使用Twython流式传输推文时不断出现chunkedencodingerror

Question

我刚开始学习编程，正在尝试通过使用Twython这个Twitter的封装库来获取推文，想要了解一下如何处理API和数据。但是每次我这样做的时候，都会遇到一个错误，大约在获取5000条推文的时候就会出问题。我用其他的库，比如python-twitter，进行流式获取时就能顺利获取到大约80万条推文，没有遇到类似的错误。

--------------------------------------------------------------------------
ChunkedEncodingError                      Traceback (most recent call last)
<ipython-input-5-fdcf34f23648> in <module>()
      1 #[stream.statuses.filter(track='twitter')]
----> 2 stream.statuses.sample()

/Users/myusername/anaconda/lib/python2.7/site-packages/twython/streaming/types.pyc in sample(self, **params)
     75         url = 'https://stream.twitter.com/%s/statuses/sample.json' \
     76               % self.streamer.api_version
---> 77         self.streamer._request(url, params=params)
     78 
     79     def firehose(self, **params):

/Users/myusername/anaconda/lib/python2.7/site-packages/twython/streaming/api.pyc in _request(self, url, method, params)
    132             response = _send(retry_counter)
    133 
--> 134             for line in response.iter_lines(self.chunk_size):
    135                 if not self.connected:
    136                     break

/Users/myusername/anaconda/lib/python2.7/site-packages/requests/models.pyc in iter_lines(self, chunk_size, decode_unicode)
    643 
    644         for chunk in self.iter_content(chunk_size=chunk_size,
--> 645                                        decode_unicode=decode_unicode):
    646 
    647             if pending is not None:

/Users/myusername/anaconda/lib/python2.7/site-packages/requests/models.pyc in generate()
    616                         yield chunk
    617                 except IncompleteRead as e:
--> 618                     raise ChunkedEncodingError(e)
    619             except AttributeError:
    620                 # Standard file-like object.

ChunkedEncodingError: IncompleteRead(0 bytes read, 1 more expected)

我用来生成这个的代码如下。我知道这个代码缺少很多东西，包括处理这个特定错误的方法。

from twython import TwythonStreamer
import numpy as np
import pandas as pd
from requests.exceptions import ChunkedEncodingError

counter = 0
class MyStreamer(TwythonStreamer):
    def on_success(self, data):
        global counter
        #if 'text' in data:
            #print data['text'].encode('utf-8')
        counter+=1
        print counter

    def on_error(self, status_code, data):
        print status_code

        # Want to stop trying to get data because of the error?
        # Uncomment the next line!
        # self.disconnect()

stream = MyStreamer(APP_KEY, APP_SECRET, OAUTH_TOKEN, OAUTH_TOKEN_SECRET)
stream.statuses.sample()

其实现在我并不打算对推文做什么，只是想看看能否在没有错误的情况下让这个程序正常运行。

Answer 1

嗯，我没搞清楚是什么导致了这个错误，但如果我改一下最后几行代码，它就能正常工作了。像这样：

def streamed():
    while True:
        try:
            stream = MyStreamer(APP_KEY, APP_SECRET, OAUTH_TOKEN, OAUTH_TOKEN_SECRET)
            stream.statuses.filter(track='twitter', stall_warnings=True)
        except  ChunkedEncodingError:
            continue

streamed()

这样可以让程序在出错时重新启动。 我还需要在

return True

这个类里的函数中添加一些东西。

之前这个回答也帮了我：如果出错，如何重新启动tweepy脚本？