#!/usr/bin/env python #-*- coding: utf-8 -*- import sys import os import urllib import string import codecs import contextlib import re import sqlite3 import itertools import collections import textwrap import argparse import threading import lxml.html class NovelInfo(object): def __init__(self, name, author, article_keys): self.name = name self.author = author self.article_keys = article_keys[:] def __str__(self): return u"<{}>:作者:{}, article_keys:{}".format(self.name, self.author, self.article_keys).encode('utf-8') class INovelSite(object): """ 與網站相關資訊 * 取得有那些小說的資訊 * 取得小說相關資訊 * 取得小說內容 """ def get_site_name(self): """ 取得小說網站的名稱 """ raise NotImplementedError def get_novel_keys(self): """ 取得所有能代表小說的 key """ raise NotImplementedError def get_novel_info(self, novel_key): """ 取得小說基本資訊 * 小說名稱 * 作者 * article_keys """ raise NotImplementedError def get_article(self, article_key): """ 取得小說文章內容 """ raise NotImplementedError class NCHNovelSite(INovelSite): """ 小說頻道的內容 """ WEBSITE_URL = "http://www.nch.com.tw" def get_site_name(self): return u"小說頻道" def get_novel_keys(self): """ 取得所有能代表小說的 key """ for i in itertools.count(1): hot_list_page = "{}/hot05.php?pages={}".format(self.WEBSITE_URL, i) with contextlib.closing(urllib.urlopen(hot_list_page)) as fp: html = codecs.decode(fp.read(), "big5", "ignore") start_index = html.find("") + len("") end_index = html.find("") html = html[start_index: end_index] document_element = lxml.html.document_fromstring(html) element_list = document_element.findall("body/table[3]/tr/td[2]/a") if not element_list: break for e in element_list: #novel_keys.append("{}/{}".format(self.WEBSITE_URL, e.attrib['href'])) yield u"{}/{}".format(self.WEBSITE_URL, e.attrib['href']) def get_novel_info(self, novel_key): """ 取得小說基本資訊 * 作者 * 更新時間 * article_keys """ with contextlib.closing(urllib.urlopen(novel_key)) as fp: html = codecs.decode(fp.read(), "big5", "ignore") name = re.search(u"([^資]+)資料大全<\/b>", html).group(1) author = re.search(u"作　者[\s\S]+?]+>([^<]+)", html).group(1) article_keys = map(lambda s: "{}/{}".format(self.WEBSITE_URL, s), re.findall(u'觀看全集', html)) return NovelInfo(name, author, article_keys) def get_article(self, article_key): """ 取得小說文章內容