Fix table of contents not preserved in PDF/A

src/ocrmypdf/pdfinfo.py

@@ -29,7 +29,7 @@ from enum import Enum
 from contextlib import contextmanager
 
 import PyPDF2 as pypdf
-from fitz import Document
+import fitz
 
 from .helpers import universal_open
 
@@ -555,7 +555,7 @@ def borrow_stream(stream):
 
 
 def _page_has_text(infile, pageno):
-    doc = Document(infile)
+    doc = fitz.Document(infile)
     text = doc.getPageText(pageno)
     if text.strip() != '':
         return True
@@ -692,6 +692,7 @@ class PdfInfo:
     def __init__(self, infile):
         self._infile = infile
         self._pages = _pdf_get_all_pageinfo(infile)
+        self._toc = fitz.Document(infile).getToC()
 
     @property
     def pages(self):
@@ -712,6 +713,10 @@ class PdfInfo:
             raise NotImplementedError("can't get filename from stream")
         return self._infile
 
+    @property
+    def table_of_contents(self):
+        return self._toc
+
     def __getitem__(self, item):
         return self._pages[item]
 

src/ocrmypdf/pipeline.py

@@ -963,11 +963,14 @@ def merge_pages_ghostscript(
     ghostscript.generate_pdfa(
         pdf_version=input_pdfinfo.min_version,
         pdf_pages=pdf_pages,
-        output_file=output_file,
+        output_file=output_file + '_toc.pdf',
         compression=options.pdfa_image_compression,
         log=log,
         threads=options.jobs or 1,
         pdfa_part=('1' if options.output_type == 'pdfa-1' else '2'))
+    doc = fitz.Document(output_file + '_toc.pdf')
+    doc.setToC(input_pdfinfo.table_of_contents)
+    doc.save(output_file)
 
 
 def merge_pages_qpdf(

tests/resources/README.rst

@@ -128,6 +128,7 @@ Assemblies
 
 These test resources are assemblies or derivatives from other previously mentioned files, released under the same license terms as their input files.
 
+- baiona_gray.png (from baiona.png)
 - cardinal.pdf (four cardinal directions, baked-in rotated copies of LinnSequencer.jpg)
 - ccitt.pdf (LinnSequencer.jpg, converted to CCITT encoding)
 - encrypted_algo4.pdf (congress.jpg, encrypted with algorithm 4 - not supported by PyPDF2)
@@ -135,12 +136,11 @@ These test resources are assemblies or derivatives from other previously mention
 - jbig2.pdf (congress.jpg, converted to JBIG2 encoding)
 - multipage.pdf (from several other files)
 - palette.pdf (congress.jpg, converted to a 256-color palette)
-- rotated_skew.pdf (a /Rotate'd and skewed document from LinnSequencer.jpg)
-- skew.pdf (from LinnSequencer.jpg, skew simulated by adjusting the transformation matrix)
-- skew-encrypted.pdf (skew.pdf with encryption - access supported by PyPDF2, password is "password")
-- baiona_gray.png (from baiona.png)
 - poster.pdf (from LinnSequencer.jpg)
-
+- rotated_skew.pdf (a /Rotate'd and skewed document from LinnSequencer.jpg)
+- skew-encrypted.pdf (skew.pdf with encryption - access supported by PyPDF2, password is "password")
+- skew.pdf (from LinnSequencer.jpg, skew simulated by adjusting the transformation matrix)
+- toc.pdf (from formxobject.pdf, trivial.pdf)
 
 
 .. _`Wikimedia: LinnSequencer`: https://upload.wikimedia.org/wikipedia/en/b/b7/LinnSequencer_hardware_MIDI_sequencer_brochure_page_2_300dpi.jpg

tests/test_metadata.py

@@ -18,6 +18,7 @@
 
 import pytest
 import PyPDF2 as pypdf
+import fitz
 
 from ocrmypdf.pdfa import file_claims_pdfa
 from ocrmypdf.exceptions import ExitCode
@@ -94,3 +95,22 @@ def test_high_unicode(spoof_tesseract_noop, resources, no_outpdf):
         env=spoof_tesseract_noop)
 
     assert p.returncode == ExitCode.bad_args, err
+
+
+@pytest.mark.parametrize('ocr_option', ['--skip-text', '--force-ocr'])
+@pytest.mark.parametrize('output_type', ['pdf', 'pdfa'])
+def test_bookmarks_preserved(spoof_tesseract_noop, output_type, ocr_option,
+                             resources, outpdf):
+    input_file = resources / 'toc.pdf'
+    before_toc = fitz.Document(str(input_file)).getToC()
+
+    check_ocrmypdf(
+        input_file, outpdf,
+        ocr_option,
+        '--output-type', output_type,
+        env=spoof_tesseract_noop)
+
+    after_toc = fitz.Document(str(outpdf)).getToC()
+    print(before_toc)
+    print(after_toc)
+    assert before_toc == after_toc