7 files changed, 184 insertions, 0 deletions

diff --git a/main.py b/main.py
new file mode 100644
index 0000000..e4157e6
--- /dev/null
+++ b/main.py
@@ -0,0 +1,46 @@
+import sys
+from shutil import copyfile
+import argparse
+
+from src.parsers import pdf
+
+
+def create_arg_parser():
+    parser = argparse.ArgumentParser(description='Metadata anonymisation toolkit 2')
+    parser.add_argument('files', nargs='*')
+
+    info = parser.add_argument_group('Information')
+    info.add_argument('-c', '--check', action='store_true',
+                      help='check if a file is free of harmful metadatas')
+    info.add_argument('-l', '--list', action='store_true',
+                      help='list all supported fileformats')
+    info.add_argument('-s', '--show', action='store_true',
+                      help='list all the harmful metadata of a file without removing them')
+    return parser
+
+def show_meta(file_name:str):
+    p = pdf.PDFParser(file_name)
+    for k,v in p.get_meta().items():
+        print("%s: %s" % (k, v))
+
+def main():
+    argparser = create_arg_parser()
+    args = argparser.parse_args()
+
+    if args.show:
+        for f in args.files:
+            show_meta(f)
+        return 0
+    elif not args.files:
+        return parser.show_help()
+
+    copyfile(sys.argv[1] + '.bak', sys.argv[1])
+    p = pdf.PDFParser(sys.argv[1])
+    p.remove_all()
+    p = pdf.PDFParser('OUT_clean.pdf')
+    print("ok")
+
+
+if __name__ == '__main__':
+
+    main()
diff --git a/src/__init__.py b/src/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/src/__init__.py
diff --git a/src/parsers/__init__.py b/src/parsers/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/src/parsers/__init__.py
diff --git a/src/parsers/abstract.py b/src/parsers/abstract.py
new file mode 100644
index 0000000..a9129cc
--- /dev/null
+++ b/src/parsers/abstract.py
@@ -0,0 +1,10 @@
+class AbstractParser(object):
+    def __init__(self, filename: str):
+        self.filename = filename
+        self.meta_list = set()
+
+    def get_meta(self):
+        raise NotImplementedError
+
+    def remove_all(self):
+        raise NotImplementedError
diff --git a/src/parsers/pdf.py b/src/parsers/pdf.py
new file mode 100644
index 0000000..c25b324
--- /dev/null
+++ b/src/parsers/pdf.py
@@ -0,0 +1,106 @@
+""" Handle PDF
+
+"""
+
+import os
+import logging
+import tempfile
+import shutil
+import io
+
+import cairo
+import gi
+gi.require_version('Poppler', '0.18')
+from gi.repository import Poppler
+
+try:
+    from PIL import Image
+except ImportError:
+    Image = None
+
+from . import abstract
+
+logging.basicConfig(level=logging.DEBUG)
+
+
+class PDFParser(abstract.AbstractParser):
+    def __init__(self, filename):
+        super().__init__(filename)
+        self.meta_list = {'title', 'author', 'subject',
+            'keywords', 'creator', 'producer', 'metadata'}
+        self.uri = 'file://' + os.path.abspath(self.filename)
+        self.password = None
+
+    def __optimize_image_size(self, img: io.BytesIO) -> io.BytesIO:
+        """ This is useless as fuck. """
+        if Image is None:
+            return img
+        ret = io.BytesIO()
+        im = Image.open(img)
+        w, h = im.size
+        resized = im.resize((w, h), Image.ANTIALIAS)
+        resized.save(ret, optimize=True, format="PNG")
+        ret.seek(0)
+
+        return ret
+
+
+    def remove_all(self):
+        """
+            Load the document into Poppler, render pages on PNG,
+            and shove those PNG into a new PDF. Metadata from the new
+            PDF are removed via Poppler, because there is no way to tell
+            cairo to not add "created by cairo" during rendering.
+
+            TODO: Improve the resolution
+            TODO: Don't use a temp file
+        """
+        document = Poppler.Document.new_from_file(self.uri, self.password)
+
+        pdf_surface = cairo.PDFSurface("OUT.pdf", 128, 128)
+        pdf_context = cairo.Context(pdf_surface)
+
+        for pagenum in range(document.get_n_pages()):
+            page = document.get_page(pagenum)
+            page_width, page_height = page.get_size()
+            logging.info("Rendering page %d/%d", pagenum + 1, document.get_n_pages())
+
+            img_surface = cairo.ImageSurface(cairo.FORMAT_ARGB32, int(page_width)*2, int(page_height)*2)
+            img_context = cairo.Context(img_surface)
+
+            img_context.scale(2, 2)
+            page.render_for_printing_with_options(img_context, Poppler.PrintFlags.DOCUMENT)
+            img_context.show_page()
+
+            buf = io.BytesIO()
+            img_surface.write_to_png(buf)
+            img_surface.finish()
+            buf.seek(0)
+
+            #buf = self.__optimize_image_size(buf)
+
+            img = cairo.ImageSurface.create_from_png(buf)
+            pdf_surface.set_size(page_width*2, page_height*2)
+            pdf_context.set_source_surface(img, 0, 0)
+            pdf_context.paint()
+            pdf_context.show_page()
+
+        pdf_surface.finish()
+
+        document = Poppler.Document.new_from_file('file://' + os.path.abspath('OUT.pdf'), self.password)
+        document.set_producer('totally not MAT2 ;)')
+        document.set_creator('')
+        document.save('file://' + os.path.abspath("OUT_clean.pdf"))
+
+        return True
+
+    def get_meta(self):
+        """ Return a dict with all the meta of the file
+        """
+        print("URI: %s", self.uri)
+        document = Poppler.Document.new_from_file(self.uri, self.password)
+        metadata = {}
+        for key in self.meta_list:
+            if document.get_property(key):
+                metadata[key] = document.get_property(key)
+        return metadata
diff --git a/tests/data/dirty.pdf b/tests/data/dirty.pdf
new file mode 100644
index 0000000..0d88779
--- /dev/null
+++ b/tests/data/dirty.pdf
diff --git a/tests/main.py b/tests/main.py
new file mode 100644
index 0000000..52828af
--- /dev/null
+++ b/tests/main.py
@@ -0,0 +1,22 @@
+#!/usr/bin/python3
+
+import unittest
+
+class TestCleaning(unittest.TestCase):
+    def test_pdf(self):
+        self.assertEqual('foo'.upper(), 'FOO')
+
+    def test_isupper(self):
+        self.assertTrue('FOO'.isupper())
+        self.assertFalse('Foo'.isupper())
+
+    def test_split(self):
+        s = 'hello world'
+        self.assertEqual(s.split(), ['hello', 'world'])
+        # check that s.split fails when the separator is not a string
+        with self.assertRaises(TypeError):
+            s.split(2)
+
+
+if __name__ == '__main__':
+    unittest.main()