Added html parsing to remove returned scripts, added logo

.gitignore

@@ -2,4 +2,5 @@ venv/
 __pycache__/
 *.pyc
 *.pem
+*.xml
 run.sh

app/routes.py

@@ -1,7 +1,9 @@
 from app import app
+from bs4 import BeautifulSoup
 from flask import request, redirect, Response, render_template
 import os
 import pycurl
+import re
 from .url import url_parse
 from io import BytesIO
 
@@ -24,7 +26,12 @@ def search():
     if 'tbm' in request.args:
         tbm = '&tbm=' + request.args.get('tbm')
 
+    start = ''
+    if 'start' in request.args:
+        start = '&start=' + request.args.get('start')
+
     user_agent = request.headers.get('User-Agent')
+    full_query = url_parse(q) + tbm + start
 
     google_ua = DESKTOP_UA
     if 'Android' in user_agent or 'iPhone' in user_agent:
@@ -32,17 +39,33 @@ def search():
 
     b_obj = BytesIO()
     crl = pycurl.Curl()
-    crl.setopt(crl.URL, 'https://www.google.com/search?q=' + url_parse(q) + tbm)
+    crl.setopt(crl.URL, 'https://www.google.com/search?gbv=1&q=' + full_query)
     crl.setopt(crl.USERAGENT, google_ua)
     crl.setopt(crl.WRITEDATA, b_obj)
     crl.perform()
     crl.close()
-    get_body = b_obj.getvalue()
+    get_body = b_obj.getvalue().decode('utf-8', 'ignore')
-    return render_template('search.html', response=get_body.decode("utf-8", 'ignore'))
+    get_body = get_body.replace('data-src', 'src').replace('.001', '1').replace('visibility:hidden', 'visibility:visible').replace('>G<', '>Bl<')
+
+    pattern = re.compile('4285f4|ea4335|fbcc05|34a853|fbbc05', re.IGNORECASE)
+    get_body = pattern.sub('0000ff', get_body)
+
+    soup = BeautifulSoup(get_body, 'html.parser')
+    try:
+        for script in soup("script"):
+            script.decompose()
+        soup.find('div', id='sfooter').decompose()
+    except Exception:
+        pass
+
+    return render_template('search.html', response=soup)
 
 
 @app.route('/url', methods=['GET'])
 def url():
+    if 'url' in request.args:
+        return redirect(request.args.get('url'))
+
     q = request.args.get('q')
     if len(q) > 0 and 'http' in q:
         return redirect(q)
@@ -50,5 +73,10 @@ def url():
         return render_template('error.html')
 
 
+@app.route('/imgres')
+def imgres():
+    return redirect(request.args.get('imgurl'))
+
+
 if __name__ == '__main__':
     app.run(debug=True, host='0.0.0.0')

app/static/css/main.css

@@ -5,6 +5,13 @@ body {
     font-family: 'Open Sans', sans-serif;
 }
 
+.logo {
+    width: 80%;
+    display: block;
+    margin: auto;
+    padding-bottom: 10px;
+}
+
 .search-container {
     width: 80%;
     position: absolute;

app/static/opensearch.template

@@ -0,0 +1,14 @@
+<OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/"
+                       xmlns:moz="http://www.mozilla.org/2006/browser/search/">
+  <ShortName>[SNK]</ShortName>
+  <Description>[Search engine full name and summary]</Description>
+  <InputEncoding>[UTF-8]</InputEncoding>
+  <Image width="16" height="16" type="image/x-icon">[https://example.com/favicon.ico]</Image>
+  <Url type="text/html" template="[searchURL]">
+    <Param name="[key name]" value="{searchTerms}"/>
+    <!-- other Params if you need them… -->
+    <Param name="[other key name]" value="[parameter value]"/>
+  </Url>
+  <Url type="application/x-suggestions+json" template="[suggestionURL]"/>
+  <moz:SearchForm>[https://example.com/search]</moz:SearchForm>
+</OpenSearchDescription>

app/templates/index.html

@@ -1,8 +1,11 @@
+
+<link rel="search" href="/static/opensearch.xml" type="application/opensearchdescription+xml" title="Bloogle Search">
 <meta name="viewport" content="width=device-width, initial-scale=1.0">
 <script src="/static/js/controller.js"></script>
 <link rel="stylesheet" href="/static/css/main.css">
 
 <div class="search-container">
+    <img class="logo" src="/static/img/logo.png">
     <div class="search-fields">
         <input type="text" id="search-bar">
         <button type="submit" id="search-submit">Search</button>