From acb9b2d14e3173029aa44e853662d11efbe4fa5e Mon Sep 17 00:00:00 2001
From: jvoisin <julien.voisin@dustri.org>
Date: Sun, 18 Mar 2018 23:48:14 +0100
Subject: [PATCH] Clean metadata
---
src/parsers/pdf.py | 37 ++++++++++++++++---------------------
tests/test_libmat2.py | 5 ++---
2 files changed, 18 insertions(+), 24 deletions(-)
diff --git a/src/parsers/pdf.py b/src/parsers/pdf.py
index a77eabd..26985c6 100644
--- a/src/parsers/pdf.py
+++ b/src/parsers/pdf.py
@@ -7,17 +7,13 @@ import logging
import tempfile
import shutil
import io
+import tempfile
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)
@@ -26,8 +22,9 @@ 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.meta_list = {'author', 'creation-date', 'creator', 'format', 'keywords',
+ 'metadata', 'mod-date', 'producer', 'subject', 'title',
+ 'viewer-preferences'}
self.uri = 'file://' + os.path.abspath(self.filename)
self.password = None
@@ -37,25 +34,24 @@ class PDFParser(abstract.AbstractParser):
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)
+ pages_count = document.get_n_pages()
- pdf_surface = cairo.PDFSurface(self.output_filename, 128, 128)
+ _, tmp_path = tempfile.mkstemp()
+ pdf_surface = cairo.PDFSurface(tmp_path, 128, 128)
pdf_context = cairo.Context(pdf_surface)
- for pagenum in range(document.get_n_pages()):
+ for pagenum in range(pages_count):
page = document.get_page(pagenum)
page_width, page_height = page.get_size()
- logging.info("Rendering page %d/%d", pagenum + 1, document.get_n_pages())
+ logging.info("Rendering page %d/%d", pagenum + 1, pages_count)
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)
+ page.render_for_printing(img_context)
img_context.show_page()
buf = io.BytesIO()
@@ -63,8 +59,6 @@ class PDFParser(abstract.AbstractParser):
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)
@@ -73,11 +67,12 @@ class PDFParser(abstract.AbstractParser):
pdf_surface.finish()
- # This is removing metadata
- #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"))
+ # This is removing metadata added by Poppler
+ document = Poppler.Document.new_from_file('file://' + tmp_path)
+ document.set_producer('')
+ document.set_creator('')
+ document.save('file://' + os.path.abspath(self.output_filename))
+ os.remove(tmp_path)
return True
diff --git a/tests/test_libmat2.py b/tests/test_libmat2.py
index 4751aa4..4b36270 100644
--- a/tests/test_libmat2.py
+++ b/tests/test_libmat2.py
@@ -31,6 +31,5 @@ class TestCleaning(unittest.TestCase):
self.assertTrue(ret)
p = pdf.PDFParser('./tests/data/clean.pdf.cleaned')
- remaining_meta = {'creator': 'cairo 1.14.10 (http://cairographics.org)',
- 'producer': 'cairo 1.14.10 (http://cairographics.org)'}
- self.assertEqual(p.get_meta(), remaining_meta)
+ expected_meta = {'creation-date': -1, 'format': 'PDF-1.5', 'mod-date': -1}
+ self.assertEqual(p.get_meta(), expected_meta)
--
GitLab