Google Colab වලින් PDF-to-Image Converter එකක් හදමු

# 1. Setup & Install
!apt-get install -y poppler-utils > /dev/null 2>&1
!pip install pdf2image > /dev/null 2>&1

import ipywidgets as widgets
from IPython.display import display, clear_output
from pdf2image import convert_from_path, pdfinfo_from_path
import os
import shutil
from google.colab import files

# 2. GUI Widgets
upload_widget = widgets.FileUpload(accept='.pdf', multiple=False, description='Upload PDF')
dpi_widget = widgets.IntSlider(value=300, min=72, max=600, step=10, description='DPI:')
format_widget = widgets.Dropdown(options=['JPEG', 'PNG'], value='JPEG', description='Format:')

# Text input for flexible page selection
pages_widget = widgets.Text(
    placeholder='e.g., 1, 3, 5 or 1-5 (Leave blank for all)',
    description='Pages:',
    layout=widgets.Layout(width='350px')
)

convert_button = widgets.Button(description="Convert & Download", button_style='success', icon='download')
output_log = widgets.Output()

# Global variable to store total pages for validation
total_pdf_pages = 0

def update_pdf_info(change):
    global total_pdf_pages
    if upload_widget.value:
        val = upload_widget.value
        file_name = list(val.keys())[0] if isinstance(val, dict) else val[0]['name']
        with open(file_name, "wb") as f:
            f.write(val[file_name]['content'] if isinstance(val, dict) else val[0]['content'])
        
        info = pdfinfo_from_path(file_name)
        total_pdf_pages = info["Pages"]
        print(f"✅ '{file_name}' loaded. Total pages: {total_pdf_pages}")

upload_widget.observe(update_pdf_info, names='value')

# Helper function to parse strings like "1,3,5" and "1-5"
def parse_page_selection(input_str, max_pages):
    if not input_str.strip():
        return list(range(1, max_pages + 1)) # Default to all pages
    
    pages = set()
    # Split by commas first
    parts = input_str.split(',')
    for part in parts:
        part = part.strip()
        if '-' in part:
            # Handle ranges like 1-5
            try:
                start, end = map(int, part.split('-'))
                # Ensure the range is valid and within PDF boundaries
                start = max(1, start)
                end = min(max_pages, end)
                if start <= end:
                    pages.update(range(start, end + 1))
            except ValueError:
                raise ValueError(f"Invalid range format: '{part}'")
        else:
            # Handle single digits
            try:
                p = int(part)
                if 1 <= p <= max_pages:
                    pages.add(p)
            except ValueError:
                raise ValueError(f"Invalid page number: '{part}'")
                
    return sorted(list(pages))

# 3. Main Conversion Logic
def process_pdf(b):
    with output_log:
        clear_output()
        if not upload_widget.value:
            print("⚠️ Please upload a PDF first.")
            return

        # Define unique folders/files
        output_folder = "final_output"
        zip_name = "converted_pdf_pages"
        
        # Parse the custom page string safely
        try:
            selected_pages = parse_page_selection(pages_widget.value, total_pdf_pages)
            if not selected_pages:
                print("⚠️ No valid pages selected within the document's page range.")
                return
        except ValueError as e:
            print(f"❌ Input Error: {e}")
            return

        # Hard Cleanup: Wipe out old folders/ZIPs to stop browser from getting confused
        if os.path.exists(output_folder): shutil.rmtree(output_folder)
        if os.path.exists(f"{zip_name}.zip"): os.remove(f"{zip_name}.zip")
        os.makedirs(output_folder, exist_ok=True)

        val = upload_widget.value
        file_name = list(val.keys())[0] if isinstance(val, dict) else val[0]['name']
        
        print(f"🔄 Converting pages: {selected_pages}...")
        for p in selected_pages:
            images = convert_from_path(file_name, dpi=dpi_widget.value, first_page=p, last_page=p)
            if images:
                images[0].save(f"{output_folder}/page_{p}.{format_widget.value.lower()}", format_widget.value)

        # Create exactly ONE Archive
        shutil.make_archive(zip_name, 'zip', output_folder)
        print("🎉 Conversion successful! Bundling complete. Downloading...")
        files.download(f"{zip_name}.zip")

convert_button.on_click(process_pdf)

# 4. Layout Display
display(widgets.VBox([
    widgets.HTML("<h3>PDF to Image Converter</h3>"),
    upload_widget,
    dpi_widget,
    format_widget,
    pages_widget,
    convert_button,
    output_log
]))

# 1. Setup & Install
!apt-get install -y poppler-utils > /dev/null 2>&1
!pip install pdf2image > /dev/null 2>&1

import ipywidgets as widgets
from IPython.display import display, clear_output
from pdf2image import convert_from_path, pdfinfo_from_path
import os
import shutil
from google.colab import files

# 2. GUI Widgets
upload_widget = widgets.FileUpload(accept='.pdf', multiple=False, description='Upload PDF')
dpi_widget = widgets.IntSlider(value=300, min=72, max=600, step=10, description='DPI:')
format_widget = widgets.Dropdown(options=['JPEG', 'PNG'], value='JPEG', description='Format:')

# Text input for flexible page selection
pages_widget = widgets.Text(
    placeholder='e.g., 1, 3, 5 or 1-5 (Leave blank for all)',
    description='Pages:',
    layout=widgets.Layout(width='350px')
)

convert_button = widgets.Button(description="Convert & Download", button_style='success', icon='download')
output_log = widgets.Output()

# Global variable to store total pages for validation
total_pdf_pages = 0

def update_pdf_info(change):
    global total_pdf_pages
    if upload_widget.value:
        val = upload_widget.value
        file_name = list(val.keys())[0] if isinstance(val, dict) else val[0]['name']
        with open(file_name, "wb") as f:
            f.write(val[file_name]['content'] if isinstance(val, dict) else val[0]['content'])
        
        info = pdfinfo_from_path(file_name)
        total_pdf_pages = info["Pages"]
        print(f"✅ '{file_name}' loaded. Total pages: {total_pdf_pages}")

upload_widget.observe(update_pdf_info, names='value')

# Helper function to parse strings like "1,3,5" and "1-5"
def parse_page_selection(input_str, max_pages):
    if not input_str.strip():
        return list(range(1, max_pages + 1)) # Default to all pages
    
    pages = set()
    # Split by commas first
    parts = input_str.split(',')
    for part in parts:
        part = part.strip()
        if '-' in part:
            # Handle ranges like 1-5
            try:
                start, end = map(int, part.split('-'))
                # Ensure the range is valid and within PDF boundaries
                start = max(1, start)
                end = min(max_pages, end)
                if start <= end:
                    pages.update(range(start, end + 1))
            except ValueError:
                raise ValueError(f"Invalid range format: '{part}'")
        else:
            # Handle single digits
            try:
                p = int(part)
                if 1 <= p <= max_pages:
                    pages.add(p)
            except ValueError:
                raise ValueError(f"Invalid page number: '{part}'")
                
    return sorted(list(pages))

# 3. Main Conversion Logic
def process_pdf(b):
    with output_log:
        clear_output()
        if not upload_widget.value:
            print("⚠️ Please upload a PDF first.")
            return

        # Define unique folders/files
        output_folder = "final_output"
        zip_name = "converted_pdf_pages"
        
        # Parse the custom page string safely
        try:
            selected_pages = parse_page_selection(pages_widget.value, total_pdf_pages)
            if not selected_pages:
                print("⚠️ No valid pages selected within the document's page range.")
                return
        except ValueError as e:
            print(f"❌ Input Error: {e}")
            return

        # Hard Cleanup: Wipe out old folders/ZIPs to stop browser from getting confused
        if os.path.exists(output_folder): shutil.rmtree(output_folder)
        if os.path.exists(f"{zip_name}.zip"): os.remove(f"{zip_name}.zip")
        os.makedirs(output_folder, exist_ok=True)

        val = upload_widget.value
        file_name = list(val.keys())[0] if isinstance(val, dict) else val[0]['name']
        
        print(f"🔄 Converting pages: {selected_pages}...")
        for p in selected_pages:
            images = convert_from_path(file_name, dpi=dpi_widget.value, first_page=p, last_page=p)
            if images:
                images[0].save(f"{output_folder}/page_{p}.{format_widget.value.lower()}", format_widget.value)

        # Create exactly ONE Archive
        shutil.make_archive(zip_name, 'zip', output_folder)
        print("🎉 Conversion successful! Bundling complete. Downloading...")
        files.download(f"{zip_name}.zip")

convert_button.on_click(process_pdf)

# 4. Layout Display
display(widgets.VBox([
    widgets.HTML("<h3>PDF to Image Converter</h3>"),
    upload_widget,
    dpi_widget,
    format_widget,
    pages_widget,
    convert_button,
    output_log
]))