Script that converts a bunch of separate markdown files into a single continuous PDFs that retains assets (images), tables of contents as well as links.

README.md

Requires the following brew packages to be installed in MacOS, actual list and names will vary depending on operating system:

pandoc
texlive

markdown-to-pdf.sh

#!/bin/bash

# Create a temporary directory for our work
TEMP_DIR=$(mktemp -d)
COMBINED_MD="$TEMP_DIR/combined.md"
OUTPUT_PDF="output.pdf"

echo "Creating combined markdown file..."

# Start with the index file if it exists
if [ -f "index.md" ]; then
  cat index.md > "$COMBINED_MD"
  echo -e "\n\n" >> "$COMBINED_MD"
fi

# Add a table of contents marker
echo -e "\\\\newpage\n# Table of Contents\n\n" >> "$COMBINED_MD"
echo -e "{:toc}\n\n\\\\newpage\n" >> "$COMBINED_MD"

# Function to process markdown files in a directory
process_directory() {
  local dir=$1
  local prefix=$2
  
  # Process files in the current directory first
  for file in "$dir"/*.md; do
    if [ -f "$file" ] && [ "$(basename "$file")" != "index.md" ]; then
      echo "Adding $(basename "$file")..."
      
      # Add a section header based on directory structure
      if [ -n "$prefix" ]; then
        echo -e "\\\\newpage\n# $prefix - $(basename "${file%.md}")\n\n" >> "$COMBINED_MD"
      else
        echo -e "\\\\newpage\n# $(basename "${file%.md}")\n\n" >> "$COMBINED_MD"
      fi
      
      # Add the content with image path correction
      # Get the relative directory path for image reference correction
      local rel_dir=$(dirname "$file")
      if [ "$rel_dir" = "." ]; then
        # File is in the root directory, no path adjustment needed
        cat "$file" >> "$COMBINED_MD"
      else
        # Adjust image references to account for the file's original location
        sed 's|!\[\(.*\)\](\([^/][^:]*\))|![\1]('"$rel_dir"'/\2)|g' "$file" >> "$COMBINED_MD"
      fi
      echo -e "\n\n" >> "$COMBINED_MD"
    fi
  done
  
  # Then process subdirectories
  for subdir in "$dir"/*/; do
    if [ -d "$subdir" ]; then
      local dirname=$(basename "$subdir")
      local new_prefix
      
      if [ -n "$prefix" ]; then
        new_prefix="$prefix / $dirname"
      else
        new_prefix="$dirname"
      fi
      
      process_directory "$subdir" "$new_prefix"
    fi
  done
}

# Process the root directory first (excluding index.md which we already added)
for file in *.md; do
  if [ -f "$file" ] && [ "$file" != "index.md" ] && [ "$file" != "README.md" ] && [ "$file" != "convert_to_pdf.sh" ]; then
    echo "Adding $file..."
    echo -e "\\\\newpage\n# $(basename "${file%.md}")\n\n" >> "$COMBINED_MD"
    
    # Add the content with image path correction for root files
    rel_dir=$(dirname "$file")
    if [ "$rel_dir" = "." ]; then
      # File is in the root directory, no path adjustment needed
      cat "$file" >> "$COMBINED_MD"
    else
      # Adjust image references to account for the file's original location
      sed 's|!\[\(.*\)\](\([^/][^:]*\))|![\1]('"$rel_dir"'/\2)|g' "$file" >> "$COMBINED_MD"
    fi
    echo -e "\n\n" >> "$COMBINED_MD"
  fi
done

# Process each subdirectory
for dir in */; do
  if [ -d "$dir" ]; then
    dirname=$(basename "$dir")
    process_directory "$dir" "$dirname"
  fi
done

echo "Converting to PDF..."

# Use pandoc to convert the combined markdown to PDF
pandoc "$COMBINED_MD" \
  --pdf-engine=xelatex \
  --toc \
  --toc-depth=3 \
  --highlight-style=tango \
  -V geometry:margin=1in \
  -V linkcolor:blue \
  -V toccolor:blue \
  -V mainfont="Arial Unicode MS" \
  -o "$OUTPUT_PDF"

echo "Done! PDF created as $OUTPUT_PDF"