Added stream to CLI

* Save stream to file
* Parse file and get the most important hashtags

bitter/VERSION

@@ -1 +1 @@
-0.6.0
+0.6.1

bitter/cli.py

@@ -6,6 +6,7 @@ import time
 import sqlalchemy.types
 import threading
 import sqlite3
+from tqdm import tqdm
 
 from sqlalchemy import exists
 
@@ -333,14 +334,49 @@ def stream(ctx):
     pass
 
 @stream.command('get')
+@click.option('-l', '--locations', default=None)
+@click.option('-t', '--track', default=None)
+@click.option('-f', '--file', help='File to store the stream of tweets')
 @click.pass_context 
-def get_stream(ctx):
+def get_stream(ctx, locations, track, file):
     wq = crawlers.StreamQueue.from_credentials(bconf.CREDENTIALS, 1)
 
-    iterator = wq.statuses.sample()
+    query_args = {}
+    if locations:
+        query_args['locations'] = locations
+    if track:
+        query_args['track'] = track
+    if not query_args:
+        iterator = wq.statuses.sample()
+    else:
+        iterator = wq.statuses.filter(**query_args)#"-4.25,40.16,-3.40,40.75")
+
+    if not file:
+        file = sys.stdout
+    else:
+        file = open(file, 'a')
+
+    for tweet in tqdm(iterator):
+        print(json.dumps(tweet), file=file)
+    if file != sys.stdout:
+        file.close()
 
-    for tweet in iterator:
-        print(tweet)
+@stream.command('read')
+@click.option('-f', '--file', help='File to read the stream of tweets from')
+@click.pass_context 
+def read_stream(ctx, file):
+    for tweet in utils.read_file(file, tail=True):
+        print('{timestamp_ms}- @{screen_name}: {text}'.format(timestamp_ms=tweet['timestamp_ms'], screen_name=tweet['user']['screen_name'], text=tweet['text']))
+
+@stream.command('tags')
+@click.option('-f', '--file', help='File to read the stream of tweets from')
+@click.argument('limit', required=False, default=None, type=int)
+@click.pass_context 
+def tags_stream(ctx, file, limit):
+    c = utils.get_hashtags(utils.read_file(file))
+    for count, tag in c.most_common(limit):
+        print('{} - {}'.format(count, tag))
+    
 
 if __name__ == '__main__':
     main()

bitter/utils.py

@@ -12,6 +12,7 @@ from multiprocessing.pool import ThreadPool
 from itertools import islice
 from contextlib import contextmanager
 from itertools import zip_longest
+from collections import Counter
 
 from twitter import TwitterHTTPError
 
@@ -86,6 +87,26 @@ def add_credentials(credfile=None, **creds):
             f.write('\n')
 
 
+def get_hashtags(iter_tweets, best=None):
+    c = Counter()
+    for tweet in iter_tweets:
+        c.update(tag['text'] for tag in tweet.get('entities', {}).get('hashtags', {}))
+    return c
+
+def read_file(filename, tail=False):
+    with open(filename) as f:
+        while True:
+            line = f.readline()
+            if line not in (None, '', '\n'):
+                tweet = json.loads(line.strip())
+                yield tweet
+            else:
+                if tail:
+                    time.sleep(1)
+                else:
+                    return line
+    
+
 def get_users(wq, ulist, by_name=False, queue=None, max_users=100):
     t = 'name' if by_name else 'uid'
     logger.debug('Getting users by {}: {}'.format(t, ulist))

requirements.txt

@@ -1,3 +1,4 @@
 sqlalchemy
 twitter
 click
+tqdm