"""Re-run translation with the new B3.5 code and save to disk.""" import sys sys.path.insert(0, '.') import shutil from pathlib import Path import fitz src = Path('sample_files/test_corpus/test_pdf.pdf') dst = Path('sample_files/test_corpus/test_pdf_translated.pdf') from translators.pdf_translator import PDFTranslator PDFTranslator().translate_file(src, dst, target_language='fr', source_language='en') o = fitz.open(src) n = fitz.open(dst) print(f'Original: {len(o)} pages, {src.stat().st_size} bytes') print(f'Translated: {len(n)} pages, {dst.stat().st_size} bytes') print() print('Page 1 font sizes:') o_sizes = set() n_sizes = set() for b in o[0].get_text("dict").get("blocks", []): for l in b.get("lines", []): for s in l.get("spans", []): o_sizes.add(round(s.get("size", 0), 1)) for b in n[0].get_text("dict").get("blocks", []): for l in b.get("lines", []): for s in l.get("spans", []): n_sizes.add(round(s.get("size", 0), 1)) print(' Original: ', sorted(o_sizes)) print(' Translated:', sorted(n_sizes)) o.close() n.close()