xmlvm-users — For current and interested XMLVM users, for public discussions and anouncements.

Re: [xmlvm-users] problem with arrays

[Panayotis K. <pan...@pa...>]

I think I found another problem with arrays.

When using 
XMLVMArray createSingleDimensionWithType:size:andData
in a multi-dimension array, it clearly states that each entry of the array is of type XMLVMElem, (which is of size 8), while the type (and thus the type size) could be smaller.
I think the implementation of this method should change, so that it could "trim" the data when this method is called automatically and to convert it to the correct size for each element.

I have a question here: is this method automatically inserted in the source code *only* for the multi-dimensional arrays, in order to define the higher orders of array length? 

Re: [xmlvm-users] problem with arrays

[Arno P. <ar...@pu...>]

yep

On 12/17/10 11:29 PM, Panayotis Katsaloulis wrote:
>
> On Dec 18, 2010, at 6:34 AM, Arno Puder wrote:
>
>>
>> looks good, however, you made an interesting point by saying that there
>> are no const arrays in Java. I think the current implementation of
>> fill-array is not correct: the data needs to be copied there as well in
>> case the application modifies the initial values.
>>
>> I think the data needs to be copied in any case and the flag ownsData
>> should always be TRUE (meaning, it can be removed).
>>
>> Arno
>
> ... which also means, that XMLVMArray->ownsData should be removed, since there is no method to set it to NO ?
> ------------------------------------------------------------------------------
> Lotusphere 2011
> Register now for Lotusphere 2011 and learn how
> to connect the dots, take your collaborative environment
> to the next level, and enter the era of Social Business.
> http://p.sf.net/sfu/lotusphere-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users

Re: [xmlvm-users] problem with arrays

[Panayotis K. <pan...@pa...>]

On Dec 18, 2010, at 6:34 AM, Arno Puder wrote:

> 
> looks good, however, you made an interesting point by saying that there 
> are no const arrays in Java. I think the current implementation of 
> fill-array is not correct: the data needs to be copied there as well in 
> case the application modifies the initial values.
> 
> I think the data needs to be copied in any case and the flag ownsData 
> should always be TRUE (meaning, it can be removed).
> 
> Arno

... which also means, that XMLVMArray->ownsData should be removed, since there is no method to set it to NO ? 

Re: [xmlvm-users] problem with arrays

[Arno P. <ar...@pu...>]

looks good, however, you made an interesting point by saying that there 
are no const arrays in Java. I think the current implementation of 
fill-array is not correct: the data needs to be copied there as well in 
case the application modifies the initial values.

I think the data needs to be copied in any case and the flag ownsData 
should always be TRUE (meaning, it can be removed).

Arno


On 12/17/10 2:10 AM, Panayotis Katsaloulis wrote:
> I have created a new patch about it, please have a look here:
>
> http://xmlvm-reviews.appspot.com/102001
>
> ------------------------------------------------------------------------------
> Lotusphere 2011
> Register now for Lotusphere 2011 and learn how
> to connect the dots, take your collaborative environment
> to the next level, and enter the era of Social Business.
> http://p.sf.net/sfu/lotusphere-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users

Re: [xmlvm-users] problem with arrays

[Panayotis K. <pan...@pa...>]

I have created a new patch about it, please have a look here:

http://xmlvm-reviews.appspot.com/102001

Re: [xmlvm-users] problem with arrays

[Panayotis K. <pan...@pa...>]

On Dec 17, 2010, at 2:19 AM, Arno Puder wrote:

> 
> I agree in principle with your proposed first solution. However, there 
> are also instances where it is not necessary to copy the array (e.g., 
> the dex:fill-array instruction). I would suggest to add another boolean 
> parameter to that method that states if 'data' should be copied or not 
> and then go through the code and decide individually where the data 
> needs to be copied or not. If we can avoid unnecessary copying, we should.
> 
> Arno

You mean this one?
dex:filled-new-array|dex:filled-new-array-range
in line 3500 of xmlvm2objc.xsl
 
I am not at all experienced with dex, and I have a question!
From my understanding this is the only location where createSingleDimensionWithType...andData is called from the ObjC backend.
And there is only one more, called by NSData.
So, should I replace this generic call of the function with a new one, passing "copyData:NO" for the dex instruction and "copyData:Yes" for NSData ?

I am trying to understand the memory management of the produced dex instruction. Why is no copying of data required at that point?

Re: [xmlvm-users] problem with arrays

[Arno P. <ar...@pu...>]

I agree in principle with your proposed first solution. However, there 
are also instances where it is not necessary to copy the array (e.g., 
the dex:fill-array instruction). I would suggest to add another boolean 
parameter to that method that states if 'data' should be copied or not 
and then go through the code and decide individually where the data 
needs to be copied or not. If we can avoid unnecessary copying, we should.

Arno


On 12/16/10 10:37 AM, Panayotis Katsaloulis wrote:
> Hello!
>
> I think I found a problem with the current implementation of XMLVM and java arrays.
> I am talking about this:
>
> + (XMLVMArray*) createSingleDimensionWithType:(int) type size:(int) size andData:(void*) data;
>
> In the currently implementation, data is not retained or copied anywhere. Thus, if data was automatically released at some time, the array has serious problems (and it took me a whole day to find it).
> The problem appears for example when NSData.bytes(); is used, where data is released back to the system.
>
> I can think of two possible solutions to this problem:
> 1) create a memory copy of the original data (i.e. do something similar to what clone__ does)
> 2) add a new member variable to XMLVMArray, which will keep hold of the creator object, and retain it (and release it when dealloc of array is found)
>
> Since this method is only used with NSData and AFAICS with initializing of multi-dimension arrays with "new" (only for their first dimension), and since NSData clearly states that this is a const structure, while with Java you can more or less do whatever you want with it, I'd suggest to fix this issue with the first solution.
> When we agree, I could submit a patch.
>
> What do you think?
>
>
> ------------------------------------------------------------------------------
> Lotusphere 2011
> Register now for Lotusphere 2011 and learn how
> to connect the dots, take your collaborative environment
> to the next level, and enter the era of Social Business.
> http://p.sf.net/sfu/lotusphere-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users

[xmlvm-users] problem with arrays

[Panayotis K. <pan...@pa...>]

Hello!

I think I found a problem with the current implementation of XMLVM and java arrays.
I am talking about this:

+ (XMLVMArray*) createSingleDimensionWithType:(int) type size:(int) size andData:(void*) data;

In the currently implementation, data is not retained or copied anywhere. Thus, if data was automatically released at some time, the array has serious problems (and it took me a whole day to find it). 
The problem appears for example when NSData.bytes(); is used, where data is released back to the system.

I can think of two possible solutions to this problem:
1) create a memory copy of the original data (i.e. do something similar to what clone__ does)
2) add a new member variable to XMLVMArray, which will keep hold of the creator object, and retain it (and release it when dealloc of array is found)

Since this method is only used with NSData and AFAICS with initializing of multi-dimension arrays with "new" (only for their first dimension), and since NSData clearly states that this is a const structure, while with Java you can more or less do whatever you want with it, I'd suggest to fix this issue with the first solution.
When we agree, I could submit a patch.

What do you think?

[xmlvm-users] Homepage Update: Improved Documentation

[Sascha H. <sa...@gm...>]

Hi everyone,

I want to inform you that we just updated the XMLVM homepage in order to
improve the documentation.

It is now easier to find the documentation you are looking for. A click on
"documentation" in the main menu leads you directly to our up-to-date usage
documentation. A new main menu entry called "contribute" holds all the
documentation you need to get started developing on the core components of
XMLVM itself.
Both of these pages are powered by Google Sites and a group of editors is
updating it all the time so you get the most up-to-date information.

If you see an error in the documentation or you think something is missing,
let us know on this mailing list. We'll either change it right away or give
you access to the documentation itself.

Have a good weekend!
// Sascha

http://xmlvm.org

Re: [xmlvm-users] XMLVM and JUnit

[Sascha H. <sa...@xm...>]

Hi guys,

very good point Ilya. We are actually current working on a strategy to add
tests to XMLVM. It has been on my list for quite a while, but I didn't get
to it.

Cross-compiling tests does make sense, for the reasons you have mentioned.
Not only will it test whether the cross-compilation is working correctly,
but also whether hans-written compatibility classes are doing what they are
supposed to do.

Right now we have a few demos and we ask committers to make sure that these
still work. But it's tedious. Being able to run one command that does all
the testing automatically and returns either "Go" or "No go" would be much
better.

// Sascha


On Wed, Dec 8, 2010 at 3:15 PM, Troy Gaines <tro...@gm...> wrote:

> I guess one issue would be if the application didn't compile correctly,
> then it's likely your tests might not either such that the tests would be
> invalidated anyway.
>
> But if you wanted to do this, I'm sure it's possible.  One of the issues
> after compilation is how you would actually execute the tests.
>
> Best of luck.
>
>
> On Wed, Dec 8, 2010 at 8:12 AM, Ilya Shinkarenko <sh...@gm...>wrote:
>
>> this is clear, but wouldn't it be nice to see if there were problems
>> in cross-compilation? just in case, especially for mission-critical
>> parts of code
>>
>> On 8 December 2010 15:06, Troy Gaines <tro...@gm...> wrote:
>> > Curiously, why are you wanting to cross compile JUnit tests?  IMHO, I
>> would
>> > think that if they did fail on the Objective C side, then this would
>> > actually be just a bug in the cross compilation and not necessarily an
>> > application problem....
>> >
>> >
>> > On Wed, Dec 8, 2010 at 5:47 AM, Miggi <mig...@go...> wrote:
>> >>
>> >> Hello
>> >>
>> >> Is someone out there who's already tried to cross-compile JUnit-tests
>> in
>> >> Objective-C? I'm working on a project where I have to cross-compile an
>> >> Android-project to an iPhone-project and it would be very nice if I
>> >> where able to use the already existing JUnit-tests.
>> >>
>> >> Or can someone give me a how-to on adding new stuff to xmlvm. I tried
>> it
>> >> but it's to complicated for myself without any help. I don't know which
>> >> parts of JUnit I have to cross-compile, to use, to rewrite or what else
>> ;)
>> >>
>> >> Greets, Miggi
>> >>
>> >>
>> >>
>> ------------------------------------------------------------------------------
>> >> What happens now with your Lotus Notes apps - do you make another
>> costly
>> >> upgrade, or settle for being marooned without product support? Time to
>> >> move
>> >> off Lotus Notes and onto the cloud with Force.com, apps are easier to
>> >> build,
>> >> use, and manage than apps on traditional platforms. Sign up for the
>> Lotus
>> >> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
>> >> _______________________________________________
>> >> xmlvm-users mailing list
>> >> xml...@li...
>> >> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>> >
>> >
>> >
>> ------------------------------------------------------------------------------
>> > What happens now with your Lotus Notes apps - do you make another costly
>> > upgrade, or settle for being marooned without product support? Time to
>> move
>> > off Lotus Notes and onto the cloud with Force.com, apps are easier to
>> build,
>> > use, and manage than apps on traditional platforms. Sign up for the
>> Lotus
>> > Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
>> > _______________________________________________
>> > xmlvm-users mailing list
>> > xml...@li...
>> > https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>> >
>> >
>>
>
>
>
> ------------------------------------------------------------------------------
> What happens now with your Lotus Notes apps - do you make another costly
> upgrade, or settle for being marooned without product support? Time to move
> off Lotus Notes and onto the cloud with Force.com, apps are easier to
> build,
> use, and manage than apps on traditional platforms. Sign up for the Lotus
> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>
>

Re: [xmlvm-users] XMLVM and JUnit

[Troy G. <tro...@gm...>]

I guess one issue would be if the application didn't compile correctly, then
it's likely your tests might not either such that the tests would be
invalidated anyway.

But if you wanted to do this, I'm sure it's possible.  One of the issues
after compilation is how you would actually execute the tests.

Best of luck.

On Wed, Dec 8, 2010 at 8:12 AM, Ilya Shinkarenko <sh...@gm...> wrote:

> this is clear, but wouldn't it be nice to see if there were problems
> in cross-compilation? just in case, especially for mission-critical
> parts of code
>
> On 8 December 2010 15:06, Troy Gaines <tro...@gm...> wrote:
> > Curiously, why are you wanting to cross compile JUnit tests?  IMHO, I
> would
> > think that if they did fail on the Objective C side, then this would
> > actually be just a bug in the cross compilation and not necessarily an
> > application problem....
> >
> >
> > On Wed, Dec 8, 2010 at 5:47 AM, Miggi <mig...@go...> wrote:
> >>
> >> Hello
> >>
> >> Is someone out there who's already tried to cross-compile JUnit-tests in
> >> Objective-C? I'm working on a project where I have to cross-compile an
> >> Android-project to an iPhone-project and it would be very nice if I
> >> where able to use the already existing JUnit-tests.
> >>
> >> Or can someone give me a how-to on adding new stuff to xmlvm. I tried it
> >> but it's to complicated for myself without any help. I don't know which
> >> parts of JUnit I have to cross-compile, to use, to rewrite or what else
> ;)
> >>
> >> Greets, Miggi
> >>
> >>
> >>
> ------------------------------------------------------------------------------
> >> What happens now with your Lotus Notes apps - do you make another costly
> >> upgrade, or settle for being marooned without product support? Time to
> >> move
> >> off Lotus Notes and onto the cloud with Force.com, apps are easier to
> >> build,
> >> use, and manage than apps on traditional platforms. Sign up for the
> Lotus
> >> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
> >> _______________________________________________
> >> xmlvm-users mailing list
> >> xml...@li...
> >> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
> >
> >
> >
> ------------------------------------------------------------------------------
> > What happens now with your Lotus Notes apps - do you make another costly
> > upgrade, or settle for being marooned without product support? Time to
> move
> > off Lotus Notes and onto the cloud with Force.com, apps are easier to
> build,
> > use, and manage than apps on traditional platforms. Sign up for the Lotus
> > Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
> > _______________________________________________
> > xmlvm-users mailing list
> > xml...@li...
> > https://lists.sourceforge.net/lists/listinfo/xmlvm-users
> >
> >
>

Re: [xmlvm-users] XMLVM and JUnit

[Ilya S. <sh...@gm...>]

this is clear, but wouldn't it be nice to see if there were problems
in cross-compilation? just in case, especially for mission-critical
parts of code

On 8 December 2010 15:06, Troy Gaines <tro...@gm...> wrote:
> Curiously, why are you wanting to cross compile JUnit tests?  IMHO, I would
> think that if they did fail on the Objective C side, then this would
> actually be just a bug in the cross compilation and not necessarily an
> application problem....
>
>
> On Wed, Dec 8, 2010 at 5:47 AM, Miggi <mig...@go...> wrote:
>>
>> Hello
>>
>> Is someone out there who's already tried to cross-compile JUnit-tests in
>> Objective-C? I'm working on a project where I have to cross-compile an
>> Android-project to an iPhone-project and it would be very nice if I
>> where able to use the already existing JUnit-tests.
>>
>> Or can someone give me a how-to on adding new stuff to xmlvm. I tried it
>> but it's to complicated for myself without any help. I don't know which
>> parts of JUnit I have to cross-compile, to use, to rewrite or what else ;)
>>
>> Greets, Miggi
>>
>>
>> ------------------------------------------------------------------------------
>> What happens now with your Lotus Notes apps - do you make another costly
>> upgrade, or settle for being marooned without product support? Time to
>> move
>> off Lotus Notes and onto the cloud with Force.com, apps are easier to
>> build,
>> use, and manage than apps on traditional platforms. Sign up for the Lotus
>> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
>> _______________________________________________
>> xmlvm-users mailing list
>> xml...@li...
>> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>
>
> ------------------------------------------------------------------------------
> What happens now with your Lotus Notes apps - do you make another costly
> upgrade, or settle for being marooned without product support? Time to move
> off Lotus Notes and onto the cloud with Force.com, apps are easier to build,
> use, and manage than apps on traditional platforms. Sign up for the Lotus
> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>
>

Re: [xmlvm-users] XMLVM and JUnit

[Troy G. <tro...@gm...>]

Curiously, why are you wanting to cross compile JUnit tests?  IMHO, I would
think that if they did fail on the Objective C side, then this would
actually be just a bug in the cross compilation and not necessarily an
application problem....




On Wed, Dec 8, 2010 at 5:47 AM, Miggi <mig...@go...> wrote:

> Hello
>
> Is someone out there who's already tried to cross-compile JUnit-tests in
> Objective-C? I'm working on a project where I have to cross-compile an
> Android-project to an iPhone-project and it would be very nice if I
> where able to use the already existing JUnit-tests.
>
> Or can someone give me a how-to on adding new stuff to xmlvm. I tried it
> but it's to complicated for myself without any help. I don't know which
> parts of JUnit I have to cross-compile, to use, to rewrite or what else ;)
>
> Greets, Miggi
>
>
> ------------------------------------------------------------------------------
> What happens now with your Lotus Notes apps - do you make another costly
> upgrade, or settle for being marooned without product support? Time to move
> off Lotus Notes and onto the cloud with Force.com, apps are easier to
> build,
> use, and manage than apps on traditional platforms. Sign up for the Lotus
> Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>

[xmlvm-users] XMLVM and JUnit

[Miggi <mig...@go...>]

Hello

Is someone out there who's already tried to cross-compile JUnit-tests in 
Objective-C? I'm working on a project where I have to cross-compile an 
Android-project to an iPhone-project and it would be very nice if I 
where able to use the already existing JUnit-tests.

Or can someone give me a how-to on adding new stuff to xmlvm. I tried it 
but it's to complicated for myself without any help. I don't know which 
parts of JUnit I have to cross-compile, to use, to rewrite or what else ;)

Greets, Miggi

Re: [xmlvm-users] Support for new Android/iPhone API

[Panayotis K. <pan...@pa...>]

> 
> Note that you need the latest iOS 4.2 SDK now otherwise you will get 
> "Missing base SDK" in your Xcode project.
> 

Btw, from now on, there will be no need to update the project file with every iOS update.
In Xcode there is a new generic target, which refers to latest (installed) SDK.

[xmlvm-users] Support for new Android/iPhone API

[Arno P. <ar...@pu...>]

I have just committed another nice patch from Panayotis Katsaloulis. 
Here is the changelog:

1) Support of iOS 4.2 SDK
2) Support of CoreLocation and MapKit (iPhone)
3) Basic support of Location and Maps (Android)
4) Support of ListActivity, BaseAdapter, ArrayAdapter, ListView (Android)
5) Redesigned Toast, Typeface (Android)
6) UINavigationView demo (iPhone)
7) Improved templates and project migration for --skeleton targets
8) Additions to java.util.HashMap java.util.LinkedList java.util.Set
9) Android and iPhone bug fixes

Note that you need the latest iOS 4.2 SDK now otherwise you will get 
"Missing base SDK" in your Xcode project.

Thanks, Panayotis!

Arno

Re: [xmlvm-users] Issue with Java long

[Arno P. <ar...@pu...>]

Java long is now mapped to 'long long' in Objective-C again. There were 
several minor problems that are fixed now. Thanks to Troy and Paul for 
helping out.

Arno


On 12/3/10 5:09 AM, Troy Gaines wrote:
> Without wasting too much of your time, what problems are we seeing?  I
> would be willing to invest some time investigating.  I used the change
> you committed and haven't seen any issues.  Is there a certain situation?
>
> Thanks.
>
> On Thu, Dec 2, 2010 at 11:47 PM, Arno Puder <ar...@pu...
> <mailto:ar...@pu...>> wrote:
>
>     Troy recently pointed out that a Java long should be mapped to a long
>     long in Objective-C. I committed a small patch that seemed to fix
>     things. However, we are now experiencing some "weird" problems because
>     of this patch. I have just introduced a typedef long JAVA_LONG in
>     xmlvm.h so that the definition of Java long can quickly be changed
>     system-wide. For now I changed it back to a long (not a long long)
>     until we have figured out what the problem is.
>
>     Arno
>
>     ------------------------------------------------------------------------------
>     Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
>     Tap into the largest installed PC base & get more eyes on your game by
>     optimizing for Intel(R) Graphics Technology. Get started today with the
>     Intel(R) Software Partner Program. Five $500 cash prizes are up for
>     grabs.
>     http://p.sf.net/sfu/intelisp-dev2dev
>     _______________________________________________
>     xmlvm-users mailing list
>     xml...@li...
>     <mailto:xml...@li...>
>     https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>
>

Re: [xmlvm-users] Issue with Java long

[Troy G. <tro...@gm...>]

Without wasting too much of your time, what problems are we seeing?  I would
be willing to invest some time investigating.  I used the change you
committed and haven't seen any issues.  Is there a certain situation?

Thanks.

On Thu, Dec 2, 2010 at 11:47 PM, Arno Puder <ar...@pu...> wrote:

> Troy recently pointed out that a Java long should be mapped to a long
> long in Objective-C. I committed a small patch that seemed to fix
> things. However, we are now experiencing some "weird" problems because
> of this patch. I have just introduced a typedef long JAVA_LONG in
> xmlvm.h so that the definition of Java long can quickly be changed
> system-wide. For now I changed it back to a long (not a long long)
> until we have figured out what the problem is.
>
> Arno
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>

[xmlvm-users] Issue with Java long

[Arno P. <ar...@pu...>]

Troy recently pointed out that a Java long should be mapped to a long
long in Objective-C. I committed a small patch that seemed to fix
things. However, we are now experiencing some "weird" problems because
of this patch. I have just introduced a typedef long JAVA_LONG in
xmlvm.h so that the definition of Java long can quickly be changed
system-wide. For now I changed it back to a long (not a long long)
until we have figured out what the problem is.

Arno

Re: [xmlvm-users] Implementing new class in Objective-C

[Tomas V. <tom...@gm...>]

Hi,

Arno:
Yes, that's what I thought. So right now I'm trying to do a demo using only
Java and cross compiling it to Obj-C code. I'm also trying to implement
missing classes (now UITextViewDelegate) by first writing the stub in Java,
cc:ing it to Obj-C and filling in the methods.

Personally I found the C backend code much harder to understand but I
haven't done a lot of C or Obj-C coding. How exactly is supposed to be used?

Panayotis:
Thank you, that is a perfect example. I'm currently trying to implement
everything in Java but I think I will try something similar to the Zip
plugin after to evaluate the options.


If I implement classes on my own I would of course like to contribute and
upload them. What worries me is that as I haven't done any Obj-C before, I'm
currently reading about it and how it works so I'm not completely lost. When
I'm implementing a new class I'm also looking at other similar classes that
are implemented. But I can't be 100% sure that it's correct, how should I
proceed then?

Best regards,
Tomas Vestelind

On 30 November 2010 19:37, Panayotis Katsaloulis <pan...@pa...>wrote:

>
> On Nov 30, 2010, at 8:18 PM, Arno Puder wrote:
>
> >
> > the preferred way is to do as much as possible in Java and then
> > cross-compile to Objective-C. If you already have an Objective-C class
> > that you want to expose in Java, you need to create a Java stub and then
> > write some "glue" code in Objective-C that bridges the Java API to the
> > Objective-C API. This is necessary because of the certain differences
> > between the languages. E.g., we have to use name mangling when
> > cross-compiling from Java to Objective-C and the "glue" code needs to do
> > a mapping from the mangled version of a name to the original version.
>
> You can also take advantage of the "plugin" mechanism.
> To see a demo of this convenient way, have a look at the directory
> demo/iphome/plugins/Zip
>
>
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users
>



-- 
Quand on veut un mouton, c'est la preuve qu'on existe

Re: [xmlvm-users] status and (L)GPL

[Hansi R. <su...@su...>]

hey!

> ok, guys, the generated code of gcc, bison, xmlvm do not fall under GPL
> or L-GPL. The clarification for bison is that since there is a bit of
> library portion that needs to be linked against the generated code
> automatically makes the whole application GPL (note that bison is GPL,
> not L-GPL). Since xmlvm is licensed under L-GPL, this problem does not
> occur.
i agree to 99%. the 1% i don't agree on is for instance the file
"Application.js.template"
that is part of xmlvm (thus lgpl):

to make my point clear let me quote the file here:


/* ************************************************************************
#asset(temp_qx_app/*)
************************************************************************ */
/**
 * This is the main application class of your custom application "temp_qx_app"
 */
qx.Class.define("{{XMLVM_TEMP_PROJECT_NAME}}.Application",
{
  extend : qx.application.Standalone,
  members :
  {
    main : function()
    {
      // Call super class
      this.base(arguments);

      // Enable logging in debug variant
      if (qx.core.Variant.isSet("qx.debug", "on"))
      {
        // support native logging capabilities, e.g. Firebug for Firefox
        qx.log.appender.Native;
        // support additional cross-browser console. Press F7 to
toggle visibility
        qx.log.appender.Console;
      }
      {{XMLVM_MAIN_METHOD_CALL}}
    }
  }
});



this is lgpl code and the transformed code is copied into it,
which imho is a modification (not inheritance or linking) of the
original code, and should thus
be made available under the lgpl as well.

i might be wrong about this and as long as the intentions of the
project are clear
this is not going to be a problem at all.


best, hansi.

Re: [xmlvm-users] status and (L)GPL

[Panayotis K. <pan...@pa...>]

> 
> i would say that most code-generating programs will have a problem like this.
> for example, look at the bison (parser generator) exception for the
> generated code:
> http://www.gnu.org/software/bison/manual/html_node/Conditions.html
> (really worth the read, too long to quote)
> 

...

> more to this can found on the gpl-faq, which answers to "can i force
> the output of a gpl-program to be gpl as well"?
> http://www.gnu.org/licenses/gpl-faq.html#GPLOutput
> it says generally no; this is impossible because the output belongs to
> the user, however, it also says explicitly
> "For instance, part of the output of Bison (see above) would be
> covered by the GNU GPL, if we had not made an exception in this
> specific case."
> this is exactly what i think the problem with xmlvm output is (at this point).
> 


Actually these thoughts have been discussed a lot in the past, and thus the XMLVM project changed license from GPL to LGPL.
Not surprisingly, these were the actual arguments at that time.

From my point of view, still I can't see what the problem is with LGPL (which form my understanding is less restrictive than GPL (even with classpath exeption).
Btw, everybody knows why GNU.org and friends doesn't like LGPL and prefer the more restrictive GPL
http://www.gnu.org/licenses/why-not-lgpl.html

Re: [xmlvm-users] status and (L)GPL

[Arno P. <ar...@pu...>]

ok, guys, the generated code of gcc, bison, xmlvm do not fall under GPL 
or L-GPL. The clarification for bison is that since there is a bit of 
library portion that needs to be linked against the generated code 
automatically makes the whole application GPL (note that bison is GPL, 
not L-GPL). Since xmlvm is licensed under L-GPL, this problem does not 
occur.

IMHO, the idea of L-GPL is to allow commercial usage. Let me be clear 
about our intentions: we (the xmlvm core team) want developers to be 
able to use xmlvm for their commercial products. My understanding of the 
L-GPL is that is like the GPL with a Classpath Linking Exception.

Arno


On 12/1/10 10:13 AM, Hansi Raber wrote:
>> Please have a look:
>> http://www.gnu.org/licenses/lgpl.html
>> Section 4d
>>
>> If my rough interpretation above is correct, then I do not see how any
>> user, even with developer knowledge, would be able to modify a
>> distributed iPhone app (as one example) and be able to continue to use it.
>
> my feeling kinda agrees with thorsten,
> i'm not sure lgpl is "lose" enough because the resulting files contain
> a mix of "your" translated code that is stuffed into pre-written
> (lgpl) skeletons.
> this is neither inheritance nor linking, the two ways i know of in
> which the lgpl doesn't force the main application to be lgpl as well.
>
> i would say that most code-generating programs will have a problem like this.
> for example, look at the bison (parser generator) exception for the
> generated code:
> http://www.gnu.org/software/bison/manual/html_node/Conditions.html
> (really worth the read, too long to quote)
>
> imho this is mostly the same situation. adding a similar exception or
> providing the
> skeleton files with a more liberal license (e.g. bsd) would be a safe path.
>
>
> more to this can found on the gpl-faq, which answers to "can i force
> the output of a gpl-program to be gpl as well"?
> http://www.gnu.org/licenses/gpl-faq.html#GPLOutput
> it says generally no; this is impossible because the output belongs to
> the user, however, it also says explicitly
> "For instance, part of the output of Bison (see above) would be
> covered by the GNU GPL, if we had not made an exception in this
> specific case."
> this is exactly what i think the problem with xmlvm output is (at this point).
>
>
>
> best, hansi.
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App&  Earn a Chance To Win $500!
> Tap into the largest installed PC base&  get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> xmlvm-users mailing list
> xml...@li...
> https://lists.sourceforge.net/lists/listinfo/xmlvm-users

Re: [xmlvm-users] status and (L)GPL

[Hansi R. <su...@su...>]

> Please have a look:
> http://www.gnu.org/licenses/lgpl.html
> Section 4d
>
> If my rough interpretation above is correct, then I do not see how any
> user, even with developer knowledge, would be able to modify a
> distributed iPhone app (as one example) and be able to continue to use it.

my feeling kinda agrees with thorsten,
i'm not sure lgpl is "lose" enough because the resulting files contain
a mix of "your" translated code that is stuffed into pre-written
(lgpl) skeletons.
this is neither inheritance nor linking, the two ways i know of in
which the lgpl doesn't force the main application to be lgpl as well.

i would say that most code-generating programs will have a problem like this.
for example, look at the bison (parser generator) exception for the
generated code:
http://www.gnu.org/software/bison/manual/html_node/Conditions.html
(really worth the read, too long to quote)

imho this is mostly the same situation. adding a similar exception or
providing the
skeleton files with a more liberal license (e.g. bsd) would be a safe path.


more to this can found on the gpl-faq, which answers to "can i force
the output of a gpl-program to be gpl as well"?
http://www.gnu.org/licenses/gpl-faq.html#GPLOutput
it says generally no; this is impossible because the output belongs to
the user, however, it also says explicitly
"For instance, part of the output of Bison (see above) would be
covered by the GNU GPL, if we had not made an exception in this
specific case."
this is exactly what i think the problem with xmlvm output is (at this point).



best, hansi.

Re: [xmlvm-users] status and (L)GPL

[Thorsten S. <co...@pa...>]

> I'm not sure I understand your question. The "L" in L-GPL is similar in
> nature to the Classpath Exception. You can write your own application
> with XMLVM without having to license your application under the GPL (and
> therefore keeping your application proprietary). If you make
> modifications to XMLVM, you have to contribute those back to the community.

Thanks for your response. Yes, I figured that was the intention. But 
what worries me is that the LGPL talks about allowing the user to modify 
the app by replacing the LGPL library part with his or her customized 
version. Or something like that, the text is not exactly easy to read 
for me. Please have a look:
http://www.gnu.org/licenses/lgpl.html
Section 4d

If my rough interpretation above is correct, then I do not see how any 
user, even with developer knowledge, would be able to modify a 
distributed iPhone app (as one example) and be able to continue to use it.

Best regards,
Thorsten.