from __future__ import annotations from bs4 import BeautifulSoup from typing import TYPE_CHECKING if TYPE_CHECKING: from io import TextIOWrapper class XMLLoader: def load(file_: TextIOWrapper) -> dict: schema = {} bs = BeautifulSoup(file_.read(), "xml") structElmts = bs.findAll("structure") structures = {} for structElmt in structElmts: structures[structElmt.get("id")] = XMLLoader.parseStructure(structElmt) schema["structures"] = structures return schema def parseStructure(structElmt: any) -> dict: struct = {} struct["bits"] = structElmt.get("bits") ranges = {} rangeElmts = structElmt.findAll("range") for rangeElmt in rangeElmts: span = rangeElmt.get("end") + "-" + rangeElmt.get("start") ranges[span] = XMLLoader.parseRange(rangeElmt) struct["ranges"] = ranges return struct def parseRange(rangeElmt: any) -> dict: range_ = {} range_["name"] = rangeElmt.get("name") desc = rangeElmt.find("description") if desc is not None: range_["description"] = desc.getText() valuesElmt = rangeElmt.find("values") if valuesElmt is not None: range_["values"] = XMLLoader.parseValues(valuesElmt) if rangeElmt.get("depends-on"): range_["depends-on"] = rangeElmt.get("depends-on") return range_ def parseValues(valuesElmt: any) -> dict: values = {} caseElmts = valuesElmt.findAll("case") for caseElmt in caseElmts: val = caseElmt.get("value") desc = caseElmt.getText() struct = caseElmt.get("structure") if struct: values[val] = { "description": desc, "structure": struct } else: values[val] = desc return values