gnu-social/classes/File.php

689 lines
29 KiB
PHP
Raw Normal View History

<?php
/*
2009-08-26 07:14:12 +09:00
* StatusNet - the distributed open-source microblogging tool
* Copyright (C) 2008, 2009, StatusNet, Inc.
*
* 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 <http://www.gnu.org/licenses/>.
*/
if (!defined('GNUSOCIAL')) { exit(1); }
/**
* Table Definition for file
*/
2011-08-23 06:52:02 +09:00
class File extends Managed_DataObject
{
public $__table = 'file'; // table name
public $id; // int(4) primary_key not_null
public $urlhash; // varchar(64) unique_key
public $url; // text
public $filehash; // varchar(64) indexed
public $mimetype; // varchar(50)
public $size; // int(4)
public $title; // varchar(191) not 255 because utf8mb4 takes more space
public $date; // int(4)
public $protected; // int(4)
public $filename; // varchar(191) not 255 because utf8mb4 takes more space
public $width; // int(4)
public $height; // int(4)
public $modified; // timestamp() not_null default_CURRENT_TIMESTAMP
const URLHASH_ALG = 'sha256';
const FILEHASH_ALG = 'sha256';
2011-08-23 06:52:02 +09:00
public static function schemaDef()
{
return array(
'fields' => array(
'id' => array('type' => 'serial', 'not null' => true),
'urlhash' => array('type' => 'varchar', 'length' => 64, 'not null' => true, 'description' => 'sha256 of destination URL (url field)'),
'url' => array('type' => 'text', 'description' => 'destination URL after following possible redirections'),
'filehash' => array('type' => 'varchar', 'length' => 64, 'not null' => false, 'description' => 'sha256 of the file contents, only for locally stored files of course'),
2011-08-23 06:52:02 +09:00
'mimetype' => array('type' => 'varchar', 'length' => 50, 'description' => 'mime type of resource'),
'size' => array('type' => 'int', 'description' => 'size of resource when available'),
'title' => array('type' => 'varchar', 'length' => 191, 'description' => 'title of resource when available'),
2011-08-23 06:52:02 +09:00
'date' => array('type' => 'int', 'description' => 'date of resource according to http query'),
'protected' => array('type' => 'int', 'description' => 'true when URL is private (needs login)'),
'filename' => array('type' => 'varchar', 'length' => 191, 'description' => 'if a local file, name of the file'),
'width' => array('type' => 'int', 'description' => 'width in pixels, if it can be described as such and data is available'),
'height' => array('type' => 'int', 'description' => 'height in pixels, if it can be described as such and data is available'),
2011-08-23 06:52:02 +09:00
'modified' => array('type' => 'timestamp', 'not null' => true, 'description' => 'date this record was modified'),
),
'primary key' => array('id'),
'unique keys' => array(
'file_urlhash_key' => array('urlhash'),
2011-08-23 06:52:02 +09:00
),
'indexes' => array(
'file_filehash_idx' => array('filehash'),
),
2011-08-23 06:52:02 +09:00
);
}
function isProtected($url) {
return 'http://www.facebook.com/login.php' === $url;
}
/**
* Save a new file record.
*
* @param array $redir_data lookup data eg from File_redirection::where()
* @param string $given_url
* @return File
*/
public static function saveNew(array $redir_data, $given_url) {
// I don't know why we have to keep doing this but I'm adding this last check to avoid
// uniqueness bugs.
$file = File::getKV('urlhash', self::hashurl($given_url));
if (!$file instanceof File) {
$file = new File;
$file->urlhash = self::hashurl($given_url);
$file->url = $given_url;
if (!empty($redir_data['protected'])) $file->protected = $redir_data['protected'];
if (!empty($redir_data['title'])) $file->title = $redir_data['title'];
if (!empty($redir_data['type'])) $file->mimetype = $redir_data['type'];
if (!empty($redir_data['size'])) $file->size = intval($redir_data['size']);
if (isset($redir_data['time']) && $redir_data['time'] > 0) $file->date = intval($redir_data['time']);
$file_id = $file->insert();
if ($file_id === false) {
throw new ServerException('File/URL metadata could not be saved to the database.');
}
}
Event::handle('EndFileSaveNew', array($file, $redir_data, $given_url));
assert ($file instanceof File);
return $file;
}
/**
2010-11-16 04:00:42 +09:00
* Go look at a URL and possibly save data about it if it's new:
* - follow redirect chains and store them in file_redirection
* - if a thumbnail is available, save it in file_thumbnail
* - save file record with basic info
* - optionally save a file_to_post record
* - return the File object with the full reference
*
* @fixme refactor this mess, it's gotten pretty scary.
2010-11-16 04:00:42 +09:00
* @param string $given_url the URL we're looking at
* @param Notice $notice (optional)
2010-11-16 04:00:42 +09:00
* @param bool $followRedirects defaults to true
*
* @return mixed File on success, -1 on some errors
*
* @throws ServerException on failure
*/
public static function processNew($given_url, Notice $notice=null, $followRedirects=true) {
if (empty($given_url)) {
throw new ServerException('No given URL to process');
}
$given_url = File_redirection::_canonUrl($given_url);
if (empty($given_url)) {
throw new ServerException('No canonical URL from given URL to process');
}
$file = null;
try {
$file = File::getByUrl($given_url);
} catch (NoResultException $e) {
// First check if we have a lookup trace for this URL already
try {
$file_redir = File_redirection::getByUrl($given_url);
$file = File::getKV('id', $file_redir->file_id);
if (!$file instanceof File) {
// File did not exist, let's clean up the File_redirection entry
$file_redir->delete();
}
} catch (NoResultException $e) {
// We just wanted to doublecheck whether a File_thumbnail we might've had
// actually referenced an existing File object.
}
}
// If we still don't have a File object, let's create one now!
if (!$file instanceof File) {
// @fixme for new URLs this also looks up non-redirect data
// such as target content type, size, etc, which we need
// for File::saveNew(); so we call it even if not following
// new redirects.
$redir_data = File_redirection::where($given_url);
if (is_array($redir_data)) {
$redir_url = $redir_data['url'];
} elseif (is_string($redir_data)) {
$redir_url = $redir_data;
$redir_data = array();
} else {
// TRANS: Server exception thrown when a URL cannot be processed.
throw new ServerException(sprintf(_("Cannot process URL '%s'"), $given_url));
}
if ($redir_url === $given_url || !$followRedirects) {
// Save the File object based on our lookup trace
$file = File::saveNew($redir_data, $given_url);
} else {
// This seems kind of messed up... for now skipping this part
// if we're already under a redirect, so we don't go into
// horrible infinite loops if we've been given an unstable
// redirect (where the final destination of the first request
// doesn't match what we get when we ask for it again).
//
// Seen in the wild with clojure.org, which redirects through
// wikispaces for auth and appends session data in the URL params.
$file = self::processNew($redir_url, $notice, /*followRedirects*/false);
File_redirection::saveNew($redir_data, $file->id, $given_url);
}
if (!$file instanceof File) {
// This should only happen if File::saveNew somehow did not return a File object,
// though we have an assert for that in case the event there might've gone wrong.
// If anything else goes wrong, there should've been an exception thrown.
throw new ServerException('URL processing failed without new File object');
2009-07-31 05:55:09 +09:00
}
}
if ($notice instanceof Notice) {
File_to_post::processNew($file, $notice);
}
return $file;
}
public static function respectsQuota(Profile $scoped, $fileSize) {
if ($fileSize > common_config('attachments', 'file_quota')) {
// TRANS: Message used to be inserted as %2$s in the text "No file may
// TRANS: be larger than %1$d byte and the file you sent was %2$s.".
// TRANS: %1$d is the number of bytes of an uploaded file.
$fileSizeText = sprintf(_m('%1$d byte','%1$d bytes',$fileSize),$fileSize);
$fileQuota = common_config('attachments', 'file_quota');
// TRANS: Message given if an upload is larger than the configured maximum.
// TRANS: %1$d (used for plural) is the byte limit for uploads,
// TRANS: %2$s is the proper form of "n bytes". This is the only ways to have
// TRANS: gettext support multiple plurals in the same message, unfortunately...
throw new ClientException(
sprintf(_m('No file may be larger than %1$d byte and the file you sent was %2$s. Try to upload a smaller version.',
'No file may be larger than %1$d bytes and the file you sent was %2$s. Try to upload a smaller version.',
$fileQuota),
$fileQuota, $fileSizeText));
}
$file = new File;
$query = "select sum(size) as total from file join file_to_post on file_to_post.file_id = file.id join notice on file_to_post.post_id = notice.id where profile_id = {$scoped->id} and file.url like '%/notice/%/file'";
$file->query($query);
$file->fetch();
$total = $file->total + $fileSize;
if ($total > common_config('attachments', 'user_quota')) {
// TRANS: Message given if an upload would exceed user quota.
2010-11-05 02:33:39 +09:00
// TRANS: %d (number) is the user quota in bytes and is used for plural.
throw new ClientException(
sprintf(_m('A file this large would exceed your user quota of %d byte.',
2010-11-05 02:33:39 +09:00
'A file this large would exceed your user quota of %d bytes.',
common_config('attachments', 'user_quota')),
common_config('attachments', 'user_quota')));
}
$query .= ' AND EXTRACT(month FROM file.modified) = EXTRACT(month FROM now()) and EXTRACT(year FROM file.modified) = EXTRACT(year FROM now())';
$file->query($query);
$file->fetch();
$total = $file->total + $fileSize;
if ($total > common_config('attachments', 'monthly_quota')) {
// TRANS: Message given id an upload would exceed a user's monthly quota.
2010-11-05 02:33:39 +09:00
// TRANS: $d (number) is the monthly user quota in bytes and is used for plural.
throw new ClientException(
sprintf(_m('A file this large would exceed your monthly quota of %d byte.',
2010-11-05 02:33:39 +09:00
'A file this large would exceed your monthly quota of %d bytes.',
common_config('attachments', 'monthly_quota')),
common_config('attachments', 'monthly_quota')));
}
return true;
}
public function getFilename()
{
if (!self::validFilename($this->filename)) {
// TRANS: Client exception thrown if a file upload does not have a valid name.
throw new ClientException(_("Invalid filename."));
}
return $this->filename;
}
// where should the file go?
static function filename(Profile $profile, $origname, $mimetype)
{
$ext = self::guessMimeExtension($mimetype);
// Normalize and make the original filename more URL friendly.
2014-06-23 00:03:27 +09:00
$origname = basename($origname, ".$ext");
if (class_exists('Normalizer')) {
// http://php.net/manual/en/class.normalizer.php
// http://www.unicode.org/reports/tr15/
$origname = Normalizer::normalize($origname, Normalizer::FORM_KC);
}
$origname = preg_replace('/[^A-Za-z0-9\.\_]/', '_', $origname);
2014-06-23 00:03:27 +09:00
$nickname = $profile->getNickname();
$datestamp = strftime('%Y%m%d', time());
do {
// generate new random strings until we don't run into a filename collision.
$random = strtolower(common_confirmation_code(16));
$filename = "$nickname-$datestamp-$origname-$random.$ext";
} while (file_exists(self::path($filename)));
return $filename;
}
static function guessMimeExtension($mimetype)
{
try {
$ext = common_supported_mime_to_ext($mimetype);
} catch (Exception $e) {
// We don't support this mimetype, but let's guess the extension
$ext = substr(strrchr($mimetype, '/'), 1);
}
return strtolower($ext);
}
/**
* Validation for as-saved base filenames
*/
static function validFilename($filename)
{
return preg_match('/^[A-Za-z0-9._-]+$/', $filename);
}
/**
* @throws ClientException on invalid filename
*/
static function path($filename)
{
if (!self::validFilename($filename)) {
// TRANS: Client exception thrown if a file upload does not have a valid name.
throw new ClientException(_("Invalid filename."));
}
$dir = common_config('attachments', 'dir');
if ($dir[strlen($dir)-1] != '/') {
$dir .= '/';
}
return $dir . $filename;
}
static function url($filename)
{
if (!self::validFilename($filename)) {
// TRANS: Client exception thrown if a file upload does not have a valid name.
throw new ClientException(_("Invalid filename."));
}
if (common_config('site','private')) {
return common_local_url('getfile',
array('filename' => $filename));
}
if (GNUsocial::useHTTPS()) {
$sslserver = common_config('attachments', 'sslserver');
if (empty($sslserver)) {
// XXX: this assumes that background dir == site dir + /file/
// not true if there's another server
if (is_string(common_config('site', 'sslserver')) &&
mb_strlen(common_config('site', 'sslserver')) > 0) {
$server = common_config('site', 'sslserver');
} else if (common_config('site', 'server')) {
$server = common_config('site', 'server');
}
$path = common_config('site', 'path') . '/file/';
} else {
$server = $sslserver;
$path = common_config('attachments', 'sslpath');
if (empty($path)) {
$path = common_config('attachments', 'path');
}
}
$protocol = 'https';
} else {
$path = common_config('attachments', 'path');
$server = common_config('attachments', 'server');
if (empty($server)) {
$server = common_config('site', 'server');
}
2010-02-12 07:06:57 +09:00
$ssl = common_config('attachments', 'ssl');
2010-02-12 07:06:57 +09:00
$protocol = ($ssl) ? 'https' : 'http';
}
2010-02-12 07:06:57 +09:00
if ($path[strlen($path)-1] != '/') {
$path .= '/';
}
if ($path[0] != '/') {
$path = '/'.$path;
}
return $protocol.'://'.$server.$path.$filename;
}
function getEnclosure(){
$enclosure = (object) array();
foreach (array('title', 'url', 'date', 'modified', 'size', 'mimetype') as $key) {
$enclosure->$key = $this->$key;
}
$needMoreMetadataMimetypes = array(null, 'application/xhtml+xml');
if (!isset($this->filename) && in_array(common_bare_mime($enclosure->mimetype), $needMoreMetadataMimetypes)) {
// This fetches enclosure metadata for non-local links with unset/HTML mimetypes,
// which may be enriched through oEmbed or similar (implemented as plugins)
Event::handle('FileEnclosureMetadata', array($this, &$enclosure));
}
if (empty($enclosure->mimetype) || in_array(common_bare_mime($enclosure->mimetype), $needMoreMetadataMimetypes)) {
// This means we either don't know what it is, so it can't
// be shown as an enclosure, or it is an HTML link which
// does not link to a resource with further metadata.
throw new ServerException('Unknown enclosure mimetype, not enough metadata');
}
return $enclosure;
}
/**
* Get the attachment's thumbnail record, if any.
* Make sure you supply proper 'int' typed variables (or null).
*
* @param $width int Max width of thumbnail in pixels. (if null, use common_config values)
* @param $height int Max height of thumbnail in pixels. (if null, square-crop to $width)
* @param $crop bool Crop to the max-values' aspect ratio
*
* @return File_thumbnail
*
* @throws UseFileAsThumbnailException if the file is considered an image itself and should be itself as thumbnail
* @throws UnsupportedMediaException if, despite trying, we can't understand how to make a thumbnail for this format
* @throws ServerException on various other errors
*/
public function getThumbnail($width=null, $height=null, $crop=false, $force_still=true)
{
// Get some more information about this file through our ImageFile class
$image = ImageFile::fromFileObject($this);
if ($image->animated && !common_config('thumbnail', 'animated')) {
// null means "always use file as thumbnail"
// false means you get choice between frozen frame or original when calling getThumbnail
if (is_null(common_config('thumbnail', 'animated')) || !$force_still) {
throw new UseFileAsThumbnailException($this->id);
}
}
return $image->getFileThumbnail($width, $height, $crop);
}
public function getPath()
{
$filepath = self::path($this->filename);
if (!file_exists($filepath)) {
throw new FileNotFoundException($filepath);
}
return $filepath;
}
public function getUrl()
{
if (!empty($this->filename)) {
// A locally stored file, so let's generate a URL for our instance.
$url = self::url($this->filename);
if (self::hashurl($url) !== $this->urlhash) {
// For indexing purposes, in case we do a lookup on the 'url' field.
// also we're fixing possible changes from http to https, or paths
$this->updateUrl($url);
}
return $url;
}
// No local filename available, return the URL we have stored
return $this->url;
}
static public function getByUrl($url)
{
$file = new File();
$file->urlhash = self::hashurl($url);
if (!$file->find(true)) {
throw new NoResultException($file);
}
return $file;
}
/**
* @param string $hashstr String of (preferrably lower case) hexadecimal characters, same as result of 'hash_file(...)'
*/
static public function getByHash($hashstr, $alg=File::FILEHASH_ALG)
{
$file = new File();
$file->filehash = strtolower($hashstr);
if (!$file->find(true)) {
throw new NoResultException($file);
}
return $file;
}
public function updateUrl($url)
{
$file = File::getKV('urlhash', self::hashurl($url));
if ($file instanceof File) {
throw new ServerException('URL already exists in DB');
}
$sql = 'UPDATE %1$s SET urlhash=%2$s, url=%3$s WHERE urlhash=%4$s;';
$result = $this->query(sprintf($sql, $this->tableName(),
$this->_quote((string)self::hashurl($url)),
$this->_quote((string)$url),
$this->_quote((string)$this->urlhash)));
if ($result === false) {
common_log_db_error($this, 'UPDATE', __FILE__);
throw new ServerException("Could not UPDATE {$this->tableName()}.url");
}
return $result;
}
2010-12-29 05:57:31 +09:00
/**
* Blow the cache of notices that link to this URL
*
* @param boolean $last Whether to blow the "last" cache too
*
* @return void
*/
2010-12-29 04:58:55 +09:00
function blowCache($last=false)
{
self::blow('file:notice-ids:%s', $this->id);
2010-12-29 04:58:55 +09:00
if ($last) {
self::blow('file:notice-ids:%s;last', $this->id);
2010-12-29 04:58:55 +09:00
}
self::blow('file:notice-count:%d', $this->id);
2010-12-29 04:58:55 +09:00
}
/**
* Stream of notices linking to this URL
*
* @param integer $offset Offset to show; default is 0
* @param integer $limit Limit of notices to show
* @param integer $since_id Since this notice
* @param integer $max_id Before this notice
*
* @return array ids of notices that link to this file
*/
function stream($offset=0, $limit=NOTICES_PER_PAGE, $since_id=0, $max_id=0)
{
$stream = new FileNoticeStream($this);
return $stream->getNotices($offset, $limit, $since_id, $max_id);
2010-12-29 04:58:55 +09:00
}
function noticeCount()
{
$cacheKey = sprintf('file:notice-count:%d', $this->id);
$count = self::cacheGet($cacheKey);
if ($count === false) {
$f2p = new File_to_post();
$f2p->file_id = $this->id;
$count = $f2p->count();
self::cacheSet($cacheKey, $count);
}
return $count;
}
public function isLocal()
{
return !empty($this->filename);
}
public function delete($useWhere=false)
{
// Delete the file, if it exists locally
if (!empty($this->filename) && file_exists(self::path($this->filename))) {
$deleted = @unlink(self::path($this->filename));
if (!$deleted) {
common_log(LOG_ERR, sprintf('Could not unlink existing file: "%s"', self::path($this->filename)));
}
}
// Clear out related things in the database and filesystem, such as thumbnails
if (Event::handle('FileDeleteRelated', array($this))) {
$thumbs = new File_thumbnail();
$thumbs->file_id = $this->id;
if ($thumbs->find()) {
while ($thumbs->fetch()) {
$thumbs->delete();
}
}
$f2p = new File_to_post();
$f2p->file_id = $this->id;
if ($f2p->find()) {
while ($f2p->fetch()) {
$f2p->delete();
}
}
}
// And finally remove the entry from the database
return parent::delete($useWhere);
}
public function getTitle()
{
$title = $this->title ?: $this->filename;
return $title ?: null;
}
static public function hashurl($url)
{
if (empty($url)) {
throw new Exception('No URL provided to hash algorithm.');
}
return hash(self::URLHASH_ALG, $url);
}
static public function beforeSchemaUpdate()
{
$table = strtolower(get_called_class());
$schema = Schema::get();
$schemadef = $schema->getTableDef($table);
// 2015-02-19 We have to upgrade our table definitions to have the urlhash field populated
2015-02-20 03:36:59 +09:00
if (isset($schemadef['fields']['urlhash']) && isset($schemadef['unique keys']['file_urlhash_key'])) {
// We already have the urlhash field, so no need to migrate it.
return;
}
echo "\nFound old $table table, upgrading it to contain 'urlhash' field...";
$file = new File();
$file->query(sprintf('SELECT id, LEFT(url, 191) AS shortenedurl, COUNT(*) AS c FROM %1$s WHERE LENGTH(url)>191 GROUP BY shortenedurl HAVING c > 1', $schema->quoteIdentifier($table)));
print "\nFound {$file->N} URLs with too long entries in file table\n";
while ($file->fetch()) {
// We've got a URL that is too long for our future file table
// so we'll cut it. We could save the original URL, but there is
// no guarantee it is complete anyway since the previous max was 255 chars.
$dupfile = new File();
// First we find file entries that would be duplicates of this when shortened
// ... and we'll just throw the dupes out the window for now! It's already so borken.
$dupfile->query(sprintf('SELECT * FROM file WHERE LEFT(url, 191) = "%1$s"', $file->shortenedurl));
// Leave one of the URLs in the database by using ->find(true) (fetches first entry)
if ($dupfile->find(true)) {
print "\nShortening url entry for $table id: {$file->id} [";
$orig = clone($dupfile);
$dupfile->url = $file->shortenedurl; // make sure it's only 191 chars from now on
$dupfile->update($orig);
print "\nDeleting duplicate entries of too long URL on $table id: {$file->id} [";
// only start deleting with this fetch.
while($dupfile->fetch()) {
print ".";
$dupfile->delete();
}
print "]\n";
} else {
print "\nWarning! URL suddenly disappeared from database: {$file->url}\n";
}
}
echo "...and now all the non-duplicates which are longer than 191 characters...\n";
$file->query('UPDATE file SET url=LEFT(url, 191) WHERE LENGTH(url)>191');
echo "\n...now running hacky pre-schemaupdate change for $table:";
// We have to create a urlhash that is _not_ the primary key,
// transfer data and THEN run checkSchema
$schemadef['fields']['urlhash'] = array (
'type' => 'varchar',
'length' => 64,
'not null' => false, // this is because when adding column, all entries will _be_ NULL!
'description' => 'sha256 of destination URL (url field)',
);
$schemadef['fields']['url'] = array (
'type' => 'text',
'description' => 'destination URL after following possible redirections',
);
unset($schemadef['unique keys']);
$schema->ensureTable($table, $schemadef);
echo "DONE.\n";
$classname = ucfirst($table);
$tablefix = new $classname;
// urlhash is hash('sha256', $url) in the File table
echo "Updating urlhash fields in $table table...";
// Maybe very MySQL specific :(
$tablefix->query(sprintf('UPDATE %1$s SET %2$s=%3$s;',
$schema->quoteIdentifier($table),
'urlhash',
// The line below is "result of sha256 on column `url`"
'SHA2(url, 256)'));
echo "DONE.\n";
echo "Resuming core schema upgrade...";
}
}