View Full Version : Epub exactly as PDF

04-02-2013, 12:26 PM
Is it possible in any way at all to do that?
I have been trying to replicate this pdf with no success:
Thanks in advace guys!

04-02-2013, 12:53 PM
That seems likely to be an unauthorised copy of a copyright work. Link deleted.
Do not post such things to MobileRead.

04-02-2013, 01:16 PM
First, you need to figure out how to extract the text, with all the bolds and italics (don't worry about the chapter titles, footnotes, etc). Copy-pasting won't work. You may have to run the PDF through ABBYY FineReader and OCR it. Then you need to recreate the layout, either in Adobe InDesign, Microsoft Word, or whatever.

Adobe InDesign will export it as ePub. Recreating it in Microsoft Word will also work, but then you should export as Filtered HTML and import that into Sigil. There's also the fact that FineReader itself can export as ePub (FineReader 11). Don't forget to proofread it at least once. Tracking down the fonts or matching them with something that looks similar is optional, but adds a nice touch. You can find the name of the fonts if you open up the PDF in Adobe Reader and go to File - Properties - Fonts.

04-02-2013, 01:44 PM
The idea behind PDF is to be static. Think of digital sheets of paper. Code WYSIWYG as device independent as possible. The idea behind epub is to be a data container to be processed by the reading system. Code WYSIWYM with regard for limits of the reading system and individual settings. Yes there are limited functionality addons for the contrary concepts within both specs (reflow in PDF and "fixed" property in epub) but it isn't the purpose of a PDF to be dynamic or of an epub to be fixed.

04-02-2013, 03:45 PM
Oh, and don't bother about the page numbers; they're calculated on-the-fly, according to the resolution of the device, the font selection and the font size. So yeah, it won't be "exactly" like the PDF. Same goes for the footnotes! The third revision of ePub (ePub 3.0) promises "real" footnotes (at the bottom of the page, through HTML5), except that the vast majority of e-readers out there cannot read this spec, so you're stuck with the second revision (ePub 2.01), meaning that you will need to convert the footnotes as endnotes.

Some devices may have issues with tables, so it's recommended that you save those as images - especially if they have vertical text or anything fancy like that.

04-05-2013, 03:18 AM
Thanks a lot for the replys!
I'm not sure if the content was copywrited or not (I got legal premission to use it).
I'll do my best to replicate the PDF in InDesign.
Do any of you have any experience with exporting as ePub 3.0 with layout (some adobe experimental format)?
Will I have better results with it if I open my ePubs in the correct ePub reader?

04-05-2013, 07:05 AM
Many devices read PDFs directly, so you are better off just leaving it in that format if the devices allow and it isn't unreadable because of multiple columns, etc.

There are few devices that will display epub 3.0, so identify your target device and make sure of its limitations before you spend a lot of time making a beautiful compliant epub that the device can not fully use.

Right now, this is the bleeding edge, a fascinating, but sometimes not so pleasant, place to be.

04-18-2013, 04:32 PM
Thanks a lot for the replys!
I'll do my best to replicate the PDF in InDesign.

First, ask yourself: if you want it to look exactly like a PDF, why can't you leave it in PDF format? Do you really need it to be EPUB (3?) ?

(BTW, InDesign is a poor tool to create eBooks.)

04-19-2013, 02:12 AM
Hmmm... InDesign is the publishing industry standard. Sure, there are other tools, perhaps with a shorter learning curve, but I wouldn't exactly call it a "poor" tool.

For someone who wants the ePub to look "exactly" like the PDF, I suggest that they start in InDesign with a PDF output in mind, export as PDF, then save an extra copy and "dumb" it down for ePub export.

04-19-2013, 03:29 AM
"Poor" w.r.t. the production of eBooks, for which better tools are out there. The fact that it is the industry standard, given it was conceived for authoring printed material, does not make it automatically good for eBooks.

04-19-2013, 06:55 AM
I need to make it the same as .pdf and make it an ePub is because these books will be used as school learning material and we want to enable students to bookmark their text, highlight the important stuff and make notes.

If not for InDesign, which tools would you recommend for ePubs?

04-19-2013, 07:25 AM
You certainly don't need PDF to bookmark or highlight...any "good" reading device or app will allow that...I recommend Marvin - it has great bookmarking/highlighting and importing/exporting of notes! (And it's free!) it is currently available for ipad...iphone will be out in 2-3months and android shortly after that. If that is the only reason you want a PDF style, then you can rest easy in a normal, reflowable, ePub.

For ease of use and making a normal/reflowable ePub quickly/well I would use Sigil. You really only need some basic HTML/CSS skills to use it. I found it much easier to use than indesign and the learning curve for HTML/CSS is much easier than indesign's.


04-19-2013, 12:41 PM
make it an ePub is because these books will be used as school learning material and we want to enable students to bookmark their text, highlight the important stuff and make notes.

FYI, you can do those things with PDFs too (even with better support than most EPUB Reading Systems...)

04-19-2013, 12:43 PM
If not for InDesign, which tools would you recommend for ePubs?

It depends on your workflow. If you want to stay cheap, Sigil is ok. (Any XHTML editori will do the job --- I use pure Vim.) Otherwise, oXygen or BlueGriffon.

02-08-2014, 11:57 AM
Sigil does not support PDFs directly. They must be converted to HTML first. Much of the WYSIWYG formatting is lost in that conversion. Don't get me wrong, I love Sigil, but it is not the ultimate solution for this type of conversion. This is the voice of experience speaking. I have struggled with this problem for quite some time.

02-08-2014, 01:29 PM
There is no good solution for PDF conversion. Get back to the basics and recreate it as an ePUB.

02-08-2014, 04:33 PM
There is no good solution for PDF conversion. Get back to the basics and recreate it as an ePUB.


Sorry, but this is square peg--round holing. It just is. There's no magic bullet for making PDF's into ePUBs. You can do it fairly simply if you just have fiction PDF's, but even then...there's a difference between the image layer and the content layer. Want to see it in action? Take ANY PDF that is searchable, and find, say, 5-10 sentences that have punctuation above the normal. Question marks, semi-colons, and the like. Also make sure you get some words with f's in them. Then copy lines from the PDF. Open Word, and paste the text NOT INTO the document, but into the Search box, and see what you get. THAT is what is really being put into the Word document, and the underlying HTML: not the text one sees with their eyes.

I just had some guy on Google+ tell some other poor schlub with "hundreds" of PDF's to convert to put the PDF's into GoogleDocs, and then use the resulting text to make ePUBs. At that point, I just gave up, because of course, the schlub came back raving about how it was so easy, and he'd made two books in a day! (I am assuming he hasn't proofed them yet). {shrug}. Image layer, content layer. Not the same. I presume that someday, Adobe will come out with an ACTUAL way to export PDF into something usable, but I'm not holding my breath. (not to mention: lots of PDF's in the world NOT made with Adobe, and those are even worse. Ever try to 'convert' a Quark-created PDF? Oh, frabjous day.)


02-09-2014, 03:52 PM
Yeah, the problem is indeed the way a given PDF is output.

In general the problem is so hard, that starting from scratch (sometimes even the plain text cannot be copied properly!) is the best way to go.

(Not to mention the embarassing number of "Conversion services! For cheap! Any language, Groundhoggese included! PDF to EPUB3!" which create EPUBs which are a sequence of images inside an EPUB3 container (!) or they use stuff like pdf2htmlEX to get a "sort of EPUB3 FXL, but you need Javascript to actually see it"...)