[Beetzart]

When you convert a pdf to epub, the result often is rather poor so it needs some cleaning up. What is the best and easiest way to do this? I know that Calibre can edit ebooks but I find it rather difficult and time consuming. Maybe that is what it is but if any of you have tips or tricks, or advice regarding other tools I'd be obliged.

it's always like that as far as I know when PDF converts to other formats

You can use Sigil to create and edit an epub ebook.

[Narasinha]

Like Constantine111 mentioned, Sigil is an excellent EPUB editor. It does require some knowledge of HTML and CSS for the best results. It has a plugin framework for validation, formatting, etc. There are Windows executables available from their GitHub page, but if you're on Mac or Linux (like me) they have instructions for downloading the source files and compiling it.

PDFs are notoriously hard to convert. The results can be bad enough that it's better to use an OCR package rather than try to convert the text directly. I've scanned a few books and used OCR and then Sigil to create the final ebooks. If you can get the plain text out of the PDF to start with, you're halfway there.

Some PDFs that are created commercially are set up in a way to purposely make them difficult to convert.

[kumarashantanu]

I would like to add further to narasinha that, sigil has further extension for those who have no basic knowledge about css and html. Its called page edit. Its available in github sigil page. you need to install it. to edit any text of ebook, just load .epub in sigil and click on external html editor. you need to enable it in setting. It will be super easy to edit further and handy too.

[marooned2]

Is there a way to check a epub for errors in Sigil, as in Calibre?
I've almost done creating/formatting a book, and noticed one chapter which looks fine in book preview but in code view, the lines appear split as it were a poem.

[Beetzart]

As far as converting pdf's to epub I have found Abby Finereader to give reasonable results.

[marooned2]

@Beetzart

I've used Abbyy only for OCR'ing but now I need to convert an OCR'd PDF to html and preserve all the formatting. Can Abby do that?

[Beetzart]

Yes I think so. You will come a long way. After OCR'ing you can save as html with options:
- formatted text
- keep pictures
- keep line breaks
- keep text and background colors
- keep headers and footers
- generate a table of contents
- preserve document metadata