基于scrapy实现的简单蜘蛛采集程序

# Standard Python library imports
# 3rd party imports
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.selector import HtmlXPathSelector
# My imports
from poetry_analysis.items import PoetryAnalysisItem
HTML_FILE_NAME = r'.+\.html'
class PoetryParser(object):
  """
  Provides common parsing method for poems formatted this one specific way.
  """
  date_pattern = r'(\d{2} \w{3,9} \d{4})'
 
  def parse_poem(self, response):
    hxs = HtmlXPathSelector(response)
    item = PoetryAnalysisItem()
    # All poetry text is in pre tags
    text = hxs.select('//pre/text()').extract()
    item['text'] = ''.join(text)
    item['url'] = response.url
    # head/title contains title - a poem by author
    title_text = hxs.select('//head/title/text()').extract()[0]
    item['title'], item['author'] = title_text.split(' - ')
    item['author'] = item['author'].replace('a poem by', '')
    for key in ['title', 'author']:
      item[key] = item[key].strip()
    item['date'] = hxs.select("//p[@class='small']/text()").re(date_pattern)
    return item
class PoetrySpider(CrawlSpider, PoetryParser):
  name = 'example.com_poetry'
  allowed_domains = ['www.example.com']
  root_path = 'someuser/poetry/'
  start_urls = ['http://www.example.com/someuser/poetry/recent/',
         'http://www.example.com/someuser/poetry/less_recent/']
  rules = [Rule(SgmlLinkExtractor(allow=[start_urls[0] + HTML_FILE_NAME]),
                  callback='parse_poem'),
       Rule(SgmlLinkExtractor(allow=[start_urls[1] + HTML_FILE_NAME]),
                  callback='parse_poem')]