def compare_documents(feed):
"""
Compare a list of documents by pair.
Pairs of duplicates are sorted by "retrieved date".
"""
duplicates = []
for pair in itertools.combinations(feed.articles, 2):
date1, date2 = pair[0].date, pair[1].date
if clear_string(pair[0].title) == clear_string(pair[1].title) and \
(date1 - date2) < timedelta(days = 1):
if pair[0].retrieved_date < pair[1].retrieved_date:
duplicates.append((pair[0], pair[1]))
else:
duplicates.append((pair[1], pair[0]))
return duplicates
def compare_documents(feed):
"""
Compare a list of documents by pair.
Pairs of duplicates are sorted by "retrieved date".
"""
duplicates = []
for pair in itertools.combinations(feed.articles, 2):
date1, date2 = pair[0].date, pair[1].date
if clear_string(pair[0].title) == clear_string(pair[1].title) and \
(date1 - date2) < timedelta(days = 1):
if pair[0].retrieved_date < pair[1].retrieved_date:
duplicates.append((pair[0], pair[1]))
else:
duplicates.append((pair[1], pair[0]))
return duplicates
def top_words(articles, n=10, size=5):
"""
Return the n most frequent words in a list.
"""
stop_words = load_stop_words()
words = Counter()
wordre = re.compile(r'\b\w{%s,}\b' % size, re.I)
for article in articles:
for word in [elem.lower() for elem in
wordre.findall(clear_string(article.content)) \
if elem.lower() not in stop_words]:
words[word] += 1
return words.most_common(n)
def top_words(articles, n=10, size=5):
"""
Return the n most frequent words in a list.
"""
stop_words = load_stop_words()
words = Counter()
wordre = re.compile(r'\b\w{%s,}\b' % size, re.I)
for article in articles:
for word in [elem.lower() for elem in
wordre.findall(clear_string(article.content)) \
if elem.lower() not in stop_words]:
words[word] += 1
return words.most_common(n)
def article(article_id=None):
"""
Presents the content of an article.
"""
article = ArticleController(g.user.id).get(id=article_id)
previous_article = article.previous_article()
if previous_article is None:
previous_article = article.source.articles[0]
next_article = article.next_article()
if next_article is None:
next_article = article.source.articles[-1]
return render_template('article.html',
head_titles=[clear_string(article.title)],
article=article,
previous_article=previous_article,
next_article=next_article)
