pdf

name: pdf description: Generate a PDF from a markdown note file via pandoc + xelatex with mermaid diagram support.

When the user asks to generate a PDF from a markdown note file, follow this workflow:

1. Determine the input file: The user will provide a path or dendron note name. Resolve it to a filename relative to notes/ (e.g., swanki.processing.pdf_processor.md).

2. Determine the output filename: Strip the .md extension from the input filename to use as the output name (e.g., swanki.processing.pdf_processor).

3. Run the PDF generation script:

   bash notes/assets/publish/scripts/bib_tex_pdf.sh <input_filename> . <output_filename>
   

   • First arg: markdown filename (relative to notes/)
   • Second arg: always . (output dir, relative to notes/ after the script's internal cd ./notes)
   • Third arg: output PDF name without extension

4. Verify the output: Check that notes/assets/pdf-output/<output_filename>.pdf exists.

5. Report the result: Tell the user the path to the generated PDF and note any warnings (e.g., missing citations).

Important Notes

• The script uses SCRIPT_DIR to derive NOTES_DIR, so all paths are resolved automatically
• The script uses mermaid-filter for diagram rendering -- if mermaid diagrams produce empty PNGs, the env vars MERMAID_FILTER_LOC, MERMAID_FILTER_SCALE, MERMAID_FILTER_WIDTH may need to be set (the script handles this)
• Unicode emojis will cause xelatex to fail -- if the PDF build fails, check for emojis and replace with plain text alternatives before retrying
• Use a 120s timeout for the bash command since PDF generation can be slow

Example Invocations

• "/pdf notes/swanki.processing.pdf_processor.md"
• "/pdf swanki.processing.pdf_processor.md"
• "generate pdf for the processing module note"