From 21012f5265dcc1541e5e296bcbb6ff54661f95a7 Mon Sep 17 00:00:00 2001
From: Ben Busby <benbusby@protonmail.com>
Date: Sun, 24 May 2020 14:03:11 -0600
Subject: [PATCH] Feature: autocomplete/search suggestions (#72)

Basic autocomplete/search suggestion functionality added

* Adds new GET and POST routes for '/autocomplete' that accept a string query and returns an array of suggestions

* Adds new autoscript.js file for handling queries on the main page and results view

* Updated requests class to include autocomplete method

* Updated opensearch template to handle search suggestions

* Added header template to allow for autocomplete on results view

* Updated readme to mention autocomplete feature
---
 README.md                     |  7 +--
 app/filter.py                 |  6 ++-
 app/request.py                | 21 ++++++--
 app/routes.py                 | 24 ++++++++-
 app/static/css/header.css     | 55 ++++++++++++++++++++
 app/static/css/search.css     | 34 ++++++++++++
 app/static/js/autocomplete.js | 98 +++++++++++++++++++++++++++++++++++
 app/static/js/controller.js   |  2 +
 app/templates/display.html    |  4 ++
 app/templates/header.html     | 53 +++++++++++++++++++
 app/templates/index.html      |  8 ++-
 app/templates/opensearch.xml  |  4 +-
 requirements.txt              |  1 +
 test/test_autocomplete.py     | 12 +++++
 14 files changed, 315 insertions(+), 14 deletions(-)
 create mode 100644 app/static/css/header.css
 create mode 100644 app/static/css/search.css
 create mode 100644 app/static/js/autocomplete.js
 create mode 100644 app/templates/header.html
 create mode 100644 test/test_autocomplete.py

diff --git a/README.md b/README.md
index f1876a0..a344f56 100644
--- a/README.md
+++ b/README.md
@@ -25,7 +25,8 @@ Contents
 - No AMP links
 - No URL tracking tags (i.e. utm=%s)
 - No referrer header
-- POST request search queries (when possible)
+- Autocomplete/search suggestions
+- POST request search and suggestion queries (when possible)
 - View images at full res without site redirect (currently mobile only)
 - Dark mode
 - Randomly generated User Agent
@@ -107,7 +108,7 @@ Description=Whoogle
 Type=simple
 User=root
 WorkingDirectory=<whoogle_directory>
-ExecStart=<whoogle_directory>/venv/bin/python3 -um app --host 0.0.0.0 --port 5000 
+ExecStart=<whoogle_directory>/venv/bin/python3 -um app --host 0.0.0.0 --port 5000
 ExecReload=/bin/kill -HUP $MAINPID
 Restart=always
 RestartSec=3
@@ -185,7 +186,7 @@ To filter by a range of time, append ":past <time>" to the end of your search, w
 
 ## Extra Steps
 ### Set Whoogle as your primary search engine
-*Note: If you're using a reverse proxy to run Whoogle Search, make sure the "Root URL" config option on the home page is set to your URL before going through these steps.* 
+*Note: If you're using a reverse proxy to run Whoogle Search, make sure the "Root URL" config option on the home page is set to your URL before going through these steps.*
 
 Update browser settings:
   - Firefox (Desktop)
diff --git a/app/filter.py b/app/filter.py
index 5c9ce2e..54029c5 100644
--- a/app/filter.py
+++ b/app/filter.py
@@ -96,9 +96,13 @@ class Filter:
             st_card.decompose()
 
         footer = soup.find('div', id='sfooter')
-        if footer is not None:
+        if footer:
             footer.decompose()
 
+        header = soup.find('header')
+        if header:
+            header.decompose()
+
         return soup
 
     def remove_ads(self, soup):
diff --git a/app/request.py b/app/request.py
index 07b6351..7ecd887 100644
--- a/app/request.py
+++ b/app/request.py
@@ -1,10 +1,12 @@
 from io import BytesIO
+from lxml import etree
 import pycurl
 import random
 import urllib.parse as urlparse
 
-# Base search url
+# Core Google search URLs
 SEARCH_URL = 'https://www.google.com/search?gbv=1&q='
+AUTOCOMPLETE_URL = 'https://suggestqueries.google.com/complete/search?client=toolbar&'
 
 MOBILE_UA = '{}/5.0 (Android 0; Mobile; rv:54.0) Gecko/54.0 {}/59.0'
 DESKTOP_UA = '{}/5.0 (X11; {} x86_64; rv:75.0) Gecko/20100101 {}/75.0'
@@ -13,9 +15,7 @@ DESKTOP_UA = '{}/5.0 (X11; {} x86_64; rv:75.0) Gecko/20100101 {}/75.0'
 VALID_PARAMS = ['tbs', 'tbm', 'start', 'near']
 
 
-def gen_user_agent(normal_ua):
-    is_mobile = 'Android' in normal_ua or 'iPhone' in normal_ua
-
+def gen_user_agent(normal_ua, is_mobile):
     mozilla = random.choice(['Moo', 'Woah', 'Bro', 'Slow']) + 'zilla'
     firefox = random.choice(['Choir', 'Squier', 'Higher', 'Wire']) + 'fox'
     linux = random.choice(['Win', 'Sin', 'Gin', 'Fin', 'Kin']) + 'ux'
@@ -64,8 +64,9 @@ def gen_query(query, args, config, near_city=None):
 
 class Request:
     def __init__(self, normal_ua, language='lang_en'):
-        self.modified_user_agent = gen_user_agent(normal_ua)
         self.language = language
+        self.mobile = 'Android' in normal_ua or 'iPhone' in normal_ua
+        self.modified_user_agent = gen_user_agent(normal_ua, self.mobile)
 
     def __getitem__(self, name):
         return getattr(self, name)
@@ -76,6 +77,16 @@ class Request:
         else:
             return 'unicode-escape'
 
+    def autocomplete(self, query):
+        ac_query = dict(hl=self.language, q=query)
+        response = self.send(base_url=AUTOCOMPLETE_URL, query=urlparse.urlencode(ac_query))
+
+        if response:
+            dom = etree.fromstring(response)
+            return dom.xpath('//suggestion/@data')
+
+        return []
+
     def send(self, base_url=SEARCH_URL, query='', return_bytes=False):
         response_header = []
 
diff --git a/app/routes.py b/app/routes.py
index 3f0d04c..7a4f4d2 100644
--- a/app/routes.py
+++ b/app/routes.py
@@ -6,7 +6,7 @@ import argparse
 import base64
 from bs4 import BeautifulSoup
 from cryptography.fernet import Fernet, InvalidToken
-from flask import g, make_response, request, redirect, render_template, send_file
+from flask import g, jsonify, make_response, request, redirect, render_template, send_file
 from functools import wraps
 import io
 import json
@@ -88,6 +88,19 @@ def opensearch():
     return response
 
 
+@app.route('/autocomplete', methods=['GET', 'POST'])
+def autocomplete():
+    request_params = request.args if request.method == 'GET' else request.form
+    q = request_params.get('q')
+
+    if not q and not request.data:
+        return jsonify({'?': []})
+    elif request.data:
+        q = urlparse.unquote_plus(request.data.decode('utf-8').replace('q=', ''))
+
+    return jsonify([q, g.user_request.autocomplete(q)])
+
+
 @app.route('/search', methods=['GET', 'POST'])
 @auth_required
 def search():
@@ -121,7 +134,14 @@ def search():
     else:
         formatted_results = content_filter.clean(dirty_soup)
 
-    return render_template('display.html', query=urlparse.unquote(q), response=formatted_results)
+    return render_template(
+        'display.html',
+        query=urlparse.unquote(q),
+        response=formatted_results,
+        search_header=render_template(
+            'header.html',
+            q=urlparse.unquote(q),
+            mobile=g.user_request.mobile))
 
 
 @app.route('/config', methods=['GET', 'POST'])
diff --git a/app/static/css/header.css b/app/static/css/header.css
new file mode 100644
index 0000000..c3eebf8
--- /dev/null
+++ b/app/static/css/header.css
@@ -0,0 +1,55 @@
+header {
+    font-family: Roboto,HelveticaNeue,Arial,sans-serif;
+    font-size: 14px;
+    line-height: 20px;
+    color: #3C4043;
+    word-wrap: break-word;
+}
+
+.logo-link, .logo-letter {
+    text-decoration: none !important;
+    letter-spacing: -1px;
+    text-align: center;
+    border-radius: 2px 0 0 0;
+}
+
+.mobile-logo {
+    font: 22px/36px Futura, Arial, sans-serif;
+    padding-left: 5px;
+}
+
+.logo-div {
+    letter-spacing: -1px;
+    text-align: center;
+    font: 22pt Futura, Arial, sans-serif;
+    padding: 10px 0 5px 0;
+    height: 37px;
+    font-smoothing: antialiased;
+}
+
+.search-div {
+    border-radius: 8px 8px 0 0;
+    box-shadow: 0 1px 6px rgba(32, 33, 36, 0.18);
+    margin-top: 10px;
+}
+
+.search-form {
+    height: 39px;
+    display: flex;
+    width: 100%;
+}
+
+.search-input {
+    background: none;
+    margin: 2px 4px 2px 8px;
+    display: block;
+    font-size: 16px;
+    padding: 0 0 0 8px;
+    flex: 1;
+    height: 35px;
+    outline: none;
+    border: none;
+    width: 100%;
+    -webkit-tap-highlight-color: rgba(0,0,0,0);
+    overflow: hidden;
+}
\ No newline at end of file
diff --git a/app/static/css/search.css b/app/static/css/search.css
new file mode 100644
index 0000000..a79522b
--- /dev/null
+++ b/app/static/css/search.css
@@ -0,0 +1,34 @@
+.autocomplete {
+    position: relative;
+    display: inline-block;
+    width: 100%;
+}
+
+.autocomplete-items {
+    position: absolute;
+    border: 1px solid #d4d4d4;
+    border-bottom: none;
+    border-top: none;
+    z-index: 99;
+
+    /*position the autocomplete items to be the same width as the container:*/
+    top: 100%;
+    left: 0;
+    right: 0;
+}
+
+.autocomplete-items div {
+    padding: 10px;
+    cursor: pointer;
+    background-color: #fff;
+    border-bottom: 1px solid #d4d4d4;
+}
+
+.autocomplete-items div:hover {
+    background-color: #e9e9e9;
+}
+
+.autocomplete-active {
+    background-color: #685e79 !important;
+    color: #ffffff;
+}
\ No newline at end of file
diff --git a/app/static/js/autocomplete.js b/app/static/js/autocomplete.js
new file mode 100644
index 0000000..316f8c4
--- /dev/null
+++ b/app/static/js/autocomplete.js
@@ -0,0 +1,98 @@
+const handleUserInput = searchBar => {
+    let xhrRequest = new XMLHttpRequest();
+    xhrRequest.open("POST", "/autocomplete");
+    xhrRequest.setRequestHeader("Content-type", "application/x-www-form-urlencoded");
+    xhrRequest.onload = function() {
+        if (xhrRequest.readyState === 4 && xhrRequest.status !== 200) {
+            alert("Error fetching autocomplete results");
+            return;
+        }
+
+        // Fill autocomplete with fetched results
+        let autocompleteResults = JSON.parse(xhrRequest.responseText);
+        autocomplete(searchBar, autocompleteResults[1]);
+    };
+
+    xhrRequest.send('q=' + searchBar.value);
+};
+
+const autocomplete = (searchInput, autocompleteResults) => {
+    let currentFocus;
+
+    searchInput.addEventListener("input", function () {
+        let autocompleteList, autocompleteItem, i, val = this.value;
+        closeAllLists();
+
+        if (!val || !autocompleteResults) {
+            return false;
+        }
+
+        currentFocus = -1;
+        autocompleteList = document.createElement("div");
+        autocompleteList.setAttribute("id", this.id + "-autocomplete-list");
+        autocompleteList.setAttribute("class", "autocomplete-items");
+        this.parentNode.appendChild(autocompleteList);
+
+        for (i = 0; i < autocompleteResults.length; i++) {
+            if (autocompleteResults[i].substr(0, val.length).toUpperCase() === val.toUpperCase()) {
+                autocompleteItem = document.createElement("div");
+                autocompleteItem.innerHTML = "<strong>" + autocompleteResults[i].substr(0, val.length) + "</strong>";
+                autocompleteItem.innerHTML += autocompleteResults[i].substr(val.length);
+                autocompleteItem.innerHTML += "<input type=\"hidden\" value=\"" + autocompleteResults[i] + "\">";
+                autocompleteItem.addEventListener("click", function () {
+                    searchInput.value = this.getElementsByTagName("input")[0].value;
+                    closeAllLists();
+                    document.getElementById("search-form").submit();
+                });
+                autocompleteList.appendChild(autocompleteItem);
+            }
+        }
+    });
+
+    searchInput.addEventListener("keydown", function (e) {
+        let suggestion = document.getElementById(this.id + "-autocomplete-list");
+        if (suggestion) suggestion = suggestion.getElementsByTagName("div");
+        if (e.keyCode === 40) { // down
+            currentFocus++;
+            addActive(suggestion);
+        } else if (e.keyCode === 38) { //up
+            currentFocus--;
+            addActive(suggestion);
+        } else if (e.keyCode === 13) { // enter
+            e.preventDefault();
+            if (currentFocus > -1) {
+                if (suggestion) suggestion[currentFocus].click();
+            }
+        }
+    });
+
+    const addActive = suggestion => {
+        if (!suggestion || !suggestion[currentFocus]) return false;
+        removeActive(suggestion);
+
+        if (currentFocus >= suggestion.length) currentFocus = 0;
+        if (currentFocus < 0) currentFocus = (suggestion.length - 1);
+
+        suggestion[currentFocus].classList.add("autocomplete-active");
+    };
+
+    const removeActive = suggestion => {
+        for (let i = 0; i < suggestion.length; i++) {
+            suggestion[i].classList.remove("autocomplete-active");
+        }
+    };
+
+    const closeAllLists = el => {
+        let suggestions = document.getElementsByClassName("autocomplete-items");
+        for (let i = 0; i < suggestions.length; i++) {
+            if (el !== suggestions[i] && el !== searchInput) {
+                suggestions[i].parentNode.removeChild(suggestions[i]);
+            }
+        }
+    };
+
+    // Close lists and search when user selects a suggestion
+    document.addEventListener("click", function (e) {
+        closeAllLists(e.target);
+    });
+};
\ No newline at end of file
diff --git a/app/static/js/controller.js b/app/static/js/controller.js
index b3b4f3f..4817195 100644
--- a/app/static/js/controller.js
+++ b/app/static/js/controller.js
@@ -11,6 +11,8 @@ const setupSearchLayout = () => {
         if (event.keyCode === 13) {
             event.preventDefault();
             searchBtn.click();
+        } else {
+            handleUserInput(searchBar);
         }
     });
 };
diff --git a/app/templates/display.html b/app/templates/display.html
index 03ed151..e66de6f 100644
--- a/app/templates/display.html
+++ b/app/templates/display.html
@@ -5,9 +5,13 @@
         <link rel="search" href="/opensearch.xml" type="application/opensearchdescription+xml" title="Whoogle Search">
         <meta name="viewport" content="width=device-width, initial-scale=1.0">
         <meta name="referrer" content="no-referrer">
+        <script type="text/javascript" src="/static/js/autocomplete.js"></script>
+        <link rel="stylesheet" href="/static/css/search.css">
+        <link rel="stylesheet" href="/static/css/header.css">
         <title>{{ query }} - Whoogle Search</title>
     </head>
     <body>
+        {{ search_header|safe }}
         {{ response|safe }}
     </body>
 </html>
diff --git a/app/templates/header.html b/app/templates/header.html
new file mode 100644
index 0000000..e106ffe
--- /dev/null
+++ b/app/templates/header.html
@@ -0,0 +1,53 @@
+{% if mobile %}
+    <header>
+        <div class="bz1lBb">
+            <form class="Pg70bf" id="search-form" method="POST">
+                <a class="logo-link mobile-logo"
+                   href="/"
+                   style="display:flex; justify-content:center; align-items:center; color:#685e79; font-size:18px; ">
+                    <span class="V6gwVd">Wh</span><span class="iWkuvd">o</span><span class="cDrQ7">o</span><span
+                        class="V6gwVd">g</span><span class="ntlR9">l</span><span
+                        class="iWkuvd tJ3Myc">e</span>
+                </a>
+                <div class="H0PQec" style="width: 100%;">
+                    <div class="sbc esbc autocomplete">
+                        <input id="search-bar" autocapitalize="none" autocomplete="off" class="noHIxc" name="q"
+                               spellcheck="false" type="text" value="{{ q }}">
+                        <div class="sc"></div>
+                    </div>
+                </div>
+            </form>
+        </div>
+    </header>
+{% else %}
+    <header>
+        <div class="logo-div">
+            <a class="logo-link" href="/">
+                <span class="V6gwVd logo-letter">Wh</span><span class="iWkuvd logo-letter">o</span><span
+                    class="cDrQ7 logo-letter">o</span><span class="V6gwVd logo-letter">g</span><span
+                    class="ntlR9 logo-letter">l</span><span class="iWkuvd tJ3Myc logo-letter">e</span>
+            </a>
+        </div>
+        <div class="search-div">
+            <form id="search-form" class="search-form" id="sf" method="POST">
+                <div class="autocomplete" style="width: 100%; flex: 1">
+                    <div style="width: 100%; display: flex">
+                        <input id="search-bar" autocapitalize="none" autocomplete="off" class="noHIxc" name="q"
+                               spellcheck="false" type="text" value="{{ q }}">
+                        <div class="sc"></div>
+                    </div>
+                </div>
+            </form>
+        </div>
+    </header>
+{% endif %}
+
+<script>
+    const searchBar = document.getElementById("search-bar");
+
+    searchBar.addEventListener("keyup", function (event) {
+        if (event.keyCode !== 13) {
+            handleUserInput(searchBar);
+        }
+    });
+</script>
\ No newline at end of file
diff --git a/app/templates/index.html b/app/templates/index.html
index af1879f..ab46526 100644
--- a/app/templates/index.html
+++ b/app/templates/index.html
@@ -17,18 +17,22 @@
         <meta name="referrer" content="no-referrer">
         <meta name="msapplication-TileColor" content="#ffffff">
         <meta name="msapplication-TileImage" content="/static/img/favicon/ms-icon-144x144.png">
+        <script type="text/javascript" src="/static/js/autocomplete.js"></script>
         <script type="text/javascript" src="/static/js/controller.js"></script>
         <link rel="search" href="/opensearch.xml" type="application/opensearchdescription+xml" title="Whoogle Search">
         <meta name="viewport" content="width=device-width, initial-scale=1.0">
+        <link rel="stylesheet" href="/static/css/search.css">
         <link rel="stylesheet" href="/static/css/main.css">
 	<title>Whoogle Search</title>
     </head>
     <body id="main" style="display: none; background-color: {{ bg }}">
         <div class="search-container">
             <img class="logo" src="/static/img/logo.png">
-            <form action="/search" method="{{ request_type }}">
+            <form id="search-form" action="/search" method="{{ request_type }}">
                 <div class="search-fields">
-                    <input type="text" name="q" id="search-bar" autofocus="autofocus">
+                    <div class="autocomplete">
+                        <input type="text" name="q" id="search-bar" autofocus="autofocus">
+                    </div>
                     <input type="submit" id="search-submit" value="Search">
                 </div>
             </form>
diff --git a/app/templates/opensearch.xml b/app/templates/opensearch.xml
index bf44b00..b737be7 100644
--- a/app/templates/opensearch.xml
+++ b/app/templates/opensearch.xml
@@ -7,7 +7,9 @@
   <Url type="text/html" method="{{ request_type }}" template="{{ main_url }}/search">
     <Param name="q" value="{searchTerms}"/>
   </Url>
-  <Url type="application/x-suggestions+json" template="{{ main_url }}/search"/>
+  <Url type="application/x-suggestions+json" method="{{ request_type }}" template="{{ main_url }}/autocomplete">
+    <Param name="q" value="{searchTerms}"/>
+  </Url>
   <moz:SearchForm>{{ main_url }}/search</moz:SearchForm>
 </OpenSearchDescription>
 
diff --git a/requirements.txt b/requirements.txt
index eac1d8b..030780c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -6,6 +6,7 @@ cryptography==2.8
 Flask==1.1.1
 itsdangerous==1.1.0
 Jinja2==2.10.3
+lxml==4.5.1
 MarkupSafe==1.1.1
 pycparser==2.19
 pycurl==7.43.0.4
diff --git a/test/test_autocomplete.py b/test/test_autocomplete.py
new file mode 100644
index 0000000..4f55f6d
--- /dev/null
+++ b/test/test_autocomplete.py
@@ -0,0 +1,12 @@
+def test_autocomplete_get(client):
+    rv = client.get('/autocomplete?q=green+eggs+and')
+    assert rv._status_code == 200
+    assert len(rv.data) >= 1
+    assert b'green eggs and ham' in rv.data
+
+
+def test_autocomplete_post(client):
+    rv = client.post('/autocomplete', data=dict(q='the+cat+in+the'))
+    assert rv._status_code == 200
+    assert len(rv.data) >= 1
+    assert b'the cat in the hat' in rv.data