I have this extremely efficient object oriented hierarchical
construct (working parallel) under GAE:
class *my_object_Root*(BaseModel):
*total_counter *= db.IntegerProperty()
class *my_object*(db.Model):
*name *= db.StringProperty()
#That's it ! Super fast !
@classmethod
def *get_profile*(self, name):
if not name:
return None
root = my_object_Root.get_profile()
#cache
self._cache = get_cache("memcached://")
skey = name
key_cache = 'my_object' + skey + str(root.key())
profile = self._cache.get(key_cache)
#cache
if not profile:
key = db.Key.from_path('my_object',
skey,
parent=root.key())
profile = db.get(key)
# not found - create
if not profile:
root = my_object_Root.get_profile_del_cache()
profile = my_object(key_name=skey,
parent=root.key())
profile.put()
return profile
#cache
self._cache.set(key_cache, profile, timeout=CACHE_TIMEOUT)
#cache
return profile
*Usage:*
my_object_name = "Test Name"
my_name = models.my_object.get_profile(my_object_name)
name = my_name.name
*total_counter* is in the root: *my_object_Root* of the object
*my_object. *We can have many children of *my_object *under
*my_object_Root *and* *additional code is updating *total_counter*
when I am adding another *my_object. *That way one very difficult
to implement and slow operation - counting is done very efficiently
on the fly and it works very quick.
This and other construct are impossible to achieve the same efficiency
under SQL so the only way I see to achieve this under web2py is web2py
abstraction layer to have a way transparently to use GAE.
Are there currently a way to achieve this?
Thanks in advance,
