def _merge_infobox(self, infobox):
add_infobox = True
infobox_id = infobox.get('id', None)
if infobox_id is not None:
for existingIndex in self.infoboxes:
if compare_urls(urlparse(existingIndex.get('id', '')), urlparse(infobox_id)):
merge_two_infoboxes(existingIndex, infobox)
add_infobox = False
if add_infobox:
self.infoboxes.append(infobox)
def _merge_result(self, result, position):
if result.__contains__('showurl'):
result['parsed_url'] = urlparse(result['showurl'])
else:
result['parsed_url'] = urlparse(result['url'])
# if the result has no scheme, use http as default
if not result['parsed_url'].scheme:
result['parsed_url'] = result['parsed_url']._replace(scheme="http")
result['url'] = urlparse(result['url']).geturl()
result['engines'] = set([result['engine']])
# strip multiple spaces and cariage returns from content
if result.get('content'):
result['content'] = WHITESPACE_REGEX.sub(' ', result['content'])
# check for duplicates
duplicated = False
for merged_result in self._merged_results:
if compare_urls(result['parsed_url'], merged_result['parsed_url'])\
and result.get('template') == merged_result.get('template'):
duplicated = merged_result
break
# merge duplicates together
if duplicated:
# using content with more text
if result_content_len(result.get('content', '')) >\
result_content_len(duplicated.get('content', '')):
duplicated['content'] = result['content']
# merge all result's parameters not found in duplicate
for key in result.keys():
if not duplicated.get(key):
duplicated[key] = result.get(key)
# add the new position
duplicated['positions'].append(position)
# add engine to list of result-engines
duplicated['engines'].add(result['engine'])
# using https if possible
if duplicated['parsed_url'].scheme != 'https' and result['parsed_url'].scheme == 'https':
duplicated['url'] = result['parsed_url'].geturl()
duplicated['parsed_url'] = result['parsed_url']
# if there is no duplicate found, append result
else:
result['positions'] = [position]
with RLock():
self._merged_results.append(result)
def _merge_infobox(self, infobox):
add_infobox = True
infobox_id = infobox.get('id', None)
infobox['engines'] = set([infobox['engine']])
if infobox_id is not None:
parsed_url_infobox_id = urlparse(infobox_id)
for existingIndex in self.infoboxes:
if compare_urls(urlparse(existingIndex.get('id', '')), parsed_url_infobox_id):
merge_two_infoboxes(existingIndex, infobox)
add_infobox = False
if add_infobox:
self.infoboxes.append(infobox)
def response(resp):
dom = html.fromstring(resp.text)
results = []
for element in dom.xpath('//div[@id="search"] //td'):
link = element.xpath('./a')[0]
google_url = urlparse(link.xpath('.//@href')[0])
query = parse_qs(google_url.query)
source_url = next(iter(query.get('q', [])), None)
title_parts = element.xpath('./cite//following-sibling::*/text()')
title_parts.extend(element.xpath('./cite//following-sibling::text()')[:-1])
result = {
'title': ''.join(title_parts),
'content': '',
'template': 'images.html',
'url': source_url,
'img_src': source_url,
'thumbnail_src': next(iter(link.xpath('.//img //@src')), None)
}
if not source_url or not result['thumbnail_src']:
continue
results.append(result)
return results
def response(resp):
dom = html.fromstring(resp.text)
results = []
for element in dom.xpath('//div[@id="search"] //td'):
link = element.xpath('./a')[0]
google_url = urlparse(link.xpath('.//@href')[0])
query = parse_qs(google_url.query)
source_url = next(iter(query.get('q', [])), None)
title_parts = element.xpath('./cite//following-sibling::*/text()')
title_parts.extend(
element.xpath('./cite//following-sibling::text()')[:-1])
result = {
'title': ''.join(title_parts),
'content': '',
'template': 'images.html',
'url': source_url,
'img_src': source_url,
'thumbnail_src': next(iter(link.xpath('.//img //@src')), None)
}
if not source_url or not result['thumbnail_src']:
continue
results.append(result)
return results
def clean_url(url):
parsed = urlparse(url)
query = [(k, v) for (k, v) in parse_qsl(parsed.query)
if k not in ['ixid', 's']]
return urlunparse(
(parsed.scheme, parsed.netloc, parsed.path, parsed.params,
urlencode(query), parsed.fragment))
def on_result(request, search, result):
doi = extract_doi(result['parsed_url'])
if doi and len(doi) < 50:
for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):
if doi.endswith(suffix):
doi = doi[:-len(suffix)]
result['url'] = get_doi_resolver(request.args, request.preferences.get_value('doi_resolver')) + doi
result['parsed_url'] = urlparse(result['url'])
return True
def on_result(request, search, result):
doi = extract_doi(result['parsed_url'])
if doi and len(doi) < 50:
for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):
if doi.endswith(suffix):
doi = doi[:-len(suffix)]
result['url'] = 'http://doai.io/' + doi
result['parsed_url'] = urlparse(result['url'])
return True
def on_result(request, search, result):
doi = extract_doi(result['parsed_url'])
if doi and len(doi) < 50:
for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):
if doi.endswith(suffix):
doi = doi[:-len(suffix)]
result['url'] = get_doi_resolver(request.args, request.preferences.get_value('doi_resolver')) + doi
result['parsed_url'] = urlparse(result['url'])
return True
def clean_url(url):
parsed = urlparse(url)
query = [(k, v) for (k, v) in parse_qsl(parsed.query) if k not in ['ixid', 's']]
return urlunparse((parsed.scheme,
parsed.netloc,
parsed.path,
parsed.params,
urlencode(query),
parsed.fragment))
def parse_url(url_string, google_hostname):
# sanity check
if url_string is None:
return url_string
# normal case
parsed_url = urlparse(url_string)
if (parsed_url.netloc in [google_hostname, '']
and parsed_url.path == redirect_path):
query = dict(parse_qsl(parsed_url.query))
return query['q']
else:
return url_string
def parse_url(url_string, google_hostname):
# sanity check
if url_string is None:
return url_string
# normal case
parsed_url = urlparse(url_string)
if (parsed_url.netloc in [google_hostname, '']
and parsed_url.path == redirect_path):
query = dict(parse_qsl(parsed_url.query))
return query['q']
else:
return url_string
def https_url_rewrite(result):
skip_https_rewrite = False
# check if HTTPS rewrite is possible
for target, rules, exclusions in https_rules:
# check if target regex match with url
if target.match(result['parsed_url'].netloc):
# process exclusions
for exclusion in exclusions:
# check if exclusion match with url
if exclusion.match(result['url']):
skip_https_rewrite = True
break
# skip https rewrite if required
if skip_https_rewrite:
break
# process rules
for rule in rules:
try:
new_result_url = rule[0].sub(rule[1], result['url'])
except:
break
# parse new url
new_parsed_url = urlparse(new_result_url)
# continiue if nothing was rewritten
if result['url'] == new_result_url:
continue
# get domainname from result
# TODO, does only work correct with TLD's like
# asdf.com, not for asdf.com.de
# TODO, using publicsuffix instead of this rewrite rule
old_result_domainname = '.'.join(
result['parsed_url'].hostname.split('.')[-2:])
new_result_domainname = '.'.join(
new_parsed_url.hostname.split('.')[-2:])
# check if rewritten hostname is the same,
# to protect against wrong or malicious rewrite rules
if old_result_domainname == new_result_domainname:
# set new url
result['url'] = new_result_url
# target has matched, do not search over the other rules
break
return result
def https_url_rewrite(result):
skip_https_rewrite = False
# check if HTTPS rewrite is possible
for target, rules, exclusions in https_rules:
# check if target regex match with url
if target.match(result['parsed_url'].netloc):
# process exclusions
for exclusion in exclusions:
# check if exclusion match with url
if exclusion.match(result['url']):
skip_https_rewrite = True
break
# skip https rewrite if required
if skip_https_rewrite:
break
# process rules
for rule in rules:
try:
new_result_url = rule[0].sub(rule[1], result['url'])
except:
break
# parse new url
new_parsed_url = urlparse(new_result_url)
# continiue if nothing was rewritten
if result['url'] == new_result_url:
continue
# get domainname from result
# TODO, does only work correct with TLD's like
# asdf.com, not for asdf.com.de
# TODO, using publicsuffix instead of this rewrite rule
old_result_domainname = '.'.join(
result['parsed_url'].hostname.split('.')[-2:])
new_result_domainname = '.'.join(
new_parsed_url.hostname.split('.')[-2:])
# check if rewritten hostname is the same,
# to protect against wrong or malicious rewrite rules
if old_result_domainname == new_result_domainname:
# set new url
result['url'] = new_result_url
# target has matched, do not search over the other rules
break
return result
def extract_url(xpath_results, search_url):
if xpath_results == []:
raise Exception('Empty url resultset')
url = extract_text(xpath_results)
if url.startswith('//'):
# add http or https to this kind of url //example.com/
parsed_search_url = urlparse(search_url)
url = u'{0}:{1}'.format(parsed_search_url.scheme, url)
elif url.startswith('/'):
# fix relative url to the search engine
url = urljoin(search_url, url)
# normalize url
url = normalize_url(url)
return url
def extract_url(xpath_results, search_url):
if xpath_results == []:
raise Exception('Empty url resultset')
url = extract_text(xpath_results)
if url.startswith('//'):
# add http or https to this kind of url //example.com/
parsed_search_url = urlparse(search_url)
url = u'{0}:{1}'.format(parsed_search_url.scheme or 'http', url)
elif url.startswith('/'):
# fix relative url to the search engine
url = urljoin(search_url, url)
# normalize url
url = normalize_url(url)
return url
def normalize_url(url):
parsed_url = urlparse(url)
# add a / at this end of the url if there is no path
if not parsed_url.netloc:
raise Exception('Cannot parse url')
if not parsed_url.path:
url += '/'
# FIXME : hack for yahoo
if parsed_url.hostname == 'search.yahoo.com'\
and parsed_url.path.startswith('/r'):
p = parsed_url.path
mark = p.find('/**')
if mark != -1:
return unquote(p[mark + 3:]).decode('utf-8')
return url
def normalize_url(url):
parsed_url = urlparse(url)
# add a / at this end of the url if there is no path
if not parsed_url.netloc:
raise Exception('Cannot parse url')
if not parsed_url.path:
url += '/'
# FIXME : hack for yahoo
if parsed_url.hostname == 'search.yahoo.com'\
and parsed_url.path.startswith('/r'):
p = parsed_url.path
mark = p.find('/**')
if mark != -1:
return unquote(p[mark + 3:]).decode('utf-8')
return url
def fetch_firefox_versions():
resp = requests.get(URL, timeout=2.0)
if resp.status_code != 200:
raise Exception("Error fetching firefox versions, HTTP code " + resp.status_code)
else:
dom = html.fromstring(resp.text)
versions = []
for link in dom.xpath('//a/@href'):
url = urlparse(urljoin(URL, link))
path = url.path
if path.startswith(RELEASE_PATH):
version = path[len(RELEASE_PATH):-1]
if NORMAL_REGEX.match(version):
versions.append(LooseVersion(version))
list.sort(versions, reverse=True)
return versions
def fetch_firefox_versions():
resp = requests.get(URL, timeout=2.0)
if resp.status_code != 200:
raise Exception("Error fetching firefox versions, HTTP code " +
resp.status_code)
else:
dom = html.fromstring(resp.text)
versions = []
for link in dom.xpath('//a/@href'):
url = urlparse(urljoin(URL, link))
path = url.path
if path.startswith(RELEASE_PATH):
version = path[len(RELEASE_PATH):-1]
if NORMAL_REGEX.match(version):
versions.append(LooseVersion(version))
list.sort(versions, reverse=True)
return versions
def response(resp):
img_results = []
text_results = []
search_results = json.loads(resp.text)
# return empty array if there are no results
if 'data' not in search_results:
return []
posts = search_results.get('data', {}).get('children', [])
# process results
for post in posts:
data = post['data']
# extract post information
params = {
'url': urljoin(base_url, data['permalink']),
'title': data['title']
}
# if thumbnail field contains a valid URL, we need to change template
thumbnail = data['thumbnail']
url_info = urlparse(thumbnail)
# netloc & path
if url_info[1] != '' and url_info[2] != '':
params['img_src'] = data['url']
params['thumbnail_src'] = thumbnail
params['template'] = 'images.html'
img_results.append(params)
else:
created = datetime.fromtimestamp(data['created_utc'])
content = data['selftext']
if len(content) > 500:
content = content[:500] + '...'
params['content'] = content
params['publishedDate'] = created
text_results.append(params)
# show images first and text results second
return img_results + text_results
def url_proxy():
"""get real url for baidu sogou and 360sousuo"""
url = request.args.get('proxyurl')
token = request.args.get('token')
if token != new_hmac(settings['result_proxy']['key'], url.encode('utf-8')):
return render('404.html'), 404
if "www.baidu.com/link?url" in url:
try:
resp = requests.head(url, timeout=1)
except requests.exceptions.Timeout:
return redirect(url)
if resp.status_code == 200:
realurl = resp.url
else:
realurl = url
return redirect(realurl)
else:
try:
resp = requests.get(url, timeout=1)
except requests.exceptions.Timeout:
return redirect(url)
if resp.status_code == 200:
if "http:" not in resp.text and "https:" not in resp.text:
# try to fix response with host in window.location.replace function
resp_content = resp.text.strip()
count = resp_content.index("window.location.replace(")
str_content = list(resp_content)
# 25 is len("window.location.replace(")+1
str_content.insert(count + 25, "https://" + urlparse(url)[1])
resp_content = "".join(str_content)
return resp_content
else:
# to get url from html response
return resp.content
else:
return redirect(url)
def response(resp):
results = []
dom = html.fromstring(resp.text)
# parse results
for img in dom.xpath('//a'):
r = {
'title': u' '.join(img.xpath('.//div[class="rg_ilmbg"]//text()')),
'content': '',
'template': 'images.html',
}
url = urlparse(img.xpath('.//@href')[0])
query = parse_qs(url.query)
r['url'] = query['imgrefurl'][0]
r['img_src'] = query['imgurl'][0]
r['thumbnail_src'] = r['img_src']
# append result
results.append(r)
# return results
return results
def __merge_url_result(self, result, position):
result['parsed_url'] = urlparse(result['url'])
# if the result has no scheme, use http as default
if not result['parsed_url'].scheme:
result['parsed_url'] = result['parsed_url']._replace(scheme="http")
result['url'] = result['parsed_url'].geturl()
result['engines'] = set([result['engine']])
# strip multiple spaces and cariage returns from content
if result.get('content'):
result['content'] = WHITESPACE_REGEX.sub(' ', result['content'])
duplicated = self.__find_duplicated_http_result(result)
if duplicated:
self.__merge_duplicated_http_result(duplicated, result, position)
return
# if there is no duplicate found, append result
result['positions'] = [position]
with RLock():
self._merged_results.append(result)
def response(resp):
"""Get response from google's search request"""
results = []
# detect google sorry
resp_url = urlparse(resp.url)
if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect':
raise RuntimeWarning('sorry.google.com')
if resp_url.path.startswith('/sorry'):
raise RuntimeWarning(gettext('CAPTCHA required'))
# which subdomain ?
# subdomain = resp.search_params.get('google_subdomain')
# convert the text to dom
dom = html.fromstring(resp.text)
img_bas64_map = scrap_out_thumbs(dom)
# parse results
#
# root element::
# <div id="islmp" ..>
# result div per image::
# <div jsmodel="tTXmib"> / <div jsaction="..." data-id="..."
# The data-id matches to a item in a json-data structure in::
# <script nonce="I+vqelcy/01CKiBJi5Z1Ow">AF_initDataCallback({key: 'ds:1', ... data:function(){return [ ...
# In this structure the ling to the origin PNG, JPG or whatever is given
# (we do not blow out the link there, you could still implement that)
# first link per image-div contains a <img> with the data-iid for bas64 encoded image data::
# <img class="rg_i Q4LuWd" data-iid="0"
# second link per image-div is the target link::
# <a class="VFACy kGQAp" href="https://en.wikipedia.org/wiki/The_Sacrament_of_the_Last_Supper">
# the second link also contains two div tags with the *description* and *publisher*::
# <div class="WGvvNb">The Sacrament of the Last Supper ...</div>
# <div class="fxgdke">en.wikipedia.org</div>
root = eval_xpath(dom, '//div[@id="islmp"]')
if not root:
logger.error("did not find root element id='islmp'")
return results
root = root[0]
for img_node in eval_xpath(root, './/img[contains(@class, "rg_i")]'):
try:
img_alt = eval_xpath(img_node, '@alt')[0]
img_base64_id = eval_xpath(img_node, '@data-iid')
if img_base64_id:
img_base64_id = img_base64_id[0]
thumbnail_src = img_bas64_map[img_base64_id]
else:
thumbnail_src = eval_xpath(img_node, '@src')
if not thumbnail_src:
thumbnail_src = eval_xpath(img_node, '@data-src')
if thumbnail_src:
thumbnail_src = thumbnail_src[0]
else:
thumbnail_src = ''
link_node = eval_xpath(img_node, '../../../a[2]')[0]
url = eval_xpath(link_node, '@href')[0]
pub_nodes = eval_xpath(link_node, './div/div')
pub_descr = img_alt
pub_source = ''
if pub_nodes:
pub_descr = extract_text(pub_nodes[0])
pub_source = extract_text(pub_nodes[1])
results.append({
'url': url,
'title': img_alt,
'content': pub_descr,
'source': pub_source,
'img_src': url,
# 'img_format': img_format,
'thumbnail_src': thumbnail_src,
'template': 'images.html'
})
except Exception as e: # pylint: disable=broad-except
logger.error(e, exc_info=True)
# from lxml import etree
# logger.debug(etree.tostring(img_node, pretty_print=True))
# import pdb
# pdb.set_trace()
continue
return results
def response(resp):
results = []
# detect google sorry
resp_url = urlparse(resp.url)
if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect':
raise RuntimeWarning('sorry.google.com')
if resp_url.path.startswith('/sorry'):
raise RuntimeWarning(gettext('CAPTCHA required'))
# which hostname ?
google_hostname = resp.search_params.get('google_hostname')
google_url = "https://" + google_hostname
# convert the text to dom
dom = html.fromstring(resp.text)
instant_answer = dom.xpath('//div[contains(@id, "ires")]//div[contains(@class, "hp-xpdbox")]')
if instant_answer:
answer_re = r'(?P<prefix><a\s+href=")\/url\?q=(?P<url>[^"]+?)\&\;[^"]*(?P<suffix>"\s*>)'
answer_subst = "\\g<prefix>\\g<url>\\g<suffix>"
answer_html = ['<br>']
for element in instant_answer:
answer_html.append(etree.tostring(element, method="html"))
answer_str = u' '.join(answer_html)
answer_fixed = re.sub(answer_re, answer_subst, answer_str, 0, re.MULTILINE)
results.append({'answer': answer_fixed})
try:
results_num = int(dom.xpath('//div[@id="resultStats"]//text()')[0]
.split()[1].replace(',', ''))
results.append({'number_of_results': results_num})
except:
pass
# parse results
for result in dom.xpath(results_xpath):
try:
title = extract_text(result.xpath(title_xpath)[0])
url = parse_url(extract_url(result.xpath(url_xpath), google_url), google_hostname)
parsed_url = urlparse(url, google_hostname)
# map result
if parsed_url.netloc == google_hostname:
# TODO fix inside links
continue
# if parsed_url.path.startswith(maps_path) or parsed_url.netloc.startswith(map_hostname_start):
# print "yooooo"*30
# x = result.xpath(map_near)
# if len(x) > 0:
# # map : near the location
# results = results + parse_map_near(parsed_url, x, google_hostname)
# else:
# # map : detail about a location
# results = results + parse_map_detail(parsed_url, result, google_hostname)
# # google news
# elif parsed_url.path == search_path:
# # skipping news results
# pass
# # images result
# elif parsed_url.path == images_path:
# # only thumbnail image provided,
# # so skipping image results
# # results = results + parse_images(result, google_hostname)
# pass
else:
# normal result
content = extract_text_from_dom(result, content_xpath)
if content is None:
continue
content_misc = extract_text_from_dom(result, content_misc_xpath)
if content_misc is not None:
content = content_misc + "<br />" + content
# append result
results.append({'url': url,
'title': title,
'content': content
})
except:
logger.debug('result parse error in:\n%s', etree.tostring(result, pretty_print=True))
continue
# parse suggestion
for suggestion in dom.xpath(suggestion_xpath):
# append suggestion
results.append({'suggestion': extract_text(suggestion)})
for correction in dom.xpath(spelling_suggestion_xpath):
results.append({'correction': extract_text(correction)})
# return results
return results
def response(resp):
"""Get response from google's search request"""
results = []
# detect google sorry
resp_url = urlparse(resp.url)
if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect':
raise RuntimeWarning('sorry.google.com')
if resp_url.path.startswith('/sorry'):
raise RuntimeWarning(gettext('CAPTCHA required'))
# which subdomain ?
# subdomain = resp.search_params.get('google_subdomain')
# convert the text to dom
dom = html.fromstring(resp.text)
# results --> answer
answer = eval_xpath(dom, '//div[contains(@class, "LGOjhe")]//text()')
if answer:
results.append({'answer': ' '.join(answer)})
else:
logger.debug("did not found 'answer'")
# results --> number_of_results
try:
_txt = eval_xpath(dom, '//div[@id="result-stats"]//text()')[0]
_digit = ''.join([n for n in _txt if n.isdigit()])
number_of_results = int(_digit)
results.append({'number_of_results': number_of_results})
except Exception as e: # pylint: disable=broad-except
logger.debug("did not 'number_of_results'")
logger.error(e, exc_info=True)
# parse results
for result in eval_xpath(dom, results_xpath):
# google *sections*
if extract_text(eval_xpath(result, g_section_with_header)):
logger.debug("ingoring <g-section-with-header>")
continue
try:
title = extract_text(eval_xpath(result, title_xpath)[0])
url = eval_xpath(result, href_xpath)[0]
content = extract_text_from_dom(result, content_xpath)
results.append({'url': url, 'title': title, 'content': content})
except Exception as e: # pylint: disable=broad-except
logger.error(e, exc_info=True)
# from lxml import etree
# logger.debug(etree.tostring(result, pretty_print=True))
# import pdb
# pdb.set_trace()
continue
# parse suggestion
for suggestion in eval_xpath(dom, suggestion_xpath):
# append suggestion
results.append({'suggestion': extract_text(suggestion)})
for correction in eval_xpath(dom, spelling_suggestion_xpath):
results.append({'correction': extract_text(correction)})
# return results
return results
def image_url_cleanup(url_string):
parsed_url = urlparse(url_string)
if parsed_url.netloc.endswith('bing4.com') and parsed_url.path == '/th':
query = dict(parse_qsl(parsed_url.query))
return "https://www.bing.com/th?id=" + query.get('id')
return url_string
def url_cleanup(url_string):
parsed_url = urlparse(url_string)
if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx':
query = dict(parse_qsl(parsed_url.query))
return query.get('url', None)
return url_string
def extract_domain(url):
return urlparse(url)[1]
def merge_two_infoboxes(infobox1, infobox2):
# get engines weights
if hasattr(engines[infobox1['engine']], 'weight'):
weight1 = engines[infobox1['engine']].weight
else:
weight1 = 1
if hasattr(engines[infobox2['engine']], 'weight'):
weight2 = engines[infobox2['engine']].weight
else:
weight2 = 1
if weight2 > weight1:
infobox1['engine'] = infobox2['engine']
if 'urls' in infobox2:
urls1 = infobox1.get('urls', None)
if urls1 is None:
urls1 = []
for url2 in infobox2.get('urls', []):
unique_url = True
for url1 in infobox1.get('urls', []):
if compare_urls(urlparse(url1.get('url', '')),
urlparse(url2.get('url', ''))):
unique_url = False
break
if unique_url:
urls1.append(url2)
infobox1['urls'] = urls1
if 'img_src' in infobox2:
img1 = infobox1.get('img_src', None)
img2 = infobox2.get('img_src')
if img1 is None:
infobox1['img_src'] = img2
elif weight2 > weight1:
infobox1['img_src'] = img2
if 'attributes' in infobox2:
attributes1 = infobox1.get('attributes', None)
if attributes1 is None:
attributes1 = []
infobox1['attributes'] = attributes1
attributeSet = set()
for attribute in infobox1.get('attributes', []):
if attribute.get('label', None) not in attributeSet:
attributeSet.add(attribute.get('label', None))
for attribute in infobox2.get('attributes', []):
if attribute.get('label', None) not in attributeSet:
attributes1.append(attribute)
if 'content' in infobox2:
content1 = infobox1.get('content', None)
content2 = infobox2.get('content', '')
if content1 is not None:
if result_content_len(content2) > result_content_len(content1):
infobox1['content'] = content2
else:
infobox1['content'] = content2
def url_cleanup(url_string):
parsed_url = urlparse(url_string)
if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx':
query = dict(parse_qsl(parsed_url.query))
return query.get('url', None)
return url_string
def response(resp):
results = []
# detect google sorry
resp_url = urlparse(resp.url)
if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect':
raise RuntimeWarning('sorry.google.com')
if resp_url.path.startswith('/sorry'):
raise RuntimeWarning(gettext('CAPTCHA required'))
# which hostname ?
google_hostname = resp.search_params.get('google_hostname')
google_url = "https://" + google_hostname
# convert the text to dom
dom = html.fromstring(resp.text)
instant_answer = dom.xpath('//div[@id="_vBb"]//text()')
if instant_answer:
results.append({'answer': u' '.join(instant_answer)})
try:
results_num = int(dom.xpath('//div[@id="resultStats"]//text()')[0]
.split()[1].replace(',', ''))
results.append({'number_of_results': results_num})
except:
pass
# parse results
for result in dom.xpath(results_xpath):
try:
title = extract_text(result.xpath(title_xpath)[0])
url = parse_url(extract_url(result.xpath(url_xpath), google_url), google_hostname)
parsed_url = urlparse(url, google_hostname)
# map result
if parsed_url.netloc == google_hostname:
# TODO fix inside links
continue
# if parsed_url.path.startswith(maps_path) or parsed_url.netloc.startswith(map_hostname_start):
# print "yooooo"*30
# x = result.xpath(map_near)
# if len(x) > 0:
# # map : near the location
# results = results + parse_map_near(parsed_url, x, google_hostname)
# else:
# # map : detail about a location
# results = results + parse_map_detail(parsed_url, result, google_hostname)
# # google news
# elif parsed_url.path == search_path:
# # skipping news results
# pass
# # images result
# elif parsed_url.path == images_path:
# # only thumbnail image provided,
# # so skipping image results
# # results = results + parse_images(result, google_hostname)
# pass
else:
# normal result
content = extract_text_from_dom(result, content_xpath)
if content is None:
continue
content_misc = extract_text_from_dom(result, content_misc_xpath)
if content_misc is not None:
content = content_misc + "<br />" + content
# append result
results.append({'url': url,
'title': title,
'content': content
})
except:
logger.debug('result parse error in:\n%s', etree.tostring(result, pretty_print=True))
continue
# parse suggestion
for suggestion in dom.xpath(suggestion_xpath):
# append suggestion
results.append({'suggestion': extract_text(suggestion)})
for correction in dom.xpath(spelling_suggestion_xpath):
results.append({'correction': extract_text(correction)})
# return results
return results
def image_url_cleanup(url_string):
parsed_url = urlparse(url_string)
if parsed_url.netloc.endswith('bing4.com') and parsed_url.path == '/th':
query = dict(parse_qsl(parsed_url.query))
return "https://www.bing.com/th?id=" + query.get('id')
return url_string
def extract_domain(url):
return urlparse(url)[1]
def response(resp):
results = []
# detect google sorry
resp_url = urlparse(resp.url)
if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect':
raise RuntimeWarning('sorry.google.com')
# which hostname ?
google_hostname = resp.search_params.get('google_hostname')
google_url = "https://" + google_hostname
# convert the text to dom
dom = html.fromstring(resp.text)
instant_answer = dom.xpath('//div[@id="_vBb"]//text()')
if instant_answer:
results.append({'answer': u' '.join(instant_answer)})
try:
results_num = int(
dom.xpath('//div[@id="resultStats"]//text()')[0].split()
[1].replace(',', ''))
results.append({'number_of_results': results_num})
except:
pass
# parse results
for result in dom.xpath(results_xpath):
try:
title = extract_text(result.xpath(title_xpath)[0])
url = parse_url(extract_url(result.xpath(url_xpath), google_url),
google_hostname)
parsed_url = urlparse(url, google_hostname)
# map result
if parsed_url.netloc == google_hostname:
# TODO fix inside links
continue
# if parsed_url.path.startswith(maps_path) or parsed_url.netloc.startswith(map_hostname_start):
# print "yooooo"*30
# x = result.xpath(map_near)
# if len(x) > 0:
# # map : near the location
# results = results + parse_map_near(parsed_url, x, google_hostname)
# else:
# # map : detail about a location
# results = results + parse_map_detail(parsed_url, result, google_hostname)
# # google news
# elif parsed_url.path == search_path:
# # skipping news results
# pass
# # images result
# elif parsed_url.path == images_path:
# # only thumbnail image provided,
# # so skipping image results
# # results = results + parse_images(result, google_hostname)
# pass
else:
# normal result
content = extract_text_from_dom(result, content_xpath)
if content is None:
continue
content_misc = extract_text_from_dom(result,
content_misc_xpath)
if content_misc is not None:
content = content_misc + "<br />" + content
# append result
results.append({
'url': url,
'title': title,
'content': content
})
except:
logger.debug('result parse error in:\n%s',
etree.tostring(result, pretty_print=True))
continue
# parse suggestion
for suggestion in dom.xpath(suggestion_xpath):
# append suggestion
results.append({'suggestion': extract_text(suggestion)})
for correction in dom.xpath(spelling_suggestion_xpath):
results.append({'correction': extract_text(correction)})
# return results
return results