Markdownify upgrade is causing problems

[harshitgtypeface]

With the recent version of markdownify, the library no longer converts HTML content into proper Markdown. Instead, it often returns the original HTML as-is. This issue is especially noticeable when handling HTML extracted from PDFs, where the expected Markdown formatting is incorrect.

[afourney]

Is there a sample document you could provide?

Also, to clarify, if HTML is extracted from a PDF, it won't automatically be converted to markdown. This type of recursive processing currently is not automatic.