1 # -*- coding: utf-8 -*-
  2 """
  3 @author: findyou
  4 @contact: albert.peng@foxmail.com
  5 @version: 1.0
  6 @license: Apache Licence
  7 @file: get_idiom_from_baidu.py
  8 @time: 2018/10/21 20:35
  9 """
 10
 11 __author__ = 'albert'
 12 __version__ = '1.0'
 13
 14 import json
 15 import sqlite3
 16
 17 import os
 18 import requests
 19 import socket
 20 import time
 21 import sys
 22
 23 # 总页数，直接手动，不去获取了
 24 page_count = 1546
 25
 26 # 组装请求头
 27 header = {
 28     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
 29     'Accept-Encoding': 'gzip, deflate, sdch',
 30     'Accept-Language': 'zh-CN,zh;q=0.9',
 31     'Connection': 'keep-alive',
 32     'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Safari/537.36'
 33 }
 34 # 默认数据库
 35 db_filename = 'idiom.sqlite3'
 36
 37
 38 def get_idiom_data(start_pagenum=1, end_pagenum=10, all=False):
 39     '''
 40       爬取百度成语数据，解析并保存到数据到数据库
 41       :param start_pagenum: 默认从第1页开始
 42       :param end_pagenum: 默认第10页结束
 43       '''
 44     global page_count, header
 45
 46     # 统计成语条数
 47     idiom_count = 0
 48     # 进度条
 49     page_num = 0
 50     get_url = 'https://hanyu.baidu.com/hanyu/ajax/search_list?wd=%E6%88%90%E8%AF%AD&from=poem&pn={}&_={}'.format(1, int(round(time.time() * 1000)))
 51
 52     # 获取全部成语
 53     if all:
 54         end_pagenum = page_count
 55
 56     for i in range(start_pagenum, end_pagenum + 1):
 57         # 连接数据库
 58         conn = sqlite3.connect(db_filename)
 59         cursor = conn.cursor()
 60
 61         # 当前时间戳
 62         # t = int(round(time.time() * 1000))
 63         # 模拟请求获取json数据
 64         try:
 65             # 自动保存cookie
 66             s = requests.session()
 67             header['Referrer'] = get_url
 68
 69             # 百度 ajax 成语请求API
 70             get_url = 'https://hanyu.baidu.com/hanyu/ajax/search_list?wd=%E6%88%90%E8%AF%AD&from=poem&pn={}&_={}'.format(i, int(round(time.time() * 1000)))
 71
 72             # 模拟请求
 73             result = s.get(get_url, headers=header)
 74             # 判断请求是否成功
 75             if result.status_code == 200:
 76                 res = json.loads(result.text)
 77                 page_count = res['extra']['total-page']
 78                 # 得到返回的成语
 79                 for a in range(len(res['ret_array'])):
 80                     txt = res['ret_array'][a]
 81                     # 条数递增
 82                     idiom_count += 1
 83
 84                     # 目前只需：成语、拼音
 85                     name = (get_vaule(txt, 'name'))[0].strip()
 86                     pinyin = (get_vaule(txt, 'pinyin'))[0].strip()
 87                     pinyin_list = pinyin.split(" ")
 88
 89                     # 例句
 90                     # liju = get_vaule(txt, 'liju')[0]
 91                     # 出处
 92                     # tmp = get_vaule(txt, 'source')[0]
 93                     # if len(tmp) > 0:
 94                     #     source = tmp.replace('"', '').replace("'", "").replace('\n', '')
 95                     # else:
 96                     #     source = ''
 97                     # 同义词
 98                     # get_vaule(txt, 'synonym')
 99                     # tmp = get_vaule(txt, 'term_synonym')
100                     # if len(tmp) > 0:
101                     #     synonym = tmp
102                     # else:
103
104                     # ID，成语，拼音，成语首字，尾字，首拼，尾拼
105                     cursor.execute(
106                         'insert into IDIOM (ID,NAME,PINYIN,NAMEF,NAMEL,PINYINF,PINYINL) values (NULL,"%s","%s","%s","%s","%s","%s")' %
107                         (name, pinyin, name[0], name[len(name) - 1], pinyin_list[0], pinyin_list[len(pinyin_list) - 1]))
108             else:
109                 print("获取数据失败：第"+i+"页")
110         except requests.exceptions.ConnectTimeout as e:
111             print("http请求超时！" + str(e))
112         except socket.timeout as e:
113             print("请求超时！ " + str(e))
114         except socket.error as e:
115             print("请求错误！" + str(e))
116         finally:
117             # 每获取一页，保存一次
118             cursor.close()
119             conn.commit()
120             conn.close()
121         # 进度条
122         page_num += 1
123         view_bar(page_num, end_pagenum)
124     print('\n本次爬取[百度成语] :  第 ' + str(start_pagenum) + ' 至 ' +
125           str(end_pagenum) + ' 页，共计 ' + str(idiom_count) + ' 条')
126
127 def view_bar(num, total):
128     '''进度条'''
129     rate = num / total
130     rate_num = int(rate * 100)
131     # r = '\r %d%%' %(rate_num)
132     r = '\r[%s>] %d%%' % ('=' * rate_num, rate_num)
133     sys.stdout.write(r)
134     sys.stdout.flush
135
136
137 def get_vaule(idiom_dict, key_vaule):
138     if key_vaule in idiom_dict.keys():
139         return idiom_dict[key_vaule]
140     else:
141         return [' ']
142
143
144 def init_db(filename=db_filename):
145     '''
146      如果数据库不存在，自动在当前目录创建idiom.sqlite3:
147     '''
148     if not os.path.exists(filename):
149         conn = sqlite3.connect(filename)
150         # 创建一个Cursor:
151         cursor = conn.cursor()
152         # 建表ID 自增长key
153         cursor.execute(
154             'CREATE TABLE IDIOM (ID INTEGER PRIMARY KEY AUTOINCREMENT, NAME VARCHAR(100),NAMEF VARCHAR(10),NAMEL VARCHAR(10),\
155               PINYIN VARCHAR(100),PINYINF VARCHAR(10),PINYINL VARCHAR(10))')
156         # 关闭Cursor:
157         cursor.close()
158         # 提交事务:
159         conn.commit()
160         # 关闭Connection:
161         conn.close()
162
163
164 if __name__ == '__main__':
165     # 初始化数据库
166     init_db()
167
168     # 获取1-10页的数据，数据库只会往里一直加数据，未做去重，所以以下三个方式，用最后一种。
169     # get_idiom_data()
170
171     # 获取x-x页的数据
172     # get_idiom_data(start_pagenum=10,end_pagenum=30)
173
174     # 获取所有数据
175     get_idiom_data(all=True)