get rid of the captcha stuff, put proxies in data
All checks were successful
docker / docker (push) Successful in 11s

Signed-off-by: ngn <ngn@ngn.tf>
This commit is contained in:
ngn
2025-07-11 16:28:11 +03:00
parent 54b3588c3a
commit 5aff5e9277
9 changed files with 51 additions and 55 deletions

4
.gitignore vendored
View File

@ -1,5 +1,5 @@
/compose.yml
/docker-compose.yml
/banners
/captcha
/banner
/favicon.ico
/config.php

View File

@ -6,8 +6,8 @@ services:
- 80:8080
volumes:
- ./config.php:/srv/4get/data/config.php:ro
- ./banners:/srv/4get/banner:ro
- ./captcha:/srv/4get/data/captcha:ro
- ./favicon.ico:/srv/4get/favicon.ico:ro
- ./banner:/srv/4get/banner:ro
- type: tmpfs
target: /tmp/icons
cap_drop:

7
src/.gitignore vendored
View File

@ -21,9 +21,8 @@ scraper/soundcloud.json
scraper/mp3-pm.html
scraper/curlie.html
icons/*
banner/*
!banner/*default*
data/*
data/captcha
data/config.php
!banner/*default*
!data/*.def.*

View File

@ -37,8 +37,8 @@ class config{
// 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
// 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

Binary file not shown.

View File

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

Binary file not shown.

Before

Width:  |  Height:  |  Size: 753 B

View File

@ -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