You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
--doc-author <STR>, --doc-subject <STR>, --doc-keywords <STR> —
populate PDF document metadata for --md-to-pdf / --html-to-pdf.
Implemented via post-generation binary patch of the PDF Info dictionary
(Chrome's printToPDF does not read <meta> tags for author/subject/keywords).
HTML <meta name="author|description|keywords"> tags are also injected
in the generated HTML for maximum compatibility.
Verifiable via pdfinfo <output>.pdf.
Changed
The recon manual PDF (docs/MANUAL.pdf) is now regenerated with all
four --doc-* metadata fields populated, dogfooding the new flags.
OUT-OF-SCOPE.md: removed "PDF metadata beyond title" from the
Document conversions Waiting list. Other-markup → PDF (reST, AsciiDoc,
Org) remains deferred — no production-ready pure-Rust parsers for those
formats as of mid-2025.