def note_iter(total_notes, notes_per_user, course_ids):
"""
Return an iterable of random notes data of length `total_notes`.
Arguments:
total_notes (int): total number of notes models to yield
notes_per_user (int): number of notes to attribute to any one user
course_ids (list): list of course_id strings to which notes will be
randomly attributed
Returns:
generator: An iterable of note models.
"""
DATA_DIRECTORY = os.path.join(os.path.dirname(__file__), "data/")
with open(os.path.join(DATA_DIRECTORY, 'basic_words.txt')) as f:
notes_text = [word for line in f for word in line.split()]
def weighted_get_words(weighted_num_words):
"""
Return random words of of a length of weighted probability.
`weighted_num_words` should look like [(word_count, weight), (word_count, weight) ...]
"""
return random.sample(
notes_text,
random.choice([
word_count for word_count, weight in weighted_num_words
for i in range(weight)
]))
get_new_user_id = lambda: uuid.uuid4().hex
user_id = get_new_user_id()
for note_count in range(total_notes):
if note_count % notes_per_user == 0:
user_id = get_new_user_id()
# Notice that quote and ranges are arbitrary
yield Note(user_id=user_id,
course_id=random.choice(course_ids),
usage_id=uuid.uuid4().hex,
quote='foo bar baz',
text=' '.join(
weighted_get_words([(10, 5), (25, 3), (100, 2)])),
ranges=json.dumps([{
"start": "/div[1]/p[1]",
"end": "/div[1]/p[1]",
"startOffset": 0,
"endOffset": 6
}]),
tags=json.dumps(
weighted_get_words([(1, 40), (2, 30), (5, 15), (10, 10),
(15, 5)])))
def test_create_invalid_note(self):
note = Note()
for empty_type in (None, '', []):
with self.assertRaises(ValidationError):
note.create(empty_type)
