Re: request for help



Ali,

It's seem's that the bug (the main one) is pthread related.
If i understand corrcetly, when I call popen I duplicate the whole process
with all threads and mutex and this leads to a deadlock (need to be
clarify).

Christophe

Le mar, 21 aoû 2001 12:06:21, Ali Akcaagac a écrit :
> On 2001.08.21 11:01:23 +0200 christophe barbé wrote:
> > > I don't understand why you are using the uname command to get
> > > the system name. After all uname() is a system call and it is
> > > Posix. Why not take the simpler approach and use that?  You also
> > > avoid the overhead of creating pipes, fork and exec and then
> > > reading from a file descriptor.  Lost of system calls replaced
> > > with just one.  Got to be better.
> > 
> > My goal was to allow custom system identification string.
> > I've not looked in the Mutt code but seeing various mails headers I
> > guess that it's customisable.
> 
> hi,
> 
> well the idea with that custom string to show other people what os you
> prefer is indeed a good idea. i have to share this and i would be glad
> you gonna implement this. but yes, i must agree to the statement of
> brian s. using the systemcall is the better solution than executing an
> external program and grep the string then.
> 
> a) the command may not be present on the system,
> b) doesnt support all the flags,
> c) command may be outdated and never got replaced
>    with a newer version for a bunch of years.
> 
> > > #include <sys/utsname.h>
> > > 
> > > char *
> > > get_system_name (void)
> > > {
> > >   struct utsname buf;
> > >   char *sysname;
> > > 
> > >   uname (&buf);
> > >   sysname = malloc (strlen (buf.sysname) + strlen (buf.release) + 2);
> > >   sprintf (sysname, "%s %s", buf.sysname, buf.release);
> > >   return sysname;
> > > }
> > > 
> > > obviously that needs error checks but its the right sequence of
> actions.
> > 
> > I've not think about this function. Perhaps I can let the user choose a
> > format string and provide substitution (%n : system name, %r release,
> ...).
> > This would certainly be better.
> 
> yeah this one is the right way to go.
> 
> > BTW I'm also interesting in knowing where my patch fail.
> 
> i am into this right now. i will report within the next couple of mins.
> but my first impression is the callbacks.. but i am not sure right now
> please hold for my next reply.
> 
> -- 
> Name....: Ali Akcaagac
> Status..: Student Of Computer & Economic Science
> E-Mail..: mailto:ali.akcaagac@stud.fh-wilhelmshaven.de
> WWW.....: http://www.fh-wilhelmshaven.de/~akcaagaa
> 
> _______________________________________________
> balsa-list mailing list
> balsa-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/balsa-list
> 
-- 
Christophe Barbé
Software Engineer - christophe.barbe@lineo.fr
Lineo France - Lineo High Availability Group
42-46, rue Médéric - 92110 Clichy - France
phone (33).1.41.40.02.12 - fax (33).1.41.40.02.01
http://www.lineo.com




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]