Extract tagged tables from pdf using 'StructTreeRoot'

[mllife]

So, I am working with pdfs and now I need to see, if I can parse the tables in pdfs (if tagged)
Reference - https://accessible-pdf.info/en/basics/general/overview-of-the-pdf-tags/#table-elements

import os
import pymupdf


def has_struct_tree_root(pdf_path):
    """
    Check if a PDF has a valid StructTreeRoot reference.
    """
    try:
        doc = pymupdf.open(pdf_path)
        catxref = doc.pdf_catalog()
        structroot = doc.xref_get_key(catxref, "StructTreeRoot")

        # Check if the StructTreeRoot reference exists
        if structroot[0] != "null":
            return structroot[1].split()[0]  # Return xref of StructTreeRoot
        else:
            return None
    except Exception as e:
        print(f"Error processing {pdf_path}: {e}")
        return None


def find_tables_in_struct_tree(doc, struct_tree_xref):
    """
    Parse the StructTreeRoot and identify elements related to tables in a PDF.
    """
    struct_tree_root_dict = doc.xref_object(int(struct_tree_xref))  # Dereference StructTreeRoot

    if "/K" in struct_tree_root_dict:
        # Extract the /K array (references to structure elements)
        if "/K [" in struct_tree_root_dict:
            k_array = struct_tree_root_dict.split("/K [")[1].split("]")[0].strip().split(" ")
        elif "/K " in struct_tree_root_dict:
            k_array = struct_tree_root_dict.split("/K ")[1].split(" ")[0]
        k_refs = [int(ref) for ref in k_array if ref.isdigit()]
        # Iterate over all objects in the /K array to find table-related tags
        for k_ref in k_refs:
            if k_ref == 0:
                continue
            # print(f"{k_ref=}")
            struct_elem = doc.xref_object(k_ref)
            # Check for table-related tags
            if any(tag in struct_elem for tag in ["/Table", "/TR", "/TH", "/TD", "/THead", "/TBody", "/TFoot"]):
                print(f"Table element found in object {k_ref}:")
                print(struct_elem)  # Output the table element details
    else:
        print("No /K array found in StructTreeRoot")


if __name__ == "__main__":
    pdf_file = "all_pdfs/new_pdfs/FishTables-only.pdf"
    struct_tree_xref = has_struct_tree_root(pdf_path=pdf_file)

    if struct_tree_xref:
        print(f"StructTreeRoot found in {pdf_file}.")
        doc = pymupdf.open(pdf_file)
        find_tables_in_struct_tree(doc, struct_tree_xref)

Attaching a sample pdf with tagged tables;
FishTables-only.pdf

[mllife]

How make the code parse the tables from the StructTreeRoot and extract json or csv using this information?