menekis/OCRmyPDF
Public Access
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

pipeline: refactor metadata_fixup

Browse Source
This commit is contained in:
James R. Barlow
2020-12-29 01:47:32 -08:00
parent babc76fa74
commit 96d68c2413
1 changed files with 10 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ocrmypdf/_pipeline.py
+10 -12
View File
@@ -756,19 +756,17 @@ def metadata_fixup(working_file: Path, context: PdfContext):
if 'xmp:CreateDate' not in meta:
meta['xmp:CreateDate'] = meta.get('xmp:ModifyDate', '')
# Ghostscript likes to set title to Untitled if omitted from input.
# Reverse this, because PDF/A TechNote 0003:Metadata in PDF/A-1
# and the XMP Spec do not make this recommendation.
if meta.get('dc:title') == 'Untitled':
with original.open_metadata(
set_pikepdf_as_editor=False, update_docinfo=False
) as original_meta:
if 'dc:title' not in original_meta:
with original.open_metadata(
set_pikepdf_as_editor=False, update_docinfo=False, strict=False
) as meta_original:
if meta.get('dc:title') == 'Untitled':
# Ghostscript likes to set title to Untitled if omitted from input.
# Reverse this, because PDF/A TechNote 0003:Metadata in PDF/A-1
# and the XMP Spec do not make this recommendation.
if 'dc:title' not in meta_original:
del meta['dc:title']
meta_original = original.open_metadata()
missing = set(meta_original.keys()) - set(meta.keys())
report_on_metadata(missing)
missing = set(meta_original.keys()) - set(meta.keys())
report_on_metadata(missing)
pdf.save(
output_file,