init commit

baidu.py

@@ -0,0 +1,119 @@
+import requests
+from bs4 import BeautifulSoup
+
+def fetch_related_words(keyword):
+    cookies = {
+        'PSTM': '1764302604',
+        'BAIDUID': '17E56B6A4915D5B98222C8D7A7CFF059:FG=1',
+        'BD_HOME': '1',
+        'H_PS_PSSID': '63140_64007_65866_66117_66218_66194_66236_66243_66168_66362_66281_66264_66393_66395_66479_66510_66529_66553_66589_66590_66602_66614_66647_66679_66692_66695_66687',
+        'delPer': '0',
+        'BD_CK_SAM': '1',
+        'PSINO': '3',
+        'BAIDUID_BFESS': '17E56B6A4915D5B98222C8D7A7CFF059:FG=1',
+        'PAD_BROWSER': '1',
+        'BD_UPN': '12314753',
+        'BDORZ': 'B490B5EBF6F3CD402E515D22BCDA1598',
+        'BA_HECTOR': 'ala10k2421a0ag25a5a40524a10l8n1kii7of24',
+        'BIDUPSID': 'C047CB4D757AC8632D7B5792A4254C89',
+        'ZFY': 'hLpeh2:BHPDeKfEN3yuM7C:A7dmFl03pP:AkeekLlPw5J4:C',
+        'channel': 'baidusearch',
+        'H_WISE_SIDS': '63140_64007_65866_66117_66218_66194_66236_66243_66168_66362_66281_66264_66393_66395_66479_66510_66529_66553_66589_66590_66602_66614_66647_66679_66692_66695_66687',
+        'baikeVisitId': '7f510782-16ce-4371-ad1d-cc8c0ba5ccc8',
+        'COOKIE_SESSION': '0_0_1_0_0_0_1_0_1_1_7462_1_0_0_0_0_0_0_1764302605%7C1%230_0_1764302605%7C1',
+        'H_PS_645EC': 'c7554ktAJah5Z6fmLi0RDEpB3a2TvS0rgHEQ7JP12K2UeBuFhGHrlxODIbY',
+        'BDSVRTM': '16',
+        'WWW_ST': '1764310101036',
+    }
+
+    from urllib.parse import quote
+    headers = {
+        'Accept': '*/*',
+        'Accept-Language': 'zh-CN,zh;q=0.9',
+        'Connection': 'keep-alive',
+        'Referer': f'https://www.baidu.com/s?ie=utf-8&f=8&rsv_bp=1&tn=baidu&wd={quote(keyword)}',
+        'Sec-Fetch-Dest': 'empty',
+        'Sec-Fetch-Mode': 'cors',
+        'Sec-Fetch-Site': 'same-origin',
+        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/142.0.0.0 Safari/537.36',
+        'X-Requested-With': 'XMLHttpRequest',
+        # 'is_pbs': quote(keyword),  # 中文字段已去除
+        # 'is_referer': f'https://www.baidu.com/s?ie=utf-8&f=8&rsv_bp=1&tn=baidu&wd={quote(keyword)}',
+        'is_xhr': '1',
+        'sec-ch-ua': '"Chromium";v="142", "Google Chrome";v="142", "Not_A Brand";v="99"',
+        'sec-ch-ua-mobile': '?0',
+        'sec-ch-ua-platform': '"Windows"',
+    }
+
+    params = [
+        ('ie', 'utf-8'),
+        ('mod', '1'),
+        ('isbd', '1'),
+        ('isid', 'b857f27e00205440'),
+        ('ie', 'utf-8'),
+        ('f', '8'),
+        ('rsv_bp', '1'),
+        ('tn', 'baidu'),
+        ('wd', keyword),
+        ('oq', keyword),
+        ('rsv_pq', 'b857f27e00205440'),
+        ('rsv_t', 'c7554ktAJah5Z6fmLi0RDEpB3a2TvS0rgHEQ7JP12K2UeBuFhGHrlxODIbY'),
+        ('rqlang', 'cn'),
+        ('rsv_enter', '1'),
+        ('rsv_dl', 'tb'),
+        ('rsv_btype', 't'),
+        ('inputT', '23852'),
+        ('rsv_sug2', '0'),
+        ('rsv_sug3', '15'),
+        ('rsv_sug1', '23'),
+        ('rsv_sug7', '100'),
+        ('rsv_sug4', '23852'),
+        ('bs', keyword),
+        ('rsv_sid', 'undefined'),
+        ('_ss', '1'),
+        ('clist', 'ddad409c4a1855aa'),
+        ('hsug', ''),
+        ('f4s', '1'),
+        ('csor', '3'),
+        ('_cr1', '35542'),
+    ]
+
+    response = requests.get('https://www.baidu.com/s', params=params, cookies=cookies, headers=headers)
+    soup = BeautifulSoup(response.text, 'html.parser')
+
+    result = []
+    div = soup.find('div', class_='list_1V4Yg')
+    if div:
+        for a in div.find_all('a', class_='item_3WKCf'):
+            spans = a.find_all('span')
+            if len(spans) > 1:
+                result.append(spans[1].get_text(strip=True))
+
+    # 相关搜索内容
+    related_search = []
+    rs_label = soup.find('div', class_='c-color-t rs-label_ihUhK')
+    if rs_label:
+        rs_table = rs_label.find_next('table', class_='rs-table_3RiQc')
+        if rs_table:
+            for a in rs_table.find_all('a', class_='rs-link_2DE3Q'):
+                span = a.find('span', class_='rs-text_3K5mR')
+                if span:
+                    related_search.append(span.get_text(strip=True))
+
+    # 保存所有内容到一个文件
+    with open('baidu_result.txt', 'w', encoding='utf-8') as f:
+        f.write('大家都在搜:\n')
+        for item in result:
+            f.write(item + '\n')
+        f.write('\n相关搜索:\n')
+        for item in related_search:
+            f.write(item + '\n')
+
+    return result, related_search
+
+if __name__ == '__main__':
+    keyword = input('请输入关键词：')
+    words, related_search = fetch_related_words(keyword)
+    print("大家都在搜：", words)
+    print("相关搜索：", related_search)
+    print("所有内容已保存到 baidu_result.txt")