def _extract_links(self, selector, response_url, response_encoding, base_url):
links = []
# hacky way to get the underlying lxml parsed document
for el, attr, attr_val in self._iter_links(selector._root):
if not self._is_valid_link(el, attr, attr_val):
continue
attr_val = urljoin(base_url, attr_val)
url = self.process_attr(attr_val)
if url is None:
continue
if isinstance(url, unicode):
url = url.encode(response_encoding)
# to fix relative links after process_value
url = urljoin(response_url, url)
if el.tag != 'a':
link = AssetLink(url, _collect_string_content(el) or u'',
nofollow=True if el.get('rel') == 'nofollow' else False)
else:
link = PageLink(url, _collect_string_content(el) or u'',
nofollow=True if el.get('rel') == 'nofollow' else False)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.unique else links
def _extract_links(self, selector, response_url, response_encoding,
base_url):
links = []
# hacky way to get the underlying lxml parsed document
for el, attr, attr_val in self._iter_links(selector._root):
# pseudo lxml.html.HtmlElement.make_links_absolute(base_url)
try:
attr_val = urljoin(base_url, attr_val)
except ValueError:
continue # skipping bogus links
else:
url = self.process_attr(attr_val)
if url is None:
continue
if isinstance(url, unicode):
url = url.encode(response_encoding)
# to fix relative links after process_value
url = urljoin(response_url, url)
link = Link(
url,
_collect_string_content(el) or u'',
nofollow=True if el.get('rel') == 'nofollow' else False)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.unique else links
def extract_links(self, response):
result = json.loads(response.text)
for pattern in self.patterns:
extractors = pattern.get('extractors')
format = pattern.get('format')
data = result
for extractor in extractors:
type = extractor.get('type')
if isinstance(data, dict):
if type == 'value':
data = self.get_value(*([data] +
extractor.get('args')))
elif isinstance(data, list):
if type == 'value':
data = [
self.get_value(*([item] + extractor.get('args')))
for item in data
]
elif type == 'slice':
data = self.get_slice(*([data] +
extractor.get('args')))
if not isinstance(data, list):
data = [data]
all_links = [
Link(response.urljoin(format.format(
*[item]))) if not isinstance(item, list) else Link(
response.urljoin(format.format(*item)))
for item in data
]
return unique_list(all_links)
def _extract_links(self, selector, response_url, response_encoding, base_url):
'''
Pretty much the same function, just added 'ignore' to url.encode
'''
links = []
# hacky way to get the underlying lxml parsed document
for el, attr, attr_val in self._iter_links(selector.root):
# pseudo lxml.html.HtmlElement.make_links_absolute(base_url)
try:
attr_val = urljoin(base_url, attr_val)
except ValueError:
continue # skipping bogus links
else:
url = self.process_attr(attr_val)
if url is None:
continue
if isinstance(url, unicode):
# add 'ignore' to encoding errors
url = url.encode(response_encoding, 'ignore')
# to fix relative links after process_value
url = urljoin(response_url, url)
link = Link(url, _collect_string_content(el) or u'',
nofollow=True if el.get('rel') == 'nofollow' else False)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.unique else links
def _process_links(self, links):
""" Normalize and filter extracted links
The subclass should override it if necessary
"""
links = unique_list(links, key=lambda link: link.url) if self.unique else links
return links
def _process_links(self, links):
""" Normalize and filter extracted links
The subclass should override it if neccessary
"""
links = unique_list(links, key=lambda link: link.url) if self.unique else links
return links
def extract_links(self, response):
if not self.base_url:
self.base_url = get_base_url(response)
items = re.findall(self.restrict_re, response.text)
all_links = [
Link(response.urljoin(self.base_url.format(str(item))))
for item in items
]
return unique_list(all_links)
def extract_links(self, response):
base_url = get_base_url(response)
if self.restrict_xpaths:
docs = [subdoc for x in self.restrict_xpaths for subdoc in response.xpath(x)]
else:
docs = [response.selector]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding, base_url)
all_links.extend(self._process_links(links))
return unique_list(all_links)
def _extract_links(self, response_text, response_url):
html = lxml.html.fromstring(response_text)
html.make_links_absolute(response_url)
for e, a, l, p in html.iterlinks():
if self.scan_tag(e.tag):
if self.scan_attr(a):
link = Link(self.process_attr(l), text=e.text)
self.links.append(link)
links = unique_list(self.links, key=lambda link: link.url) if self.unique else self.links
return links
def _extract_links(self, response_text, response_url):
html = lxml.html.fromstring(response_text)
html.make_links_absolute(response_url)
for e, a, l, p in html.iterlinks():
if self.scan_tag(e.tag):
if self.scan_attr(a):
link = Link(self.process_attr(l), text=e.text)
self.links.append(link)
links = unique_list(self.links, key=lambda link: link.url) \
if self.unique else self.links
return links
def extract_links(self, response):
base_url = get_base_url(response)
if self.restrict_xpaths:
docs = [subdoc
for x in self.restrict_xpaths
for subdoc in response.xpath(x)]
else:
docs = [response.selector]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding, base_url)
all_links.extend(self._process_links(links))
return unique_list(all_links)
def _extract_links(self, response_text, response_url, response_encoding):
self.base_url, self.links = etree.HTML(response_text, self.parser)
links = unique_list(self.links, key=lambda link: link.url) if self.unique else self.links
ret = []
base_url = urljoin_rfc(response_url, self.base_url) if self.base_url else response_url
for link in links:
link.url = urljoin_rfc(base_url, link.url, response_encoding)
link.url = safe_url_string(link.url, response_encoding)
link.text = str_to_unicode(link.text, response_encoding)
ret.append(link)
return ret
def extract_links(self, response):
html = Selector(response)
try:
base_url = response.xpath("//base/@href").extract()[0]
except IndexError:
base_url = get_base_url(response)
if self.restrict_xpaths:
docs = [subdoc for x in self.restrict_xpaths for subdoc in html.xpath(x)]
else:
docs = [html]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding, base_url)
all_links.extend(self._process_links(links))
return unique_list(all_links)
def merge_clusters(merged, all_urls, min_cluster_size):
res = {'clusters': {}, 'unclustered': []}
unclustered = all_urls
for regex in merged:
matches = apply_reg_ex_to_urls(regex, all_urls)
matches = unique_list(matches)
if len(matches) >= min_cluster_size:
for match in matches:
try:
unclustered.remove(match)
except:
continue
human = regex.replace('([^/]+)', '[...]').replace('([^&=?]+)', '[...]').replace('(\d+)', '[NUMBER]')
res['clusters'].update({(regex, human): matches})
res['unclustered'] = unclustered
return res
def _extract_links(self, response_text, response_url, response_encoding):
self.reset()
self.feed(response_text)
self.close()
links = unique_list(self.links, key=lambda link: link.url) if self.unique else self.links
ret = []
base_url = urljoin_rfc(response_url, self.base_url) if self.base_url else response_url
for link in links:
link.url = urljoin_rfc(base_url, link.url, response_encoding)
link.url = safe_url_string(link.url, response_encoding)
link.text = link.text.decode(response_encoding)
ret.append(link)
return ret
def get_urls(_url, _html=None, headers=None):
if _html is None:
response = requests.get(_url, verify=False, headers=headers)
_html = response.content
page = str(BeautifulSoup(_html))
url_list = []
while True:
url, n = get_url(page)
page = page[n:]
if url:
url = urljoin(_url, url)
url_list.append(url)
else:
break
url_list = unique_list(url_list)
return url_list
def _extract_links(self, response_text, response_url, response_encoding):
self.reset()
self.feed(response_text)
self.close()
links = unique_list(self.links, key=lambda link: link.url) if self.unique else self.links
ret = []
base_url = self.base_url if self.base_url else response_url
for link in links:
link.url = urljoin_rfc(base_url, link.url, response_encoding)
link.url = safe_url_string(link.url, response_encoding)
link.text = link.text.decode(response_encoding)
ret.append(link)
return ret
def extract_links(self, response):
base_url = get_base_url(response)
if self.restrict_xpaths:
links = [link
for xpath in self.restrict_xpaths
for link in response.xpath(xpath)]
else:
links = [response.selector,]
all_links = [Link(response.url),]
for link in links:
new_link = self._extract_links(link, response.url, response.encoding, base_url)
all_links.extend(self._process_links(new_link))
return unique_list(all_links)
def _extract_links(self, response_text, response_url, response_encoding):
links = []
html = lxml.html.fromstring(response_text)
html.make_links_absolute(response_url)
for e, a, l, p in html.iterlinks():
if self.tag_func(e.tag):
if self.attr_func(a):
l = safe_url_string(l, response_encoding)
text = u''
if e.text:
text = str_to_unicode(e.text, response_encoding, errors='replace').strip()
link = Link(self.process_func(l), text=text)
links.append(link)
links = unique_list(links, key=lambda link: link.url) \
if self.unique else links
return links
def _extract_links(self, response_text, response_url, response_encoding):
self.base_url, self.links = etree.HTML(response_text, self.parser)
links = unique_list(
self.links,
key=lambda link: link.url) if self.unique else self.links
ret = []
base_url = urljoin_rfc(
response_url, self.base_url) if self.base_url else response_url
for link in links:
link.url = urljoin_rfc(base_url, link.url, response_encoding)
link.url = safe_url_string(link.url, response_encoding)
link.text = str_to_unicode(link.text,
response_encoding,
errors='replace')
ret.append(link)
return ret
def _extract_links(self, selector, response_url, response_encoding, base_url):
links = []
# hacky way to get the underlying lxml parsed document
for el, attr, attr_val in self._iter_links(selector._root):
# pseudo lxml.html.HtmlElement.make_links_absolute(base_url)
attr_val = urljoin(base_url, attr_val)
url = self.process_attr(attr_val)
if url is None:
continue
if isinstance(url, unicode):
url = url.encode(response_encoding, 'ignore')
# to fix relative links after process_value
url = urljoin(response_url, url)
link = Link(url, _collect_string_content(el) or u'',
nofollow=True if el.get('rel') == 'nofollow' else False)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.unique else links
def extract_links(self, response): base_url = get_base_url(response) print "base_url", base_url domain_name=tldextract.extract(base_url).domain if domain_name in self.crawledPagesPerSite and self.crawledPagesPerSite[domain_name]>self.maximumPagesPerSite: return [] if self.restrict_xpaths: docs = [subdoc for x in self.restrict_xpaths for subdoc in response.xpath(x)] else: docs = [response.selector] all_links = [] for doc in docs: links = self._extract_links(doc, response.url, response.encoding, base_url) all_links.extend(self._process_links(links)) all_links=unique_list(all_links) new_all_links=[] for link in all_links: url=link.url domain_name=tldextract.extract(url).domain suffix=tldextract.extract(url).suffix domain_and_suffix=domain_name+"."+suffix if domain_and_suffix not in self.allow_domains: continue if domain_name in self.crawledPagesPerSite: self.crawledPagesPerSite[domain_name]+=1 else: self.crawledPagesPerSite[domain_name]=1 if self.crawledPagesPerSite[domain_name]>self.maximumPagesPerSite: break else: print "have crawled " , self.crawledPagesPerSite[domain_name], "pages" new_all_links.append(link) return new_all_links
def _extract_links(self, response_text, response_url, response_encoding):
self.reset()
self.feed(response_text)
self.close()
links = unique_list(self.links, key=lambda link: link.url) if self.unique else self.links
ret = []
base_url = urljoin(response_url, self.base_url) if self.base_url else response_url
for link in links:
if isinstance(link.url, unicode):
link.url = link.url.encode(response_encoding)
try:
link.url = urljoin(base_url, link.url)
except ValueError:
continue
link.url = safe_url_string(link.url, response_encoding)
link.text = link.text.decode(response_encoding)
ret.append(link)
return ret
def extract_links(self, response):
# TODO: remove debug code
with open('/export/home/asanakoy/tmp/response.txt', 'w') as f:
f.write(response.body)
assert False, 'enough ;)'
base_url = self.base_url if self.base_url else get_base_url(response)
if self.restrict_xpaths:
docs = [
subdoc for x in self.restrict_xpaths
for subdoc in response.xpath(x)
]
else:
docs = [response.selector]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding,
base_url)
print 'Num links before filter:', len(links)
all_links.extend(self._process_links(links))
print 'Num links:', len(all_links)
return unique_list(all_links)
def extract_links(self, response):
"""Returns a list of :class:`~scrapy.link.Link` objects from the
specified :class:`response <scrapy.http.Response>`.
Only links that match the settings passed to the ``__init__`` method of
the link extractor are returned.
Duplicate links are omitted.
"""
base_url = get_base_url(response)
if self.restrict_xpaths:
docs = [subdoc
for x in self.restrict_xpaths
for subdoc in response.xpath(x)]
else:
docs = [response.selector]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding, base_url)
all_links.extend(self._process_links(links))
return unique_list(all_links)
def extract_links(self, response):
"""Returns a list of :class:`~scrapy.link.Link` objects from the
specified :class:`response <scrapy.http.Response>`.
Only links that match the settings passed to the ``__init__`` method of
the link extractor are returned.
Duplicate links are omitted.
"""
base_url = get_base_url(response)
if self.restrict_xpaths:
docs = [
subdoc
for x in self.restrict_xpaths #大循环 有多少条restrict_xpaths
for subdoc in response.xpath(x) #小循环 符合这个xpath的条目的 有多少个字条目
]#[a for i in range(3) for a in range(10)]》》》[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
else:
docs = [response.selector]
all_links = []
for doc in docs:
links = self._extract_links(doc, response.url, response.encoding, base_url) #实际上是调用 LxmlParserLinkExtractor._extract_links
all_links.extend(self._process_links(links))
return unique_list(all_links)
def _extract_links(self, selector, response_url, response_encoding, base_url):
links = []
# hacky way to get the underlying lxml parsed document
for el, attr, attr_val in self._iter_links(selector.root):
if self.scan_tag(el.tag) and self.scan_attr(attr):
# pseudo root.make_links_absolute(base_url)
# START PATCH: Added check to filter links before making absolute
if not _is_valid_link(attr_val):
continue
# END PATCH
attr_val = urljoin(base_url, attr_val)
url = self.process_attr(attr_val)
if url is None:
continue
# to fix relative links after process_value
url = urljoin(response_url, url)
link = Link(
url, _collect_string_content(el) or '',
nofollow=True if el.get('rel') == 'nofollow' else False
)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.unique else links
def _extract_from_js(self, doc, response_url, response_encoding, base_url):
data = doc.re('siblings:\s*(?P<data>.*)\s*\, registryURL')
if not data:
return []
links = []
articles = json.loads(data[0])
for article in articles['articleList']:
try:
attr_val = urljoin(base_url, article['uri'])
except ValueError:
continue # skipping bogus links
else:
url = self.link_extractor.process_attr(attr_val)
if url is None:
continue
if isinstance(url, unicode):
url = url.encode(response_encoding)
url = urljoin(response_url, url)
link = Link(url, u'', nofollow=False)
links.append(link)
return unique_list(links, key=lambda link: link.url) \
if self.link_extractor.unique else links
def _deduplicate_if_needed(self, links):
if self.unique:
return unique_list(links, key=lambda link: link.url)
return links
def extract_links(self, response):
links = super(RmDupliLinkExtractor, self).extract_links(response)
base_url = get_base_url(response)
return unique_list([link for link in links if not (link.url.startswith(base_url) and base_url.endswith(link.url[len(base_url):]))])
def _deduplicate_if_needed(self, links):
if self.unique:
return unique_list(links, key=self.link_key)
return links
def _process_links(self, links):
""" Normalize and filter extracted links
The subclass should override it if necessary
"""
return unique_list(links, key=self.link_key) if self.unique else links
def _process_links(self, links):
""" Normalize and filter extracted links
The subclass should override it if necessary
"""
return unique_list(links, key=self.link_key) if self.unique else links
def _deduplicate_if_needed(self, links):
if self.unique:
return unique_list(links, key=self.link_key)
return links
def extract_links(self, response):
all_links = []
for json_path in self.json_paths:
links = self._extract_links(json_path, response)
all_links.extend(self._process_links(links))
return unique_list(all_links)
def _deduplicate_if_needed(self, links):
if self.unique:
return unique_list(links, key=lambda link: link.url)
return links
