使用Tweepy调用Twitter API v2提取地点字段遇'places'错误求助

尝试用Tweepy Paginator提取推文的地点字段时，Python报出和places相关的错误，但用同样方式处理用户字段完全正常，相关代码如下：

import tweepy
import time
import pandas as pd

client = tweepy.Client(bearer_token='____')

puma_tweets = []
for response in tweepy.Paginator(client.search_all_tweets, 
                                 query = 'puma -is:retweet lang:en place_country:US',
                                 user_fields = ['id', 'username', 'name'],
                                 place_fields = ['id','full_name', 'country', 'geo', 'name', 'place_type'],
                                 tweet_fields = ['id', 'created_at', 'geo', 'public_metrics', 'text'],
                                 expansions = 'author_id',
                                 max_results=500, limit = 20):
    time.sleep(1)
    puma_tweets.append(response)


result = []
user_dict = {}
place_dict = {}

for response in puma_tweets:

    for user in response.includes['users']:
        user_dict[user.id] = {'id': user.id,
                              'username': user.username, 
                              'name': user.name,
                              'followers': user.public_metrics['followers_count'],
                             }
        
    
    for place in response.includes['places']:
        place_dict[place.id] = {'id': place.id,
                                'full_name': place.full_name
                             }

    for tweet in response.data:

        author_info = user_dict[tweet.author_id]

        result.append({'author_id': tweet.author_id,
                       'username': author_info['username'],
                       'text': tweet.text,
                       'geo': tweet.geo
                      }) 


df = pd.DataFrame(result) 

核心问题 #

你没在expansions参数里添加geo.place_id，Twitter API不会主动返回地点数据到response.includes['places']中。而用户字段能正常工作，是因为你把author_id加入了expansions列表。另外，部分分页响应可能没有地点数据，直接访问response.includes['places']会触发KeyError。

修复步骤 #

1. 更新expansions参数，把geo.place_id加进去：

expansions = ['author_id', 'geo.place_id']

注意这里要改成列表形式，因为要指定多个扩展字段。

2. 给地点遍历加存在性判断，避免无地点数据时报错：

# 替换原来的place遍历代码
if 'places' in response.includes:
    for place in response.includes['places']:
        place_dict[place.id] = {'id': place.id,
                                'full_name': place.full_name
                             }

3. （可选）把地点信息关联到结果里，如果需要在最终DataFrame中显示地点名称：

# 在tweet循环中添加地点信息
for tweet in response.data:
    author_info = user_dict[tweet.author_id]
    # 获取地点信息（如果有的话）
    place_full_name = None
    if tweet.geo and 'place_id' in tweet.geo:
        place_info = place_dict.get(tweet.geo['place_id'])
        if place_info:
            place_full_name = place_info['full_name']
    
    result.append({'author_id': tweet.author_id,
                   'username': author_info['username'],
                   'text': tweet.text,
                   'geo': tweet.geo,
                   'place_full_name': place_full_name
                  })

修复后关键代码片段 #

修改后的Paginator调用部分：

for response in tweepy.Paginator(client.search_all_tweets, 
                                 query = 'puma -is:retweet lang:en place_country:US',
                                 user_fields = ['id', 'username', 'name'],
                                 place_fields = ['id','full_name', 'country', 'geo', 'name', 'place_type'],
                                 tweet_fields = ['id', 'created_at', 'geo', 'public_metrics', 'text'],
                                 expansions = ['author_id', 'geo.place_id'],  # 这里修改
                                 max_results=500, limit = 20):
    time.sleep(1)
    puma_tweets.append(response)

内容的提问来源于stack exchange，提问作者user17424407