def build(self, url=None, wanted_list=None, html=None, request_args=None):
"""
Automatically constructs a set of rules to scrape the specified target[s] from a web page.
The rules are represented as stack_list.
Parameters:
----------
url : str, optional
URL of the target web page. You should either pass url or html.
wanted_list : list, optional
A list of needed contents to be scraped.
AutoScraper learns a set of rules to scrape these targets.
html : str, optional
An HTML string can also be passed instead of URL.
You should either pass url or html.
request_args : dict, optional
A dictionary used to specify a set of additional request parameters used by requests
module. You can specify proxy URLs, custom headers etc.
Returns:
--------
None
"""
self.url = url
soup = self._get_soup(url=url, html=html, request_args=request_args)
result_list = []
stack_list = []
for wanted in wanted_list:
children = self._get_children(soup, wanted)
for child in children:
result, stack = self._get_result_for_child(child, soup)
result_list += result
stack_list.append(stack)
result_list = unique(result_list)
if all(w in result_list for w in wanted_list):
self.stack_list = unique(stack_list)
return result_list
return None
def build(self, url=None, wanted_list=None, html=None, request_args=None):
self.url = url
soup = self._get_soup(url=url, html=html, request_args=request_args)
result_list = []
stack_list = []
for wanted in wanted_list:
children = self._get_children(soup, wanted)
for child in children:
result, stack = self._get_result_for_child(child, soup)
result_list += result
stack_list.append(stack)
result_list = unique(result_list)
if all(w in result_list for w in wanted_list):
self.stack_list = unique(stack_list)
return result_list
return None
def get_result_similar(self,
url=None,
html=None,
soup=None,
request_args=None):
if url:
self.url = url
if not soup:
soup = self._get_soup(url=url,
html=html,
request_args=request_args)
result = []
for stack in self.stack_list:
result += self._get_result_with_stack(stack, soup)
return unique(result)
def _get_result_by_func(self, func, url, html, soup, request_args,
grouped):
if url:
self.url = url
if not soup:
soup = self._get_soup(url=url,
html=html,
request_args=request_args)
result_list = []
grouped_result = defaultdict(list)
for stack in self.stack_list:
result = func(stack, soup)
if not grouped:
result_list += result
continue
stack_id = stack['stack_id']
grouped_result[stack_id] += result
return dict(grouped_result) if grouped else unique(result_list)
def get_result_exact(self,
url=None,
html=None,
soup=None,
request_args=None):
if url:
self.url = url
if not soup:
soup = self._get_soup(url=url,
html=html,
request_args=request_args)
result = []
for stack in self.stack_list:
try:
result.append(
self._get_result_with_stack_index_based(stack, soup))
except IndexError:
continue
return unique(result)
