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(