def get_clean_urls(table_name):
raw_data = list(mongo_driver.get_all(table_name))
urls = list(filter(lambda item: 'url' in item, raw_data))
def clean_link(data):
link = data['url'].lower().replace('http://', '').replace('https://', '').replace(
'www.', '').replace((' '), '')
if link.endswith('/'):
return link[:-1], data
else:
return link, data
return dict(list(map(lambda item: clean_link(item), urls)))
def get_clean_urls(table_name):
raw_data = list(mongo_driver.get_all(table_name))
urls = list(filter(lambda item: 'url' in item, raw_data))
def clean_link(data):
link = data['url'].lower().replace('http://',
'').replace('https://',
'').replace('www.',
'').replace(
(' '),
'')
if link.endswith('/'):
return link[:-1], data
else:
return link, data
return dict(list(map(lambda item: clean_link(item), urls)))
def flags_articles_gen():
for i, _ in enumerate(mongo_driver.get_all('articles')):
yield _
def corpus_gen():
for i, _ in enumerate(mongo_driver.get_all('articles_cleaned')):
if _['article']:
yield _
def corpus_gen():
for i, _ in enumerate(mongo_driver.get_all('articles_cleaned')):
if _['article']:
yield _
def threadpool():
pool = Pool(30)
x = pool.imap_unordered(go, batch)
while True:
try:
x.next(timeout=10)
except multiprocessing.context.TimeoutError:
print('timeout!')
except AttributeError as e:
print(e)
except StopIteration:
print('batch finished.')
pool.close()
break
except EOFError:
pass
if __name__ == '__main__':
news_sources = mongo_driver.get_all('all_sources')
while True:
try:
batch = itertools.islice(news_sources, 90)
threadpool()
except StopIteration:
print('finished.')
exit()
def flags_articles_gen():
for i, _ in enumerate(mongo_driver.get_all('articles')):
yield _
def corpus_gen():
for i, _ in enumerate(mongo_driver.get_all('articles_cleaned')):
if _['article']: #and _['flag'] != 'satire':
yield _