Does it support extracting structured information from the results recognized by OCR

Suppose I now use a certain OCR model to extract MarkDown files from product information PDFS or some contract form-based PDFS. Can langextract reasonably parse structured information from md files? If possible, I think this will be a really great project