def _create_link_from_element(
anchor, # type: HTMLElement
page_url, # type: str
base_url, # type: str
):
# type: (...) -> Optional[Link]
"""
Convert an anchor element in a simple repository page to a Link.
"""
href = anchor.get("href")
if not href:
return None
url = _clean_link(urllib_parse.urljoin(base_url, href))
pyrequire = anchor.get('data-requires-python')
pyrequire = unescape(pyrequire) if pyrequire else None
yanked_reason = anchor.get('data-yanked')
if yanked_reason:
# This is a unicode string in Python 2 (and 3).
yanked_reason = unescape(yanked_reason)
link = Link(
url,
comes_from=page_url,
requires_python=pyrequire,
yanked_reason=yanked_reason,
)
return link
def links(self):
"""
Return the URLs of all the links on a page together with information
about their "rel" attribute, for determining which ones to treat as
downloads and which ones to queue for further scraping.
"""
def clean(url):
"Tidy up an URL."
scheme, netloc, path, params, query, frag = urlparse(url)
return urlunparse((scheme, netloc, quote(path),
params, query, frag))
result = set()
for match in self._href.finditer(self.data):
d = match.groupdict('')
rel = (d['rel1'] or d['rel2'] or d['rel3'] or
d['rel4'] or d['rel5'] or d['rel6'])
url = d['url1'] or d['url2'] or d['url3']
url = urljoin(self.base_url, url)
url = unescape(url)
url = self._clean_re.sub(lambda m: '%%%2x' % ord(m.group(0)), url)
result.add((url, rel))
# We sort the result, hoping to bring the most recent versions
# to the front
result = sorted(result, key=lambda t: t[0], reverse=True)
return result
def links(self):
"""
Return the URLs of all the links on a page together with information
about their "rel" attribute, for determining which ones to treat as
downloads and which ones to queue for further scraping.
"""
def clean(url):
"Tidy up an URL."
scheme, netloc, path, params, query, frag = urlparse(url)
return urlunparse(
(scheme, netloc, quote(path), params, query, frag))
result = set()
for match in self._href.finditer(self.data):
d = match.groupdict('')
rel = (d['rel1'] or d['rel2'] or d['rel3'] or d['rel4']
or d['rel5'] or d['rel6'])
url = d['url1'] or d['url2'] or d['url3']
url = urljoin(self.base_url, url)
url = unescape(url)
url = self._clean_re.sub(lambda m: '%%%2x' % ord(m.group(0)), url)
result.add((url, rel))
# We sort the result, hoping to bring the most recent versions
# to the front
result = sorted(result, key=lambda t: t[0], reverse=True)
return result
def links(self):
"""Yields all links in the page"""
for anchor in self.parsed.findall(".//a"):
if anchor.get("href"):
href = anchor.get("href")
url = self.clean_link(urllib_parse.urljoin(
self.base_url, href))
pyrequire = anchor.get('data-requires-python')
pyrequire = unescape(pyrequire) if pyrequire else None
yield Link(url, self, requires_python=pyrequire)
def links(self):
"""Yields all links in the page"""
for anchor in self.parsed.findall(".//a"):
if anchor.get("href"):
href = anchor.get("href")
url = self.clean_link(
urllib_parse.urljoin(self.base_url, href)
)
pyrequire = anchor.get('data-requires-python')
pyrequire = unescape(pyrequire) if pyrequire else None
yield Link(url, self, requires_python=pyrequire)
def iter_links(self):
"""Yields all links in the page"""
document = html5lib.parse(
self.content,
transport_encoding=_get_encoding_from_headers(self.headers),
namespaceHTMLElements=False,
)
base_url = _determine_base_url(document, self.url)
for anchor in document.findall(".//a"):
if anchor.get("href"):
href = anchor.get("href")
url = _clean_link(urllib_parse.urljoin(base_url, href))
pyrequire = anchor.get('data-requires-python')
pyrequire = unescape(pyrequire) if pyrequire else None
yield Link(url, self.url, requires_python=pyrequire)
def iter_links(self):
"""Yields all links in the page"""
document = html5lib.parse(
self.content,
transport_encoding=_get_encoding_from_headers(self.headers),
namespaceHTMLElements=False,
)
base_url = _determine_base_url(document, self.url)
for anchor in document.findall(".//a"):
if anchor.get("href"):
href = anchor.get("href")
url = _clean_link(urllib_parse.urljoin(base_url, href))
pyrequire = anchor.get('data-requires-python')
pyrequire = unescape(pyrequire) if pyrequire else None
yield Link(url, self.url, requires_python=pyrequire)
