Fix exception on traversing corrupt ToC entries

src/ocrmypdf/_weave.py

@@ -25,7 +25,7 @@ from .helpers import flatten_groups, page_number
 
 
 def _update_page_resources(*, page, font, font_key, procset):
-    "Update this page's fonts with a reference to the Glyphless font"
+    """Update this page's fonts with a reference to the Glyphless font"""
 
     if '/Resources' not in page:
         page['/Resources'] = pikepdf.Dictionary({})
@@ -160,7 +160,7 @@ def _weave_layers_graft(
 
 
 def _find_font(text, pdf_base):
-    "Copy a font from the filename text into pdf_base"
+    """Copy a font from the filename text into pdf_base"""
 
     font, font_key = None, None
     possible_font_names = ('/f-0-0', '/F1')
@@ -245,12 +245,14 @@ def _fix_toc(pdf_base, pageref_remap, log):
         Inner helper function: change the objgen for any page from the old we
         invalidated to its new one.
         """
-        if not isinstance(dest_node, pikepdf.Array):
-            return
+        try:
         pageref = dest_node[0]
         if pageref['/Type'] == '/Page' and pageref.objgen in pageref_remap:
             new_objgen = pageref_remap[pageref.objgen]
             dest_node[0] = pdf_base.get_object(new_objgen)
+        except (IndexError, TypeError) as e:
+            log.warning("This file may contain invalid table of contents entries")
+            log.debug(e)
 
     def visit_remap_dest(pdf_base, node, log):
         """

tests/test_weave.py

@@ -0,0 +1,40 @@
+# © 2019 James R. Barlow: github.com/jbarlow83
+#
+# This file is part of OCRmyPDF.
+#
+# OCRmyPDF is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# OCRmyPDF is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
+
+from unittest.mock import MagicMock
+import logging
+
+import pytest
+
+import pikepdf
+from ocrmypdf._weave import _fix_toc
+
+def test_invalid_toc(resources, tmpdir, caplog):
+    pdf = pikepdf.open(resources / 'toc.pdf')
+
+    # Corrupt a TOC entry
+    pdf.Root.Outlines.Last.Dest = pikepdf.Array([None, 0.0, 0.1, 0.2])
+    pdf.save(tmpdir / 'test.pdf')
+
+    pdf = pikepdf.open(tmpdir / 'test.pdf')
+    remap = {}
+    remap[pdf.pages[0].objgen] = pdf.pages[0].objgen  # Dummy remap
+
+    # Confirm we complain about the TOC and don't throw an exception
+    log = logging.getLogger()
+    _fix_toc(pdf, remap, log)
+    assert 'invalid table of contents entries' in caplog.text