Base64 to PowerPoint (PPTX)

How to decode Base64 to a PowerPoint file

1. Paste the Base64 into Base64 or data URI. A data:application/vnd.openxmlformats-officedocument.presentationml.presentation;base64, prefix is fine, and Open .txt file handles long strings saved as text.
2. Slide text lists every slide under its number (Slide 1, Slide 2 and so on), with the text of each paragraph on its own line. A slide without text shows ….
3. The entry list underneath shows all parts of the file with View and Save.
4. Press Download file to save the .pptx.

The text is extracted from the slide XML with the browser’s XML parser. Nothing in the presentation is executed, and the page sends no network requests besides Cloudflare’s page-view count.

What the slide text includes

Each slide is a separate XML file inside the .pptx, and the preview takes the text of every paragraph in it: titles, body placeholders, text boxes, shapes, grouped shapes and table cells. Slides marked as hidden are included. Up to 200 slides are read.

What is stored elsewhere does not appear:

Slides are listed by the number in their file name, slide1.xml, slide2.xml and so on. That is normally the order of the deck, but a presentation rearranged by a script can keep old file numbers, so treat the numbering as a guide.

Checking that it is a PPTX

Every Office Open XML file is a ZIP archive, so a PowerPoint deck starts with UEsDB in Base64 just like a Word document or an Excel workbook. The tool looks for ppt/ folders among the entries to tell them apart. When the content is a different format, a note says so and offers the right extension:

A deck with macros (.pptm) contains ppt/vbaProject.bin. Its text previews normally, but PowerPoint opens it only with the .pptm extension, so type that before downloading.

Read the slide text in Python

The standard library is enough to list the text per slide, in the same order the preview uses:

import base64, io, re, zipfile
import xml.etree.ElementTree as ET

A = "{http://schemas.openxmlformats.org/drawingml/2006/main}"

with open("deck.b64") as f:
    pptx = base64.b64decode(f.read())

with zipfile.ZipFile(io.BytesIO(pptx)) as z:
    slides = [n for n in z.namelist() if re.fullmatch(r"ppt/slides/slide\d+\.xml", n)]
    slides.sort(key=lambda n: int(re.search(r"\d+", n).group()))
    for name in slides:
        root = ET.fromstring(z.read(name))
        lines = ["".join(t.text or "" for t in p.iter(A + "t")) for p in root.iter(A + "p")]
        print(name, [s for s in lines if s.strip()])

For the sample deck on this page it prints ppt/slides/slide1.xml ['Hello from Base64', 'A two-slide deck built in your browser.'] and the same for slide 2. The real slide order is in ppt/presentation.xml, in the <p:sldIdLst> element, if you need it exactly. More on Base64 in Python; the archive side of things is covered on Base64 to ZIP.

FAQ