From 2cba152e7c00ff2c422d5e1c911f17ea07f346ed Mon Sep 17 00:00:00 2001
From: jvoisin
Date: Mon, 6 Feb 2012 02:05:05 +0100
Subject: Merge the two processing mode into a unique one

---
 lib/office.py | 68 +++++++++++++----------------------------------------------
 1 file changed, 15 insertions(+), 53 deletions(-)

(limited to 'lib/office.py')

diff --git a/lib/office.py b/lib/office.py
index e1d738e..82b817e 100644
--- a/lib/office.py
+++ b/lib/office.py
@@ -49,7 +49,7 @@ class OpenDocumentStripper(archive.GenericArchiveStripper):
             logging.debug('%s has no opendocument metadata' % self.filename)
         return metadata
 
-    def _remove_all(self, method):
+    def _remove_all(self):
         '''
             FIXME ?
             There is a patch implementing the Zipfile.remove()
@@ -84,10 +84,7 @@ class OpenDocumentStripper(archive.GenericArchiveStripper):
                     try:
                         cfile = mat.create_class_file(name, False,
                             self.add2archive)
-                        if method == 'normal':
-                            cfile.remove_all()
-                        else:
-                            cfile.remove_all_strict()
+                        cfile.remove_all()
                         logging.debug('Processing %s from %s' % (item,
                             self.filename))
                         zipout.write(name, item)
@@ -137,20 +134,17 @@ class PdfStripper(parser.GenericParser):
             Check if the file is clean from harmful metadatas
         '''
         for key in self.meta_list:
-            if self.document.get_property(key) is not None and \
-                self.document.get_property(key) != '':
+            if self.document.get_property(key) != None:
                 return False
         return True
 
-
     def remove_all(self):
         '''
             Remove supperficial
         '''
         return self._remove_meta()
 
-
-    def remove_all_strict(self):
+    def _remove_meta(self):
         '''
             Opening the PDF with poppler, then doing a render
             on a cairo pdfsurface for each pages.
@@ -166,54 +160,26 @@ class PdfStripper(parser.GenericParser):
         for pagenum in xrange(self.document.get_n_pages()):
             page = self.document.get_page(pagenum)
             context.translate(0, 0)
-            page.render(context)  # render the page on context
+            page.render_for_printing(context)  # render the page on context
             context.show_page()  # draw context on surface
         surface.finish()
-        return self._remove_meta()
 
-    def _remove_meta(self):
-        '''
-            Remove superficial/external metadata
-            from a PDF file, using exiftool,
-            of pdfrw if exiftool is not installed
-        '''
-        processed = False
-        try:# try with pdfrw
+        try:
             import pdfrw
             #For now, poppler cannot write meta, so we must use pdfrw
             logging.debug('Removing %s\'s superficial metadata' % self.filename)
             trailer = pdfrw.PdfReader(self.output)
-            trailer.Info.Producer = trailer.Author = trailer.Info.Creator = None
+            trailer.Info.Producer = None
+            trailer.Info.Creator = None
             writer = pdfrw.PdfWriter()
             writer.trailer = trailer
             writer.write(self.output)
             self.do_backup()
-            processed = True
-        except:
-            pass
-
-        try:  # try with exiftool
-            subprocess.Popen('exiftool', stdout=open('/dev/null'))
-            import exiftool
-            # Note: '-All=' must be followed by a known exiftool option.
-            if self.backup:
-                process = subprocess.Popen(['exiftool', '-m', '-All=',
-                    '-out', self.output, self.filename], stdout=open('/dev/null'))
-                process.wait()
-            else:
-                # Note: '-All=' must be followed by a known exiftool option.
-                process = subprocess.Popen(
-                    ['exiftool', '-All=', '-overwrite_original', self.filename],
-                    stdout=open('/dev/null'))
-                process.wait()
-            processed = True
+            return True
         except:
-            pass
-
-        if processed is False:
-            logging.error('Please install either pdfrw, or exiftool to\
-                    fully handle PDF files')
-        return processed
+            print('Unable to remove all metadata from %s, please install\
+                    pdfrw' % self.output)
+            return False
 
     def get_meta(self):
         '''
@@ -221,8 +187,7 @@ class PdfStripper(parser.GenericParser):
         '''
         metadata = {}
         for key in self.meta_list:
-            if self.document.get_property(key) is not None and \
-                self.document.get_property(key) != '':
+            if self.document.get_property(key) is not None:
                 metadata[key] = self.document.get_property(key)
         return metadata
 
@@ -234,7 +199,7 @@ class OpenXmlStripper(archive.GenericArchiveStripper):
         It contains mostly xml, but can have media blobs, crap, ...
         (I don't like this format.)
     '''
-    def _remove_all(self, method):
+    def _remove_all(self):
         '''
             FIXME ?
             There is a patch implementing the Zipfile.remove()
@@ -258,10 +223,7 @@ class OpenXmlStripper(archive.GenericArchiveStripper):
                     try:
                         cfile = mat.create_class_file(name, False,
                             self.add2archive)
-                        if method == 'normal':
-                            cfile.remove_all()
-                        else:
-                            cfile.remove_all_strict()
+                        cfile.remove_all()
                         logging.debug('Processing %s from %s' % (item,
                             self.filename))
                         zipout.write(name, item)
-- 
cgit v1.3