Adds to Melange a tags framework based on taggable-mixin.
The taggable-mixin allowed only tag per Datastore model. This is extended
framework allows any arbitrary number of tags per Datastore model. Also,
now one can define different models for different Tag types which are all
inherited from the base Tag model provided by taggable-mixin.
The GHOPTask model makes use of 2 tags per model, one for difficulty and the
other for task_type, both using the tags framework.
Reviewed by: Paweł Sołyga
"""
Common checksum routines (used in multiple localflavor/ cases, for example).
"""
__all__ = ['luhn',]
LUHN_ODD_LOOKUP = (0, 2, 4, 6, 8, 1, 3, 5, 7, 9) # sum_of_digits(index * 2)
def luhn(candidate):
"""
Checks a candidate number for validity according to the Luhn
algorithm (used in validation of, for example, credit cards).
Both numeric and string candidates are accepted.
"""
if not isinstance(candidate, basestring):
candidate = str(candidate)
try:
evens = sum([int(c) for c in candidate[-1::-2]])
odds = sum([LUHN_ODD_LOOKUP[int(c)] for c in candidate[-2::-2]])
return ((evens + odds) % 10 == 0)
except ValueError: # Raised if an int conversion fails
return False