Serrano comes with a base resource that provide structure for the exposed resources. This is described here to make it easier to define other Avocado-backed resources for the most common use cases.


  • Every request is checked for authentication if required. This also includes token-based access if enabled.
  • If cross-origin resource sharing is enabled, every response gets two additional headers Access-Control-Allow-Origin and Access-Control-Allow-Methods.



Takes the request object and returns a cleaned dict of GET parameters. This uses a restlib2 Parametizer class to clean the parameters.

resource = BaseResource()
params = resource.get_params(request)

To customize the clean of parameters and for setting defaults, create a subclass of restlib2.params.Parametizer and set it to the parametizer on the resource class. For example:

from restlib2.params import Parametizer, param_cleaners
from serrano.resources.base import BaseResource

class MyParametizer(Parametizer):
    query = ''
    page = 1
    per_page = 10

    def clean_page(self, value):
        return param_cleaners.clean_int(value)

    def clean_per_page(self, value):
        return param_cleaners.clean_int(value)    

class MyResource(BaseResource):
    parametizer = MyParametizer


Takes the request object and returns a DataContext object that is currently applicable to the request. For convenience, the method also takes an attrs keyword argument that would be the context nodes themselves. A series of checks are performed:

  • If attrs are not supplied, derive attrs from request object using the context key
    • For instance, if a POST request was being handled and the POST data contained a key context, that would be processed
  • If attrs is a dict, validate and return
  • If attrs is an integer, lookup the context object for the requesting user and return
  • Attempt to current the session context for the user if one exists
  • Fallback to an empty context
resource = BaseResource()
context = resource.get_context(request)
# Supply explicit context data
context = resource.get_context(request, attrs={'field': 1, 'operator': 'exact', 'value': 5})


This method is identical to the above get_context regarding the processing steps. The difference, of course, is that is assumes valid DataView data and it returns a DataView instance.

resource = BaseResource()
view = resource.get_view(request)
# Supply explicit view data
view = resource.get_view(request, attrs={'columns': [3, 2, 1]})