It's definitely far easier to emit a controlled, useful subset of PDF than it is to parse PDF documents. I wrote a small PDF library for the Decker ecosystem th

It's definitely far easier to emit a controlled, useful subset of PDF than it is to parse PDF documents. I wrote a small PDF library for the Decker ecosystem that just focuses on bitmaps and page layout; roughly 4kb and 135 LoC. docs/demos: https://beyondloom.com/decker/pdf.html https://beyondloom.com/decker/pdf.html browsable source: https://github.com/JohnEarnest/Decker/blob/main/examples/decks/pdf.deck#L142-L276 https://github.com/JohnEarnest/Decker/blob/main/examples/dec...