frontend-slides

Use python-pptx with python3 to extract text, images, and speaker notes from the .ppt or .pptx file while preserving slide order. After extraction, run the same style-selection workflow as a new presentation to rebuild the deck as a self-contained HTML file.

Build a single HTML file with inline CSS and JS containing a presentation controller class for keyboard, wheel, and touch navigation plus Intersection Observer for reveal animations. No external frameworks are needed for a fully functional deck.

Set every slide to height: 100vh and 100dvh with overflow: hidden, and scale all typography and spacing with clamp(). When content overflows, split it into multiple slides rather than shrinking text below readable sizes.

New presentations from notes or drafts require no dependencies at all. For PowerPoint conversion, if python-pptx is unavailable, the workflow asks whether to install it or fall back to a manual export-based approach.

Browsers silently ignore negated CSS functions like -clamp() or -min(). Wrap them in calc() instead, for example calc(-1 * clamp(28px, 3.5vw, 44px)), so the negation is computed correctly.

Single-file decks avoid external assets, so extracted or user-supplied images require a separate assets folder. Very dense content must be split across slides since internal scrolling and tiny text are not permitted.