PDF Processing Skill

This skill provides guidance on working with PDF files in various ways.

Reading PDFs

Using pdftotext (Fast, Text-Only)

For quick text extraction from PDFs:

pdftotext input.pdf output.txt

Or output to stdout:

pdftotext input.pdf -

Using PyMuPDF (Python, More Features)

For more control and metadata:

import fitz  # PyMuPDF

# Open PDF
doc = fitz.open("input.pdf")

# Extract text from all pages
text = ""
for page in doc:
    text += page.get_text()

# Close document
doc.close()

Using Go (pdfcpu library)

For Go projects:

import "github.com/pdfcpu/pdfcpu/pkg/api"

// Extract text
text, err := api.ExtractTextFile("input.pdf", "", nil)

Creating PDFs

From HTML (wkhtmltopdf)

wkhtmltopdf input.html output.pdf

Using Go (gofpdf)

import "github.com/jung-kurt/gofpdf"

pdf := gofpdf.New("P", "mm", "A4", "")
pdf.AddPage()
pdf.SetFont("Arial", "B", 16)
pdf.Cell(40, 10, "Hello World!")
err := pdf.OutputFileAndClose("output.pdf")

Merging PDFs

Using pdfunite

pdfunite file1.pdf file2.pdf file3.pdf output.pdf

Using Go (pdfcpu)

import "github.com/pdfcpu/pdfcpu/pkg/api"

err := api.MergeFile([]string{"file1.pdf", "file2.pdf"}, "output.pdf", nil)

Best Practices

1. Check file existence before processing
2. Handle errors gracefully - PDFs can be corrupted
3. Consider file size - large PDFs may need streaming
4. Respect encoding - PDFs may contain non-UTF8 text
5. Test with real files - PDF standards vary widely

Common Issues

• Password-protected PDFs: Require password parameter
• Scanned PDFs: May need OCR (tesseract) for text extraction
• Complex layouts: Table extraction may need specialized tools
• Fonts: Embedding fonts may be needed for consistent rendering