Too many page breaks when converting from .azw3 to epub

[Kaetrin]

Hi,

I've noticed recently that books I've bought from Amazon and converted to ePub in Calibre are showing up on my reader with a page break at the end of each paragraph. So, I'm getting a book with 2756 pages instead of 290 and I have to swipe after the end of each paragraph - when the paragraph consists of ""Yes" he said" it is especially frustrating.

Is there some code I can use in the edit book function to fix the conversion or can you suggest some other way to make the ePub version look better?

thx so much!

[pdurrant]

Do not edit the epub. Tweak the conversion parameters. I'd guess that the paragraphs in the AZW3 are marked up in a way that the default calibre settings interpret as requiring a new page.

[Kaetrin]

How would I tweak the conversion parameters? Any suggestions? (I've never had to do it before - I've just clicked "convert"!)

[Divingduck]

For me a better solution is most time the plugin Kindle unpack (KF8 to EPUB).

https://www.mobileread.com/forums/sho...d.php?t=171529

[pdurrant]

Originally Posted by Kaetrin

How would I tweak the conversion parameters? Any suggestions? (I've never had to do it before - I've just clicked "convert"!)

In the conversion dialog there's a section 'structure detection'. I suspect some paragraph attribute is matching the 'detect chapters' expression.

[JSWolf]

Originally Posted by Divingduck

For me a better solution is most time the plugin Kindle unpack (KF8 to EPUB).

https://www.mobileread.com/forums/sho...d.php?t=171529

After the format shift from KF8 to ePub, you will need to validate the code because it will contain errors.Not many and it's easy to fix.

[Divingduck]

Yes, that is correct. Thanks for the hint (I forgot to mention this point). I usually open each of these files with the editor and adjust the findings on the fly. As you wrote, easy and fast to fix. For me this is in most cases a better way then to screw up the whole book structure with a standard EPUB conversion.

[Kaetrin]

@JS Wolf I'm sorry, I'm not sure what you mean by "validate the code".

I was using Kindle unpack to convert amazon files to epub but my reader didn't like it much (although I have a new one now so maybe that's changed). Even though I could see the epub was fine in Calibre it wouldn't open on my reader. I found that if I convered .azw3 to mobi and then converted to epub from the mobi file just using the standard convert button in Calibre it worked fine. Perhaps a little more unwieldy but I was happy enough and I never had trouble getting books on my reader. But suddenly there are a subset of the .azw3 files which are converting to epub strangely.

@PDurrant
I don't know much about html code or anything like that so I'm hoping the solution will be really simple. Many of the suggestions so far have gone over my head I'm afraid. You are probably right about some paragraph attribute but I don't have the first clue what I'm looking for here or how to fix it. Hence I came to the experts!

[JSWolf]

Originally Posted by Kaetrin

@JS Wolf I'm sorry, I'm not sure what you mean by "validate the code".

I was using Kindle unpack to convert amazon files to epub but my reader didn't like it much (although I have a new one now so maybe that's changed). Even though I could see the epub was fine in Calibre it wouldn't open on my reader. I found that if I convered .azw3 to mobi and then converted to epub from the mobi file just using the standard convert button in Calibre it worked fine. Perhaps a little more unwieldy but I was happy enough and I never had trouble getting books on my reader. But suddenly there are a subset of the .azw3 files which are converting to epub strangely.

Um.. don't do that. It makes a right mess of things. Mobi is obsolete and going KF8 > Mobi > ePub is a rather bad idea.

Validate the code means opening the eBook in the Calibre eBook editor and validating the code and then fixing anything found. I should then work on your Reader.

[Kaetrin]

@JSWolf. You see, I don't understand what you've just said. I don't know what "validating the code" means. I don't know what codes to look for and what is "good" code and what is "bad" code. I don't know what instructions to put in to change something from one way to another. I'm not experienced in ebook editing at all. All I want to be able to do is not have a page break at the end of every paragraph when I convert from .azw3 to epub.

As for converting to mobi first, I still buy books in mobi format and I've never had any trouble converting from .azw3 to mobi to epub before. Even if I just convert from .azw3 to epub, I'm still getting the same issue however (I tested it). In any event, it's not mobi that's causing the issue, I'm quite confident of that. There's something in the .azw3 file but I don't know what it is, how to identify it or how to get rid of it.

[jackie_w]

@Kaetrin,

It sounds to me as if you are using calibre's default setting for page splits and that this particular book is totally unsuited for that default. Try the following:

Make sure you're converting from the original source azw3 and on calibre's Convert - Structure Detection page try the following 2 settings:
Detect chapters at: /
Insert page breaks before: /

and on the Convert - Table of Contents page, make sure the Level 1, Level 2, Level 3 TOC boxes are all blank

The calibre defaults were probably created for the early ebook days when it was necessary to split a solid wall of text into sensible chapter chunks. These days a well-designed retail quality ebook should (but don't bet real money on it ) already be split into the correct chunks with a matching TOC so you don't need calibre to try and guess where the chapters start/end.

ETA: Also make sure that on the Convert - Heuristic Processing page the 'Enable heuristic processing' is UNCHECKED

[Divingduck]

Originally Posted by Kaetrin

@JS Wolf I'm sorry, I'm not sure what you mean by "validate the code".
I was using Kindle unpack to convert amazon files to epub but my reader didn't like it much (although I have a new one now so maybe that's changed). Even though I could see the epub was fine in Calibre it wouldn't open on my reader.

As I brought this solution up to discussion with no further explanations, here a short introduction how to go forward with Kindle unpack and error correction of the resulting EPUB. The ZIP contains a word file with explanations, some screenshots and a saved search example for import used in the explanation for the calibre editor. Hope you will find it useful for your needs.

Best regards, DD

[Kaetrin]

Hi Jackie,

Here is what Calibre defaults to:

Detect chapters at (x path expression)
//*[((name()='h1' or name()='h2') and re:test(., '\s*((chapter|book|section|part)\s+)|((prolog|prol ogue|epilogue)(\s+|$))', 'i')) or @class = 'chapter']

and

Insert page breaks before (x path expression)
//*[name()='h1' or name()='h2']

Should I delete one of the / symbols?

[Kaetrin]

thx, I'll take a look and see if I can get my head around it.

[jackie_w]

Set both of them to a single forward-slash.