Source code for energy_data_retrieval.wind_catalogue.fosen

"""Offline Fosen Vind fact-card inventory; reported output is not measured yield."""

from __future__ import annotations

import hashlib
import math
import re
from html.parser import HTMLParser
from typing import Any


class _Cards(HTMLParser):
    def __init__(self) -> None:
        super().__init__()
        self.depth = 0
        self.heading = False
        self.item = False
        self.name: list[str] = []
        self.words: list[str] = []
        self.items: list[str] = []
        self.cards: list[tuple[str, list[str]]] = []

    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
        if tag == "div":
            if self.depth:
                self.depth += 1
            elif "featured-content-block" in (dict(attrs).get("class") or "").split():
                self.depth = 1
                self.name, self.items = [], []
        if self.depth and tag == "h2":
            self.heading = True
        if self.depth and tag == "li":
            if self.item:
                raise ValueError("Nested fact-card list item")
            self.item, self.words = True, []

    def handle_data(self, data: str) -> None:
        if self.heading:
            self.name.append(data)
        if self.item:
            self.words.append(data)

    def handle_endtag(self, tag: str) -> None:
        if tag == "h2":
            self.heading = False
        if tag == "li" and self.item:
            self.items.append(" ".join("".join(self.words).split()))
            self.item = False
        if tag == "div" and self.depth:
            self.depth -= 1
            if not self.depth:
                if self.item or self.heading:
                    raise ValueError("Unclosed card field")
                self.cards.append((" ".join("".join(self.name).split()), self.items))


[docs] def parse_fosen_cards( original: bytes, *, sha256: str, expected_names: list[str] ) -> list[dict[str, Any]]: """Require an exact complete card roster and retain units and source labels. Tower height is not silently called hub height. Annual generation has no calendar period, gross/net boundary or measured/modeled classification in these cards. No capacity factor or operational performance is inferred. """ if hashlib.sha256(original).hexdigest() != sha256: raise ValueError("Original checksum mismatch") parser = _Cards() parser.feed(original.decode("utf-8")) parser.close() if parser.depth or parser.item or parser.heading: raise ValueError("Truncated fact-card document") names = [n for n, _ in parser.cards] if not names or len(names) != len(set(names)) or sorted(names) != sorted(expected_names): raise ValueError("Incomplete or duplicate fact-card roster") fields = { "generation_gwh": ("Produksjon (GWh)", "", "GWh"), "installed_mw": ("Installert effekt (MW)", "", "MW"), "rotor_m": ("Rotordiameter", "meter", "m"), "tower_height_m": ("Tårnhøyde", "meter", "m"), "turbines": ("Vindturbiner", "", "count"), "road_length_km": ("Veier, adkomst og interne (km)", "", "km"), "planning_area_km2": ("Planområde", "km2", "km2"), "construction_year": ("Byggestart", "", "year"), "assembly_year": ("Turbinmontasje", "", "year"), "commissioning_year": ("Idriftsettelse", "", "year"), } results = [] for name, items in parser.cards: facts: dict[str, Any] = {"name": name, "source_items": items} for key, (label, suffix, unit) in fields.items(): matches = [i for i in items if re.match(re.escape(label) + r"(?:\s|:|$)", i)] if len(matches) != 1: raise ValueError(f"Missing or duplicate {label}") number = matches[0][len(label) :].lstrip(": ").removesuffix(suffix).strip() if not re.fullmatch(r"\d+(?:,\d+)?", number): raise ValueError(f"Invalid Norwegian number for {label}") value = float(number.replace(",", ".")) if not math.isfinite(value): raise ValueError("Nonfinite fact-card number") if key.endswith("year") or key == "turbines": if not value.is_integer(): raise ValueError("Fractional count or year") value = int(value) facts[key] = {"value": value, "unit": unit, "original": matches[0]} models = [i for i in items if re.match(r"Turbin(?:\s|:)", i)] if len(models) != 1: raise ValueError("Missing or duplicate turbine model label") facts["model_label"] = models[0].removeprefix("Turbin").lstrip(": ") facts["generation_basis"] = {"period": None, "gross_net": None, "kind": None} facts["exact_equipment_revision"] = None results.append(facts) return sorted(results, key=lambda r: r["name"])