Model/epc_data/attributes/WallAttributes.py

import re
from typing import Dict, Union
from epc_data.attributes.attribute_utils import extract_component_types


class WallAttributes:
    WALL_TYPES = ['cavity wall', 'filled cavity', 'solid brick', 'system built', 'timber frame', 'granite or whinstone',
                  'as built', 'cob', 'assumed', 'sandstone or limestone']

    def __init__(self, description: str):
        """
        :param description: Description of the walls.
        """
        self.description: str = description

    def process(self) -> Dict[str, Union[float, str, bool, None]]:
        result: Dict[str, Union[float, str, bool, None]] = {}
        description = self.description.lower()

        # thermal transmittance - it can be negative which is errneous however we'll still pull it out
        match = re.search(r"average thermal transmittance (-?\d+\.\d+)\s(w/m-¦k)", description)
        if match:
            result['thermal_transmittance'] = float(match.group(1))
            result['thermal_transmittance_unit'] = match.group(2)
        else:
            result['thermal_transmittance'] = None
            result['thermal_transmittance_unit'] = None

        # wall type
        result, description = extract_component_types(result, description, list_of_components=self.WALL_TYPES)

        # insulation thickness - this is far from a perfect approach and we'd likely need to use nlp to do this
        # generally however this is sufficient for mvp
        thickness_map = {
            "external insulation": "average",
            "internal insulation": "average",
            "partial insulation": "below average",
            "no insulation": "none",
            "additional insulation": "above average",
            "insulated": "average"
        }
        for key, value in thickness_map.items():
            if key in description:
                result['insulation_thickness'] = value
                break
        else:
            result['insulation_thickness'] = None

        # insulation type
        result['external_insulation'] = 'external insulation' in description
        result['internal_insulation'] = 'internal insulation' in description

        return result