class BaseSpider(CrawlSpider):
name = "Base"
site__id = None
extractors = None
allowed_domains = []
start_urls = []
session = None
rules = (
Rule(RegexLinkExtractor(),callback='parse_item'),
Rule(SgmlLinkExtractor(),callback='parse_item'),
Rule(LxmlParserLinkExtractor(),callback='parse_item'),
)
def process_results(self, response, results):
return chain(results, self.parse_item(response))
def parse_item(self, response):
for extractor in self.extractors:
values = {
'URL_PROD': response.url,
}
extract = {}
for e in extractor(response):
extract.update(e) # TODO: check relevance if overwriting
for k,v in extract.iteritems():
values[k] = v[0]
name = values.get('NAME_PROD')
if name:
yield ScraperItem(name=name, site=self.site__id, values=values.iteritems())
def __init__(self, skip_text=False):
lx = LxmlParserLinkExtractor()
self.linkextractor = FilteringLinkExtractor(lx,
allow=(),
deny=(),
allow_domains=(),
deny_domains=(),
restrict_xpaths=(),
canonicalize=True,
deny_extensions=None)
self.skip_text = skip_text
def parse(self, response):
ret = []
title = "".join(
response.xpath(".//span[@id='sites-page-title']/text()").extract())
# print "Page Title", title
url = response.url
path = url.replace(" ", "").replace("/", "_").replace(":", "")
content = response.xpath(".//*[@id='sites-canvas-main-content']/table")
links = []
domains = []
#print "Page Content", content
l = LxmlParserLinkExtractor()
for e in content:
for x in e.xpath('//*//@href'):
y = x.extract()
links.append(y)
domres = tldextract.extract(y)
sd = domres.subdomain
dm = domres.domain
tld = domres.suffix
if dm:
#print domres
d = "{0}.{1}.{2}".format(sd, dm, tld)
if d not in domains:
domains.append(d)
content = "".join(content.extract())
p = Page(title=title,
path=path,
url=url,
content=content,
links=links,
domains=domains)
ret.append(p)
# spider the other local pages
for sel in response.xpath('//ul/li/div'):
title = sel.xpath('a/text()').extract()
link = sel.xpath('a/@href').extract()
desc = sel.xpath('text()').extract()
if link:
l = self.base + link[0]
#print title, l, desc
ret.append(scrapy.http.Request(l))
return ret
