Split at multiple locations : Help Needed

[gtriever]

I have a book with one large HTML file. Common chapter numbering in the Preview window is [Roman Numeral](space)[dash][dash](space)[Subtitle], for example II -- The Mystery Quest

Can anybody give me a regex expression to split that file at multiple locations, or do I have to continue to do it all manually? So far I'm entering the [specific Roman Numeral](space)[dash][dash] as my search term and cutting it at each spot using the split-click method.

[Karellen]

What is the code that surrounds "II -- The Mystery Quest"?

Is it a heading, like <h2>II -- The Mystery Quest</h2>?

[gtriever]

Originally Posted by Karellen

What is the code that surrounds "II -- The Mystery Quest"?

Is it a heading, like <h2>II -- The Mystery Quest</h2>?

No. There isn't any separate chapter formatting, the entire file content is coded as <p class="calibre3">. Any file splitting will have to be based off the text string [Roman Numeral](space)[dash][dash]

[theducks]

Is that the only place with 2 dashes in a P tag

<p class="calibre3">(.+?\s--\s.+?)

Replace:

 <hr class="sigil_split_marker" /> <h3 class="chapno">\1</h3>

I happen to use Sigil for this task, but Calibre will work fine with a small adjustment.

The Marker is what you split on.
I made it a H3 tag (chapno is the stylesheet entry)

[gtriever]

Originally Posted by theducks

Is that the only place with 2 dashes in a P tag

<p class="calibre3">(.+?\s--\s.+?)

Replace:

 <hr class="sigil_split_marker" /> <h3 class="chapno">\1</h3>

I happen to use Sigil for this task, but Calibre will work fine with a small adjustment.

The Marker is what you split on.
I made it a H3 tag (chapno is the stylesheet entry)

My apologies, but I'm embarassingly lost by what you just wrote... in fact, in my original post I incorrectly asked for a regex expression when it should have been an XPath expression.

[Karellen]

Originally Posted by gtriever

My apologies, but I'm embarassingly lost by what you just wrote... in fact, in my original post I incorrectly asked for a regex expression when it should have been an XPath expression.

What @theducks is trying to accomplish is to make your headings unique so you can easily split the pages.
Currently your headings have the same styling as your paragraphs, so there is no way to distinguish one from the other.
Except for the double dash/hyphen. Because of this unique use, and if the use is limited to those headings only, then you can use that uniqueness to capture those headings and change the coding to <h2>heading</h2>. From there you can easily split the chapters into separate files.

So insert that regex into the Search box and make sure you have selected "regex" as the search type.
Though for the replace I would use a simpler <h2>\1</h2>
Step through the first few one at a time to make sure it works correctly, then if you have a lot of chapters you can just replace all.

[gtriever]

Unfortunately, the double dash is used in the paragraphs as well. Oh well... thanks to all for the replies.

[Karellen]

How many chapters?

[theducks]

<p class="calibre3">([CLXVI]{1,7}\s--\s.+?)</p>

My Roman search added

[Quoth]

Originally Posted by gtriever

Unfortunately, the double dash is used in the paragraphs as well. Oh well... thanks to all for the replies.

They should use en or em dash. But irrelevant.

[phossler]

Originally Posted by gtriever

No. There isn't any separate chapter formatting, the entire file content is coded as <p class="calibre3">. Any file splitting will have to be based off the text string [Roman Numeral](space)[dash][dash]

You could try Diap's Toolbag plug in to convert <p class="calibre3 ... to h4 and then use Split at Multiple locations with Xpath = //h:h4

[Quoth]

Originally Posted by phossler

You could try Diap's Toolbag plug in to convert <p class="calibre3 ... to h4 and then use Split at Multiple locations with Xpath = //h:h4

But apparently ALL the regular paragraphs are <p class="calibre3 etc.

And calibre will split an entire ebook into a file per paragraph. It happened once when an added book had a class for the body paragraphs that looked like a chapter heating to calibre. Reverted to original, renamed class and reconverted.

[theducks]

The OP asked in Editor, So I gave a REGEX solution.
AFAIK xpath is for conversion

[gtriever]

Let me rephrase what I'm looking for:

Is it possible to automatically split at multiple locations based off of a text string?

With all the paragraphs and chapter titles being the same <p class = "calibre3">, and my programming skills without the Wizard being nil, I don't see any way do it.

The common "chapter" reference is [Roman Numeral](space)[dash], which is what I've been using in the Editor Preview window to search and then manually split one chapter at a time.

[theducks]

I gave you the tools to locate and insert a unique Marker.

You the simply use the split at multiple locations feature of the editor using the markers code as the criteria.
Use the Wizard in the tool to setup the x-path with ease