"""AgroStar табличные форматы (PDF, xlsx) → те же AgrostarSample, что и XML."""

from __future__ import annotations

import re
from typing import Any

from app.lab.etl.agrostar_xml_import import (
    AgrostarParseResult,
    AgrostarSample,
    _parse_float,
    _pct_dm_to_g_per_kg_sv,
    finalize_agrostar_sample,
)

# Подпись в отчёте → ключ AgroStar XML (дальше — общий _AGROSTAR_MAP)
_TABULAR_LABEL_TO_AGRO: dict[str, str] = {
    "Сухое вещество (DM)": "DM",
    "Сырой протеин (Crude Protein)": "CP",
    "КДК (ADF)": "ADF",
    "НДК (aNDF)": "NDF",
    "НДК по орг. веществу (aNDFom)": "aNDFom",
    "Сырой жир (Fat EE)": "Fat_EE",
    "Сырая зола (Ash)": "Ash",
    "Кальций (Ca)": "Ca",
    "Фосфор (P)": "P",
    "Магний (Mg)": "Mg",
    "Калий (K)": "K",
    "Сера (S)": "S",
    "Хлор (Cl)": "Cl",
    "Водорастворимый сахар (WSC)": "Sugar_WSC",
    "Cпирторастворимый сахар (ESC)": "Sugar_ESC",
    "Крахмал (Starch)": "Starch",
    "Безволокнистые углеводы (NFC)": "NFC",
    "Переваримые питательные вещества (TDN)": "TDN",
    "НДК протеин (ND-ICP)": "NDICP_CP",
    "Лизин": "Lys",
    "Метионин": "Met",
    "Лейцин": "Leu",
    "Изолейцин": "Ile",
    "Валин": "Val",
    "Усваиваемость НДК 30 ч (NDFD30)": "NDFDom_IV_30hr",
}

_SECTION_HEADERS = frozenset(
    {
        "Протеин:",
        "Клетчатка:",
        "Сахара и крахмал:",
        "Жир и жирные кислоты:",
        "Минералы:",
        "Энергия:",
    }
)


def map_tabular_label(label: str) -> str | None:
    text = (label or "").strip()
    if not text or text in _SECTION_HEADERS:
        return None
    if text in _TABULAR_LABEL_TO_AGRO:
        return _TABULAR_LABEL_TO_AGRO[text]
    lowered = text.lower()
    for key, ag_key in sorted(_TABULAR_LABEL_TO_AGRO.items(), key=lambda item: len(item[0]), reverse=True):
        if key.lower() in lowered:
            return ag_key
    return None


def tabular_rows_to_agro_fields(rows: list[tuple[str, str]]) -> dict[str, float]:
    """[(label, raw_value), ...] → raw_fields AgroStar."""
    out: dict[str, float] = {}
    for label, raw in rows:
        ag_key = map_tabular_label(label)
        if not ag_key:
            continue
        val = _parse_float(raw)
        if val is not None:
            out[ag_key] = val
    return out


def sample_from_agro_fields(
    *,
    sample_no: str,
    farm_name: str,
    farm_id: str = "",
    date_printed: str = "",
    feed_type: str = "Mixed haylage",
    desc_1: str = "",
    desc_2: str = "",
    desc_3: str = "",
    agro_fields: dict[str, float],
) -> AgrostarSample:
    sample = AgrostarSample(
        sample_no=sample_no,
        farm_name=farm_name,
        farm_id=farm_id,
        date_printed=date_printed,
        feed_type=feed_type,
        desc_1=desc_1,
        desc_2=desc_2,
        desc_3=desc_3,
        dry_matter_pct=agro_fields.get("DM"),
    )
    sample.raw_fields = {k: v for k, v in agro_fields.items() if k != "DM"}
    return finalize_agrostar_sample(sample)


def parse_tabular_triplet_lines(lines: list[str]) -> list[tuple[str, str, str]]:
    """Строки PDF: name, unit (%DM|%CP|%), value."""
    rows: list[tuple[str, str, str]] = []
    i = 0
    units = frozenset({"%DM", "%CP", "%"})
    while i < len(lines):
        line = lines[i]
        if line in units and i >= 1:
            name = lines[i - 1]
            val = lines[i + 1] if i + 1 < len(lines) else ""
            if val and (val[0].isdigit() or val.startswith("<")):
                rows.append((name, line, val))
                i += 2
                continue
        i += 1
    return rows


def rows_to_label_values(triplets: list[tuple[str, str, str]]) -> list[tuple[str, str]]:
    return [(name, val) for name, _unit, val in triplets]