feat: add batch PDF-to-PNG conversion endpoint — current implementation only converts the first page, forcing users to make N separate requests for an N-page PDF

[divyanshim27]

🚀 Problem Statement

The README documents: "Convert PDF (first page) to PNG." The POST /convertPng endpoint in backend/blueprints/pdf.py uses PyMuPDF (fitz) to render only page = doc[0] — the first page — and returns a single PNG.

A user with a 10-page PDF who needs all pages as PNGs must submit 10 separate requests. This is:

• Slow (10 round-trips vs 1)
• Error-prone (no atomicity — page 7 might fail while 1-6 succeeded)
• Not how file manipulation tools work (e.g., ImageMagick, LibreOffice batch-convert all pages)

Proposed Fix

Add a POST /convertPngBatch endpoint (respecting Project Rule 1: no disk storage) that returns a ZIP archive of all pages as PNGs, generated entirely in memory:

# backend/blueprints/pdf.py — new batch endpoint

import io
import zipfile
import fitz  # PyMuPDF
from flask import Blueprint, request, send_file

@pdf_bp.route('/convertPngBatch', methods=['POST'])
def convert_pdf_all_pages_to_png():
    """
    Convert all pages of an uploaded PDF to PNG images.
    Returns a ZIP archive (in memory, no disk storage) containing
    one PNG per page, named page_001.png, page_002.png, etc.
    """
    if 'file' not in request.files:
        return {'error': 'No file provided'}, 400

    file = request.files['file']
    if not file.filename.lower().endswith('.pdf'):
        return {'error': 'File must be a PDF'}, 415

    pdf_bytes = file.read()
    doc = fitz.open(stream=pdf_bytes, filetype='pdf')

    zip_buffer = io.BytesIO()
    with zipfile.ZipFile(zip_buffer, 'w', zipfile.ZIP_DEFLATED) as zf:
        for page_num in range(len(doc)):
            page = doc[page_num]
            mat = fitz.Matrix(2.0, 2.0)  # 2x scale for quality
            pix = page.get_pixmap(matrix=mat)
            png_bytes = pix.tobytes('png')
            zf.writestr(f'page_{page_num + 1:03d}.png', png_bytes)

    zip_buffer.seek(0)
    return send_file(
        zip_buffer,
        mimetype='application/zip',
        as_attachment=True,
        download_name='converted_pages.zip'
    )

This strictly follows Project Rules 1 and 2: all processing is in memory, no external APIs used.

Files to Modify

File	Change
backend/blueprints/pdf.py	Add POST /convertPngBatch endpoint
frontend/src/pages/PdfPng.jsx	Add UI toggle for "Convert All Pages" mode
Readme.md	Document the new /convertPngBatch endpoint

Suggested labels: enhancement, backend, frontend

I would like to work on this. Could you please assign it to me?

[divyanshim27]

Hey @Durgeshwar-AI ,
Please assign it to me under GSSoC 2026.
/claim /assign /claim-assignment

[Durgeshwar-AI]

@divyanshim27 I want to bring the conversion to frontend if it is still in backend.