lolcat
/
4get
6
13
Fork 4
Code Issues 13 Pull Requests Packages Projects Releases Wiki Activity

soundcloud fix, for good this time

This commit is contained in:
lolcat 2024-06-29 12:55:28 -04:00
parent 77931f3ee9
commit e252bf4fce
1 changed files with 81 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

103
scraper/sc.php
View File

@ -6,6 +6,9 @@ class sc{
include "lib/backend.php"; include "lib/backend.php";
$this->backend = new backend("sc"); $this->backend = new backend("sc");
include "lib/fuckhtml.php";
$this->fuckhtml = new fuckhtml();
} }
public function getfilters($page){ public function getfilters($page){
@ -25,7 +28,7 @@ class sc{
]; ];
} }
private function get($proxy, $url, $get = []){ private function get($proxy, $url, $get = [], $web_req = false){
$curlproc = curl_init(); $curlproc = curl_init();
@ -37,19 +40,42 @@ class sc{
curl_setopt($curlproc, CURLOPT_URL, $url); curl_setopt($curlproc, CURLOPT_URL, $url);
curl_setopt($curlproc, CURLOPT_ENCODING, ""); // default encoding curl_setopt($curlproc, CURLOPT_ENCODING, ""); // default encoding
curl_setopt($curlproc, CURLOPT_HTTPHEADER,
["User-Agent: " . config::USER_AGENT, // use http2
"Accept: application/json, text/javascript, */*; q=0.01", curl_setopt($curlproc, CURLOPT_HTTP_VERSION, CURL_HTTP_VERSION_2_0);
"Accept-Language: en-US,en;q=0.5",
"Accept-Encoding: gzip", if($web_req === false){
"Referer: https://soundcloud.com/",
"Origin: https://soundcloud.com", curl_setopt($curlproc, CURLOPT_HTTPHEADER,
"DNT: 1", ["User-Agent: " . config::USER_AGENT,
"Connection: keep-alive", "Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8",
"Sec-Fetch-Dest: empty", "Accept-Language: en-US,en;q=0.5",
"Sec-Fetch-Mode: cors", "Accept-Encoding: gzip",
"Sec-Fetch-Site: same-site"] "Referer: https://soundcloud.com/",
); "Origin: https://soundcloud.com",
"DNT: 1",
"Connection: keep-alive",
"Sec-Fetch-Dest: empty",
"Sec-Fetch-Mode: cors",
"Sec-Fetch-Site: same-site"]
);
}else{
curl_setopt($curlproc, CURLOPT_HTTPHEADER,
["User-Agent: " . config::USER_AGENT,
"Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8",
"Accept-Language: en-US,en;q=0.5",
"Accept-Encoding: gzip",
"DNT: 1",
"Connection: keep-alive",
"Upgrade-Insecure-Requests: 1",
"Sec-Fetch-Dest: document",
"Sec-Fetch-Mode: navigate",
"Sec-Fetch-Site: cross-site",
"Priority: u=1",
"TE: trailers"]
);
}
curl_setopt($curlproc, CURLOPT_RETURNTRANSFER, true); curl_setopt($curlproc, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curlproc, CURLOPT_SSL_VERIFYHOST, 2); curl_setopt($curlproc, CURLOPT_SSL_VERIFYHOST, 2);
@ -396,13 +422,47 @@ class sc{
$token = apcu_fetch("sc_token"); $token = apcu_fetch("sc_token");
if($token === false){ if($token !== false){
return $token;
}
// search through all javascript components on the main page
try{
$html =
$this->get(
$proxy,
"https://soundcloud.com",
[]
);
}catch(Exception $error){
throw new Exception("Failed to fetch front page");
}
$this->fuckhtml->load($html);
$scripts =
$this->fuckhtml
->getElementsByTagName(
"script"
);
foreach($scripts as $script){
if(
!isset($script["attributes"]["src"]) ||
strpos($script["attributes"]["src"], "sndcdn.com") === false
){
continue;
}
try{ try{
$js = $js =
$this->get( $this->get(
$proxy, $proxy,
"https://a-v2.sndcdn.com/assets/0-a901c1e0.js", $script["attributes"]["src"],
[] []
); );
}catch(Exception $error){ }catch(Exception $error){
@ -416,16 +476,15 @@ class sc{
$token $token
); );
if(!isset($token[1])){ if(isset($token[1])){
throw new Exception("Failed to get search token"); apcu_store("sc_token", $token[1]);
return $token[1];
break;
} }
apcu_store("sc_token", $token[1]);
return $token[1];
} }
return $token; throw new Exception("Did not find a Soundcloud token in the Javascript blobs");
} }
private function limitstrlen($text){ private function limitstrlen($text){