import os import argostranslate.package import argostranslate.translate from tkinter import filedialog, Tk import pdfplumber from reportlab.lib.pagesizes import letter from reportlab.pdfgen import canvas from reportlab.pdfbase import pdfmetrics from reportlab.pdfbase.ttfonts import TTFont from PyPDF2 import PdfReader, PdfWriter from io import BytesIO from tqdm import tqdm import re import sys import traceback # Limbile disponibile SOURCE_LANGUAGES = { "1": ("en", "Engleză"), "2": ("fr", "Franceză"), "3": ("ru", "Rusă"), "4": ("el", "Greacă"), "5": ("grc", "Greacă veche"), "6": ("es", "Spaniolă"), "7": ("de", "Germană"), "8": ("it", "Italiană"), "9": ("la", "Latină") } TARGET_LANGUAGES = { "1": ("ro", "Română"), "2": ("en", "Engleză"), "3": ("fr", "Franceză") } # Configurare traducere Argos def setup_argos_translation(from_lang, to_lang): try: print(f"Se configurează traducerea {from_lang} -> {to_lang}...") argostranslate.package.update_package_index() available_packages = argostranslate.package.get_available_packages() package = next((p for p in available_packages if p.from_code == from_lang and p.to_code == to_lang), None) if package: if package not in argostranslate.package.get_installed_packages(): package_path = package.download() argostranslate.package.install_from_path(package_path) return True if from_lang != "en" and to_lang != "en": package1 = next((p for p in available_packages if p.from_code == from_lang and p.to_code == "en"), None) package2 = next((p for p in available_packages if p.from_code == "en" and p.to_code == to_lang), None) if package1 and package2: for pkg in [package1, package2]: if pkg not in argostranslate.package.get_installed_packages(): package_path = pkg.download() argostranslate.package.install_from_path(package_path) return True return False except Exception as e: print(f"Eroare la configurarea traducătorului: {str(e)}") return False # Selectare limbă def select_language(prompt, languages): print(prompt) for num, (_, name) in languages.items(): print(f"{num}: {name}") choice = input("Introdu numărul limbii: ").strip() return languages.get(choice, (None, None))[0] # Extragere text și proprietăți def extract_text_with_properties(pdf_file): text_lines = [] try: with pdfplumber.open(pdf_file) as pdf: for page_num, page in enumerate(pdf.pages, 1): words = page.extract_words(keep_blank_chars=True) for word in words: text_lines.append({ "text": word["text"], "x0": word["x0"], "top": word["top"], "x1": word["x1"], "bottom": word["bottom"], "page": page_num, "width": page.width, "height": page.height, "fontname": word.get("fontname", "DejaVuSans"), "size": word.get("size", 12) }) return text_lines except Exception as e: print(f"Eroare la extragerea textului: {str(e)}") return None # Gestionare formule matematice def handle_math_formulas(text): math_pattern = r'\$\$?.*?\$\$?|\$.*?\$' formulas = re.findall(math_pattern, text) text_with_placeholder = re.sub(math_pattern, "###MATH###", text) return formulas, text_with_placeholder # Traducere text def translate_with_formulas(text, from_lang, to_lang): try: formulas, text_placeholder = handle_math_formulas(text) translated = argostranslate.translate.translate(text_placeholder, from_lang, to_lang) for formula in formulas: translated = translated.replace("###MATH###", formula, 1) return translated except Exception as e: print(f"Eroare la traducere: {str(e)}") return text # Calcul lățime text def get_text_width(text, fontname, fontsize): try: font = pdfmetrics.getFont(fontname) return font.stringWidth(text, fontsize) except: return pdfmetrics.stringWidth(text, fontname, fontsize) # Generare PDF tradus def generate_translated_pdf(text_data, translated_texts, input_pdf, output_file): try: pdfmetrics.registerFont(TTFont("DejaVuSans", "DejaVuSans.ttf")) pdfmetrics.registerFont(TTFont("DejaVuSans-Bold", "DejaVuSans-Bold.ttf")) pdfmetrics.registerFont(TTFont("DejaVuSans-Oblique", "DejaVuSans-Oblique.ttf")) pdf_reader = PdfReader(input_pdf) pdf_writer = PdfWriter() temp_pdf = BytesIO() pdf_canvas = canvas.Canvas(temp_pdf, pagesize=letter) current_page = 1 for i, item in enumerate(text_data): if item["page"] != current_page: pdf_canvas.showPage() current_page = item["page"] pdf_canvas.setPageSize((item["width"], item["height"])) translated_text = translated_texts[i] x0 = item["x0"] y_pos = item["height"] - item["bottom"] fontname = "DejaVuSans" if "Bold" in item["fontname"]: fontname = "DejaVuSans-Bold" elif "Italic" in item["fontname"] or "Oblique" in item["fontname"]: fontname = "DejaVuSans-Oblique" font_size = item["size"] original_width = item["x1"] - item["x0"] text_width = get_text_width(translated_text, fontname, font_size) if text_width > original_width: font_size = font_size * (original_width / text_width) font_size = max(font_size, 6) pdf_canvas.setFont(fontname, font_size) pdf_canvas.drawString(x0, y_pos, translated_text) pdf_canvas.save() temp_pdf.seek(0) translated_pdf = PdfReader(temp_pdf) for page_num in range(len(pdf_reader.pages)): original_page = pdf_reader.pages[page_num] translated_page = translated_pdf.pages[page_num] if page_num < len(translated_pdf.pages) else None new_page = original_page if translated_page: new_page.merge_page(translated_page) pdf_writer.add_page(new_page) with open(output_file, "wb") as f: pdf_writer.write(f) print(f"PDF generat: {output_file}") except Exception as e: print(f"Eroare la generarea PDF-ului: {str(e)}") # Funcția principală def process_pdf_translation(): try: from_lang = select_language("Selectați limba sursă:", SOURCE_LANGUAGES) if not from_lang: return to_lang = select_language("Selectați limba țintă:", TARGET_LANGUAGES) if not to_lang: return if not setup_argos_translation(from_lang, to_lang): print("Eroare la configurarea traducătorului.") return root = Tk() root.withdraw() pdf_file = filedialog.askopenfilename(title="Selectează PDF", filetypes=[("PDF files", "*.pdf")]) root.destroy() if not pdf_file: return output_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "translated") os.makedirs(output_dir, exist_ok=True) text_data = extract_text_with_properties(pdf_file) if not text_data: return translated_texts = [translate_with_formulas(item["text"], from_lang, to_lang) for item in text_data] output_file = os.path.join(output_dir, f"{os.path.splitext(os.path.basename(pdf_file))[0]}_translated_{to_lang}.pdf") generate_translated_pdf(text_data, translated_texts, pdf_file, output_file) except Exception as e: print(f"Eroare: {str(e)}") if __name__ == "__main__": process_pdf_translation()