nextgenpdf/pdf_renamer.py

import os
import tempfile
import zipfile
from flask import Flask, request, render_template_string, send_from_directory
from werkzeug.utils import secure_filename
from PyPDF2 import PdfReader

app = Flask(__name__)
app.config['UPLOAD_FOLDER'] = tempfile.mkdtemp()
app.config['MAX_CONTENT_LENGTH'] = 100 * 1024 * 1024  # 100MB limit

ALLOWED_EXTENSIONS = {'pdf', 'zip'}

def allowed_file(filename):
    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS

def extract_names_from_pdf(pdf_path):
    try:
        reader = PdfReader(pdf_path)
        fields = reader.get_fields()
        last_name = fields.get('Last Name', {}).get('/V', '')
        first_name = fields.get('First Name', {}).get('/V', '')
        if last_name and first_name:
            return f"{last_name}_{first_name}.pdf"
        else:
            return None
    except Exception as e:
        return None

@app.route('/', methods=['GET', 'POST'])
def upload_files():
    if request.method == 'POST':
        files = request.files.getlist('files')
        processed_files = []
        temp_extract_dirs = []
        for file in files:
            if file and allowed_file(file.filename):
                filename = secure_filename(file.filename)
                filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
                file.save(filepath)
                if filename.lower().endswith('.zip'):
                    # Extract ZIP
                    extract_dir = os.path.join(app.config['UPLOAD_FOLDER'], 'extract_' + os.path.splitext(filename)[0])
                    os.makedirs(extract_dir, exist_ok=True)
                    temp_extract_dirs.append(extract_dir)
                    with zipfile.ZipFile(filepath, 'r') as zip_ref:
                        zip_ref.extractall(extract_dir)
                    # Process PDFs in extract_dir
                    for root, dirs, files_in_zip in os.walk(extract_dir):
                        for f in files_in_zip:
                            if f.lower().endswith('.pdf'):
                                pdf_path = os.path.join(root, f)
                                new_name = extract_names_from_pdf(pdf_path)
                                if new_name:
                                    new_path = os.path.join(root, new_name)
                                    counter = 1
                                    while os.path.exists(new_path):
                                        base, ext = os.path.splitext(new_name)
                                        new_path = os.path.join(root, f"{base}_{counter}{ext}")
                                        counter += 1
                                    os.rename(pdf_path, new_path)
                                    processed_files.append(('success', os.path.relpath(new_path, extract_dir), new_path))
                                else:
                                    processed_files.append(('skipped', f, None))
                else:
                    # Single PDF
                    new_name = extract_names_from_pdf(filepath)
                    if new_name:
                        new_filepath = os.path.join(app.config['UPLOAD_FOLDER'], new_name)
                        counter = 1
                        while os.path.exists(new_filepath):
                            base, ext = os.path.splitext(new_name)
                            new_filepath = os.path.join(app.config['UPLOAD_FOLDER'], f"{base}_{counter}{ext}")
                            counter += 1
                        os.rename(filepath, new_filepath)
                        processed_files.append(('success', os.path.basename(new_filepath), new_filepath))
                    else:
                        processed_files.append(('skipped', filename, None))
        
        # Create ZIP of all successful renames
        successful_files = [p for p in processed_files if p[0] == 'success']
        if successful_files:
            zip_output = os.path.join(app.config['UPLOAD_FOLDER'], 'renamed_files.zip')
            with zipfile.ZipFile(zip_output, 'w') as zip_out:
                for _, rel_name, full_path in successful_files:
                    zip_out.write(full_path, rel_name)
            zip_name = 'renamed_files.zip'
        else:
            zip_name = None
        
        skipped = [p[1] for p in processed_files if p[0] == 'skipped']
        
        return render_template_string('''
        <!DOCTYPE html>
        <html>
        <head>
            <title>Results</title>
            <style>
                body { font-family: Arial, sans-serif; background-color: #f4f4f4; margin: 0; padding: 20px; text-align: center; }
                img { margin-bottom: 20px; }
                h1 { color: #333; }
                p { font-size: 18px; }
                a { color: #007bff; text-decoration: none; padding: 10px 20px; background-color: #007bff; color: white; border-radius: 5px; }
                a:hover { background-color: #0056b3; }
                ul { list-style-type: none; padding: 0; }
                li { background-color: #fff; margin: 5px 0; padding: 10px; border-radius: 5px; box-shadow: 0 0 5px rgba(0,0,0,0.1); }
            </style>
        </head>
        <body>
            <img src="/static/trbc-logo-full.png" alt="Logo" width="200">
            <h1>Processing Results</h1>
            {% if zip_name %}
            <p><a href="/download/{{ zip_name }}">Download Renamed PDFs (ZIP)</a></p>
            {% endif %}
            <h2>Skipped Files:</h2>
            <ul>
            {% for file in skipped %}
                <li>{{ file }}</li>
            {% endfor %}
            </ul>
            <a href="/">Upload more files</a>
        </body>
        </html>
        ''', zip_name=zip_name, skipped=skipped)
    return render_template_string('''
    <!DOCTYPE html>
    <html>
    <head>
        <title>PDF Renamer</title>
        <style>
            body { font-family: Arial, sans-serif; background-color: #f4f4f4; margin: 0; padding: 20px; text-align: center; }
            img { margin-bottom: 20px; }
            h1 { color: #333; }
            form { background-color: #fff; padding: 20px; border-radius: 10px; box-shadow: 0 0 10px rgba(0,0,0,0.1); display: inline-block; }
            input[type="file"] { margin: 10px 0; }
            input[type="submit"] { background-color: #28a745; color: white; border: none; padding: 10px 20px; border-radius: 5px; cursor: pointer; }
            input[type="submit"]:hover { background-color: #218838; }
        </style>
    </head>
    <body>
        <img src="/static/trbc-logo-full.png" alt="Logo" width="200">
        <h1>Upload PDF Files or ZIP to Rename</h1>
        <form method="post" enctype="multipart/form-data">
            <input type="file" name="files" multiple accept=".pdf,.zip">
            <br><br>
            <input type="submit" value="Process PDFs">
        </form>
    </body>
    </html>
    ''')

@app.route('/download/<filename>')
def download_file(filename):
    return send_from_directory(app.config['UPLOAD_FOLDER'], filename, as_attachment=True)

if __name__ == '__main__':
    app.run(debug=True, port=3000)