TXT to EPUB: How to Make Chapter Markers as Titles

[citrate]

Hello there. I have a novel in .txt format and I want to convert it to .epub format. Since there's no tags decorating chapter markers (hashtags or <h2> tag) in the original text, I used regular expression (`//*[re.test(., "<regexp>", "i")]`) in structure detection to find out all the chapters.

The chapters are detected correctly, however, the chapter markers are treated as ordinary text. Is it possible to let Calibrie understand that the detected chapter marker are titles/headers?

For example, If I have something like:

```
Chapter 1 Chapter Name

some text some text
```

I want it be something like

```
# Chapter 1 Chapter Name

some text some text
```

in the .epub output. That is, give the chapter marker/title a different format from ordinary text.

[kovidgoyal]

txt is internally converted to HTML and that is what the expression applies too. You can see the HTML using the debug section of the conversion dialog.

[JSWolf]

Originally Posted by citrate

Hello there. I have a novel in .txt format and I want to convert it to .epub format. Since there's no tags decorating chapter markers (hashtags or <h2> tag) in the original text, I used regular expression (`//*[re.test(., "<regexp>", "i")]`) in structure detection to find out all the chapters.

The chapters are detected correctly, however, the chapter markers are treated as ordinary text. Is it possible to let Calibrie understand that the detected chapter marker are titles/headers?

For example, If I have something like:

```
Chapter 1 Chapter Name

some text some text
```

I want it be something like

```
# Chapter 1 Chapter Name

some text some text
```

in the .epub output. That is, give the chapter marker/title a different format from ordinary text.

What novel is this?

[citrate]

Originally Posted by kovidgoyal

txt is internally converted to HTML and that is what the expression applies too. You can see the HTML using the debug section of the conversion dialog.

Yes, the .txt file is internally converted to HTML file. It becomes something like

```
...
<p>final sentence of last chapter</p>
<p class="whitespace" style="text-align:center; margin-top:0em; margin-bottom:0em">*</p>
<p>Chapter X</p>
<p>first sentence of current chapter</p>
...
```

After detecting structure, it becomes

```
...
<p>final sentence of last chapter</p>
<p class="whitespace" style="text-align:center; margin-top:0em; margin-bottom:0em">*</p>
<div style="display: block; page-break-after: always"></div>
<p id="calibre_toc_X">Chapter X</p>
<p>first sentence of current chapter</p>
...
```

But what I want is something like

```
...
<p>final sentence of last chapter</p>
<p class="whitespace" style="text-align:center; margin-top:0em; margin-bottom:0em">*</p>
<div style="display: block; page-break-after: always"></div>
<h2 id="calibre_toc_X">Chapter X</h2>
<p>first sentence of current chapter</p>
...
```

Pay attention to the line of Chapter X, I want it to become a title. Is it possible to achieve this in Calibrie?

[citrate]

Originally Posted by JSWolf

What novel is this?

Well... It's a Chinese web novel and the text I provided is just an example of what I want rather than the real content of it.

[kovidgoyal]

you want something like

//h:p[re.test(., '^Chapter.*', 'i')]

[citrate]

Originally Posted by kovidgoyal

you want something like

//h:p[re.test(., '^Chapter.*', 'i')]

Well it doesn't work. I think above XPath expression is "find all <p>-tag text that matching the regular expression", but what I want is

1. Find all text matching the regular expression
2. Change the <p>-tag of the matching text to an <h>-tag

[kovidgoyal]

You cannot change tags, but using the correct expression will allow you to mark those tags as chapter starts with either a pagebreak or a rule above.

[Ghostcat]

Why not use <b>Search and Replace</b> to change <p> to <h2>?

E.g:

search for "<p>(Chapter [0-9]+)</p>"

replace with "<h2>\1</h2>"