Python 请求 - ConnectionError - 超出最大重试次数

Question

我正在使用 rauth 和 requests 来调用 Beatport API。通话有效，但我偶尔会收到以下错误ConnectionError: HTTPSConnectionPool(host='oauth-api.beatport.com', port=443): Max retries exceeded with url

这是回溯。

Traceback (most recent call last):
  File "<console>", line 1, in <module>
  File "scraper/songlist_top100.py", line 88, in <module>
    'sortBy': 'releaseDate ASC'})
  File "C:\Python27\lib\site-packages\requests\sessions.py", line 347, in get
    return self.request('GET', url, **kwargs)
  File "C:\Python27\lib\site-packages\rauth\session.py", line 208, in request
    return super(OAuth1Session, self).request(method, url, **req_kwargs)
  File "C:\Python27\lib\site-packages\requests\sessions.py", line 335, in reques
t
    resp = self.send(prep, **send_kwargs)
  File "C:\Python27\lib\site-packages\requests\sessions.py", line 438, in send
    r = adapter.send(request, **kwargs)
  File "C:\Python27\lib\site-packages\requests\adapters.py", line 327, in send
    raise ConnectionError(e)
ConnectionError: HTTPSConnectionPool(host='oauth-api.beatport.com', port=443): 
Max retries exceeded with url: /catalog/3/tracks?perPage=150&
oauth_nonce=xxxxx&oauth_timestamp=xxxxx&facets=artistName%3A
Avicii&oauth_signature_method=HMAC-SHA1&oauth_version=1.0&
oauth_consumer_key=xxxxx&oauth_token=xxxxxx&sortBy=releaseDate+ASC
&oauth_signature=xxxxx%3D&page=3 (Caused by <class 'httplib.BadStatusLine'>: '')

这是我的脚本

from rauth import OAuth1Service
import requests
from hunt.models import DJ, Song

def get_obj_or_none(model, **kwargs):
    try:
        return model.objects.get(**kwargs)
    except model.DoesNotExist:
        return None

beatport_login = 'xxx'
beatport_pass = 'xxx'

beatport = OAuth1Service(
    name='beatport',
    consumer_key='xxxxx',
    consumer_secret='xxxxx',
    request_token_url= 'https://oauth-api.beatport.com/identity/1/oauth/request-token',
    access_token_url='https://oauth-api.beatport.com/identity/1/oauth/access-token',
    authorize_url='https://oauth-api.beatport.com/identity/1/oauth/authorize',
    base_url='https://oauth-api.beatport.com/json/catalog')

request_token, request_token_secret = beatport.get_request_token(method='POST', data={
    'oauth_callback': 'http://www.edmhunters.com'})

authorize_url = beatport.get_authorize_url(request_token)

values = {
    'oauth_token': request_token,
    'username': beatport_login,
    'password': beatport_pass,
    'submit' : 'Login',
}

r = requests.post('https://oauth-api.beatport.com/identity/1/oauth/authorize-submit', data=values)

verifier = r.url.split("oauth_verifier=",1)[1]

tokens = beatport.get_raw_access_token(request_token, request_token_secret, method='POST', data={
    'oauth_verifier': verifier})

token_string = tokens.content

access_token = token_string[token_string.find('=')+1:token_string.find('&')]
access_token_secret = token_string[token_string.find('t=')+2:token_string.rfind('&s')]

session = beatport.get_session((access_token, access_token_secret))

for dj in DJ.objects.all():
    r = session.get('https://oauth-api.beatport.com/catalog/3/tracks', params={'facets': "artistName:"+dj.name, 'perPage': 150})
    count_response = r.json()
    results = []
    for i in range(1, count_response['metadata']['totalPages']+1):
        r1 = session.get('https://oauth-api.beatport.com/catalog/3/tracks', params={'facets': "artistName:"+dj.name,
                                                                                    'page': i,
                                                                                    'perPage': 150,
                                                                                    'sortBy': 'releaseDate ASC'})
        json_response = r1.json()
        results += json_response['results']

    song_list = []      
    for song in results:
        artists =  [artist['name'] for artist in song['artists'] if str(artist['type'])=='artist']
        remixers = [artist['name'] for artist in song['artists'] if str(artist['type'])=='remixer']
        if not ((dj.name in artists) and ((dj.name not in remixers) if len(remixers)>0 else False)):
            song_list.append(song)

    for song in song_list:
        artists =  [artist['name'] for artist in song['artists'] if str(artist['type'])=='artist']
        remixers = [artist['name'] for artist in song['artists'] if str(artist['type'])=='remixer']
        artist_list = ', '.join(artists)
        remixer_list = ', '.join(remixers)
        song_name = song['name']
        if not(song_name.lower().find("feat.") == -1 ):
            normal_name=song_name[0:song_name.lower().find("feat.")].rstrip()
        else:
            normal_name=song_name

        genre_list=[]
        for genre in song['genres']:
            genre_list.append(genre['name'])
        genres = ', '.join(genre_list)

        if not get_obj_or_none(Song, name__iexact=song_name, artist=dj):
            s = Song(song_id=song['id'],
                     name=song_name,
                     title=song['title'],
                     normalized_name=normal_name,
                     artist=dj,
                     artists=artist_list,
                     remixers=remixer_list,
                     release_date=song['releaseDate'],
                     slug=song['slug'],
                     artwork=song['images']['large']['url'],
                     genres=genres)
            s.save()
            print "Added song:", s.song_id, s.artist

为什么会出现上述错误？

Answer 1

Beatport API 似乎过载并且有时会过早关闭连接。您的第一组请求成功了，因为响应为空，是第 3 页引发了错误。

您确实应该将此报告给 Beatport，但您或许可以通过指示 requests 模块重试请求来解决此问题：

from requests.adapters import HTTPAdapter

# ....

session = beatport.get_session((access_token, access_token_secret))
session.mount('https://oauth-api.beatport.com', HTTPAdapter(max_retries=5))

如果发生错误，会再重试几次您的请求。