Hi Jeevaka,

On 03/30/2011 08:33 AM, Jeevaka.Badrappan(a)elektrobit.com wrote:
> Hi Denis,
> 
> Denis Kenzior wrote:
>> Hi Jeevaka,
>>
>>> void emit_en_list_changed(struct ofono_voicecall *vc)  static void
>>> set_new_ecc(struct ofono_voicecall *vc)  {
>>>  	int i = 0;
>>> +	GSList *l;
>>>
>>>  	g_slist_foreach(vc->en_list, (GFunc) g_free, NULL); 
>>> g_slist_free(vc->en_list); @@ -2079,6 +2081,12 @@ static void
>>>  	set_new_ecc(struct ofono_voicecall *vc) vc->en_list =
>>>  	vc->new_en_list; vc->new_en_list = NULL;
>>>
>>> +	for (l = vc->sim_en_list; l; l = l->next) {
>>> +		if (!g_slist_find_custom(vc->en_list, l->data,
> number_compare))
>>> +			vc->en_list = g_slist_prepend(vc->en_list,
>>> +
>> g_strdup(l->data));
>>> +	}
>>> +
>>>  	while (default_en_list[i]) {
>>>  		GSList *l;
>>>
>>> @@ -2120,10 +2128,11 @@ static void ecc_g2_read_cb(int ok, int
>>> total_length, int record,  		data += 3; 
>>>
>>>  		if (en[0] != '\0')
>>> -			vc->new_en_list =
>> g_slist_prepend(vc->new_en_list,
>>> +			vc->sim_en_list =
>> g_slist_prepend(vc->sim_en_list,
>>>
>> g_strdup(en));
>>
>> To preserve the old behavior, can we also check that the
>> number is not on the default_en_list?
> 
> SIM ecc number check with default_en_list is already taken care inside
> set_new_ecc function. 
> 

Thinking some more about it, yes you're absolutely right that it works
out in the end.

It might still be a good idea to use a Hashtable to fake a Set (e.g.
uniqueness) behavior of ECCs.  That would make the driver
implementations a bit simpler as well.

Regards,
-Denis