Re: [PATCH v3 16/18] ui: Split hmp_mouse_set() and move the HMP part to ui/

[Markus Armbruster]

Philippe Mathieu-Daudé <philmd@linaro.org> writes:

> On 20/12/22 10:06, Markus Armbruster wrote:
>> Signed-off-by: Markus Armbruster <armbru@redhat.com>
>> ---
>>   include/monitor/hmp.h | 1 +
>>   include/ui/console.h  | 2 +-
>>   monitor/misc.c        | 1 -
>>   ui/input.c            | 5 +----
>>   ui/ui-hmp-cmds.c      | 8 ++++++++
>>   5 files changed, 11 insertions(+), 6 deletions(-)
>
>
>> diff --git a/include/ui/console.h b/include/ui/console.h
>> index e400ee9fa7..e7f375312c 100644
>> --- a/include/ui/console.h
>> +++ b/include/ui/console.h
>> @@ -65,7 +65,7 @@ void qemu_remove_led_event_handler(QEMUPutLEDEntry *entry);
>>   
>>   void kbd_put_ledstate(int ledstate);
>>   
>> -void hmp_mouse_set(Monitor *mon, const QDict *qdict);
>> +void qemu_mouse_set(int index, Error **err);
>
> Can we return a boolean to indicate eventual failure?

The only caller isn't interested.  But we could do it anyway to conform
to the guidance in qapi/error.h.

>> @@ -594,10 +592,9 @@ MouseInfoList *qmp_query_mice(Error **errp)
>>       return mice_list;
>>   }
>>   
>> -void hmp_mouse_set(Monitor *mon, const QDict *qdict)
>> +void qemu_mouse_set(int index, Error **err)
>>   {
>>       QemuInputHandlerState *s;
>> -    int index = qdict_get_int(qdict, "index");
>>       int found = 0;
>>   
>>       QTAILQ_FOREACH(s, &handlers, node) {
>
> However no failure is reported, errp isn't used... Do we really
> want to pass it along?

Daniel spotted this, too.  See my reply there.

Thanks!