Added backend site alternative redirects/text replacement

By default, twitter/instagram/youtube will redirect to
nitter/bibliogram/invidious respectively. Currently missing a front end
for enabling/disabling this feature, but may want to set this as enabled
until a user manually disables it.

Also refactored util naming a bit, added filter utils to hold all
non-class methods from filter.py

app/__init__.py

@@ -1,4 +1,4 @@
-from app.utils.misc import generate_user_keys
+from app.utils.session_utils import generate_user_keys
 from flask import Flask
 from flask_session import Session
 import os

app/filter.py

@@ -1,56 +1,11 @@
 from app.request import VALID_PARAMS
-from app.utils.misc import BLACKLIST
-from bs4 import BeautifulSoup
+from app.utils.filter_utils import *
 from bs4.element import ResultSet
 from cryptography.fernet import Fernet
 import re
 import urllib.parse as urlparse
 from urllib.parse import parse_qs
 
-SKIP_ARGS = ['ref_src', 'utm']
-FULL_RES_IMG = '<br/><a href="{}">Full Image</a>'
-GOOG_IMG = '/images/branding/searchlogo/1x/googlelogo'
-LOGO_URL = GOOG_IMG + '_desk'
-BLANK_B64 = '''
-data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAQAAAAnOwc2AAAAD0lEQVR42mNkwAIYh7IgAAVVAAuInjI5AAAAAElFTkSuQmCC
-'''
-
-
-def get_first_link(soup):
-    # Replace hrefs with only the intended destination (no "utm" type tags)
-    for a in soup.find_all('a', href=True):
-        # Return the first search result URL
-        if 'url?q=' in a['href']:
-            return filter_link_args(a['href'])
-
-
-def filter_link_args(query_link):
-    parsed_link = urlparse.urlparse(query_link)
-    link_args = parse_qs(parsed_link.query)
-    safe_args = {}
-
-    if len(link_args) == 0 and len(parsed_link) > 0:
-        return query_link
-
-    for arg in link_args.keys():
-        if arg in SKIP_ARGS:
-            continue
-
-        safe_args[arg] = link_args[arg]
-
-    # Remove original link query and replace with filtered args
-    query_link = query_link.replace(parsed_link.query, '')
-    if len(safe_args) > 0:
-        query_link = query_link + urlparse.urlencode(safe_args, doseq=True)
-    else:
-        query_link = query_link.replace('?', '')
-
-    return query_link
-
-
-def has_ad_content(element: str):
-    return element.upper() in (value.upper() for value in BLACKLIST) or 'ⓘ' in element
-
 
 class Filter:
     def __init__(self, user_keys: dict, mobile=False, config=None):
@@ -61,6 +16,7 @@ class Filter:
         self.dark = config['dark'] if 'dark' in config else False
         self.nojs = config['nojs'] if 'nojs' in config else False
         self.new_tab = config['new_tab'] if 'new_tab' in config else False
+        self.alt_redirect = config['alts'] if 'alts' in config else True
         self.mobile = mobile
         self.user_keys = user_keys
         self.main_divs = ResultSet('')
@@ -232,11 +188,11 @@ class Filter:
         else:
             link['href'] = href
 
+        # Replace link location
+        if self.alt_redirect:
+            link['href'] = get_site_alt(link['href'])
+            link_desc = link.find_all(text=re.compile('|'.join(SITE_ALTS.keys())))
+            if len(link_desc) == 0:
+                return
 
-def gen_nojs(sibling):
-    nojs_link = BeautifulSoup().new_tag('a')
-    nojs_link['href'] = '/window?location=' + sibling['href']
-    nojs_link['style'] = 'display:block;width:100%;'
-    nojs_link.string = 'NoJS Link: ' + nojs_link['href']
-    sibling.append(BeautifulSoup('<br><hr><br>', 'html.parser'))
-    sibling.append(nojs_link)
+            link_desc[0].replace_with(get_site_alt(link_desc[0]))

app/routes.py

@@ -15,7 +15,7 @@ from requests import exceptions
 from app import app
 from app.models.config import Config
 from app.request import Request
-from app.utils.misc import valid_user_session
+from app.utils.session_utils import valid_user_session
 from app.utils.routing_utils import *
 
 

app/utils/filter_utils.py

@@ -0,0 +1,79 @@
+from bs4 import BeautifulSoup
+import urllib.parse as urlparse
+from urllib.parse import parse_qs
+
+SKIP_ARGS = ['ref_src', 'utm']
+FULL_RES_IMG = '<br/><a href="{}">Full Image</a>'
+GOOG_IMG = '/images/branding/searchlogo/1x/googlelogo'
+LOGO_URL = GOOG_IMG + '_desk'
+BLANK_B64 = '''
+data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAoAAAAKCAQAAAAnOwc2AAAAD0lEQVR42mNkwAIYh7IgAAVVAAuInjI5AAAAAElFTkSuQmCC
+'''
+
+BLACKLIST = [
+    'ad', 'anuncio', 'annuncio', 'annonce', 'Anzeige', '广告', '廣告', 'Reklama', 'Реклама', 'Anunț', '광고',
+    'annons', 'Annonse', 'Iklan', '広告', 'Augl.', 'Mainos', 'Advertentie', 'إعلان', 'Գովազդ', 'विज्ञापन', 'Reklam',
+    'آگهی', 'Reklāma', 'Reklaam', 'Διαφήμιση', 'מודעה', 'Hirdetés'
+]
+
+SITE_ALTS = {
+    'twitter.com': 'nitter.net',
+    'youtube.com': 'invidio.us',
+    'instagram.com': 'bibliogram.art/u'
+}
+
+
+def has_ad_content(element: str):
+    return element.upper() in (value.upper() for value in BLACKLIST) or 'ⓘ' in element
+
+
+def get_first_link(soup):
+    # Replace hrefs with only the intended destination (no "utm" type tags)
+    for a in soup.find_all('a', href=True):
+        # Return the first search result URL
+        if 'url?q=' in a['href']:
+            return filter_link_args(a['href'])
+
+
+def get_site_alt(link: str):
+    for site_key in SITE_ALTS.keys():
+        if site_key not in link:
+            continue
+
+        link = link.replace(site_key, SITE_ALTS[site_key])
+        break
+
+    return link
+
+
+def filter_link_args(query_link):
+    parsed_link = urlparse.urlparse(query_link)
+    link_args = parse_qs(parsed_link.query)
+    safe_args = {}
+
+    if len(link_args) == 0 and len(parsed_link) > 0:
+        return query_link
+
+    for arg in link_args.keys():
+        if arg in SKIP_ARGS:
+            continue
+
+        safe_args[arg] = link_args[arg]
+
+    # Remove original link query and replace with filtered args
+    query_link = query_link.replace(parsed_link.query, '')
+    if len(safe_args) > 0:
+        query_link = query_link + urlparse.urlencode(safe_args, doseq=True)
+    else:
+        query_link = query_link.replace('?', '')
+
+    return query_link
+
+
+def gen_nojs(sibling):
+    nojs_link = BeautifulSoup().new_tag('a')
+    nojs_link['href'] = '/window?location=' + sibling['href']
+    nojs_link['style'] = 'display:block;width:100%;'
+    nojs_link.string = 'NoJS Link: ' + nojs_link['href']
+    sibling.append(BeautifulSoup('<br><hr><br>', 'html.parser'))
+    sibling.append(nojs_link)

app/utils/routing_utils.py

@@ -1,5 +1,5 @@
 from app.filter import Filter, get_first_link
-from app.utils.misc import generate_user_keys
+from app.utils.session_utils import generate_user_keys
 from app.request import gen_query
 from bs4 import BeautifulSoup
 from cryptography.fernet import Fernet, InvalidToken

app/utils/session_utils.py

@@ -2,11 +2,6 @@ from cryptography.fernet import Fernet
 from flask import current_app as app
 
 REQUIRED_SESSION_VALUES = ['uuid', 'config', 'fernet_keys']
-BLACKLIST = [
-    'ad', 'anuncio', 'annuncio', 'annonce', 'Anzeige', '广告', '廣告', 'Reklama', 'Реклама', 'Anunț', '광고',
-    'annons', 'Annonse', 'Iklan', '広告', 'Augl.', 'Mainos', 'Advertentie', 'إعلان', 'Գովազդ', 'विज्ञापन', 'Reklam',
-    'آگهی', 'Reklāma', 'Reklaam', 'Διαφήμιση', 'מודעה', 'Hirdetés'
-]
 
 
 def generate_user_keys(cookies_disabled=False) -> dict:

test/conftest.py

@@ -1,5 +1,5 @@
 from app import app
-from app.utils.misc import generate_user_keys
+from app.utils.session_utils import generate_user_keys
 import pytest
 
 

test/test_misc.py

@@ -1,4 +1,4 @@
-from app.utils.misc import generate_user_keys, valid_user_session
+from app.utils.session_utils import generate_user_keys, valid_user_session
 
 
 def test_generate_user_keys():

test/test_results.py

@@ -1,6 +1,6 @@
 from bs4 import BeautifulSoup
 from app.filter import Filter
-from app.utils.misc import generate_user_keys
+from app.utils.session_utils import generate_user_keys
 from datetime import datetime
 from dateutil.parser import *