from datatypes.epc.domain.epc_property_data import EpcPropertyData
from datatypes.epc.domain.mapper import EpcPropertyDataMapper

from backend.documents_parser.extractor import PasHubRdSapSiteNotesExtractor
from backend.documents_parser.pdf import pdf_to_text_list


def parse_pashub_site_notes(file_path: str) -> EpcPropertyData:
    with open(file_path, "rb") as f:
        pdf_bytes = f.read()
    tokens = pdf_to_text_list(pdf_bytes)
    site_notes = PasHubRdSapSiteNotesExtractor(tokens).extract()
    return EpcPropertyDataMapper.from_site_notes(site_notes)