Large book with a 2-column structure, PDF -> EPUB .. any tips on this one?

[cow]

Hello

I'm looking for a tips converting this PDF to EPUB (if it is even realistically possible). My Kobo Sage can handle PDFs just fine, but they are much slower than EPUB

https://theholychristianchurch.org/d...iffe-bible.pdf

The file is very clean and well-structured, but I'm having difficulties getting a nice epub conversion out of it.

The entire document is comprised of two columns, which the epub converter doesn't seem to like. Any tips on converting this?

Also, is it possible to only convert, for example, the first 50 pages so I can get an idea of how how the conversion is going without waiting for the entire 1100-page book to process? I've looked through all the settings over and over and I don't see any option to do this. It takes quite a long time

Any tips appreciated. Thank you

[cow]

I can't seem to edit my post, I wanted to update before anyone here spent time replying.

I may have found a way to pull this off. I found a utility on Linux called "pdf2htmlex" and got a brilliant, near-perfect HTML conversion out of it. Maybe this can help someone else in the future.

I am now converting the HTML to EPUB in Calibre to see if that works better (I assume it should). I will post again with screenshots again once it finishes and report back - taking a very long time.

This is how the HTML conversion came out with pdf2htmlex:

[retiredbiker]

In general, pdfs are the worst files to convert. See the sticky post for a lot of details: Read this before Posting PDF Questions

This particular file does seem to have excellent text. You can select and paste it, and it seems to be complete and in order. Unlike so many pdfs. So in theory it might convert well.

But the two columns are the killer. Even if the text comes out of the two columns un-garbled in a conversion, epub does not have any easy two column display. And it's pretty obvious you want this book to preserve the columns as they are, not only as columns but with the entries aligned as in a table.

So to make this into an epub you would have to put the text into tables. That would be a manual job, a huge one, (unless someone knows of a tool to automate it). And tables have real troubles on an e-reader device if a user changes text size, for example, or if the table entries don't fit the reader screen size.

A book like this is best left as a pdf, IMHO.

[BetterRed]

@cow - you need make some more posts before you can edit your own posts - play some games in the lounge. Your second post went into the moderation queue. Re the image size see ==>> Guideline #9

BR

[cow]

Originally Posted by BetterRed

@cow - you need make some more posts before you can edit your own posts - play some games in the lounge. Your second post went into the moderation queue. Re the image size see ==>> Guideline #9

BR

Understood, thanks!

Originally Posted by retiredbiker

In general, pdfs are the worst files to convert. See the sticky post for a lot of details: Read this before Posting PDF Questions

This particular file does seem to have excellent text. You can select and paste it, and it seems to be complete and in order. Unlike so many pdfs. So in theory it might convert well.

But the two columns are the killer. Even if the text comes out of the two columns un-garbled in a conversion, epub does not have any easy two column display. And it's pretty obvious you want this book to preserve the columns as they are, not only as columns but with the entries aligned as in a table.

So to make this into an epub you would have to put the text into tables. That would be a manual job, a huge one, (unless someone knows of a tool to automate it). And tables have real troubles on an e-reader device if a user changes text size, for example, or if the table entries don't fit the reader screen size.

A book like this is best left as a pdf, IMHO.

Yeah, the only reason I wanted to keep the columns is because of the side-by-side comparison format that the book is built upon. My PDF conversion did indeed fail spectacularly. Once I got the epub added to my reader, I realized why the side by side format wouldn't work.. still makes for tiny text and requires zooming, same as the PDF. I guess I was expecting the impossible and didn't think enough about it.

I suppose one possibility would be to break these columns out into alternating rows and use white background color for Wycliffe and light gray background color for KJV.. or something to that nature. If I was able to get a coherent conversion from PDF to EPUB, I could probably do this work in Sigil with regular expressions and other tools.

But the problem is first getting that initial conversion so I can start editing it manually.

It turns out that the converter I used put EVERY SINGLE LETTER IN THE BOOK into individual <span>S</span> objects, each with their own pixel-perfect absolute positioning. So there wasn't much I could do with it.

Maybe I'd be better off just buying a hard copy