Save PDF from Revealjs HTML by python script

Save-PDF.md

Save PDF from Revealjs HTML by Pure Python

This is demo for new feature of sphinx-revealjs

Usage

uv run save-pdf.py URL OUTPUT

This is supporing PEP723 (Inline script metadata). It may work without venv.

Refs

• attakei/sphinx-revealjs#179

save-pdf.py

#!/usr/bin/env python
# /// script
# dependencies = [
#     "click>=8.1.7",
#     "playwright>=1.47.0",
#     "pypdf>=5.0.1",
# ]
# ///
import io
import subprocess
from pathlib import Path

import click
from playwright.sync_api import Page, sync_playwright
from pypdf import PdfReader, PdfWriter


def collect_slides(page: Page, url: str) -> list[bytes]:
    slides = []
    page.emulate_media(media="screen")
    page.goto(url)
    # NOTE: Optional - Hide component not need for pdf
    page.evaluate("Reveal.configure({progress: false});")
    size = page.viewport_size
    if size is None:
        raise Exception("Viewport is None")
    while True:
        content = page.pdf(
            width=str(size["width"]),
            height=str(size["height"]),
            print_background=True,
        )
        if slides and slides[-1] == content:
            break
        slides.append(content)
        page.evaluate("Reveal.next();")
    return slides


@click.command()
@click.argument("url", type=str)
@click.argument("out", type=click.Path(path_type=Path))
def main(url: str, out: Path):
    with sync_playwright() as p:
        browser = p.chromium.launch()
        page = browser.new_page()
        slides = collect_slides(page, url)
        browser.close()

    writer = PdfWriter()
    for slide in slides:
        reader = PdfReader(io.BytesIO(slide))
        writer.add_page(reader.pages[0])
    writer.write(out)
    writer.close()


if __name__ == "__main__":
    subprocess.run("playwright install chromium".split())
    main()