From 9e0f6cf0ea0a992450c12aec73b459403de5f3c2 Mon Sep 17 00:00:00 2001
From: jvoisin
Date: Tue, 21 Jun 2011 17:27:11 +0200
Subject: Pdf metadata support

---
 lib/images.py |  4 +++-
 lib/mat.py    |  8 +++++++-
 lib/parser.py | 30 +++++-------------------------
 3 files changed, 15 insertions(+), 27 deletions(-)

diff --git a/lib/images.py b/lib/images.py
index 21229c2..76696fd 100644
--- a/lib/images.py
+++ b/lib/images.py
@@ -11,7 +11,9 @@ class JpegStripper(parser.Generic_parser):
 
 class PngStripper(parser.Generic_parser):
     def _should_remove(self, field):
-        if field.name in ('comment'):
+        if field.name.startswith("text["):
+            return True
+        elif field.name is "time":
             return True
         else:
             return False
diff --git a/lib/mat.py b/lib/mat.py
index 5641c62..3cbd81b 100644
--- a/lib/mat.py
+++ b/lib/mat.py
@@ -1,4 +1,4 @@
-#!/usr/bin/python
+#!/usr/bin/env python
 
 '''
     Metadata anonymisation toolkit library
@@ -12,6 +12,8 @@ import hachoir_parser
 import hachoir_editor
 
 import images
+import audio
+import misc
 
 __version__ = "0.1"
 __author__ = "jvoisin"
@@ -19,6 +21,8 @@ __author__ = "jvoisin"
 strippers = {
     hachoir_parser.image.JpegFile: images.JpegStripper,
     hachoir_parser.image.PngFile: images.PngStripper,
+    hachoir_parser.audio.MpegAudioFile: audio.MpegAudioStripper,
+    hachoir_parser.misc.PDFDocument: misc.PdfStripper,
 }
 
 def create_class_file(name):
@@ -50,4 +54,6 @@ def create_class_file(name):
         #Place for another lib than hachoir
         print("Don't have stripper for file type: %s" % editor.description)
         sys.exit(1)
+    if editor.input.__class__ == hachoir_parser.misc.PDFDocument:
+        return stripper_class(filename)
     return stripper_class(realname, filename, parser, editor)
diff --git a/lib/parser.py b/lib/parser.py
index 6c9f6f3..1084de5 100644
--- a/lib/parser.py
+++ b/lib/parser.py
@@ -4,7 +4,6 @@
 
 import hachoir_core.error
 import hachoir_parser
-import hachoir_metadata
 import hachoir_editor
 import sys
 
@@ -12,32 +11,10 @@ POSTFIX = ".cleaned"
 
 class Generic_parser():
     def __init__(self, realname, filename, parser, editor):
-        self.meta = {}
         self.filename = filename
         self.realname = realname
         self.parser = parser
         self.editor = editor
-        #self.meta = self.__fill_meta()
-
-    def __fill_meta(self):
-        metadata = {}
-        try:
-            meta = hachoir_metadata.extractMetadata(self.parser)
-        except hachoir_core.error.HachoirError, err:
-            print("Metadata extraction error: %s" % err)
-
-        if not meta:
-            print("Unable to extract metadata from the file %s" % self.filename)
-            #sys.exit(1)
-
-        for title in meta:
-            #fixme i'm so dirty
-            if title.values != []: #if the field is not empty
-                value = ""
-                for item in title.values:
-                    value = item.text
-                metadata[title.key] = value
-        return metadata
 
     def is_clean(self):
         '''
@@ -84,10 +61,13 @@ class Generic_parser():
         '''
             return a dict with all the meta of the file
         '''
-        metadata = []
+        metadata = {}
         for field in self.editor:
             if self._should_remove(field):
-                metadata.append(field.name)
+                try:
+                    metadata[field.name] = field.value
+                except:
+                    metadata[field.name] = "harmful content"
         return metadata
 
     def _should_remove(self, key):
-- 
cgit v1.3