friendica-github/src/Module/Proxy.php

354 lines
10 KiB
PHP
Raw Normal View History

<?php
/**
2021-03-29 06:40:20 +00:00
* @copyright Copyright (C) 2010-2021, the Friendica project
*
* @license GNU AGPL version 3 or any later version
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as
* published by the Free Software Foundation, either version 3 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <https://www.gnu.org/licenses/>.
*
*/
namespace Friendica\Module;
use Friendica\BaseModule;
use Friendica\Core\Logger;
use Friendica\Core\System;
use Friendica\DI;
use Friendica\Model\Photo;
use Friendica\Object\Image;
use Friendica\Util\HTTPSignature;
use Friendica\Util\Proxy as ProxyUtils;
/**
2020-01-19 06:05:23 +00:00
* Module Proxy
2018-12-09 11:33:38 +00:00
*
* urls:
* /proxy/[sub1/[sub2/]]<base64url image url>[.ext][:size]
* /proxy?url=<image url>
*/
class Proxy extends BaseModule
{
/**
2020-01-19 06:05:23 +00:00
* Initializer method for this class.
*
* Sets application instance and checks if /proxy/ path is writable.
*
*/
public static function rawContent(array $parameters = [])
{
// Set application instance here
$a = DI::app();
/*
* Pictures are stored in one of the following ways:
*
* 1. If a folder "proxy" exists and is writeable, then use this for caching
* 2. If a cache path is defined, use this
* 3. If everything else failed, cache into the database
*
* Question: Do we really need these three methods?
*/
if (isset($_SERVER['HTTP_IF_MODIFIED_SINCE']) && isset($_SERVER['HTTP_IF_NONE_MATCH'])) {
header('HTTP/1.1 304 Not Modified');
header('Last-Modified: ' . gmdate('D, d M Y H:i:s', time()) . ' GMT');
header('Etag: ' . $_SERVER['HTTP_IF_NONE_MATCH']);
header('Expires: ' . gmdate('D, d M Y H:i:s', time() + (31536000)) . ' GMT');
header('Cache-Control: max-age=31536000');
if (function_exists('header_remove')) {
header_remove('Last-Modified');
header_remove('Expires');
header_remove('Cache-Control');
}
/// @TODO Stop here?
exit();
}
if (function_exists('header_remove')) {
header_remove('Pragma');
header_remove('pragma');
}
2018-12-09 11:33:38 +00:00
$direct_cache = self::setupDirectCache();
2018-12-09 11:33:38 +00:00
$request = self::getRequestInfo();
if (empty($request['url'])) {
throw new \Friendica\Network\HTTPException\BadRequestException();
}
if (!local_user()) {
Logger::info('Redirecting not logged in user to original address', ['url' => $request['url']]);
System::externalRedirect($request['url']);
}
2018-12-09 11:33:38 +00:00
// Webserver already tried direct cache...
// Try to use filecache;
$cachefile = self::responseFromCache($request);
// Try to use photo from db
self::responseFromDB($request);
//
// If script is here, the requested url has never cached before.
// Let's fetch it, scale it if required, then save it in cache.
//
// It shouldn't happen but it does - spaces in URL
$request['url'] = str_replace(' ', '+', $request['url']);
2020-12-15 22:56:46 +00:00
$fetchResult = HTTPSignature::fetchRaw($request['url'], local_user(), ['timeout' => 10]);
2018-12-09 11:33:38 +00:00
$img_str = $fetchResult->getBody();
// If there is an error then return a blank image
if ((substr($fetchResult->getReturnCode(), 0, 1) == '4') || empty($img_str)) {
Logger::info('Error fetching image', ['image' => $request['url'], 'return' => $fetchResult->getReturnCode(), 'empty' => empty($img_str)]);
self::responseError();
2018-12-09 11:33:38 +00:00
// stop.
}
$tempfile = tempnam(get_temppath(), 'cache');
file_put_contents($tempfile, $img_str);
$mime = mime_content_type($tempfile);
unlink($tempfile);
2018-12-09 11:33:38 +00:00
$image = new Image($img_str, $mime);
if (!$image->isValid()) {
Logger::info('The image is invalid', ['image' => $request['url'], 'mime' => $mime]);
self::responseError();
2018-12-09 11:33:38 +00:00
// stop.
}
2019-01-03 17:34:18 +00:00
$basepath = $a->getBasePath();
$filepermission = DI::config()->get('system', 'proxy_file_chmod');
2018-12-09 11:33:38 +00:00
// Store original image
if ($direct_cache) {
// direct cache , store under ./proxy/
$filename = $basepath . '/proxy/' . ProxyUtils::proxifyUrl($request['url'], true);
file_put_contents($filename, $image->asString());
if (!empty($filepermission)) {
chmod($filename, $filepermission);
}
2018-12-09 11:33:38 +00:00
} elseif($cachefile !== '') {
// cache file
file_put_contents($cachefile, $image->asString());
} else {
// database
Photo::store($image, 0, 0, $request['urlhash'], $request['url'], '', 100);
}
// reduce quality - if it isn't a GIF
if ($image->getType() != 'image/gif') {
$image->scaleDown($request['size']);
}
// Store scaled image
if ($direct_cache && $request['sizetype'] != '') {
$filename = $basepath . '/proxy/' . ProxyUtils::proxifyUrl($request['url'], true) . $request['sizetype'];
file_put_contents($filename, $image->asString());
if (!empty($filepermission)) {
chmod($filename, $filepermission);
}
2018-12-09 11:33:38 +00:00
}
self::responseImageHttpCache($image);
// stop.
}
/**
2020-01-19 06:05:23 +00:00
* Build info about requested image to be proxied
2018-12-09 11:33:38 +00:00
*
* @return array
* [
* 'url' => requested url,
* 'urlhash' => sha1 has of the url prefixed with 'pic:',
* 'size' => requested image size (int)
* 'sizetype' => requested image size (string): ':micro', ':thumb', ':small', ':medium', ':large'
* ]
2019-01-06 21:06:53 +00:00
* @throws \Exception
2018-12-09 11:33:38 +00:00
*/
private static function getRequestInfo()
{
$a = DI::app();
$size = ProxyUtils::PIXEL_LARGE;
2018-12-09 11:33:38 +00:00
$sizetype = '';
// Look for filename in the arguments
2019-05-01 19:29:04 +00:00
// @TODO: Replace with parameter from router
2018-12-09 11:33:38 +00:00
if (($a->argc > 1) && !isset($_REQUEST['url'])) {
if (isset($a->argv[3])) {
$url = $a->argv[3];
} elseif (isset($a->argv[2])) {
$url = $a->argv[2];
} else {
$url = $a->argv[1];
}
2018-12-09 11:33:38 +00:00
/// @TODO: Why? And what about $url in this case?
2019-05-01 19:29:04 +00:00
/// @TODO: Replace with parameter from router
if (isset($a->argv[3]) && ($a->argv[3] == 'thumb')) {
$size = 200;
}
// thumb, small, medium and large.
if (substr($url, -6) == ':micro') {
$size = ProxyUtils::PIXEL_MICRO;
$sizetype = ':micro';
$url = substr($url, 0, -6);
} elseif (substr($url, -6) == ':thumb') {
$size = ProxyUtils::PIXEL_THUMB;
$sizetype = ':thumb';
$url = substr($url, 0, -6);
} elseif (substr($url, -6) == ':small') {
$size = ProxyUtils::PIXEL_SMALL;
$url = substr($url, 0, -6);
$sizetype = ':small';
} elseif (substr($url, -7) == ':medium') {
$size = ProxyUtils::PIXEL_MEDIUM;
$url = substr($url, 0, -7);
$sizetype = ':medium';
} elseif (substr($url, -6) == ':large') {
$size = ProxyUtils::PIXEL_LARGE;
$url = substr($url, 0, -6);
$sizetype = ':large';
}
$pos = strrpos($url, '=.');
if ($pos) {
$url = substr($url, 0, $pos + 1);
}
$url = str_replace(['.jpg', '.jpeg', '.gif', '.png'], ['','','',''], $url);
$url = base64_decode(strtr($url, '-_', '+/'), true);
} else {
$url = $_REQUEST['url'] ?? '';
}
2018-12-09 11:33:38 +00:00
return [
'url' => $url,
'urlhash' => 'pic:' . sha1($url),
'size' => $size,
'sizetype' => $sizetype,
];
}
2019-01-06 21:06:53 +00:00
2018-12-09 11:33:38 +00:00
/**
2020-01-19 06:05:23 +00:00
* setup ./proxy folder for direct cache
2018-12-09 11:33:38 +00:00
*
* @return bool False if direct cache can't be used.
2019-01-06 21:06:53 +00:00
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
2018-12-09 11:33:38 +00:00
*/
private static function setupDirectCache()
{
$a = DI::app();
2018-12-09 11:33:38 +00:00
$basepath = $a->getBasePath();
2018-12-09 11:33:38 +00:00
// If the cache path isn't there, try to create it
if (!is_dir($basepath . '/proxy') && is_writable($basepath)) {
mkdir($basepath . '/proxy');
}
2018-12-09 11:33:38 +00:00
// Checking if caching into a folder in the webroot is activated and working
$direct_cache = (is_dir($basepath . '/proxy') && is_writable($basepath . '/proxy'));
// we don't use direct cache if image url is passed in args and not in querystring
2018-12-09 11:33:38 +00:00
$direct_cache = $direct_cache && ($a->argc > 1) && !isset($_REQUEST['url']);
2018-12-09 11:33:38 +00:00
return $direct_cache;
}
2019-01-06 21:06:53 +00:00
2018-12-09 11:33:38 +00:00
/**
2020-01-19 06:05:23 +00:00
* Try to reply with image in cachefile
2018-12-09 11:33:38 +00:00
*
2019-01-06 21:06:53 +00:00
* @param array $request Array from getRequestInfo
2018-12-09 11:33:38 +00:00
*
* @return string Cache file name, empty string if cache is not enabled.
2019-01-06 21:06:53 +00:00
*
2018-12-09 11:33:38 +00:00
* If cachefile exists, script ends here and this function will never returns
2019-01-06 21:06:53 +00:00
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
* @throws \ImagickException
2018-12-09 11:33:38 +00:00
*/
private static function responseFromCache(&$request)
{
$cachefile = get_cachefile(hash('md5', $request['url']));
if ($cachefile != '' && file_exists($cachefile)) {
$img = new Image(file_get_contents($cachefile), mime_content_type($cachefile));
self::responseImageHttpCache($img);
// stop.
}
2018-12-09 11:33:38 +00:00
return $cachefile;
}
2019-01-06 21:06:53 +00:00
2018-12-09 11:33:38 +00:00
/**
2020-01-19 06:05:23 +00:00
* Try to reply with image in database
2018-12-09 11:33:38 +00:00
*
2019-01-06 21:06:53 +00:00
* @param array $request Array from getRequestInfo
2018-12-09 11:33:38 +00:00
*
* If the image exists in database, then script ends here and this function will never returns
2019-01-06 21:06:53 +00:00
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
* @throws \ImagickException
2018-12-09 11:33:38 +00:00
*/
private static function responseFromDB(&$request)
{
2018-12-09 11:33:38 +00:00
$photo = Photo::getPhoto($request['urlhash']);
if ($photo !== false) {
$img = Photo::getImageForPhoto($photo);
self::responseImageHttpCache($img);
// stop.
}
2018-12-09 11:33:38 +00:00
}
2018-12-09 11:33:38 +00:00
/**
* In case of an error just stop. We don't return content to avoid caching problems
2018-12-09 11:33:38 +00:00
*
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
2018-12-09 11:33:38 +00:00
*/
private static function responseError()
{
throw new \Friendica\Network\HTTPException\InternalServerErrorException();
2018-12-09 11:33:38 +00:00
}
2019-01-06 21:06:53 +00:00
2018-12-09 11:33:38 +00:00
/**
2020-01-19 06:05:23 +00:00
* Output the image with cache headers
2018-12-09 11:33:38 +00:00
*
2019-01-06 21:06:53 +00:00
* @param Image $img
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
2018-12-09 11:33:38 +00:00
*/
private static function responseImageHttpCache(Image $img)
{
if (is_null($img) || !$img->isValid()) {
Logger::info('The cached image is invalid');
2018-12-09 11:33:38 +00:00
self::responseError();
// stop.
}
2018-12-09 11:33:38 +00:00
header('Content-type: ' . $img->getType());
header('Last-Modified: ' . gmdate('D, d M Y H:i:s', time()) . ' GMT');
header('Etag: "' . md5($img->asString()) . '"');
header('Expires: ' . gmdate('D, d M Y H:i:s', time() + (31536000)) . ' GMT');
header('Cache-Control: max-age=31536000');
echo $img->asString();
exit();
}
}