Markdown documents not supported?

[sammcj]

I went to try out Omniparse (looks great!) but when I went to upload my documents I was met with an error stating markdown documents aren't supported.

This really surprised me given most wikis, knowledge bases, code / library / project documentation / PKI data etc... are markdown (aka plain text).

Why would I want to parse data that's already Markdown you might ask?

• You might have a directory / git repo full of markdown across various depths of directories.
• Markdown could contain base64 or linked images that need to be parsed.
• Markdown could contain code blocks or inline HTML.
• Markdown could be very long and need to be chunked.

[qyou]

So far only doc/docx/ppt/pptx/pdf file formats are supported.

[adithya-s-k]

As mentioned by @qyou, we only support doc/docx/ppt/pptx/pdf. All the points you mentioned are valid, @sammcj. We will integrate normal text-based and table-based file types as well.

For chunking, please indicate which of the following is your preferred chunking strategy:

• Topic/Heading-Based Chunking
• Fixed-Size Sentence-Based Chunking
• Semantic Chunking

Your feedback will greatly help, @sammcj and @qyou.