def test_html_encoded_chars(self):
res = cleanup_title("OBAMA's TEAM WELCOMES CASTRO & "
"CUBA TO AMERICA & TO RECeive CRITICISM: "
"'WOULDN'T DISAGREE'...")
self.assertEqual(
res, "Obama's Team Welcomes Castro "
"& Cuba To America & To Receive "
"Criticism: 'Wouldn't Disagree'...")
def test_acronym_check(self):
text = "'What's this another new US " \
"Usa U.S. U.s.a. title it's a miracle...'"
res = cleanup_title(text)
self.assertEqual(
res, "What's This Another New "
"US USA U.S. U.S.A. Title It's A "
"Miracle...")
def validate_title(self, value):
if not only_roman_chars(value):
raise ValidationError("Can only have roman characters")
# Clean up the title
value = cleanup_title(value)
# Check if title is in our excluded list or close to it
for excluded_article in settings.DEFAULT_EXCLUDE_ARTICLES:
# If the title has something similar to an article
# we don't want to include, remove it
# Not a list comprehension to ease readability. Not a
# huge issue since this is normally called in a task.
if SequenceMatcher(a=excluded_article.lower(),
b=value.lower()).ratio() > .70:
raise ValidationError("Contains content that is "
"not allowed")
# Check if title already exists
query = 'MATCH (news:NewsArticle {title: "%s"}) ' \
'RETURN news' % value
res, _ = db.cypher_query(query)
if res.one is not None:
raise ValidationError("This field must be unique")
return value
def test_lowercase(self):
text = "Friends of Israel - The New Yorker"
res = cleanup_title(text)
self.assertEqual(res, "Friends Of Israel - The New Yorker")
def test_quotes(self):
text = "\"Yet another title! What is this!\""
res = cleanup_title(text)
self.assertEqual(res, "Yet Another Title! What Is This!")