def parse(self):
doc = lxml_parse(StringIO(self.f.read()), base_url=self.f.geturl()).getroot()
if doc is None:
return None
# Make all relative links absolute
doc.make_links_absolute()
links = list(doc.cssselect("a"))
logging.debug("(TID:%d) %s: %d links" % (self.thread_id, "links_"+self.f.geturl(), len(links)))
for link in links:
if not link.get("href"):
continue
scheme, netloc, path, query, fragment = urlsplit(link.get("href"))
if scheme.lower() in ["http"]:
# Add the link to the queue to be processed
URLHandler.add_url(self.db, link.get("href"))
# Add the link to the page's link set
self.db.sadd("links_"+self.f.geturl(), link.get("href"))
def parse(self):
doc = lxml_parse(StringIO(self.f.read()),
base_url=self.f.geturl()).getroot()
if doc is None:
return None
# Make all relative links absolute
doc.make_links_absolute()
links = list(doc.cssselect("a"))
logging.debug("(TID:%d) %s: %d links" %
(self.thread_id, "links_" + self.f.geturl(), len(links)))
for link in links:
if not link.get("href"):
continue
scheme, netloc, path, query, fragment = urlsplit(link.get("href"))
if scheme.lower() in ["http"]:
# Add the link to the queue to be processed
URLHandler.add_url(self.db, link.get("href"))
# Add the link to the page's link set
self.db.sadd("links_" + self.f.geturl(), link.get("href"))
def grab_content(url, original_description):
logging.debug("grabbing the content of %s" % url)
try:
site = urlopen(url)
except Exception as e:
logging.debug("Error: can't access %s for grabbing: %s" % (url, e))
return original_description + "\n<p><b>WarGod error</b>: I could not access this url</p>"
site_url = "/".join(site.geturl().split("/")[:3]) + "/"
path = get_grabber(url)
logging.debug("xpath query is %s" % path)
try:
xml = lxml_parse(site)
content = xml.xpath(path)[0]
except Unparseable:
logging.debug("Error: can't parse %s using %s" % (url, path))
return original_description + "\n<p><b>WarGod error</b>: I could not parse this url</p>"
except IndexError:
logging.debug("Error: can't get content of %s using %s" % (url, path))
return original_description + "\n<p><b>WarGod error</b>: I could not parse this url</p>"
except Exception as e:
sys.stderr.write("Unpredicted Error: can't get content of %s using %s" % (url, path))
return original_description + "\n<p><b>WarGod error</b>: I could not parse this url</p>"
content.make_links_absolute(site_url)
return etree.tostring(content, encoding="Utf-8").decode("Utf-8")
def parse(self, parser=None, base_url=None):
"""Parses the underlying html source using `lxml` library.
This parsed tree is stored in :attr:`root` of this object.
which could be used to perform numerous operations.
Returns
-------
ElementTree
"""
utx = self._get_utx()
assert utx is not None, "UrlTransformer not Implemented." # internal error
assert utx.base_path is not None, "Base Path is not set!"
assert utx.base_url is not None, "Base url is not Set!"
if not isinstance(parser, HTMLParser):
TypeError("Expected instance of <%r>, got <%r>" %
(HTMLParser, parser))
if not parser:
parser = HTMLParser(encoding=self.encoding, collect_ids=False)
source = self.get_source()
assert source is not None, "Source is not Set!"
assert hasattr(source, 'read'), "File like object is required!"
# assert self._element_factory is not None
# assert hasattr(self._element_factory, 'make_element')
LOGGER.info(
'Parsing tree with source: <%r> encoding <%s> and parser <%r>' %
(self._source, self.encoding, parser))
context_tree = lxml_parse(source, parser=parser, base_url=base_url)
# The tree generated by the parse is stored in the self.root
# variable and can be utilised further for any number of use cases
self._tree = context_tree
self.root = context_tree.getroot()
if self.root is not None:
# WaterMarking :)
self.root.insert(
0, Comment(MARK.format('', __version__, utx.url, utc_now(),
'')))
# There are internal links present on the html page which are files
# that includes `#` and `javascript:` and 'data:base64;` type links
# or a simple `/` url referring anchor tag
# thus these links needs to be left as is.
factory = getattr(self, 'make_element', None)
assert callable(factory), "Element generator is not callable!"
# Modify the tree elements
for el in context_tree.iter():
# A element can contain multiple urls
for pack in self._handle_lxml_elem(el):
if pack is not None:
elem, attr, url, pos = pack
else: # pragma: no cover
continue
if elem is not None:
o = factory(elem, attr, url, pos)
if o is not None:
self._stack.append(o)
self._parseComplete = True
return self.root
