view src/protocols/zephyr/ZNewLocU.c @ 11044:8bf376579177

[gaim-migrate @ 12958] Patch #1229320 from Bleeter "I couldn't get anyone to confirm that doing this is correct, however to me it would seem that I'm leaking on a couple of the privacy_checks inside the yahoo prpl. Here's a fix. This also includes privacy wrapper for the conference decline function, used in some booters." (00:37:39) Bleeter: ... we're returning out of the function, and those things are freed further down if the function isn't returned (00:38:18) Bleeter: so it makes sense to free them before the return. I'm just unsure because they'reprevious patches which were applied without these checks being picked up That makes sense to me. It looks good. committer: Tailor Script <[e-mail]>

/* This file is part of the Project Athena Zephyr Notification System.
 * It contains source for the ZNewLocateUser function.
 *
 *	Created by:	Robert French
 *
 *	$Source$
 *	$Author: chipx86 $
 *
 *	Copyright (c) 1987,1988,1991 by the Massachusetts Institute of Technology.
 *	For copying and distribution information, see the file
 *	"mit-copyright.h". 
 */
/* $Header$ */

#include "internal.h"

#ifndef lint
static const char rcsid_ZNewLocateUser_c[] =
    "$Id: ZNewLocU.c 9554 2004-04-24 09:02:28Z chipx86 $";
#endif

Code_t ZLocateUser(user, nlocs, auth)
    char *user;
    int *nlocs;
    Z_AuthProc auth;
{
    Code_t retval;
    ZNotice_t notice;
    ZAsyncLocateData_t zald;

    (void) ZFlushLocations();	/* ZFlushLocations never fails (the library
				   is allowed to know this). */

    if ((retval = ZRequestLocations(user, &zald, UNACKED, auth)) != ZERR_NONE)
	return(retval);

    retval = Z_WaitForNotice (&notice, ZCompareALDPred, &zald, SRV_TIMEOUT);
    if (retval == ZERR_NONOTICE)
	return ETIMEDOUT;
    if (retval != ZERR_NONE)
	return retval;

    if ((retval = ZParseLocations(&notice, &zald, nlocs, NULL)) != ZERR_NONE) {
	ZFreeNotice(&notice);
	return(retval);
    }

    ZFreeNotice(&notice);
    ZFreeALD(&zald);
    return(ZERR_NONE);
}