General hash_join function

hash_join.py

def hash_join(left, right, join_key, attr_keys, outer=True):
    """
    Allows you to join the keys/values from a list of dictionaries as the 
    attributes of a given list of objects.
    
    Params::
    
    - left - list of objects to join to
    - right - list of dictionaries from which to get joined data
    - join_key - attribute of both object and dictionary with which we can match
    - attr_keys - the list of keys on the right dict you want to join
    - outer - if a matching key is not found in right, outer = True would keep
    the row in left and set all keys in attr_keys to None. outer = False would
    delete the row in left if no matching join_key is found in right
    
    """
    # convert list of dicts to a dict of dicts
    # 2.7 only - right = {r.get(join_key): r for r in right}
    right = dict((r.get(join_key), r) for r in right)

    if not outer:
        left[:] = filter(lambda x: right.get(getattr(x, join_key)), left)

    # attach the inner dict values to the attr on the object
    for l in left:
        if not hasattr(l, join_key):
            continue
        else:
            # check if the dict has a matching record
            right_vals = right.get(getattr(l, join_key), {})
            # attach attributes to objects
            for key in attr_keys:
                setattr(l, key, right_vals.get(key))

    return left

class thingy(object):
    pass
    
def build_things():
    things = []
    thing = thingy()
    setattr(thing, 'group_id', 1)
    things.append(thing)
    thing2 = thingy()
    setattr(thing2, 'group_id', 2)
    things.append(thing2)
    return things

stuff = [{'group_id': 1, 'group_name': 'Alex', 'group_size': 100}, {'group_name': 'Kevin', 'group_size': 200}]

whatsa = hash_join(build_things(), stuff, 'group_id', ['group_name', 'group_size'], outer=False)

for what in whatsa:
    print getattr(what, 'group_id', None), getattr(what, 'group_name', None), getattr(what, 'group_size', None)
    
bobs = hash_join(build_things(), stuff, 'group_id', ['group_size'], outer=True)

for bob in bobs:
    print getattr(bob, 'group_id', None), getattr(bob, 'group_name', None), getattr(bob, 'group_size', None)