每周
2020-10-16 本文已影响0人
月夜星空下
# encoding:utf-8
import os
import re
import json
import time
import random
# import json
import codecs
import pymysql
import requests
import os, pprint
import mysql.connector # 先安装mysql-connector-python-1.0.12-py3.3,再引入包
# from aip import AipNlp
###图片路径###
path_url = '/Users/lilong/Desktop/likeshuo/url.txt'
op_url = open(path_url, encoding='utf-8')
pool_op_url = list(op_url)
#############
# data = readJSON.read_JSON("data.json")
# non = data['bosh'] #代表文章主要废话来源
# duplication = 1
dir_name = str(input("请输入生成文件夹名称："))
# page_num = int(input("请输入生成文章篇数："))
# dir_name = 'test'
all_link = '/Users/lilong/Desktop/{}'.format(dir_name)
dir = os.makedirs(all_link)
config = {'host': '127.0.0.1',
'user': 'root',
'password': 'yz1028959',
'port': 3306,
'database': 'data_likeshuo',
'charset': 'utf8'
}
try:
cnn = mysql.connector.connect(**config) # connect方法加载config的配置进行数据库的连接，完成后用一个变量进行接收
except mysql.connector.Error as e:
print('数据库链接失败！', str(e))
else: # try没有异常的时候才会执行
print("sucessfully!")
cursor = cnn.cursor(buffered=True) # 获取查询的标记位
######接口#######
# APP_ID = '17895362'
# API_KEY = 'LDvWo01ypc5umRTk6B3q1hjK'
# SECRET_KEY = 'ywEQtOsx4WVVPg291Y9rmOA7uqG9GDcs'
# client = AipNlp(APP_ID, API_KEY, SECRET_KEY)
# maxSummaryLen = 300
######分割线#####
""" 调用新闻摘要接口 """
# 查询数据库
n = random.randint(3, 5) # 随机生成自然段
# print("「简版」—— 用于测试环境") # 随机生成自然段+导读自然段
# data = (input("请输入关键词："), n)
data = (input("请输入关键词："), n)
t_key = data[0] # 只取关键词，不取随机生成自然段
cx_sql = 'select t1.content from children_title t LEFT JOIN children_english t1 on t.ceid=t1.id where t.ckey=%s ORDER BY RAND() LIMIT %s'
s_title = "select title from lks_title where title like '%%%%%s%%%%' ORDER BY RAND() LIMIT 1" % t_key
# link = pymysql.Connect(host='localhost', port=3306, user='root', password='yz1028959', db='dede1', charset='utf8')
# cursorde = link.cursor()
count_dd = 'select count(1) from dede_addonarticle'
# cursorde.execute(count_dd)
# rand_x = cursorde.fetchall()[0][0]
# cursorde.close()
# link.close()
# print(rand_x)
content_page_num = int(input("请输入生成文章篇数："))
page_num = 1
article = 0
all_rand_title = []
# all_n_piecewise_content = []
# all_one_content = []
p_lable = "<p>"
nbsp = " "
p_lable_last = "</p>"
all_one_middle = []
# randint_x = rand_x + content_page_num
randint_x = content_page_num
all_list_con_all = []
while True:
if article == content_page_num:
break
a_num = random.randint(0, randint_x)
all_con_str = ''
all_list_con = []
random.shuffle(pool_op_url)
con_url = pool_op_url[0]
# href_a = 'Click on the surprise：<body><a href="http://127.0.0.1/baike/{}.html" target="_blank">{}</a></body>'.format(a_num, t_key)
cursor.execute(cx_sql, data, )
all = cursor.fetchall()
cursor.execute(s_title)
rand_title = cursor.fetchall()
rand_title = rand_title[0][0] # 在列表中的元组取出字符串
all_rand_title.append(rand_title)
# con_url = "<p style={center}><img src={url} alt={rt}></p>".format(center="text-align:center", url=con_url, rt=rand_title )
p1 = re.compile(r'\s+') # 正则
add_list = []
content_list = []
for i in all:
content = ''
tupe1 = i[0]
add_list.append(tupe1)
content += i[0]
content = p1.sub('', content) # 去除让内容空格
content_list.append(content) # 新内容添加到content_list
# print(content_list) # 内容中部装在一个列表[]里面，以字符串,分开
n_piecewise_content = ''
for i in content_list:
# n_piecewise_content += p_lable + nbsp + i + p_lable_last + '\r\n'
n_piecewise_content += " " + i + '\r\n'
# header = " " + str(tmp) + lks_db(n_piecewise_content) + '\r\n' # 文章第一自然段，概括内容（对n_piecewise_content总结）
center_title = rand_title.center(100) + '\r\n' # 文章标题居中
# middle = con_url + n_piecewise_content + href_a # 文章中部
middle = n_piecewise_content
# middle = n_piecewise_content
# headers = {'Content-Type': 'application/x-www-form-urlencoded',
# 'Authorization': '7DFC27E2A121441A9C9CF348CC202D03'} # 消息头，根据实际需要添加
# url = "http://apis.5118.com/wyc/title" # 地址
# print('1')
# body = middle
# Body = {'txt': body} # 入参
# res = requests.post(url, headers=headers, data=Body) # 接口调用
# resTest = json.loads(res.text)
# rand_title = resTest['data']
# center_title = rand_title.center(100) + '\r\n'
# all_text = center_title + header + middle # 标题+开头+中部
all_text = center_title + middle
all_one_middle.append(middle)
one_content = all_text + '\r\n'
###生成###
# txt = os.path.join(all_link, str(article + 1) + '.txt')
txt = os.path.join(all_link, rand_title + '.txt')
w_txt = open(txt, 'w')
f = codecs.open(txt, 'r+', encoding='utf-8')
f.write(one_content)
f.close()
article += 1
print('ok')
