[Vala] How to make shared library in win32?

[san hoi]

I try to build libdvw shared library. But following error occur.

https://live.gnome.org/Vala/SharedLibSample
http://gitorious.org/libdvw/libdvw

toddw@TODDW-PC /C/msys/home/toddw/libdvw/dvw (master)
$ valac --pkg config --pkg gtk+-2.0 --pkg gee-1.0 --pkg libxml-2.0
--vapidir=../vapi/ *.vala --disable-warnings --library dvw-0.1

c:/msys/home/toddw/libdvw/dvw/util.vala.c: In function
`_vala_dvw_point_get_property':
c:/msys/home/toddw/libdvw/dvw/util.vala.c:586: warning: passing arg 2 of `g_valu
e_set_pointer' discards qualifiers from pointer target type
c:/msys/home/toddw/libdvw/dvw/util.vala.c:589: warning: passing arg 2 of `g_valu
e_set_pointer' discards qualifiers from pointer target type
c:/vala-0.12.0/lib/libmingw32.a(main.o):main.c:(.text+0xbd): undefined reference
 to `WinMain@16'
collect2: ld returned 1 exit status
error: cc exited with status 1
Compilation failed: 1 error(s), 0 warning(s)

I try to use Carl's 13.1 valac environment too. But it's a same result.
Thanks in advance for any suggetions.
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

[Vala] Error attempting to use WebKit-3.0.gir

[bsquared]

How can I use this gir directly.
I get the following errors:

$ valac --pkg gtk+-3.0 --pkg json-glib-1.0 --pkg JSCore-3.0 --pkg
WebKit-3.0   --vapidir . --pkg config gss_js_vala.vala
JSCore-3.0.gir:0.0-0.0: error: The type name `JSGlobalContextRef'
could not be found
JSCore-3.0.gir:0.0-0.0: error: The type name `JSObjectRef' could not be found
gss_js_vala.vala:31.3-31.8: error: The symbol `WebKit' could not be found
WebKit.WebView webview = new WebKit.WebView();
^^
Compilation failed: 3 error(s), 0 warning(s)


What am I doing wrong?

-- 
Regards,
-Brian
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala logo symbol voting

[Dru Moore]

Probably far too late in the process to come back and comment on this
but I have a few general comments.

I really like the concept of wings etc and think of the logo marks,
all but #4 are really very strong, to me #4 is the weakest and is only
good.

However, I really dislike the font used for VALA.

I totally get that by reducing the number of strokes required to form
the letters you are trying to evoke simplicity and lightness as shared
qualities. For me it actually suggests a different, less accurate set
of ideas:
Only a few basic tools - 4 unique strokes build the entire word
Very direct and angular - fast yes, elegant no.

These would seem more to be the qualities of a low level language
(assembler or c) rather than a modern, high level language. Vala
provides lot's of tools and high level objects to it's users that
should let us express ourselves in a much more elegant fashion.

Appreciate these comments may be coming at the least helpful time and
regardless the outcome will clearly be to a very high standard.

On 31 August 2011 18:20, Tobias Bernard  wrote:
> I totally agree with you that war isn't a good symbolism for this, it is one
> of the reasons I deleted that proposal.
>
> The reason why I created it in the first place was the sound of "Vala": it
> reminds "of Valkyrie" or "Valhalla" (and if you've read the Silmarillion it
> reminds of the "Valar"), and ancient weapons are strongly connected to those
> topics, so I wanted to try working on that. But since there are better
> symbolisms (which also look better graphically :D), I deleted that proposal.
>
> I don't know if you've taken a look at the pdf, but there are several
> proposals which communicate exactly what you said: lightness, easiness,
> flying etc.
> A flying carpet isn't a bad idea but personally, I like a wing better,
> because it's less complex and easier to understand.
> in case you've missed it: symbol proposals pdf
> download
>
> kind regards
> tobias
>
> 2011/8/31 Jim Peters 
>
>> Sam Liddicott wrote:
>> > >2) The symbolism of "ancient warfare" is hard to
>> > >understand. Probably too hard to be understood instantly by most
>> > >people, so I put that proposal away too.
>> >
>> > The presentation is very good.
>> >
>> > I think that symbols of weapons, killing and/or warfare should be
>> > avoided, having very current and unpleasant connotations for many
>> > people in specific locales.
>>
>> War happens -- deal with it.  But anyway, Vala is not about war, not
>> even about posing with weapons, nor indulging in feelings of
>> superiority or anything like that.  Vala is practical and gets the job
>> done.  It is not an academic language made by purists or theorists.
>> It doesn't seem to attract zealots or fanatics -- it is relatively
>> humble in that sense.  It makes life easier for those who use it --
>> compared to the alternative tools available to do the same job.
>>
>> Getting a good visual logo can be hard.  When I was involved in
>> OpenEEG, someone sent us a drawing of a levitating, meditating,
>> EEG-wired GNU, which was really perfect and said everything:
>>
>>  http://openeeg.sf.net/
>>
>> However, a programming language is more abstract.  Linux has a penguin
>> mascot pictured with a bellyful of fish (according to Linus), which I
>> guess represents a feeling of comfort.
>>
>> I wonder whether we can find a feeling connection for Vala?
>>
>> To me it is about flying above all the nightmare complexity of pages
>> of boilerplate code that is required at the C level.  So maybe
>> Aladdin's flying carpet?  Or something else to represent the ease and
>> feeling of flying above complexity?
>>
>> Any other way we can represent this feeling?
>>
>> Or is there some other feeling we can associate with Vala?
>>
>> Jim
>>
>> --
>>  Jim Peters                  (_)/=\~/_(_)                 j...@uazu.net
>>                          (_)  /=\  ~/_  (_)
>>  Uazú                  (_)    /=\    ~/_    (_)                http://
>>  in Peru            (_)  /=\  ~/_  (_)            uazu.net
>> ___
>> vala-list mailing list
>> vala-list@gnome.org
>> http://mail.gnome.org/mailman/listinfo/vala-list
>>
>
> ___
> vala-list mailing list
> vala-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/vala-list
>
>
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

[Vala] Google Native Client and Vala

[Andrea Del Signore]

Just in case that's of some interest to someone, I'm doing some
experiments binding the Google Native Client API [1] and writing a nacl
module in vala.

I pushed my results in this git repos:
https://gitorious.org/ideas/nacl-vala

Regards,
Andrea

[1] http://code.google.com/chrome/nativeclient/

___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala completion support for vim

[Damien Radtke]

I believe I've found the problem. =)

I figured out how to enable debugging and discovered that the segfault
occurred in the CodeDom destructor. Part of the destructor runs a foreach
loop on "source_files", which was null. In order to make sure that object
creation worked correctly, my test included creating a CompletionEngine but
adding nothing to it; hence, a null list and the cause of the segfault. It
should be easily fixed by making sure that the list always exists, even if
it's empty.

On Wed, Aug 31, 2011 at 10:25 AM, Damien Radtke wrote:

> I don't believe the segfault bug is limited to the CompletionEngine
> destructor. I ran a quick test by commenting out the entire destructor, and
> it didn't help (remember I also ran into the problem when I tried creating a
> CodeDom). I did, however, notice that it adds some log messages via
> Utils.trace; how do I check the log? It looks like it just passes it along
> to GLib.logv(), but I don't know where this method sends the information.
>
>
> On Wed, Aug 31, 2011 at 1:09 AM, Andrea Del Signore wrote:
>
>> Hi
>>
>> On Tue, 2011-08-30 at 09:29 -0700, Damien Radtke wrote:
>>
>> >
>> > This program ends with ** Message: symbolresolver.vala:54: file
>> > program.vala - can't resolve using directive: Afrodite. I'm assuming
>> > this means that I would have to add libafrodite-0.12.vapi to the
>> > engine manually, but it's installed in the right place, so I shouldn't
>> > have to.
>>
>> No you have to add the vapi to the engine. Afrodite just parse and build
>> the codedom it doesn't know anything about the project structure.
>>
>> In vala-toys another library, vbf (vala build framework), is used to
>> parse the project structure.
>>
>> Anyway when you have the vapi file name, you can use the
>> Afrodite.Utils.get_package_paths to get real full path and the vapi
>> dependecies.
>>
>>
>> >
>> > If I comment out everything between the creation of the engine and
>> > "return 0", then it again exits with a segfault.
>>
>> I think that is a bug in the completionengine object desctructor.
>>
>> > I'd be happy to help contribute to afrodite, but it might take me a
>> > while to learn how it works and where it needs improvements. Any help
>> > or links to good resources would be great.
>>
>> I'm sorry but there isn't such documentation ;) All we have is the
>> afrodite source code and vala compiler source code.
>>
>> So feel free to ask, you can also find me on the vala irc channel I'm
>> sejerpz.
>>
>> > Thanks again,
>> > ~Damien
>>
>> Ciao,
>> Andrea
>>
>>
>>
>
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala completion support for vim

[Andrea Del Signore]

On Wed, 2011-08-31 at 10:25 -0700, Damien Radtke wrote:
> I don't believe the segfault bug is limited to the CompletionEngine
> destructor. I ran a quick test by commenting out the entire
> destructor, and it didn't help (remember I also ran into the problem
> when I tried creating a CodeDom). I did, however, notice that it adds
> some log messages via Utils.trace; how do I check the log? It looks
> like it just passes it along to GLib.logv(), but I don't know where
> this method sends the information.
> 

The log messages should be displayed in the terminal, just make sure
that you built afrodite with --enable-debug in the configure switch.

HTH,
Andrea

___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala completion support for vim

[Damien Radtke]

I don't believe the segfault bug is limited to the CompletionEngine
destructor. I ran a quick test by commenting out the entire destructor, and
it didn't help (remember I also ran into the problem when I tried creating a
CodeDom). I did, however, notice that it adds some log messages via
Utils.trace; how do I check the log? It looks like it just passes it along
to GLib.logv(), but I don't know where this method sends the information.

On Wed, Aug 31, 2011 at 1:09 AM, Andrea Del Signore  wrote:

> Hi
>
> On Tue, 2011-08-30 at 09:29 -0700, Damien Radtke wrote:
>
> >
> > This program ends with ** Message: symbolresolver.vala:54: file
> > program.vala - can't resolve using directive: Afrodite. I'm assuming
> > this means that I would have to add libafrodite-0.12.vapi to the
> > engine manually, but it's installed in the right place, so I shouldn't
> > have to.
>
> No you have to add the vapi to the engine. Afrodite just parse and build
> the codedom it doesn't know anything about the project structure.
>
> In vala-toys another library, vbf (vala build framework), is used to
> parse the project structure.
>
> Anyway when you have the vapi file name, you can use the
> Afrodite.Utils.get_package_paths to get real full path and the vapi
> dependecies.
>
>
> >
> > If I comment out everything between the creation of the engine and
> > "return 0", then it again exits with a segfault.
>
> I think that is a bug in the completionengine object desctructor.
>
> > I'd be happy to help contribute to afrodite, but it might take me a
> > while to learn how it works and where it needs improvements. Any help
> > or links to good resources would be great.
>
> I'm sorry but there isn't such documentation ;) All we have is the
> afrodite source code and vala compiler source code.
>
> So feel free to ask, you can also find me on the vala irc channel I'm
> sejerpz.
>
> > Thanks again,
> > ~Damien
>
> Ciao,
> Andrea
>
>
>
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala logo symbol voting

[Tobias Bernard]

I totally agree with you that war isn't a good symbolism for this, it is one
of the reasons I deleted that proposal.

The reason why I created it in the first place was the sound of "Vala": it
reminds "of Valkyrie" or "Valhalla" (and if you've read the Silmarillion it
reminds of the "Valar"), and ancient weapons are strongly connected to those
topics, so I wanted to try working on that. But since there are better
symbolisms (which also look better graphically :D), I deleted that proposal.

I don't know if you've taken a look at the pdf, but there are several
proposals which communicate exactly what you said: lightness, easiness,
flying etc.
A flying carpet isn't a bad idea but personally, I like a wing better,
because it's less complex and easier to understand.
in case you've missed it: symbol proposals pdf
download

kind regards
tobias

2011/8/31 Jim Peters 

> Sam Liddicott wrote:
> > >2) The symbolism of "ancient warfare" is hard to
> > >understand. Probably too hard to be understood instantly by most
> > >people, so I put that proposal away too.
> >
> > The presentation is very good.
> >
> > I think that symbols of weapons, killing and/or warfare should be
> > avoided, having very current and unpleasant connotations for many
> > people in specific locales.
>
> War happens -- deal with it.  But anyway, Vala is not about war, not
> even about posing with weapons, nor indulging in feelings of
> superiority or anything like that.  Vala is practical and gets the job
> done.  It is not an academic language made by purists or theorists.
> It doesn't seem to attract zealots or fanatics -- it is relatively
> humble in that sense.  It makes life easier for those who use it --
> compared to the alternative tools available to do the same job.
>
> Getting a good visual logo can be hard.  When I was involved in
> OpenEEG, someone sent us a drawing of a levitating, meditating,
> EEG-wired GNU, which was really perfect and said everything:
>
>  http://openeeg.sf.net/
>
> However, a programming language is more abstract.  Linux has a penguin
> mascot pictured with a bellyful of fish (according to Linus), which I
> guess represents a feeling of comfort.
>
> I wonder whether we can find a feeling connection for Vala?
>
> To me it is about flying above all the nightmare complexity of pages
> of boilerplate code that is required at the C level.  So maybe
> Aladdin's flying carpet?  Or something else to represent the ease and
> feeling of flying above complexity?
>
> Any other way we can represent this feeling?
>
> Or is there some other feeling we can associate with Vala?
>
> Jim
>
> --
>  Jim Peters  (_)/=\~/_(_) j...@uazu.net
>  (_)  /=\  ~/_  (_)
>  Uazú  (_)/=\~/_(_)http://
>  in Peru(_)  /=\  ~/_  (_)uazu.net
> ___
> vala-list mailing list
> vala-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/vala-list
>
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Is a Direct GCallback to signals possible in vala?

[Tal Hadad]

Sorry about my mistake, I should wrote at the end:
widget.hide_on_delete()

Tal

From: tal...@hotmail.com
To: vala-list@gnome.org
Date: Wed, 31 Aug 2011 17:23:30 +0300
Subject: [Vala] Is a Direct GCallback to signals possible in vala?

 
I take a look in the generated C code, and I thought, does signal handling
have to be two stages? It possible to do it on 1 stage, right?
I'll give an example that shows this behavior:
Take a look on gtk_widget_hide_on_delete method in GTK. This method
were ment to be used as direct callback to delete-event, for popup hiding
popup windows when user click something outside of this window.
I thought I'll embrace it in my Vala code, but I run into a little mess- access
to this method denied. This is because of vala behavior a.foo (...) instead
of ..._foo (a, ...) .
I "fixed" this by adding Gtk.widget_hide_on_delete to my low_binding.vapi
file. So this is how it looked:
[CCode (cheader_filename = "gtk/gtk.h", cname_prefix = 
"gtk_widget_hide_on_delete")]
public static bool widget_hide_on_delete (Widget widget);
 
But then I get:
error: Argument 1: Cannot convert from`Gtk.widget_hide_on_delete' to 
`Gtk.Widget.delete_event'
I tried to add a parameter named "Gdk.Event event" to the method above.
I get error from CC that too many arguments in this method, which were only
logical to me after I looked on the generated code. So far, the only way to use
this method is by writing another method or lambda that calls to:
Widget.hide_on_delete (widget)
Is it possible to do it by a single GCallback? I'm talking only about static
methods, because that for the others have the "self" order problem.
  

___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list
  ___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

[Vala] Is a Direct GCallback to signals possible in vala?

[Tal Hadad]

I take a look in the generated C code, and I thought, does signal handling
have to be two stages? It possible to do it on 1 stage, right?
I'll give an example that shows this behavior:
Take a look on gtk_widget_hide_on_delete method in GTK. This method
were ment to be used as direct callback to delete-event, for popup hiding
popup windows when user click something outside of this window.
I thought I'll embrace it in my Vala code, but I run into a little mess- access
to this method denied. This is because of vala behavior a.foo (...) instead
of ..._foo (a, ...) .
I "fixed" this by adding Gtk.widget_hide_on_delete to my low_binding.vapi
file. So this is how it looked:
[CCode (cheader_filename = "gtk/gtk.h", cname_prefix = 
"gtk_widget_hide_on_delete")]
public static bool widget_hide_on_delete (Widget widget);

But then I get:
error: Argument 1: Cannot convert from`Gtk.widget_hide_on_delete' to 
`Gtk.Widget.delete_event'
I tried to add a parameter named "Gdk.Event event" to the method above.
I get error from CC that too many arguments in this method, which were only
logical to me after I looked on the generated code. So far, the only way to use
this method is by writing another method or lambda that calls to:
Widget.hide_on_delete (widget)
Is it possible to do it by a single GCallback? I'm talking only about static
methods, because that for the others have the "self" order problem.
  ___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Was GNOME Panel applet example removed?

[Frederik]

On 08/31/2011 12:52 PM, Alessandro Pellizzari wrote:
> Il giorno mar, 30/08/2011 alle 22.40 +0300, Tal Hadad ha scritto:
> 
>> However, GNOME3 has different opinions among users.
>> So I think that it's wrong to remove examples just because GNOME moved on.
> 
> Totally agree.
> 
> Moreover, there is a fork of gnome2 (mate), and I think gnome2 and
> eventually mate will be used for at least 3-4 years from now.
> 
> Bye.
> 

Hi,

due to some demand on this mailing list I have re-added the Panel
applet example and created a page where deprecated stuff can go:

  https://live.gnome.org/Vala/DeprecatedSamples

Also, I have added a note + link on the GTK+ example page with
information on how to find the old GTK+ 2 sample code:

  https://live.gnome.org/Vala/GTKSample


Best regards,

Frederik
___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Was GNOME Panel applet example removed?

[Alessandro Pellizzari]

Il giorno mar, 30/08/2011 alle 22.40 +0300, Tal Hadad ha scritto:

> However, GNOME3 has different opinions among users.
> So I think that it's wrong to remove examples just because GNOME moved on.

Totally agree.

Moreover, there is a fork of gnome2 (mate), and I think gnome2 and
eventually mate will be used for at least 3-4 years from now.

Bye.

-- 
Alessandro Pellizzari

___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list

Re: [Vala] Vala completion support for vim

[Andrea Del Signore]

Hi

On Tue, 2011-08-30 at 09:29 -0700, Damien Radtke wrote:

> 
> This program ends with ** Message: symbolresolver.vala:54: file
> program.vala - can't resolve using directive: Afrodite. I'm assuming
> this means that I would have to add libafrodite-0.12.vapi to the
> engine manually, but it's installed in the right place, so I shouldn't
> have to.

No you have to add the vapi to the engine. Afrodite just parse and build
the codedom it doesn't know anything about the project structure.

In vala-toys another library, vbf (vala build framework), is used to
parse the project structure.

Anyway when you have the vapi file name, you can use the
Afrodite.Utils.get_package_paths to get real full path and the vapi
dependecies.


> 
> If I comment out everything between the creation of the engine and
> "return 0", then it again exits with a segfault.

I think that is a bug in the completionengine object desctructor.

> I'd be happy to help contribute to afrodite, but it might take me a
> while to learn how it works and where it needs improvements. Any help
> or links to good resources would be great.

I'm sorry but there isn't such documentation ;) All we have is the
afrodite source code and vala compiler source code.

So feel free to ask, you can also find me on the vala irc channel I'm
sejerpz.

> Thanks again,
> ~Damien

Ciao,
Andrea


___
vala-list mailing list
vala-list@gnome.org
http://mail.gnome.org/mailman/listinfo/vala-list