|
| 1 | +import tkinter as tk |
| 2 | +from tkinter import filedialog, scrolledtext, messagebox, simpledialog, font |
| 3 | +from tkinter import PhotoImage # Import PhotoImage for handling images |
| 4 | +import torch |
| 5 | +from transformers import T5Tokenizer, T5ForConditionalGeneration |
| 6 | +from pdfminer.high_level import extract_text |
| 7 | +import requests |
| 8 | +from io import BytesIO |
| 9 | +from fpdf import FPDF # Import FPDF for PDF generation |
| 10 | + |
| 11 | +model = T5ForConditionalGeneration.from_pretrained('t5-small') |
| 12 | +tokenizer = T5Tokenizer.from_pretrained('t5-small') |
| 13 | +device = torch.device('cpu') |
| 14 | + |
| 15 | +def pdf_to_text(pdf_path): |
| 16 | + try: |
| 17 | + text = extract_text(pdf_path) |
| 18 | + return text |
| 19 | + except Exception as e: |
| 20 | + print(f"Error extracting text from PDF: {e}") |
| 21 | + return None |
| 22 | + |
| 23 | +def summarize_text(text): |
| 24 | + preprocess_text = text.strip().replace("\n", " ") |
| 25 | + t5_prepared_text = "summarize: " + preprocess_text |
| 26 | + tokenized_text = tokenizer.encode(t5_prepared_text, return_tensors="pt").to(device) |
| 27 | + summary_ids = model.generate(tokenized_text, |
| 28 | + num_beams=4, |
| 29 | + no_repeat_ngram_size=2, |
| 30 | + min_length=30, |
| 31 | + max_length=100, |
| 32 | + early_stopping=True) |
| 33 | + output = tokenizer.decode(summary_ids[0], skip_special_tokens=True) |
| 34 | + return output |
| 35 | + |
| 36 | +def download_pdf(url): |
| 37 | + try: |
| 38 | + response = requests.get(url) |
| 39 | + response.raise_for_status() |
| 40 | + return BytesIO(response.content) |
| 41 | + except requests.RequestException as e: |
| 42 | + messagebox.showerror("Download Error", f"Failed to download PDF: {e}") |
| 43 | + return None |
| 44 | + |
| 45 | +def select_files(): |
| 46 | + file_paths = filedialog.askopenfilenames(filetypes=[("PDF files", "*.pdf")]) |
| 47 | + if file_paths: |
| 48 | + summaries = [summarize_text(pdf_to_text(pdf_file)) for pdf_file in file_paths if |
| 49 | + pdf_to_text(pdf_file) is not None] |
| 50 | + output_text.delete('1.0', tk.END) |
| 51 | + output_text.insert(tk.END, "\n\n".join(summaries)) |
| 52 | + messagebox.showinfo("Completion", "Summarization completed!") |
| 53 | + else: |
| 54 | + messagebox.showinfo("No Files", "No PDF files were selected.") |
| 55 | + |
| 56 | +def fetch_from_url(): |
| 57 | + url = simpledialog.askstring("Enter URL", "Please enter the PDF URL:") |
| 58 | + if url: |
| 59 | + pdf_file = download_pdf(url) |
| 60 | + if pdf_file: |
| 61 | + summary = summarize_text(pdf_to_text(pdf_file)) |
| 62 | + output_text.delete('1.0', tk.END) |
| 63 | + output_text.insert(tk.END, summary) |
| 64 | + messagebox.showinfo("Completion", "Summarization completed!") |
| 65 | + |
| 66 | +def save_summary_as_pdf(): |
| 67 | + text_to_save = output_text.get('1.0', tk.END) |
| 68 | + if not text_to_save.strip(): |
| 69 | + messagebox.showerror("No Data", "There is no summary to save.") |
| 70 | + return |
| 71 | + |
| 72 | + file_path = filedialog.asksaveasfilename(defaultextension=".pdf", filetypes=[("PDF files", "*.pdf")]) |
| 73 | + if file_path: |
| 74 | + pdf = FPDF() |
| 75 | + pdf.add_page() |
| 76 | + |
| 77 | + # Calculate center position for logo and title |
| 78 | + logo_x = (pdf.w - 30) / 2 # Centering logo horizontally |
| 79 | + pdf.image("p.png", x=logo_x, y=10, w=30) |
| 80 | + pdf.ln(28) |
| 81 | + # Add header title centered |
| 82 | + pdf.set_font("Arial", size=12) |
| 83 | + pdf.cell(0, 10, "MGM's College Of Engineering", ln=True, align='C') |
| 84 | + |
| 85 | + # Add line separator |
| 86 | + pdf.set_line_width(0.5) |
| 87 | + pdf.line(10, 50, pdf.w - 10, 50) |
| 88 | + |
| 89 | + # Add summary text |
| 90 | + pdf.set_font("Arial", size=10) |
| 91 | + pdf.multi_cell(0, 10, text_to_save) |
| 92 | + |
| 93 | + pdf.output(file_path) |
| 94 | + |
| 95 | + messagebox.showinfo("Saved as PDF", "The summary has been saved as PDF!") |
| 96 | + |
| 97 | +app = tk.Tk() |
| 98 | +app.title("PDF Summarizer") |
| 99 | +app.geometry("800x600") |
| 100 | +app.config(bg="light blue") |
| 101 | +customFont = font.Font(family="Helvetica", size=12) |
| 102 | + |
| 103 | +btn_load = tk.Button(app, text="Load PDFs", command=select_files, font=customFont, bg="navy", fg="white", padx=10, |
| 104 | + pady=5) |
| 105 | +btn_load.pack(pady=10) |
| 106 | + |
| 107 | +btn_fetch = tk.Button(app, text="Fetch PDF from URL", command=fetch_from_url, font=customFont, bg="navy", fg="white", |
| 108 | + padx=10, pady=5) |
| 109 | +btn_fetch.pack(pady=10) |
| 110 | + |
| 111 | +btn_save_pdf = tk.Button(app, text="Save Summary as PDF", command=save_summary_as_pdf, font=customFont, bg="navy", |
| 112 | + fg="white", padx=10, pady=5) |
| 113 | +btn_save_pdf.pack(pady=10) |
| 114 | + |
| 115 | +output_text = scrolledtext.ScrolledText(app, width=70, height=20, font=customFont, padx=10, pady=10, wrap=tk.WORD) |
| 116 | +output_text.pack(pady=20) |
| 117 | + |
| 118 | +app.mainloop() |
0 commit comments