tensorlakeai / indexify

A realtime and indexing and structured extraction engine for Unstructured Data to build Generative AI Applications
https://docs.getindexify.ai
Apache License 2.0
842 stars 93 forks source link

Add errors if files dont get boiled down properly by extractors #810

Open sadath-12 opened 1 month ago

sadath-12 commented 1 month ago

for example pptx wunt work with pdfextractor but it doesnt error tho while uploading , we get to know only while retrieving and get confused

diptanu commented 1 month ago

We need to look at the top level extraction policies of a graph and throw an error if the mime-type of the uploaded file doesn't match with any of them