What is the current status?

Desktop Ruby-Gnome 2
1

I am looking into languages for linux programming and came upon the ruby
language. I use gnome so I was looking for gtk bindings. When I started
using the bindings included in edgy eft (the released ones) I kept
running into errors. It seems like the bindings are way out of date. I
compiled CVS and it helped some, but it is still pretty unusable. I
filed a bug about the problem I had.

The question is, what is the current status of this project? Is it being
maintained? Should I look into wxruby?

Thanks

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

2

Hi,

On 11/29/06, William J. [email protected] wrote:

I am looking into languages for linux programming and came upon the ruby
language. I use gnome so I was looking for gtk bindings. When I started
using the bindings included in edgy eft (the released ones) I kept
running into errors. It seems like the bindings are way out of date. I
compiled CVS and it helped some, but it is still pretty unusable. I
filed a bug about the problem I had.

The question is, what is the current status of this project? Is it being
maintained? Should I look into wxruby?

As a long time user, the status:

  • 0.15.x versions have tried to fix some hard (involving closures) to
    fix memory leaks, which cause some destabilization, but which is
    slowly disappearing; as long as errors are fixed and they are being.
    Report one with an example to reproduce it if you please. :slight_smile:

  • GTK and Gnome are in full development, constantly. Thus, even though
    the Ruby-GNOME2 has tried to keep with the times, it’s harder than it
    seems. Masao and now Kouhei have always tried their best (given their
    free times) to maintain the project. I think Masao has worked on some
    new GTK 2.10+ bindings, but it’s an endless job, because new versions
    of GTK+ come every few months, and there’s even documentation to
    maintain… Kouhei is the author/maintainer of the RCairo which is
    used in the newest GTK+ support as well. Last time I did a grep for
    Cairo in the project files, it showed a large list of hits.

  • wxruby might be interesting, but I for one don’t like it that much.
    YMMV. :slight_smile:

Lastly, it seems that there are always new users of Ruby-GNOME2
showing up around here. But some of them did have some problems with
the 0.15.x destabilization, even though the CVS HEAD version seems to
be more stable for everybody.

Cheers,
Joao

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

3

Hi,

On Wed, 29 Nov 2006 10:23:57 -0600
William J. [email protected] wrote:

I am looking into languages for linux programming and came upon the ruby
language. I use gnome so I was looking for gtk bindings. When I started
using the bindings included in edgy eft (the released ones) I kept
running into errors. It seems like the bindings are way out of date. I
compiled CVS and it helped some, but it is still pretty unusable. I
filed a bug about the problem I had.

Could you explain about your bug ?

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

4

Masao M. wrote:

compiled CVS and it helped some, but it is still pretty unusable. I
filed a bug about the problem I had.

Could you explain about your bug ?

http://sourceforge.net/tracker/index.php?func=detail&atid=470969&aid=1602773&group_id=53614

Here is the bug report. If you need any more info please ask.

Thanks

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

5

Hi,

Add to that, (No
offense Masao or Kouhei), neither speak much English. So it’s kinda a
language barrier there to. Though we’re slowly educating them to the
way us American’s talk, I’m sure.
WTF! Educating them?? What about you learning Japanese? You know, you
Americans are not the only people on earth. A (free software) community
is about people from all over the world, with different backgrounds,
cultures, and languages…

Unlike you, I don’t feel that barrier and I think that their English
skills are enough to communicate on this list and carry out their
maintainer’s duties. Besides, as this thread shows, there are always
other people to answer. What you said is irrelevant.

I think this is more of a technical issue. 0.14.1 bindings were very
good and quite stable despite those memory leaks. In an attempt to fix
them, 0.15 unfortunately introduced some new problems and the bindings
got much more unstable. As it was pointed out many times on this list,
the point is that those GC problems are very tough to debug. It is
difficult to isolate the problems from big existing applications (seg
faults, uninformative warnings etc).

CVS version is slowly getting better and we can hope that in a near
future all major problems will be fixed. If you can’t wait, I recommend
you use the 0.14.1 version. The only problem with that is that you won’t
be able to take part to the debugging effort.

The bindings remain for me the best bindings so far. Making UI with ruby
flavor is so great!

I hope you guys Masao and Kouhei won’t get pissed off by those comments.
Keep on the good work!

Mathieu

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

6

Joao P. wrote:

Hi,

On 11/29/06, William J. [email protected] wrote:

> free times) to maintain the project. I think Masao has worked on some > the 0.15.x destabilization, even though the CVS HEAD version seems to > be more stable for everybody. > > Cheers, > Joao I just wanted to add a couple of things to Joao's excellent explanation of the current status of Ruby-GNOME 2's project. It is very much still being actively worked on, though, as much as can be expected from only 2 developers working on the C[++] backend of the deal. Which is where the biggest deals of Ruby-GNOME2 has to be worked on. Add to that, (No offense Masao or Kouhei), neither speak much English. So it's kinda a language barrier there to. Though we're slowly educating them to the way us American's talk, I'm sure.

Plus, the difference between the bindings of Gnome2 and wxRuby, are
major, and substantial. wxRuby is going through a complete re-work of
the entire code base, for the bindings. They are converting from
hand-coded wrappers, to a SWIG interface. So there are still quite a
few instabilities with wxRuby, and it’s not guaranteed that the API is
going to remain the same. Where as, with Gnome2 bindings, there’s a lot
of foundation there already, that will work with most newer versions of
GTK+ API, however, the changes the GTK+ Development team are doing to
the memory handling, and such, is causing a lot of headaches, but a lot
of patches to be added to the bindings, from other authors, to fix these
problems. Ruby-GNOME2 will become stable again, as these patches come
along, and people help test the newest code. But, till such time, kinda
have to go with the flow of things, and understand, that it’s mostly a 2
man team that is working on the backend of this thing.

I only wish, there was a GTK-Slim for Windows, that would make the
distro a lot smaller for applications that I write, and make certain
things a bit faster, then I’d come back to Ruby-GNOME in a heartbeat.

As for the comment about the certain things a bit faster, I’m mainly
talking about the fact that Ruby-GNOME2’s Gtk::Entry in one of my
applications seems to be a bit slow to respond, when typing. I’ll end
up having to try the newer version of Ruby-GNOME2 to see if it has
improved. Which reminds me, anyone have a mswin32 compile of
Ruby-GNOME2 bindings for CVS?

L8ers,

Mario S.

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

7

Hi,

2006/11/30, William J. [email protected]:

Ruby-GNOME 2 / Bugs / #72 glade fails to work with gtknotebook

You need to add a widget to each tab.
It seems that this is a problem of glade.

Thanks,

kou

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

8

Mathieu B. wrote:

cultures, and languages…

Unlike you, I don’t feel that barrier and I think that their English
skills are enough to communicate on this list and carry out their
maintainer’s duties. Besides, as this thread shows, there are always
other people to answer. What you said is irrelevant.

The rest isn’t applicable here, cause I agree with it, and it doesn’t
apply to what I wanted to reply about. I had a feeling something like
this would come out, if I made the remark, and was trying to light step
it. But, I did want to point out a few things.

1.) Educating them was meant to be a light talk, wasn’t meant to be
derogatory.
2.) I have been learning some Japanese, but it’s been sparse and in
between. I love to learn their language, and their history more
in-depth, cause I am completely fascinated by their culture.
3.) Yes, I know American’s aren’t the only ones on the face of the
Earth, there are many cultures that make up the Earth, many religions,
and so forth. I believe in the right to be different, and have one’s own
view of the world, being Wiccan and all. So you won’t get no argument
from me there.
4.) I don’t see it as a barrier, for what Masao and Kouhei does, is very
admirable. Communication is one of the hardest things to do, when there
are so many perspectives, I meant it only as a warning for others, that
may get upset when they look at their replies, and may start to flame.
Hence for the part of the No offense to Masao and Kouhei. And they are
learning I am sure, by reading what we say, and translating it back into
their own natural language. It’s a learning experience for anyone who
speaks another language, Programming or not.
5.) American isn’t even a language per say, it doesn’t conform to the
English dialect as it was truely meant to be spoken in England. It is,
in all terms of the word, slang.
6.) Lastly, I didn’t mean this or the last as an offensive email to
anyone, I am sorry you took it that way. I only meant to point out to
be patient with the developers. They are doing what they can, with what
little time they have. Also, note that this is an English speaking
mailing list, designated by the ruby-gnome2-devel-en part of the email
address.

Again, I apologize for the offensiveness of my last email that anyone
took.

Apologetically yours,
Mario S.

9

Hi again,

As it was pointed out many times on this list,
the point is that those GC problems are very tough to debug.

This makes me think that RG2 does (or it seems to) not have unit tests.
I would be great to have a battery of tests that could be run every time
new code is committed to the CVS. It would also allow to detect
regressions.

Mathieu

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

10

On 12/1/06, Mario S. [email protected] wrote:

5.) American isn’t even a language per say, it doesn’t conform to the
English dialect as it was truely meant to be spoken in England. It is, in
all terms of the word, slang.

Gaah! And you just keep on doing it! American English is in every
respect as much a language as English English. The two languages have
grown from the same English used during the colonization of America.
One might consider American English a /dialect/ of English English,
but considering the amount of time they’ve been separate and the
amount of change incurred on both languages it’s reasonable to
describe them as two separate languages with the same origins.

You can find more information here:

nikolai

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

11

Hi,

Thanks folks.

I should explain the status of Ruby-GNOME2 and me ;).

At first, I need to comment some points.

On Wed, 29 Nov 2006 20:15:46 -0600
Mario S. [email protected] wrote:

slowly disappearing; as long as errors are fixed and they are being.
Report one with an example to reproduce it if you please. :slight_smile:

A lot of GC problem have been fixed by Kouhei.
The next version of Ruby-GNOME2 will be stable again.

But we can’t test the whole of APIs yet, I hope your helps more.

The most causes to delay developing Ruby-GNOME2 is that I
couldn’t spend my private time to develop Ruby-GNOME2.

  • My business has been so busy for a long time.
    You may surprise this thing, I work 10:00am to 23:00pm now.
    It’s not so special thing in Japan. But actually I’ve not have enough
    spare times.
    Unfortunately, it will continue until next March.

    My business is not related both of Ruby and Ruby-GNOME2.

  • Develop Ruby-GetText-Package for Ruby on Rails.
    I needed to make it as the most useful L10n library for Rails.
    So I concentrated to improve it on 4th quoters.
  • Helped some sections of “The Ruby Way second edition”
    I wrote Ruby-GetText section and updated Ruby-GNOME2 section.
    It was valuable experience but it consumed my private times.
  • Helped some other Japanese books for Ruby on Rails.

I just wanted to add a couple of things to Joao’s excellent explanation
of the current status of Ruby-GNOME 2’s project. It is very much still
being actively worked on, though, as much as can be expected from only 2
developers working on the C[++] backend of the deal. Which is where the
biggest deals of Ruby-GNOME2 has to be worked on. Add to that, (No
offense Masao or Kouhei), neither speak much English. So it’s kinda a
language barrier there to. Though we’re slowly educating them to the
way us American’s talk, I’m sure.

Hmm. I think there is no language barrier here.
Japanese developers join this ML and we discussed a lot of things here.
And also many sub-libraries are maintained by non-Japanese.

I admit I’m not good at English, though.

If you feel there is a few replying messages recently from me,
it’s just my problem which I describe it above.

Second, I explain the status of Ruby-GNOME2.

The last month, I’ve been back to develop Ruby-GNOME2 again.

Already, Ruby/GTK, GdkPixbuf, GDK, Pango have been updated completely
against the newest version of GTK+, Pango.

Now I’m working to update Ruby/ATK, and next, I’ll update Ruby/GLib and
other gnome stuff.

Then, I’ll release the new version of Ruby-GNOME2. It will be the end of
this year.

But I’ve not have enough time for developing Ruby-GNOME2 yet.
Any helps are welcomed.

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

12

Kouhei S. wrote:

Thanks,

kou

It seems to have worked. Thanks!

I say seem because I ran into another bug while doing it, but I will
figure this one out.

On another note, I am not a C programmer so I have no clue what the
difficulty is, but has anyone ported or considered porting libsexy?

Thanks

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

13

On Fri, Dec 01, 2006 at 08:34:34AM -0600, William J. wrote:

On another note, I am not a C programmer so I have no clue what the
difficulty is, but has anyone ported or considered porting libsexy?

As someone who has tried to port bits of libsexy to gtk# (the C#
language binding of gtk+), I can say from experience that libsexy makes
use of some private fields in GtkEntry, etc to manage to do what it does
for things like the validating/input masked entry. Mainly for accessing
private GdkWindows to resize them, etc, iirc.

There’s probably lots of libsexy that could be ported, but beware of
things like this which are not so easily done from a language binding of
gtk+.

-pete

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

14

Hi,

I noticed glade-2 creates some helper source codes for
GtkNotebook.
I’ll try that ruby-glade-create-template creates them.
But now, you need to write the same code in Ruby by yourself.

On Thu, 30 Nov 2006 16:22:58 +0900
“Kouhei S.” [email protected] wrote:

–
ruby-gnome2-devel-en List Signup and Options

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

15

On 12/2/06, Mario S. [email protected] wrote:

Many, normal words that are
perfectly good words, have been given derogative meanings to them. One
perfect example being the word “Gay” as such. It was originally an
adjective meaning “carefree”, “happy”, or “bright and showy”, however,
it’s been turned into a derogative meaning for Homosexuality, and meant
to be a curse word.

Considering that the word “gay” was initially used by homosexuals to
describe themselves, this isn’t quite the turn of events. It’s even
noted in the Wikipedia article I’m guessing you got the meaning from.
And just because some idiots have a hard time with homosexuals doesn’t
mean that “gay” should be considered derogative or a curse word in the
large.

Finally, languages evolve. As soon as a language stops changing, it’s
dead.

nikolai

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

16

Already, Ruby/GTK, GdkPixbuf, GDK, Pango have been updated completely
against the newest version of GTK+, Pango.

This is very good news! I feel sorry I began working on adding new
features of gtk+2.10 but couldn’t complete. It’s fortunate you could
complete it.

Then, I’ll release the new version of Ruby-GNOME2. It will be the end of this year.

I’ve just built latest CVS and tried to toy around with my program
booh to see if there was still problems. Unfortunately, the program
eventually aborted with the “[BUG] Segmentation fault” ruby message.
And even more unfortunately, it happened after quite a lot of
operations in the program and was not reproduceable with the last
thing I did when it happened :confused: The suggestion to track it down to a
small reproduceable program is very hard in that kind of scenario.

So I tried to use a different approach: run the program through
valgrind, it is often a very useful and clever tool to hunt down
memory allocation problems. It showed me some invalid reads in
rbgobj_closure.c (read within a memory chunk which is already free’d).
I had a tough time understanding the intrinsics of rbgobj_closure.c
and understand the scenario of the problem, but I came up with
something:

  1. at a point of time, Ruby’s GC decides that a closure holder’s
    memory needs to be reclaimed, thus calls gr_closure_holder_free, which
    in turn calls rclosure_invalidate, which doesn’t call rclosure_unref
    because the counter is at 2 at that time

  2. at a later point of time, GTK’s GC sees that the closure
    corresponding the the closure holder’s of step #1 needs to be
    finalized, thus calls rclosure_invalidate, with the same outcome as
    upper (no rclosure_unref), then finalizes the closure (frees the
    memory allocated for it)

  3. at an even later point of time, GTK’s GC finalizes an object on
    which the closure was attached, so calls rclosure_weak_notify, which
    calls rclosure_alive_p, which performs two invalids memory reads (to
    get the count and the holder), as the memory allocated for the closure
    was freed at step #2

A possible fix might be to not use rclosure_alive_p in
rclosure_invalidate, in favor of only testing whether the counter is
superior to 0, so that rclosure_invalidate would really perform the
necessary weak_unref calls - this seems to fix the problems I could
see.

–
Guillaume C. - Guillaume Cottenceau

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

17

Nikolai W. wrote:

English language - Wikipedia
American English - Wikipedia
English language in England - Wikipedia

Thank you for the Articles Nikolai.

But, I’m not going to digress back into this subject. I still believe
in the fact that a lot of the English words used today amongst the North
America, have been turned into slang quite a bit. Not saying so much
that English isn’t a Language, or American English isn’t a language,
hence for the part about “per say” in my last email. You have to admit,
that a lot of cultural complexities have diverged the barrier of Proper
English, Germanic English, and so forth. Many, normal words that are
perfectly good words, have been given derogative meanings to them. One
perfect example being the word “Gay” as such. It was originally an
adjective meaning “carefree”, “happy”, or “bright and showy”, however,
it’s been turned into a derogative meaning for Homosexuality, and meant
to be a curse word.

But, I digress, and therefore will no longer post any more messages
concerning this. Once again, if anyone is offended by my emails
concerning this subject, it was in no way meant to be that way, Masao
has admitted that he doesn’t speak English well, so that was my only
point to this entire matter.

L8ers,

Mario

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

18

see.
So, it seems that 3. can’t call rclosure_weak_notify().
Please check again my explanation: rclosure_unref is not called in
step #1 or #2.

Hmm… Could you make a patch for your idea?

I explained it in the last paragraph of my previous message, but I can
show it as code if you prefer:

— glib/src/rbgobj_closure.c 12 Oct 2006 13:30:22 -0000 1.37
+++ glib/src/rbgobj_closure.c 6 Dec 2006 11:25:31 -0000
@@ -260,7 +263,7 @@
{
GRClosure rclosure = (GRClosure)closure;

  • if (rclosure_alive_p(rclosure)) {
  • if (rclosure->count > 0) {
    rclosure->count = 1;
    rclosure_unref(rclosure);
    }

I can’t commit your idea until I can reproduce your problem
and confirm your idea fixes the problem…

I understand, unfortunately, as I explained, it would be very hard to
make a small testcase exhausting a segfault.

But maybe we can discuss the principles of the problem, that’s what
I’ve tried to explain in my 1/2/3 scenario upper. Maybe with the
additional information I’ve replied upper, you would agree with the
scenario? In my opinion, the scenario I’ve explained clearly
demonstrates a flaw which should amyway be fixed.

Else, if you’re willing to use valgrind and to try with my program
booh, I can explain a way for you to reproduce the valgrind-reported
memory problem (though it’s a bit long to setup, and then it takes a
couple of minutes each time because of valgrind overhead).

–
Guillaume C. - Guillaume Cottenceau

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

19

Hi,

In [email protected]
“Re: [ruby-gnome2-devel-en] What is the current status?” on Wed, 6 Dec
2006 11:05:24 +0100,
“Guillaume C.” [email protected] wrote:

see.
In rclosure_unref() (from rclosure_invalidate()),
rclosure_weak_notify() is removed from all attached objects:

    for (next = rclosure->objects; next; next = next->next) {
        GObject *object;
        object = G_OBJECT(next->data);
        g_object_weak_unref(object, rclosure_weak_notify, rclosure);
    }

So, it seems that 3. can’t call rclosure_weak_notify().

Hmm… Could you make a patch for your idea?
I can’t commit your idea until I can reproduce your problem
and confirm your idea fixes the problem…

Thanks,

kou

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

20

Hi,

In [email protected]
“Re: [ruby-gnome2-devel-en] What is the current status?” on Wed, 6 Dec
2006 12:30:13 +0100,
“Guillaume C.” [email protected] wrote:

Please check again my explanation: rclosure_unref is not called in
step #1 or #2.

Both #1 and #2 calls rclosure_invalidate() but don’t call
rclosure_unref(). Is it OK?

If it’s true, it seems that the first rclosure_invalidate()
call invokes rclosure_unref(). Because rclosure_invalidate()
force to decrease the counter to 1 when rclosure is alive:

if (rclosure_alive_p(rclosure)) {
    rclosure->count = 1;
    rclosure_unref(rclosure);
}

Ah. Does what you said mean that rcosure is already dead
when the first rclosure_invalidate() call?

I explained it in the last paragraph of my previous message, but I can
show it as code if you prefer:

Thanks. I’m good in programming language rather than
English. X<

 }

rcosure isn’t freed but rclosure->rb_holder is freed. Is it
OK? If it’s true, does the following patch solve your
problem:

Index: src/rbgobj_closure.c

RCS file: /cvsroot/ruby-gnome2/ruby-gnome2/glib/src/rbgobj_closure.c,v
retrieving revision 1.37
diff -u -p -r1.37 rbgobj_closure.c
— src/rbgobj_closure.c 12 Oct 2006 13:30:22 -0000 1.37
+++ src/rbgobj_closure.c 6 Dec 2006 13:12:18 -0000
@@ -251,6 +251,7 @@ rclosure_unref(GRClosure *rclosure)
GRClosureHolder *holder;
Data_Get_Struct(rclosure->rb_holder, GRClosureHolder,
holder);
holder->closure = NULL;

  •        rclosure->rb_holder = Qnil;
   }
    }
    }

Thanks,

kou

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net’s Techsay panel and you’ll get the chance to share
your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV