open.source/whoogle-search
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

PEP-8: Fix filter class formatting

Browse Source
This commit is contained in:
Ben Busby 2020-12-17 15:24:18 -05:00
parent dd56ca9bfc
commit 92d94d9ea7
No known key found for this signature in database
GPG Key ID: 3B08611DF6E62ED2
1 changed files with 47 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

73
app/filter.py
View File

@ -32,20 +32,27 @@ class Filter:
def reskin(self, page): def reskin(self, page):
# Aesthetic only re-skinning # Aesthetic only re-skinning
if self.dark: if self.dark:
page = page.replace('fff', '000').replace('202124', 'ddd').replace('1967D2', '3b85ea') page = page.replace(
'fff', '000').replace(
'202124', 'ddd').replace(
'1967D2', '3b85ea')
return page return page
def encrypt_path(self, msg, is_element=False): def encrypt_path(self, msg, is_element=False):
# Encrypts path to avoid plaintext results in logs # Encrypts path to avoid plaintext results in logs
if is_element: if is_element:
# Element paths are tracked differently in order for the element key to be regenerated # Element paths are encrypted separately from text, to allow key
# once all elements have been loaded # regeneration once all items have been served to the user
enc_path = Fernet(self.user_keys['element_key']).encrypt(msg.encode()).decode() enc_path = Fernet(
self.user_keys['element_key']
).encrypt(msg.encode()).decode()
self._elements += 1 self._elements += 1
return enc_path return enc_path
return Fernet(self.user_keys['text_key']).encrypt(msg.encode()).decode() return Fernet(
self.user_keys['text_key']
).encrypt(msg.encode()).decode()
def clean(self, soup): def clean(self, soup):
self.main_divs = soup.find('div', {'id': 'main'}) self.main_divs = soup.find('div', {'id': 'main'})
@ -74,8 +81,8 @@ class Filter:
footer = soup.find('footer') footer = soup.find('footer')
if footer: if footer:
# Remove divs that have multiple links beyond just page navigation # Remove divs that have multiple links beyond just page navigation
[_.decompose() for _ in footer.find_all('div', recursive=False) [_.decompose() for _ in footer.find_all('div', recursive=False)
if len(_.find_all('a', href=True)) > 3] if len(_.find_all('a', href=True)) > 3]
header = soup.find('header') header = soup.find('header')
if header: if header:
@ -88,8 +95,9 @@ class Filter:
return return
for div in [_ for _ in self.main_divs.find_all('div', recursive=True)]: for div in [_ for _ in self.main_divs.find_all('div', recursive=True)]:
has_ad = len([_ for _ in div.find_all('span', recursive=True) if has_ad_content(_.text)]) div_ads = [_ for _ in div.find_all('span', recursive=True)
_ = div.decompose() if has_ad else None if has_ad_content(_.text)]
_ = div.decompose() if len(div_ads) else None
def fix_question_section(self): def fix_question_section(self):
if not self.main_divs: if not self.main_divs:
@ -97,14 +105,14 @@ class Filter:
question_divs = [_ for _ in self.main_divs.find_all( question_divs = [_ for _ in self.main_divs.find_all(
'div', recursive=False 'div', recursive=False
) if len(_.find_all('h2')) > 0] ) if len(_.find_all('h2')) > 0]
if len(question_divs) == 0: if len(question_divs) == 0:
return return
# Wrap section in details element to allow collapse/expand # Wrap section in details element to allow collapse/expand
details = BeautifulSoup(features='lxml').new_tag('details') details = BeautifulSoup('html.parser').new_tag('details')
summary = BeautifulSoup(features='lxml').new_tag('summary') summary = BeautifulSoup('html.parser').new_tag('summary')
summary.string = question_divs[0].find('h2').text summary.string = question_divs[0].find('h2').text
question_divs[0].find('h2').decompose() question_divs[0].find('h2').decompose()
details.append(summary) details.append(summary)
@ -113,7 +121,7 @@ class Filter:
for question_div in question_divs: for question_div in question_divs:
questions = [_ for _ in question_div.find_all( questions = [_ for _ in question_div.find_all(
'div', recursive=True 'div', recursive=True
) if _.text.endswith('?')] ) if _.text.endswith('?')]
for question in questions: for question in questions:
question['style'] = 'padding: 10px; font-style: italic;' question['style'] = 'padding: 10px; font-style: italic;'
@ -131,11 +139,15 @@ class Filter:
element['src'] = BLANK_B64 element['src'] = BLANK_B64
return return
element['src'] = 'element?url=' + self.encrypt_path(element_src, is_element=True) + \ element['src'] = 'element?url=' + self.encrypt_path(
'&type=' + urlparse.quote(mime) element_src,
# TODO: Non-mobile image results link to website instead of image is_element=True) + '&type=' + urlparse.quote(mime)
# FIXME: Non-mobile image results link to website instead of image
# if not self.mobile: # if not self.mobile:
# img.append(BeautifulSoup(FULL_RES_IMG.format(element_src), 'html.parser')) # img.append(
# BeautifulSoup(FULL_RES_IMG.format(element_src),
# 'html.parser'))
def update_styling(self, soup): def update_styling(self, soup):
# Remove unnecessary button(s) # Remove unnecessary button(s)
@ -149,8 +161,9 @@ class Filter:
# Update logo # Update logo
logo = soup.find('a', {'class': 'l'}) logo = soup.find('a', {'class': 'l'})
if logo and self.mobile: if logo and self.mobile:
logo['style'] = 'display:flex; justify-content:center; align-items:center; color:#685e79; ' \ logo['style'] = ('display:flex; justify-content:center; '
'font-size:18px; ' 'align-items:center; color:#685e79; '
'font-size:18px; ')
# Fix search bar length on mobile # Fix search bar length on mobile
try: try:
@ -163,7 +176,7 @@ class Filter:
# Replace href with only the intended destination (no "utm" type tags) # Replace href with only the intended destination (no "utm" type tags)
href = link['href'].replace('https://www.google.com', '') href = link['href'].replace('https://www.google.com', '')
if 'advanced_search' in href or 'tbm=shop' in href: if 'advanced_search' in href or 'tbm=shop' in href:
# TODO: The "Shopping" tab requires further filtering (see #136) # FIXME: The "Shopping" tab requires further filtering (see #136)
# Temporarily removing all links to that tab for now. # Temporarily removing all links to that tab for now.
link.decompose() link.decompose()
return return
@ -171,20 +184,26 @@ class Filter:
link['target'] = '_blank' link['target'] = '_blank'
result_link = urlparse.urlparse(href) result_link = urlparse.urlparse(href)
query_link = parse_qs(result_link.query)['q'][0] if '?q=' in href else '' query_link = parse_qs(
result_link.query
)['q'][0] if '?q=' in href else ''
if query_link.startswith('/'): if query_link.startswith('/'):
# Internal google links (i.e. mail, maps, etc) should still be forwarded to Google # Internal google links (i.e. mail, maps, etc) should still
# be forwarded to Google
link['href'] = 'https://google.com' + query_link link['href'] = 'https://google.com' + query_link
elif '/search?q=' in href: elif '/search?q=' in href:
# "li:1" implies the query should be interpreted verbatim, so we wrap it in double quotes # "li:1" implies the query should be interpreted verbatim,
# which is accomplished by wrapping the query in double quotes
if 'li:1' in href: if 'li:1' in href:
query_link = '"' + query_link + '"' query_link = '"' + query_link + '"'
new_search = 'search?q=' + self.encrypt_path(query_link) new_search = 'search?q=' + self.encrypt_path(query_link)
query_params = parse_qs(urlparse.urlparse(href).query) query_params = parse_qs(urlparse.urlparse(href).query)
for param in VALID_PARAMS: for param in VALID_PARAMS:
param_val = query_params[param][0] if param in query_params else '' if param not in query_params:
continue
param_val = query_params[param][0]
new_search += '&' + param + '=' + param_val new_search += '&' + param + '=' + param_val
link['href'] = new_search link['href'] = new_search
elif 'url?q=' in href: elif 'url?q=' in href:
@ -199,9 +218,11 @@ class Filter:
# Replace link location if "alts" config is enabled # Replace link location if "alts" config is enabled
if self.alt_redirect: if self.alt_redirect:
# Search and replace all link descriptions with alternative location # Search and replace all link descriptions
# with alternative location
link['href'] = get_site_alt(link['href']) link['href'] = get_site_alt(link['href'])
link_desc = link.find_all(text=re.compile('|'.join(SITE_ALTS.keys()))) link_desc = link.find_all(
text=re.compile('|'.join(SITE_ALTS.keys())))
if len(link_desc) == 0: if len(link_desc) == 0:
return return