Folks, so I figured it out: turns out that there is what I believe a bug (or
a feature) in g_list_insert_sorted -- if the list is null, even then it goes
off to sort. This should not happen. If the list is NULL, it should just add
to the list without comparing.
Many thanks and best wishes,
Trotter
--- John Wendel <john.wendel@xxxxxxxxxxxxxxx> wrote:
> Globe Trotter wrote:
> > Hi,
> >
> > OK, I know this is not a list for this question, but I have posted on the
> GTK
> > list, but that is very low volume, and the answer is killing me. So, I
> > apologize for this question, but after struggling with this for over a day,
> I
> > wanted to ask you for your help regarding usage for g_list_insert_sorted.
> Can
> > someone please help. Here are the operative parts of my program:
> >
> > I have the following struct:
> >
> > typedef struct _message_info
> > {
> > char *from;
> > char *subject;
> > char *face;
> > long *ctime;
> > } MESSAGE_INFO;
> >
> > I have the following comparison function:
> >
> > static gint compare_ctime (gconstpointer a, gconstpointer b)
> > {
> > const MESSAGE_INFO *p1=a,*p2=b;
> >
> > return ((p1->ctime > p2->ctime) ? +1: (p1->ctime == p2->ctime) ? 0 : -1);
> > }
> >
> > and the following calling part:
> >
> > Glist *headers;
> > .....
> > MESSAGE_INFO *m;
> > m = g_new(MESSAGE_INFO,1);
> > m->from = safe_strdup (from);
> > m->subject = safe_strdup (subject);
> > m->face = safe_strdup (xface);
> > m->ctime = ctime;
> > headers = g_list_insert_sorted (headers, m,(GCompareFunc)compare_ctime);
> > ....
> >
> > I get segmentation fault with this, and was wondering how I could fix it. I
> > guess my question is how to use g_list_insert_sorted.
> >
> > I have Googled quite a bit on this, but have not been able to devise the
> fix.
> > This is a programming error, I think, based on my inadequate knowledge. Any
> > help would be very appreciated.
> >
> > Many thanks and best wishes,
> > trotter
>
> DISCLAIMER: I don't know what I'm writing about!
>
> "m->ctime = ctime;" looks suspect. I don't see the definition of
> "ctime" in your code fragment, but it looks like you're assigning a
> long to a pointer.
>
> Either in MESSAGE_INFO ctime should be declared "long ctime" or the
> assignment should be m->ctime = & ctime;
>
> Hope this isn't totally off base.
>
> Regards,
>
> John
>
>
>
> --
> fedora-list mailing list
> fedora-list@xxxxxxxxxx
> To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
