How to use
- Drop an EPUB or TXT file, or click to choose one. The file type is recognized automatically.
- For an EPUB you get EPUB → TXT and EPUB → PDF; for a TXT file you get TXT → EPUB.
- Set the options, then press the download (or print) button.
Conversions
| Conversion | Result | Main options |
|---|---|---|
| EPUB → TXT | One text file in reading order | # headings, collected footnotes, UTF-8 / UTF-8 (BOM) / CP949, CRLF |
| EPUB → PDF | A PDF saved through the browser print dialog | Paper (A5, A4), margin 5-40 mm, text 8-20 pt |
| TXT → EPUB | An EPUB 3 book with chapters and contents | Title, author, language, cover, chapter rule |
EPUB → TXT puts a blank line between paragraphs and turns <br> into a line break. Entities such as & are decoded, and scripts, styles, the document head and ruby pronunciation marks are left out. Chapters are separated by two blank lines, so when you turn that TXT back into an EPUB the “2 or more blank lines” rule fits as is. Footnotes appear as [1] in the text and are collected at the end of the chapter. When saving as CP949 for old Korean Windows programs, characters CP949 does not have (such as emoji) become ?, and you are told how many beforehand.
EPUB → PDF goes through printing
Writing a PDF file directly would mean embedding CJK fonts into it, which makes files large and text prone to breaking. So the cleaned text is joined into one document, and the Print → Save as PDF button hands it to the browser’s own PDF export. In the print dialog choose “Save as PDF” and leave margins on “Default” so the @page margin set here applies. Each chapter starts on a new page. Scripts inside the book do not run and external addresses are removed while it is drawn.
Making an EPUB from a TXT novel
The text encoding is detected from a BOM, a strict UTF-8 check and a Korean-text score; if it is wrong, choose UTF-8 or CP949 yourself. There are four chapter rules:
| Rule | What counts as a chapter title |
|---|---|
| Chapter 1 / 제1장 heading lines | Lines of up to 120 characters starting with Chapter 1, Part IV, 제1장, 第12章, Prologue, 프롤로그 and similar |
| 2 or more blank lines | A new chapter after two or more blank lines. With a “title, blank line, text” shape the first line is the title |
| Every N characters | A new chapter at the paragraph break once N characters (minimum 500) are reached. Titles are numbers |
| Regex | A line matching the expression is the title (case-insensitive). Stopped if it takes over 3 seconds |
For example, the text after a Chapter 1 line belongs to that chapter, and text before the first heading becomes an untitled first chapter. Every line becomes one paragraph (<p>) with surrounding spaces removed. The finished EPUB stores mimetype first and uncompressed, then container.xml, an OPF (with dcterms:modified), a nav.xhtml table of contents, one XHTML file per chapter and a basic stylesheet. A cover adds a cover page. You can open the result right away in EPUB Reader.
What is not supported
- MOBI and AZW3 are not supported. Convert them to EPUB with a program such as Calibre first.
- DRM-protected EPUB files cannot be converted.
- Pictures are dropped from TXT output, and PDF page breaks can differ from the original for books with many figures or tables.
To change a text file’s encoding on its own use Text Encoding Converter, to measure length Character Counter, and to view PDFs PDF Viewer.
Files are processed only in your browser and are never sent to a server.