jandieman
/
searxng
mirror of https://github.com/searxng/searxng.git
2
0
Fork 0
Code Issues Projects Releases Wiki Activity

updated formatting

This commit is contained in:
Fmstrat 2024-08-14 10:06:49 -04:00
parent 8ae66a7eaa
commit 5e11d697ab
1 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
searx/engines/google.py
View File

@ -324,12 +324,12 @@ def _eval_answers(results, dom, xpath):
answer_list = eval_xpath(dom, xpath)
drop_elements = [
'.//div[@class="nnFGuf"]',
'.//script', # Scripts like the calculator
'.//table[@class="HOoTuc"]', # The actual calculator controls
'.//table[@class="ElumCf"]', # The actual calculator buttons
'.//div[@class="mDRaHd"]', # Instructions with links
'.//script', # Scripts like the calculator
'.//table[@class="HOoTuc"]', # The actual calculator controls
'.//table[@class="ElumCf"]', # The actual calculator buttons
'.//div[@class="mDRaHd"]', # Instructions with links
'.//span[@class="W7GCoc CNbPnc"]', # Feedback
'.//*[@style="display:none"]', # Hidden elements
'.//*[@style="display:none"]', # Hidden elements
]
for item in answer_list:
for element in eval_xpath(item, ' | '.join(drop_elements)):
@ -340,20 +340,20 @@ def _eval_answers(results, dom, xpath):
if table_elements:
extracted_table = table_elements[0]
extracted_table.attrib.clear()
for element in extracted_table.xpath(f'.//*'):
for element in extracted_table.xpath('.//*'):
element.attrib.clear()
extracted_table.set('cellpadding', '2')
extracted_table.set('cellspacing', '2')
extracted_table.set('border', '0')
extracted_table_html = html.tostring(extracted_table, pretty_print=True, encoding='unicode')
is_safe = True
for element in eval_xpath(item, './/table'): # Drop all remaining tables
for element in eval_xpath(item, './/table'): # Drop all remaining tables
element.drop_tree()
answer_content = extract_text(item)
if extracted_table_html:
answer_content += '<p>' + extracted_table_html
url = (eval_xpath(item, '../..//a/@href') + [None])[0]
if url and url.startswith('/search?'): # If the answer is a Google search link, don't use it
if url and url.startswith('/search?'): # If the answer is a Google search link, don't use it
url = None
results.append(
{
@ -376,21 +376,21 @@ def response(resp):
dom = html.fromstring(resp.text)
data_image_map = _parse_data_images(dom)
results = _eval_answers(results, dom, '//div[contains(@class, "card-section")]') # Look for cards first
results = _eval_answers(results, dom, '//div[contains(@class, "card-section")]') # Look for cards first
if not results:
results = _eval_answers(results, dom, '//div[contains(@class, "LGOjhe")]') # Look for rendered answers next
results = _eval_answers(results, dom, '//div[contains(@class, "LGOjhe")]') # Look for rendered answers next
# Look for JS DOM encoded string answers last
if not results:
pattern = r"'\\x3c.*?\\x3e'" # These are DOM encoded strings that Google will push into the DOM
pattern = r"'\\x3c.*?\\x3e'" # These are DOM encoded strings that Google will push into the DOM
matches = re.findall(pattern, resp.text)
for match in matches:
decoded_html = match.encode().decode('unicode_escape')
encoded_dom = html.fromstring(decoded_html)
sub_doms = eval_xpath(encoded_dom, '//div[contains(@class, "LGOjhe")]')
if sub_doms:
if '<span class="hgKElc"><b>' in decoded_html: # Main answers start with a bold
if '<span class="hgKElc"><b>' in decoded_html: # Main answers start with a bold
results = _eval_answers(results, encoded_dom, '//div[contains(@class, "LGOjhe")]')
break # If it's a JS encoded answer, we only want the first one if it has bold above
break # If it's a JS encoded answer, we only want the first one if it has bold above
# parse results