def get_sites_by_seed_sites(account: MajesticCom, seed_domains: [], catagories: [], fresh_data=False, index=0,
iteration=1, loop_count=0, count_per_domain=100, callback=None, current_count=0,
max_count=-1, tf=20) -> []:
if iteration < 0:
raise ValueError("get_sites_by_seed_sites: iteration should >= 0.")
sub_domains = [LinkChecker.get_root_domain(x, use_www=False)[4] for x in seed_domains[index:]]
if len(sub_domains) == 0:
return []
backlinks = []
# counter = index
if max_count > 0 and current_count >= max_count:
return backlinks
temp_sub_domains = []
temp = []
# target_catagories = []
# for catagory in catagories:
# target_catagories.append(str(CategoryManager.decode_sub_category(catagory, False)))
for sub_domain in sub_domains:
print("doing backlinks of domain:", sub_domain, "seed len:", len(temp_sub_domains))
try:
temp = account.get_backlinks(sub_domain, count_per_domain, topic="", is_dev=False,
fresh_data=fresh_data)
current_count += 1
except Exception as ex:
print(ex)
for item in temp:
if isinstance(item, MajesticBacklinkDataStruct):
# item_catagory = str(CategoryManager.decode_sub_category(item.src_topic, False))
domain = LinkChecker.get_root_domain(item.backlink, use_www=False)[4]
item.ref_domain = domain
# if callback is not None:
# callback(item)
# if len(target_catagories) > 0 and item_catagory not in target_catagories:
# continue
if domain not in sub_domains and domain not in temp_sub_domains:
if len(catagories) > 0:
is_in = False
if len(item.src_topic) > 0:
decoded = str(CategoryManager.decode_sub_category(item.src_topic, False))
for cate in catagories:
if cate in decoded:
is_in = True
break
if is_in and item.src_tf >= tf:
temp_sub_domains.append(domain)
elif item.src_tf >= tf:
temp_sub_domains.append(domain)
item.ref_domain = domain
if callback is not None:
callback(item)
if max_count > 0 and current_count >= max_count:
break
if loop_count >= iteration:
return backlinks
else:
return backlinks + GoogleMajestic.get_sites_by_seed_sites(account, sub_domains + temp_sub_domains, catagories, fresh_data, len(seed_domains),
iteration, loop_count+1, count_per_domain, callback, current_count, max_count, tf)
def get_sites(keyword: str, page_number: int=1, result_per_page: int=100,
index: int=0, length: int=100, use_browser=False,
source_type="", filter_list=[], country_code='', return_domain_home_only=True, days_ago=0,
**kwargs) -> []:
# def get_sites(keyword: str, page: int=1, index: int=0, length: int=100,
# history=SeedSiteSettings.TIME_NOW, blog=False) -> []:
assert page_number > 0, "page number should greater than 0"
assert index >= 0, "index should greater or equal to 0"
assert length > 0, "length should greater than 0"
search_query = BingConst.SearchLink.format(quote(keyword), quote(keyword), (page_number-1)*length + index + 1, length)
user_agent = WebRequestCommonHeader.webpage_agent
try:
req = BingCom._get_response(request_link=search_query, user_agent=user_agent, **kwargs)
# req = requests.get(search_query, timeout=30, headers=WebRequestCommonHeader.get_html_header())
result = req.text
soup = bs4.BeautifulSoup(result)
tags = soup.select(BingConst.SitePath)
domains = []
for tag in tags:
try:
domain = tag.attrs["href"].strip().replace(" ", "")
if return_domain_home_only:
domain = LinkChecker.get_root_domain(domain, use_www=False)[2] # get the link
else:
domain = LinkChecker.get_root_domain(domain, use_www=False)[3]
if len(domain) > 0:
domains.append(domain)
except:
pass
new_list = []
if isinstance(domains, list):
if len(filter_list) > 0:
for domain in domains:
if isinstance(domain, str):
temp = domain.lower().strip()
if not any(x in temp for x in filter_list):
new_list.append(temp)
else:
new_list = domains
end = index + length
data_len = len(new_list)
if domains is not None and index < data_len:
if data_len >= end:
return new_list[index:end]
else:
return new_list[index:]
else:
return []
except Exception as ex:
print(ex)
return None
def get_search_results(keyword: str, page_number: int, proxy: ProxyStruct=None, result_per_page: int=GoogleConst.Result100, timeout=5,
return_domain_home_only=True, use_forbidden_filter=True, days_ago=0, addtional_query_parameter: str="",
country_code="us", use_browser=False) -> list:
"""
generic normal search, get a list of domains form page
:param keyword:
:param page_number: > 0
:param resultPerPage:
:param timeout:
:param return_domain_home_only: return root domain name if True, else return protocol suffix + domain name
:param use_forbidden_filter:
:param days_ago: specify how many days ago before when results were indexed.
:return:
"""
assert page_number > 0, "page number should be greater than 0."
page_range = GoogleCom.get_result_per_page_range()
assert result_per_page in page_range, "result per page should be one of those values:" + str(page_range)
sub_domain = "www"
request_link = GoogleUtility.get_local_endpoint(country_code, sub_domain) \
+ GoogleConst.CommonSearchPath.format(quote(keyword), result_per_page, (page_number - 1) * result_per_page, country_code) \
+ addtional_query_parameter+GoogleUtility.get_query_for_days(days_ago)
try:
user_agent = WebRequestCommonHeader.webpage_agent
if not use_browser:
response = GoogleCom._get_response(request_link, proxy=proxy, timeout=timeout, user_agent=user_agent)
if not response.status_code == 200:
# if response.status_code == 503:
# print(response.text)
raise ConnectionRefusedError("error getting result, with status code:", response.status_code)
result = response.text
else:
result = GoogleCom._get_response_browser(request_link, proxy=proxy, timeout=timeout, user_agent=user_agent)
soup = bs4.BeautifulSoup(result)
tags = soup.select(GoogleConst.SitePath)
domains = []
for tag in tags:
try:
domain = tag.text.strip().replace(" ", "")
if return_domain_home_only:
domain = LinkChecker.get_root_domain(domain, use_www=False)[2] # get the link
else:
domain = LinkChecker.get_root_domain(domain, use_www=False)[3]
if use_forbidden_filter and LinkChecker.is_domain_forbidden(domain):
continue
if len(domain) > 0:
domains.append(domain)
except:
pass
return domains
except Exception as ex:
print(ex)
return None
def _check_whois_v1(self, domain_data: OnSiteLink):
root_domain = domain_data.link
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
real_response_code = domain_data.response_code
whois = LinkChecker.check_whois(root_domain) # check whois record
if whois[0]:
if whois[2]: # domain is expired
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
if real_response_code == ResponseCode.Expired:
#if ResponseCode.domain_might_be_expired(real_response_code):
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# if isinstance(self._queue_lock, multiprocessing.RLock):
with self._queue_lock:
self._output_q.put(
(domain_data.link, domain_data.response_code))
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex,
"_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def check_whois_with_dns(page: OnSiteLink):
real_response_code = ResponseCode.DNSError
skip_whois_check = False
try:
root_result = LinkChecker.get_root_domain(page.link)
root_domain = root_result[1]
sub_domain = root_result[4]
suffix = root_result[5]
if len(sub_domain) == 0 or suffix not in TldUtility.TOP_TLD_LIST:
skip_whois_check = True
else:
if LinkChecker.is_domain_DNS_OK(sub_domain): # check DNS first
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
elif not sub_domain.startswith("www."):
if LinkChecker.is_domain_DNS_OK("www." + root_domain):
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
# response = LinkChecker.get_response(page.link, timeout) # check 404 error
page.response_code = real_response_code
page.link_type = OnSiteLink.TypeOutbound
page.link = root_domain
except Exception as ex:
# ErrorLogger.log_error("WhoisChecker", ex, "_check_whois_with_dns() " + page.link)
skip_whois_check = True
finally:
if not skip_whois_check and real_response_code == ResponseCode.DNSError:
return check_whois(page)
else:
return page.link, page.response_code
def _check_whois(self, domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if not self._is_debug:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(
domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(
root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
self._put_output_result_in_queue(domain_data)
else:
self._put_output_result_in_queue(domain_data)
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex,
"_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def testScrapePageBatch(self):
save_path = "/Users/superCat/Desktop/PycharmProjectPortable/test/profile_archive_downloaded.csv"
file_path = "/Users/superCat/Desktop/PycharmProjectPortable/test/profile_test_links.txt"
CsvLogger.log_to_file_path(save_path, [ArchiveDetail.get_title()])
domains_links = FileHandler.read_lines_from_file(file_path)
for link in domains_links:
# link = "http://web.archive.org/web/20140711025724/http://susodigital.com/"
#link ="http://web.archive.org/web/20130415001342/http://www.bbc.co.uk/"
stop_event = multiprocessing.Event()
inner_link, domain, path, link_class, ext, fragment = LinkUtility.get_link_detail(
link)
root_domain = LinkChecker.get_root_domain(domain)[1]
path = "/index.html"
link_s = LinkAttrs(link=link,
path=path,
ref_link="/",
shadow_ref_link="/",
source=path,
res_type=LinkUtility.EXT_WEBPAGE,
level=0)
explorer = ArchiveExplorer(
original_domain=root_domain,
link=link,
external_stop_event=stop_event,
download_base_dir=FilePath.get_default_archive_dir(),
max_thread=10,
max_level=2)
explorer.run()
archive_detail = explorer.get_archive_detail()
CsvLogger.log_to_file_path(save_path, [archive_detail.to_tuple()])
def check_whois_with_dns(page: OnSiteLink):
real_response_code = ResponseCode.DNSError
skip_whois_check = False
try:
root_result = LinkChecker.get_root_domain(page.link)
root_domain = root_result[1]
sub_domain = root_result[4]
suffix = root_result[5]
if len(sub_domain) == 0 or suffix not in TldUtility.TOP_TLD_LIST:
skip_whois_check = True
else:
if LinkChecker.is_domain_DNS_OK(sub_domain): # check DNS first
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
elif not sub_domain.startswith("www."):
if LinkChecker.is_domain_DNS_OK("www." + root_domain):
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
# response = LinkChecker.get_response(page.link, timeout) # check 404 error
page.response_code = real_response_code
page.link_type = OnSiteLink.TypeOutbound
page.link = root_domain
except Exception as ex:
# ErrorLogger.log_error("WhoisChecker", ex, "_check_whois_with_dns() " + page.link)
skip_whois_check = True
finally:
if not skip_whois_check and real_response_code == ResponseCode.DNSError:
return check_whois(page)
else:
return page.link, page.response_code
def test_get_sub_domains(self):
full_link = "http://blogspot.co.uk/"
domain_data = LinkChecker.get_root_domain(full_link, False)
root_domain = domain_data[1]
sub_domain = domain_data[4]
domain_suffix = domain_data[5]
sub_domain_no_local = sub_domain.strip(domain_suffix)
print(sub_domain_no_local)
def test_get_sub_domains(self):
full_link = "http://blogspot.co.uk/"
domain_data = LinkChecker.get_root_domain(full_link, False)
root_domain = domain_data[1]
sub_domain = domain_data[4]
domain_suffix = domain_data[5]
sub_domain_no_local = sub_domain.strip(domain_suffix)
print(sub_domain_no_local)
def testScrapePage(self):
# link = "http://web.archive.org/web/20111102054835/http://www.agfdh.org:80/"
link = "http://web.archive.org/web/20150425143742/http://susodigital.com/"
#link ="http://web.archive.org/web/20130415001342/http://www.bbc.co.uk/"
stop_event = multiprocessing.Event()
inner_link, domain, path, link_class, ext, fragment = LinkUtility.get_link_detail(link)
root_domain = LinkChecker.get_root_domain(domain)[1]
path = "/index.html"
link_s = LinkAttrs(link=link, path=path, ref_link="/", shadow_ref_link="/", source=path, res_type=LinkUtility.EXT_WEBPAGE, level=0)
explorer = ArchiveExplorer(original_domain=root_domain, link=link,
external_stop_event=stop_event,
download_base_dir=FilePath.get_default_archive_dir(), max_thread=10, max_level=2)
explorer.run()
save_path = "/Users/superCat/Desktop/PycharmProjectPortable/test/profile_archive_downloaded.csv"
CsvLogger.log_to_file_path(save_path, [ArchiveDetail.get_title()])
archive_detail = explorer.get_archive_detail()
CsvLogger.log_to_file_path(save_path, [archive_detail.to_tuple()])
def check_external_page(checker: SiteChecker, page: OnSiteLink, timeout=10):
"""
check DNS Error Only
:param checker:
:param page:
:param timeout:
:return:
"""
# response = LinkChecker.get_response(page.link, timeout)
#real_response_code = response[0]
#real_response_code = ResponseCode.LinkOK
#print("-------checking external " + page.link)
try:
root_result = LinkChecker.get_root_domain(page.link)
root_domain = root_result[1]
sub_domain = root_result[4]
if len(sub_domain) == 0 or root_domain in checker.external_cache_list:
return
else:
if len(checker.external_cache_list) < checker.external_cache_size:
checker.external_cache_list.append(root_domain)
real_response_code = page.response_code
if LinkChecker.is_domain_DNS_OK(sub_domain): # check DNS first
real_response_code = ResponseCode.NoDNSError
elif not sub_domain.startswith("www."):
if LinkChecker.is_domain_DNS_OK("www." + root_domain):
real_response_code = ResponseCode.NoDNSError
# response = LinkChecker.get_response(page.link, timeout) # check 404 error
page.response_code = real_response_code
page.link_type = OnSiteLink.TypeOutbound
page.link = root_domain
#print(" ready to output external:", str(page))
if checker.output_all_external or ResponseCode.domain_might_be_expired(real_response_code):
# if checker.delegate is not None:
# checker.delegate(new_page)
if checker.output_queue is not None:
with checker._queue_lock:
checker.output_queue.put(page)
except Exception as ex:
PrintLogger.print(ex)
ErrorLogger.log_error("PageChecker", ex, "check_external_page() " + page.link)
def check_whois(domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# self._output_q.put((domain_data.link, domain_data.response_code))
except Exception as ex:
print(ex)
finally:
return domain_data.link, domain_data.response_code
def check_whois(domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(
root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# self._output_q.put((domain_data.link, domain_data.response_code))
except Exception as ex:
print(ex)
finally:
return domain_data.link, domain_data.response_code
def _check_whois(self, domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if not self._is_debug:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
self._put_output_result_in_queue(domain_data)
else:
self._put_output_result_in_queue(domain_data)
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex, "_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def _get_back_link_thread(account: MajesticCom, sub_domain: str, count_per_domain: int, fresh_data: bool,
sub_domains: [], temp_sub_domains: [], categories: [], callback, tf=20, bad_country_list=[]):
temp = []
print("doing backlinks of domain:", sub_domain, " domain len:", len(temp_sub_domains))
try:
temp = account.get_backlinks(sub_domain, count_per_domain, topic="", is_dev=False, fresh_data=fresh_data)
except Exception as ex:
print(ex)
for item in temp:
if isinstance(item, MajesticBacklinkDataStruct):
# item_catagory = str(CategoryManager.decode_sub_category(item.src_topic, False))
domain = LinkChecker.get_root_domain(item.backlink, use_www=False)[4]
item.ref_domain = domain
# if callback is not None:
# callback(item)
# if len(target_catagories) > 0 and item_catagory not in target_catagories:
# continue
if domain not in sub_domains and domain not in temp_sub_domains:
if len(categories) > 0:
is_in = False
if len(item.src_topic) > 0:
decoded = str(CategoryManager.decode_sub_category(item.src_topic, False))
for cate in categories:
if cate in decoded:
is_in = True
break
if is_in and item.src_tf >= tf:
temp_sub_domains.append(domain)
elif item.src_tf >= tf:
temp_sub_domains.append(domain)
item.ref_domain = domain
if callback is not None:
callback(item)
time.sleep(1)
def _check_whois_v1(self, domain_data: OnSiteLink):
root_domain = domain_data.link
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
real_response_code = domain_data.response_code
whois = LinkChecker.check_whois(root_domain) # check whois record
if whois[0]:
if whois[2]: # domain is expired
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
if real_response_code == ResponseCode.Expired:
#if ResponseCode.domain_might_be_expired(real_response_code):
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# if isinstance(self._queue_lock, multiprocessing.RLock):
with self._queue_lock:
self._output_q.put((domain_data.link, domain_data.response_code))
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex, "_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def get_sites_by_seed_sites_muti_threads(account: MajesticCom, seed_domains: [], catagories: [], fresh_data=False, index=0,
iteration=1, loop_count=0, count_per_domain=100, callback=None, current_count=0,
max_count=-1, tf=20, thread_pool_size=20, get_backlinks=True, bad_country_list=[]):
"""
:param account:
:param seed_domains:
:param catagories:
:param fresh_data:
:param index:
:param iteration:
:param loop_count:
:param count_per_domain:
:param callback:
:param current_count:
:param max_count:
:param tf:
:param thread_pool_size:
:param get_backlinks: it will get backlinks of domains if True, else it will get ref domains instead, which is cheaper.
:return:
"""
target_func = GoogleMajestic._get_back_link_thread if get_backlinks else GoogleMajestic._get_ref_domain_thread
if iteration < 0:
raise ValueError("get_sites_by_seed_sites: iteration should >= 0.")
sub_domains = [LinkChecker.get_root_domain(x, use_www=False)[4] for x in seed_domains[index:]]
if len(sub_domains) == 0:
print("sub_domains is len 0.")
return
# counter = index
process_len = len(sub_domains)
if max_count > 0:
if current_count >= max_count:
print("exceeded seed len.")
return
elif current_count + process_len > max_count:
process_len = max_count - current_count
#target_catagories = []
# for catagory in catagories:
# target_catagories.append(str(CategoryManager.decode_sub_category(catagory, False)))
temp_sub_domains = []
thread_pool = ThreadPool(processes=thread_pool_size)
processes = [thread_pool.apply_async(target_func,
args=(account, x, count_per_domain, fresh_data, sub_domains,
temp_sub_domains, catagories, callback, tf, bad_country_list))
for x in sub_domains[0: process_len]]
results = [y.get() for y in processes]
thread_pool.terminate()
current_count += process_len
if loop_count >= iteration:
return
else:
new_seeds = sub_domains + temp_sub_domains
print("going to next level with seeds:", len(new_seeds))
return GoogleMajestic.get_sites_by_seed_sites_muti_threads(account, new_seeds,
catagories, fresh_data, len(seed_domains),
iteration, loop_count+1, count_per_domain,
callback, current_count, max_count,
thread_pool_size=10, tf=tf,
get_backlinks=get_backlinks,
bad_country_list=bad_country_list)
def __init__(self, full_link: str="", data_source: SiteTempDataSrcInterface=None,
controller: SiteCheckerController=None,
max_level=10, max_page=1000, delegate=None, output_buff_size=2000,
output_queue=None, output_all_external=False, result_delegate=None,
memory_control_terminate_event=None, check_robot_text=True,
**kwargs):
"""
:param full_link: The full link of a domain, e.g: https://www.google.co.uk
:param domain: domain to crawl
:param max_level: stop crawling if it reaches this level
:param max_page: maximum pages to check within a site, also stop crawling
:param delegate: if this is not None, then it will send the latest result of external domain of ResponseCode==404 or 999
:param result_delegate: send site_info upon finish
:param memory_control_terminate_event: if this is not None and being set, it will be able to terminate an external memory controlled process.
:return:
"""
FeedbackInterface.__init__(self, **kwargs)
#super(SiteChecker, self).__init__(**kwargs)
if full_link is None or len(full_link) == 0:
raise ValueError()
original_path = ""
try:
paras = urlsplit(full_link)
self.scheme, self.domain, original_path = paras[0], paras[1], paras[2]
except:
pass
domain_data = LinkChecker.get_root_domain(full_link, False)
self.root_domain = domain_data[1]
self.sub_domain = domain_data[4]
self.domain_suffix = domain_data[5]
self.sub_domain_no_local = self.sub_domain.strip(self.domain_suffix)
if self.scheme == "":
self.scheme = "http"
if self.domain == "":
self.domain = self.root_domain
self.orginal_link = full_link
self.domain_link = LinkChecker.get_valid_link(self.root_domain, full_link, self.scheme)
self.max_level = max_level
self.max_page = max_page
self.page_count = 0 # keep track page done
self._page_count_shadow = 0 # track previous count
self._all_page_count_shadow = 0 #track previous count in datasource
self.internal_page_count = 0
self.internal_page_last_count = 0
self.page_allocated = 0
self.current_level = 0 # if this = 0, it is root domain/home_page
self._stop_event = Event()
valid_file_name = SiteTempDataSrcInterface.get_valid_file_name(self.domain_link)
self._external_db_buffer = ExternalTempDataDiskBuffer(valid_file_name+".ext.db", self,
stop_event=self._stop_event,
buf_size=int(output_buff_size/2),
dir_path=get_db_buffer_default_dir(),
convert_output=False)
self._external_db_buffer.append_to_buffer([(self.root_domain, ResponseCode.DNSError),], convert_tuple=False)
self._memory_control_terminate_event = memory_control_terminate_event
self.task_control_lock = threading.RLock()
if data_source is None:
#self.data_source = SiteTempDataDisk(self.root_domain, ref_obj=self)
self.data_source = SiteTempDataDiskWithBuff(ref=self.domain_link, output_buff_size=output_buff_size, ref_obj=self)
else:
self.data_source = data_source # a list of OnSiteLink
self.delegate = delegate
if LinkChecker.might_be_link_html_page(original_path):
self.data_source.append(OnSiteLink(self.domain_link, response_code=ResponseCode.LinkOK, link_level=1)) # add the root domain as a starting point
self.data_source.append(OnSiteLink(self.scheme + "://www."+self.sub_domain, ResponseCode.LinkOK, link_level=1))
self.data_source.append(OnSiteLink(self.scheme + "://" + self.domain, ResponseCode.LinkOK, link_level=1))
self.cache_list = [] # internal page cache
self.page_need_look_up_temp = 0
self.cache_list.append(self.domain_link)
if "www." not in self.sub_domain:
self.cache_list.append(self.scheme + "://www."+self.sub_domain)
self.cache_list.append(self.scheme + "://" + self.domain)
self.page_need_look_up = self.data_source.count_all()
self.cache_size = 500 # create a small cache list to avoid going to check link in file system with lots of read and write
self._double_check_cache_lock = threading.RLock()
self._double_check_cache = deque(maxlen=self.cache_size)
self.external_cache_list = []
self.external_cache_size = 500 # cache that hold external sites
self.external_links_checked = 0
self.add_internal_page_OK_only = True
self.output_queue = output_queue
self.output_all_external = output_all_external
self.controller = controller
self.result_delegate = result_delegate
self.page_count_lock = threading.RLock()
self.internal_page_count_lock = threading.RLock()
self.level_lock = threading.RLock()
self.page_look_up_lock = threading.RLock()
self.external_link_check_lock = threading.RLock()
self._finihsed = False
self.task_control_max = 1
self.agent = "VegeBot (we follow your robots.txt settings before crawling, you can slow down the bot by change the Crawl-Delay parameter in the settings." \
"if you have an enquiry, please email to: [email protected])"
self.agent_from = "*****@*****.**"
if check_robot_text:
self.robot_agent = LinkChecker.get_robot_agent(self.sub_domain, protocol=self.scheme)
else:
self.robot_agent = None
self.site_crawl_delay = 0.60
if isinstance(self.robot_agent, Rules):
delay_temp = self.robot_agent.delay(self.agent)
if delay_temp is not None and delay_temp != self.site_crawl_delay:
self.site_crawl_delay = delay_temp
self.task_control_counter = 1
self._speed_penalty_count = 0
self._speed_penalty_threshold = 10
self._progress_logging_speed = 120
self._output_period = 120
self._output_batch_size = 100
self._death_wish_sent = False
SiteChecker._is_lxml_parser_exist()
self._output_thread = None
self._output_queue = None
self.progress_logger = ProgressLogger(self._progress_logging_speed, self, self._stop_event)
self._status = "Start"
self._populate_with_state() # restore laste known state
def get_search_results(keyword: str,
page_number: int,
proxy: ProxyStruct = None,
result_per_page: int = GoogleConst.Result100,
timeout=5,
return_domain_home_only=True,
use_forbidden_filter=True,
days_ago=0,
addtional_query_parameter: str = "",
country_code="us",
use_browser=False) -> list:
"""
generic normal search, get a list of domains form page
:param keyword:
:param page_number: > 0
:param resultPerPage:
:param timeout:
:param return_domain_home_only: return root domain name if True, else return protocol suffix + domain name
:param use_forbidden_filter:
:param days_ago: specify how many days ago before when results were indexed.
:return:
"""
assert page_number > 0, "page number should be greater than 0."
page_range = GoogleCom.get_result_per_page_range()
assert result_per_page in page_range, "result per page should be one of those values:" + str(
page_range)
sub_domain = "www"
request_link = GoogleUtility.get_local_endpoint(country_code, sub_domain) \
+ GoogleConst.CommonSearchPath.format(quote(keyword), result_per_page, (page_number - 1) * result_per_page, country_code) \
+ addtional_query_parameter+GoogleUtility.get_query_for_days(days_ago)
try:
user_agent = WebRequestCommonHeader.webpage_agent
if not use_browser:
response = GoogleCom._get_response(request_link,
proxy=proxy,
timeout=timeout,
user_agent=user_agent)
if not response.status_code == 200:
# if response.status_code == 503:
# print(response.text)
raise ConnectionRefusedError(
"error getting result, with status code:",
response.status_code)
result = response.text
else:
result = GoogleCom._get_response_browser(request_link,
proxy=proxy,
timeout=timeout,
user_agent=user_agent)
soup = bs4.BeautifulSoup(result)
tags = soup.select(GoogleConst.SitePath)
domains = []
for tag in tags:
try:
domain = tag.text.strip().replace(" ", "")
if return_domain_home_only:
domain = LinkChecker.get_root_domain(
domain, use_www=False)[2] # get the link
else:
domain = LinkChecker.get_root_domain(domain,
use_www=False)[3]
if use_forbidden_filter and LinkChecker.is_domain_forbidden(
domain):
continue
if len(domain) > 0:
domains.append(domain)
except:
pass
return domains
except Exception as ex:
print(ex)
return None
