Fix blank pages bug (fixes #98)

5 files changed, 63 insertions, 13 deletions

diff --git a/paper2remarkable/pdf_ops.py b/paper2remarkable/pdf_ops.py
index 93f1200..e0a67d2 100644
--- a/paper2remarkable/pdf_ops.py
+++ b/paper2remarkable/pdf_ops.py
@@ -44,16 +44,15 @@ def blank_pdf(filepath):
     """Add blank pages to PDF"""
     logger.info("Adding blank pages")
     pdf = Pdf.open(filepath)
-
-    previous_pages = pdf.pages
-    pdf.pages = []
-
-    for page in previous_pages:
-        pdf.pages.append(page)
-        pdf.add_blank_page()
-
+    # Note: creating a new file doesn't keep the table of contents, but it
+    # would anyway be incorrect when adding blank pages
+    dst = Pdf.new()
+    for page in pdf.pages:
+        dst.pages.append(page)
+        x0, y0, x1, y1 = page.MediaBox
+        dst.add_blank_page(page_size=(x1 - x0, y1 - y0))
     output_file = os.path.splitext(filepath)[0] + "-blank.pdf"
-    pdf.save(output_file)
+    dst.save(output_file)
     return output_file
 
 
diff --git a/setup.py b/setup.py
index fb7d21c..7a2ebe3 100644
--- a/setup.py
+++ b/setup.py
@@ -19,7 +19,7 @@ VERSION = None
 
 # What packages are required for this module to be executed?
 REQUIRED = [
-    "pikepdf>=2.8.0",
+    "pikepdf>=2.9.0",
     "beautifulsoup4>=4.8",
     "html2text>=2020.1.16",
     "markdown>=3.1.1",
diff --git a/tests/_constants.py b/tests/_constants.py
new file mode 100644
index 0000000..e0ef956
--- /dev/null
+++ b/tests/_constants.py
@@ -0,0 +1,14 @@
+# -*- coding: utf-8 -*-
+
+TEST_FILE = (
+    "%PDF-1.1\n%¿÷¢þ\n1 0 obj\n<< /Pages 2 0 R /Type /Catalog >>\n"
+    "endobj\n2 0 obj\n<< /Count 1 /Kids [ 3 0 R ] /MediaBox [ 0 0 300 600 ] "
+    "/Type /Pages >>\nendobj\n3 0 obj\n<< /Contents 4 0 R /Parent 2 0 R "
+    "/Resources << /Font << /F1 << /BaseFont /Times-Roman /Subtype /Type1 "
+    "/Type /Font >> >> >> /Type /Page >>\nendobj\n4 0 obj\n<< /Length 44 >>\n"
+    "stream\nBT /F1 18 Tf 80 80 Td (Hello World 1) Tj ET\nendstream\nendobj\n"
+    "xref\n0 5\n0000000000 65535 f \n0000000019 00000 n \n0000000067 00000 n "
+    "\n0000000153 00000 n \n0000000306 00000 n \ntrailer << /Root 1 0 R /Size "
+    "5 /ID [<015d3b8119c73f2291496f4b9d03fe4f>"
+    "<015d3b8119c73f2291496f4b9d03fe4f>] >>\nstartxref\n399\n%%EOF"
+)
diff --git a/tests/test_pdf_ops.py b/tests/test_pdf_ops.py
new file mode 100644
index 0000000..bed763a
--- /dev/null
+++ b/tests/test_pdf_ops.py
@@ -0,0 +1,37 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+"""Tests for the PDF operations"""
+
+import os
+import shutil
+import tempfile
+import unittest
+
+from pikepdf import Pdf
+
+from paper2remarkable.providers import LocalFile
+
+from _constants import TEST_FILE
+
+
+class PdfOpsTestCase(unittest.TestCase):
+    def setUp(self):
+        self._tmpdir = tempfile.mkdtemp(prefix="p2r_test_blank_")
+
+    def tearDown(self):
+        shutil.rmtree(self._tmpdir)
+
+    def test_blank_pages(self):
+        local_filename = os.path.join(self._tmpdir, "test_blank.pdf")
+        with open(local_filename, "w") as fp:
+            fp.write(TEST_FILE)
+        prov = LocalFile(upload=False, blank=True)
+        out_filename = os.path.join(self._tmpdir, "test_blank1.pdf")
+        filename = prov.run(local_filename, filename=out_filename)
+        pdf = Pdf.open(filename)
+        self.assertEqual(len(pdf.pages), 2)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_providers.py b/tests/test_providers.py
index af69c64..e4ab2a7 100644
--- a/tests/test_providers.py
+++ b/tests/test_providers.py
@@ -11,6 +11,7 @@ import pdfplumber
 import shutil
 import tempfile
 import unittest
+
 from pikepdf import Pdf
 
 from paper2remarkable.providers import (
@@ -36,6 +37,7 @@ from paper2remarkable.providers import (
     TandFOnline,
 )
 from paper2remarkable.utils import download_url
+from _constants import TEST_FILE
 
 VERBOSE = False
 
@@ -158,9 +160,7 @@ class TestProviders(unittest.TestCase):
     def test_local(self):
         local_filename = "test.pdf"
         with open(local_filename, "w") as fp:
-            fp.write(
-                "%PDF-1.1\n%¿÷¢þ\n1 0 obj\n<< /Pages 2 0 R /Type /Catalog >>\nendobj\n2 0 obj\n<< /Count 1 /Kids [ 3 0 R ] /MediaBox [ 0 0 300 600 ] /Type /Pages >>\nendobj\n3 0 obj\n<< /Contents 4 0 R /Parent 2 0 R /Resources << /Font << /F1 << /BaseFont /Times-Roman /Subtype /Type1 /Type /Font >> >> >> /Type /Page >>\nendobj\n4 0 obj\n<< /Length 44 >>\nstream\nBT /F1 18 Tf 80 80 Td (Hello World 1) Tj ET\nendstream\nendobj\nxref\n0 5\n0000000000 65535 f \n0000000019 00000 n \n0000000067 00000 n \n0000000153 00000 n \n0000000306 00000 n \ntrailer << /Root 1 0 R /Size 5 /ID [<015d3b8119c73f2291496f4b9d03fe4f><015d3b8119c73f2291496f4b9d03fe4f>] >>\nstartxref\n399\n%%EOF"
-            )
+            fp.write(TEST_FILE)
         prov = LocalFile(upload=False, verbose=VERBOSE)
         filename = prov.run(local_filename)
         self.assertEqual("test_.pdf", os.path.basename(filename))