javaArgs system properties to tune the
> forked JVM (memory if it's a memory issue).
> Or you could possibly simply use fewer workers.
>
>
> On Tuesday, January 24, 2017 at 4:01:47 PM UTC+1, Ed wrote:
>>
>> Anybody any idea what to do with the following error:
>> I haven't seen this before, I see it since GWT 2.8. It occurs sometimes
>> between permutation 15-16 and sometimes between 11-12 like the exception
>> below.
>> However, it ends with success, strange, and the application does work.
>>
>> I receive it when it's been build by Jenkins.
>> Any idea what it is and how to solve it ?
>>
>> - Ed
>>
>> ---
>>
>> [INFO] Compiling permutation 10...
>> [INFO]  Compiling
>> [INFO] Compiling permutation 11...
>> [INFO]   [WARN] Lost communication with remote process
>> [INFO] java.io.EOFException
>> [INFO]   at 
>> java.io.ObjectInputStream$BlockDataInputStream.peekByte(ObjectInputStream.java:2601)
>> [INFO]   at 
>> java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1319)
>> [INFO]   at 
>> java.io.ObjectInputStream.readObject(ObjectInputStream.java:371)
>> [INFO]   at 
>> com.google.gwt.dev.ExternalPermutationWorkerFactory$ExternalPermutationWorker.compile(ExternalPermutationWorkerFactory.java:154)
>> [INFO]   at 
>> com.google.gwt.dev.PermutationWorkerFactory$Manager$WorkerThread.run(PermutationWorkerFactory.java:74)
>> [INFO]   at java.lang.Thread.run(Thread.java:745)
>> [INFO]   Compiling permutation 12...
>> [INFO]  Compiling
>> [INFO] Compiling permutation 13...
>> [INFO]   Compiling permutation 14...
>> [INFO]  Compiling
>> [INFO] Compiling permutation 15...
>> [INFO]   Compiling permutation 16...
>> [INFO]  Compiling
>> [INFO] Compiling permutation 17...
>> [INFO]   Compiling permutation 18...
>> [INFO]  Compiling
>> [INFO] Compiling permutation 10...
>> [INFO]Compile of permutations succeeded
>> [INFO]Compilation succeeded -- 866.139s
>> [INFO] Linking into /home/develop/gwt/work/prod/total/plus
>> [INFO]Link succeeded
>> [INFO]Linking succeeded -- 14.454s
>>
>>
>>
>> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/
> topic/google-web-toolkit/zIYAvNMnGtU/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: GoogleBot is using the wrong permutation?

[Ed Bras]

Thanks, but I already have that. If I wouldn't have that, it would result
in another exception like I had in the past. Thanks.

On 29 February 2016 at 16:24, Michael Joyner  wrote:

> You need to have a "default" permutation for Google Bot:
>
> We use the following here:
>
> 
>  
>
> On 02/27/2016 10:10 AM, Rogelio Flores wrote:
>
> If you compile your app with the "PRETTY" compiler flag and then deploy it
> and run the same test, you might know exactly what "Kg" is (what its real
> name is on your Java code) which hopefully will let you know exactly what
> is failing and why.
>
>
> On Friday, February 26, 2016 at 2:29:21 AM UTC-7, Ed wrote:
>>
>> Ok, Still googlebot comes a long and after I deployed a new version I
>> always get google bot exceptions.
>>
>> I checked the permutation, and that often seem to be ok (not always).
>> That is: the one I receive as http header is the same as the one that I
>> send a long to the backend when an exception occurs.
>>
>> The error I get:
>> ---
>>
>> (TypeError) : Cannot call method 'Kg' of null
>>
>> at Unknown.vSb( https://www.
>> leuker.nl/app/aangifte/start/)
>>
>>
>>
>>
>>
>>
>> ---
>>
>> Any idea ?
>> I think it might come with the JS engine that googlebot uses. For
>> example: I don't support IE9, as I use JS features that aren't supported by
>> IE9 that result in similar errors.
>> But how to work around this? It's hard to see the exact error that occurs.
>>
>> - Ed
>>
> --
> You received this message because you are subscribed to the Google Groups
> "GWT Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/ixo_9kd-joU/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: GoogleBot is using the wrong permutation?

[Ed Bras]

Ofcourse that is possible, but that is not an option in production.

On 27 February 2016 at 16:10, Rogelio Flores 
wrote:

> If you compile your app with the "PRETTY" compiler flag and then deploy it
> and run the same test, you might know exactly what "Kg" is (what its real
> name is on your Java code) which hopefully will let you know exactly what
> is failing and why.
>
>
> On Friday, February 26, 2016 at 2:29:21 AM UTC-7, Ed wrote:
>>
>> Ok, Still googlebot comes a long and after I deployed a new version I
>> always get google bot exceptions.
>>
>> I checked the permutation, and that often seem to be ok (not always).
>> That is: the one I receive as http header is the same as the one that I
>> send a long to the backend when an exception occurs.
>>
>> The error I get:
>> ---
>>
>> (TypeError) : Cannot call method 'Kg' of null
>>
>> at Unknown.vSb(https://www.leuker.nl/app/aangifte/start/)
>>
>>
>>
>>
>>
>>
>> ---
>>
>> Any idea ?
>> I think it might come with the JS engine that googlebot uses. For
>> example: I don't support IE9, as I use JS features that aren't supported by
>> IE9 that result in similar errors.
>> But how to work around this? It's hard to see the exact error that occurs.
>>
>> - Ed
>>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/ixo_9kd-joU/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: Click even on outer HTMLPanel on existing html ?

[Ed Bras]

Yes I do (as DomHandler: ClickEvent.getType()) that's why I don't
understand I don't catch any.
I get the idea the events aren't bubbled up (passed through), because the
loaded html in the HTMLPanel exists of existing html instead of creating it
through gwt, such that events are passed through...


On 26 January 2016 at 22:02, Vassilis Virvilis  wrote:

> stupid suggestion but do you sink the events you want to listen for?
>
> On Tue, Jan 26, 2016 at 11:00 PM, Ed  wrote:
>
>> Can somebody please give me some tips on how to solve this ?
>> How to get the click events on the outer div html, loaded in an HTMLPanel?
>> Thank you, Ed
>>
>>
>> On Tuesday, 12 January 2016 11:53:03 UTC+1, Ed wrote:
>>>
>>> Please some advice on this issue ?
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "GWT Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to google-web-toolkit+unsubscr...@googlegroups.com.
>> To post to this group, send email to google-web-toolkit@googlegroups.com.
>> Visit this group at https://groups.google.com/group/google-web-toolkit.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>
> --
> Vassilis Virvilis
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/60RGI3WN1W0/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: Click even on outer HTMLPanel on existing html ?

[Ed Bras]

Hmmm,, so it should work for existing html, that is loaded from an xml file
(so no wrap is needed).


On 26 January 2016 at 23:31, Jens  wrote:

>
> stupid suggestion but do you sink the events you want to listen for?
>>
>
> addDomHandler() automatically sinks the corresponding native event type.
> You only need to sink events yourself when you override onBrowserEvent() to
> handle native events without registering a handler using addDomHandler().
>
> I know that HTMLPanel.addDomHandler(ClickHandler, ClickEvent.getType())
> works as expected as our app uses it. However it does not use
> HTMLPanel.wrap().
>
> -- J.
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/60RGI3WN1W0/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: How to solve RPC error: Character Encoding is 'text/'. Expected 'UTF-8'

[Ed Bras]

@Thomas: thanks for the tips, I will perform some logging of the
Content-Type to see where it's coming from.​

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

@Axel: thanks for sharing and you are 100% spot on
I also like to know if/how SDBG can be improved to benefit the full
debugging power
​

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

Good to know, @DavidN, thanks for sharing...

On 22 December 2015 at 10:45, DavidN  wrote:

> On Friday, December 18, 2015 at 10:17:27 AM UTC+1, Ed wrote:
>
>> I was reluctant at first to move over to SDM, but after some time with
>>> it, I got used to it. I combine it with SDBG in eclipse and it works mostly
>>> ok.
>>>
>> What is exactly not woking well with the SDBG eclipse plugin?
>> And why is this working well in IntelliJ ?
>>
>>
> Sometimes it works, sometimes it doesn't that is a bit annoying. Some
> constructs like for loops were not properly mapped to the java sources
> which made stepping through them a bit painful. But I am running behind on
> SDBG, I did not do GWT development in the last year - just starting again.
>
> In Chrome I do not always seem to get the mapping to work for all classes
> in my project.
>
> I have no experience with IntelliJ, but I am interested in using it since
> maven/eclipse integration seems to be really slow and buggy. But I am
> forced to use ClearCase for the moment and we don't have IntelliJ support
> for it. I hope we can switch to something git based next year.
>
>
>> What SDM brings is much faster prototyping of the GUI, especially in
>>> combination with Thomas Broyer's maven plugin it's really nice to have both
>>> server and client refresh so easily while developing.
>>>
>> How do you use the maven plugin to speed this up ?
>>
>
>
> I just launch tomcat and codeserver from maven instead of depending on my
> IDE. It works really well. Any change I make to server classes gets loaded
> automatically by Tomcat and any change on client code gets picked up by the
> SDM codeserver. In the browser I just click on reload and I have the latest
> version, it takes just a second or 2. (OK I seem to have an issue with GWT
> 2.8.0-beta1, but I'm sure that will get resolved somehow).
>
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/IfLWnnCf9rM/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

>
> I was reluctant at first to move over to SDM, but after some time with it,
> I got used to it. I combine it with SDBG in eclipse and it works mostly ok.
>
What is exactly not woking well with the SDBG eclipse plugin?
And why is this working well in IntelliJ ?


What SDM brings is much faster prototyping of the GUI, especially in
> combination with Thomas Broyer's maven plugin it's really nice to have both
> server and client refresh so easily while developing.
>
How do you use the maven plugin to speed this up ?

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

@Frank: you are absolutely right, but it's difficult to generalize this, as
it highly depends on your situation.

I have seen hello-world kind of code that was very easy to understand, but
also very complex, but very clean code to meet a lot of financial
(security) requirement, or code with many visitor patterns to be able to
handle different kind of products (subclasses). In these latter cases, it
can cost you a lot of time to understand, even when it's very clean and
good coding... Often analyzing the code cost more then making the
change... (according to statistics: solving bug consists 70% of code/issue
analyzing).

On 10 December 2015 at 18:01, Frank  wrote:

> If you have to read code for hours to solve a problem without a debugger I
> think you have some bigger problems with your code than some bugs... (code
> structure)
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/IfLWnnCf9rM/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

Interesting discussion about breakpoints ;)​
It all depends on your situations and requirements...
I don't like breakpoints but need them:
1) Solve a problem quickly instead of reading the code for hours (like
mentioned).
2) Easily understand the flow (stack) without setting many print stuff.
3) Improve performance: loops that are done to much, code that is called
too much, which can be a burden with big data (you don't want to wait till
productions runs with bigger data then in test)


Just my 50 cents...

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

Thanks for sharing @Steve. Good to read that you do have "ok" experience
with larger codebase.
> SDM is the ability to debug a large codebase in reasonable time -
startups for Dev mode can be very slow.
I am always surprised when people say this.
 run a code base of total more then 2.5MB total (after compilation in soyc
report), pretty fast on windows and mac. Startup in about 5-8s. When
developers came to work on the project they ware always surprised it starts
fast.
Maybe it's because I debug for performance ;-)



On 10 December 2015 at 11:33, steve Zara  wrote:

> SDM takes some getting used to, but now I wouldn't want to go back to Dev
> mode.  The real advantage I have found with SDM is the ability to debug a
> large codebase in reasonable time - startups for Dev mode can be very
> slow.  Source maps are now supported pretty much everywhere, and work very
> well in Chrome.  I used to complain about SDM but I can honestly say that
> after months of use I found it a reasonable way to debug.  Not perfect, but
> it did mean that I could debug graphics-intensive code in reasonable time,
> which was just not possible in Dev mode.
>
> On Wednesday, 9 December 2015 18:35:38 UTC, Liraz S wrote:
>>
>> I get that this is impossible to keep maintaining the previous Dev mode,
>> but without it GWT cannot be debugged properly anymore.
>> See the following for example:
>>
>> http://stackoverflow.com/questions/25105469/how-to-debug-using-superdev-mode
>>
>> In SDM You cannot watch, it's impossible to check which type or what's
>> the problem exactly during debug - it basically made working with GWT
>> impossible.
>> The only thing that helps me get going with GWT development is the
>> regular Dev mode - which you can use the full features of debugging, and
>> solving the problem in seconds - not hours
>>
>> And as far as i can see, this will not be improved anytime soon - since
>> Chrome is simply translating source maps, it's not debugging a JVM program
>> - and will be the same for all other browsers that will support source maps.
>> Oh.. and what about IE? will they support source maps - you just killed
>> it there i guess..
>>
>> You have to get the Dev mode back, and to make the legacy Dev mode better
>> - not the super dev mode that is aimed to source maps - i think it's a
>> mistake.
>> Super dev mode is a nice idea and feature, but it doesn't work - tried it
>> for some time...
>> Also tried it on 2.8.0 - still useless.. same problems - far from the
>> basic needs of debugging
>>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "GWT Users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/google-web-toolkit/IfLWnnCf9rM/unsubscribe
> .
> To unsubscribe from this group and all its topics, send an email to
> google-web-toolkit+unsubscr...@googlegroups.com.
> To post to this group, send email to google-web-toolkit@googlegroups.com.
> Visit this group at http://groups.google.com/group/google-web-toolkit.
> For more options, visit https://groups.google.com/d/optout.
>

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

Re: Getting rid of Dev mode for future GWT releases

[Ed Bras]

@Steve: thanks for sharing your experience, good to know.​
When I have some more time the next coming month I want to try again Super
DevMode, I really would love to use it. It's just easy not to use it when
you work a lot and the DevMode works fine in my case...

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

Re: GoogleBot is using the wrong permutation?

[Ed Bras]

What I did that probably solved it:​ instead of reading the permutation
name through the HTTP header using RpcRequestBuilder.STRONG_NAME_HEADER, I
now send the permutation name along from with the exception occurred in the
client.
At least I am not using any old cached http header.
Let's see if that works, I think it does as I didn't see the exception just
yet ;)

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

>
> It should be in Elemental 2.0 (I suppose); there's no need to bring an
> abstraction on top, JsInterop should suffice.

@Thomas, thanks, looking forward to it.​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

​@Thomas: besides the fallback hash support, will pushState not be
supported at all in the gwt core in the future ?

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

@Thomas: thanks for your input.  Yes, I remember I also looked at your
Html5Historian code as example.
Do you know if or how, the Html5 pushState will be supported in the gwt
core?
I think it should be possible to support it the way I implemented it: use
pushState if the browser supports it, else the old hash fragment method is
used.

​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

@Pablo: cool, thanks for your input, however, i can't use GWTP framework
and not even allowed to use it (giving the dependency issues we had in the
past). We can use it if it's part of the core ;)
​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

​It's rather straight forward and the nice thing is that you can start your
app in different way's, by inspecting the url it was started way. Example:
I change my landing page on the client side, depending on the url. Compare
www.leuker.nl and www.leuker.nl/trainmore or www.leuker.nl/flks (all same
app, just showing other content that have stored in xml files).

Anybody, first you need to ensure you all url's end up in the index.html. I
apache config, I did that through some rewrite rules:

  RewriteRule ^$ - [NC,L]  RewriteRule ^index.html - [NC,L]

  RewriteRule ^resources/ - [NC,L]

  RewriteCond %{REQUEST_FILENAME} !-f

  RewriteCond %{REQUEST_FILENAME} !-d

  RewriteRule .* index.html [NC,L]

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

Oeps, I wasn't finished yet, and pressed send. Let's continue were I left
off:
BTW: the above rewrite rule with resource in it, can be removed if you
don't have that path. Ensure to put the above in the  tag, else
the REQUEST_FILENAME will not be filled yet and will contain the URI, and
that's not what you want, such that it doesn't work... (took me some time
to figger that out ;)

Then in the front-end, do what ever you want with the url it's start with,
show the context what you want.

Then the pushState thing. I started with the gwt-pushstate project on
github, but it doesn't work with gwt 2.7 due to changes in the gwt History
class.
As such, I replaced the History class with my own version that I use. It
can use the old url fragment style or the push state if supported (not
tested all well I have to admit).
Have a look at the code attached, start with the BrowserHistory class, that
is the replacement class of the gwt History class, and also contains a lot
of code of that class. It will talk to a SimpleBrowserHistory class that is
created through deferred binding and will use the pushState subclass if
pushState is supported. The support is detected though the attached gwt
file (thanks to code from gwt-pushstate on github).
I might change some thinks in the future if I have a moment and optimize
thinks, but for now it works well.
Just look around in the code, and if you have any questions, just shoot.





​

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


SimpleBrowserHistory.java
Description: Binary data


BrowserHistory.java
Description: Binary data


PushStateHistory.java
Description: Binary data



	
	
	
	
	

	

	
	

	
	

	
	

	

	
	
		
			
			
		
	

	
	
		
		
			
			

			
			
		
	

	
	
		
		
			
		
	
	
		
		
			
		
	

	
	
		
		
			
		
	

	
	
		
	
	
		
	

Re: GWT browser plugin not used with Web rewrite rules?

[Ed Bras]

One solution that seem to work:
Use of the old fashion base tag in the index.html:
   
I am just not so font of making the index.html location/url dependent, such
that you have to edit the index.html during deployment.. but you have
to tell the browser which base url to use when performing some kind of url
dispatching that results in changed url paths...

I noticed it's a well known issue with using html 5 pushstate, see: LINK


Would be nice to be able to solve it in apache, such that it's set per app
(location)... I will have a look...
Any ideas ?


​

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

Re: GWT browser plugin not used with Web rewrite rules?

[Ed Bras]

@Thomas: the plus.nocache.js is my script that is included in the
index.html:
  

Let me see if I understand you correctly:
You think that it will look for /plus/bla1/plus.nocache.js which doesn't
exists such that the server returns index.html ?
H, just checked and I am afraid you are right :(, see screenshot...
I have to think about thiscome up with a better solution: use absolute
paths or use relative and handle it correctly on the server

Some side info:
+ I am running GWT in no-server mode.
+ I don't use the eclipse plugin, but use Eclipse launch files to run Dev
mode.


[image: Inline images 1]
​

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

Re: GWT browser plugin not used with Web rewrite rules?

[Ed Bras]

BTW: I was just testing some more and for some reason the last rewrite rule
"RewriteRule plus/(.*) plus/index.html [NC,L]" is changing something (even
do it shouldn't). I should copy the query params.
If I change this rule to "RewriteRule plus/(.*) - [NC,L]" such that nothing
is changed and create a sub dir "bla1" with all required content, it does
work...
So what is the first rewrite rule changing such that the gwt plugin not
work ?... :(


​

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

Re: GWT browser plugin not used with Web rewrite rules?

[Ed Bras]

What I forgot: I think that the GWT plugin is exactly reading the browser
URL and use it. A such it will look in the /plus/bla1 location.
If I use the rewrite rule "RewriteRule plus/(.*) plus/index.html [NC,L]"
with the sub dir "bal1" that contains the same files as the root, it works.

I think I always have to create this sub dir with the root content, not ?
Else, how will the GWT plugin knows in which URL location to look not ?
But why does the GWT plugin exactly look in that location? What does it do?

- Ed
​

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

Re: With ie9 I get: XMLParserImplIE6 Could not find appropriate version of DOMDocument ?

[Ed Bras]

Thanks for the insight @Jens.​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

> I have seen this page before. Where did you see that google requires #!
(vs plain #) in its new scheme?
Where can I find the new scheme details ?
I find it hard to find the exact details, often you will find the old
details.

I notice in the Google Webmaster tools that the ur's with the # url
fragment aren't indexed, which is also mentioned in the following article:
https://www.deepcrawl.com/knowledge/best-practice/guide-to-url-design/

So I am thinking about using #! instead of # and see what happens, or using
the History.pushState() (I not support IE 9 anyway), as I get the idea that
these url's (created with the pushState()), can be indexed by Google. But I
am not sure.

What do you think? Are your ur's with the # fragment indexed by Google?

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

​BTW: i noticed in this link, that I should use #! still such that Google
will crawl it, else it will not crawl it:
 https://support.google.com/webmasters/answer/174993?hl=en

Anybody any experience with it?

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

@​Vassilis: I created a page sitemap.xml (exactly a sitemap index with a
video sitemap and a page sitemap), that contains pages like:



https://www.leuker.nl/index.html#DcNvUsg


And if I submit this to Google it reports no errors and does indicate that
17 pages are found/indexed

I think I now have to wait and check the log file when the sitemap pages
are indexed. I am not sure when that occurs, I checked the log, noticed
google bot coming by but not crawling the sitemap pages yet...

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

I just noticed that GoogleBot is coming by and resulted in an error when
the site tried to perform an animation through a javascript function that
isn't present.
It's because I use GreenSock animation that is loaded in the index.html,
but apparently the function $wnd.TweenLite function isn't present :(...
Strange, any idea why this is ? Why isn't this created on the "Dom" when
GoogleBot visits the website ?


​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

Thanks Vassilis.
I just solved part of the issue: I commented a line robots.txt file such
that it can crawl the /resouces folder that contains all images/css/js.
So google was able to get further, and show correctly the first screen.
I just have to find out how it's able to crawl the rest of the screens. It
hink through a sitemap, but I am a bit confused how the sitemap should look
? Should I just use the full ur's with the history token ? example:
www.leuker.nl/index.html#OtherPage ?
- Ed
​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

@Thomas: I just added:
  
The exception popup is now not shown during startup, but the google bot (in
google webmaster tools) will not get passed the "loading..." text, that is
shown in the index.html and then removed by the GWT app when started.
H
Please help on how to solve this?
- Ed
​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

Thanks Thomas, I was just reading about it in the 8079
.

Do you think it's the cause that Google doesn't correctly crawl my site ?
As I notice that my dynamic pages that are directly accessible through
leuker.nl#Example don't show up in the google result.
- Ed
​

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

Re: SEO Google Crawlable experience ?

[Ed Bras]

My gwt app isn't packaged in the war, it's separated from the backend so
you don't see the touch issue.
However, yes, I have the tric before, creating the undefined.nocache.js
file, but never used it.
I will add the property to supress the warning and see if google can crawl
it normally.​

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

Re: GWT and GA and analytics.js usage ?

[Ed Bras]

@Gilberto: How do you use the Server side analytics class without GIN ?

​

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

Re: GWT and GA and analytics.js usage ?

[Ed Bras]

Thanks, I am looking at the Arbees classes at the moment and I think it's
my best bet.

​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

I have some problem with code ending up in the left over because it's
reached from a split point, while it shouldn't be reached through that
split point.
Why?

Details:
Below a snippet from the Soyc report that shows why the Login component is
alive from a split point in the app Ztoeslag.
The Login component shouldn't be alive through this split point, but it
occurs because controller ProfileControllerDefault is shown in the soyc
snippet below.

I don't understand why controller "ProfileControllerDefault" appears in the
alive-list below, as it is created in a RunAsync GeneralSplitPoint10 (the
same as the Login component).
The only place in the whole code where "ProfileControllerDefault" appears
is at the place it's created, which is called in the
GeneralSplitPoint10 split point, that calls the following method in the
ProfileFacade:

public ProfileController createController() {
   return new ProfileControllerDefault();
}
Because the Split points work method-based, that should be fine, however it
shouldn't be touched/reached in the soyc snippet below.

Some more method flow:
The LazyProfileController.showProfile is listed below. The call findActor()
will return the ProfileControllerDefault instance if already loaded. If not
loaded yet, it will start loading and in the mean while the showProfile()
action in put on the queue through the method below:
GeneralActionQueueFacade.showProfile(). When the ProfileControllerDefault
is created, the queue is emptied.

So can you please tell me why the ProfileControllerDefault appears in the
list below that shows why the Login component is alive.
How can I further analyze this? How can I confirm this is a bug ? (I hope
not :(

- Ed

-- Snippet of why the Login component is alive.
...
- com.profile.impl.ProfileControllerDefault::$getEnsureView
- com.profile.impl.ProfileControllerDefault::$beforeShow
- com.actionqueue.GeneralActionQueueFacade::$showProfile
- com.profile.impl.LazyProfileController::$showProfile
- com.profile.impl.ProfileFacade::$showProfile
- com.profile.ProfileHistory::runIntern
..

   -
   com.ztoeslag.ZtoeslagLazyLoaderDefault::$createCodeSplitZtoeslagGeneral10

--

 The LazyProfileController.showProfile() method:

public void showProfile(final FlexShowContext context) {
   if (!isCommandPresentAndEnsureLoadingStarted(getShowProfileId())) { //
will check the queued command isn't present yet, and ensure the will be
loaded if not present.
 getGeneralActionQueueFacade().showProfile(findActor(), this,
getShowProfileId(), context); // (findActor(): will return the
ProfileControllerDefault if loaded)
   }
}


--- The GeneralActionQueueFacade.showProfile() method:

public final >> T showProfile(final T item, final Q queue,
final String id, final FlexShowContext context) {
 if (isNull(item)) { // buffer the action
 queue.addToQueue(new ActionCommandBase(id) {
 public void execute(final T controller) {
  controller.showProfile(context);
  }
});
}
else {
   item.showProfile(context);
}
return item;
}

---


​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

BTW: you think it's possible to define different split points per app build?
Why? Declare and Ztoeslag build have both an initial download of about
400K, but in total the initial download is about 800K.
So I think in Total I would like to have less split points in Declare and
Ztoeslag to reduce the initial download by making code exclusive to a split
point.

I was thinking about using deferred binding to override the class that lazy
load the components through runAsync and change the number of runAsync
methods.

What do you think of this idea ?
​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

@Jens: thanks a lot for al the insight.
I thought I understand it, but apparently not.
I did some reading and started to optimize thinks, like the thinks you
suggested and reviewing my split points. I think I can merge some of them
such that some code becomes exclusive to that split point that will reduce
the left over.
Also I will define the initial download, after I am using class name
literals for the split points (I wasn't aware of this 2.6.0 class literal
split point shared naming feature).
I also removed some sprite images that are used at the end of the app, but
ended up left over. I use background-image css instead now.
Thanks, I will continue optimizing ;)
​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

"One" more thing (sorry for asking so much):
I noticed that my login component is contained in the left-over fragment
instead of being exclusive to it's own splitpoint.
The Soyc report shows that it's because it's referenced by the Account
component that is loaded in another splitpoint. The Account will try to
retrieve the login name from the Login component in case it's now filled in
yet.
Anyway: The Account will ask the LoginFacade singleton for the login name
that will return null in case the LoginController isn't loaded, which is
loaded in a split point.

Why is the Account pushing the Login component in the left over? as the
login component is isolated through a split point, and it's like it just
"ignores" when looking at the soyc as it goes from the LoginFacade straight
in the LoginController.
Note: the LoginController is the interface and the LoginControllerDefault
is the implementation that is created through runAsync.

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

@Jens: thanks.
> Personally I probably wouldn't have created a Total app,
Yes, I wanted to do the same, and for the short time, it's the best. But
seeing the coming feature requirements, I need to share info between the
apps while running next to each other.

Let's see how it goes, it's difficult to predict the future. Looking back,
I would set it up independently for now because of the big download, but
also because letting them run next to each other was a bit more pain then I
expected (scoping issues: app and global scope when switching apps, but
that has nothing to do with the lazy loading issue here)

I will use deferred binding for now, it's easy to make the switch (I think
;) (I use all app deferred binding in a central place: one deferred binding
of a application context class that changes all used factory creation done
in the context class.)

​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

@Jens: thanks again for the quick response. Almost there I think

> As your if-statement can not be evaluated to either true or false at
compile time / static code analysis
Hmmm interesting...
Any idea how I can optimize this such that it's "left alone"? (I can't
really think of one).

What I do: the LoginController is created in a split point and put in a
ControllerRegister.
When the Account Component access the login name through the LoginFacade
singleton (getLoginFacade().getLoginName()), the facade will lookup the
LoginController in the ControllerRegister. If it's present, it will pass on
the method, else it will return null (the IF statement)

​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

​Thanks, I understand what you mean.
However, I feels awkward, and will not improve maintainability.
I get the feeling I am working around a bug.
The component facade is especially there to hide the component internals,
like the controller implementation, that is package protected and only
referenced in a runAsync call. So it's very strange that it's included in
the split point that is used by Account

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

@Jens: thanks for the reponse.
> You have used "fragmentCount" during compilation. When using this setting
you are
> allowing the compiler to merge split points if the compiler thinks that
merging will decrease the
> size of your left over fragment.
Indeed, I just removed it, and it's larger (about 15% I guess).

Anyway: Yes, your scenario description is correct and thanks for better
understanding it.
So I was reading it wrong to check why it wasn't exclusive to split point
#4.

But let's start with the original problem: I want the Total app to be
small, and merely/not contain/share any code of the other apps. How to
accomplish this?

I think I have to make sure that the apps Declare and Ztoeslag are allowed
to share code through the code fragments, but not share any code with the
Total app. But how to realize/configure this?

(I could then load the apps in the background when the Total app is
started, but that is for latter optimization).

The alternative would be that I make 3 separate GWT apps with their own
html index.html page and simple call these pages when needed. But that is
old fashion and I think I am loosing the power of GWT hen ...



​

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

Re: Why code ends up in Left over Soyc report? How to analyze?

[Ed Bras]

Thanks @Jens for your quick response.
Complicated.. ;)

> Its not really possible to keep Total really small because you compile
everything into a single app.
Yes, I understand that, but it would be nice to configure what is
shared/downloaded and when.

> probably keep small in your Total app is the initial code download
because everything else is behind a split point.
What do mean by this and how to do this?
I don't understand what mean, as currently I put everything behind a split
point, and it's far from small.
A little side note: The Total app will fire the HomeController that is lazy
loaded through a general split point (the split point isn't shared
currently). The total controller will then show the page with the 2
app-start buttons.

> IMHO you actually haven't won anything because both your app's split
points have been merged by GWT
> to minimize left over.
Yes, I understand what you mean, but I though that in GWT 2.7.0, GWT will
create smaller code fragments that are shared between split points, and
don't necessary be in the left over fragment (= initial download
eventually).
So why does GWT not create a small left over for Total and then a big
"left-over" shared code fragment that is shared between Ztoeslag and
Declare and is loaded when one of these apps are loaded (as a left-over
fragment), so it's not part of the initial left-over.
This is for me the expected behavior, especially as Total is very little
and hardly share any code with Ztoeslag and Declare like explained in the
first post.
What am I not understanding here?

I still don't understand why this class "InkomstenOverzichtCalculator2011"
(as example) is contained in the initial left over as it's ONLY used by the
Declare app... Why? :(

> The only downside might be that the client downloads common GWT SDK code
(widgets, etc.) multiple times.
Yes, that is one of the main  reasons I put them together.
BTW: the apps can still run completely independent and being deployed
independently.

> Choose your poison ;)
So basically you are saying it's (currently) not possible to fine tune the
current GWT lazy loading mechanism to have control on what is loaded
and when?
In my case I would like to indicate on for example package level what is
allowed to be shared. How can this be possible (hidden split point features
maybe...)?
And can I realize a better left over by changing the fragmentCount?


​

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

Re: Most code in SOYC report cannot be attributed

[Ed Bras]

Interesting , as I don't see that part below in your screenshot :( (and
don't remember ever seeing 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+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.

Re: GSS Media Query Variables

[Ed Bras]

I think your snippet must work, if not, check if this was a bug and already
solved but not yet released.
I remember that some bugs are solved in 2.8, but not yet released. This
might be one of them.

I also notice that GSS is still bit restricted​, but it's still a bit new
and I am sure they will add more features when it gets popular.

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

Re: How to search object instances by their type ?

[Ed Bras]

Anybody any idea please ?​

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

 Graag gedaan.
Hahhaha, grappig :)
​

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

BTW: thanks all for the hulp, tips and ticks, it helped a lot.

​

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

I did run IE11 in VirtualBox as @Jens suggested and with style on DETAILS.
I quickly noticed that IE11 doesn't like the $doc.contains(Element) method.
I changed it to $doc.body.contains(Element) and that seem to work.
See also: https://github.com/skatejs/skatejs/issues/153

​

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

Thanks @David, you sure gave me some good pointers

 Does your app have JSNI code or some code that does try {} catch(...)
without reporting the issue ?​
Not as far as I know, but maybe the lib's I use. I use mainly GreenSock as
external lib and they don't do try/catch for performance reasons. I had
this discussion with them in the forum some time ago when some exceptions
got lost in async JS execution.
I use it for animations like slide in/out, so it could be that that is the
reason the web site isn't shown.

 When I run it through IE11 with break on every exception it report a lot
of contains is undefined exceptions ...
Sorry, I completely forgot to use that :( thanks...
I just looked at it, and yes, some ugly exceptions come by that I can't
explain yet.
As we speak, I am compiling the gwt app with DETAILS style output to get
some more info during debugging.

I hope that works (I will get back on it). If not, I will probably setup
@Jens his suggestion: Virtual box with IE11 with Super Dev.

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

@Jens: thanks for your feedback.

 Can't you just debug the app using SuperDevMode?
I thought about it, but I am not sure how as I develop on a Mac?
And shouldn't the IE11 support the source mapping, or it does..
(I played with super dev mode in the past, but still not using it)

 Maybe try GWTs UncaughtExceptionHandler
I do, that is the weird think, and it's set directly when the app starts
and it will show a friendly popup that something went wrong and will
email the error. And in the dev version (on the dev server), the error is
also send in a develop div that I can show through a shortcut But it
does catch nothing :(... strange...

It must be something stupid, but can't find it and don't really know were
to search :(
Any ideas?
(I think of adding a bunch of log statements in the dev version and see
what it does in de dev console dev)

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

Re: Website not working in IE11. How to solve?

[Ed Bras]

@Paul: thanks for your details feedback and pointers.

 Can you confirm you are using 2.7.1, do you mean 2.7.0?
Yes, sorry, you are right, it's 2.7.0


 Do you possibly include a module that overrides how the user agent
permutation is calculated from the user agent string?
I don't think so, how can I check that ?
I use derived user agent properties like presto, webkit, gecko.
And I use a a property gwt.os that has values like mobile, desktop, ipad,
etc.. I derive that from JS with a property-provider, and use it in gss
files.

 For ie11 when I add the meta tag you've added I do not get any issues.
What value did you exactly use and how did you change it such that it has
effect?
When I change the value to IE=10 with the IE dev tools, I don't see any
change, and doubt if it has effect, as I don't see some kind of refresh...
But if I change it on the server in the index.html, it also has no effect.

Thanks for the debugging hints, I will dive in to them.







​

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

Re: Include a GSS file in another GSS file?

[Ed Bras]

Thanks for your insight @Jens, I already was afraid it wasn't possible.
Have a nice weekend all ;)
​

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

Re: Include a GSS file in another GSS file?

[Ed Bras]

Thanks, I know the construction, but how does this solve my issue ?
(including a gss file in another gss file like an import mechanism, such
that others only include one gss file)

​

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

Re: How to do smooth scrolling on browser back/next button on same-page navigation?

[Ed Bras]

@Jens: I don't think your solution isn't working.

You mentioned I should not use id attribute, and instead use something like
data-scroll-id.
I change it now to for example: div data-id=blaMarker.../div, but
still the browser jumps on pressing the history back button before I can
process my history marker.

Details:
As soon as a person arrives as a certain part of the page I set a history
marker (as hash) in the url, like example:
#NvWrkDcUsgSt?id=IdUsgSelect​
The first part indicates the page to be shown and the id parts the section
on the page.
The target Element containing the id part: div
data-id=idUsgSelect.../div

So when the History change event is fired with the above history value, I
stop in the GWT class History.onHashChanged() through a breakpoint, but
then the browser already jumped to the position when the history marker was
added to the url.

Apparently the browser will record the scroll position of the page when the
url is changed and jumps back to it when it appears in the browser history
during back/forward actions.
But is there anyway to ensure the browser will scroll smoothly instead of
jumping ...?

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

Re: How to do smooth scrolling on browser back/next button on same-page navigation?

[Ed Bras]

​@Jens: thanks, i hoped I missed something simple.
I think I just leave it for now. It's more a nice-2-have and there are
more important thinks now,a and I remember messing with the browser history
can become a bit hairy..

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

Re: How to do smooth scrolling on browser back/next button on same-page navigation?

[Ed Bras]

@Jaga: I already have the smooth scrolling for free through the components
I use.
The problem is probably that I use the id attribute like @Jens mention,
which I will also do in LayoutPanel ;)..
I am about to test it, I will come back on 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+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.

Re: How to do smooth scrolling on browser back/next button on same-page navigation?

[Ed Bras]

@Jens, thanks I think this is the problem:
 Since your site should not have an element with id=section-1
Interesting... wasn't aware of that, that is what I do and will probably be
the problem .  I will change it, using the data- attribute as you suggest
and let you know how it works.

I tried to search for the info about when/how the browser does add a
History token, but can't find it. If you have a link, please share.

The Smooth scrolling is already done through (the same) History markers
through click events/navigation (menu), so it's nice if it also happens in
case of back/forward. It looks weird if it only works half...


​

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

Re: How to do smooth scrolling on browser back/next button on same-page navigation?

[Ed Bras]

@Gilbert, thanks, but don't think I made myself clear, sorry for that.
What you are writing is exactly what I do, the problem is that before I can
perform the scrolling myself because I capture a History event the browser
did already the scrolling for me like explained in my original post.

btw: I don't need to store scolling positions as the scrolling sections
contain unique id's that are used for scrolling, but also as navigation
from other places in the app.
​

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

Re: GSS mixin restricted ?

[Ed Bras]

@Julien: thanks for the quick answer.
Stupid of mine not to think about your approach, so simple ;)

Anyway: I tried it, but the GWT/GSS compiler doesn't seem to like the @if
conditions inside the @defmixin {}. I get an error :(
My tryout code:

@defmixin slideInToRight() {
@if is(browser.engine,  webkit) {
-webkit-transform: translate(-40%, 0);
-webkit-transition: transform 1s, opacity 2s;
}
}


.Content {
@mixin slideInToRight();
}

When I remove the inside @if, it works however. But with the inside @if
(like above) I do get the following exception:
What to do ? ;) (BTW: I use GWT 2.7.0, sorry, can't try it on the trunc at
the moment)

--
00:07:29,829 [ERROR] Generator
'com.google.gwt.resources.rebind.context.InlineClientBundleGenerator' threw
an exception while rebinding
'com.bla.loginflow.resources.LoginFlowResources'
java.lang.IllegalStateException: null at
com.google.gwt.thirdparty.guava.common.base.Preconditions.checkState(Preconditions.java:161)
at
com.google.gwt.thirdparty.common.css.compiler.passes.CreateConditionalNodes.createConditionalRuleNode(CreateConditionalNodes.java:122)
at
com.google.gwt.thirdparty.common.css.compiler.passes.CreateConditionalNodes.leaveUnknownAtRule(CreateConditionalNodes.java:93)
at
com.google.gwt.thirdparty.common.css.compiler.ast.DefaultVisitController$VisitUnknownAtRuleState.doVisit(DefaultVisitController.java:1774)
at
com.google.gwt.thirdparty.common.css.compiler.ast.DefaultVisitController.startVisit(DefaultVisitController.java:1995)
at
com.google.gwt.thirdparty.common.css.compiler.passes.CreateConditionalNodes.runPass(CreateConditionalNodes.java:184)
at
com.google.gwt.resources.rg.GssResourceGenerator.finalizeTree(GssResourceGenerator.java:596)
at
com.google.gwt.resources.rg.GssResourceGenerator.parseResources(GssResourceGenerator.java:770)
at
com.google.gwt.resources.rg.GssResourceGenerator.prepare(GssResourceGenerator.java:479)
at
com.google.gwt.resources.rg.CssResourceGenerator.prepare(CssResourceGenerator.java:527)
at
com.google.gwt.resources.rebind.context.AbstractClientBundleGenerator.initAndPrepare(AbstractClientBundleGenerator.java:1013)
at
com.google.gwt.resources.rebind.context.AbstractClientBundleGenerator.initAndPrepare(AbstractClientBundleGenerator.java:1039)
at
com.google.gwt.resources.rebind.context.AbstractClientBundleGenerator.generateIncrementally(AbstractClientBundleGenerator.java:419)
at
com.google.gwt.dev.javac.StandardGeneratorContext.runGeneratorIncrementally(StandardGeneratorContext.java:754)
at
com.google.gwt.dev.cfg.RuleGenerateWith.realize(RuleGenerateWith.java:160) at
com.google.gwt.dev.shell.StandardRebindOracle$Rebinder.rebind(StandardRebindOracle.java:79)
at
com.google.gwt.dev.shell.StandardRebindOracle.rebind(StandardRebindOracle.java:276)
at
com.google.gwt.dev.shell.ShellModuleSpaceHost.rebind(ShellModuleSpaceHost.java:152)
at com.google.gwt.dev.shell.ModuleSpace.rebind(ModuleSpace.java:613) at
com.google.gwt.dev.shell.ModuleSpace.rebindAndCreate(ModuleSpace.java:483) at
com.google.gwt.dev.shell.GWTBridgeImpl.create(GWTBridgeImpl.java:49) at
com.google.gwt.core.shared.GWT.createImpl(GWT.java:83) at
com.google.gwt.core.client.GWT.create(GWT.java:86)
--



​

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

Re: LESS/ SASS/... style usage in GWT ?

[Ed Bras]

​There are several options to fly to the moon.
There is no silver bullet. .. Just pick the best of all worlds for your
situation...
If your Css needs to be very flexible, I don't think that CssResource is
the correct choice...
I would then use Errai with a good html parser, that is able to create
behavior from the html attributes (like frameworks like
bootstrap/foundation do). Example: in my html, through data-* attribute I
define behavior like the carousel delay, or the internal navigation to
other parts of the app. Works very well, no need to change gwt code through
these kind of syntax

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

Re: GSS mixin restricted ?

[Ed Bras]

Thanks @Jens for your feedback.
It's true, also thought about just removing these @if's such that I only
use the correct browser css rule. It's simple getting a bit too much, the
advantage isn't that big, and it's not improving maintainability. So
reasons enough to keep it simple like you explained.

Thanks for the exception feedback and noticing the error. Adding the
required prefixLess css rules even results in less benefit :(
Have a nice weekend, Ed
​

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

Re: LESS/ SASS/... style usage in GWT ?

[Ed Bras]

Have a look at the RedHat Errai templating mechanism.
Errai is setup in modules (jar's), so you can use only the parts you need.
And easy to extend/fork parts you need.

The idea is that you load the css/html dynamically when needed in the app
(through annotations). These html/css snippets/pages can be
controlled/provided by Customers/third party companies. It works well,
efficient with a good productivity: you can easily change the css/html
without recompiling the GWT app. It works good when clients want other
colors or provider their own html snippets/pages..
GWT loads the css/html snippets and adds logic to the required html
snippets, like buttons and form processing, etc...

With all the css/html framework and developers out there these days, it a
very friendly and productive integration.
As a GWT developer I love it and makes me happy. I don't want to be busy
with missing pixels and jumping css behaviors. Everybody has to do what he
is good at ;) Especially these days, css/js is changing a lot, difficult to
keep up with...

Just my 50 cent...


​

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

Re: LESS/ SASS/... style usage in GWT ?

[Ed Bras]

Concerning the Theming part: I did a gwt project like @Jens explained and
works well.

Details: like @Jens mention, retrieve values (often: colors, logos, etc..
and not layout like height/width) from static java files. The static java
(preference bridge) class will then forward the call to a contained java
(preference) class that it loads through GWT, as such that in GWT config
you specify the file that it need to load - Theming ;)... Works well,
another GWT config file results in another (client) theme.

Below Some copy/paste code to make it more clear (the Css part is like Jens
mentioned above). It concern a Standard theme that has a corresponding
Standard.gwt.xml file. So every client/theme would have it's own gwt xml
file. The Theme file you include in your app and compile it... So in Maven
you can define different gwt compiler jobs for every Theme...
Works well for white label kind of web apps..

The bridge class (I think I posted it before in the GWT forum some years
ago):

public final class ActivateCodePreferencesBridge implements
ResourcesPreferencesBridge {

private static final ActivateCodePreferences BRIDGE =
GWT.create(ActivateCodePreferences.class);

private ActivateCodePreferencesBridge() {
}

private static ActivateCodePreferences getBridge() {
return BRIDGE;
}

public static String activateFontColor() {
return getBridge().activateFontColor();
}

public static String activateFontWeight() {
return getBridge().activateFontWeight();
}

public static String activateFontSize() {
return getBridge().activateFontSize();
}
}



An implementation of the Preference instance:
--
public final class StandardActivationCodePreferences extends
AppStandardPreferencesBase implements ActivateCodePreferences {

public String activateFontColor() {
return color006699();
}

public String activateFontWeight() {
return fontWeightBold();
}

public String activateFontSize() {
return fontSize14();
}
}

--

And the GWT config Standard.gwt.xml:

module
   replace-with class=com.bla.standard.StandardActivateCodePreferences
  when-type-is class=com.bla.ActivateCodePreferences /
   /replace-with
/module


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

Re: GSS condition with Variable not working ?

[Ed Bras]

@Jens: Yes, I know that, but the exact syntax of the @if conditions isn't
completely clear to me from the docs, so it tried several thinks.
Any idea how it can work ? (what is allowed to be contained in the @if
condition ?)
​

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

Re: GSS condition with Variable not working ?

[Ed Bras]

@Jens: I am aware of the docs, but I think it's missing some details to
setup.

I just tried it again:
1) The property part in the gwt xml file:

define-configuration-property name=BLA is-multi-valued=false /
set-configuration-property name=BLA value=true /
The doc is mentioning *set-configuration-property * but if I use that,
the compiler i complaining. Shouldn't this be the
define-configuration-property?

2) The gss part:

@if (BLA) {
  html {background-color: #FF00FF;}
}

And the error:  The expression [BLA] is not valid condition.
So what am i doing wrong ?

Another thing: how to make the property true in the gwt config file in
case the property browser.engine equals webkit ?

​

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

Re: GSS condition with Variable not working ?

[Ed Bras]

@Julien: thanks for the explanation.​

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

Re: GSS condition with Variable not working ?

[Ed Bras]

 Seems like you have to stick with @if(is(browser.engine, webkit))
then.​
Thanks Jens. I was afraid of that.
Any idea this will be supported in the future ? (or maybe already planned)

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

Re: GSS condition with Variable not working ?

[Ed Bras]

@Julien:
Btw: is it possible to define the value of the BLA property when a certain
browser.engine value is met ?

So I am looking for support for something like (not gwt correct syntax):
define-configuration-property name=BLA is-multi-valued=false /
set-configuration-property name=BLA value=true 
  when-property-is name=user.agent value=webkit /
/set-configuration-property

Or:
set-property name=BLA value=true
  when-property-is name=user.agent value=safari /
/set-property

@if (BLA) {
  html {background-color: #FF00FF;}
}
​
But the above doesn't seem to be supported.
So currently it's only supported by manually setting it manually in the gwt
config file?

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

Re: [gwt-contrib] Re: idea: i18n dynamic templating ? (like wordpress), feedback

[Ed Bras]

@Jens: thanks for your feedback. I looked at it. I like the component-based
annotation approach.
It sure gives me inspiration. It would be nice if this templating mechanism
can be used independently of the rest of the platform (haven't looked at it
yet), and that it's flexible to extend it (It extends Widget which for me
is obsolete, don't use it anymore except for the root)
​

-- 
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.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit-contributors/CABzbGkL-%3DUSgNodYMKi2829iZwmfcGqQbeFMmS2dj-MuBrh0dw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [ERROR] Only either css files or gss files are supported on one interface

[Ed Bras]

Thanks for the explanation @Jens. I think I understand why the extension is
important. So we can't  let go if that requirement.
But let's go back to the original error (subject of the topic).
If I understand you correctly, I could be able to mix css and gss files by
using the correct extension. In case it concerns a css file, the gss
mechanism will sue the css2gss converter to create the gss file before
passing it to to the closure compiler. But that is what I am doing, so why
isn't that allowed?
- Ed
​

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

Re: [ERROR] Only either css files or gss files are supported on one interface

[Ed Bras]

Thanks @Jens.

 Because GWT can not convert it on a file-by-file basis.
Why not ?
​

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

Re: [ERROR] Only either css files or gss files are supported on one interface

[Ed Bras]

Ahhh, yes, you explained that before, sorry for that.
All clear now, I have to think about how to use 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+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.

Re: [ERROR] Only either css files or gss files are supported on one interface

[Ed Bras]

Thanks @Jens, I understand what you mean.
However, I still not see a could reason to force the gss file extension. I
think we can assume that the developer knows what he is doing and it's his
responsibility to feed the gss engine with the correct files, giving him
the freedom to use any extension he wants.

I understand that it's a nice separation, and it's how other css generation
tools work.
But Gwt is a bit different and does more then just css generation, so in
this case it has some blocking disadvantages.
Advantages of allowing the css extension: shared code use, merging to gss
step by step.

Let me explain it through my use case:
I have a gwt code base that exists of many widget having their own resource
bundle and css files (empty mostly, following to the adapter pattern).
New projects want to use css3 features, as such need the gss mechanism.
However, these shared resource bundles can't be changed, as it results in
breaking other projects using them. Neither do I want to change them as
mostly they concern empty css files that aren't gss dependent, which they
would look like it if they have the gss extension.

To solve this I see 2 options: duplicating the shared css files and giving
them the gss extension - a lot of work, messy maintainable code.
Patching the gwt gss engine to allow for css files with the css extension.
I think I go for the last option as it seems an easy one looking at the gss
code.

I would suggest to allow for the css extension. Optional with a console
warning that can be suppressed (like done in other parts of the gwt code).

I understand that the css files might contain some gwt syntax like @def
that might result in unwanted results when processed by gss. But isn't that
the same as letting Spring eat some hibernate syntax that it understands
and results in strange outcome It's the responsibility of the developer
to overcome that So give him that freedom

What do you think?
Just my 50 cents.


​

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

Re: [ERROR] Only either css files or gss files are supported on one interface

[Ed Bras]

I am very sorry Jens, but don't really understand your explanation.
Why can't the css file be used as gss file?
If I look at the examples of gss files, they can contain standard css
content.
​

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

Re: div / parsing not work in dom Element through innerHtml ?

[Ed Bras]

Thanks @Jens, I was afraid of that.. :(
I am not sure what doctype I should use at the moment with all these html5
support.
I currently use: !DOCTYPE html.
I am also not sure if I would break other things.

I think it's better to use valid html...

​

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

Re: GWT error no instantiable subtypes with interface in RPC method with 1 known concrete class?

[Ed Bras]

@Rhmoller: thanks. Yes they are marked as Serializable. And yes they are
visible to gwt, when I create a instance in gwt code, it results in no
error.​ I hardly ever use client, shared, server packages..

I get that impression that GWT does't allow interfaces (anymore) in RPC
methods. This is the new 2.7.0 code that shows me the error: LINKE
https://gwt-review.googlesource.com/#/c/7500/

@Jens: thanks. I am about to try that (using concrete classes), let me get
back on that.

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

Re: Types not resolved when upgrading to GWT 2.7.0 ?

[Ed Bras]

Cool, thanks ;)

​

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

Re: Compiling permutation hangs ?

[Ed Bras]

@Jens, thanks for the tip.
I tried localWorkers 1 before, but no luck. Just did it again and after
31 minutes without any progress (compiling permutation 0 of 10), I toke the
following dump with Jstack.
I haven't used jstack before, but can't really see anything helpful I
think, you ?
What to do next?

---JSTACK dump: --
2014-11-18 21:24:12
Full thread dump Java HotSpot(TM) 64-Bit Server VM (23.25-b01 mixed mode):

Attach Listener daemon prio=10 tid=0x7f62a4001000 nid=0x24eb waiting
on condition [0x]
   java.lang.Thread.State: RUNNABLE

   Locked ownable synchronizers:
- None

In-process PermutationWorker 0 prio=10 tid=0x7f62d8324000 nid=0x249f
runnable [0x7f62c5b98000]
   java.lang.Thread.State: RUNNABLE
at
com.google.gwt.thirdparty.guava.common.base.Equivalence$Equals.doEquivalent(Equivalence.java:327)
at
com.google.gwt.thirdparty.guava.common.base.Equivalence.equivalent(Equivalence.java:71)
at
com.google.gwt.thirdparty.guava.common.collect.MapMakerInternalMap$Segment.getEntry(MapMakerInternalMap.java:2413)
at
com.google.gwt.thirdparty.guava.common.collect.MapMakerInternalMap.getEntry(MapMakerInternalMap.java:3446)
at
com.google.gwt.thirdparty.guava.common.collect.Interners$WeakInterner.intern(Interners.java:76)
at com.google.gwt.dev.js.ast.JsScope.maybeMangleKeyword(JsScope.java:52)
at com.google.gwt.dev.js.ast.JsScope.findExistingName(JsScope.java:105)
at
com.google.gwt.dev.js.JsInliner$InliningVisitor.process(JsInliner.java:1264)
at
com.google.gwt.dev.js.JsInliner$InliningVisitor.endVisit(JsInliner.java:1036)
at com.google.gwt.dev.js.ast.JsInvocation.traverse(JsInvocation.java:73)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at
com.google.gwt.dev.js.ast.JsBinaryOperation.traverse(JsBinaryOperation.java:118)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at
com.google.gwt.dev.js.ast.JsBinaryOperation.traverse(JsBinaryOperation.java:118)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at
com.google.gwt.dev.js.ast.JsBinaryOperation.traverse(JsBinaryOperation.java:116)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAcceptList(JsModVisitor.java:183)
at com.google.gwt.dev.js.ast.JsVisitor.acceptList(JsVisitor.java:113)
at com.google.gwt.dev.js.ast.JsInvocation.traverse(JsInvocation.java:71)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAcceptList(JsModVisitor.java:183)
at com.google.gwt.dev.js.ast.JsVisitor.acceptList(JsVisitor.java:113)
at com.google.gwt.dev.js.ast.JsInvocation.traverse(JsInvocation.java:71)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at com.google.gwt.dev.js.ast.JsReturn.traverse(JsReturn.java:51)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$ListContext.traverse(JsModVisitor.java:80)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAcceptWithInsertRemove(JsModVisitor.java:197)
at
com.google.gwt.dev.js.ast.JsVisitor.acceptWithInsertRemove(JsVisitor.java:121)
at com.google.gwt.dev.js.ast.JsBlock.traverse(JsBlock.java:48)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at com.google.gwt.dev.js.ast.JsFunction.traverse(JsFunction.java:202)
at com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at 

Re: Compiling permutation hangs ?

[Ed Bras]

@Jens, thanks​
I just tried again with more memory, namely:

extraJvmArgs-Xms2G -Xmx2G -Xss1024k -XX:PermSize=512m
-XX:MaxPermSize=1024m -Dgwt.persistentunitcache=false/extraJvmArgs


And got the following output snippet below with jstack after about 30 min.
It's still busy, but doesn't complet...

I am not sure if I need to set the MAVEN_OPTS during CLI maven usage ? (I
do when running the job through Jenkins). I tired it before, but still not
luck, maybe I used it wrong...
I am not sure if the gwt maven plugin does fork a new thread/process that
uses part of the total memory assigned to maven, or that it uses the memory
as specified in the plugin config (as above).
I can imagine that if the assigned total maven mem is Max let say 512MB,
setting the mem max 2G in the plugin config might have no effect as it's
more then maven can allow the plugin to use...



In-process PermutationWorker 0 prio=10 tid=0x7f4bd86dc800 nid=0x6fcc
runnable [0x7f4bc60be000]
   java.lang.Thread.State: RUNNABLE
at
com.google.gwt.dev.js.ast.JsConditional.traverse(JsConditional.java:87)
at
com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at
com.google.gwt.dev.js.ast.JsBinaryOperation.traverse(JsBinaryOperation.java:118)
at
com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
at com.google.gwt.dev.js.ast.JsExprStmt.traverse(JsExprStmt.java:42)
at
com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$ListContext.traverse(JsModVisitor.java:80)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAcceptWithInsertRemove(JsModVisitor.java:197)
at
com.google.gwt.dev.js.ast.JsVisitor.acceptWithInsertRemove(JsVisitor.java:121)
at com.google.gwt.dev.js.ast.JsBlock.traverse(JsBlock.java:48)
at
com.google.gwt.dev.js.ast.JsVisitor.doTraverse(JsVisitor.java:468)
at
com.google.gwt.dev.js.ast.JsModVisitor$NodeContext.traverse(JsModVisitor.java:153)
at
com.google.gwt.dev.js.ast.JsModVisitor.doAccept(JsModVisitor.java:176)
at com.google.gwt.dev.js.ast.JsVisitor.accept(JsVisitor.java:109)
---

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

Re: Text markup (XML) parser in GWT?

[Ed Bras]

Jens: thanks for your feedback.

 Then override onBrowserEvent()
Yes, I could change it. You mean overriding the onBrowserEvent() of the
parent of the anchor element I suppose?
I would have to dig a bit deeper to see if it would cover all my
requirements, but sure a good idea. I remember doing this in the past but
without overriding the onBrowser event which didn't work well..

  I would parse it using GWT xml parser
​But how can I use the gwt xml parser to specify the opening/closing tags
as they are different then the xml tags ([, ] vs , )..

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

Re: Text markup (XML) parser in GWT?

[Ed Bras]

​@Jens: cool, thanks, I will have a look at the parser refs.

I though more about using the anchor/onBrowserEvent and it's to restricted
in some cases (I think that was the reason I didn't use it in the past as
well)
Example: through the tag name I can specify the widget that needs to be
created that can consists of several div elements.. (pretty powerful, I
like using 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+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/d/optout.

Re: Text markup (XML) parser in GWT?

[Ed Bras]

 HTMLPanel.addAndReplaceElement()
This is basically what I do, but then through customized markup.

But I think you are right, i can probably directly  use div/span as marker
with data-* attributes that hold extra details like references through
other cms text snippets that should be shown in for example a popup.
I think I got a bit tunnel vision through history of the legacy markup. I
will give it a try through direct div/span usages.
Thanks.
​

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

Re: Best practices for reducing generated RPC Serialization code?

[Ed Bras]

Thanks for the input.
Could you give me some more insight on how you use the command pattern?
(how it appears in the front/backend interface/classes, and flow)..

I am also using the command pattern in the front-end, not the backend. The
command is handled by a central ServiceHandler that run all commands, and a
command will call a rpc method directly.

I am also using concrete classes, like ArrayList instead of List.
But what about using the issue mentioned in SO (duplicated serialization
code for same class in different method in different Async interfaces)?

​

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

Re: Development Mode will not be supported in Firefox 27+

[Ed Bras]

Can't agree more. Same experience, even contributed to mygwt (previous name
of GXT)...
The nice thing about GWT: making easy/fast small lib's, is also a risk; not
being maintained..

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

Re: CssResource interface conflict?

[Ed Bras]

Cool, thanks for the test case, it makes it a lot easier.

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

Re: CssResource interface conflict?

[Ed Bras]

Thanks Thomas, I will gather some more info.

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

Re: Development Mode will not be supported in Firefox 27+

[Ed Bras]

 They have to change, they have to update their tools and move forward, or
quit doing Web dev.
Agree, that's why I spend a lot of grey hairs in using SDM, but it's simple
not there yet saidly... Don't get me wrong, I wish it would as I certainly
see the advantages.

I noticed that it's hard to understand these reasons by people that have
their focus on Hello world app's or hardly touch enterprise alike GWT
app's...

BTW: for none enterprise app's, I use others JS frameworks like JQuery.
GWT: heavier business logica (= call it Enterprise app's or some other
beautiful trendy name), share with backend...

Just my 50 cents to have a good understanding what we talk about..

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

Re: Development Mode will not be supported in Firefox 27+

[Ed Bras]

 I'm surprised more folks are not excited to jump ship to SuperDevMode.
I think because many dev peeps use GWT for enterprise projects.(and
they should of course)...
And for bigger projects, SDM is (still) hard to use (as explained in this
post, and I really tried/used it several times).
Really hoping this will change soon...

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

Re: Best alternative for Character.getNumericValue(ch)?

[Ed Bras]

@Thomas: thanks for the details, it makes it more clear.

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

Re: Best alternative for Character.getNumericValue(ch)?

[Ed Bras]

Thanks thomas, I will have a look at it. I am not sure which situations
this covers, and where you get this from, please give me a hint, I am not
so well known with this character encoding.

In the mean time, I super-sourced the Character class and implemented the
isWhiteSpace() and getNumericValue() method for latin support only (no
noticable code increase), in other cases I throw an exception, and will
probably never occur in my case. I like this solution as I have shared
classes that use these methods also in none-gwt code.

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

Re: How to super-source a single method in a JRE emulated class?

[Ed Bras]

Clear, thanks. @Jens

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

Re: sibling CSS selector ~ workaround?

[Ed Bras]

 ...because it uses the Flute CSS parser, that only supports CSS 2 syntax.
Have a look at gss.gwt if you want to use CSS 3:
  https://github.com/jDramaix/gss.gwt

@Thomas, thanks, just give it a try, after temporarily  removing all @def,
@eval, @if statements, It will run, but it will not inject my styles, the
elements simple have no styles anymore. Thought the ensureInjected() method
do got called way before usage... Strange... Can't seem to solve this..

Is there another way? Maybe Overriding the Generator to ignore this `
symbol? I looked at it, but couldn't really find 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+unsubscr...@googlegroups.com.
To post to this group, send email to google-web-toolkit@googlegroups.com.
Visit this group at http://groups.google.com/group/google-web-toolkit.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Development Mode will not be supported in Firefox 27+

[Ed Bras]

  The Chrome debugger is quite good and I recommend learning it well;
It depends on your project. I been down this road, tried it several times
with several projects (few months ago last one).
Currently the Super Dev mode is Hello World ready, but not Enterprise
ready

With new projects, I like to start with Super dev mode, but when it becomes
bigger... I am forced to jump back to FF/Chrome plugin usage.

Using it with bigger code base makes it slow and a code puzzle in chrome
debug.
For bigger projects, Chrome debugger should offer more IDE features that
make debugging easy and friendly, just like we are used to using IDE tools
like Eclipse.
Of course this will change, and I hope this will be soon..

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

Re: Excluding permutation compilation combinations?

[Ed Bras]

Thanks Thomas, that seem to work.

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

Re: Development Mode will not be supported in Firefox 27+

[Ed Bras]

 Remote debug works great too ;-)
Yes, if you have a Mac ;)  (I assume you mean Safari remote debugging ?
Else let me know, device independent (iphone, android))


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

Re: seedTable error in web mode?

[Ed Bras]

Thanks, Let's hope you right (I first reviewed/improved the split points,
but that didn't help).
BTW: I found it a bit scary that this error doesn't show itself in dev mode.

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

Re: Why are GWT widgets heavy?

[Ed Bras]

@Thomas: thanks for the insight, exactly what I was looking for. I wasn't
aware that DOM manipulations could have such a cost.
It's could to be more aware of this during development, especially for
mobile apps.

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