streaming.py 6.38 KB
Newer Older
1
# Tweepy
2
3
# Copyright 2009-2010 Joshua Roesslein
# See LICENSE for details.
4
5

import httplib
6
from socket import timeout
7
from threading import Thread
8
from time import sleep
9
import urllib
10

Josh Roesslein's avatar
Josh Roesslein committed
11
from tweepy.auth import BasicAuthHandler
12
from tweepy.models import Status
Josh Roesslein's avatar
Josh Roesslein committed
13
14
from tweepy.api import API
from tweepy.error import TweepError
15

16
17
from tweepy.utils import import_simplejson
json = import_simplejson()
18

19
20
STREAM_VERSION = 1

Josh Roesslein's avatar
Josh Roesslein committed
21

22
23
class StreamListener(object):

24
25
    def __init__(self, api=None):
        self.api = api or API()
26
27
28
29
30
31
32
33
34

    def on_data(self, data):
        """Called when raw data is received from connection.

        Override this method if you wish to manually handle
        the stream data. Return False to stop stream and close connection.
        """

        if 'in_reply_to_status_id' in data:
35
            status = Status.parse(self.api, json.loads(data))
36
37
38
39
40
41
42
43
44
45
            if self.on_status(status) is False:
                return False
        elif 'delete' in data:
            delete = json.loads(data)['delete']['status']
            if self.on_delete(delete['id'], delete['user_id']) is False:
                return False
        elif 'limit' in data:
            if self.on_limit(json.loads(data)['limit']['track']) is False:
                return False

Josh Roesslein's avatar
Josh Roesslein committed
46
47
48
49
50
51
52
    def on_status(self, status):
        """Called when a new status arrives"""
        return

    def on_delete(self, status_id, user_id):
        """Called when a delete notice arrives for a status"""
        return
53

Josh Roesslein's avatar
Josh Roesslein committed
54
55
56
    def on_limit(self, track):
        """Called when a limitation notice arrvies"""
        return
57

Josh Roesslein's avatar
Josh Roesslein committed
58
59
60
    def on_error(self, status_code):
        """Called when a non-200 status code is returned"""
        return False
61

Josh Roesslein's avatar
Josh Roesslein committed
62
63
64
    def on_timeout(self):
        """Called when stream connection times out"""
        return
65

66

67
68
class Stream(object):

Josh Roesslein's avatar
Josh Roesslein committed
69
70
    host = 'stream.twitter.com'

71
    def __init__(self, auth, listener, **options):
72
        self.auth = auth
Josh Roesslein's avatar
Josh Roesslein committed
73
        self.listener = listener
74
75
76
77
78
79
80
81
82
83
84
        self.running = False
        self.timeout = options.get("timeout") or 5.0
        self.retry_count = options.get("retry_count")
        self.retry_time = options.get("retry_time") or 10.0
        self.snooze_time = options.get("snooze_time") or 5.0
        self.buffer_size = options.get("buffer_size") or 1500
        if options.get("secure"):
            self.scheme = "https"
        else:
            self.scheme = "http"

Josh Roesslein's avatar
Josh Roesslein committed
85
        self.api = API()
86
        self.headers = headers or {}
87
        self.body = None
Josh Roesslein's avatar
Josh Roesslein committed
88
89
90

    def _run(self):
        # setup
91
        url = "%s://%s%s" % (self.scheme, self.host, self.url)
92
        self.auth.apply_auth(url, 'GET', self.headers, None)
Josh Roesslein's avatar
Josh Roesslein committed
93
94
95
96

        # enter loop
        error_counter = 0
        conn = None
97
        exception = None
Josh Roesslein's avatar
Josh Roesslein committed
98
99
100
101
102
103
104
105
        while self.running:
            if self.retry_count and error_counter > self.retry_count:
                # quit if error count greater than retry count
                break
            try:
                conn = httplib.HTTPConnection(self.host)
                conn.connect()
                conn.sock.settimeout(self.timeout)
106
                conn.request('POST', self.url, self.body, headers=self.headers)
Josh Roesslein's avatar
Josh Roesslein committed
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
                resp = conn.getresponse()
                if resp.status != 200:
                    if self.listener.on_error(resp.status) is False:
                        break
                    error_counter += 1
                    sleep(self.retry_time)
                else:
                    error_counter = 0
                    self._read_loop(resp)
            except timeout:
                if self.listener.on_timeout() == False:
                    break
                if self.running is False:
                    break
                conn.close()
                sleep(self.snooze_time)
123
            except Exception, exception:
Josh Roesslein's avatar
Josh Roesslein committed
124
125
126
127
128
129
130
131
                # any other exception is fatal, so kill loop
                break

        # cleanup
        self.running = False
        if conn:
            conn.close()

132
133
134
        if exception:
            raise exception

Josh Roesslein's avatar
Josh Roesslein committed
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
    def _read_loop(self, resp):
        data = ''
        while self.running:
            if resp.isclosed():
                break

            # read length
            length = ''
            while True:
                c = resp.read(1)
                if c == '\n':
                    break
                length += c
            length = length.strip()
            if length.isdigit():
                length = int(length)
            else:
                continue

154
            # read data and pass into listener
Josh Roesslein's avatar
Josh Roesslein committed
155
            data = resp.read(length)
156
157
            if self.listener.on_data(data) is False:
                self.running = False
Josh Roesslein's avatar
Josh Roesslein committed
158

159
160
161
162
163
164
165
166
    def _start(self, async):
        self.running = True
        if async:
            Thread(target=self._run).start()
        else:
            self._run()

    def firehose(self, count=None, async=False):
Josh Roesslein's avatar
Josh Roesslein committed
167
168
169
170
171
        if self.running:
            raise TweepError('Stream object already connected!')
        self.url = '/%i/statuses/firehose.json?delimited=length' % STREAM_VERSION
        if count:
            self.url += '&count=%s' % count
172
        self._start(async)
Josh Roesslein's avatar
Josh Roesslein committed
173

174
    def retweet(self, async=False):
175
176
177
        if self.running:
            raise TweepError('Stream object already connected!')
        self.url = '/%i/statuses/retweet.json?delimited=length' % STREAM_VERSION
178
        self._start(async)
179

180
    def sample(self, count=None, async=False):
Josh Roesslein's avatar
Josh Roesslein committed
181
182
183
184
185
        if self.running:
            raise TweepError('Stream object already connected!')
        self.url = '/%i/statuses/sample.json?delimited=length' % STREAM_VERSION
        if count:
            self.url += '&count=%s' % count
186
        self._start(async)
Josh Roesslein's avatar
Josh Roesslein committed
187

188
    def filter(self, follow=None, track=None, async=False, locations=None):
189
190
        params = {}
        self.headers['Content-type'] = "application/x-www-form-urlencoded"
Josh Roesslein's avatar
Josh Roesslein committed
191
192
193
194
        if self.running:
            raise TweepError('Stream object already connected!')
        self.url = '/%i/statuses/filter.json?delimited=length' % STREAM_VERSION
        if follow:
195
            params['follow'] = ','.join(map(str, follow))
Josh Roesslein's avatar
Josh Roesslein committed
196
        if track:
197
            params['track'] = ','.join(map(str, track))
198
199
        if locations and len(locations) > 0:
            assert len(locations) % 4 == 0
200
            params['locations'] = ','.join(['%.2f' % l for l in locations])
201
        self.body = urllib.urlencode(params)
202
        self._start(async)
Josh Roesslein's avatar
Josh Roesslein committed
203
204

    def disconnect(self):
205
        if self.running is False:
Josh Roesslein's avatar
Josh Roesslein committed
206
207
208
            return
        self.running = False