Don't Understand Error Message: ... but found '"liga" 0'

[enuddleyarbl]

Originally Posted by JSWolf

@DaveLessnau Would you mind shifting from KF8 to ePub using KindleUnpack and then scrambling it and posting so we can see the original code? Thanks.

Here's the epub (Scrambled) that KindleUnpack produced using the KF8 to epub option:

https://drive.google.com/file/d/1stC...ew?usp=sharing

I started from the azw3 from my new download, turned on most of the heuristic processing in Calibre's conversion routine (which didn't really help), converted it to an epub and have been working on that. I think I've got most everything cleared up but the <div id=...> stuff at the start of every paragraph.

[enuddleyarbl]

I think I've figured out the regex to search/replace those <div id=...> things out:

Search for: <(div id="\S+?") (class="para-normal">.+?)(<\/div>)
Replace with: <p \2</p>

"Wrap" and "Dot All" are checked in the Search dialog. That looks for the first occurrence of:

<div id=blahblahblah class="para-normal">blahblahblah_until_it_reaches</div>

it replaces that with:

<p class="para-normal">blahblahblah</p>

I'd reduced all the paragraph styles down to that para-normal and paracenter (one of mine). I just had to do another run with paracenter in the place of para-normal.

That appears to have worked. Next, I've got to get rid of the <br ....> things that are at the end of most paragraphs.

[JSWolf]

Originally Posted by DaveLessnau

I think I've figured out the regex to search/replace those <div id=...> things out:

Search for: <(div id="\S+?") (class="para-normal">.+?)(<\/div>)
Replace with: <p \2</p>

"Wrap" and "Dot All" are checked in the Search dialog. That looks for the first occurrence of:

<div id=blahblahblah class="para-normal">blahblahblah_until_it_reaches</div>

it replaces that with:

<p class="para-normal">blahblahblah</p>

I'd reduced all the paragraph styles down to that para-normal and paracenter (one of mine). I just had to do another run with paracenter in the place of para-normal.

That appears to have worked. Next, I've got to get rid of the <br ....> things that are at the end of most paragraphs.

Before you get rid of the <br you'll need to code for the spacing to replace the <br.

[enuddleyarbl]

My replacement regex for the <div... stuff added a </p> to the end of all those paragraphs. So, all I did was delete all the <br...> tags and left the </p> tags to handle the paragraph spacing. I had to manually adjust some of the paragraphs on a couple of pages (on, for instance, the copyright page). But, in general, that part was easy.

I've got a reasonable looking ebook now and am reading it for issues and to look for places I can put my own heading/chapter marks and scene breaks. For some reason, Zelazny and/or the publishers didn't bother with things like that.

I've been scratching my head over why the publisher would have put that ridiculous html and css stuff in there. My guess is that they started with either a scanned copy of the paper book (or a PDF of one) and stuck those styles in there because of variations in how that came out instead of how it should have looked. Why else would they have lines/words in a single paragraph changing their height? I'd have thought that someone might have actually looked at the finished product and realized they were trying to reproduce scanning issues in CSS.

And, BTW, the original issue I started this thread with (liga 0) is now OBE: I deleted almost all of the stuff in those areas. Sorry I so quickly caused this thread to stray from an Editor issue to a Conversion one.

[JSWolf]

I've had a quick look at the code and yes, it is a mess. Most of the CSS files are duplicates which you can delete. Then you can move all the CSS code from the remaining CSS to the first CSS and delete the rest. Just remember to fix all the HTML to link to that CSS.

[DNSB]

Originally Posted by DaveLessnau

I've got a reasonable looking ebook now and am reading it for issues and to look for places I can put my own heading/chapter marks and scene breaks. For some reason, Zelazny and/or the publishers didn't bother with things like that.

Going by my paper copy, the original was one rather long chapter. I had a few other science fiction books that were the same. One of those fashion decisions that turned out bad?

[enuddleyarbl]

I've noticed that a lot of Zelazny and Pratchett books don't bother with chapters.

[enuddleyarbl]

I'm running EPUBCheck from within the Editor and it's pointing out duplicates id= lines. Something like this at the start of several of the files:

<body class="normbody">
<div class="page" id="section2_1">
<div id="box1_1" class="normbody">
...

These are leftovers from the original since I basically don't know what they're for. According to:

https://www.w3docs.com/learn-css/css-id-and-class.html

An ID selector is a unique identifier of the HTML element to which a particular style must be applied. It is used only when a single HTML element on the web page must have a specific style.

Both in Internal and External Style Sheets we use hash (#) for an id selector

So, even ignoring the duplicates, there ought to be a style somewhere where #section2_1 and #box1_1 have some attribute applied. But, I can find no reference to them. I suppose it's possible I've deleted it already without realizing it. Can I just delete both of those <div... lines?

[JSWolf]

Originally Posted by DaveLessnau

I've noticed that a lot of Zelazny and Pratchett books don't bother with chapters.

Discworld has no chapters.

[JSWolf]

Originally Posted by DaveLessnau

I'm running EPUBCheck from within the Editor and it's pointing out duplicates id= lines. Something like this at the start of several of the files:

<body class="normbody">
<div class="page" id="section2_1">
<div id="box1_1" class="normbody">
...

These are leftovers from the original since I basically don't know what they're for. According to:

https://www.w3docs.com/learn-css/css-id-and-class.html



So, even ignoring the duplicates, there ought to be a style somewhere where #section2_1 and #box1_1 have some attribute applied. But, I can find no reference to them. I suppose it's possible I've deleted it already without realizing it. Can I just delete both of those <div... lines?

A lot of the iDs are for footnotes as Discworld books can have a number of footnotes.