Python爬取推特(X)的各种数据

🌟 Hello，我是蒋星熠Jaxonic！

🌈 在浩瀚无垠的技术宇宙中，我是一名执着的星际旅人，用代码绘制探索的轨迹。

🚀 每一个算法都是我点燃的推进器，每一行代码都是我航行的星图。

🔭 每一次性能优化都是我的天文望远镜，每一次架构设计都是我的引力弹弓。

🎻 在数字世界的协奏曲中，我既是作曲家也是首席乐手。让我们携手，在二进制星河中谱写属于极客的壮丽诗篇！

一、简介

推特作为全球最大的实时信息平台之一，蕴含着海量的价值数据。在过去几年中，我设计并实现了多个推特数据爬取与分析系统，推特提供了多种API接口，用于开发者获取平台数据。了解这些API的特点和限制，是进行有效数据爬取的第一步。但是推特的反爬机制特别强，需要进行特殊的处理才可以。

二、结果展示

先开门见山了：

这是爬取到的数据，包含了很多内容，像推文内容，发布时间，点赞数，转推数，评论数，用户名，用户简介，用户ip，用户贴子数，关注量等等

三、代码部分：

引入需要的包：

import json
import os
import time
import requests
import pandas as pd
import datetime
import uuid
import base64
import random
from urllib.parse import quote

定义爬虫类和headers等信息：

# 禁用警告
requests.packages.urllib3.disable_warnings()

class TwitterKeywordSearch:
    def __init__(self, saveFileName, cookie_str):
        self.saveFileName = saveFileName
        self.cookie_str = cookie_str
        self.operation_hash = "NiZ1seU-Qm1TUiThEaWXKA"  # 固定 hash
        self.bearer_token = "Bearer AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA"

        # 从 Cookie 字符串解析
        self.cookies = {}
        for item in cookie_str.split("; "):
            if "=" in item:
                k, v = item.split("=", 1)
                self.cookies[k.strip()] = v.strip()

        # 提取 ct0 用于 x-csrf-token
        self.ct0 = self.cookies.get('ct0', '')

        # 基础 headers（完全复刻浏览器）
        self.headers = {
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/139.0.0.0 Safari/537.36 Edg/139.0.0.0',
            'Authorization': self.bearer_token,
            'x-twitter-client-language': 'en',
            'x-twitter-active-user': 'yes',
            'x-twitter-auth-type': 'OAuth2Session',
            'x-csrf-token': self.ct0,
            'Accept': '*/*',
            'sec-fetch-site': 'same-origin',
            'sec-fetch-mode': 'cors',
            'Referer': 'https://x.com/search',
            'Accept-Encoding': 'gzip, deflate, br, zstd',
            'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6,zh-TW;q=0.5',
            'Priority': 'u=1, i',
            'sec-ch-ua': '"Not;A=Brand";v="99", "Microsoft Edge";v="139", "Chromium";v="139"',
            'sec-ch-ua-arch': '"x86"',
            'sec-ch-ua-bitness': '"64"',
            'sec-ch-ua-full-version': '"139.0.3405.86"',
            'sec-ch-ua-full-version-list': '"Not;A=Brand";v="99.0.0.0", "Microsoft Edge";v="139.0.3405.86", "Chromium";v="139.0.7258.67"',
            'sec-ch-ua-mobile': '?0',
            'sec-ch-ua-platform': '"Windows"',
            'sec-ch-ua-platform-version': '"19.0.0"',
            'Content-Type': 'application/json',
            'x-client-transaction-id': self.generate_transaction_id(),  # 动态生成
        }

        self.resultList = []

定义各种爬取函数：

  def generate_transaction_id(self):
        """生成 x-client-transaction-id"""
        timestamp = str(int(time.time() * 1000))
        rand_str = str(uuid.uuid4())
        raw = f"{timestamp}:{rand_str}"
        return base64.b64encode(raw.encode()).decode()

    def get_params(self, raw_query, cursor=None):
        """构造 variables 和 features（完全复刻浏览器）"""
        variables = {
            "rawQuery": raw_query,
            "count": 20,
            "querySource": "recent_search_click",
            "product": "Latest",
            "withGrokTranslatedBio": False
        }
        if cursor:
            variables["cursor"] = cursor

        features = {
            "rweb_video_screen_enabled": False,
            "payments_enabled": False,
            "rweb_xchat_enabled": False,
            "profile_label_improvements_pcf_label_in_post_enabled": True,
            "rweb_tipjar_consumption_enabled": True,
            "verified_phone_label_enabled": False,
            "creator_subscriptions_tweet_preview_api_enabled": True,
            "responsive_web_graphql_timeline_navigation_enabled": True,
            "responsive_web_graphql_skip_user_profile_image_extensions_enabled": False,
            "premium_content_api_read_enabled": False,
            "communities_web_enable_tweet_community_results_fetch": True,
            "c9s_tweet_anatomy_moderator_badge_enabled": True,
            "responsive_web_grok_analyze_button_fetch_trends_enabled": False,
            "responsive_web_grok_analyze_post_followups_enabled": True,
            "responsive_web_jetfuel_frame": True,
            "responsive_web_grok_share_attachment_enabled": True,
            "articles_preview_enabled": True,
            "responsive_web_edit_tweet_api_enabled": True,
            "graphql_is_translatable_rweb_tweet_is_translatable_enabled": True,
            "view_counts_everywhere_api_enabled": True,
            "longform_notetweets_consumption_enabled": True,
            "responsive_web_twitter_article_tweet_consumption_enabled": True,
            "tweet_awards_web_tipping_enabled": False,
            "responsive_web_grok_show_grok_translated_post": False,
            "responsive_web_grok_analysis_button_from_backend": True,
            "creator_subscriptions_quote_tweet_preview_enabled": False,
            "freedom_of_speech_not_reach_fetch_enabled": True,
            "standardized_nudges_misinfo": True,
            "tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled": True,
            "longform_notetweets_rich_text_read_enabled": True,
            "longform_notetweets_inline_media_enabled": True,
            "responsive_web_grok_image_annotation_enabled": True,
            "responsive_web_grok_imagine_annotation_enabled": True,
            "responsive_web_grok_community_note_auto_translation_is_enabled": False,
            "responsive_web_enhance_cards_enabled": False
        }

        return {
            "variables": json.dumps(variables, separators=(',', ':')),
            "features": json.dumps(features, separators=(',', ':'))
        }

    def get(self, raw_query, cursor=None):
        """发送请求"""
        url = f"https://x.com/i/api/graphql/{self.operation_hash}/SearchTimeline"
        params = self.get_params(raw_query, cursor)

        # 每次请求更新 transaction-id
        self.headers['x-client-transaction-id'] = self.generate_transaction_id()

        # 动态更新 Referer
        self.headers['Referer'] = f"https://x.com/search?q={quote(raw_query)}&src=recent_search_click&f=live"

        try:
            response = requests.get(
                url,
                headers=self.headers,
                cookies=self.cookies,
                params=params,
                timeout=(10, 20),
                verify=False
            )

            if response.status_code == 429:
                reset_time = int(response.headers.get('x-rate-limit-reset', time.time() + 900))
                wait_time = max(reset_time - int(time.time()), 60)
                print(f"【限流】等待 {wait_time} 秒...")
                time.sleep(wait_time)
                return None

            if response.status_code == 200:
                try:
                    return response.json()
                except json.JSONDecodeError:
                    print("❌ 响应不是 JSON 格式")
                    print("响应内容:", response.text[:300])
                    return None
            else:
                print(f"请求失败: {response.status_code}, 响应: {response.text[:200]}")
                return None

        except Exception as e:
            print(f"请求异常: {e}")
            time.sleep(5)
            return None

    def get_cursor(self, response_json):
        """从响应中提取 next_cursor"""
        try:
            entries = response_json['data']['search_by_raw_query']['search_timeline']['timeline']['instructions']
            next_cursor = None

            for entry in entries:
                if entry['type'] == 'TimelineAddEntries':
                    for item in entry['entries']:
                        if 'cursor-bottom' in item['entryId']:
                            content = item['content']
                            if content.get('cursorType') == 'Bottom':
                                next_cursor = content['value']
                                break
                elif entry['type'] == 'TimelineReplaceEntry':
                    if 'cursor-bottom' in entry['entry']['entryId']:
                        next_cursor = entry['entry']['content']['value']
                        break
            return next_cursor
        except Exception as e:
            print(f"提取 cursor 失败: {e}")
            return None

    def parse_data(self, response_json):
        """解析推文数据"""
        def trans_time(created_at):
            dt = datetime.datetime.strptime(created_at, '%a %b %d %H:%M:%S +0000 %Y')
            return dt.strftime('%Y-%m-%d %H:%M:%S')

        items = []
        try:
            entries = response_json['data']['search_by_raw_query']['search_timeline']['timeline']['instructions']
            for entry in entries:
                if entry['type'] == 'TimelineAddEntries':
                    for item in entry['entries']:
                        entry_id = item['entryId']
                        if 'tweet-' in entry_id or 'profile-conversation' in entry_id:
                            content = item['content']['itemContent']
                            if 'tweet_results' in content:
                                tweet = content['tweet_results']['result']
                                legacy = tweet.get('legacy') or tweet.get('tweet', {}).get('legacy')
                                core = tweet.get('core') or tweet.get('tweet', {}).get('core')
                                if not legacy or not core:
                                    continue

                                user = core['user_results']['result']['legacy']
                                full_text = legacy.get('full_text', '')
                                if 'note_tweet' in tweet:
                                    full_text = tweet['note_tweet']['result']['text']

                                items.append({
                                    "内容": full_text,
                                    "时间": trans_time(legacy['created_at']),
                                    "点赞": legacy.get('favorite_count', 0),
                                    "评论": legacy.get('reply_count', 0),
                                    "转发": legacy.get('retweet_count', 0) + legacy.get('quote_count', 0),
                                    "简介": user.get('description', ''),
                                    "粉丝量": user.get('followers_count', 0),
                                    "关注量": user.get('friends_count', 0)
                                })
        except Exception as e:
            print(f"解析失败: {e}")

        if items:
            self.save_data(items)
            print(f"✅ 保存 {len(items)} 条推文")
        return items

    def save_data(self, data_list):
        """追加保存到 CSV"""
        df = pd.DataFrame(data_list)
        file_path = f'./{self.saveFileName}.csv'
        header = not os.path.exists(file_path)
        df.to_csv(file_path, index=False, mode='a', encoding='utf_8_sig', header=header)

    def run(self, raw_query):
        """主爬取逻辑"""
        cursor = None
        page = 1
        while True:
            print(f"📌 第 {page} 页...")
            response = self.get(raw_query, cursor)
            if not response:
                print("❌ 请求失败或被限流")
                break

            new_tweets = self.parse_data(response)
            if not new_tweets:
                print("📭 无新数据，可能已到底")
                break

            next_cursor = self.get_cursor(response)
            if not next_cursor or next_cursor == cursor:
                print("🔚 已到最后一页")
                break

            cursor = next_cursor
            page += 1
            time.sleep(random.uniform(3, 7))  # 随机延时

    def main(self, keywords_list):
        """批量执行"""
        for keyword in keywords_list:
            print(f"\n🔍 开始搜索: {keyword}")
            self.run(keyword)

用户配置，这里有像token,ct0,cookie等信息，需要自己配置，至于获取的方法见下文：

# ========================
# ✅ 用户配置区（请替换）
# ========================
if __name__ == '__main__':
    # 🔐 从浏览器复制的完整 Cookie（请确保有效）
    COOKIE_STR = "__cuid=0c6eafb48bca44bfa53c96ecfc34766e; g_state={\"i_l\":0}; kdt=DjYN3dU6uqZUFbK0Z2yCGIMfxs3tBeg26PMxUJ4a; att=1-yhCypM1zy0pRa4hZkYP5WuKabl5TxYyRiLioSlOW; lang=en; dnt=1; guest_id=v1%3A175516465513586295; guest_id_marketing=v1%3A175516465513586295; guest_id_ads=v1%3A175516465513586295; personalization_id=\"v1_i54Wl/43qBb3wyWw5rTpnw==\"; gt=1955928444279382425; _twitter_sess=BAh7CSIKZmxhc2hJQzonQWN0aW9uQ29udHJvbGxlcjo6Rmxhc2g6OkZsYXNo%250ASGFzaHsABjoKQHVzZWR7ADoPY3JlYXRlZF9hdGwrCLNX96eYAToMY3NyZl9p%250AZCIlYTk5MTdiYmU5N2MwYWNlMDcyNDNlYmU3NGE5YzNmYTg6B2lkIiU2ZDhk%250AZGQwZmVkYjgxOWM4N2JlOTI2MzZlNjQyMDIyMg%253D%253D--32e8b8bbc3092f636d7dfdf0870d7bc3cc9c1143; auth_token=a512f85cc3d5473e0efe957edd507ef70a795325; ct0=646e440727f56a3ae48ff5c117073ff5b225a9e04a13bee1ffc4765ced1e8024ff957ed7bae4dc1098a8e9216176f2ee03d631a6a7b6ad12f518baf47f966114005d09981cb8382103691b32255af6c6; twid=u%3D1955928620771520512; ph_phc_TXdpocbGVeZVm5VJmAsHTMrCofBQu3e0kN8HGMNGTVW_posthog=%7B%22distinct_id%22%3A%220198a7ff-c624-71c1-bbfd-28d98ef752a9%22%2C%22%24sesid%22%3A%5B1755165267989%2C%220198a7ff-c623-7586-8e40-0293329c615b%22%2C1755165214243%5D%7D; cf_clearance=Hg4mn9dhWvdVIcA5VcFR1Fw8TjX4aqyESpH1HYyBzE0-1755165970-1.2.1.1-Ese9AP5TQuv.LAU6Ah_iLxp0axDGXgahTEGQI_XJ6U_ezjTFFLzo9zg1WiRbX_6yrBwRFGRpp6MFt21Ruqza0LfY_qow25dGGt5Q7Yk.mEIH13Ao5vCkYx6vUBEh0hTDw9KWlZu_158u.L0qr9YSMactfbbobwbxV1wYuTPEIdxdVN6yxxgR17Fo.7ofspnvwFv_OuM8AEgkUc8IAmbQuUINolhFHTXctXrAPJqKrDw"

    # 🔍 搜索关键词（注意：中文逗号 %EF%BC%8C 已保留）
    KEYWORDS = ["MeToo，AzizAnsari"]  # 可添加更多，如 ["covid", "ai"]

    # 💾 保存文件名
    SAVE_FILE = "twitter_mee_too_aziz"

    # ✅ 创建爬虫并运行
    crawler = TwitterKeywordSearch(saveFileName=SAVE_FILE, cookie_str=COOKIE_STR)
    crawler.main(KEYWORDS)

四、获取自己推特Cookie,token，URL等信息的方法：

连接魔法后按F12跳转到开发者工具，然后点击网络，因为很多网页请求信息都存在searchtimeline里：

自己去找到相关信息改代码，也可以下载浏览器插件：

由于推特是有反爬机制的，searchtimeline会实时更新，每次往下翻页会出现新的searchtimeline，像这样：

所以我在代码里设置了更新读入的机制，还有推迟爬取防止被限流检测到不是真人

五、完整代码：

import json
import os
import time
import requests
import pandas as pd
import datetime
import uuid
import base64
import random
from urllib.parse import quote

# 禁用警告
requests.packages.urllib3.disable_warnings()

class TwitterKeywordSearch:
    def __init__(self, saveFileName, cookie_str):
        self.saveFileName = saveFileName
        self.cookie_str = cookie_str
        self.operation_hash = "NiZ1seU-Qm1TUiThEaWXKA"  # 固定 hash
        self.bearer_token = "Bearer AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA"

        # 从 Cookie 字符串解析
        self.cookies = {}
        for item in cookie_str.split("; "):
            if "=" in item:
                k, v = item.split("=", 1)
                self.cookies[k.strip()] = v.strip()

        # 提取 ct0 用于 x-csrf-token
        self.ct0 = self.cookies.get('ct0', '')

        # 基础 headers（完全复刻浏览器）
        self.headers = {
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/139.0.0.0 Safari/537.36 Edg/139.0.0.0',
            'Authorization': self.bearer_token,
            'x-twitter-client-language': 'en',
            'x-twitter-active-user': 'yes',
            'x-twitter-auth-type': 'OAuth2Session',
            'x-csrf-token': self.ct0,
            'Accept': '*/*',
            'sec-fetch-site': 'same-origin',
            'sec-fetch-mode': 'cors',
            'Referer': 'https://x.com/search',
            'Accept-Encoding': 'gzip, deflate, br, zstd',
            'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6,zh-TW;q=0.5',
            'Priority': 'u=1, i',
            'sec-ch-ua': '"Not;A=Brand";v="99", "Microsoft Edge";v="139", "Chromium";v="139"',
            'sec-ch-ua-arch': '"x86"',
            'sec-ch-ua-bitness': '"64"',
            'sec-ch-ua-full-version': '"139.0.3405.86"',
            'sec-ch-ua-full-version-list': '"Not;A=Brand";v="99.0.0.0", "Microsoft Edge";v="139.0.3405.86", "Chromium";v="139.0.7258.67"',
            'sec-ch-ua-mobile': '?0',
            'sec-ch-ua-platform': '"Windows"',
            'sec-ch-ua-platform-version': '"19.0.0"',
            'Content-Type': 'application/json',
            'x-client-transaction-id': self.generate_transaction_id(),  # 动态生成
        }

        self.resultList = []

    def generate_transaction_id(self):
        """生成 x-client-transaction-id"""
        timestamp = str(int(time.time() * 1000))
        rand_str = str(uuid.uuid4())
        raw = f"{timestamp}:{rand_str}"
        return base64.b64encode(raw.encode()).decode()

    def get_params(self, raw_query, cursor=None):
        """构造 variables 和 features（完全复刻浏览器）"""
        variables = {
            "rawQuery": raw_query,
            "count": 20,
            "querySource": "recent_search_click",
            "product": "Latest",
            "withGrokTranslatedBio": False
        }
        if cursor:
            variables["cursor"] = cursor

        features = {
            "rweb_video_screen_enabled": False,
            "payments_enabled": False,
            "rweb_xchat_enabled": False,
            "profile_label_improvements_pcf_label_in_post_enabled": True,
            "rweb_tipjar_consumption_enabled": True,
            "verified_phone_label_enabled": False,
            "creator_subscriptions_tweet_preview_api_enabled": True,
            "responsive_web_graphql_timeline_navigation_enabled": True,
            "responsive_web_graphql_skip_user_profile_image_extensions_enabled": False,
            "premium_content_api_read_enabled": False,
            "communities_web_enable_tweet_community_results_fetch": True,
            "c9s_tweet_anatomy_moderator_badge_enabled": True,
            "responsive_web_grok_analyze_button_fetch_trends_enabled": False,
            "responsive_web_grok_analyze_post_followups_enabled": True,
            "responsive_web_jetfuel_frame": True,
            "responsive_web_grok_share_attachment_enabled": True,
            "articles_preview_enabled": True,
            "responsive_web_edit_tweet_api_enabled": True,
            "graphql_is_translatable_rweb_tweet_is_translatable_enabled": True,
            "view_counts_everywhere_api_enabled": True,
            "longform_notetweets_consumption_enabled": True,
            "responsive_web_twitter_article_tweet_consumption_enabled": True,
            "tweet_awards_web_tipping_enabled": False,
            "responsive_web_grok_show_grok_translated_post": False,
            "responsive_web_grok_analysis_button_from_backend": True,
            "creator_subscriptions_quote_tweet_preview_enabled": False,
            "freedom_of_speech_not_reach_fetch_enabled": True,
            "standardized_nudges_misinfo": True,
            "tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled": True,
            "longform_notetweets_rich_text_read_enabled": True,
            "longform_notetweets_inline_media_enabled": True,
            "responsive_web_grok_image_annotation_enabled": True,
            "responsive_web_grok_imagine_annotation_enabled": True,
            "responsive_web_grok_community_note_auto_translation_is_enabled": False,
            "responsive_web_enhance_cards_enabled": False
        }

        return {
            "variables": json.dumps(variables, separators=(',', ':')),
            "features": json.dumps(features, separators=(',', ':'))
        }

    def get(self, raw_query, cursor=None):
        """发送请求"""
        url = f"https://x.com/i/api/graphql/{self.operation_hash}/SearchTimeline"
        params = self.get_params(raw_query, cursor)

        # 每次请求更新 transaction-id
        self.headers['x-client-transaction-id'] = self.generate_transaction_id()

        # 动态更新 Referer
        self.headers['Referer'] = f"https://x.com/search?q={quote(raw_query)}&src=recent_search_click&f=live"

        try:
            response = requests.get(
                url,
                headers=self.headers,
                cookies=self.cookies,
                params=params,
                timeout=(10, 20),
                verify=False
            )

            if response.status_code == 429:
                reset_time = int(response.headers.get('x-rate-limit-reset', time.time() + 900))
                wait_time = max(reset_time - int(time.time()), 60)
                print(f"【限流】等待 {wait_time} 秒...")
                time.sleep(wait_time)
                return None

            if response.status_code == 200:
                try:
                    return response.json()
                except json.JSONDecodeError:
                    print("❌ 响应不是 JSON 格式")
                    print("响应内容:", response.text[:300])
                    return None
            else:
                print(f"请求失败: {response.status_code}, 响应: {response.text[:200]}")
                return None

        except Exception as e:
            print(f"请求异常: {e}")
            time.sleep(5)
            return None

    def get_cursor(self, response_json):
        """从响应中提取 next_cursor"""
        try:
            entries = response_json['data']['search_by_raw_query']['search_timeline']['timeline']['instructions']
            next_cursor = None

            for entry in entries:
                if entry['type'] == 'TimelineAddEntries':
                    for item in entry['entries']:
                        if 'cursor-bottom' in item['entryId']:
                            content = item['content']
                            if content.get('cursorType') == 'Bottom':
                                next_cursor = content['value']
                                break
                elif entry['type'] == 'TimelineReplaceEntry':
                    if 'cursor-bottom' in entry['entry']['entryId']:
                        next_cursor = entry['entry']['content']['value']
                        break
            return next_cursor
        except Exception as e:
            print(f"提取 cursor 失败: {e}")
            return None

    def parse_data(self, response_json):
        """解析推文数据"""
        def trans_time(created_at):
            dt = datetime.datetime.strptime(created_at, '%a %b %d %H:%M:%S +0000 %Y')
            return dt.strftime('%Y-%m-%d %H:%M:%S')

        items = []
        try:
            entries = response_json['data']['search_by_raw_query']['search_timeline']['timeline']['instructions']
            for entry in entries:
                if entry['type'] == 'TimelineAddEntries':
                    for item in entry['entries']:
                        entry_id = item['entryId']
                        if 'tweet-' in entry_id or 'profile-conversation' in entry_id:
                            content = item['content']['itemContent']
                            if 'tweet_results' in content:
                                tweet = content['tweet_results']['result']
                                legacy = tweet.get('legacy') or tweet.get('tweet', {}).get('legacy')
                                core = tweet.get('core') or tweet.get('tweet', {}).get('core')
                                if not legacy or not core:
                                    continue

                                user = core['user_results']['result']['legacy']
                                full_text = legacy.get('full_text', '')
                                if 'note_tweet' in tweet:
                                    full_text = tweet['note_tweet']['result']['text']

                                items.append({
                                    "内容": full_text,
                                    "时间": trans_time(legacy['created_at']),
                                    "点赞": legacy.get('favorite_count', 0),
                                    "评论": legacy.get('reply_count', 0),
                                    "转发": legacy.get('retweet_count', 0) + legacy.get('quote_count', 0),
                                    "简介": user.get('description', ''),
                                    "粉丝量": user.get('followers_count', 0),
                                    "关注量": user.get('friends_count', 0)
                                })
        except Exception as e:
            print(f"解析失败: {e}")

        if items:
            self.save_data(items)
            print(f"✅ 保存 {len(items)} 条推文")
        return items

    def save_data(self, data_list):
        """追加保存到 CSV"""
        df = pd.DataFrame(data_list)
        file_path = f'./{self.saveFileName}.csv'
        header = not os.path.exists(file_path)
        df.to_csv(file_path, index=False, mode='a', encoding='utf_8_sig', header=header)

    def run(self, raw_query):
        """主爬取逻辑"""
        cursor = None
        page = 1
        while True:
            print(f"📌 第 {page} 页...")
            response = self.get(raw_query, cursor)
            if not response:
                print("❌ 请求失败或被限流")
                break

            new_tweets = self.parse_data(response)
            if not new_tweets:
                print("📭 无新数据，可能已到底")
                break

            next_cursor = self.get_cursor(response)
            if not next_cursor or next_cursor == cursor:
                print("🔚 已到最后一页")
                break

            cursor = next_cursor
            page += 1
            time.sleep(random.uniform(3, 7))  # 随机延时

    def main(self, keywords_list):
        """批量执行"""
        for keyword in keywords_list:
            print(f"\n🔍 开始搜索: {keyword}")
            self.run(keyword)


# ========================
# ✅ 用户配置区（请替换）
# ========================
if __name__ == '__main__':
    # 🔐 从浏览器复制的完整 Cookie（请确保有效）
    COOKIE_STR = "__cuid=0c6eafb48bca44bfa53c96ecfc34766e; g_state={\"i_l\":0}; kdt=DjYN3dU6uqZUFbK0Z2yCGIMfxs3tBeg26PMxUJ4a; att=1-yhCypM1zy0pRa4hZkYP5WuKabl5TxYyRiLioSlOW; lang=en; dnt=1; guest_id=v1%3A175516465513586295; guest_id_marketing=v1%3A175516465513586295; guest_id_ads=v1%3A175516465513586295; personalization_id=\"v1_i54Wl/43qBb3wyWw5rTpnw==\"; gt=1955928444279382425; _twitter_sess=BAh7CSIKZmxhc2hJQzonQWN0aW9uQ29udHJvbGxlcjo6Rmxhc2g6OkZsYXNo%250ASGFzaHsABjoKQHVzZWR7ADoPY3JlYXRlZF9hdGwrCLNX96eYAToMY3NyZl9p%250AZCIlYTk5MTdiYmU5N2MwYWNlMDcyNDNlYmU3NGE5YzNmYTg6B2lkIiU2ZDhk%250AZGQwZmVkYjgxOWM4N2JlOTI2MzZlNjQyMDIyMg%253D%253D--32e8b8bbc3092f636d7dfdf0870d7bc3cc9c1143; auth_token=a512f85cc3d5473e0efe957edd507ef70a795325; ct0=646e440727f56a3ae48ff5c117073ff5b225a9e04a13bee1ffc4765ced1e8024ff957ed7bae4dc1098a8e9216176f2ee03d631a6a7b6ad12f518baf47f966114005d09981cb8382103691b32255af6c6; twid=u%3D1955928620771520512; ph_phc_TXdpocbGVeZVm5VJmAsHTMrCofBQu3e0kN8HGMNGTVW_posthog=%7B%22distinct_id%22%3A%220198a7ff-c624-71c1-bbfd-28d98ef752a9%22%2C%22%24sesid%22%3A%5B1755165267989%2C%220198a7ff-c623-7586-8e40-0293329c615b%22%2C1755165214243%5D%7D; cf_clearance=Hg4mn9dhWvdVIcA5VcFR1Fw8TjX4aqyESpH1HYyBzE0-1755165970-1.2.1.1-Ese9AP5TQuv.LAU6Ah_iLxp0axDGXgahTEGQI_XJ6U_ezjTFFLzo9zg1WiRbX_6yrBwRFGRpp6MFt21Ruqza0LfY_qow25dGGt5Q7Yk.mEIH13Ao5vCkYx6vUBEh0hTDw9KWlZu_158u.L0qr9YSMactfbbobwbxV1wYuTPEIdxdVN6yxxgR17Fo.7ofspnvwFv_OuM8AEgkUc8IAmbQuUINolhFHTXctXrAPJqKrDw"

    # 🔍 搜索关键词（注意：中文逗号 %EF%BC%8C 已保留）
    KEYWORDS = ["MeToo，AzizAnsari"]  # 可添加更多，如 ["covid", "ai"]

    # 💾 保存文件名
    SAVE_FILE = "twitter_mee_too_aziz"

    # ✅ 创建爬虫并运行
    crawler = TwitterKeywordSearch(saveFileName=SAVE_FILE, cookie_str=COOKIE_STR)
    crawler.main(KEYWORDS)

■ 我是蒋星熠Jaxonic！如果这篇文章在你的技术成长路上留下了印记
■ 👁 【关注】与我一起探索技术的无限可能，见证每一次突破
■ 👍 【点赞】为优质技术内容点亮明灯，传递知识的力量
■ 🔖 【收藏】将精华内容珍藏，随时回顾技术要点
■ 💬 【评论】分享你的独特见解，让思维碰撞出智慧火花
■ 🗳 【投票】用你的选择为技术社区贡献一份力量
■ 技术路漫漫，让我们携手前行，在代码的世界里摘取属于程序员的那片星辰大海！