How to obtain a new page for every chapter in regular expression (search/replace)?

Ravenik · 05-23-2012, 05:32 PM

Hi, I am a beginner in Calibre. I convert .epub -> .epub in order to break to a new page for every chapter.
I have chapters' heads in the format: ". XX ." where XX is a number.
My question is if I can force a new page for the destination epub before every chapter, using regular expression? What will I put in Replacement Text to result in a new page?

Currently I have:

Search:

Code:

<p class="calibre2"><span>[\s]*(\. \d\d? \.)

The Test search results OK, finding 77 occurances (correctly=the number of chapters)!!!!!

Replacement:

Code:

<div class="mbp_pagebreak"/></div><p class="mojrozdz1"><span>\1

I thought that <div class="mbp_pagebreak"/></div> will result in new pages (probably also in separate 77 .html files), but it does not.

Generally the above search/replace reg.exp. fails in two ways:

"calibre2" class name is not replaced with "mojrozdz". "mojrozdz" does not occur in the result html files!
<div class="mbp_pagebreak"/></div> does not occur in result file and furthermore there is no new page and no new .html file for each chapter

Please help me to understand what is wrong in my procedure as regards aforementioned failures 1. and 2. ? Calibre 0.8.52

ldolse · 05-24-2012, 12:55 AM

The pagebreak you're trying to insert is specific to mobi, not ePub. Try this:

Search Pattern:

Code:

<p[^>]*>\s*<span[^>]*>\s*(\.\s*\d+\s*\.)\s*</span>\s*</p>

Replacement:

Code:

<h2>\1</h2>

Then you need to go to Structure detection and change the chapter detection xpath to:

Code:

//h:h2

Calbre probably flattens the css in a way that mojrozdz1 disappears from your perspective (although it's also possible your regex was bad) - the correct way to do a custom style is to use extra css under Look and Feel. So take whatever you styles you want in mojrozdz1 and add them there for h2:

Code:

h2{text-align:center;}

When your doing these kinds of replacements it's good to enable debugging so you can see what the html looks like at each stage of the conversion. You want to look at html in the parsed and/or structure debug folders to see the results of your search and replace.

Ravenik · 05-24-2012, 07:04 AM

Hi, Thanks a lot, your solution works! Now I think I will study a bit about xpath syntax.

05-23-2012, 05:32 PM	#1
Ravenik Junior Member Posts: 4 Karma: 10 Join Date: May 2012 Device: Sony PRS-T1	How to obtain a new page for every chapter in regular expression (search/replace)? Hi, I am a beginner in Calibre. I convert .epub -> .epub in order to break to a new page for every chapter. I have chapters' heads in the format: ". XX ." where XX is a number. My question is if I can force a new page for the destination epub before every chapter, using regular expression? What will I put in Replacement Text to result in a new page? Currently I have: Search: Code: <p class="calibre2"><span>[\s]*(\. \d\d? \.) The Test search results OK, finding 77 occurances (correctly=the number of chapters)!!!!! Replacement: Code: <div class="mbp_pagebreak"/></div><p class="mojrozdz1"><span>\1 I thought that <div class="mbp_pagebreak"/></div> will result in new pages (probably also in separate 77 .html files), but it does not. Generally the above search/replace reg.exp. fails in two ways: "calibre2" class name is not replaced with "mojrozdz". "mojrozdz" does not occur in the result html files! <div class="mbp_pagebreak"/></div> does not occur in result file and furthermore there is no new page and no new .html file for each chapter Please help me to understand what is wrong in my procedure as regards aforementioned failures 1. and 2. ? Calibre 0.8.52

05-24-2012, 12:55 AM	#2
ldolse Wizard Posts: 1,337 Karma: 123455 Join Date: Apr 2009 Location: Malaysia Device: PRS-650, iPhone	The pagebreak you're trying to insert is specific to mobi, not ePub. Try this: Search Pattern: Code: <p[^>]>\s<span[^>]>\s(\.\s\d+\s\.)\s</span>\s</p> Replacement: Code: <h2>\1</h2> Then you need to go to Structure detection and change the chapter detection xpath to: Code: //h:h2 Calbre probably flattens the css in a way that mojrozdz1 disappears from your perspective (although it's also possible your regex was bad) - the correct way to do a custom style is to use extra css under Look and Feel. So take whatever you styles you want in mojrozdz1 and add them there for h2: Code: h2{text-align:center;} When your doing these kinds of replacements it's good to enable debugging so you can see what the html looks like at each stage of the conversion. You want to look at html in the parsed and/or structure debug folders to see the results of your search and replace. Last edited by ldolse; 05-24-2012 at 01:04 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Help with regular expression search/replace	bfollowell	Sigil	12	06-20-2013 07:36 PM
Regular Expression: does not replace with spaces	murphycc	Conversion	3	02-11-2012 01:31 PM
Search & Replace - Regular expression	oldbwl	Calibre	2	01-09-2011 09:33 AM
Regular expression for chapter combining	bremler	Sigil	2	12-23-2010 09:06 AM
Find/Replace with regular expression	hydrolith	Sigil	6	03-01-2010 08:42 PM

05-24-2012, 07:04 AM	#3
Ravenik Junior Member Posts: 4 Karma: 10 Join Date: May 2012 Device: Sony PRS-T1	Hi, Thanks a lot, your solution works! Now I think I will study a bit about xpath syntax.

Advert