Nabble - groovy - user

Re: default button in swingbuilder

2009-11-09T11:06:03Z

In that case move the code to set the focus inside the dialog as a do later, and do a requestFocusInWindow(), and use the doLater closure...

dialog( ..., show:true, modal:true, ...) {
//...
myButton = button( ... defaultButton:true ...)
//...
doLater {button.requestFocusInWindow()}
}

The show handler runs after the complete execution of the window. So assuming you are running the builder in the EDT the doLater will result in a event being fired on the EDT. I haven't tried this yet so there may be some wierdness with the way modal dialogs intercept the EDT.

Another option is the the listeners. windowActivated and windowGainedFocus are the two you would want to try.

dialog ( ..., show:true, windowGainedFocus {myButton.requestFocusInWindow}, ...) {
myButton = button( ... defaultButton:true ...)
}

as long as myButton is unbound the above should work.

On Mon, Nov 9, 2009 at 7:14 AM, David Rosenstark <David.Rosenstark@...> wrote:

Sorry, but I am still not clear on this.

If I have a modal dialog (as is my case) then I will not be able to run additional code to set focus. Is there on onFocus of the window?

How do I catch such events in groovy?

Thanks,

David

From: Danno Ferrin [mailto:danno.ferrin@...]
Sent: Thursday, November 05, 2009 9:57 PM

To: user@...
Subject: Re: [groovy-user] default button in swingbuilder

I would do the request after the window deceleration...

dialog( ..., show:true ...) {
myButton = button( ... defaultButton:true ...)
}

button.requestFocus()

On Wed, Nov 4, 2009 at 1:35 PM, David Rosenstark <David.Rosenstark@...> wrote:

Thanks. I am still not clear – I want to open a dialog on top of main window and set focus to this button. So, when I hit show, I need to have already requested focus for it?

From: Danno Ferrin [mailto:danno.ferrin@...]
Sent: Wednesday, November 04, 2009 8:00 PM

To: user@...
Subject: Re: [groovy-user] default button in swingbuilder

The "default button" only works when the focus is already on the window but not on any other button, for example:

new groovy.swing.SwingBuilder().edt {
frame(pack:true, show:true, defaultCloseOperation:javax.swing.JFrame.DISPOSE_ON_CLOSE) {
flowLayout()
textField("Yo ho ho!")
button("one")
button("two", defaultButton:true, actionPerformed: {e->println 'yo!'})
button("Three")
}
}

if you place the cursor in the text component and press enter, 'yo' will fire. But if it is on any other button or not in the frame, nothing happens.

What you probably want is keyboard focus and window focus. To do this you need to track the button and possibly the dialog/frame. The swing requestFocusInWindow and requestFocus methods can help there:

frame {
   //....
   myButton = button(action: okAction, defaultButton) ...
}

myButton.requestFocusInWindow()
// Or...
myButton.requestFocus()

The difference is that requestFocusInWindow() will not attempt to cause the root level component to gain focus, whill requestFocus() will attempt. And by attempt, some OS window managers may suppress that call and simply cause the doc icon to bounce or the taskbar to flash for that task, I don't rememeber the specific rules and specific cases. If your java app already has focus, it should shift to the reqesting window.

On Wed, Nov 4, 2009 at 5:01 AM, David Rosenstark <David.Rosenstark@...> wrote:

Hi,

I have a simple window created using swingbuilder

I have a few buttons and would like to have the focus on the last button so that the user can hit enter and it closes the window with all the defaults as set.

I have tried defaultButton and selected but neither of these seems to work. Any ideas?

button(action: action(name: 'Select All', closure: {

            list1.selectedIndices = (0 .. list1.model.size - 1) as int[]

          }))

          button(action: action(name: 'DeSelect All', closure:{

                              list1.selectedIndices = [] as int[]

             }))

           button(action: action(name: 'OK', defaultButton: true, selected: true,

                  closure:{

                          println("button clicked")

                          list1.selectedValues.each {

                            println(it)

                          dispose()

                        } }))

--
------------------------------------------------------
"But you didn't." - Jim Halpert, The Office S05E23

--
------------------------------------------------------
"But you didn't." - Jim Halpert, The Office S05E23

--
------------------------------------------------------
"But you didn't." - Jim Halpert, The Office S05E23

Re: On static compilation of Groovy

2009-11-09T10:58:27Z

Alex Tkachman schrieb:
> Last experiments with GPars (results can be probably found in GPars
> list or easily reproduced) shows that groovy brings to serious
> overhead when using multi-thread-ly.

I agree partially. It is not because of Groovy performance in general,
it is because for each method invocation there will be a synchronization
and that leads to real bad performance if many Threads are running. In
other words the current runtime imposes a massive communication overhead
per Thread. In that regard it would not matter if single thread
performance of Groovy would be as fast as Java. But I am positive, that
those synchronizations can be removed and then the performance for
multithreading will be much better. Also the issue with the overhead is
partially related to how gpars is doing the processing and for example
Grails is much less influenced. So if Groovy does scale or not depends
on how you do things.

bye blackdrag

--
Jochen "blackdrag" Theodorou
The Groovy Project Tech Lead (http://groovy.codehaus.org)
http://blackdragsview.blogspot.com/

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: On static compilation of Groovy

2009-11-09T10:25:21Z

Last experiments with GPars (results can be probably found in GPars
list or easily reproduced) shows that groovy brings to serious
overhead when using multi-thread-ly.
I don't want to go in any "negative" discussions as my approach is
100% positive - both static and dynamic code can and should coexist
together

On Mon, Nov 9, 2009 at 8:09 PM, David Clark <davidclark@...> wrote:

> In your article you make the assertion, "due to performance reasons Groovy is
> almost useless today for multi-core programming" without giving any reasons or
> data to defend the assertion. I can't think of why this assertion should
> hold. Of course Groovy is inappropriate for computation intensive code, but
> this holds for any number of threads and any number of cores. Why would
> properly parallelized Groovy code be inappropriate for code that engages in
> lots of IO, which is the bulk of what business oriented code does?
>
> As for statically compiling Groovy code I'm all for it provided that 1) It's
> entirely transparent for end users, 2) It does not increase the complexity of
> the Groovy compiler thereby increasing bugs and slowing down development of
> other features, and 3) It changes none of the semantics of current Groovy
> code. I doubt those three things are possible, but I am willing to be proved
> wrong.
>
> Also, how many of the speed problems with dynamic dispatch will be alleviated
> with the new invokedynamic that is supposed to start appearing in JVM's
> sometime in the future?
>
> Also, if the objections to static compiling that I raised above cannot be met
> there are alternatives to speeding up Groovy code that seem easier to
> implement but would give big speedups to performance critical code. One would
> be to allow embedding java or scala code like C/C++ compiles allow embedding
> of ASM code. Granted it is already possible to mix and match Groovy, Java,
> and Scala, so this doesn't provide anything that can't already be done, it
> simply provides convenience to the programmer. Another solution would be to
> provide hints (via Annotations or the like) like Common Lisp allows. This
> would allow small compute intensive sections of code to go really fast, but
> seems much simpler than overhauling the compiler to produce statically
> dispatched code.
>
> On Monday 09 November 2009, Alex Tkachman wrote:
>> Hi guys,
>>
>> I've wrote small article summarizing my thoughts and prototype
>> development of optional static compilation of Groovy
>> http://groovy.dzone.com/articles/groovy-and-static-compilation
>>
>> I am really interested in opinions.
>>
>> Best regards
>> Alex
>>
>> ---------------------------------------------------------------------
>> To unsubscribe from this list, please visit:
>>
>> http://xircles.codehaus.org/manage_email
>>
>
> --
> David
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email
>
>
>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: On static compilation of Groovy

2009-11-09T10:09:48Z

In your article you make the assertion, "due to performance reasons Groovy is
almost useless today for multi-core programming" without giving any reasons or
data to defend the assertion. I can't think of why this assertion should
hold. Of course Groovy is inappropriate for computation intensive code, but
this holds for any number of threads and any number of cores. Why would
properly parallelized Groovy code be inappropriate for code that engages in
lots of IO, which is the bulk of what business oriented code does?

As for statically compiling Groovy code I'm all for it provided that 1) It's
entirely transparent for end users, 2) It does not increase the complexity of
the Groovy compiler thereby increasing bugs and slowing down development of
other features, and 3) It changes none of the semantics of current Groovy
code. I doubt those three things are possible, but I am willing to be proved
wrong.

Also, how many of the speed problems with dynamic dispatch will be alleviated
with the new invokedynamic that is supposed to start appearing in JVM's
sometime in the future?

Also, if the objections to static compiling that I raised above cannot be met
there are alternatives to speeding up Groovy code that seem easier to
implement but would give big speedups to performance critical code. One would
be to allow embedding java or scala code like C/C++ compiles allow embedding
of ASM code. Granted it is already possible to mix and match Groovy, Java,
and Scala, so this doesn't provide anything that can't already be done, it
simply provides convenience to the programmer. Another solution would be to
provide hints (via Annotations or the like) like Common Lisp allows. This
would allow small compute intensive sections of code to go really fast, but
seems much simpler than overhauling the compiler to produce statically
dispatched code.

On Monday 09 November 2009, Alex Tkachman wrote:

> Hi guys,
>
> I've wrote small article summarizing my thoughts and prototype
> development of optional static compilation of Groovy
> http://groovy.dzone.com/articles/groovy-and-static-compilation
>
> I am really interested in opinions.
>
> Best regards
> Alex
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email
>

--
David

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: On static compilation of Groovy

2009-11-09T09:35:28Z

Alex Tkachman schrieb:
> Hi guys,
>
> I've wrote small article summarizing my thoughts and prototype
> development of optional static compilation of Groovy
> http://groovy.dzone.com/articles/groovy-and-static-compilation
>
> I am really interested in opinions.

I wonder how you would make something like Grails without meta
programming. If you say you would not, then what exactly would be the
benefit for for example Grails.

Our builders for example are surely possible in a static language, but
the details are what makes it a problem. For example:

foo { bar() }

In Groovy bar might be on the delegate, on this or some dynamic stuff.
You can control it by setting the resolve strategy. In a static language
with still a delegate you would have to route the method call to a
method missing method and from there you would have to do the method
call to the real method. But if you do so, you will have to use
Reflection with a performance even worse than what groovy now has. Also
what if bar does not exist? If you don't know the delegate then how to
check it? the call would have to be modified to something like:

foo <U,V>{-> bar() }

With for example U for the return type and V for the delegate type. Only
then you can check. Or the alternative is that foo gives the types and
you save the this here. But then the method does have to exist on either
this or the delegate and the resolving strategy is preset by the
compiler too. Something like StreamingMarkupBuilder would become
impossible.

So either you decide for method missing and loose performance or you
don't, then you loose method missing and get problems with the resolving
strategy as well as having to have methods on the builder for each
possibility.

I should maybe also mention that there is a major difference between
{1+2} and {bar()}
The first case is more or less an anonymous function. They don't need an
outer scope to close over and are not closures. They won't have the
problem with the delegate, but they are of no use for builder. The other
form needs an outside scope to define bar somehow. In Groovy this scope
is applied at runtime, making it more or less a closure.

So I claim that runtime metaprogramming, multimethods (as they are now
in Groovy), and builder do require dynamic dispatch.

In your article you mention a mixed mode. Well that is more or less what
I target in 1.8. There static dispatch will be done if possible and
dynamic dispatch when needed. The use of runtime meta programming will
then in some cases turn of the static dispatch. If most of the dispatch
is still done static, the performance should still be great. And this
approach will not need any annotations.

Of course I don't know yet if it will work as good as I think it could
(overhead under 100%), but if it does it will be more than good enough.

bye blackdrag

--
Jochen "blackdrag" Theodorou
The Groovy Project Tech Lead (http://groovy.codehaus.org)
http://blackdragsview.blogspot.com/

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

On static compilation of Groovy

2009-11-09T08:21:14Z

Hi guys,

I've wrote small article summarizing my thoughts and prototype
development of optional static compilation of Groovy
http://groovy.dzone.com/articles/groovy-and-static-compilation

I am really interested in opinions.

Best regards
Alex

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-09T06:56:04Z

tog wrote:
> The initial file does not go through the CXF wsdl2java :( Is this the
> one having an extra " ?

That would be the udiws10.ws file that has the extra " .

> If yes can you tell me where is is located ?

http://www.linuxha.com/common/udiws10.ws

Normally it would be referenced from:

http://isy.uucp/services.wsdl

Now I've noted that at least one person attempted to use GroovyWS
to access the services.wsdl and the udiws10.ws via my linuxha.com
site (thanks, I really do appreciate the effort) so I also put
the files in the following locations that should work:

http://www.linuxha.com/services.wsdl

and

http://www.linuxha.com/WEB/udiws10.ws

I think that will work and I just tested it failed similarly to
the actual local isy.uucp device.

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: GroovyWS proxy.create() - class package determination?

2009-11-09T06:34:37Z

not sure on your environment. If you are in grails, you can change your /conf/config.groovy to something like this. Otherwise google for log4j configuration.

// log4j configuration
log4j = {
// Example of changing the log pattern for the default console
// appender:
//
appenders {
console name:'stdout', layout:pattern(conversionPattern: '%c{2} %m%n')
}

error 'org.codehaus.groovy.grails.web.servlet', // controllers
'org.codehaus.groovy.grails.web.pages', // GSP
'org.codehaus.groovy.grails.web.sitemesh', // layouts
'org.codehaus.groovy.grails.web.mapping.filter', // URL mapping
'org.codehaus.groovy.grails.web.mapping', // URL mapping
'org.codehaus.groovy.grails.commons', // core / classloading
'org.codehaus.groovy.grails.plugins', // plugins
'org.codehaus.groovy.grails.orm.hibernate', // hibernate integration
'org.springframework',
'org.hibernate'

debug 'grails.app',
'groovyx.net.ws',
'org.apache.cxf' // <<<<<<<<

warn 'org.mortbay.log',
'org.codehaus.groovy.grails.plugins.starksecurity'

rootLogger="stdout"
}

chad.

Neil Cherry-3 wrote:

casmall wrote:
> found an answer to this one.
>
> Just make sure you have info log turned on for org.apache.cxf ...
>
> debug 'grails.app',
> 'groovyx.net.ws',
> 'org.apache.cxf'
>
> and out will pop a nice log message giving the available classes with
> package (as part of proxy.initialize()):
>
> dynamic.DynamicClientFactory Created classes:
> com.company.caqh.services.model.beans.realtime.CAQHRealtimeInput, etc...
>
> chad.

Chad, how do I turn on debug?

Thanks

--
Linux Home Automation Neil Cherry ncherry@linuxha.com
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

RE: default button in swingbuilder

2009-11-09T06:14:41Z

Sorry, but I am still not clear on this.

If I have a modal dialog (as is my case) then I will not be able to run additional code to set focus. Is there on onFocus of the window?

How do I catch such events in groovy?

Thanks,

David

From: Danno Ferrin [mailto:danno.ferrin@...]
Sent: Thursday, November 05, 2009 9:57 PM
To: user@...
Subject: Re: [groovy-user] default button in swingbuilder

I would do the request after the window deceleration...

dialog( ..., show:true ...) {
myButton = button( ... defaultButton:true ...)
}

button.requestFocus()

On Wed, Nov 4, 2009 at 1:35 PM, David Rosenstark <David.Rosenstark@...> wrote:

Thanks. I am still not clear – I want to open a dialog on top of main window and set focus to this button. So, when I hit show, I need to have already requested focus for it?

From: Danno Ferrin [mailto:danno.ferrin@...]
Sent: Wednesday, November 04, 2009 8:00 PM

To: user@...
Subject: Re: [groovy-user] default button in swingbuilder

The "default button" only works when the focus is already on the window but not on any other button, for example:

new groovy.swing.SwingBuilder().edt {
frame(pack:true, show:true, defaultCloseOperation:javax.swing.JFrame.DISPOSE_ON_CLOSE) {
flowLayout()
textField("Yo ho ho!")
button("one")
button("two", defaultButton:true, actionPerformed: {e->println 'yo!'})
button("Three")
}
}

if you place the cursor in the text component and press enter, 'yo' will fire. But if it is on any other button or not in the frame, nothing happens.

What you probably want is keyboard focus and window focus. To do this you need to track the button and possibly the dialog/frame. The swing requestFocusInWindow and requestFocus methods can help there:

frame {
//....
myButton = button(action: okAction, defaultButton) ...
}

myButton.requestFocusInWindow()
// Or...
myButton.requestFocus()

The difference is that requestFocusInWindow() will not attempt to cause the root level component to gain focus, whill requestFocus() will attempt. And by attempt, some OS window managers may suppress that call and simply cause the doc icon to bounce or the taskbar to flash for that task, I don't rememeber the specific rules and specific cases. If your java app already has focus, it should shift to the reqesting window.

On Wed, Nov 4, 2009 at 5:01 AM, David Rosenstark <David.Rosenstark@...> wrote:

Hi,

I have a simple window created using swingbuilder

I have a few buttons and would like to have the focus on the last button so that the user can hit enter and it closes the window with all the defaults as set.

I have tried defaultButton and selected but neither of these seems to work. Any ideas?

button(action: action(name: 'Select All', closure: {

list1.selectedIndices = (0 .. list1.model.size - 1) as int[]

}))

button(action: action(name: 'DeSelect All', closure:{

list1.selectedIndices = [] as int[]

}))

button(action: action(name: 'OK', defaultButton: true, selected: true,

closure:{

println("button clicked")

list1.selectedValues.each {

println(it)

dispose()

} }))

--
------------------------------------------------------
"But you didn't." - Jim Halpert, The Office S05E23

RE: FW: [grails-user] What's the proper way to invoke super.beforeInsert() for a Domain class?

2009-11-09T05:05:13Z

Good idea – I’ll give that a shot. On the Grails list, I also got the following suggestion from Burt Beckwith:

Have beforeInsert delegate to a real method:

class A {

//Some properties

protected void onBeforeInsert() {

//Do something interesting

}

def beforeInsert = {

onBeforeInsert()

}

class B extends A {

//Some added properties

@Override

protected void onBeforeInsert() {

super.onBeforeInsert()

//Do additional interesting things

}

Both seem obvious now that I see them. Thanks!

Matt

From: roshandawrani@... [mailto:roshandawrani@...] On Behalf Of Roshan Dawrani
Sent: Sunday, November 08, 2009 7:54 AM
To: user@...
Subject: Re: [groovy-user] FW: [grails-user] What's the proper way to invoke super.beforeInsert() for a Domain class?

You can do it like:

=================================
class A {
    def beforeInsert = {
        println "A->beforeInsert()"
    }
}

class B extends A {
    def getSuperClosure() {
        super.getBeforeInsert()
    }
    def beforeInsert = {
        getSuperClosure()()
        println "B->beforeInsert()"
    }
}

new B().beforeInsert()
=================================

Hope it helps.

-- Roshan

On Sat, Nov 7, 2009 at 2:40 AM, <Matthew.Lachman@...> wrote:

I asked this question on the Grails user list but perhaps it's more of a
Groovy question. How to I invoke a super class's closure that has the
same name as "this" class's closure?

-----Original Message-----
From: Lachman, Matt (West)
Sent: Friday, November 06, 2009 3:01 PM
To: user@...
Subject: [grails-user] What's the proper way to invoke
super.beforeInsert() for a Domain class?

I have domain classes similar to the following:

class A {
//Some properties

def beforeInsert = {
//Do something interesting
}
}

class B extends A {
//Some added properties

def beforeInsert = {
super.beforeInsert() //doesn't actually call
A.beforeInsert()
//Do additional interesting things
}
}

What's the proper way to achieve the effect I'm looking for?

Thanks,
Matt

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Creating a List

2009-11-09T03:37:45Z

I'd use JsonSlurper (part of json-lib):

@Grab('net.sf.json-lib:json-lib:2.3:jdk15')
import net.sf.json.groovy.JsonSlurper

JsonSlurper slurper = new JsonSlurper()

def res = slurper.parseText( "[{\"name\":\"A new member\",\"age\":21,\"cover\":\"Single\",\"londonUpgrade\":0},{\"name\":\"A new member\",\"age\":21,\"cover\":\"Single\",\"excess\":\"\"}]" )

println res

Tim

On Mon, Nov 9, 2009 at 9:55 AM, douggiefox <paul@...> wrote:

Hi

I have a string that's come from a HTML page and is encoded like this:

"[{\"name\":\"A new
member\",\"age\":21,\"cover\":\"Single\",\"londonUpgrade\":0},{\"name\":\"A
new member\",\"age\":21,\"cover\":\"Single\",\"excess\":\"\"}]"

Is there a simple way I can convert this to a List?

Best

DF
--
View this message in context: http://old.nabble.com/Creating-a-List-tp26263681p26263681.html
Sent from the groovy - user mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Creating a List

2009-11-09T01:55:19Z

Hi

I have a string that's come from a HTML page and is encoded like this:

"[{\"name\":\"A new member\",\"age\":21,\"cover\":\"Single\",\"londonUpgrade\":0},{\"name\":\"A new member\",\"age\":21,\"cover\":\"Single\",\"excess\":\"\"}]"

Is there a simple way I can convert this to a List?

Best

DF

Re: Problems with GroovyWS

2009-11-09T00:35:33Z

The initial file does not go through the CXF wsdl2java :( Is this the one having an extra " ?
If yes can you tell me where is is located ?

On Mon, Nov 9, 2009 at 10:56 AM, Neil Cherry <ncherry@...> wrote:

tog wrote:
> Ok I guess you are missing the service part with the port in your wsdl.
> It should be something like this:
>
> <wsdl:service name="TerraService">
> <wsdl:port name="TerraServiceSoap" binding="tns:TerraServiceSoap">
> <soap:address location="http://terraservice.net/TerraService.asmx"/>
> </wsdl:port>
> </wsdl:service>

Oh, sorry, I tried so many different things that I last tried the
previous file. Here's the initial file:

http://www.linuxha.com/common/services.wsdl

Normally it would be called with:

http://isy.uucp/services.wsdl

--

Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

--
PGP KeyID: 1024D/69B00854 subkeys.pgp.net

http://cheztog.blogspot.com

Re: metaclass with enum

2009-11-08T22:46:02Z

Good to hear that. I get my time's worth, if you are continuing to go on with groovy. :-)

Have fun.
Roshan

On Mon, Nov 9, 2009 at 12:12 PM, Ista Pouss <istaous@...> wrote:

OK.

I understand the expert point of view but, from a newbee point of
view, it's a curious feature !

Planty of thanks, I go on with groovy.

2009/11/7 Roshan Dawrani <roshandawrani@...>:
> Shamelessly mailing one more version. The last one!
>
> ===================================================
> enum Ville
> {
>     TOKIO, PARIS
> }
>
> Ville.metaClass.parcouru = Integer.MAX_VALUE
> Ville.metaClass.précédent = 0
>
> // refresh metaclass of each instance with the latest class level metaclass
> Ville.values().each {enumInstance ->
>     enumInstance.metaClass = null
> }
>
> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
> assert Ville.TOKIO.précédent == 0
>
> Ville.TOKIO.parcouru = 10
> assert Ville.TOKIO.parcouru == 10
> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>
> println "Done"
> ===================================================
>
> rgds,
> Roshan
>
> On Sat, Nov 7, 2009 at 3:15 PM, Roshan Dawrani <roshandawrani@...>
> wrote:
>>
>> Slightly improved version of the last solution (removes the enum's
>> metaClass field and getMetaClass() method). Sorry for so many versions. I am
>> trying metaclass changes on enums for the first time.
>>
>> ========================================
>> enum Ville
>> {
>>     TOKIO, PARIS
>>
>>     def addProperties(propMap) {
>>         def emc = new ExpandoMetaClass(Ville, true, true);
>>         setMetaClass(emc)
>>         def itr = propMap.keySet().iterator()
>>         while(itr.hasNext()) {
>>             def propName = itr.next()
>>             this.metaClass."$propName" = propMap[propName]
>>         }
>>         emc.initialize();
>>     }
>> }
>>
>> Ville.values().each { enumInstance ->
>>     enumInstance.addProperties([parcouru:Integer.MAX_VALUE, précédent:0])
>> }
>>
>> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
>> assert Ville.TOKIO.précédent == 0
>>
>> Ville.TOKIO.parcouru = 10
>> assert Ville.TOKIO.parcouru == 10
>> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>> println "Done"
>> ========================================
>>
>> On Sat, Nov 7, 2009 at 2:52 PM, Roshan Dawrani
>> <roshandawrani@...> wrote:
>>>
>>> Here is another solution, which changes the metaClass of enum instances
>>> at runtime without needing to do a global change by calling
>>> EMC.enableGlobally() :-
>>>
>>> ================================================
>>> enum Ville
>>> {
>>>     TOKIO, PARIS
>>>     def metaClass
>>>     MetaClass getMetaClass() {
>>>         if(!metaClass) {
>>>             metaClass = new ExpandoMetaClass(Ville, true, true);
>>>             metaClass.initialize()
>>>         }
>>>         metaClass
>>>     }
>>>     def addProperty(propName, propValue) {
>>>         metaClass = new ExpandoMetaClass(Ville, true, true);
>>>         this.metaClass."$propName" = propValue
>>>         metaClass.initialize();
>>>     }
>>> }
>>>
>>> // make the metaclass changes on the enum instances
>>> Ville.values().each { enumInstance ->
>>>     enumInstance.addProperty('parcouru', Integer.MAX_VALUE)
>>>     enumInstance.addProperty('précédent', 0)
>>> }
>>>
>>> // check if the enum has now got the new Properties
>>> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
>>> assert Ville.TOKIO.précédent == 0
>>>
>>> // change runtime properties and check on different instances
>>> Ville.TOKIO.parcouru = 10
>>> assert Ville.TOKIO.parcouru == 10
>>> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>>> println "Done"
>>> =======================================
>>>
>>> Hope it helps.
>>> Roshan
>>>
>>> On Sat, Nov 7, 2009 at 2:26 PM, Roshan Dawrani
>>> <roshandawrani@...> wrote:
>>>>
>>>> Ok, now I get the question fully.
>>>>
>>>> You want to define new properties on the enum but not from inside the
>>>> enum as I first suggested but at runtime.
>>>>
>>>> I don't know at the moment a good, clean answer to that question -
>>>> because enums are a little special and their instance are created internally
>>>> when enum class is loaded, so the changes that you make to enum's metaclass
>>>> after the enum is fully loaded (and its instances are constructed) are not
>>>> seen by the enum instances like TOKIO.
>>>>
>>>> Till any better answer comes along, one possible solution is below.
>>>>
>>>> =================================
>>>> enum Ville
>>>> {
>>>>     TOKIO, PARIS
>>>> }
>>>> // after this already constructed enum instances still pick up enum
>>>> class level metaclass changes made at runtime
>>>> ExpandoMetaClass.enableGlobally()
>>>>
>>>> Ville.metaClass.cool = 10
>>>>
>>>> println Ville.PARIS.cool
>>>>
>>>> // after this enum instance will go back to its own MC again and not see
>>>> enum class level metaclass changes made at runtime
>>>> ExpandoMetaClass.disbleGlobally()
>>>>
>>>> println Ville.PARIS.cool // will fail again
>>>> =================================
>>>>
>>>> -- Roshan
>>>>
>>>> On Sat, Nov 7, 2009 at 2:06 PM, Ista Pouss <istaous@...> wrote:
>>>>>
>>>>> Yes, but how do that on the enum itself ?
>>>>>
>>>>> As an exemple, I do this code in the groovy web console :
>>>>>
>>>>> class A {}
>>>>> enum Ville {TOKIO, PARIS}
>>>>>
>>>>> assert Ville.TOKIO instanceof Ville // of course.
>>>>>
>>>>> A.metaClass.cool = 10
>>>>>
>>>>> a = new A()
>>>>> assert a.cool == 10
>>>>>
>>>>> Ville.metaClass.cool = 10
>>>>> assert Ville.TOKIO.cool == 10 // BING !
>>>>>
>>>>> I do exactly the same metaClass thing, on a "regular" class (A), and
>>>>> on an enum (Ville). It's OK for instance class A (a), not for instance
>>>>> Ville (TOKIO).
>>>>>
>>>>> The error in groovy web console (http://groovyconsole.appspot.com/) is
>>>>> :
>>>>>
>>>>> groovy.lang.MissingPropertyException: No such property: cool for class:
>>>>> Ville
>>>>> at Script1.run(Script1.groovy:12)
>>>>>
>>>>>
>>>>>
>>>>> 2009/11/7 Roshan Dawrani <roshandawrani@...>:
>>>>> > In the first mail you wanted to access Ville.PARIS.parcouru, which
>>>>> > meant
>>>>> > that you wanted to have "parcouru" on the enum constant, which was of
>>>>> > type
>>>>> > Ville.
>>>>> >
>>>>> > If you want to add property "parcouru" at runtime on some regular
>>>>> > class,
>>>>> > then you can very easily do it. Code below shows 2 ways of doing it:
>>>>> > =======================================================
>>>>> > enum Ville {TOKIO, PARIS}
>>>>> > class AlgoTown
>>>>> > {
>>>>> >     Ville ville;
>>>>> > }
>>>>> >
>>>>> > String runtimePropName = "précédent"
>>>>> > addPropertyWhoseNameIsKnown()
>>>>> > addPropertyWhoseNameIsNotKnownUntilRuntime(runtimePropName)
>>>>> >
>>>>> > def addPropertyWhoseNameIsKnown() {
>>>>> >     AlgoTown.metaClass.parcouru = Integer.MAX_VALUE
>>>>> > }
>>>>> > def addPropertyWhoseNameIsNotKnownUntilRuntime(propName) {
>>>>> >     AlgoTown.metaClass."$propName" = 0
>>>>> > }
>>>>> >
>>>>> > def algo1 = new AlgoTown()
>>>>> > def algo2 = new AlgoTown()
>>>>> >
>>>>> > println algo1.parcouru // Integer.MAX_VALUE
>>>>> > println algo1."$runtimePropName" // 0
>>>>> >
>>>>> > algo1.parcouru = 10 // change algo1's parcouru property to 10
>>>>> >
>>>>> > println algo1.parcouru // 10
>>>>> > println algo2.parcouru // algo2 still has parcouru =
>>>>> > Integer.MAX_VALUE
>>>>> > =======================================================
>>>>> >
>>>>> > Hope it helps.
>>>>> > Roshan
>>>>> >
>>>>> >
>>>>> > On Sat, Nov 7, 2009 at 12:49 PM, Ista Pouss <istaous@...>
>>>>> > wrote:
>>>>> >>
>>>>> >> 2009/11/7 Roshan Dawrani <roshandawrani@...>:
>>>>> >> > Hi,
>>>>> >> > If you don't have to add enum properties parcouru and précédent
>>>>> >> > dynamically
>>>>> >> > at run-time, then you can do it like:
>>>>> >> >
>>>>> >> > ==========================================================
>>>>> >> > enum Ville
>>>>> >> > {
>>>>> >> >     TOKIO, MAKAO, OSLO, PARIS, NANTES, ZARASTAPARTAKIS
>>>>> >> >     Ville() {
>>>>> >> >         parcouru = Integer.MAX_VALUE
>>>>> >> >         précédent = 0
>>>>> >> >     }
>>>>> >> >     Integer parcouru
>>>>> >> >     Integer précédent
>>>>> >> > }
>>>>> >>
>>>>> >> Yes, but these properties are not "ville" (ie: town)'s properties.
>>>>> >> It's properties for the algorithm.
>>>>> >>
>>>>> >> In java, I make something like that in this case :
>>>>> >>
>>>>> >> class AlgoTown // stupid name, sorry
>>>>> >> {
>>>>> >> Ville ville; // it's the town
>>>>> >> int parcouru;
>>>>> >> int précédent;
>>>>> >> }
>>>>> >>
>>>>> >> With groovy (i'm a neewbee), my impression is : it is possible to
>>>>> >> add
>>>>> >> properties at runtime with metaclass (and expando ? ) thing. Is it a
>>>>> >> rigth / good impression ? Is it The Good Thing To Do in my case ? Is
>>>>> >> it possible with "enum" ? How ?
>>>>> >>
>>>>> >>
>>>>> >> ---------------------------------------------------------------------
>>>>> >> To unsubscribe from this list, please visit:
>>>>> >>
>>>>> >> http://xircles.codehaus.org/manage_email
>>>>> >>
>>>>> >>
>>>>> >
>>>>> >
>>>>>
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe from this list, please visit:
>>>>>
>>>>> http://xircles.codehaus.org/manage_email
>>>>>
>>>>>
>>>>
>>>
>>
>
>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: metaclass with enum

2009-11-08T22:42:33Z

OK.

I understand the expert point of view but, from a newbee point of
view, it's a curious feature !

Planty of thanks, I go on with groovy.

2009/11/7 Roshan Dawrani <roshandawrani@...>:

> Shamelessly mailing one more version. The last one!
>
> ===================================================
> enum Ville
> {
>     TOKIO, PARIS
> }
>
> Ville.metaClass.parcouru = Integer.MAX_VALUE
> Ville.metaClass.précédent = 0
>
> // refresh metaclass of each instance with the latest class level metaclass
> Ville.values().each {enumInstance ->
>     enumInstance.metaClass = null
> }
>
> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
> assert Ville.TOKIO.précédent == 0
>
> Ville.TOKIO.parcouru = 10
> assert Ville.TOKIO.parcouru == 10
> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>
> println "Done"
> ===================================================
>
> rgds,
> Roshan
>
> On Sat, Nov 7, 2009 at 3:15 PM, Roshan Dawrani <roshandawrani@...>
> wrote:
>>
>> Slightly improved version of the last solution (removes the enum's
>> metaClass field and getMetaClass() method). Sorry for so many versions. I am
>> trying metaclass changes on enums for the first time.
>>
>> ========================================
>> enum Ville
>> {
>>     TOKIO, PARIS
>>
>>     def addProperties(propMap) {
>>         def emc = new ExpandoMetaClass(Ville, true, true);
>>         setMetaClass(emc)
>>         def itr = propMap.keySet().iterator()
>>         while(itr.hasNext()) {
>>             def propName = itr.next()
>>             this.metaClass."$propName" = propMap[propName]
>>         }
>>         emc.initialize();
>>     }
>> }
>>
>> Ville.values().each { enumInstance ->
>>     enumInstance.addProperties([parcouru:Integer.MAX_VALUE, précédent:0])
>> }
>>
>> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
>> assert Ville.TOKIO.précédent == 0
>>
>> Ville.TOKIO.parcouru = 10
>> assert Ville.TOKIO.parcouru == 10
>> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>> println "Done"
>> ========================================
>>
>> On Sat, Nov 7, 2009 at 2:52 PM, Roshan Dawrani
>> <roshandawrani@...> wrote:
>>>
>>> Here is another solution, which changes the metaClass of enum instances
>>> at runtime without needing to do a global change by calling
>>> EMC.enableGlobally() :-
>>>
>>> ================================================
>>> enum Ville
>>> {
>>>     TOKIO, PARIS
>>>     def metaClass
>>>     MetaClass getMetaClass() {
>>>         if(!metaClass) {
>>>             metaClass = new ExpandoMetaClass(Ville, true, true);
>>>             metaClass.initialize()
>>>         }
>>>         metaClass
>>>     }
>>>     def addProperty(propName, propValue) {
>>>         metaClass = new ExpandoMetaClass(Ville, true, true);
>>>         this.metaClass."$propName" = propValue
>>>         metaClass.initialize();
>>>     }
>>> }
>>>
>>> // make the metaclass changes on the enum instances
>>> Ville.values().each { enumInstance ->
>>>     enumInstance.addProperty('parcouru', Integer.MAX_VALUE)
>>>     enumInstance.addProperty('précédent', 0)
>>> }
>>>
>>> // check if the enum has now got the new Properties
>>> assert Ville.TOKIO.parcouru == Integer.MAX_VALUE
>>> assert Ville.TOKIO.précédent == 0
>>>
>>> // change runtime properties and check on different instances
>>> Ville.TOKIO.parcouru = 10
>>> assert Ville.TOKIO.parcouru == 10
>>> assert Ville.PARIS.parcouru == Integer.MAX_VALUE
>>> println "Done"
>>> =======================================
>>>
>>> Hope it helps.
>>> Roshan
>>>
>>> On Sat, Nov 7, 2009 at 2:26 PM, Roshan Dawrani
>>> <roshandawrani@...> wrote:
>>>>
>>>> Ok, now I get the question fully.
>>>>
>>>> You want to define new properties on the enum but not from inside the
>>>> enum as I first suggested but at runtime.
>>>>
>>>> I don't know at the moment a good, clean answer to that question -
>>>> because enums are a little special and their instance are created internally
>>>> when enum class is loaded, so the changes that you make to enum's metaclass
>>>> after the enum is fully loaded (and its instances are constructed) are not
>>>> seen by the enum instances like TOKIO.
>>>>
>>>> Till any better answer comes along, one possible solution is below.
>>>>
>>>> =================================
>>>> enum Ville
>>>> {
>>>>     TOKIO, PARIS
>>>> }
>>>> // after this already constructed enum instances still pick up enum
>>>> class level metaclass changes made at runtime
>>>> ExpandoMetaClass.enableGlobally()
>>>>
>>>> Ville.metaClass.cool = 10
>>>>
>>>> println Ville.PARIS.cool
>>>>
>>>> // after this enum instance will go back to its own MC again and not see
>>>> enum class level metaclass changes made at runtime
>>>> ExpandoMetaClass.disbleGlobally()
>>>>
>>>> println Ville.PARIS.cool // will fail again
>>>> =================================
>>>>
>>>> -- Roshan
>>>>
>>>> On Sat, Nov 7, 2009 at 2:06 PM, Ista Pouss <istaous@...> wrote:
>>>>>
>>>>> Yes, but how do that on the enum itself ?
>>>>>
>>>>> As an exemple, I do this code in the groovy web console :
>>>>>
>>>>> class A {}
>>>>> enum Ville {TOKIO, PARIS}
>>>>>
>>>>> assert Ville.TOKIO instanceof Ville // of course.
>>>>>
>>>>> A.metaClass.cool = 10
>>>>>
>>>>> a = new A()
>>>>> assert a.cool == 10
>>>>>
>>>>> Ville.metaClass.cool = 10
>>>>> assert Ville.TOKIO.cool == 10 // BING !
>>>>>
>>>>> I do exactly the same metaClass thing, on a "regular" class (A), and
>>>>> on an enum (Ville). It's OK for instance class A (a), not for instance
>>>>> Ville (TOKIO).
>>>>>
>>>>> The error in groovy web console (http://groovyconsole.appspot.com/) is
>>>>> :
>>>>>
>>>>> groovy.lang.MissingPropertyException: No such property: cool for class:
>>>>> Ville
>>>>> at Script1.run(Script1.groovy:12)
>>>>>
>>>>>
>>>>>
>>>>> 2009/11/7 Roshan Dawrani <roshandawrani@...>:
>>>>> > In the first mail you wanted to access Ville.PARIS.parcouru, which
>>>>> > meant
>>>>> > that you wanted to have "parcouru" on the enum constant, which was of
>>>>> > type
>>>>> > Ville.
>>>>> >
>>>>> > If you want to add property "parcouru" at runtime on some regular
>>>>> > class,
>>>>> > then you can very easily do it. Code below shows 2 ways of doing it:
>>>>> > =======================================================
>>>>> > enum Ville {TOKIO, PARIS}
>>>>> > class AlgoTown
>>>>> > {
>>>>> >     Ville ville;
>>>>> > }
>>>>> >
>>>>> > String runtimePropName = "précédent"
>>>>> > addPropertyWhoseNameIsKnown()
>>>>> > addPropertyWhoseNameIsNotKnownUntilRuntime(runtimePropName)
>>>>> >
>>>>> > def addPropertyWhoseNameIsKnown() {
>>>>> >     AlgoTown.metaClass.parcouru = Integer.MAX_VALUE
>>>>> > }
>>>>> > def addPropertyWhoseNameIsNotKnownUntilRuntime(propName) {
>>>>> >     AlgoTown.metaClass."$propName" = 0
>>>>> > }
>>>>> >
>>>>> > def algo1 = new AlgoTown()
>>>>> > def algo2 = new AlgoTown()
>>>>> >
>>>>> > println algo1.parcouru // Integer.MAX_VALUE
>>>>> > println algo1."$runtimePropName" // 0
>>>>> >
>>>>> > algo1.parcouru = 10 // change algo1's parcouru property to 10
>>>>> >
>>>>> > println algo1.parcouru // 10
>>>>> > println algo2.parcouru // algo2 still has parcouru =
>>>>> > Integer.MAX_VALUE
>>>>> > =======================================================
>>>>> >
>>>>> > Hope it helps.
>>>>> > Roshan
>>>>> >
>>>>> >
>>>>> > On Sat, Nov 7, 2009 at 12:49 PM, Ista Pouss <istaous@...>
>>>>> > wrote:
>>>>> >>
>>>>> >> 2009/11/7 Roshan Dawrani <roshandawrani@...>:
>>>>> >> > Hi,
>>>>> >> > If you don't have to add enum properties parcouru and précédent
>>>>> >> > dynamically
>>>>> >> > at run-time, then you can do it like:
>>>>> >> >
>>>>> >> > ==========================================================
>>>>> >> > enum Ville
>>>>> >> > {
>>>>> >> >     TOKIO, MAKAO, OSLO, PARIS, NANTES, ZARASTAPARTAKIS
>>>>> >> >     Ville() {
>>>>> >> >         parcouru = Integer.MAX_VALUE
>>>>> >> >         précédent = 0
>>>>> >> >     }
>>>>> >> >     Integer parcouru
>>>>> >> >     Integer précédent
>>>>> >> > }
>>>>> >>
>>>>> >> Yes, but these properties are not "ville" (ie: town)'s properties.
>>>>> >> It's properties for the algorithm.
>>>>> >>
>>>>> >> In java, I make something like that in this case :
>>>>> >>
>>>>> >> class AlgoTown // stupid name, sorry
>>>>> >> {
>>>>> >> Ville ville; // it's the town
>>>>> >> int parcouru;
>>>>> >> int précédent;
>>>>> >> }
>>>>> >>
>>>>> >> With groovy (i'm a neewbee), my impression is : it is possible to
>>>>> >> add
>>>>> >> properties at runtime with metaclass (and expando ? ) thing. Is it a
>>>>> >> rigth / good impression ? Is it The Good Thing To Do in my case ? Is
>>>>> >> it possible with "enum" ? How ?
>>>>> >>
>>>>> >>
>>>>> >> ---------------------------------------------------------------------
>>>>> >> To unsubscribe from this list, please visit:
>>>>> >>
>>>>> >> http://xircles.codehaus.org/manage_email
>>>>> >>
>>>>> >>
>>>>> >
>>>>> >
>>>>>
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe from this list, please visit:
>>>>>
>>>>> http://xircles.codehaus.org/manage_email
>>>>>
>>>>>
>>>>
>>>
>>
>
>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T21:26:26Z

tog wrote:
> Ok I guess you are missing the service part with the port in your wsdl.
> It should be something like this:
>
> <wsdl:service name="TerraService">
> <wsdl:port name="TerraServiceSoap" binding="tns:TerraServiceSoap">
> <soap:address location="http://terraservice.net/TerraService.asmx"/>
> </wsdl:port>
> </wsdl:service>

Oh, sorry, I tried so many different things that I last tried the
previous file. Here's the initial file:

http://www.linuxha.com/common/services.wsdl

Normally it would be called with:

http://isy.uucp/services.wsdl

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T21:06:40Z

Ok I guess you are missing the service part with the port in your wsdl. It should be something like this:

<wsdl:service name="TerraService">
    <wsdl:port name="TerraServiceSoap" binding="tns:TerraServiceSoap">
        <soap:address location="http://terraservice.net/TerraService.asmx"/>
    </wsdl:port>
</wsdl:service>

On Mon, Nov 9, 2009 at 9:35 AM, Neil Cherry <ncherry@...> wrote:

tog wrote:
> Hi
>
> Yes it might help to see the wsdl. Is a wget on the url pointing to your
> wsdl working ?

Yes, it works for the most part (found an extra " in the middle of
the file with no purpose, have to file a bug on that with the
vendor). Here's the file on my site after I did a wget:

http://www.linuxha.com/common/udiws10.ws

It's the untouched file.

--

Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

--
PGP KeyID: 1024D/69B00854 subkeys.pgp.net

http://cheztog.blogspot.com

Re: Problems with GroovyWS

2009-11-08T20:05:42Z

tog wrote:
> Hi
>
> Yes it might help to see the wsdl. Is a wget on the url pointing to your
> wsdl working ?

Yes, it works for the most part (found an extra " in the middle of
the file with no purpose, have to file a bug on that with the
vendor). Here's the file on my site after I did a wget:

http://www.linuxha.com/common/udiws10.ws

It's the untouched file.

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Handling RESTClient post() responses with no content-type or body.

2009-11-08T18:56:03Z

MMmm, you're right; in most cases, the entity is returned, not just
the Location header. Unfortunately, REST is a somewhat loosely
defined spec and I'm sure there are going to be plenty of examples of
divergances from what "should" be done. The RESTClient should be able
to handle a POST without a response entity, so I've updated the latest
SNAPSHOT to handle it.

-Tom

On Sun, Nov 8, 2009 at 3:15 PM, Jeff Gortatowsky <indanapt@...> wrote:

> Not to be picky, nor I am saying the link I am providing is authoritative, but 201 SHOULD return a response which is the locations of the entity created. I realize that is being 'opinionated' and 'technical', as opposed to "Well here's the real world..." But that has always been my understanding. CREATED means there is a new resource and a URI for it. Those representations, even if only one, should be in the entity returned with the most specific in the Location header
>
> http://restpatterns.org/HTTP_Status_Codes/201_-_Created
> ---------------------------------------
> Jeff Gortatowsky, Fullerton, CA | Twitter: JeffGortatowsky
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email
>
>
>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T18:52:25Z

Hi

Yes it might help to see the wsdl. Is a wget on the url pointing to your wsdl working ?

Regards
Guillaume

On Mon, Nov 9, 2009 at 8:10 AM, Neil Cherry <ncherry@...> wrote:

casmall wrote:
> I might have jumped the gun on your issue anyway Neil. This url should point
> to the wsdl of the web service you are calling into.
> "http://isy.uucp/WEB/udiws10.ws" I'm not sure this is the case - at least
> from the public internet, I couldn't resolve the url. And wsdl url's
> commonly end in ?WSDL.

Sorry the .uucp domain is no longer in use outside my home (I hope).
the is.uucp resolves from inside my home to 192.168.24.21. If you'd
like to see the wsdl (udiws10.ws) file I can post a link to it on
my web site. The device that serves up that file is a ISY99i
( http://www.universal-devices.com/99i.htm ). It's a web server
interface to the Insteon devices on the power line (home automation).

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

--
PGP KeyID: 1024D/69B00854 subkeys.pgp.net

http://cheztog.blogspot.com

Re: Problems with GroovyWS

2009-11-08T18:40:56Z

casmall wrote:
> I might have jumped the gun on your issue anyway Neil. This url should point
> to the wsdl of the web service you are calling into.
> "http://isy.uucp/WEB/udiws10.ws" I'm not sure this is the case - at least
> from the public internet, I couldn't resolve the url. And wsdl url's
> commonly end in ?WSDL.

Sorry the .uucp domain is no longer in use outside my home (I hope).
the is.uucp resolves from inside my home to 192.168.24.21. If you'd
like to see the wsdl (udiws10.ws) file I can post a link to it on
my web site. The device that serves up that file is a ISY99i
( http://www.universal-devices.com/99i.htm ). It's a web server
interface to the Insteon devices on the power line (home automation).

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: GroovyWS proxy.create() - class package determination?

2009-11-08T18:36:33Z

casmall wrote:

> found an answer to this one.
>
> Just make sure you have info log turned on for org.apache.cxf ...
>
> debug 'grails.app',
> 'groovyx.net.ws',
> 'org.apache.cxf'
>
> and out will pop a nice log message giving the available classes with
> package (as part of proxy.initialize()):
>
> dynamic.DynamicClientFactory Created classes:
> com.company.caqh.services.model.beans.realtime.CAQHRealtimeInput, etc...
>
> chad.

Chad, how do I turn on debug?

Thanks

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T16:07:57Z

I might have jumped the gun on your issue anyway Neil. This url should point to the wsdl of the web service you are calling into. "http://isy.uucp/WEB/udiws10.ws" I'm not sure this is the case - at least from the public internet, I couldn't resolve the url. And wsdl url's commonly end in ?WSDL.

chad.

Neil Cherry-3 wrote:

casmall wrote:
> Hey Neil,
>
> There seem to be issues with dynamic downloading of the wsdl from a url
> (http://jira.codehaus.org/browse/GMOD-96) that I ran into a couple weeks ago
> too. Every soap http header response I've worked with has more than 3
> lines. You could patch the source as the defect describes - works, I tried
> it. I've opted to download and use a local .wsdl file instead. I had to do
> something like this (I'm in grails app):
>
> def canonicalPath = (new File(".")).getCanonicalPath()
> def wsdlFilename = "wsdlFile.wsdl"
>
> def wsdlURL
> if (ApplicationHolder?.application?.isWarDeployed()) {
> wsdlURL =
> "file:///${canonicalPath}/../webapps/ROOT/WEB-INF/wsdls/${wsdlFilename}"
> } else {
> wsdlURL =
> "file:///${canonicalPath}/grails-app/conf/wsdls/${wsdlFilename}"
> }
>
> def proxy = new WSClient(wsdlURL, this.class.classLoader)
> proxy.initialize()
>
> Now I'm struggling with my proxy.create() :( Have other thread on this
> list.

I didn't understand a word you said. :-) I'm still relearning Java etal.
I'll play with the code and see what it does (I think I understand that
part). I've had no luck getting GroovyWS to compile as I have no idea
how Grails works or what the error message I'm getting means. Since I'm
not trying to learn Gradle yet I've decided to leave that one alone..

Thanks for the pointer and I am following your GroovyWS discussion.

--
Linux Home Automation Neil Cherry ncherry@linuxha.com
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: GroovyWS proxy.create() - class package determination?

2009-11-08T16:00:42Z

found an answer to this one.

Just make sure you have info log turned on for org.apache.cxf ...

debug 'grails.app',
'groovyx.net.ws',
'org.apache.cxf'

and out will pop a nice log message giving the available classes with package (as part of proxy.initialize()):

dynamic.DynamicClientFactory Created classes: com.company.caqh.services.model.beans.realtime.CAQHRealtimeInput, etc...

chad.

casmall wrote:

Howdie any GroovyWS gurus,

I'm wondering how to find out what the fully qualified class name is for a soap method using groovy-ws. I've tried about every combination I can come up with, but no luck. In my case ...

def proxy = new WSClient(wsdlURL, this.class.classLoader)
proxy.initialize()
def caqhRealtimeInput = proxy.create("com.realtime.beans.model.services.caqh.company.CAQHRealtimeInput") // ???? How do I figure out what to use here?

stack:
java.lang.ClassNotFoundException: com.realtime.beans.model.services.caqh.company.CAQHRealtimeInput

WSDL snip:

<wsdl:types>
<schema targetNamespace="http://realtime.beans.model.services.caqh.company.com" xmlns="http://www.w3.org/2001/XMLSchema" xmlns:wsdl="http://schemas.xmlsoap.org/wsdl/" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<complexType name="CAQHRealtimeInput">
<sequence>
<element name="hipaa270" nillable="true" type="xsd:string"/>
<element name="date" nillable="true" type="xsd:string"/>
</sequence>
</complexType>

thanks,
chad.

Re: Is this a bug or not?

2009-11-08T15:02:17Z

Tim Yates schrieb:
> is calling .intern() a valid way of achieving this as well?

inter() on GString would return toString().intern(), so I assume yes.

In general I advise not to use intern if there is no real need to do that.

bye blackdrag

--
Jochen "blackdrag" Theodorou
The Groovy Project Tech Lead (http://groovy.codehaus.org)
http://blackdragsview.blogspot.com/

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Is this a bug or not?

2009-11-08T14:45:58Z

is calling .intern() a valid way of achieving this as well?

Tim

On Sun, Nov 8, 2009 at 6:10 PM, Jochen Theodorou <blackdrag@...> wrote:

...If you want to use the GString just to more easily build a String, then you have to convert the GString into a String first. you can do that by assigning the value to a String, by casting with a String, by using "as String" or by simply calling toString() on it.

Licensing of the sample code and sample document for the online course

2009-11-08T14:41:32Z

I am not sure if this is the right alias to ask this question.
Hopefully someone
knows the answer.

My name is Sang Shin. I teach several free online courses through
javapassion.com below.

http://www.javapassion.com/

I am planning to start "Groovy and Grails (with Passion!)" online course
from Jan. 10th, 2010.

Now I am wondering if I can borrow the code and document from the
Groovy document site below (without violating any copyright).

http://groovy.codehaus.org/Documentation

Thanks in advance.

-Sang

--
---------------------------------------------------------------
Sang Shin, sangshinpassion@..., sang.shin@...
http://www.javapassion.com/SangShinSchedule.html
(Life is worth living... with Passion!)
---------------------------------------------------------------
"Ajax and Web 2.0 Programming" course starts from Sep. 7th, 2009
http://www.javapassion.com/ajaxcodecamp/
"Java EE Programming Advanced" course starts from Oct. 1st, 2009
http://www.javapassion.com/javaee_advanced/
---------------------------------------------------------------

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: How to get Map from String?

2009-11-08T13:24:07Z

In a news site, one nodes can refer to many nodes, can embeded many nodes, can have out-link with many nodes, I have a class named as NodeRelation with
class NodeRelation {
Node fromNode
NodeRevision fromRevision
Node toNode
Node toRevision
Integer weight
String relationType
}

In Node class, I want to store two string,
String nodeReferring
String nodeReferred

So, I can get the relationship from one query.

I just want to serialize the two string with a easy way.

I am trying groovy.util.Eval class to achieve it.

Thanks.

Martin C. Martin wrote:

There probably isn't a way, e.g. both these print exactly the same thing:
a = [b:'[c:d]']
println a.toMapString()
a = [b:[c:'d'] ]
println a.toMapString()

But if you use inspect() instead of toMapString(), you get something
parsable. However, not many objects implement inspect(), so if you're
using classes you haven't written yourself, you'll either have to add
them using something like the Expando Meta Class, or abandon inspect()
altogether.

The output of inspect() should be valid groovy source, so you could
eval() it.

What's the use case for this? Why not use Java's serialization to do it?

Best,
Martin

Hubert Chang wrote:
> def a= [:]
> a = [b:[c:'d:=e', f:'g'], h:'i']
>
> a.toMapString() ==> [b:[c:d:=e, f:g], h:i]
>
> How can I get the original linkedHashMap from a given '[b:[c:d:=e, f:g],
> h:i]' string?
>
> Thanks a lot.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Groovy Eclipse Editor Copy Paste Down

2009-11-08T13:23:17Z

Hi,

This is now working in the latest snapshot. Please update your
groovy-eclipse version.

On Sat, Nov 7, 2009 at 9:31 PM, akiliner <drypot@...> wrote:

>
> I'm using eclipse 3.5
>
> Copy & Paste in ${}, eclipse crash.
>
> for example, If I paste something in XXX position, eclipse does not
> response.
>
> "session.logged: ${session.XXX}"
>
> Turning off eclipse option, Java > Editor > Typing > Adjust Indentation,
> remove this problem.
> --
> View this message in context: http://old.nabble.com/Groovy-Eclipse-Editor-Copy-Paste-Down-tp26251171p26251171.html
> Sent from the groovy - user mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email
>
>
>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T13:22:14Z

casmall wrote:

> Hey Neil,
>
> There seem to be issues with dynamic downloading of the wsdl from a url
> (http://jira.codehaus.org/browse/GMOD-96) that I ran into a couple weeks ago
> too. Every soap http header response I've worked with has more than 3
> lines. You could patch the source as the defect describes - works, I tried
> it. I've opted to download and use a local .wsdl file instead. I had to do
> something like this (I'm in grails app):
>
> def canonicalPath = (new File(".")).getCanonicalPath()
> def wsdlFilename = "wsdlFile.wsdl"
>
> def wsdlURL
> if (ApplicationHolder?.application?.isWarDeployed()) {
> wsdlURL =
> "file:///${canonicalPath}/../webapps/ROOT/WEB-INF/wsdls/${wsdlFilename}"
> } else {
> wsdlURL =
> "file:///${canonicalPath}/grails-app/conf/wsdls/${wsdlFilename}"
> }
>
> def proxy = new WSClient(wsdlURL, this.class.classLoader)
> proxy.initialize()
>
> Now I'm struggling with my proxy.create() :( Have other thread on this
> list.

I didn't understand a word you said. :-) I'm still relearning Java etal.
I'll play with the code and see what it does (I think I understand that
part). I've had no luck getting GroovyWS to compile as I have no idea
how Grails works or what the error message I'm getting means. Since I'm
not trying to learn Gradle yet I've decided to leave that one alone..

Thanks for the pointer and I am following your GroovyWS discussion.

--
Linux Home Automation Neil Cherry ncherry@...
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: How to get Map from String?

2009-11-08T13:03:00Z

probably not the best solution:

h= new GroovyShell().evaluate("[b:[c:'d:=e', f:'g'], h:'i']")

Re: How to get Map from String?

2009-11-08T12:39:38Z

There probably isn't a way, e.g. both these print exactly the same thing:
a = [b:'[c:d]']
println a.toMapString()
a = [b:[c:'d'] ]
println a.toMapString()

But if you use inspect() instead of toMapString(), you get something
parsable. However, not many objects implement inspect(), so if you're
using classes you haven't written yourself, you'll either have to add
them using something like the Expando Meta Class, or abandon inspect()
altogether.

The output of inspect() should be valid groovy source, so you could
eval() it.

What's the use case for this? Why not use Java's serialization to do it?

Best,
Martin

Hubert Chang wrote:
> def a= [:]
> a = [b:[c:'d:=e', f:'g'], h:'i']
>
> a.toMapString() ==> [b:[c:d:=e, f:g], h:i]
>
> How can I get the original linkedHashMap from a given '[b:[c:d:=e, f:g],
> h:i]' string?
>
> Thanks a lot.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

How to get Map from String?

2009-11-08T12:23:06Z

def a= [:]
a = [b:[c:'d:=e', f:'g'], h:'i']

a.toMapString() ==> [b:[c:d:=e, f:g], h:i]

How can I get the original linkedHashMap from a given '[b:[c:d:=e, f:g], h:i]' string?

Thanks a lot.

Re: Handling RESTClient post() responses with no content-type or body.

2009-11-08T12:15:20Z

Not to be picky, nor I am saying the link I am providing is authoritative, but 201 SHOULD return a response which is the locations of the entity created. I realize that is being 'opinionated' and 'technical', as opposed to "Well here's the real world..." But that has always been my understanding. CREATED means there is a new resource and a URI for it. Those representations, even if only one, should be in the entity returned with the most specific in the Location header

http://restpatterns.org/HTTP_Status_Codes/201_-_Created
---------------------------------------
Jeff Gortatowsky, Fullerton, CA | Twitter: JeffGortatowsky

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email

Re: Problems with GroovyWS

2009-11-08T11:28:44Z

Hey Neil,

There seem to be issues with dynamic downloading of the wsdl from a url (http://jira.codehaus.org/browse/GMOD-96) that I ran into a couple weeks ago too. Every soap http header response I've worked with has more than 3 lines. You could patch the source as the defect describes - works, I tried it. I've opted to download and use a local .wsdl file instead. I had to do something like this (I'm in grails app):

def canonicalPath = (new File(".")).getCanonicalPath()
def wsdlFilename = "wsdlFile.wsdl"

def wsdlURL
if (ApplicationHolder?.application?.isWarDeployed()) {
wsdlURL = "file:///${canonicalPath}/../webapps/ROOT/WEB-INF/wsdls/${wsdlFilename}"
} else {
wsdlURL = "file:///${canonicalPath}/grails-app/conf/wsdls/${wsdlFilename}"
}

def proxy = new WSClient(wsdlURL, this.class.classLoader)
proxy.initialize()

Now I'm struggling with my proxy.create() :( Have other thread on this list.

Good Luck
chad.

Neil Cherry-3 wrote:

I'm attempting to use GroovyWS (0.5.0) with the following code:

import groovyx.net.ws.WSClient

@Grab(group='org.codehaus.groovy.modules', module='groovyws', version='0.5.0')

getProxy(wsdl, classLoader) {
new WSClient(wsdl, classLoader)
}

url = "http://isy.uucp/WEB/udiws10.ws"

proxy = new WSClient(url, this.class.classLoader)

proxy.initialize()

ERROR org.apache.cxf.service.factory.ServiceConstructionException:
NO_SERVICE_EXC
at org.apache.cxf.wsdl11.WSDLServiceFactory.create (WSDLServiceFactory.java:102)
at org.apache.cxf.endpoint.ClientImpl.<init> (ClientImpl.java:145)
at org.apache.cxf.endpoint.dynamic.DynamicClientFactory.createClient
(DynamicClientFactory.java:253)
at org.apache.cxf.endpoint.dynamic.DynamicClientFactory.createClient
(DynamicClientFactory.java:196)
at org.apache.cxf.endpoint.dynamic.DynamicClientFactory.createClient
(DynamicClientFactory.java:175)
at groovyx.net.ws.AbstractCXFWSClient.createClient (AbstractCXFWSClient.java:194)
at groovyx.net.ws.WSClient.initialize (WSClient.java:107)
at groovyx.net.ws.IWSClient$initialize.call (Unknown Source)
at groovysh_evaluate.run (groovysh_evaluate:3)
...

I've tried the simplier services from the GrovvyWS page and they work
fine. But what I have noticed with the above is that the rather large
wsdl file gets truncated while sending over the network. I'm unable to
diagnose the problem any further because I'm a bit new to Java, WSDL
and Groovy.

One last data point I was able to test the wsdl with SoupUI and I was able
to validate the wsdl and test the service.

Any pointers what to do next?

--
Linux Home Automation Neil Cherry ncherry@linuxha.com
http://www.linuxha.com/ Main site
http://linuxha.blogspot.com/ My HA Blog
Author of: Linux Smart Homes For Dummies

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

http://xircles.codehaus.org/manage_email