Using Ebooklib in Django project, so links in book are not working

[leonfitz]

Hi,
I'm using Ebooklib inside a Django project, i.e. extracting ITEM_DOCUMENT and ITEM_STYLE from a given epub file and displaying the styled xhtml content in the browser. This works great, but any links inside the epub document (to images in a folder, or to chapters) don't work, obviously, because the url clashes with the django url.

I'm thinking about solutions, and I'm wondering if there was an easier solution than filtering the whole xhtml content of the ebook with regular expressions and modifying the links so the root is the epub folder and not the Django url. Is there a built-in function in ebooklib to extract links?

This isn't actually an issue of ebooklib, I hope it's ok to ask about it here. Amazing project, thanks a lot!

[c1924959470]

   您好！您的来信我已接受，我会尽快回复您。

[aerkalov]

Booktype was online book editing tool which used Ebooklib. It was written in Django and has all kind of export/import/editing/view parts which might help you with how to use Ebooklib to change the structure of the document.

This is for instance importer which does opposite of what you need to do.

This code reads EPUB and modifes all kind of links and everything else to make it editor friendly and you need to do something similar:
https://github.com/booktype/Booktype/blob/master/lib/booktype/importer/epub/epubimporter.py

This code creates EPUB file from some internal structure:
https://github.com/booktype/Booktype/blob/master/lib/booktype/convert/epub/converter.py

This ones is used for creating XHTML output and is probably something similar to what you need to do:
https://github.com/booktype/Booktype/blob/master/lib/booktype/convert/xhtml/converter.py

What you need to do is use lxml (don't use regular expressions) to find all existing links to other pages, images and other resources and rewrite them to use new paths required by your app. These are converters and are part of a bigger architecture of Booktype and you will probaby need to convert chapter and not entire book in one go.

Hope it helps.

Aco

[aerkalov]

Oh yes, just wanted to add.

Use this function ebooklib.utils.parse_html_string to parse the content of the chapter. Use lxml methods on the document to find elements you need (xpath, iter, ...). When you find what you need to modify use .get method on the element to fetch value of the src/href/... elements and use .set method to set new values. When you are done just use etree.tostring method to serialize lxml object to html string.

Not sure if these are some EPUB files you created and you know they are valid, but always use urllib.unquote, urlparse.urlparse, os.path.basename like you will find in the examples I pasted. You can find all kind of strange things in EPUB files so always be sure and verify everything. Never assume. Code i pasted was written for Python 2 so just modify it to use correct Python 3 modules.