gmane.comp.gnome.devtools.gob.general

[PATCH] Don't output #line 0 directives for prototypes.

Nick Bowler — 2011-03-31T01:25:28

It is not valid for the integer argument of a #line directive to be 0.
However, the generated C source will contain such a directive around the
prototypes of private functions that do not appear in the .gob file.
For example, the trivial

  class My:Test from G:Object {
  }

generates the invalid C code:

  #line 0 "my-test.gob"
  static void my_test_init (MyTest * o) G_GNUC_UNUSED;
  /* ... */
  #line 0 "my-test.gob"
  static void my_test_class_init (MyTestClass * c) G_GNUC_UNUSED;

Fix this up and add an assertion to out_addline_infile to catch future
problems of this nature.
---

Hopefully this is the right place to send patches!  The gob2 website
doesn't make any mention of where they should be sent.

 src/main.c |    6 ++++--
 src/out.c  |    2 ++
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/main.c b/src/main.c
index 5d2d4c9..cd8edef 100644
--- a/src/main.c
+++ b/src/main.c
< at >< at > -578,7 +578,8 < at >< at > put_priv_method_prot(const Method *m)
     m->cbuf)) {
 /* add unique ID */
 char *

Re: Call order of parrent methods in dispose() and finalize()

Jiri Lebl — 2011-03-06T07:09:45

Not that I remember, I'll take a look at it.

Jiri

On Sat, Mar 5, 2011 at 10:08 PM, Dmitri Toubelis
<dmitri.toubelis< at >alkeron.com> wrote:

Re: Building on ubuntu 10.04 LTS fails

Jiri Lebl — 2011-03-06T07:07:55

Install flex.  It should fail configure ... I'll take a look at that.

Jiri

On Sat, Mar 5, 2011 at 9:47 PM, Dmitri Toubelis
<dmitri.toubelis< at >alkeron.com> wrote:

Call order of parrent methods in dispose() and finalize()

Dmitri Toubelis — 2011-03-06T04:08:39

Hi,

I noticed that implementation of __dispose() and __finalize() methods generated by GOB includes calls corresponding dispose() and finalize() methods of the parent BEFORE deallocating local objects. GObject specification states to call them AFTER local resources have been freed. In most cases that should not be an issue, but the fact that it is inconsistent with the specification makes me cautious. Is there any reason for this behavior?

Regards,
Dmitri



--
to unsubscribe:
send mail to minimalist< at >5z.com with "unsubscribe gob-list" in the subject

Building on ubuntu 10.04 LTS fails

Dmitri Toubelis — 2011-03-06T03:47:34

Hi, 

I'm trying to build gob-2.0.18 from source tar ball on Ubuntu 10.04 LTS and it fails with the following messages: 

dev02:~/src/gob2-2.0.18> ./configure 
checking for a BSD-compatible install... /usr/bin/install -c 
checking whether build environment is sane... yes 
checking for a thread-safe mkdir -p... /bin/mkdir -p 
checking for gawk... gawk 
checking whether make sets $(MAKE)... yes 
checking whether to enable maintainer-specific portions of Makefiles... no 
checking for style of include used by make... GNU 
checking for gcc... gcc 
checking whether the C compiler works... yes 
checking for C compiler default output file name... a.out 
checking for suffix of executables... 
checking whether we are cross compiling... no 
checking for suffix of object files... o 
checking whether we are using the GNU C compiler... yes 
checking whether gcc accepts -g... yes 
checking for gcc option to accept ISO C89... none needed 
checking dependency style of gcc... gcc3 
checking for library containing strerror...

ANNOUNCE: gob2 2.0.18 the "bit torture" release

Jiri Lebl — 2011-01-06T19:48:52

UGUGUFEEEEEEEEEEEEEEEEEEEEEEEEEE

Perfection can always be improved: Time for a new gob version.

So what is this gob thing?  Well besides being the cure for cancer, it also
generates GObjects (or GTK+ objects).  GOB2 is a replacement for the version
1 GOB, that was for GTK+ 1.x mostly.  GOB2 can handle pretty much most of the
GObject features.  At least most of the ones that anyone will ever use.  It
only requires GLib 2.0 (or higher for some features) and can generate
arbitrary GObjects.  You can have both versions installed at the same time if
you wish, but if anyone is still using gob version 1 and GTK+ 1.x, they
should get their head examined.

Here are the news in 2.0.18:

* Add %ctop{ %} to do what alltop does but only for the C file
* Buildfixes for Cygwin (Roland Clobus)
* gob2.m4 exits if gob is not found (fixes ubuntu #317801)
          (Stephen Kitt)
* Escape minus signs in the webpage and fix some typos (Stephen Kitt)
* clarify README as to the public domainness of the generated code

ftp:/

Re: Need for %ctop{ %}

Jiri Lebl — 2011-01-03T20:07:11

Hmmm, that's not a bad idea.  Here's a patch.  Make sure to run "make
distclelan" and rerun the autogen.sh to make sure that everything gets
rebuilt.  I'll make a new release of gob at some point soon.

The patch adds %ct{ and %ctop{

Jiri

On Thu, Dec 30, 2010 at 9:12 AM, Dmitri Toubelis
<dmitri.toubelis< at >alkeron.com> wrote:

Need for %ctop{ %}

Dmitri Toubelis — 2010-12-30T17:12:19

Hi, 

I'm new to this list but I was using GOB with autotools for several years now. GOB is a excellent tool and it is saved us a lot of time. However, all this time I was having intermittent problems with our builds. I was fixing these problems by adding temporary hacks to a standard autotool's build process. Finally I had some time to investigate them properly and here is my observations: 

- We use autotools to generate config.h file that needs to be included as a first thing in *.c files.

- I would normally put it at the top of %{ %} section but then I was having problems and in particular with large file support that didn't work on GOB generated files and worked flawlessly on manually created ones. 

- It appears that top of %{ %} section does not translate to the top of corresponding *.c file and GOB includes its headers before this point when it is too late for config.h.

- To prove my suspicion I tried to include config.h in %alltop{ %} section and it worked alright. However, config.h should not be

Re: Build gob2 on Cygwin

Jiri Lebl — 2010-10-09T19:18:42

Not sure exactly why the patch is needed.  Though now I am not
completely understanding what the Makefile.am was doing.  I fixed
things up and removed anything unused in git.  So this should be
included in the next release.

Jiri

On Thu, Oct 7, 2010 at 11:23 AM, Roland Clobus
<rclobus< at >users.sourceforge.net> wrote:

Build gob2 on Cygwin

Roland Clobus — 2010-10-07T18:23:55

--
to unsubscribe:
send mail to minimalist< at >5z.com with "unsubscribe gob-list" in the subject

Re: New Debian maintainer, patches and licensing

Jiri Lebl — 2010-06-01T07:37:07

Thanks for the patches, applied everything to git, so whenever next
release happens they'll be in.

Anyway, the README is wrong, I forgot to update it when I updated
COPYING.generated-code.  The correct license for all generated code is
public domain.   Fixed that in git too.

Jiri

On Tue, Jun 1, 2010 at 12:58 AM, Stephen Kitt <steve< at >sk2.org> wrote:

New Debian maintainer, patches and licensing

Stephen Kitt — 2010-06-01T05:58:49

Hi,

I'm adopting gob2 from Mark Brown in Debian. I'm currently preparing an
upload of version 2.0.17, but I have a few questions first.
* The manpage doesn't escape minus signs; does the attached patch seem OK to
  you?
* The m4 macro GOB2_CHECK only issues a warning if gob2 is not found;
  shouldn't it produce an error? See
  https://bugs.launchpad.net/ubuntu/+source/gob2/+bug/317801 for the source
  of the attached patch.
* The README mentions that generated code is under LGPL, yet
  COPYING.generated-code specifies that it is in the public domain. Which is
  correct?

I'm also attaching a patch which fixes some typos.

Best regards,

Stephen
--
to unsubscribe:
send mail to minimalist< at >5z.com with "unsubscribe gob-list" in the subject

Re: Gob2 sigfault with dispose

Dan Saul — 2010-04-01T20:49:15

--
to unsubscribe:
send mail to minimalist< at >5z.com with "unsubscribe gob-list" in the subject

Re: Gob2 sigfault with dispose

Jiri Lebl — 2010-03-30T03:40:54

There is a bug with the new dispose/finalize/construct methods from
the last version.  I have it fixed in git now and I'll release a new
version this week.

Jiri

On Thu, Mar 25, 2010 at 7:28 PM, Dan Saul <daniel.saul< at >gmail.com> wrote:

Re: Gob2 sigfault with dispose

Jiri Lebl — 2010-03-26T03:11:48

I will investigate as soon as I get back home next week

Jiri

On Thu, Mar 25, 2010 at 7:28 PM, Dan Saul <daniel.saul< at >gmail.com> wrote:

Gob2 sigfault with dispose

Dan Saul — 2010-03-26T01:28:07

--
to unsubscribe:
send mail to minimalist< at >5z.com with "unsubscribe gob-list" in the subject

Re: Patch for the bug on remove_sep().

Jiri Lebl — 2009-12-23T20:03:33

Sorry for the late reply.

I have applied the patch in git.  I'll make a new release soon, thanks.

Jiri

On Mon, Nov 23, 2009 at 10:11 PM, Ding Yi Chen <dchen< at >redhat.com> wrote:

Patch for the bug on remove_sep().

Ding Yi Chen — 2009-11-24T04:11:43

Hi,

I am using gob2 to develop various projects such as WritRecogn, ibus-chewing, and MakerDialog.
Recently I found that those project are failed to compile. After investigation, I think the problem 
might reside in remove_sep(), src/util.c.

In remove_sep(), the code that remove ':' is:

while((p = strchr(s, ':')))
strcpy(p, p+1);

However, according to the man page of strcpy:
   "The strings may not overlap, and the destination string dest must  be
       large enough to receive the copy."

In fact, this code block produces unexpected results in some systems.
For example:
  Ma:Class is converted to Ma:Clsss
  Ma:Placement is converted to Ma:Plccemnnt

The patch that address this issue is attached.

Regards,

gmane.comp.gnome.devtools.gob.general

[PATCH] Don't output #line 0 directives for prototypes.

Re: Call order of parrent methods in __dispose() and __finalize()

Re: Building on ubuntu 10.04 LTS fails

Call order of parrent methods in __dispose() and __finalize()

Building on ubuntu 10.04 LTS fails

ANNOUNCE: gob2 2.0.18 the "bit torture" release

Re: Need for %ctop{ %}

Need for %ctop{ %}

Re: Build gob2 on Cygwin

Build gob2 on Cygwin

Re: New Debian maintainer, patches and licensing

New Debian maintainer, patches and licensing

Re: Gob2 sigfault with dispose

Re: Gob2 sigfault with dispose

Re: Gob2 sigfault with dispose

Gob2 sigfault with dispose

Re: Patch for the bug on remove_sep().

Patch for the bug on remove_sep().

Re: Call order of parrent methods in dispose() and finalize()

Call order of parrent methods in dispose() and finalize()