I've done only two small projects with pyramid, but from the beginning i've
decided to go with traversal. because with traversal i don't have to decide
about every urls of my site(which url must represent which resource and
...).
So far my project structure is like this:
*Models:* i have a models folder which contains my models definitions(Duh!).
*Resources:* i have a resources folder which contains my root object. also
for each model i have a resource file which has two classes: one for
resource collection(i.e. *CustomersListResource*) and other one for single
resource(i.e* CustomerResource*).
With pb help i figured out how to write my view/resource to minimize db
queries. for example:
customer_rc.py:
...
class CustomersListResource(Resource):
def __init__(self, name, parent):
Resource.__init__(self, name, parent)
def __getitem__(self, key):
if key.is_digit():
# check db if customer id exists:
c_id = DBSession.query(models.Customer.id).filter(models.Customer
.id == int(key)).scalar()
if c_id:
return Customer(c_id, self, None)
...
raise KeyError(key)
def get_list(self):
# Query and return all customers.
customers = []
for c in DBSession.query(models.Customer):
customers.append(CustomerResource(c.id, self, c))
return customers
class CustomerResource(Resource):
def __init__(name, parent, model):
Resource.__init__(self, name, parent)
self.customer_model = model
def __getitem__(self, key):
if key == "orders":
return OrdersListResource("orders", self)
...
def load_model(self):
self.customer_model =
DBSession.query(models.Customer).get(self.__name__)
def get_orders(self, year, month):
...
def __json__(self, request):
d = {}
if self.customer_model is None:
d["id"] = self.__name__
else:
# Fill up dict with all customer data needed to passed to
client side.
return d
And this is my view example:
@view_config(context=resources.CustomersListResource, renderer="json")
def customers_view(context, request):
return {
"context": context
}
@view_config(context=resources.CustomersListResource, name="list",
renderer="json")
def customers_list_view(context, request):
return {
"context": context.get_list()
}
@view_config(context=resources.CustomerResource, renderer="json")
def customer_view(context, request):
return {
"context": context
}
@view_config(context=resources.CustomerResource, name="get",
renderer="json")
def customer_view(context, request):
context.load_model()
return {
"context": context
}
@view_config(context=resources.CustomerResource, name="orders",
renderer="json")
def customer_view(context, request):
year = request.params.get("year", None)
month = request.params.get("month", None)
return {
"context": context,
"orders": context.get_orders(year, month)
}
I would like to know your opinions about this approach.
Thanks.
On Friday, August 28, 2015 at 2:31:51 PM UTC+4:30, pb wrote:
>
> Hi Mehdi,
>
> traversing your resource tree does not necessarily mean running any
> queries. In
> your example, it wouldn't have to run any queries at all.
> Basically, it is there to bring you to the view and possibly gather some
> information
> on the way. For example (simplified and certainly not working):
>
> class TopContext(object):
> def __init__(self, request):
> self.request = request
> def __getitem__(self, entity_name):
> if "customers" == entity_name:
> return CustomerContext(self.request)
>
> class CustomersContext(object):
> def __init__(self, request):
> self.request = request
> def get_data(self):
> # run the select and collect the data ...
> def __getitem__(self, ent_id):
> if ent_id.isdigit():
> return OneCustomerContext(self.request)
>
> class OneCustomerContext(object):
> def __init__(self, request):
> self.request = request
> def get_data(self):
> # run the select and collect the data ...
>
> Now, when pyramid traverses the resource tree, it only uses constructors
> and __getitem__,
> so if there are no queries in there, they are not run. And you usually do
> not need them.
>
> If it was up to me, I would not define OrderContext. The traversal would
> get me to
> OneCustomerContext, and *order* would be the view name (parameter name of
> view_config).
>
> --
> Petr
>
>
>
> On Fri, Aug 28, 2015 at 9:54 AM, Mehdi <[email protected] <javascript:>>
> wrote:
>
>> Hi
>> Let's get straight to my question:
>> Consider this url: *http://mysite.com/customers/137/orders?year=2015&month=5
>> <http://mysite.com/customers/137/orders?year=2015&month=5>*
>> Now in pyramid traversal mechanism, how many queries will be run against
>> database?
>> I guess it would be three:
>> 1. */customers* : select * from customers
>> 2. */customers/137* : select * from customers
>> where id = 137
>> 3. */customers/137/**orders?year=2015&month=5: *select * from orders
>> where customer_id = 137 and year=:year and month=:month
>>
>> If so is there anyway to reduce the db queries down to one last query?
>> Thanks.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "pylons-discuss" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected] <javascript:>.
>> To post to this group, send email to [email protected]
>> <javascript:>.
>> Visit this group at http://groups.google.com/group/pylons-discuss.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/pylons-discuss.
For more options, visit https://groups.google.com/d/optout.
