Re: [gnome-db] More fun with apps
- From: will lashell net
- To: Rodrigo Moya <rodrigo gnome-db org>
- Cc: gnome-db-list gnome org
- Subject: Re: [gnome-db] More fun with apps
- Date: Wed, 5 Sep 2001 09:04:21 -0700
<SNIP>
> >
> this is not needed, since the call to gda_command_execute below will
> create the GdaRecordset
Hrmm, your example code and documentation should be changed to reflect
this then.
>
> > cmd = gda_command_new ();
> > gda_command_set_connection (cmd, conn);
> > gda_command_set_text (cmd, "SELECT * FROM message where recipient = 'wlashell' order by date DESC");
> > rs = gda_command_execute (cmd, &reccount, blah);
> >
> > gtk_clist_clear( GTK_CLIST( msg_list ) );
> > printf( "query successful\n" );
> > while( ! gda_recordset_eof( GDA_RECORDSET(rs) ) )
> > {
> > /* foo[0] = row_data[4];
> > foo[1] = row_data[2];
> > foo[2] = row_data[5];
> > gtk_clist_append( GTK_CLIST( msg_list ), foo); */
> > gda_recordset_move_next( rs );
> > }
> > }
> > }
> >
> instead of this, why don't you use the GnomeDbGrid widget? It uses also
> a GtkCList, and does everything for you. You just have to associate a
> GdaRecordset with it:
>
> GtkWidget *grid;
>
> grid = gnome_db_grid_new (rs);
Well, in reply to another email, you said that if we use a grid, you basically
have to free the resultset to do anything. This is broken in my opinion because
i very likely will want to do something different to the resultset and shouldn't
have to query the dbms again to get a new one
> >
> > which compiles just fine. but.. we get a segfault when it tries to
> > execute.
> > I ran a backtrace and this is what I ended up with:
> >
> > 0x407b6dd3 in strlen () from /lib/i686/libc.so.6
> > (gdb) bt
> > #0 0x407b6dd3 in strlen () from /lib/i686/libc.so.6
> > #1 0x40460251 in GDA_Connection_open () from /usr/lib/libgda-client.so.0
> > #2 0x40466174 in gda_connection_open () from /usr/lib/libgda-client.so.0
> >
> ok, the problem seems here, in the CORBA code. Let me see.... yes, in
> this case, it was passing NULL values to the ORBit-generated functions,
> and there, a call to strlen with a NULL pointer made it crash.
>
> It's now fixed in CVS (gnome-1-x-branch CVS branch)
Cool! What's the word on the release you've been planning?
Will
--
-----------------------------------------------
"Doh.. STUPID STUPID STUPID"
--an excerpt taken from every developers cvs
commit message at some point
-----------------------------------------------
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]