ePub to Word

[AlexBell]

The background to this question is that the usual work plan when I design an ebook is

- The publisher and author work together until they have a Word version of the book which is ready for the printer,
- I design an ePub ebook from that Word document, and then make a Kindle version by converting with calibre.

For an ebook which is nearly finished we have followed an unusual work plan because the author wanted to make extensive changes, so she and I have worked together and made the changes. This means of course that the text in the original Word document is quite different from the text and format in the ebooks, and it would be a mammoth task for the publisher to read the ebook and type out a new ready for printer Word document.

Is there any software which can read the finished ePub version and create a ready for printer Word document from the ePub version?

[DSpider]

You could try converting the ePub to HTML and then importing the HTML file into Word. Styles will probably turn into direct formatting, though. Hmmm... Wouldn't it be easier to just edit the ePub file with Sigil or something?

And if they have the edited version of the Word file, can't they just send it to you? It would mean that you'd have to start from scratch... but if editing the ePub is out of the question, at least it's a start. What kind of "extensive" changes are we talking about?

[AlexBell]

Thanks, DSpider

To answer your last question first: we're talking about multiple punctuation changes, changes of tense, adding or deleting or changing many phrases or sentences ... and probably more that I've forgotten. When I say multiple or many I mean scores if not hundreds of changes. I'm not sure if the author has kept an up to date version of the Word file.

The way I make ebooks is to get the original Word file, then mark it up to be valid XHTML files, then make the appropriate content.opf and toc.ncx files, and then package the lot up into ePub. I use an HTML editor - not Sigil, but one I used when I was dabbling in website design.

I didn't know that Word could read HTML files, but I don't use Word myself, I use LibreOffice. I'll check to see if that or OpenOffice.org can read HTML and convert the files into Word.

[Toxaris]

The method I use, is create a single file in Sigil (merge...). Extract the ePUB and load the HTML in Notepad++. Change the header by making is pure HTML and set UTF-8. Load the HTML in Word.
One issue is that the style names are gone. I would have loved that those style names would be retained in Word. At least the styles are applied most of the times. Sometimes italics are gone.

[AlexBell]

Originally Posted by Toxaris

The method I use, is create a single file in Sigil (merge...). Extract the ePUB and load the HTML in Notepad++. Change the header by making is pure HTML and set UTF-8. Load the HTML in Word.
One issue is that the style names are gone. I would have loved that those style names would be retained in Word. At least the styles are applied most of the times. Sometimes italics are gone.

Thanks, Toxaris. I'll looke at Notepad++ when the time comes. At the moment we're still changing punctuation.

[Toxaris]

You can use almost any text editor instead of Notepad++. In this case I only use it to add

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>

in the header.

[helpplease]

@Toxaris
I find Calibre formatting very ugly to read. Nice team though. I like to retain author's layout where I can. So I attempted your method, and now all muddled.

Is there still no automatic program (april 2012)?

I found this free online converter from epub (or mobi) to PDF (or word):
http://ebook.online-convert.com/convert-to-pdf
I used to ipad or msreader. as output setting. although I read on my iphone. these were best result for me after testing all options.


bad news on result:
1 images cut in half (so no intelligent page cut offs)
2 images are not "images" in output
3 hyperlinks are NOT active
4 sometimes, for ipad, the bibliography would get split e.g. 1 entry per page..

good news
1 formatting is nice (much better than Calibre)
2 its 1 step process, very quick and free

-----
I have sigil, but find it very confusing to use.
1 In sigil, can I add this "merge" function on toolbar so 1 click?
2 In sigil, can I add this "extract" function on toolbar so 1 click?

I have another texteditor (also I find confusing).
3 what are steps to "Change the header" by making is pure HTML
4 How do I "set UTF-8".


I can do rest of steps
5 Load the HTML in Word.


lastly, any improvement on "retaining style names"?

you said:
One issue is that the style names are gone. I would have loved that those style names would be retained in Word. At least the styles are applied most of the times. Sometimes italics are gone.

[Toxaris]

The situation is about the same. Actually, with my method the text is not muddled, so I wonder why it is for you... Images and everything are retained (if you extract them from the ePUB of course...

Merging in Sigil is easy. Select the text files you need in the Bookbrowser, right-click and select merge. I usually then copy the new HTML file in Notepad++ for additional work. I don't save the ePUB, because I want to keep that.
Extracting the images and stylesheet I do with a zip-program. Yes, it is still manual work. BTW, either retain the file structure or adjust your links.

Changing the header is also easy. In you big HTML file, delete the first couple of lines. The first line must be <html>.
Copy the line I mentioned above between the <head></head> tags.

The stylenames will be gone. Blame Microsoft if you want.

[helpplease]

found this FREE service online.
www.zamzar.com

better than online converter
as supports intelligent page ends
e.g. does not cut pictures in half
e.g. ends table full row on one page

bad - is the cover was cut
bad - hyperlinks still lost as with online converter
bad - images as images still lost as with online converter
bad - chapter starts, bib and index do NOT start on new page. so all messy

still FREE very fast, and sends you link. 24hr to dl.
my current reccomendation is this one.

[Toxaris]

Nah, my own method works a lot better. You can actually speed it up a bit. Convert the ePUB to HTMLZ with Calibre. Basically that is a zip file with all the HTML files merged. Extract the resulting HTMLZ and read the HTML into Word.

[laji]

Originally Posted by Toxaris

The method I use, is create a single file in Sigil (merge...). Extract the ePUB and load the HTML in Notepad++. Change the header by making is pure HTML and set UTF-8. Load the HTML in Word.
One issue is that the style names are gone. I would have loved that those style names would be retained in Word. At least the styles are applied most of the times. Sometimes italics are gone.

Sorry: Word (mine is 2003) does not convert internal links, when converts HTML to DOC.

I tried: changed all occurences of <a href="../Text/mymergedbook_1.xhtml# to <a href="# before opening the html in Word, but with no success.

And, the second: after merging files with Sigil, the new file could not open in Firefox Epubreader! It says: "A Firefox nem találja a fájlt a(z) /C:/Documents and Settings/Lala/Application Data/Mozilla/Firefox/Profiles/febeprof.Lala/epub/127/OEBPS/jogilexikon_1.xhtml helyen." (Firefox does not find the file in that place.)

So, total failure...

[Toxaris]

Originally Posted by laji

Sorry: Word (mine is 2003) does not convert internal links, when converts HTML to DOC.

I tried: changed all occurences of <a href="../Text/mymergedbook_1.xhtml# to <a href="# before opening the html in Word, but with no success.

And, the second: after merging files with Sigil, the new file could not open in Firefox Epubreader! It says: "A Firefox nem találja a fájlt a(z) /C:/Documents and Settings/Lala/Application Data/Mozilla/Firefox/Profiles/febeprof.Lala/epub/127/OEBPS/jogilexikon_1.xhtml helyen." (Firefox does not find the file in that place.)

So, total failure...

As I don't work with Word 2003 anymore for a long time already, I have no opportunity to check. However, links are always cumbersome during the import, also for newer version. For example, if you have foot-/endnotes the links will work, but not considered as foot-/endnotes. It would be best to recreate the links in the Word document. What is the origin of the links? Where do they point to?

If you merge the files with Sigil (take care if you have internal stylesheets like sgc-x, things can really be messy then), you need to recreate the ncx (TOC) before using it as an ePUB again. Probably that is the reason why it doesn't work in ePUBReader.