1. def data_arrange(data):
    2.     import json
    3.     import requests
    4.     data_set = []
    5.     for ps in data:
    6.         try:
    7.             data_set.append(ps.get())
    8.         except json.decoder.JSONDecodeError:
    9.             pass
    10.         except requests.exceptions.SSLError:
    11.             pass
    12.     return data_set
    15. def data_tran(text):
    16.     loca = tuple(text.keys())[0]
    17.     data = text[loca]
    18.     res = []
    19.     contain = []
    20.     for it in data:
    21.         keys = tuple(it.keys())[0]
    22.         te = it[keys]
    23.         for t in te:
    24.             res.append(loca)
    25.             res.append(keys)
    26.             res.extend(t)
    27.             middle = res.copy()
    28.             contain.append(middle)
    29.             res.clear()
    30.     print(contain)
    31.     return contain
    34. def data_output(data):
    35.     res = []
    36.     data_sets = data_arrange(data)
    37.     for it in data_sets:
    38.         out = data_tran(it)
    39.         res.extend(out)
    40.     for g in res:
    41.         if len(g) > 4:
    42.             g.remove("NaN")
    43.     return res
    46. def output_data(all_data, output_file_name):
    47.     '''该函数将最终数据输出为CSV文件'''
    48.     import pandas as pd
    49.     name = ["location", "keyword", "date", "text"]
    50.     table = pd.DataFrame(columns=name, data=all_data)
    51.     table.to_csv("/Users/mengxinwei/Downloads/" + output_file_name + ".csv")
    52.     return table
    55. ###############################
    57. data_final = data_output(data_1)
    58. for g in data_final:
    59.     if len(g) > 4:
    60.         g.remove("NaN")
    61. s = output_data(final_2000, "Weibo_RawText_2000")
    68. def get_sentiment(text):
    69.     ssl._create_default_https_context = ssl._create_unverified_context
    70.     urllib3.disable_warnings()
    71.     ###第一步：获取access_token
    72.     host = 'https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials&client_id=pd152AVsnYc6Nbfk5Yuh9XTh&client_secret=ZTn7ASCKAGgPFEFmMxszEoQmAlVb2LRM'
    73.     request = urllib.request.Request(host)
    74.     request.add_header('Content-Type', 'application/json; charset=UTF-8')
    75.     response = urllib.request.urlopen(request)
    76.     content = response.read()
    77.     # if (content):
    78.     #     print(type(content))  # <class 'bytes'>
    79.     content_str = str(content, encoding="utf-8")
    80.     content_dir = eval(content_str)
    81.     access_token = content_dir['access_token']
    82.     ### 第二步：获取
    83.     # print(sys.getdefaultencoding())
    84.     http=urllib3.PoolManager()
    85.     url = "https://aip.baidubce.com/rpc/2.0/nlp/v1/sentiment_classify?access_token="+access_token
    86.     # print(url)
    87.     data ={"text": text}
    88.     encode_data = json.dumps(data).encode('GBK')
    89.     #JSON:在发起请求时,可以通过定义body 参数并定义headers的Content-Type参数来发送一个已经过编译的JSON数据：
    90.     request = http.request('POST',
    91.                            url,
    92.                            body=encode_data,
    93.                            headers={'Content-Type': 'application/json'}
    94.                            )
    95.     result = request.data.decode("GBK")[-4]
    96.     return result