tools-only Implementation Guide
Code Structure
How is the code organized?
pdf_to_epub/core/pdf_extractor.py
Implementation Notes
Key technical details to remember:
PyMuPDF Usage
import fitz
doc = fitz.open(path)
for page in doc:
text = page.get_text("text", sort=True)
Setting sort=True is vital for correct reading order.
Paragraph Joining
PDF text often has extra newlines. We should join lines that end without "sentence-ending" punctuation or lines that look like parts of the same flow.
Error Handling
How do we handle failures?
- Catch
fitz.FileDataErrorfor corrupted files. - Catch
RuntimeErrorfor encrypted files without password.