* the file mime.types in the includes directory.
*/
define('MM_WELL_KNOWN_MIME_TYPES',<<<END_STRING
-application/ogg ogg ogm ogv
+application/ogg ogx ogg ogm ogv oga spx
application/pdf pdf
application/vnd.oasis.opendocument.chart odc
application/vnd.oasis.opendocument.chart-template otc
audio/mpeg mpga mpa mp2 mp3
audio/x-aiff aif aiff aifc
audio/x-wav wav
-audio/ogg ogg
+audio/ogg oga spx ogg
image/x-bmp bmp
image/gif gif
image/jpeg jpeg jpg jpe
image/x-xcf xcf
text/plain txt
text/html html htm
-video/ogg ogm ogg ogv
+video/ogg ogv ogm ogg
video/mpeg mpg mpeg
END_STRING
);
audio/x-wav [AUDIO]
audio/mp3 audio/mpeg [AUDIO]
application/ogg audio/ogg video/ogg [MULTIMEDIA]
-image/x-bmp image/bmp [BITMAP]
+image/x-bmp image/x-ms-bmp image/bmp [BITMAP]
image/gif [BITMAP]
image/jpeg [BITMAP]
image/png [BITMAP]
global $wgLoadFileinfoExtension;
if ($wgLoadFileinfoExtension) {
- if(!extension_loaded('fileinfo')) dl('fileinfo.' . PHP_SHLIB_SUFFIX);
+ wfDl( 'fileinfo' );
}
/**
* Mapping of media types to arrays of mime types.
* This is used by findMediaType and getMediaType, respectively
*/
- var $mMediaTypes= NULL;
+ var $mMediaTypes= null;
/** Map of mime type aliases
*/
- var $mMimeTypeAliases= NULL;
+ var $mMimeTypeAliases= null;
/** map of mime types to file extensions (as a space seprarated list)
*/
- var $mMimeToExt= NULL;
+ var $mMimeToExt= null;
/** map of file extensions types to mime types (as a space seprarated list)
*/
- var $mExtToMime= NULL;
+ var $mExtToMime= null;
/** IEContentAnalyzer instance
*/
*/
function guessTypesForExtension( $ext ) {
$m = $this->getTypesForExtension( $ext );
- if ( is_null( $m ) ) return NULL;
+ if ( is_null( $m ) ) return null;
$m = trim( $m );
$m = preg_replace( '/\s.*$/', '', $m );
$ext = $this->getExtensionsForType( $mime );
if ( !$ext ) {
- return NULL; //unknown
+ return null; //unknown
}
$ext = explode( ' ', $ext );
'xbm',
// Formats we recognize magic numbers for
- 'djvu', 'ogg', 'ogv', 'mid', 'pdf', 'wmf', 'xcf',
+ 'djvu', 'ogx', 'ogg', 'ogv', 'oga', 'spx', 'mid', 'pdf', 'wmf', 'xcf',
// XML formats we sure hope we recognize reliably
'svg',
* but applies additional checks to determine some well known file formats that may be missed
* or misinterpreter by the default mime detection (namely xml based formats like XHTML or SVG).
*
- * @param string $file The file to check
- * @param mixed $ext The file extension, or true to extract it from the filename.
- * Set it to false to ignore the extension.
+ * @param $file String: the file to check
+ * @param $ext Mixed: the file extension, or true to extract it from the filename.
+ * Set it to false to ignore the extension.
*
* @return string the mime type of $file
*/
}
/*
- * look for PHP
- * Check for this before HTML/XML...
- * Warning: this is a heuristic, and won't match a file with a lot of non-PHP before.
- * It will also match text files which could be PHP. :)
+ * Look for PHP. Check for this before HTML/XML... Warning: this is a
+ * heuristic, and won't match a file with a lot of non-PHP before. It
+ * will also match text files which could be PHP. :)
+ *
+ * FIXME: For this reason, the check is probably useless -- an attacker
+ * could almost certainly just pad the file with a lot of nonsense to
+ * circumvent the check in any case where it would be a security
+ * problem. On the other hand, it causes harmful false positives (bug
+ * 16583). The heuristic has been cut down to exclude three-character
+ * strings like "<? ", but should it be axed completely?
*/
if( ( strpos( $head, '<?php' ) !== false ) ||
- ( strpos( $head, '<? ' ) !== false ) ||
- ( strpos( $head, "<?\n" ) !== false ) ||
- ( strpos( $head, "<?\t" ) !== false ) ||
- ( strpos( $head, "<?=" ) !== false ) ||
( strpos( $head, "<\x00?\x00p\x00h\x00p" ) !== false ) ||
( strpos( $head, "<\x00?\x00 " ) !== false ) ||
/*
* look for shell scripts
*/
- $script_type = NULL;
+ $script_type = null;
# detect by shebang
if ( substr( $head, 0, 2) == "#!" ) {
}
}
- // Check for ZIP (before getimagesize)
+ // Check for ZIP variants (before getimagesize)
if ( strpos( $tail, "PK\x05\x06" ) !== false ) {
- wfDebug( __METHOD__.": ZIP header present at end of $file\n" );
- return $this->detectZipType( $head );
+ wfDebug( __METHOD__.": ZIP header present in $file\n" );
+ return $this->detectZipType( $head, $tail, $ext );
}
wfSuppressWarnings();
/**
* Detect application-specific file type of a given ZIP file from its
- * header data. Currently works for OpenDocument types...
+ * header data. Currently works for OpenDocument and OpenXML types...
* If can't tell, returns 'application/zip'.
*
- * @param string $header Some reasonably-sized chunk of file header
+ * @param $header String: some reasonably-sized chunk of file header
+ * @param $tail String: the tail of the file
+ * @param $ext Mixed: the file extension, or true to extract it from the filename.
+ * Set it to false to ignore the extension.
+ *
* @return string
*/
- function detectZipType( $header ) {
+ function detectZipType( $header, $tail = null, $ext = false ) {
+ $mime = 'application/zip';
$opendocTypes = array(
'chart-template',
'chart',
// http://lists.oasis-open.org/archives/office/200505/msg00006.html
$types = '(?:' . implode( '|', $opendocTypes ) . ')';
$opendocRegex = "/^mimetype(application\/vnd\.oasis\.opendocument\.$types)/";
- wfDebug( __METHOD__.": $opendocRegex\n" );
-
+
+ $openxmlRegex = "/^\[Content_Types\].xml/";
+
if( preg_match( $opendocRegex, substr( $header, 30 ), $matches ) ) {
$mime = $matches[1];
wfDebug( __METHOD__.": detected $mime from ZIP archive\n" );
- return $mime;
+ } elseif( preg_match( $openxmlRegex, substr( $header, 30 ) ) ) {
+ $mime = "application/x-opc+zip";
+ if( $ext !== true && $ext !== false ) {
+ /** This is the mode used by getPropsFromPath
+ * These mime's are stored in the database, where we don't really want
+ * x-opc+zip, because we use it only for internal purposes
+ */
+ if( $this->isMatchingExtension( $ext, $mime) ) {
+ /* A known file extension for an OPC file,
+ * find the proper mime type for that file extension */
+ $mime = $this->guessTypesForExtension( $ext );
+ } else {
+ $mime = "application/zip";
+ }
+ }
+ wfDebug( __METHOD__.": detected an Open Packaging Conventions archive: $mime\n" );
+ } else if( substr( $header, 0, 8 ) == "\xd0\xcf\x11\xe0\xa1\xb1\x1a\xe1" &&
+ ($headerpos = strpos( $tail, "PK\x03\x04" ) ) !== false &&
+ preg_match( $openxmlRegex, substr( $tail, $headerpos + 30 ) ) ) {
+ if( substr( $header, 512, 4) == "\xEC\xA5\xC1\x00" ) {
+ $mime = "application/msword";
+ }
+ switch( substr( $header, 512, 6) ) {
+ case "\xEC\xA5\xC1\x00\x0E\x00":
+ case "\xEC\xA5\xC1\x00\x1C\x00":
+ case "\xEC\xA5\xC1\x00\x43\x00":
+ $mime = "application/vnd.ms-powerpoint";
+ break;
+ case "\xFD\xFF\xFF\xFF\x10\x00":
+ case "\xFD\xFF\xFF\xFF\x1F\x00":
+ case "\xFD\xFF\xFF\xFF\x22\x00":
+ case "\xFD\xFF\xFF\xFF\x23\x00":
+ case "\xFD\xFF\xFF\xFF\x28\x00":
+ case "\xFD\xFF\xFF\xFF\x29\x00":
+ case "\xFD\xFF\xFF\xFF\x10\x02":
+ case "\xFD\xFF\xFF\xFF\x1F\x02":
+ case "\xFD\xFF\xFF\xFF\x22\x02":
+ case "\xFD\xFF\xFF\xFF\x23\x02":
+ case "\xFD\xFF\xFF\xFF\x28\x02":
+ case "\xFD\xFF\xFF\xFF\x29\x02":
+ $mime = "application/vnd.msexcel";
+ break;
+ }
+
+ wfDebug( __METHOD__.": detected a MS Office document with OPC trailer\n");
} else {
wfDebug( __METHOD__.": unable to identify type of ZIP archive\n" );
- return 'application/zip';
}
+ return $mime;
}
/** Internal mime type detection, please use guessMimeType() for application code instead.
* If the mime type is still unknown, getimagesize is used to detect the mime type if the file is an image.
* If no mime type can be determined, this function returns "unknown/unknown".
*
- * @param string $file The file to check
- * @param mixed $ext The file extension, or true to extract it from the filename.
- * Set it to false to ignore the extension.
+ * @param $file String: the file to check
+ * @param $ext Mixed: the file extension, or true to extract it from the filename.
+ * Set it to false to ignore the extension.
*
* @return string the mime type of $file
* @access private
function detectMimeType( $file, $ext = true ) {
global $wgMimeDetectorCommand;
- $m = NULL;
+ $m = null;
if ( $wgMimeDetectorCommand ) {
$fn = wfEscapeShellArg( $file );
$m = `$wgMimeDetectorCommand $fn`;
$m = strtolower( $m );
if ( strpos( $m, 'unknown' ) !== false ) {
- $m = NULL;
+ $m = null;
} else {
wfDebug( __METHOD__.": magic mime type of $file: $m\n" );
return $m;
* @todo analyse file if need be
* @todo look at multiple extension, separately and together.
*
- * @param string $path full path to the image file, in case we have to look at the contents
+ * @param $path String: full path to the image file, in case we have to look at the contents
* (if null, only the mime type is used to determine the media type code).
- * @param string $mime mime type. If null it will be guessed using guessMimeType.
+ * @param $mime String: mime type. If null it will be guessed using guessMimeType.
*
* @return (int?string?) a value to be used with the MEDIATYPE_xxx constants.
*/
- function getMediaType( $path = NULL, $mime = NULL ) {
+ function getMediaType( $path = null, $mime = null ) {
if( !$mime && !$path ) return MEDIATYPE_UNKNOWN;
# If mime type is unknown, guess it
}
# Check for entry for file extension
- $e = NULL;
+ $e = null;
if ( $path ) {
$i = strrpos( $path, '.' );
$e = strtolower( $i ? substr( $path, $i + 1 ) : '' );
* Get the MIME types that various versions of Internet Explorer would
* detect from a chunk of the content.
*
- * @param string $fileName The file name (unused at present)
- * @param string $chunk The first 256 bytes of the file
- * @param string $proposed The MIME type proposed by the server
+ * @param $fileName String: the file name (unused at present)
+ * @param $chunk String: the first 256 bytes of the file
+ * @param $proposed String: the MIME type proposed by the server
*/
public function getIEMimeTypes( $fileName, $chunk, $proposed ) {
$ca = $this->getIEContentAnalyzer();
dépôts / lhc / web / wiklou.git / blobdiff