collect.sh

#!/bin/bash

# Check if required tools are installed
command -v pdfimages >/dev/null 2>&1 || { echo "pdfimages is required but not installed. Install poppler-utils."; exit 1; }
command -v img2pdf >/dev/null 2>&1 || { echo "img2pdf is required but not installed. Install img2pdf."; exit 1; }

# Create output directory if it doesn't exist
mkdir -p output_pdfs
mkdir -p temp_images

# Process each PDF file in the current directory
for pdf in *.pdf; do
    if [ -f "$pdf" ]; then
        echo "Processing: $pdf"

        # Create a temporary directory for this PDF's images
        temp_dir="temp_images/${pdf%.*}"
        mkdir -p "$temp_dir"

        # Extract images from PDF
        pdfimages -all "$pdf" "$temp_dir/image"

        # Check if any images were extracted
        if ls "$temp_dir"/image* 1> /dev/null 2>&1; then
            # Create new PDF from extracted images
            img2pdf "$temp_dir"/image* -o "output_pdfs/${pdf%.*}_images.pdf"
            echo "Created: output_pdfs/${pdf%.*}_images.pdf"
        else
            echo "No images found in $pdf"
        fi

        # Clean up temporary files
        rm -rf "$temp_dir"
    fi
done

# Clean up main temporary directory
rm -rf temp_images

echo "Processing complete!"