Il 07/03/2014 03:56, Hu Tao ha scritto:
- obj = object_new(type);
- if (qemu_opt_foreach(opts, object_set_property, obj, 1) < 0) {
- object_unref(obj);
- return -1;
+ qdict_del(pdict, "qom-type");
+ visit_type_str(opts_get_visitor(ov), &type, "qom-type", &err);
+ if (err) {
+ goto out;
}
Can be moved up right before creating qdict.
- if (!object_dynamic_cast(obj, TYPE_USER_CREATABLE)) {
- error_setg(&local_err, "object '%s' isn't supported by -object",
- id);
+ qdict_del(pdict, "id");
+ visit_type_str(opts_get_visitor(ov), &id, "id", &err);
+ if (err) {
goto out;
}
Can be moved up right before creating qdict.
In both cases I prefer to keep the qdict_del and visit_type_str together.
- user_creatable_complete(obj, &local_err);
- if (local_err) {
+ object_add(type, id, pdict, opts_get_visitor(ov), &err);
I think it's better to move object_add() from qmp.c to qom/object.c.
No, I don't think so. qom/object.c is not using QDict. It is common
for "human user interface" files (hmp.c, ui/gtk.c, in this case vl.c) to
use qmp.c, the structure should be
hmp.c, ui/gtk.c, vl.c
|
qmp.c
|
qom/, cpus.c, etc.
We could move parts of qmp.c to qom/qmp.c, that would be fine.
Paolo
+ if (err) {
goto out;
}
-
- object_property_add_child(container_get(object_get_root(), "/objects"),
- id, obj, &local_err);
+ visit_end_struct(opts_get_visitor(ov), &err);
+ if (err) {
+ qmp_object_del(id, NULL);
+ }
out:
- object_unref(obj);
- if (local_err) {
- qerror_report_err(local_err);
- error_free(local_err);
- return -1;
+ opts_visitor_cleanup(ov);
+
+ QDECREF(pdict);
+ g_free(id);
+ g_free(type);
+ g_free(dummy);
+ if (err) {
+ qerror_report_err(err);
}
return 0;
}
--
1.8.5.3
