如何處理AttributeError：'NoneType'對象在大字典中沒有屬性'get'

Question

我有一個大詞典，可以將數據存儲在推文中。 其中的許多字段都具有值None或Null 。 當我使用get()函數時，我得到一個AttributeError: 'NoneType' object has no attribute 'get' 。

我的字典很大，不可能單獨處理每個案例。 關於如何做到這一點的任何想法？ 我的代碼是：

from twython import TwythonStreamer
import json
from pymongo import MongoClient
import os
from datetime import *
client = MongoClient()
db1 = client.PWSocial
db2 = client.PWSocial
db3 = client.PWSocial
db4 = client.PWSocial
APP_KEY = 'XXXXX'
APP_SECRET = 'XXXXX'
OAUTH_TOKEN = 'XXXXX'
OAUTH_TOKEN_SECRET = 'XXXXXX'
class MyStreamer(TwythonStreamer):
    def on_success(self, data):
        self.n=self.n+1
        print self.n
        print data
        user=data.get("user")
        tweets = db1.tweets
        user_mentions = db2.user_mentions
        hash_tags = db3.hash_tags
        users = db4.users
        dict1 = {'id':data.get("id"),'contributors':data.get("contributors"),'truncated':data.get('truncated'),'text':data.get('text'),'in_reply_to_status_id':data.get("in_reply_to_status_id"),'favorite_count':data.get('favorite_count'),'source':data.get('source'),'retweeted':data.get('retweeted'),'coordinates':data.get('coordinates'),'symbols':data.get('symbols'),'urls':data.get('urls'),'in_reply_to_screen_name':data.get('in_reply_to_screen_name'),'id_str':data.get('id_str'),'retweet_count':data.get('retweet_count'),'in_reply_to_user_id':data.get('in_reply_to_user_id'),'favorited':data.get('favorited'),'geo':data.get('geo'),'in_reply_to_user_id_str':data.get('in_reply_to_user_id_str'),'lang':data.get('lang'),'created_at':data.get('created_at'),'filter_level':data.get('filter_level'),'in_reply_to_status_id_str':data.get('in_reply_to_status_id_str'),'place':data.get('place'),'user_id':user.get('user_id')}
        posts1 = db1.posts
        post_id = posts1.insert(dict1)
        um = data.get("user_mentions")
        #dict2 = {'tweet_id':data.get('id'),'id':um.get('id'),'indices':um.get('indices'),'id_str':um.get('id_str'),'screen_name':um.get('screen_name'),'name':um.get('name')}
        #posts2 = db2.posts
        #post_id2 = posts2.insert(dict2)
        ht = data.get('hashtags')
        dict3 = {'tweet_id':data.get('id'),'indices':ht.get('indices'),'text':ht.get('text')}
        posts3 = db3.posts
        post_id3 = posts3.insert(dict3)
        usr = data.get("user")
        dict4 = {'follow_request_sent':usr.get('follow_request_sent'),'profile_use_background_image':usr.get('profile_use_background_image'),'default_profile_image':usr.get('default_profile_image'),'id':usr.get('id'),'verified':usr.get('verified'),'profile_image_url_https':usr.get('profile_image_url_https'),'profile_sidebar_fill_color':usr.get("profile_sidebar_fill_color"),'profile_text_color':usr.get('profile_text_color'),'followers_count':usr.get('followers_count'),'profile_sidebar_border_color':usr.get('profile_sidebar_border_color'),'id_str':usr.get('id_str'),'profile_background_color':usr.get('profile_background_color'),'listed_count':usr.get('listed_count'),'profile_background_image_url_https':usr.get('profile_background_image_url_https'),'utc_offset':usr.get('utc_offset'),'statuses_count':usr.get('statuses_count'),'description':usr.get('description'),'friends_count':usr.get('friends_count'),'location':usr.get('location'),'profile_link_color':usr.get('profile_link_color'),'profile_image_url':usr.get('profile_image_url'),'following':usr.get('following'),'geo_enabled':usr.get('geo_enabled'),'profile_banner_url':usr.get('profile_banner_url'),'profile_background_image_url':usr.get('profile_background_image_url'),'name':usr.get('name'),'lang':usr.get('lang'),'profile_background_tile':usr.get('profile_background_tile'),'favourites_count':usr.get('favourites_count'),'screen_name':usr.get('screen_name'),'notifications':usr.get('notifications'),'url':usr.get('url'),'created_at':usr.get('created_at'),'contributors_enabled':usr.get('contributors_enabled'),'time_zone':usr.get('time_zone'),'protected':usr.get('protected'),'default_profile':usr.get('default_profile'),'is_translator':usr.get('is_translator')}
        if (datetime.now() > self.stop_time):
            self.disconnect()
    def on_error(self, status_code, data):
        print status_code
        return True # Don't kill the stream

    def on_timeout(self):
        print >> sys.stderr, 'Timeout...'
        return True # Don't kill the stream

stream = MyStreamer(APP_KEY, APP_SECRET,
                    OAUTH_TOKEN, OAUTH_TOKEN_SECRET)
#stream.statuses.filter(follow = [57947109, 183093247, 89443197, 431336956])
stream.statuses.filter(follow = 95995660)

我正在處理的字典樣本是：

data = {
    "contributors": null, 
    "truncated": false, 
    "text": "@mipaltan #MI4LakhStronger if i become admin,i will hold trivia based contest which will held fans refresh old memories", 
    "in_reply_to_status_id": 420452758068539392, 
    "id": 420455226563231744, 
    "favorite_count": 0, 
    "source": "<a href=\"https://mobile.twitter.com\" rel=\"nofollow\">Mobile Web (M2)</a>", 
    "retweeted": false, 
    "coordinates": null, 
    "entities": {
        "symbols": [], 
        "user_mentions": [
            {
                "id": 106345557, 
                "indices": [
                    0, 
                    9
                ], 
                "id_str": "106345557", 
                "screen_name": "mipaltan", 
                "name": "Mumbai Indians"
            }
        ], 
        "hashtags": [
            {
                "indices": [
                    10, 
                    26
                ], 
                "text": "MI4LakhStronger"
            }
        ], 
        "urls": []
    }, 
    "in_reply_to_screen_name": "mipaltan", 
    "id_str": "420455226563231744", 
    "retweet_count": 0, 
    "in_reply_to_user_id": 106345557, 
    "favorited": false, 
    "user": {
        "follow_request_sent": null, 
        "profile_use_background_image": true, 
        "default_profile_image": false, 
        "id": 87174680, 
        "verified": false, 
        "profile_image_url_https": "https://pbs.twimg.com/profile_images/378800000220301249/a0c7b8c5766de83b65a42ca52196c4b3_normal.jpeg", 
        "profile_sidebar_fill_color": "EADEAA", 
        "profile_text_color": "333333", 
        "followers_count": 348, 
        "profile_sidebar_border_color": "D9B17E", 
        "id_str": "87174680", 
        "profile_background_color": "8B542B", 
        "listed_count": 5, 
        "profile_background_image_url_https": "https://si0.twimg.com/profile_background_images/350203578/Photo0003.jpg", 
        "utc_offset": 19800, 
        "statuses_count": 20119, 
        "description": "Sports Lover", 
        "friends_count": 708, 
        "location": "India", 
        "profile_link_color": "9D582E", 
        "profile_image_url": "http://pbs.twimg.com/profile_images/378800000220301249/a0c7b8c5766de83b65a42ca52196c4b3_normal.jpeg", 
        "following": null, 
        "geo_enabled": true, 
        "profile_background_image_url": "http://a0.twimg.com/profile_background_images/350203578/Photo0003.jpg", 
        "name": "Ronak Baj", 
        "lang": "en", 
        "profile_background_tile": true, 
        "favourites_count": 17, 
        "screen_name": "ronakbaj", 
        "notifications": null, 
        "url": null, 
        "created_at": "Tue Nov 03 12:02:56 +0000 2009", 
        "contributors_enabled": false, 
        "time_zone": "New Delhi", 
        "protected": false, 
        "default_profile": false, 
        "is_translator": false
    }, 
    "geo": null, 
    "in_reply_to_user_id_str": "106345557", 
    "lang": "en", 
    "created_at": "Tue Jan 07 07:21:52 +0000 2014", 
    "filter_level": "medium", 
    "in_reply_to_status_id_str": "420452758068539392", 
    "place": null
}

完整的追溯是：

  File "new.py", line 51, in <module>
    stream.statuses.filter(follow = 95995660)
  File "/usr/lib/python2.7/site-packages/twython/streaming/types.py", line 65, in filter
    self.streamer._request(url, 'POST', params=params)
  File "/usr/lib/python2.7/site-packages/twython/streaming/api.py", line 148, in _request
    if self.on_success(data):  # pragma: no cover
  File "new.py", line 33, in on_success
    dict3 = {'tweet_id':data.get('id'),'indices':ht.get('indices'),'text':ht.get('text')}
AttributeError: 'NoneType' object has no attribute 'get'

PS：如果我提供的任何數據不必要，請評論，我將編輯問題。

Answer 1

您正在嘗試訪問結果data.get()調用的.get()方法：

ht = data.get('hashtags')

和

usr = data.get("user")

但如果data字典中沒有 'hashtags'或'user'鍵，則ht或usr設置為None ，並且ht.get()或usr.get()調用失敗。

使用更好的默認值或防止usr為None ：

ht = data.get('hashtags', {})  # return an empty dictionary if missing

和

usr = data.get('user', {})  # return an empty dictionary if missing

如果密鑰不存在是錯誤，請不要使用.get()但使用直接訪問：

usr = data['user']

請注意，您的data對象在任何情況下都沒有 hashtags鍵。 但是， data['entities']字典中有一個hashtags鍵：

ht = data['entities']['hashtags']

這是一個ht.get('indices') 列表 ，所以ht.get('indices')現在會因AttributeError: 'list' object has no attribute 'get'失敗AttributeError: 'list' object has no attribute 'get' ; 你需要循環遍歷所有不同的主題標簽或從列表中選擇一個 。

Answer 2

根據我找到的最終解決方案，我自己回答了這個問題。

@Martjin Peters上面說的是絕對正確的，並且返回的推文在實體字典下的hashtags字典中沒有數據，在這種情況下，ht中沒有數據。 在這種情況下，不能使用.get()函數。

但是，解決方案很簡單，只需使用即可

ht = data.get('hashtags', {})

現在，在進一步使用ht字典之前，簡單地使用if條件就可以實現這一點

.
.
.#above code    

if um:
             dict2 = um[0]
             posts2 = db2.posts
             post_id2 = posts2.insert(dict2)
.
.
.#other code

為所有可能的字典實現此功能，其中不返回任何值。 在我的例子中，這必須為ht和um 。

希望如果將來有人遇到類似的問題，這會有所幫助。