« Return to Thread: libpurple, the nullclient and AIM

Re: libpurple, the nullclient and AIM

by Jeremy Trammell - Setaris :: Rate this Message:

Reply to Author | View in Thread

glib_input_add() looks identical to pidgin_input_add() to me.  I'm pretty sure they compile to the same code.


Jeremy Trammell
Chief Technology Officer
  509 East 87th St. #1W
New York, NY 10128
Setaris Corporation cell +1.518.331.5180
www.setaris.com fax +1.646.786.4370
 
INTELLIGENT SOFTWARE





From: Daniel Atallah daniel.atallah@...
To: Jeremy Trammell - Setaris jeremy.trammell@...
CC: Daniel Atallah datallah@..., support@... support@...
Sent: 07/07/2009 1:28:35 AM -0400
Subject: libpurple, the nullclient and AIM




On Jul 7, 2009, at 0:24, Jeremy Trammell - Setaris jeremy.trammell@... wrote:

I've done a variety of things towards that end, but to keep things simple I am now focusing on simply using the example nullclient code, as provided, which does pass a set of glib io handlers in the PurpleEventLoopUiOps structure.

static PurpleEventLoopUiOps glib_eventloops =
{
    g_timeout_add,
    g_source_remove,
    glib_input_add,
    g_source_remove,
    NULL,
#if GLIB_CHECK_VERSION(2,14,0)
    g_timeout_add_seconds,
#else
    NULL,
#endif

    /* padding */
    NULL,
    NULL,
    NULL
};

That isn't going to work like that - g_input_add on glib >= 2.8 runs into the problem you're seeing. You need to use the version in libpurple (which is based on an older glib version).

-D

_______________________________________________
Support@... mailing list
Want to unsubscribe?  Use this link:
http://pidgin.im/cgi-bin/mailman/listinfo/support

 « Return to Thread: libpurple, the nullclient and AIM