PRS-T1 Exporting notes: Convert EPUB with annotations to PDF?

[martz]

Hi,
There are several excellent tools for managing annotations on PDF documents. But has someone figured out a way for converting an EPUB file with annotations and highlights to PDF?

In other words: I read my EPUB with PRS-T1, do some scribbling and highlighting on the pages and then want to save the scribbled and highlighted EPUB as PDF.

Could this be accomplished using Calibre? (I'm on Linux, so a simple command line script would be great.)

Thanks for any help!

[Schrollini]

Originally Posted by martz

Hi,
There are several excellent tools for managing annotations on PDF documents. But has someone figured out a way for converting an EPUB file with annotations and highlights to PDF?

I'm afraid part of this request is going to be nearly impossible. EPUB is, like HTML, just a string of text that the viewer is responsible for laying out. (PDF, on the other hand, fully describes the layout of each page.) So to be able to match up the handwritten annotations with the text, you'd have to write a EPUB layout engine that works exactly like the one in your Reader. Don't count on this happening. (Highlighted text doesn't have this problem, but I'm unaware of anyone who's built that part.)

As a work-around, you could convert your EPUB to a PDF before loading it onto your Reader, and then use something like PRSAnnots (shameless plug!) to get the annotations you made on the PDF. You'd lose some of the flexibility of the EPUB (changing fonts and the like), but if you make good choices when creating the PDF, you might not care.

As for converting EPUB to PDF: Apparently Calibre has this ability, but I haven't tested it out. Since you're on Linux, you probably have a print-to-PDF functionality, so any ebook software that can print could be used to make a PDF as well.

Good luck.

[idle]

I'm using PRSAnnots (thanks!) and I like it for PDF files, but I'd too appreciate something similar for EPUBs (even if it was just for highlights and typed annotations; I understand why handwritten ones would be a problem). The difference is not just flexibility in font sizes and such, but also the reader performance. Some time ago I annotated a novel-sized text (and a fat novel at that), converted the file to PDF for that purpose, but it grew awfully slow with each new annotation so that in the end I had to split it in two and still had to wait several seconds for page turns. Luckily I don't need to do this too often.

[martz]

Originally Posted by Schrollini

I'm afraid part of this request is going to be nearly impossible. EPUB is, like HTML, just a string of text that the viewer is responsible for laying out. (PDF, on the other hand, fully describes the layout of each page.) So to be able to match up the handwritten annotations with the text, you'd have to write a EPUB layout engine that works exactly like the one in your Reader. Don't count on this happening. (Highlighted text doesn't have this problem, but I'm unaware of anyone who's built that part.)

Thanks very much, this is a very good explanation. I thought something similar myself but I couldn't figure out the details.

So handwritten notes are causing trouble here. But what about converting an EPUB that only has text highlights (no scribbles)? Is this likely to cause similar problems?

(It would be great to be able to assign a PDF printer for this: it would take an EPUB from my reader and convert it, while also exporting my underlines.)

BTW, I absolutely love PRSAnnots, what a great tool! But I still prefer EPUBs to PDFs, mostly for reasons similar to what user idle lined out: PDFs are bigger and slower. Often too slow for comfortably reading longer texts.

Thanks again,
M.

[Schrollini]

Thanks, martz and idle, for the kind words about PRSAnnots. (I didn't mean to be fishing for complements, but I'll take 'em!)

Originally Posted by martz

So handwritten notes are causing trouble here. But what about converting an EPUB that only has text highlights (no scribbles)? Is this likely to cause similar problems?

No. Highlighted text should be possible, and probably easier than with PDFs, since you don't have to deal with the rather arcane PDF format.

Is there a reason you want PDF output? This requires both a EPUB layout engine and a PDF renderer, which would make this program more complicated. What about making an annotated EPUB or HTML file?

I'll think a bit about whether this could fit into PRSAnnots. But don't expect much progress in the next month or so -- I'll be busy with real life. However, if someone else wants to take a stab at this, either for inclusion with PRSAnnots or as a separate program, don't hesitate to get in touch with me with questions.

[idle]

Originally Posted by Schrollini

Is there a reason you want PDF output? This requires both a EPUB layout engine and a PDF renderer, which would make this program more complicated. What about making an annotated EPUB or HTML file?

I don't know about martz, but both epub and html would be good solutions for me. (I'm afraid I won't be of much help, though, apart from testing.)

[martz]

Originally Posted by Schrollini

Thanks, martz and idle, for the kind words about PRSAnnots. (I didn't mean to be fishing for complements, but I'll take 'em!)

You're welcome!

Originally Posted by Schrollini

Is there a reason you want PDF output? This requires both a EPUB layout engine and a PDF renderer, which would make this program more complicated. What about making an annotated EPUB or HTML file?

PRSAnnots allows me to have a very nice archive of highlighted documents. Initially, I was just looking for something similar for EPUB files, too -- but I figured having everything as PDFs would give me a more "uniform" library. Also, I'd prefer to browse all my archive documents with one app, but haven't yet found a Linux tool that reads both EPUBs and PDFs (for Windows, there is the excellent Sumatra PDF).

That said, an annotated EPUB/HTML would actually serve these purposes very well! Yes, I'd have to use Firefox for opening the files, but as long as the underlined text is shown on the page (not extracted to a separate file), it's great.

I'd be really thankful for such a feature in PRSAnnots. I suspect it would have lots of users.

I'll think a bit about whether this could fit into PRSAnnots. But don't expect much progress in the next month or so -- I'll be busy with real life. However, if someone else wants to take a stab at this, either for inclusion with PRSAnnots or as a separate program, don't hesitate to get in touch with me with questions.

That sounds great! I'm not a coder, but I'm happy to help in testing how it works. Keep us posted!

[martz]

For reference: pdf2htmlEX seems to make a quite impressive work on converting PDFs to EPUBs without losing page formatting:
https://bbs.archlinux.org/viewtopic.php?pid=1154758

So if one could use this the other way round...