Nye-TeeOff/backend/vtg_courses.py

import re
from datetime import date, datetime
from typing import Any


MONTH_MAP: dict[str, int] = {
    "januar": 1,
    "jan": 1,
    "februar": 2,
    "feb": 2,
    "mars": 3,
    "mar": 3,
    "april": 4,
    "apr": 4,
    "mai": 5,
    "juni": 6,
    "jun": 6,
    "juli": 7,
    "jul": 7,
    "august": 8,
    "aug": 8,
    "september": 9,
    "sep": 9,
    "sept": 9,
    "oktober": 10,
    "okt": 10,
    "november": 11,
    "nov": 11,
    "desember": 12,
    "des": 12,
}


def normalize_whitespace(value: str) -> str:
    return re.sub(r"\s+", " ", str(value or "")).strip()


def _to_date(year: int, month: int, day: int) -> date | None:
    try:
        return date(year, month, day)
    except ValueError:
        return None


def _infer_year(month: int, day: int, explicit_year: int | None, today: date) -> int:
    if explicit_year:
        return explicit_year

    candidate = _to_date(today.year, month, day)
    if candidate and candidate < today.replace(day=max(1, min(today.day, 28))):
        if (today - candidate).days > 7:
            return today.year + 1
    return today.year


def _parse_numeric_date(raw: str) -> date | None:
    match = re.search(r"\b(\d{1,2})[./](\d{1,2})[./](\d{2,4})\b", raw)
    if not match:
        return None

    day = int(match.group(1))
    month = int(match.group(2))
    year = int(match.group(3))
    if year < 100:
        year += 2000
    return _to_date(year, month, day)


def _parse_textual_dates(raw: str, today: date) -> list[date]:
    results: list[date] = []
    pattern = re.compile(
        r"\b(\d{1,2})\.?\s*(" + "|".join(sorted(MONTH_MAP.keys(), key=len, reverse=True)) + r")\b(?:\s+(20\d{2}))?",
        re.IGNORECASE,
    )
    for match in pattern.finditer(raw):
        day = int(match.group(1))
        month = MONTH_MAP.get(match.group(2).lower())
        if not month:
            continue
        explicit_year = int(match.group(3)) if match.group(3) else None
        year = _infer_year(month, day, explicit_year, today)
        candidate = _to_date(year, month, day)
        if candidate:
            results.append(candidate)
    return results


def parse_course_date_range(raw: str, today: date | None = None) -> tuple[date | None, date | None]:
    reference_today = today or date.today()
    normalized = normalize_whitespace(raw).lower()
    if not normalized:
        return None, None

    iso_candidate = None
    try:
        iso_candidate = datetime.fromisoformat(normalized).date()
    except ValueError:
        iso_candidate = None
    if iso_candidate:
        return iso_candidate, iso_candidate

    numeric_dates = re.findall(r"\b\d{1,2}[./]\d{1,2}[./]\d{2,4}\b", normalized)
    if len(numeric_dates) >= 2:
        start = _parse_numeric_date(numeric_dates[0])
        end = _parse_numeric_date(numeric_dates[1])
        return start, end or start
    if len(numeric_dates) == 1:
        single = _parse_numeric_date(numeric_dates[0])
        return single, single

    range_match = re.search(
        r"\b(\d{1,2})\s*\.?\s*(?:-|–|—|til)\s*(\d{1,2})\.?\s*(" + "|".join(sorted(MONTH_MAP.keys(), key=len, reverse=True)) + r")\b(?:\s+(20\d{2}))?",
        normalized,
        re.IGNORECASE,
    )
    if range_match:
        start_day = int(range_match.group(1))
        end_day = int(range_match.group(2))
        month = MONTH_MAP.get(range_match.group(3).lower())
        explicit_year = int(range_match.group(4)) if range_match.group(4) else None
        if month:
            year = _infer_year(month, end_day, explicit_year, reference_today)
            start = _to_date(year, month, start_day)
            end = _to_date(year, month, end_day)
            return start, end or start

    textual_dates = _parse_textual_dates(normalized, reference_today)
    if len(textual_dates) >= 2:
        return textual_dates[0], textual_dates[1]
    if len(textual_dates) == 1:
        return textual_dates[0], textual_dates[0]

    return None, None


def normalize_vtg_course_rows(rows: Any) -> list[dict[str, Any]]:
    if not isinstance(rows, list):
        return []

    normalized_rows: list[dict[str, Any]] = []
    for index, row in enumerate(rows):
        if not isinstance(row, dict):
            continue
        display_label = normalize_whitespace(str(row.get("dato") or row.get("display_label") or ""))
        if not display_label:
            continue
        status = normalize_whitespace(str(row.get("status") or "Ledig")) or "Ledig"
        explicit_start = row.get("start_date")
        explicit_end = row.get("end_date")
        if explicit_start:
            try:
                start_date = datetime.fromisoformat(str(explicit_start)).date()
            except ValueError:
                start_date = None
        else:
            start_date = None
        if explicit_end:
            try:
                end_date = datetime.fromisoformat(str(explicit_end)).date()
            except ValueError:
                end_date = None
        else:
            end_date = None

        if not start_date and not end_date:
            start_date, end_date = parse_course_date_range(display_label)

        normalized_rows.append(
            {
                "dato": display_label,
                "status": status,
                "start_date": start_date.isoformat() if start_date else None,
                "end_date": end_date.isoformat() if end_date else None,
                "sort_order": index,
            }
        )

    normalized_rows.sort(
        key=lambda row: (
            row.get("start_date") or row.get("end_date") or "9999-12-31",
            int(row.get("sort_order") or 0),
            row.get("dato") or "",
        )
    )
    return normalized_rows


def is_upcoming_course(row: dict[str, Any], today: date | None = None) -> bool:
    reference_today = today or date.today()
    end_value = row.get("end_date") or row.get("start_date")
    if not end_value:
        return True
    try:
        end_date = datetime.fromisoformat(str(end_value)).date()
    except ValueError:
        return True
    return end_date >= reference_today


def filter_upcoming_courses(rows: Any) -> list[dict[str, Any]]:
    normalized_rows = normalize_vtg_course_rows(rows)
    return [row for row in normalized_rows if is_upcoming_course(row)]


def get_invalid_vtg_course_labels(rows: Any) -> list[str]:
    if not isinstance(rows, list):
        return []

    invalid_labels: list[str] = []
    for row in rows:
        if not isinstance(row, dict):
            continue

        display_label = normalize_whitespace(str(row.get("dato") or row.get("display_label") or ""))
        if not display_label:
            continue

        explicit_start = row.get("start_date")
        explicit_end = row.get("end_date")
        start_date = None
        end_date = None

        if explicit_start:
            try:
                start_date = datetime.fromisoformat(str(explicit_start)).date()
            except ValueError:
                start_date = None
        if explicit_end:
            try:
                end_date = datetime.fromisoformat(str(explicit_end)).date()
            except ValueError:
                end_date = None

        if not start_date and not end_date:
            start_date, end_date = parse_course_date_range(display_label)

        if not start_date and not end_date:
            invalid_labels.append(display_label)

    return invalid_labels