# coding=utf-8
import requests
from lxml import etree
from urllib.parse import urljoin
import re
import os
# 獲取頁(yè)面，并返回解析整理好的文本
def get_page(url):
  response = requests.get(url, headers=header)
  set_encoding(response)
  text = parse_page(response.text)
  return text
# 解析頁(yè)面，將當(dāng)前頁(yè)面中的文字篩選出來(lái)
def parse_page(html):
  title = re.findall('<div class="bookname">\s+<h1>(.+?)</h1>', html)[0]
  content = re.findall('div id="content">(.*?)</div>', html, re.S)[0]
  content = content.replace('<br />', '').replace(' ', ' ').replace('\r\n\r\n', '\r\n')
  content = title + '\r\n' + content + '\r\n\r\n'
  return content
# 將文本追加到file_path對(duì)應(yīng)的txt中
def save_page(path, text):
  with open(path, 'a', encoding='utf-8') as f:
    f.write(text)
# 設(shè)置對(duì)response得到文本的解析編碼為'gbk'
def set_encoding(response):
  response.encoding = 'gbk'
# 從配置文件中獲取當(dāng)前保存的鏈接總數(shù)
def get_current_chapters_count(path):
  # （1）第一次讀配置文件可能沒(méi)有創(chuàng)建，所以要支持沒(méi)有文件創(chuàng)建文件的功能（2）如果文件存在，則不能清空，參考http://m.fzitv.net/article/158740.htm
  with open(path, 'a+') as f:
    f.seek(0)
    res = f.read()
    if res == '':
      return 0
    else:
      return int(res)
# 將保存的鏈接總數(shù)保存到配置文件中
def set_current_chapters_count(path, count):
  with open(path, 'w') as f:
    f.write(str(count))
# 需要配置的字典
config_dic = dict(start_url='http://www.booktxt.net/2_2220/', # 待下載小說(shuō)的章節(jié)首頁(yè) 
         latest_item=9, # 列出的所有章節(jié)鏈接中，前面幾個(gè)鏈接為最新章節(jié)，一般為9個(gè)，爬取時(shí)避免與最后部分重復(fù)，所以前面9個(gè)鏈接不爬取
         bookname='贅婿', # 待下載的小說(shuō)名 
         folder_path='D:\\') #保存位置
domain = 'http://www.booktxt.net' # 頂點(diǎn)網(wǎng)域名
if __name__ == '__main__':
  chapter_url_list = []
  response = requests.get(config_dic['start_url'], headers=header)
  set_encoding(response)
  html = etree.HTML(response.text)
  chapters = html.xpath('//dd')
  print('所有鏈接' + str(len(chapters)))
  chapters = chapters[config_dic['latest_item']:] # 前9章為最新章節(jié)，后面還會(huì)重復(fù)，這里去掉
  print('不重復(fù)有效章節(jié)鏈接' + str(len(chapters)))
  folder_path = config_dic['folder_path'] + config_dic['bookname']
  if not os.path.exists(folder_path):
    os.mkdir(folder_path)
  file_path = folder_path + '\\' + config_dic['bookname'] + '.txt'
  config_file_path = folder_path + '\\' + 'config.txt'
  print('小說(shuō)存儲(chǔ)路徑為：' + file_path)
  print('配置文件存儲(chǔ)路徑為：' + config_file_path)
  saved_count = get_current_chapters_count(config_file_path) # 獲取目前保存的小說(shuō)中已經(jīng)包含的章節(jié)數(shù)
  print('當(dāng)前' + file_path + '中已經(jīng)保存的章節(jié)總數(shù)' + str(saved_count))
  if saved_count < len(chapters): # 說(shuō)明有更新
    set_current_chapters_count(config_file_path, len(chapters))
    print('共更新 ' + str(len(chapters) - saved_count) + ' 章')
    for c in chapters[saved_count:]: # 從上次保存的位置開(kāi)始繼續(xù)保存
      url = c.xpath('a/@href')[0]
      url = urljoin(domain, url)
      txt = c.xpath('a/text()')[0]
      chapter_url_list.append(url)
      print(url)
      print(txt)
      save_page(file_path, get_page(url))
  else:
    print('小說(shuō)還沒(méi)有更新哦')