Preliminary support for openxml office format

2 files changed, 40 insertions, 3 deletions

diff --git a/lib/mat.py b/lib/mat.py
index 8fe6fb4..9f3f6c5 100644
--- a/lib/mat.py
+++ b/lib/mat.py
@@ -7,6 +7,7 @@
 import os
 import subprocess
 import logging
+import mimetypes
 import xml.sax
 
 import hachoir_core.cmd_line
@@ -30,9 +31,11 @@ STRIPPERS = {
     'application/x-bzip2': archive.Bzip2Stripper,
     'application/zip': archive.ZipStripper,
     'audio/mpeg': audio.MpegAudioStripper,
+    'image/gif': images.GifStripper,
     'image/jpeg': images.JpegStripper,
     'image/png': images.PngStripper,
     'application/vnd.oasis.opendocument': office.OpenDocumentStripper,
+    'application/vnd.openxmlformats-officedocument': office.OpenXmlStripper,
 }
 
 try:
@@ -140,15 +143,18 @@ def create_class_file(name, backup, add2archive):
 
     mime = parser.mime_type
 
+    if mime == 'application/zip': # some formats are zipped stuff
+        mime = mimetypes.guess_type(name)[0]
+
     if mime.startswith('application/vnd.oasis.opendocument'):
         mime = 'application/vnd.oasis.opendocument'  # opendocument fileformat
-
-    #stripper_class = STRIPPERS[mime]
+    elif mime.startswith('application/vnd.openxmlformats-officedocument'):
+        mime = 'application/vnd.openxmlformats-officedocument'
 
     try:
         stripper_class = STRIPPERS[mime]
     except KeyError:
-        logging.info('Don\'t have stripper for %s\'s format' % name)
+        logging.info('Don\'t have stripper for %s format' % mime)
         return
 
     return stripper_class(filename, parser, mime, backup, add2archive)
diff --git a/lib/office.py b/lib/office.py
index f236d09..3cbc566 100644
--- a/lib/office.py
+++ b/lib/office.py
@@ -178,3 +178,34 @@ class PdfStripper(parser.GenericParser):
                     self.document.get_property(key) != '':
                     metadata[key] = self.document.get_property(key)
         return metadata
+
+
+class OpenXmlStripper(archive.GenericArchiveStripper):
+    '''
+        Represent an office openxml document, which is like
+        an opendocument format, with some tricky stuff added.
+        It contains mostly xml, but can have media blobs, crap, ...
+        (I don't like this format.)
+    '''
+    def is_clean(self):
+        return False
+
+    def get_meta(self):
+        '''
+            Return a dict with all the meta of the file
+        '''
+        zipin = zipfile.ZipFile(self.filename, 'r')
+        metadata = {}
+        try:
+            content = zipin.read('docProps/app.xml')
+            metadata['app'] = 'harful meta'
+        except KeyError:  # no app.xml file found
+            logging.debug('%s has no app.xml metadata' % self.filename)
+        try:
+            content = zipin.read('docProps/core.xml')
+            metadata['core'] = 'harmful meta'
+        except KeyError:  # no core.xml found
+            logging.debug('%s has no core.xml metadata' % self.filename)
+        zipin.close()
+
+        return metadata