def test_boost(self):
cont_id = ContiguousID()
self.assertEquals(cont_id['a'], 0)
self.assertEquals(cont_id['b'], 1)
self.assertEquals(cont_id[0], 2)
cont_id.boost(10)
self.assertEquals(cont_id['a'], 10)
self.assertEquals(cont_id['b'], 11)
self.assertEquals(cont_id[0], 12)
def test_all(self):
cont_id = ContiguousID()
self.assertEquals(cont_id['a'], 0)
self.assertEquals(cont_id['b'], 1)
self.assertEquals(cont_id[0], 2)
self.assertEquals(cont_id['a'], 0)
self.assertEquals(cont_id[int], 3)
new_keys = range(1, 200)
random.shuffle(new_keys)
iter_vals = iter(range(4, 204))
for i in new_keys:
self.assertEquals(cont_id[i], next(iter_vals))
for i, d in enumerate(sorted(cont_id.itervalues())):
self.assertEqual(i, d)
self.assertEquals(203, len(cont_id))
def create_annotations(annotations_file):
'''
Creates annotations from a LibraryThing annotations file.
Arguments
---------
annotations_file : file
Annotations file to parse
'''
user_ids = ContiguousID()
item_ids = ContiguousID()
tag_ids = ContiguousID()
annotation_list = []
with open(annotations_file) as annotations:
for line in annotations:
spl = line.split()
#trace is: user, item, rating, tag
user = spl[0]
item = spl[1]
tag = ' '.join(spl[3:]) #deals with multiple word tags
uid = user_ids[user]
iid = item_ids[item]
tid = tag_ids[tag]
annotation = {}
annotation['user'] = uid
annotation['item'] = iid
annotation['tag'] = tid
annotation_list.append(annotation)
return annotation_list, user_ids, item_ids, tag_ids
def __reset(self):
'''Resets the parser ids to new ones. Useful for reusing
the same parser object'''
self.tag_ids = ContiguousID()
self.item_ids = ContiguousID()
self.user_ids = ContiguousID()
