get rid of old about and donate pages

Signed-off-by: ngn <ngn@ngn.tf>

.gitignore

@@ -1,28 +1,5 @@
-lib/test.html
+/compose.yml
-lib/postdata.json
+/docker-compose.yml
-lib/nextpage.json
+/banner
-scraper/brave.html
+/favicon.ico
-scraper/yandex.json
+/config.php
-scraper/marginalia.json
-banner_og/
-scraper/mojeek.html
-scraper/google.html
-scraper/google-img.html
-scraper/google-video.html
-scraper/google-news.html
-scraper/google-img-nextpage.html
-scraper/brave-image.html
-scraper/brave-video.html
-scraper/facebook.html
-scraper/facebook-nextpage.json
-scraper/yandex-video.json
-scraper/yandex.html
-scraper/soundcloud.json
-scraper/mp3-pm.html
-banner/*
-data/captcha/birds/
-data/captcha/fumo_plushies/
-data/captcha/minecraft/
-!banner/*default*
-scraper/curlie.html
-icons/*

Dockerfile

@@ -1,24 +1,20 @@
-FROM alpine:latest
+FROM alpine
 
-RUN apk update
+RUN apk update && \
-RUN apk upgrade
+  apk upgrade && \
-
+  apk add \
-RUN apk add php apache2-ssl php83-fileinfo php83-openssl    \
+    php php83-fileinfo php83-iconv php83-common php83-dom php83-sodium \
-            php83-iconv php83-common php83-dom php83-sodium \
+    php83-curl php83-pecl-apcu php83-apache2 php-mbstring \
-            php83-curl curl php83-pecl-apcu php83-apache2   \
+    php83-pecl-imagick imagemagick-webp imagemagick-jpeg
-            imagemagick php83-pecl-imagick php-mbstring     \
-            imagemagick-webp imagemagick-jpeg
 
 COPY ./docker/httpd.conf /etc/apache2/httpd.conf
 COPY ./docker/init.sh    /
 
-WORKDIR /var/www/html
+WORKDIR /srv
 COPY ./src ./4get
 
-WORKDIR /var/www/html/4get
+RUN adduser -DSH -u 1000 -h /srv runner
-COPY ./docker/gen_config.php .
+RUN chown -R runner /srv && chmod +x /init.sh
 
-RUN chmod 777 /var/www/html/4get/icons
+USER runner
-RUN chmod +x /init.sh
+CMD ["/init.sh"]
-
-CMD  ["/init.sh"]

compose.example.yml

@@ -0,0 +1,18 @@
+services:
+  fourget:
+    container_name: 4get
+    image: git.ngn.tf/ngn/4get
+    ports:
+      - 80:8080
+    volumes:
+      - ./config.php:/srv/4get/data/config.php:ro
+      - ./favicon.ico:/srv/4get/favicon.ico:ro
+      - ./banner:/srv/4get/banner:ro
+      - type: tmpfs
+        target: /tmp/icons
+    cap_drop:
+      - ALL
+    security_opt:
+      - no-new-privileges:true
+    read_only: true
+    restart: unless-stopped

docker-compose.example.yml

@@ -1,12 +0,0 @@
-services:
-  fourget:
-    container_name: 4get
-    image: git.ngn.tf/ngn/4get
-    environment:
-    - FOURGET_SERVER_NAME=example.com
-    ports:
-    - 80:80
-    volumes:
-    - ./banners:/var/www/html/4get/banner
-    - ./captcha:/var/www/html/4get/data/captcha
-    restart: unless-stopped

docker/gen_config.php

@@ -1,90 +0,0 @@
- <?php
-
-include "/var/www/html/4get/data/config.php";
-
-$refl = new ReflectionClass('config');
-$from_config = ($refl->getConstants());
-$from_env = array();
-
-$env = getenv();
-$fourget_env = array_filter($env, function($v, $k) {
-        return str_starts_with($k, "FOURGET");
-}, ARRAY_FILTER_USE_BOTH);
-
-foreach($fourget_env as $key => $val)  {
-        $target_key = preg_replace('/^FOURGET_/', '', $key);
-        $from_env[$target_key] = trim($val, '\'"');
-};
-
-$merged_config = array_merge($from_config, $from_env);
-
-function type_to_string($n) {
-        $type = gettype($n);
-        if ($type === "NULL") {
-                return "null";
-        }
-        if ($type === "boolean") {
-                return $n ? 'true' : 'false';
-        }
-        if ($type === "string") {
-                if(is_numeric($n)) {
-                        return $n;
-                }
-                return "\"$n\"";
-        }
-        if ($type === "array") {
-                return json_encode($n, JSON_UNESCAPED_SLASHES);
-        }
-        return $n;
-}
-
-
-function detect_captcha_dirs() {
-        $captcha_dir = "/var/www/html/4get/data/captcha/";
-        $categories = (array_map(function ($n) {
-                return explode("/", $n)[7];
-        }, glob($captcha_dir . "*")));
-
-
-        $result = array_map(function($category) {
-                return [$category, count(glob("/var/www/html/4get/data/captcha/" . $category . "/*" ))];
-        }, $categories);
-
-        return $result;
-}
-
-
-$special_keys = ["PROTO", "CAPTCHA_DATASET"];
-
-$output = "<?php\n // This file was generated by docker/gen_config.php\n";
-
-$output = $output . "class config {\n";
-foreach(($merged_config) as $key => $val){
-        if(!in_array($key, $special_keys)) {
-            $stored_value = $val;
-            // conversion between arrays and comma separated env value.
-            // Handle case when original type of field is array and there is a type mismatch when a comma separted string is passed,
-            // then split on comma if string (and not numeric, boolean, null, etc)
-            //
-            // except in the case where the inital value in default config is null or boolean. Assuming null and boolean
-            // in default config will be never be assigned an array
-
-            if(gettype($from_config[$key]) != gettype($val) && !is_numeric($val) && !is_null($from_config[$key]) && gettype($from_config[$key]) != "boolean") {
-                $stored_value = explode(",", $val);
-            }
-            $output = $output . "\tconst " . $key . " = " . type_to_string($stored_value) . ";\n";
-
-            continue;
-        }
-
-
-        if($key === "CAPTCHA_DATASET") {
-                $output = $output . "\tconst " . $key . " = " . type_to_string(detect_captcha_dirs()) . ";\n";
-        }
-}
-
-$output = $output . "}\n";
-$output = $output . "?>";
-
-file_put_contents("./data/config.php", $output);
-?>

docker/httpd.conf

@@ -1,16 +1,19 @@
-Listen 80
+Listen 8080
-ServerTokens OS
+
 ServerRoot /var/www
-ServerSignature On
 ServerName localhost
 
-DocumentRoot "/var/www/html/4get"
+ServerSignature Off
+ServerTokens    Prod
 
-LogLevel error
+PidFile      /dev/shm/httpd.pid
+DocumentRoot /srv/4get
+
+LogLevel  error
 CustomLog /dev/null common
-ErrorLog /dev/null
+ErrorLog  /dev/stderr
 
-<Directory "/var/www/html/4get">
+<Directory /srv/4get>
   RewriteEngine On
   RewriteCond %{THE_REQUEST} ^\w+\ /(.*)\.php(\?.*)?\ HTTP/
   RewriteRule ^ http://%{HTTP_HOST}/%1 [R=301]
@@ -22,68 +25,32 @@ ErrorLog /dev/null
 </Directory>
 
 # deny access to private resources
-<Directory "/var/www/html/4get/data">
+<Directory /srv/4get/data>
   Require all denied
-  <Files "*">
+  <Files *>
     Require all denied
   </Files>
 </Directory>
 
-LoadModule rewrite_module modules/mod_rewrite.so
+LoadModule rewrite_module       modules/mod_rewrite.so
-LoadModule mpm_prefork_module modules/mod_mpm_prefork.so
+LoadModule mpm_prefork_module   modules/mod_mpm_prefork.so
-LoadModule authn_file_module modules/mod_authn_file.so
+LoadModule authz_core_module    modules/mod_authz_core.so
-LoadModule authn_core_module modules/mod_authn_core.so
-LoadModule authz_host_module modules/mod_authz_host.so
-LoadModule authz_groupfile_module modules/mod_authz_groupfile.so
-LoadModule authz_user_module modules/mod_authz_user.so
-LoadModule authz_core_module modules/mod_authz_core.so
 LoadModule access_compat_module modules/mod_access_compat.so
-LoadModule auth_basic_module modules/mod_auth_basic.so
+LoadModule filter_module        modules/mod_filter.so
-LoadModule reqtimeout_module modules/mod_reqtimeout.so
+LoadModule mime_module          modules/mod_mime.so
-LoadModule filter_module modules/mod_filter.so
+LoadModule log_config_module    modules/mod_log_config.so
-LoadModule mime_module modules/mod_mime.so
+LoadModule unixd_module         modules/mod_unixd.so
-LoadModule log_config_module modules/mod_log_config.so
+LoadModule negotiation_module   modules/mod_negotiation.so
-LoadModule env_module modules/mod_env.so
+LoadModule dir_module           modules/mod_dir.so
-LoadModule headers_module modules/mod_headers.so
-LoadModule setenvif_module modules/mod_setenvif.so
-LoadModule version_module modules/mod_version.so
-LoadModule unixd_module modules/mod_unixd.so
-LoadModule status_module modules/mod_status.so
-LoadModule autoindex_module modules/mod_autoindex.so
-LoadModule dir_module modules/mod_dir.so
-LoadModule alias_module modules/mod_alias.so
-LoadModule negotiation_module modules/mod_negotiation.so
-
-<IfModule unixd_module>
-  User apache
-  Group apache
-</IfModule>
 
 <Directory />
   AllowOverride none
   Require all denied
 </Directory>
 
-<IfModule dir_module>
-  DirectoryIndex index.html
-</IfModule>
-
 <Files ".ht*">
   Require all denied
 </Files>
 
-<IfModule headers_module>
+Include /etc/apache2/conf.d/languages.conf
-  RequestHeader unset Proxy early
+Include /etc/apache2/conf.d/php83-module.conf
-</IfModule>
-
-<IfModule mime_module>
-  TypesConfig /etc/apache2/mime.types
-  AddType application/x-compress .Z
-  AddType application/x-gzip .gz .tgz
-</IfModule>
-
-<IfModule mime_magic_module>
-  MIMEMagicFile /etc/apache2/magic
-</IfModule>
-
-IncludeOptional /etc/apache2/conf.d/*.conf

docker/init.sh

@@ -1,11 +1,17 @@
-#!/bin/sh
+#!/bin/sh -e
-set -e
 
-if [ ! -f '/var/www/html/4get/data/config.php' ] && [ -f './gen_config.php' ]
+config='/srv/4get/data/config.php'
-then
+defconfig='/srv/4get/data/config.def.php'
-  php ./gen_config.php
+
-  rm -f ./gen_config.php
+# check for the configuration file
+if [ ! -f "${config}" ]; then
+  echo "configuration file not specified"
+  echo "here's the default configuration, modify and mount this to ${config}"
+  echo
+  cat "${defconfig}"
+  exit 1
 fi
 
-echo "Starting up apache2"
+# execute apache
-exec httpd -DFOREGROUND
+echo "starting apache web server"
+exec httpd -D FOREGROUND

src/.gitignore

@@ -0,0 +1,28 @@
+lib/test.html
+lib/postdata.json
+lib/nextpage.json
+
+scraper/brave.html
+scraper/yandex.json
+scraper/marginalia.json
+scraper/mojeek.html
+scraper/google.html
+scraper/google-img.html
+scraper/google-video.html
+scraper/google-news.html
+scraper/google-img-nextpage.html
+scraper/brave-image.html
+scraper/brave-video.html
+scraper/facebook.html
+scraper/facebook-nextpage.json
+scraper/yandex-video.json
+scraper/yandex.html
+scraper/soundcloud.json
+scraper/mp3-pm.html
+scraper/curlie.html
+
+banner/*
+data/*
+
+!banner/*default*
+!data/*.def.*

src/about.php

@@ -1,39 +0,0 @@
-<?php
-
-include "data/config.php";
-include "lib/frontend.php";
-$frontend = new frontend();
-
-echo
-	$frontend->load(
-		"header_nofilters.html",
-		[
-			"title" => "About",
-			"class" => " class=\"about\""
-		]
-	);
-
-$left =
-	explode(
-		"\n",
-		file_get_contents("template/about.html")
-	);
-
-$out = "";
-
-foreach($left as $line){
-	
-	$out .= trim($line);
-}
-
-echo
-	$frontend->load(
-		"search.html",
-		[
-			"timetaken" => null,
-			"class" => "",
-			"right-left" => "",
-			"right-right" => "",
-			"left" => $out
-		]
-	);

src/data/config.def.php

@@ -0,0 +1,87 @@
+<?php
+class config{
+	// Welcome to the 4get configuration file
+	// When updating your instance, please make sure this file isn't missing
+	// any parameters.
+
+	// 4get version. Please keep this updated
+	const VERSION = 8;
+
+	// Will be shown pretty much everywhere.
+	const SERVER_NAME = "4get";
+
+	// Will be shown in <meta> tag on home page
+	const SERVER_SHORT_DESCRIPTION = "4get is a proxy search engine that doesn't suck.";
+
+	// Will be shown in server list ping (null for no description)
+	const SERVER_LONG_DESCRIPTION = null;
+
+	// Add your own themes in "static/themes". Set to "Dark" for default theme.
+	// Eg. To use "static/themes/Cream.css", specify "Cream".
+	const DEFAULT_THEME = "black";
+
+	// Default user agent to use for scraper requests. Sometimes ignored to get specific webpages
+	// Changing this might break things.
+	const USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:134.0) Gecko/20100101 Firefox/134.0";
+
+	// Temporary directory for saving the page icons
+	const ICON_DIR = "/tmp/icons";
+
+	// List of domains that point to your servers. Include your tor/i2p
+	// addresses here! Must be a valid URL. Won't affect links placed on
+	// the homepage.
+	const ALT_ADDRESSES = [
+		//"https://4get.alt-tld",
+		//"http://4getwebfrq5zr4sxugk6htxvawqehxtdgjrbcn2oslllcol2vepa23yd.onion"
+	];
+
+	// Proxy pool assignments for each scraper
+	// false = Use server's raw IP
+	// string = will load a proxy list from data directory
+	// Eg. "tor" will load data/tor.txt
+	const PROXY_DDG = false; // duckduckgo
+	const PROXY_BRAVE = false;
+	const PROXY_FB = false; // facebook
+	const PROXY_GOOGLE = false;
+	const PROXY_GOOGLE_API = false;
+	const PROXY_GOOGLE_CSE = false;
+	const PROXY_STARTPAGE = false;
+	const PROXY_QWANT = false;
+	const PROXY_GHOSTERY = false;
+	const PROXY_MARGINALIA = false;
+	const PROXY_MOJEEK = false;
+	const PROXY_SC = false; // soundcloud
+	const PROXY_SPOTIFY = false;
+	const PROXY_SOLOFIELD = false;
+	const PROXY_WIBY = false;
+	const PROXY_CURLIE = false;
+	const PROXY_YT = false; // youtube
+	const PROXY_YEP = false;
+	const PROXY_PINTEREST = false;
+	const PROXY_SANKAKUCOMPLEX = false;
+	const PROXY_FLICKR = false;
+	const PROXY_FIVEHPX = false;
+	const PROXY_VSCO = false;
+	const PROXY_SEZNAM = false;
+	const PROXY_NAVER = false;
+	const PROXY_GREPPR = false;
+	const PROXY_CROWDVIEW = false;
+	const PROXY_MWMBL = false;
+	const PROXY_FTM = false; // findthatmeme
+	const PROXY_IMGUR = false;
+	const PROXY_YANDEX_W = false; // yandex web
+	const PROXY_YANDEX_I = false; // yandex images
+	const PROXY_YANDEX_V = false; // yandex videos
+
+	//
+	// Scraper-specific parameters
+	//
+
+	// GOOGLE CSE & GOOGLE API
+	const GOOGLE_CX_ENDPOINT = "d4e68b99b876541f0";
+
+	// MARGINALIA
+	// Use "null" to default out to HTML scraping OR specify a string to
+	// use the API (Eg: "public"). API has less filters.
+	const MARGINALIA_API_KEY = null;
+}

src/data/config.php

@@ -1,173 +0,0 @@
-<?php
-class config{
-	// Welcome to the 4get configuration file
-	// When updating your instance, please make sure this file isn't missing
-	// any parameters.
-
-	// 4get version. Please keep this updated
-	const VERSION = 8;
-
-	// Will be shown pretty much everywhere.
-	const SERVER_NAME = "4get";
-
-	// Will be shown in <meta> tag on home page
-	const SERVER_SHORT_DESCRIPTION = "4get is a proxy search engine that doesn't suck.";
-
-	// Will be shown in server list ping (null for no description)
-	const SERVER_LONG_DESCRIPTION = null;
-
-	// Add your own themes in "static/themes". Set to "Dark" for default theme.
-	// Eg. To use "static/themes/Cream.css", specify "Cream".
-	const DEFAULT_THEME = "black";
-
-	// Enable the API?
-	const API_ENABLED = true;
-
-	//
-	// BOT PROTECTION
-	//
-
-	// 0 = disabled, 1 = ask for image captcha, @TODO: 2 = invite only (users needs a pass)
-	// VERY useful against a targetted attack
-	const BOT_PROTECTION = 0;
-
-	// if BOT_PROTECTION is set to 1, specify the available datasets here
-	// images should be named from 1.png to X.png, and be 100x100 in size
-	// Eg. data/captcha/birds/1.png up to 2263.png
-	const CAPTCHA_DATASET = [
-		// example:
-		//["birds", 2263],
-		//["fumo_plushies", 1006],
-		//["minecraft", 848]
-	];
-
-	// If this regex expression matches on the user agent, it blocks the request
-	// Not useful at all against a targetted attack
-	const HEADER_REGEX = '/bot|wget|curl|python-requests|scrapy|go-http-client|ruby|yahoo|spider|qwant/i';
-
-	// Block clients who present any of the following headers in their request (SPECIFY IN !!lowercase!!)
-	// Eg: ["x-forwarded-for", "x-via", "forwarded-for", "via"];
-	// Useful for blocking *some* proxies used for botting
-	const FILTERED_HEADER_KEYS = [
-		//"x-forwarded-for",
-		//"x-cluster-client-ip",
-		//"x-client-ip",
-		//"x-real-ip",
-		//"client-ip",
-		//"real-ip",
-		//"forwarded-for",
-		//"forwarded-for-ip",
-		//"forwarded",
-		//"proxy-connection",
-		//"remote-addr",
-		//"via"
-	];
-
-	// Block SSL ciphers used by CLI tools used for botting
-	// Basically a primitive version of Cloudflare's browser integrity check
-	// ** If curl can still access the site (with spoofed headers), please make sure you use the new apache2 config **
-	// https://git.lolcat.ca/lolcat/4get/docs/apache2.md
-	const DISALLOWED_SSL = [
-		// "TLS_AES_256_GCM_SHA384" // used by WGET and CURL
-	];
-
-	// Maximal number of searches per captcha key/pass issued. Counter gets
-	// reset on every APCU cache clear (should happen once a day).
-	// Only useful when BOT_PROTECTION is NOT set to 0
-	const MAX_SEARCHES = 100;
-
-	// List of domains that point to your servers. Include your tor/i2p
-	// addresses here! Must be a valid URL. Won't affect links placed on
-	// the homepage.
-	const ALT_ADDRESSES = [
-		//"https://4get.alt-tld",
-		//"http://4getwebfrq5zr4sxugk6htxvawqehxtdgjrbcn2oslllcol2vepa23yd.onion"
-	];
-
-	// Known 4get instances. MUST use the https protocol if your instance uses
-	// it. Is used to generate a distributed list of instances.
-	// To appear in the list of an instance, contact the host and if everyone added
-	// eachother your serber should appear everywhere.
-	const INSTANCES = [
-		"https://4get.ca",
-		"https://4get.zzls.xyz",
-		"https://4getus.zzls.xyz",
-		"https://4get.silly.computer",
-		"https://4get.konakona.moe",
-		"https://4get.lvkaszus.pl",
-		"https://4g.ggtyler.dev",
-		"https://4get.perennialte.ch",
-		"https://4get.sijh.net",
-		"https://4get.hbubli.cc",
-		"https://4get.plunked.party",
-		"https://4get.seitan-ayoub.lol",
-		"https://4get.etenie.pl",
-		"https://4get.lunar.icu",
-		"https://4get.dcs0.hu",
-		"https://4get.kizuki.lol",
-		"https://4get.psily.garden",
-		"https://search.milivojevic.in.rs",
-		"https://4get.snine.nl",
-		"https://4get.datura.network",
-		"https://4get.neco.lol",
-		"https://4get.lol",
-		"https://4get.ch",
-		"https://4get.edmateo.site",
-		"https://4get.sudovanilla.org",
-		"https://search.mint.lgbt"
-	];
-
-	// Default user agent to use for scraper requests. Sometimes ignored to get specific webpages
-	// Changing this might break things.
-	const USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:134.0) Gecko/20100101 Firefox/134.0";
-
-	// Proxy pool assignments for each scraper
-	// false = Use server's raw IP
-	// string = will load a proxy list from data/proxies
-	// Eg. "onion" will load data/proxies/onion.txt
-	const PROXY_DDG = false; // duckduckgo
-	const PROXY_BRAVE = false;
-	const PROXY_FB = false; // facebook
-	const PROXY_GOOGLE = false;
-	const PROXY_GOOGLE_API = false;
-	const PROXY_GOOGLE_CSE = false;
-	const PROXY_STARTPAGE = false;
-	const PROXY_QWANT = false;
-	const PROXY_GHOSTERY = false;
-	const PROXY_MARGINALIA = false;
-	const PROXY_MOJEEK = false;
-	const PROXY_SC = false; // soundcloud
-	const PROXY_SPOTIFY = false;
-	const PROXY_SOLOFIELD = false;
-	const PROXY_WIBY = false;
-	const PROXY_CURLIE = false;
-	const PROXY_YT = false; // youtube
-	const PROXY_YEP = false;
-	const PROXY_PINTEREST = false;
-	const PROXY_SANKAKUCOMPLEX = false;
-	const PROXY_FLICKR = false;
-	const PROXY_FIVEHPX = false;
-	const PROXY_VSCO = false;
-	const PROXY_SEZNAM = false;
-	const PROXY_NAVER = false;
-	const PROXY_GREPPR = false;
-	const PROXY_CROWDVIEW = false;
-	const PROXY_MWMBL = false;
-	const PROXY_FTM = false; // findthatmeme
-	const PROXY_IMGUR = false;
-	const PROXY_YANDEX_W = false; // yandex web
-	const PROXY_YANDEX_I = false; // yandex images
-	const PROXY_YANDEX_V = false; // yandex videos
-
-	//
-	// Scraper-specific parameters
-	//
-
-	// GOOGLE CSE & GOOGLE API
-	const GOOGLE_CX_ENDPOINT = "d4e68b99b876541f0";
-
-	// MARGINALIA
-	// Use "null" to default out to HTML scraping OR specify a string to
-	// use the API (Eg: "public"). API has less filters.
-	const MARGINALIA_API_KEY = null;
-}

src/data/proxies/.gitignore

@@ -1,3 +0,0 @@
-*
-!.gitignore
-!onion.txt

src/donate.php

@@ -1,39 +0,0 @@
-<?php
-
-include "data/config.php";
-include "lib/frontend.php";
-$frontend = new frontend();
-
-echo
-	$frontend->load(
-		"header_nofilters.html",
-		[
-			"title" => "Donate to the project",
-			"class" => " class=\"about\""
-		]
-	);
-
-$left =
-	explode(
-		"\n",
-		file_get_contents("template/donate.html")
-	);
-
-$out = "";
-
-foreach($left as $line){
-	
-	$out .= trim($line);
-}
-
-echo
-	$frontend->load(
-		"search.html",
-		[
-			"timetaken" => null,
-			"class" => "",
-			"right-left" => "",
-			"right-right" => "",
-			"left" => $out
-		]
-	);

src/favicon.php

@@ -1,7 +1,7 @@
 <?php
 
 if(!isset($_GET["s"])){
-	
+
 	header("X-Error: Missing parameter (s)ite");
 	die();
 }
@@ -10,86 +10,92 @@ include "data/config.php";
 new favicon($_GET["s"]);
 
 class favicon{
-	
+
 	public function __construct($url){
 
 		header("Content-Type: image/png");
 
-		if(
+		// first check the URL with regex
-			preg_match(
+		if(preg_match('/^https?:\/\/[A-Za-z0-9.-]+$/', $url) === 0){
-				'/^https?:\/\/[A-Za-z0-9.-]+$/',
-				$url
-			) === 0
-		){
-
 			header("X-Error: Only provide the protocol and domain");
 			$this->defaulticon();
 		}
-		
+
-		$filename = str_replace(["https://", "http://"], "", $url);
+		// validate the URL
-		header("Content-Disposition: inline; filename=\"{$filename}.png\"");
+		$url = filter_var($url, FILTER_VALIDATE_URL);
-		
+		if(!$url) {
+			header("X-Error: Invalid URL");
+			$this->defaulticon();
+		}
+
+		// extract the hostname
+		$this->filename = parse_url($url, PHP_URL_HOST);
+		if(!$this->filename || is_null($this->filename) || $this->filename === ""){
+			header("X-Error: Invalid URL");
+			$this->defaulticon();
+		}
+
+		// specify the filename in content-disposition
+		header("Content-Disposition: inline; filename=\"{$this->filename}.png\"");
+
 		include "lib/curlproxy.php";
 		$this->proxy = new proxy(false);
-		
+
-		$this->filename = parse_url($url, PHP_URL_HOST);
-		
 		/*
 			Check if we have the favicon stored locally
 		*/
-		if(file_exists("icons/" . $filename . ".png")){
+		if(file_exists($this->iconpath())){
-				
+			$handle = fopen($this->iconpath(), "r");
-			$handle = fopen("icons/" . $filename . ".png", "r");
+			echo fread($handle, filesize($this->iconpath()));
-			echo fread($handle, filesize("icons/" . $filename . ".png"));
 			fclose($handle);
 			return;
 		}
-		
+
 		/*
 			Scrape html
 		*/
 		try{
-			
+
 			$payload = $this->proxy->get($url, $this->proxy::req_web, true);
-			
+
 		}catch(Exception $error){
-			
+
 			header("X-Error: Could not fetch HTML (" . $error->getMessage() . ")");
 			$this->favicon404();
 		}
 		//$payload["body"] = '<link rel="manifest" id="MANIFEST_LINK" href="/data/manifest/" crossorigin="use-credentials" />';
-		
+
 		// get link tags
 		preg_match_all(
 			'/< *link +(.*)[\/]?>/Uixs',
 			$payload["body"],
 			$linktags
 		);
-		
+
 		/*
 			Get relevant tags
 		*/
-		
+
 		$linktags = $linktags[1];
 		$attributes = [];
-		
+
 		/*
 		header("Content-Type: text/plain");
 		print_r($linktags);
 		print_r($payload);
 		die();*/
-		
+
 		for($i=0; $i<count($linktags); $i++){
-			
+
 			// get attributes
 			preg_match_all(
 				'/([A-Za-z0-9]+) *= *("[^"]*"|[^" ]+)/s',
 				$linktags[$i],
 				$tags
 			);
-			
+
 			for($k=0; $k<count($tags[1]); $k++){
-				
+
 				$attributes[$i][] = [
 					"name" => $tags[1][$k],
 					"value" => trim($tags[2][$k], "\" \n\r\t\v\x00")
@@ -101,22 +107,22 @@ class favicon{
 		unset($linktags);
 
 		$href = [];
-		
+
 		// filter out the tags we want
 		foreach($attributes as &$group){
-			
+
 			$tmp_href = null;
 			$tmp_rel = null;
 			$badtype = false;
-			
+
 			foreach($group as &$attribute){
-				
+
 				switch($attribute["name"]){
-					
+
 					case "rel":
-						
+
 						$attribute["value"] = strtolower($attribute["value"]);
-						
+
 						if(
 							(
 								$attribute["value"] == "icon" ||
@@ -126,49 +132,49 @@ class favicon{
 								$attribute["value"] == "mask-icon"
 							) === false
 						){
-							
+
 							break;
 						}
-						
+
 						$tmp_rel = $attribute["value"];
 						break;
-					
+
 					case "type":
 						$attribute["value"] = explode("/", $attribute["value"], 2);
-						
+
 						if(strtolower($attribute["value"][0]) != "image"){
-							
+
 							$badtype = true;
 							break;
 						}
 						break;
-					
+
 					case "href":
-						
+
 						// must not contain invalid characters
 						// must be bigger than 1
 						if(
 							filter_var($attribute["value"], FILTER_SANITIZE_URL) == $attribute["value"] &&
 							strlen($attribute["value"]) > 0
 						){
-							
+
 							$tmp_href = $attribute["value"];
 							break;
 						}
 						break;
 				}
 			}
-			
+
 			if(
 				$badtype === false &&
 				$tmp_rel !== null &&
 				$tmp_href !== null
 			){
-				
+
 				$href[$tmp_rel] = $tmp_href;
 			}
 		}
-		
+
 		/*
 			Priority list
 		*/
@@ -176,37 +182,37 @@ class favicon{
 		header("Content-Type: text/plain");
 		print_r($href);
 		die();*/
-		
+
 		if(isset($href["icon"])){ $href = $href["icon"]; }
 		elseif(isset($href["apple-touch-icon"])){ $href = $href["apple-touch-icon"]; }
 		elseif(isset($href["manifest"])){
-			
+
 			// attempt to parse manifest, but fallback to []
 			$href = $this->parsemanifest($href["manifest"], $url);
 		}
-		
+
 		if(is_array($href)){
-			
+
 			if(isset($href["mask-icon"])){ $href = $href["mask-icon"]; }
 			elseif(isset($href["shortcut icon"])){ $href = $href["shortcut icon"]; }
 			else{
-				
+
 				$href = "/favicon.ico";
 			}
 		}
-		
+
 		$href = $this->proxy->getabsoluteurl($href, $url);
 		/*
 		header("Content-type: text/plain");
 		echo $href;
 		die();*/
-		
+
-		
+
 		/*
 			Download the favicon
 		*/
 		//$href = "https://git.lolcat.ca/assets/img/logo.svg";
-		
+
 		try{
 			$payload =
 				$this->proxy->get(
@@ -215,24 +221,24 @@ class favicon{
 					true,
 					$url
 				);
-				
+
 		}catch(Exception $error){
-			
+
 			header("X-Error: Could not fetch the favicon (" . $error->getMessage() . ")");
 			$this->favicon404();
 		}
-		
+
 		/*
 			Parse the file format
 		*/
 		$image = null;
 		$format = $this->proxy->getimageformat($payload, $image);
-		
+
 		/*
 			Convert the image
 		*/
 		try{
-			
+
 			/*
 				@todo: fix issues with avif+transparency
 				maybe using GD as fallback?
@@ -240,32 +246,32 @@ class favicon{
 			if($format !== false){
 				$image->setFormat($format);
 			}
-			
+
 			$image->setBackgroundColor(new ImagickPixel("transparent"));
 			$image->readImageBlob($payload["body"]);
 			$image->resizeImage(16, 16, imagick::FILTER_LANCZOS, 1);
 			$image->setFormat("png");
-			
+
 			$image = $image->getImageBlob();
-			
+
 			// save favicon
-			$handle = fopen("icons/" . $this->filename . ".png", "w");
+			$handle = fopen($this->iconpath(), "w");
 			fwrite($handle, $image, strlen($image));
 			fclose($handle);
-			
+
 			echo $image;
-			
+
 		}catch(ImagickException $error){
-			
+
 			header("X-Error: Could not convert the favicon: (" . $error->getMessage() . ")");
 			$this->favicon404();
 		}
-		
+
 		return;
 	}
-	
+
 	private function parsemanifest($href, $url){
-		
+
 		if(
 			// check if base64-encoded JSON manifest
 			preg_match(
@@ -274,17 +280,17 @@ class favicon{
 				$json
 			)
 		){
-			
+
 			$json = base64_decode($json[1]);
-			
+
 			if($json === false){
-				
+
 				// could not decode the manifest regex
 				return [];
 			}
-			
+
 		}else{
-			
+
 			try{
 				$json =
 					$this->proxy->get(
@@ -293,76 +299,81 @@ class favicon{
 						false,
 						$url
 					);
-					
+
 					$json = $json["body"];
-					
+
 			}catch(Exception $error){
-				
+
 				// could not fetch the manifest
 				return [];
 			}
 		}
-		
+
 		$json = json_decode($json, true);
-		
+
 		if($json === null){
-			
+
 			// manifest did not return valid json
 			return [];
 		}
-		
+
 		if(
 			isset($json["start_url"]) &&
 			$this->proxy->validateurl($json["start_url"])
 		){
-			
+
 			$url = $json["start_url"];
 		}
-		
+
 		if(!isset($json["icons"][0]["src"])){
-			
+
 			// manifest does not contain a path to the favicon
 			return [];
 		}
-		
+
 		// horay, return the favicon path
 		return $json["icons"][0]["src"];
 	}
-	
+
+	private function iconpath() {
+		// $this->filename can be trusted
+		return config::ICON_DIR . "/" . $this->filename . ".png";
+	}
+
 	private function favicon404(){
-		
+
 		// fallback to google favicons
 		// ... probably blocked by cuckflare
 		try{
-			
+
 			$image =
 				$this->proxy->get(
 					"https://t0.gstatic.com/faviconV2?client=SOCIAL&type=FAVICON&fallback_opts=TYPE,SIZE,URL&url=http://{$this->filename}&size=16",
 					$this->proxy::req_image
 				);
 		}catch(Exception $error){
-			
+
 			$this->defaulticon();
 		}
-		
+
 		// write favicon from google
-		$handle = fopen("icons/" . $this->filename . ".png", "w");
+		$handle = fopen($this->iconpath(), "w");
 		fwrite($handle, $image["body"], strlen($image["body"]));
 		fclose($handle);
-		
+
 		echo $image["body"];
 		die();
 	}
-	
+
 	private function defaulticon(){
-		
+
 		// give 404 and fuck off
 		http_response_code(404);
-		
+
 		$handle = fopen("lib/favicon404.png", "r");
 		echo fread($handle, filesize("lib/favicon404.png"));
 		fclose($handle);
-		
+
 		die();
 	}
 }

src/lib/anubis.php

@@ -24,7 +24,7 @@ class anubis{
 				"script"
 			);
 
-		if(count($script) === 0){
+		if($script === false){
 
 			throw new Exception("Failed to scrape anubis challenge data");
 		}

src/lib/backend.php

@@ -1,27 +1,27 @@
 <?php
 class backend{
-	
+
 	public function __construct($scraper){
-		
+
 		$this->scraper = $scraper;
 	}
-	
+
 	/*
 		Proxy stuff
 	*/
 	public function get_ip(){
-		
+
 		$pool = constant("config::PROXY_" . strtoupper($this->scraper));
 		if($pool === false){
-			
+
 			// we don't want a proxy, fuck off!
 			return 'raw_ip::::';
 		}
-		
+
 		// indent
 		$proxy_index_raw = apcu_inc("p." . $this->scraper);
-		
+
-		$proxylist = file_get_contents("data/proxies/" . $pool . ".txt");
+		$proxylist = file_get_contents("data/" . $pool . ".txt");
 		$proxylist = explode("\n", $proxylist);
 
 		// ignore empty or commented lines
@@ -29,15 +29,15 @@ class backend{
 			$entry = ltrim($entry);
 			return strlen($entry) > 0 && substr($entry, 0, 1) != "#";
 		});
-		
+
 		$proxylist = array_values($proxylist);
-		
+
 		return $proxylist[$proxy_index_raw % count($proxylist)];
 	}
-	
+
 	// this function is also called directly on nextpage
 	public function assign_proxy(&$curlproc, string $ip){
-		
+
 		// parse proxy line
 		[
 			$type,
@@ -46,34 +46,34 @@ class backend{
 			$username,
 			$password
 		] = explode(":", $ip, 5);
-		
+
 		switch($type){
-			
+
 			case "raw_ip":
 				return;
 				break;
-			
+
 			case "http":
 			case "https":
 				curl_setopt($curlproc, CURLOPT_PROXYTYPE, CURLPROXY_HTTP);
 				curl_setopt($curlproc, CURLOPT_PROXY, $type . "://" . $address . ":" . $port);
 				break;
-			
+
 			case "socks4":
 				curl_setopt($curlproc, CURLOPT_PROXYTYPE, CURLPROXY_SOCKS4);
 				curl_setopt($curlproc, CURLOPT_PROXY, $address . ":" . $port);
 				break;
-			
+
 			case "socks5":
 				curl_setopt($curlproc, CURLOPT_PROXYTYPE, CURLPROXY_SOCKS5);
 				curl_setopt($curlproc, CURLOPT_PROXY, $address . ":" . $port);
 				break;
-			
+
 			case "socks4a":
 				curl_setopt($curlproc, CURLOPT_PROXYTYPE, CURLPROXY_SOCKS4A);
 				curl_setopt($curlproc, CURLOPT_PROXY, $address . ":" . $port);
 				break;
-			
+
 			case "socks5_hostname":
 			case "socks5h":
 			case "socks5a":
@@ -81,25 +81,25 @@ class backend{
 				curl_setopt($curlproc, CURLOPT_PROXY, $address . ":" . $port);
 				break;
 		}
-		
+
 		if($username != ""){
-			
+
 			curl_setopt($curlproc, CURLOPT_PROXYUSERPWD, $username . ":" . $password);
 		}
 	}
-	
+
-	
+
-	
+
 	/*
 		Next page stuff
 	*/
 	public function store(string $payload, string $page, string $proxy){
-		
+
 		$key = sodium_crypto_secretbox_keygen();
 		$nonce = random_bytes(SODIUM_CRYPTO_SECRETBOX_NONCEBYTES);
-		
+
 		$requestid = apcu_inc("requestid");
-		
+
 		apcu_store(
 			$page[0] . "." . // first letter of page name
 			$this->scraper . // scraper name
@@ -117,31 +117,31 @@ class backend{
 			900 // cache information for 15 minutes
 		);
 
-		return 
+		return
 			$this->scraper . $requestid . "." .
 			rtrim(strtr(base64_encode($key), '+/', '-_'), '=');
 	}
-	
+
 	public function get(string $npt, string $page){
-		
+
 		$page = $page[0];
 		$explode = explode(".", $npt, 2);
-		
+
 		if(count($explode) !== 2){
-			
+
 			throw new Exception("Malformed nextPageToken!");
 		}
-		
+
 		$apcu = $page . "." . $explode[0];
 		$key = $explode[1];
-		
+
 		$payload = apcu_fetch($apcu);
-		
+
 		if($payload === false){
-			
+
 			throw new Exception("The next page token is invalid or has expired!");
 		}
-		
+
 		$key =
 			base64_decode(
 				str_pad(
@@ -151,7 +151,7 @@ class backend{
 					STR_PAD_RIGHT
 				)
 			);
-		
+
 		// decrypt and decompress data
 		$payload[2] =
 			gzinflate(
@@ -161,15 +161,15 @@ class backend{
 					$key
 				)
 			);
-		
+
 		if($payload[2] === false){
-			
+
 			throw new Exception("The next page token is invalid or has expired!");
 		}
-		
+
 		// remove the key after using successfully
 		apcu_delete($apcu);
-		
+
 		return [
 			$payload[2], // data
 			$payload[1] // proxy

src/lib/frontend.php

@@ -73,7 +73,6 @@ class frontend{
 	}
 
 	public function loadheader(array $get, array $filters, string $page){
-
 		echo
 			$this->load("header.html", [
 				"title" => trim(htmlspecialchars($get["s"]) . " ({$page})"),
@@ -83,60 +82,6 @@ class frontend{
 				"tabs" => $this->generatehtmltabs($page, $get["s"]),
 				"filters" => $this->generatehtmlfilters($filters, $get)
 			]);
-
-		$headers_raw = getallheaders();
-		$header_keys = [];
-		$user_agent = "";
-		$bad_header = false;
-
-		// block bots that present X-Forwarded-For, Via, etc
-		foreach($headers_raw as $headerkey => $headervalue){
-
-			$headerkey = strtolower($headerkey);
-			if($headerkey == "user-agent"){
-
-				$user_agent = $headervalue;
-				continue;
-			}
-
-			// check header key
-			if(in_array($headerkey, config::FILTERED_HEADER_KEYS)){
-
-				$bad_header = true;
-				break;
-			}
-		}
-
-		// SSL check
-		$bad_ssl = false;
-		if(
-			isset($_SERVER["https"]) &&
-			$_SERVER["https"] == "on" &&
-			isset($_SERVER["SSL_CIPHER"]) &&
-			in_array($_SERVER["SSL_CIPHER"], config::FILTERED_HEADER_KEYS)
-		){
-
-			$bad_ssl = true;
-		}
-
-		if(
-			$bad_header === true ||
-			$bad_ssl === true ||
-			$user_agent == "" ||
-			// user agent check
-			preg_match(
-				config::HEADER_REGEX,
-				$user_agent
-			)
-		){
-
-			// bot detected !!
-			$this->drawerror(
-				"Tshh, blocked!",
-				'Your browser, IP or IP range has been blocked from this 4get instance.'
-			);
-			die();
-		}
 	}
 
 	public function drawerror($title, $error, $timetaken = null){
@@ -178,7 +123,6 @@ class frontend{
 				'<li>Remove keywords that could cause errors</li>' .
 				'<li><a href="/instances?target=' . $target . "&" . $this->buildquery($get, false) . '">Try your search on another 4get instance</a></li>' .
 			'</ul><br>' .
-			'If the error persists, please <a href="/about">contact the administrator</a>.',
 			$timetaken
 		);
 	}

src/scraper/ddg.php

@@ -719,7 +719,7 @@ class ddg{
 						->getTextContent(
 							$json["suggestion"]
 						),
-					"correction" => $json["recourseText"]
+					"correction" => html_entity_decode($json["recourseText"])
 				];
 			}
 		}

src/scraper/marginalia.php

@@ -317,14 +317,19 @@ class marginalia{
 		//
 		// Bypass anubis check
 		//
+		/*
 		if(($anubis_key = apcu_fetch("marginalia_cookie")) === false){
 
 			try{
 				$html =
 					$this->get(
 						$proxy,
-						"https://old-search.marginalia.nu/"
+						"https://old-search.marginalia.nu/search",
+						[
+							"query" => $search
+						]
 					);
+
 			}catch(Exception $error){
 
 				throw new Exception("Failed to get anubis challenge");
@@ -361,7 +366,7 @@ class marginalia{
 			}
 
 			apcu_store("marginalia_cookie", $anubis_key);
-		}
+		}*/
 
 		if($get["npt"]){
 
@@ -377,7 +382,7 @@ class marginalia{
 						$proxy,
 						"https://old-search.marginalia.nu/search?" . $params,
 						[],
-						$anubis_key
+						//$anubis_key
 					);
 			}catch(Exception $error){
 
@@ -408,7 +413,7 @@ class marginalia{
 						$proxy,
 						"https://old-search.marginalia.nu/search",
 						$params,
-						$anubis_key
+						//$anubis_key
 					);
 			}catch(Exception $error){
 

src/static/style.css

@@ -4,20 +4,20 @@
 	--282828: #282828;
 	--3c3836: #3c3836;
 	--504945: #504945;
-	
+
 	/* font */
 	--928374: #928374;
 	--a89984: #a89984;
 	--bdae93: #bdae93;
 	--8ec07c: #8ec07c;
 	--ebdbb2: #ebdbb2;
-	
+
 	/* code highlighter */
 	--comment: #9e8e73;
 	--default: #d4be98;
 	--keyword: #d8a657;
 	--string: #7daea7;
-	
+
 	/* color codes for instance list */
 	--green: #b8bb26;
 	--yellow: #d8a657;
@@ -258,7 +258,7 @@ h3,h4,h5,h6{
 	font-family:Times;
 	width:100%;
 	height:100%;
-	background:var(--282828);
+	background: none;
 	display:block;
 	object-fit:contain;
 }
@@ -1300,20 +1300,20 @@ table tr a:last-child{
 	Responsive design
 */
 @media only screen and (max-width: 1550px){
-	
+
 	.web .right-right,
 	.web .right-left{
 		float:none;
 		width:initial;
 		padding:0 0 0 15px;
 	}
-	
+
 	.web .left,
 	.searchbox,
 	#images .infobox{
 		width:60%;
 	}
-	
+
 	.web .right-wrapper{
 		width:40%;
 	}
@@ -1323,43 +1323,43 @@ table tr a:last-child{
 	form{
 		padding-top:27px;
 	}
-	
+
 	.navigation{
 		left:0;
 		right:unset;
 		line-height:22px;
 	}
-	
+
 	.nextpage.img{
 		width:initial;
 	}
-	
+
 	.web .right-right,
 	.web .right-left{
 		border:none;
 		padding:0;
 	}
-	
+
 	.web .right-wrapper{
 		float:none;
 		padding:0;
 		width:initial;
 	}
-	
+
 	.web .left,
 	.searchbox{
 		width:100%;
 	}
-	
+
 	body:not(.instances) table td{
 		display:block;
 		width:100%;
 	}
-	
+
 	table a{
 		padding:0;
 	}
-	
+
 	.web.has-answer .left::before{
 		display:block;
 		content:"Results";
@@ -1368,16 +1368,16 @@ table tr a:last-child{
 		margin-bottom:17px;
 		color:var(--bdae93);
 	}
-	
+
 	.web .answer{
 		max-height:200px;
 	}
-	
+
 	.web .wiki-head tr td:first-child,
 	.web .info-table tr td:first-child{
 		text-decoration:underline;
 	}
-	
+
 	#images .infobox{
 		width:100%;
 	}

src/web.php

@@ -84,7 +84,7 @@ if($results["spelling"]["type"] != "no_correction"){
 			'&' .
 			$frontend->buildquery($get, true) .
 			'&spellcheck=no">' .
-			$results["spelling"]["correction"] .
+			htmlspecialchars($results["spelling"]["correction"]) .
 			'</a>?' .
 		'</div>';
 }

ups.json

@@ -1,7 +1,7 @@
 {
   "upstream": "https://git.lolcat.ca/lolcat/4get",
   "provider": "gitea",
-  "commit": "a2bc1e6190bab561b7244e2e9bbda994ab0d0d31",
+  "commit": "430c0a2f0f72f1254ab65d53f13640fe02418f05",
   "scripts": [
     "s/--- a\\//--- a\\/src\\//g",
     "s/+++ b\\//+++ b\\/src\\//g"