site/WESP_REL/app/lab/etl/lab_import_router.py

"""Единый роутер импорта lab: XML / PDF / xlsx → preview API."""

from __future__ import annotations

import re
from typing import Any

from app.lab.etl.agrostar_pdf_import import looks_like_agrostar_pdf, parse_agrostar_pdf
from app.lab.etl.agrostar_xlsx_import import parse_agrostar_xlsx
from app.lab.etl.agrostar_xml_import import AgrostarParseResult, enrich_parse_with_matches, parse_agrostar_xml
from app.lab.etl.plinor_pdf_import import (
    enrich_plinor_composition,
    is_plinor_sos,
    is_plinor_zoo,
    parse_plinor_sos,
    parse_plinor_zoo,
)

SOURCE_LABELS: dict[str, str] = {
    "agrostar_xml": "AgroStar XML",
    "agrostar_pdf": "AgroStar PDF",
    "agrostar_xlsx": "AgroStar Excel",
    "plinor_sos": "ПЛИНОР — состав рациона",
    "plinor_zoo": "ПЛИНОР — показатели рациона",
}

_PDF_MAGIC = b"%PDF"
_XLSX_MAGIC = b"PK\x03\x04"
_XML_MARKERS = (b"Standard_XML_Data", b"<?xml", b"<Standard_XML_Data")


def _filename_hint_plinor_zoo(name: str) -> bool:
    n = (name or "").lower()
    return "зоо" in n or re.search(r"(^|[^a-z])zoo([^a-z]|$)", n) is not None


def _filename_hint_plinor_sos(name: str) -> bool:
    n = (name or "").lower()
    return "сос" in n or re.search(r"(^|[^a-z])sos([^a-z]|$)", n) is not None


def _looks_like_xml_agrostar(data: bytes) -> bool:
    head = data[:16384]
    return any(marker in head for marker in _XML_MARKERS)


def _looks_like_xlsx(data: bytes) -> bool:
    return data.startswith(_XLSX_MAGIC)


def _looks_like_pdf(data: bytes) -> bool:
    return data.startswith(_PDF_MAGIC)


def _pdf_import_available() -> bool:
    try:
        import fitz  # noqa: F401
    except ImportError:
        return False
    return True


def detect_source_format(filename: str, data: bytes) -> str:
    """Определение формата по содержимому и имени (расширение — запасной вариант)."""
    name = (filename or "").lower()

    if _looks_like_pdf(data):
        return _detect_pdf_format(data, name)
    if _looks_like_xlsx(data):
        return "agrostar_xlsx"
    if _looks_like_xml_agrostar(data):
        return "agrostar_xml"

    if name.endswith(".xml"):
        return "agrostar_xml"
    if name.endswith(".xlsx") or name.endswith(".xls"):
        return "agrostar_xlsx"
    if name.endswith(".pdf"):
        return _detect_pdf_format(data, name)
    return "unknown"


def _detect_pdf_format(data: bytes, name: str) -> str:
    head = data[:8192]
    if b"Standard_XML_Data" in head:
        return "agrostar_xml"
    if looks_like_agrostar_pdf(data):
        return "agrostar_pdf"

    pdf_text = ""
    if _pdf_import_available():
        try:
            pdf_text = _peek_pdf_text(data)
        except Exception:  # noqa: BLE001
            pdf_text = ""

    if pdf_text:
        if is_plinor_zoo(pdf_text):
            return "plinor_zoo"
        if is_plinor_sos(pdf_text):
            return "plinor_sos"
        if looks_like_agrostar_pdf(data):
            return "agrostar_pdf"

    if _filename_hint_plinor_zoo(name):
        return "plinor_zoo"
    if _filename_hint_plinor_sos(name):
        return "plinor_sos"
    return "unknown_pdf"


def _peek_pdf_text(data: bytes) -> str:
    try:
        import fitz
    except ImportError as exc:
        raise RuntimeError("Для PDF нужен pymupdf (pip install pymupdf)") from exc
    doc = fitz.open(stream=data, filetype="pdf")
    text = doc[0].get_text() if doc.page_count else ""
    doc.close()
    return text


def parse_lab_import(filename: str, data: bytes) -> dict[str, Any]:
    """Разбор файла → единый JSON для /lab комбайна."""
    if not data:
        return {"error": True, "message": "Пустой файл", "kind": "error"}

    fmt = detect_source_format(filename, data)
    source_label = SOURCE_LABELS.get(fmt, "Неизвестный формат")

    if fmt == "agrostar_xml":
        text = _decode_text(data)
        parsed = parse_agrostar_xml(text)
        if parsed.errors and not parsed.samples:
            return {
                "error": True,
                "message": "; ".join(parsed.errors),
                "kind": "error",
                "sourceFormat": fmt,
                **parsed.to_api_dict(),
            }
        body = enrich_parse_with_matches(parsed)
        body["kind"] = "lab_samples"
        body["sourceFormat"] = fmt
        body["sourceLabel"] = source_label
        body["fileName"] = filename
        if parsed.errors:
            body["warnings"] = parsed.errors
        return body

    if fmt == "agrostar_pdf":
        parsed = parse_agrostar_pdf(data)
        return _lab_samples_response(parsed, fmt, source_label, filename)

    if fmt == "agrostar_xlsx":
        parsed = parse_agrostar_xlsx(data)
        return _lab_samples_response(parsed, fmt, source_label, filename)

    if fmt == "plinor_sos":
        parsed = parse_plinor_sos(data)
        if parsed.errors and not parsed.feed_lines:
            return {
                "error": True,
                "message": "; ".join(parsed.errors),
                "kind": "error",
                "sourceFormat": fmt,
            }
        body = enrich_plinor_composition(parsed.to_api_dict())
        body["kind"] = "ration_composition"
        body["sourceFormat"] = fmt
        body["sourceLabel"] = source_label
        body["fileName"] = filename
        if parsed.errors:
            body["warnings"] = parsed.errors
        return body

    if fmt == "plinor_zoo":
        parsed = parse_plinor_zoo(data)
        if parsed.errors and not parsed.indicators:
            return {
                "error": True,
                "message": "; ".join(parsed.errors),
                "kind": "error",
                "sourceFormat": fmt,
            }
        body = parsed.to_api_dict()
        body["kind"] = "ration_indicators"
        body["sourceFormat"] = fmt
        body["sourceLabel"] = source_label
        body["fileName"] = filename
        if parsed.errors:
            body["warnings"] = parsed.errors
        return body

    if fmt == "unknown_pdf" and _looks_like_pdf(data) and not _pdf_import_available():
        return {
            "error": True,
            "message": "Для PDF нужен pymupdf — выполните: pip install pymupdf",
            "kind": "error",
            "sourceFormat": fmt,
            "fileName": filename,
        }

    return {
        "error": True,
        "message": "Формат не распознан. Поддерживаются: AgroStar xml/pdf/xlsx, ПЛИНОР сос/зоо pdf",
        "kind": "error",
        "sourceFormat": fmt,
        "fileName": filename,
    }


def _lab_samples_response(
    parsed: AgrostarParseResult,
    fmt: str,
    source_label: str,
    filename: str,
) -> dict[str, Any]:
    if parsed.errors and not parsed.samples:
        return {
            "error": True,
            "message": "; ".join(parsed.errors),
            "kind": "error",
            "sourceFormat": fmt,
            **parsed.to_api_dict(),
        }
    body = enrich_parse_with_matches(parsed)
    body["kind"] = "lab_samples"
    body["sourceFormat"] = fmt
    body["sourceLabel"] = source_label
    body["fileName"] = filename
    if parsed.errors:
        body["warnings"] = parsed.errors
    return body


def _decode_text(data: bytes) -> str:
    for encoding in ("utf-8", "utf-8-sig", "cp1251"):
        try:
            return data.decode(encoding)
        except UnicodeDecodeError:
            continue
    return data.decode("utf-8", errors="replace")