Browse Source

Use CURLOPT_HEADERFUNCTION to track requests

CURLOPT_HEADERFUNCTION is the recommended method for collecting the
response headers. Because every request in the redirect chain contains
headers (at the minimum a location line), we can track which URL’s
headers are being parsed and collect a history of URLs.
pull/2/head
Martijn van der Ven 5 years ago
parent
commit
a1e7c7ff03
1 changed files with 36 additions and 18 deletions
  1. +36
    -18
      src/p3k/HTTP/Curl.php

+ 36
- 18
src/p3k/HTTP/Curl.php View File

@ -6,6 +6,10 @@ class Curl implements Transport {
protected $_timeout = 4; protected $_timeout = 4;
protected $_max_redirects = 8; protected $_max_redirects = 8;
static protected $_http_version = null; static protected $_http_version = null;
private $_last_seen_url = null;
private $_last_seen_code = null;
private $_current_headers = [];
private $_current_redirects = [];
public function set_max_redirects($max) { public function set_max_redirects($max) {
$this->_max_redirects = $max; $this->_max_redirects = $max;
@ -21,16 +25,14 @@ class Curl implements Transport {
if($headers) if($headers)
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers); curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
$response = curl_exec($ch); $response = curl_exec($ch);
$header_size = curl_getinfo($ch, CURLINFO_HEADER_SIZE);
$header_str = trim(substr($response, 0, $header_size));
return [ return [
'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE), 'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE),
'header' => $header_str,
'body' => substr($response, $header_size),
'header' => implode("\r\n", array_map(function ($a) { return $a[0] . ': ' . $a[1]; }, $this->_current_headers)),
'body' => $response,
'redirects' => $this->_current_redirects,
'error' => self::error_string_from_code(curl_errno($ch)), 'error' => self::error_string_from_code(curl_errno($ch)),
'error_description' => curl_error($ch), 'error_description' => curl_error($ch),
'url' => curl_getinfo($ch, CURLINFO_EFFECTIVE_URL),
'debug' => $response
'url' => $this->_last_seen_url
]; ];
} }
@ -42,16 +44,14 @@ class Curl implements Transport {
if($headers) if($headers)
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers); curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
$response = curl_exec($ch); $response = curl_exec($ch);
$header_size = curl_getinfo($ch, CURLINFO_HEADER_SIZE);
$header_str = trim(substr($response, 0, $header_size));
return [ return [
'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE), 'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE),
'header' => $header_str,
'body' => substr($response, $header_size),
'header' => implode("\r\n", array_map(function ($a) { return $a[0] . ': ' . $a[1]; }, $this->_current_headers)),
'body' => $response,
'redirects' => $this->_current_redirects,
'error' => self::error_string_from_code(curl_errno($ch)), 'error' => self::error_string_from_code(curl_errno($ch)),
'error_description' => curl_error($ch), 'error_description' => curl_error($ch),
'url' => curl_getinfo($ch, CURLINFO_EFFECTIVE_URL),
'debug' => $response
'url' => $this->_last_seen_url
]; ];
} }
@ -64,23 +64,22 @@ class Curl implements Transport {
$response = curl_exec($ch); $response = curl_exec($ch);
return [ return [
'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE), 'code' => curl_getinfo($ch, CURLINFO_HTTP_CODE),
'header' => trim($response),
'header' => implode("\r\n", array_map(function ($a) { return $a[0] . ': ' . $a[1]; }, $this->_current_headers)),
'redirects' => $this->_current_redirects,
'error' => self::error_string_from_code(curl_errno($ch)), 'error' => self::error_string_from_code(curl_errno($ch)),
'error_description' => curl_error($ch), 'error_description' => curl_error($ch),
'url' => curl_getinfo($ch, CURLINFO_EFFECTIVE_URL),
'debug' => $response
'url' => $this->_last_seen_url
]; ];
} }
private function _set_curlopts($ch, $url) { private function _set_curlopts($ch, $url) {
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_HEADER, true);
if($this->_max_redirects > 0)
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, $this->_max_redirects > 0);
curl_setopt($ch, CURLOPT_MAXREDIRS, $this->_max_redirects); curl_setopt($ch, CURLOPT_MAXREDIRS, $this->_max_redirects);
curl_setopt($ch, CURLOPT_TIMEOUT_MS, round($this->_timeout * 1000)); curl_setopt($ch, CURLOPT_TIMEOUT_MS, round($this->_timeout * 1000));
curl_setopt($ch, CURLOPT_CONNECTTIMEOUT_MS, 2000); curl_setopt($ch, CURLOPT_CONNECTTIMEOUT_MS, 2000);
curl_setopt($ch, CURLOPT_HTTP_VERSION, $this->_http_version()); curl_setopt($ch, CURLOPT_HTTP_VERSION, $this->_http_version());
curl_setopt($ch, CURLOPT_HEADERFUNCTION, [$this, '_header_function']);
} }
private function _http_version() { private function _http_version() {
@ -96,6 +95,25 @@ class Curl implements Transport {
return static::$_http_version; return static::$_http_version;
} }
private function _header_function($curl, $header) {
$current_url = curl_getinfo($curl, CURLINFO_EFFECTIVE_URL);
if ($current_url !== $this->_last_seen_url) {
if ($this->_last_seen_url !== null) {
$this->_current_redirects[] = [$this->_last_seen_code, $this->_last_seen_url];
}
$this->_current_headers = [];
$this->_last_seen_url = $current_url;
$this->_last_seen_code = curl_getinfo($curl, CURLINFO_HTTP_CODE);
}
$length = strlen($header);
$header = explode(':', $header, 2);
if (count($header) !== 2) {
return $length;
}
$this->_current_headers[] = array_map('trim', [$header[0], $header[1]]);
return $length;
}
public static function error_string_from_code($code) { public static function error_string_from_code($code) {
switch($code) { switch($code) {
case 0: case 0:

Loading…
Cancel
Save