Question

I am facing a weird issue while trying to crawl a particular site.
If i use the basespider to crawl some pages, the code works perfectly, but if i change the code to use crawlspider, the spider completes without any errors but with no spidering

The following piece of code works fine

    from scrapy.spider import BaseSpider
    from scrapy.selector import HtmlXPathSelector
    from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
    from scrapy.contrib.spiders import CrawlSpider, Rule
    from scrapy.contrib.loader import XPathItemLoader
    from dirbot.items import Website
    from urlparse import urlparse
    from scrapy import log


class hushBabiesSpider(BaseSpider):
   name = "hushbabies"
   #download_delay = 10
   allowed_domains = ["hushbabies.com"]
   start_urls = [
       "http://www.hushbabies.com/category/toys-playgear-bath-bedtime.html",
       "http://www.hushbabies.com/category/mommy-newborn.html",
       "http://www.hushbabies.com"

   ]
   def parse(self, response):
       print response.body
       print "Inside parse Item"
       return []

The following piece of code does not work

from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.loader import XPathItemLoader
from dirbot.items import Website
from urlparse import urlparse
from scrapy import log

class hushBabiesSpider(CrawlSpider):
   name = "hushbabies"
   #download_delay = 10
   allowed_domains = ["hushbabies.com"]
   start_urls = [
       "http://www.hushbabies.com/category/toys-playgear-bath-bedtime.html",
       "http://www.hushbabies.com/category/mommy-newborn.html",
       "http://www.hushbabies.com"

   ]
   rules = (
        Rule(SgmlLinkExtractor(allow=()),
            'parseItem',
            follow=True,
        ),
    )
  def parseItem(self, response):
       print response.body
       print "Inside parse Item"
       return []

The output from the Scrapy run is as follows

scrapy crawl hushbabies
2012-07-23 18:50:37+0000 [scrapy] INFO: Scrapy 0.15.1-198-g831a450 started (bot: SKBot)
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Enabled extensions: LogStats, TelnetConsole, WebService, CoreStats, MemoryUsage, SpiderState, CloseSpider
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Enabled downloader middlewares: RobotsTxtMiddleware, HttpAuthMiddleware, DownloadTimeoutMiddleware, UserAgentMiddleware, RetryMiddleware, DefaultHeadersMiddleware, RedirectMiddleware, CookiesMiddleware, HttpCompressionMiddleware, ChunkedTransferMiddleware, DownloaderStats
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Enabled spider middlewares: HttpErrorMiddleware, OffsiteMiddleware, RefererMiddleware, UrlLengthMiddleware, DepthMiddleware
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Enabled item pipelines: SQLStorePipeline
2012-07-23 18:50:37+0000 [hushbabies] INFO: Spider opened
2012-07-23 18:50:37+0000 [hushbabies] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Telnet console listening on 0.0.0.0:6023
2012-07-23 18:50:37+0000 [scrapy] DEBUG: Web service listening on 0.0.0.0:6080
2012-07-23 18:50:37+0000 [hushbabies] DEBUG: Crawled (200) <GET http://www.hushbabies.com/robots.txt> (referer: None)
2012-07-23 18:50:39+0000 [hushbabies] DEBUG: Crawled (200) <GET http://www.hushbabies.com> (referer: None)
2012-07-23 18:50:39+0000 [hushbabies] DEBUG: Crawled (200) <GET http://www.hushbabies.com/category/mommy-newborn.html> (referer: None)
2012-07-23 18:50:39+0000 [hushbabies] INFO: Closing spider (finished)
2012-07-23 18:50:39+0000 [hushbabies] INFO: Dumping spider stats:
        {'downloader/request_bytes': 634,
         'downloader/request_count': 3,
         'downloader/request_method_count/GET': 3,
         'downloader/response_bytes': 44395,
         'downloader/response_count': 3,
         'downloader/response_status_count/200': 3,
         'finish_reason': 'finished',
         'finish_time': datetime.datetime(2012, 7, 23, 18, 50, 39, 674965),
         'scheduler/memory_enqueued': 2,
         'start_time': datetime.datetime(2012, 7, 23, 18, 50, 37, 700711)}
2012-07-23 18:50:39+0000 [hushbabies] INFO: Spider closed (finished)
2012-07-23 18:50:39+0000 [scrapy] INFO: Dumping global stats:
        {'memusage/max': 27820032, 'memusage/startup': 27652096}

Changing the site from hushbabies.com to others will make the code to work.

Answer 1

It seems there is a problem in the underlying SGML parser in SgmlLinkExtractor, sgmllib.

The following code returns zero links:

>>> from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
>>> fetch('http://www.hushbabies.com/')
>>> len(SgmlLinkExtractor().extract_links(response))
0

You can try an alternative link extractor from Slybot which depends on Scraply:

>>> from slybot.linkextractor import LinkExtractor
>>> from scrapely.htmlpage import HtmlPage
>>> p = HtmlPage(body=response.body_as_unicode())
>>> sum(1 for _ in LinkExtractor().links_to_follow(p))
314