def test_since_id():
for tweet in twarc.search('obama'):
id = tweet['id_str']
break
assert id
time.sleep(5)
for tweet in twarc.search('obama', since_id=id):
assert tweet['id_str'] > id
def test_since_id():
for tweet in twarc.search('obama'):
id = tweet['id_str']
break
assert id
time.sleep(5)
for tweet in twarc.search('obama', since_id=id):
assert tweet['id_str'] > id
def test_max_id():
for tweet in twarc.search('obama'):
id = tweet['id_str']
break
assert id
time.sleep(5)
count = 0
for tweet in twarc.search('obama', max_id=id):
count += 1
assert tweet['id_str'] <= id
if count > 100:
break
def test_max_id():
for tweet in twarc.search('obama'):
id = tweet['id_str']
break
assert id
time.sleep(5)
count = 0
for tweet in twarc.search('obama', max_id=id):
count += 1
assert tweet['id_str'] <= id
if count > 100:
break
def test_max_and_since_ids():
max_id = since_id = None
count = 0
for tweet in twarc.search('obama'):
count += 1
if not max_id:
max_id = tweet['id_str']
since_id = tweet['id_str']
if count > 500:
break
count = 0
for tweet in twarc.search('obama', max_id=max_id, since_id=since_id):
count += 1
assert tweet['id_str'] <= max_id
assert tweet['id_str'] > since_id
def test_max_and_since_ids():
max_id = since_id = None
count = 0
for tweet in twarc.search('obama'):
count += 1
if not max_id:
max_id = tweet['id_str']
since_id = tweet['id_str']
if count > 500:
break
count = 0
for tweet in twarc.search('obama', max_id=max_id, since_id=since_id):
count += 1
assert tweet['id_str'] <= max_id
assert tweet['id_str'] > since_id
def __init__(self, text, images_dir, dbname):
""" Initialize Epheme & then kick off the twarc search. """
# store tweet JSON in a db with same name as search text
if dbname is None:
dbname = text.replace(' ', '_')
self.db = MongoClient()[dbname]
self.images_dir = images_dir
self.hashtag = text
print 'Beginning to search Twitter API for', self.hashtag
# note: first run through is a special case, most_recent_id()
# returns None & twarc will return a bunch of recent tweets
# then start stepping backwards in time. It's limited by the
# Search API in that it cannot go back that far, a couple of
# weeks is the limit I think.
for tweet in twarc.search(self.hashtag,
since_id=self.most_recent_id()):
try:
self.insert_tweet_into_db(tweet)
# tweet already in db
except errors.DuplicateKeyError:
print 'Skipping duplicate Tweet', tweet['id']
continue
# done getting all the metadata? now grab the referenced images
self.download_all_imgs()
print 'Done searching Twitter for', self.hashtag, '& downloading images.'
def test_paging():
# pages are 100 tweets big so if we can get 500 paging is working
count = 0
for tweet in twarc.search('obama'):
count += 1
if count == 500:
break
assert count == 500
def test_search():
count = 0
for tweet in twarc.search('obama'):
assert tweet['id_str']
count += 1
if count == 10:
break
assert count == 10
def test_paging():
# pages are 100 tweets big so if we can get 500 paging is working
count = 0
for tweet in twarc.search('obama'):
count += 1
if count == 500:
break
assert count == 500
def test_search():
count = 0
for tweet in twarc.search('obama'):
assert tweet['id_str']
count += 1
if count == 10:
break
assert count == 10
#!/usr/bin/env python
import twarc
for tweet in twarc.search("hmd2014"):
print tweet["created_at"] + " @" + tweet["user"]["screen_name"] + ": " + tweet["text"]
