feat(epub): Add EPUB support

[0xRaduan]

Addresses #88.

Adds new converter + new test.

[l-lumin]

hi, could you check if this can use existing HtmlConverter

markitdown/src/markitdown/_markitdown.py

Lines 183 to 223 in cb66b35

	class HtmlConverter(DocumentConverter):
	"""Anything with content type text/html"""
	def convert(
	self, local_path: str, **kwargs: Any
	) -> Union[None, DocumentConverterResult]:
	# Bail if not html
	extension = kwargs.get("file_extension", "")
	if extension.lower() not in [".html", ".htm"]:
	return None
	result = None
	with open(local_path, "rt", encoding="utf-8") as fh:
	result = self._convert(fh.read())
	return result
	def _convert(self, html_content: str) -> Union[None, DocumentConverterResult]:
	"""Helper function that converts and HTML string."""
	# Parse the string
	soup = BeautifulSoup(html_content, "html.parser")
	# Remove javascript and style blocks
	for script in soup(["script", "style"]):
	script.extract()
	# Print only the main content
	body_elm = soup.find("body")
	webpage_text = ""
	if body_elm:
	webpage_text = _CustomMarkdownify().convert_soup(body_elm)
	else:
	webpage_text = _CustomMarkdownify().convert_soup(soup)
	assert isinstance(webpage_text, str)
	return DocumentConverterResult(
	title=None if soup.title is None else soup.title.string,
	text_content=webpage_text,
	)

[gagb]

+1

[0xRaduan]

addressed

[gagb]

@0xRaduan love this PR. We already have a dependency for HTML to text (markdownify) in the HTML convertor. Can you check if that would be sufficient?

[0xRaduan]

Hey @gagb, sorry was on a long vacation, going to take a look right now...

[samuelfernandez]

Thanks for this, really looking forward to it!

[dgiagio]

Can we have an update on this? Thanks

[0xRaduan]

self-review

[0xRaduan]

noticed that it doesn't have an tag, and that's used in Epub as far as I know

[0xRaduan]

cc. @gagb - do you think we can merge this?

i resolved all the merge conflicts as far as i can see

[0xRaduan]

or also cc. @afourney, since I see you've been merging the latest PRs into main

[0xRaduan]

@gagb - does this still await my response? any timeline for getting this merged?

[afourney]

@0xRaduan Apologies for the delay. We're a super small team, with several large projects (e.g., AutoGen). I'll work on getting this in, and conflicts resolved, this weekend.

[afourney]

Ok, on second glance, EbookLib is AGPL -- which is very strong copyleft. I'm not clear we can include it here. I can look for an alternative, or I can help you set it up as a 3rd party plugin that you can host. LMK

[afourney]

@0xRaduan I adapted this PR to not use ebooklib (as per above discussion). Admittedly, rather vibe-coded.

Please have a look at #1131 and let me know if it suits your need.

[afourney]

Closed in #1131

[0xRaduan]

Thanks, closing this PR.

[mad768063]

• Symferopolskaya 2L, Днепр, 49005, Украина

[mad768063]

Symferopolskaya 2L, Днепр, 49005, Украина

[mad768063]

@