Merge branch '0.9.x' of gitorious.org:statusnet/mainline into 1.0.x

lib/mediafile.php

@@ -180,7 +180,8 @@ class MediaFile
             return;
         }
 
-        $mimetype = MediaFile::getUploadedFileType($_FILES[$param]['tmp_name']);
+        $mimetype = MediaFile::getUploadedFileType($_FILES[$param]['tmp_name'],
+                                                   $_FILES[$param]['name']);
 
         $filename = null;
 
@@ -241,19 +242,41 @@ class MediaFile
         return new MediaFile($user, $filename, $mimetype);
     }
 
-    static function getUploadedFileType($f) {
+    /**
+     * Attempt to identify the content type of a given file.
+     * 
+     * @param mixed $f file handle resource, or filesystem path as string
+     * @param string $originalFilename (optional) for extension-based detection
+     * @return string
+     * 
+     * @fixme is this an internal or public method? It's called from GetFileAction
+     * @fixme this seems to tie a front-end error message in, kinda confusing
+     * @fixme this looks like it could return a PEAR_Error in some cases, if
+     *        type can't be identified and $config['attachments']['supported'] is true
+     * 
+     * @throws ClientException if type is known, but not supported for local uploads
+     */
+    static function getUploadedFileType($f, $originalFilename=false) {
         require_once 'MIME/Type.php';
+        require_once 'MIME/Type/Extension.php';
+        $mte = new MIME_Type_Extension();
 
         $cmd = &PEAR::getStaticProperty('MIME_Type', 'fileCmd');
         $cmd = common_config('attachments', 'filecommand');
 
         $filetype = null;
 
+        // If we couldn't get a clear type from the file extension,
+        // we'll go ahead and try checking the content. Content checks
+        // are unambiguous for most image files, but nearly useless
+        // for office document formats.
+
         if (is_string($f)) {
 
             // assuming a filename
 
             $filetype = MIME_Type::autoDetect($f);
+
         } else {
 
             // assuming a filehandle
@@ -262,7 +285,32 @@ class MediaFile
             $filetype = MIME_Type::autoDetect($stream['uri']);
         }
 
-        if (common_config('attachments', 'supported') === true || in_array($filetype, common_config('attachments', 'supported'))) {
+        // The content-based sources for MIME_Type::autoDetect()
+        // are wildly unreliable for office-type documents. If we've
+        // gotten an unclear reponse back or just couldn't identify it,
+        // we'll try detecting a type from its extension...
+        $unclearTypes = array('application/octet-stream',
+                              'application/vnd.ms-office',
+                              'application/zip');
+
+        if ($originalFilename && (!$filetype || in_array($filetype, $unclearTypes))) {
+            $type = $mte->getMIMEType($originalFilename);
+            if (is_string($type)) {
+                $filetype = $type;
+            }
+        }
+
+        $supported = common_config('attachments', 'supported');
+        if (is_array($supported)) {
+            // Normalize extensions to mime types
+            foreach ($supported as $i => $entry) {
+                if (strpos($entry, '/') === false) {
+                    common_log(LOG_INFO, "sample.$entry");
+                    $supported[$i] = $mte->getMIMEType("sample.$entry");
+                }
+            }
+        }
+        if ($supported === true || in_array($filetype, $supported)) {
             return $filetype;
         }
         $media = MIME_Type::getMedia($filetype);

tests/MediaFileTest.php

@@ -0,0 +1,96 @@
+<?php
+
+if (isset($_SERVER) && array_key_exists('REQUEST_METHOD', $_SERVER)) {
+    print "This script must be run from the command line\n";
+    exit();
+}
+
+define('INSTALLDIR', realpath(dirname(__FILE__) . '/..'));
+define('STATUSNET', true);
+define('LACONICA', true);
+
+require_once INSTALLDIR . '/lib/common.php';
+
+class MediaFileTest extends PHPUnit_Framework_TestCase
+{
+
+    public function setup()
+    {
+        $this->old_attachments_supported = common_config('attachments', 'supported');
+        $GLOBALS['config']['attachments']['supported'] = true;
+    }
+
+    public function tearDown()
+    {
+        $GLOBALS['config']['attachments']['supported'] = $this->old_attachments_supported;
+    }
+
+    /**
+     * @dataProvider fileTypeCases
+     *
+     */
+    public function testFileType($filename, $expectedType)
+    {
+        if (!file_exists($filename)) {
+            throw new Exception("WTF? $filename test file missing");
+        }
+
+        $type = MediaFile::getUploadedFileType($filename, basename($filename));
+        $this->assertEquals($expectedType, $type);
+    }
+
+    /**
+     * @dataProvider fileTypeCases
+     *
+     */
+    public function testUploadedFileType($filename, $expectedType)
+    {
+        if (!file_exists($filename)) {
+            throw new Exception("WTF? $filename test file missing");
+        }
+        $tmp = tmpfile();
+        fwrite($tmp, file_get_contents($filename));
+
+        $type = MediaFile::getUploadedFileType($tmp, basename($filename));
+        $this->assertEquals($expectedType, $type);
+    }
+
+    static public function fileTypeCases()
+    {
+        $base = dirname(__FILE__);
+        $dir = "$base/sample-uploads";
+        $files = array(
+            "image.png" => "image/png",
+            "image.gif" => "image/gif",
+            "image.jpg" => "image/jpeg",
+            "image.jpeg" => "image/jpeg",
+        
+            "office.pdf" => "application/pdf",
+            
+            "wordproc.odt" => "application/vnd.oasis.opendocument.text",
+            "wordproc.ott" => "application/vnd.oasis.opendocument.text-template",
+            "wordproc.doc" => "application/msword",
+            "wordproc.docx" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+            "wordproc.rtf" => "text/rtf",
+            
+            "spreadsheet.ods" => "application/vnd.oasis.opendocument.spreadsheet",
+            "spreadsheet.ots" => "application/vnd.oasis.opendocument.spreadsheet-template",
+            "spreadsheet.xls" => "application/vnd.ms-excel",
+            "spreadsheet.xlt" => "application/vnd.ms-excel",
+            "spreadsheet.xlsx" => "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+            
+            "presentation.odp" => "application/vnd.oasis.opendocument.presentation",
+            "presentation.otp" => "application/vnd.oasis.opendocument.presentation-template",
+            "presentation.ppt" => "application/vnd.ms-powerpoint",
+            "presentation.pptx" => "application/vnd.openxmlformats-officedocument.presentationml.presentation",
+        );
+
+        $dataset = array();
+        foreach ($files as $file => $type) {
+            $dataset[] = array("$dir/$file", $type);
+        }
+        return $dataset;
+    }
+
+}
+

tests/sample-uploads/wordproc.rtf

@@ -0,0 +1,16 @@
+{\rtf1\ansi\deff0\adeflang1025
+{\fonttbl{\f0\froman\fprq2\fcharset128 Times New Roman;}{\f1\froman\fprq2\fcharset128 Times New Roman;}{\f2\fswiss\fprq2\fcharset128 Arial;}{\f3\fnil\fprq2\fcharset128 DejaVu Sans;}}
+{\colortbl;\red0\green0\blue0;\red128\green128\blue128;}
+{\stylesheet{\s1\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\loch\f0\fs24\lang1033\snext1 Normal;}
+{\s2\sb240\sa120\keepn\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\afs28\lang1081\ltrch\dbch\langfe2052\hich\f2\fs28\lang1033\loch\f2\fs28\lang1033\sbasedon1\snext3 Heading;}
+{\s3\sa120\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\loch\f0\fs24\lang1033\sbasedon1\snext3 Body Text;}
+{\s4\sa120\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\loch\f0\fs24\lang1033\sbasedon3\snext4 List;}
+{\s5\sb120\sa120\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ai\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\i\loch\f0\fs24\lang1033\i\sbasedon1\snext5 caption;}
+{\s6\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\loch\f0\fs24\lang1033\sbasedon1\snext6 Index;}
+}
+{\info{\author Brion }{\creatim\yr2010\mo5\dy10\hr15\min2}{\revtim\yr0\mo0\dy0\hr0\min0}{\printim\yr0\mo0\dy0\hr0\min0}{\comment StarWriter}{\vern3200}}\deftab709
+{\*\pgdsctbl
+{\pgdsc0\pgdscuse195\pgwsxn12240\pghsxn15840\marglsxn1134\margrsxn1134\margtsxn1134\margbsxn1134\pgdscnxt0 Standard;}}
+\paperh15840\paperw12240\margl1134\margr1134\margt1134\margb1134\sectd\sbknone\pgwsxn12240\pghsxn15840\marglsxn1134\margrsxn1134\margtsxn1134\margbsxn1134\ftnbj\ftnstart1\ftnrstcont\ftnnar\aenddoc\aftnrstcont\aftnstart1\aftnnrlc
+\pard\plain \ltrpar\s1\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af3\afs24\lang1081\ltrch\dbch\af3\langfe2052\hich\f0\fs24\lang1033\loch\f0\fs24\lang1033 
+\par }