Aislo/B03_FileInput/B03_FileInput_Engine_Analyze.py

"""B03 원본 입력 파일 메타데이터 분석."""

import csv
import logging
import math
import re
from pathlib import Path
from threading import get_ident
from typing import Any

import laspy
import numpy as np
import rasterio
from pyproj import CRS

logger = logging.getLogger(__name__)

_CUSTOM_VERTICAL_AUTHORITY_PATTERN = re.compile(
    r',?\s*AUTHORITY\["EPSG","(9995|99999)"\]',
    re.IGNORECASE,
)
_CUSTOM_VERTICAL_WARNING_PATTERN = re.compile(
    r"proj_create_from_database: crs not found: EPSG:(9995|99999)",
    re.IGNORECASE,
)


class _CustomVerticalCrsWarningFilter(logging.Filter):
    """알려진 사설 수직 CRS 경고만 B03 안내 로그로 변환한다."""

    def __init__(self, source: Path) -> None:
        super().__init__()
        self.source = source
        self.thread_id = get_ident()
        self.logged_codes: set[str] = set()

    def filter(self, record: logging.LogRecord) -> bool:
        if get_ident() != self.thread_id:
            return True
        match = _CUSTOM_VERTICAL_WARNING_PATTERN.search(record.getMessage())
        if match is None:
            return True
        code = f"EPSG:{match.group(1)}"
        if code not in self.logged_codes:
            logger.info(
                "사용자 정의 수직 CRS 코드를 보존합니다: file=%s code=%s",
                self.source.name,
                code,
            )
            self.logged_codes.add(code)
        return False


def _component_metadata(crs: CRS) -> dict[str, Any]:
    """CRS 구성 요소를 JSON 저장 가능한 메타데이터로 변환한다."""
    authority = crs.to_authority()
    epsg = crs.to_epsg()
    return {
        "name": crs.name,
        "type": crs.type_name,
        "epsg": epsg,
        "authority": (
            {"name": authority[0], "code": authority[1]} if authority is not None else None
        ),
    }


def normalize_crs_metadata(crs: Any | None) -> dict[str, Any]:
    """복합 CRS를 수평·수직 구성 요소로 분리해 일관된 상태를 반환한다."""
    if crs is None:
        return {
            "crs": None,
            "epsg": None,
            "horizontal_crs": None,
            "vertical_crs": None,
            "crs_status": "missing_crs",
        }

    parsed = crs if isinstance(crs, CRS) else CRS.from_user_input(crs)
    components = parsed.sub_crs_list or [parsed]
    horizontal = next(
        (item for item in components if item.is_projected or item.is_geographic),
        None,
    )
    vertical = next((item for item in components if item.is_vertical), None)

    horizontal_metadata = _component_metadata(horizontal) if horizontal is not None else None
    vertical_metadata = _component_metadata(vertical) if vertical is not None else None
    horizontal_epsg = horizontal_metadata["epsg"] if horizontal_metadata is not None else None

    if horizontal_epsg is None:
        status = "unknown_horizontal_crs"
    elif vertical_metadata is not None and vertical_metadata["epsg"] is None:
        status = "custom_vertical_crs"
    else:
        status = "identified"

    return {
        "crs": crs.to_string(),
        "epsg": horizontal_epsg,
        "horizontal_crs": horizontal_metadata,
        "vertical_crs": vertical_metadata,
        "crs_status": status,
    }


def _log_crs_status(source: Path, metadata: dict[str, Any]) -> None:
    """정상화된 CRS 상태를 B03 도메인 로그로 남긴다."""
    if metadata["crs_status"] == "custom_vertical_crs":
        logger.info(
            "사용자 정의 수직 CRS를 보존합니다: file=%s horizontal_epsg=%s vertical=%s",
            source.name,
            metadata["epsg"],
            metadata["vertical_crs"]["name"],
        )
    elif metadata["crs_status"] == "unknown_horizontal_crs":
        logger.warning("수평 CRS를 EPSG로 식별하지 못했습니다: file=%s", source.name)


def _prepare_prj_wkt(text: str) -> tuple[str, list[str]]:
    """KNGeoid24 사설 EPSG 표식만 파싱용 WKT에서 분리한다."""
    if "KNGeoid24" not in text:
        return text, []
    codes = sorted({f"EPSG:{code}" for code in _CUSTOM_VERTICAL_AUTHORITY_PATTERN.findall(text)})
    return _CUSTOM_VERTICAL_AUTHORITY_PATTERN.sub("", text), codes


def analyze_las_metadata(path: str | Path) -> dict[str, Any]:
    """LAS/LAZ 헤더와 분류 통계를 메모리에 전체 적재하지 않고 분석한다."""
    source = Path(path)
    with laspy.open(source) as las_file:
        header = las_file.header
        point_format = header.point_format
        dimension_names = list(point_format.dimension_names)
        point_count = int(header.point_count)
        crs = header.parse_crs()
        crs_metadata = normalize_crs_metadata(crs)
        _log_crs_status(source, crs_metadata)
        metadata: dict[str, Any] = {
            "file": source.name,
            "version": f"{header.version.major}.{header.version.minor}",
            "point_format": {
                "id": point_format.id,
                "dimensions": dimension_names,
            },
            "point_count": point_count,
            "bounds": {
                "x": [float(header.mins[0]), float(header.maxs[0])],
                "y": [float(header.mins[1]), float(header.maxs[1])],
                "z": [float(header.mins[2]), float(header.maxs[2])],
            },
            "scale": [float(value) for value in header.scales],
            "offset": [float(value) for value in header.offsets],
            "has_crs": crs is not None,
            **crs_metadata,
            "has_classification": "classification" in dimension_names,
            "has_rgb": all(name in dimension_names for name in ("red", "green", "blue")),
            "has_intensity": "intensity" in dimension_names,
            "has_return_number": "return_number" in dimension_names,
        }

        if metadata["has_classification"] and point_count > 0:
            classification_counts: dict[int, int] = {}
            for chunk in las_file.chunk_iterator(500_000):
                values, counts = np.unique(
                    np.asarray(chunk.classification, dtype=np.uint8),
                    return_counts=True,
                )
                for value, count in zip(values.tolist(), counts.tolist(), strict=True):
                    classification_counts[value] = classification_counts.get(value, 0) + count
            metadata["classification_summary"] = {
                str(key): value for key, value in sorted(classification_counts.items())
            }

        return metadata


def analyze_prj_metadata(path: str | Path) -> dict[str, Any]:
    """PRJ WKT에서 좌표계 식별자와 명칭을 추출한다."""
    source = Path(path)
    text = source.read_text(encoding="utf-8", errors="replace").strip()
    metadata: dict[str, Any] = {
        "file": source.name,
        "text_preview": text[:500],
        "epsg": None,
        "name": None,
        "authority": None,
        "is_valid": False,
    }
    if not text:
        metadata["error"] = "PRJ 파일이 비어 있습니다."
        return metadata

    parse_text, custom_authority_codes = _prepare_prj_wkt(text)
    try:
        crs = CRS.from_wkt(parse_text)
    except Exception as exc:
        metadata["error"] = str(exc)
        return metadata

    crs_metadata = normalize_crs_metadata(crs)
    _log_crs_status(source, crs_metadata)
    metadata.update(
        {
            **crs_metadata,
            "name": crs.name,
            "authority": crs.to_authority(),
            "custom_authority_codes": custom_authority_codes,
            "is_valid": True,
        }
    )
    return metadata


def analyze_tfw_metadata(path: str | Path) -> dict[str, Any]:
    """TFW의 affine 변환 계수와 유효성을 분석한다."""
    source = Path(path)
    values = [
        float(line.strip())
        for line in source.read_text(encoding="utf-8", errors="replace").splitlines()
        if line.strip()
    ]
    if any(not math.isfinite(value) for value in values):
        raise ValueError("TFW 변환 계수는 유한한 숫자여야 합니다.")

    return {
        "file": source.name,
        "values": values,
        "pixel_size_x": values[0] if len(values) > 0 else None,
        "rotation_y": values[1] if len(values) > 1 else None,
        "rotation_x": values[2] if len(values) > 2 else None,
        "pixel_size_y": values[3] if len(values) > 3 else None,
        "origin_x": values[4] if len(values) > 4 else None,
        "origin_y": values[5] if len(values) > 5 else None,
        "is_valid": len(values) == 6,
    }


def analyze_tif_metadata(path: str | Path) -> dict[str, Any]:
    """TIF/GeoTIFF 데이터셋의 공간 및 밴드 메타데이터를 분석한다."""
    source = Path(path)
    rasterio_logger = logging.getLogger("rasterio._env")
    warning_filter = _CustomVerticalCrsWarningFilter(source)
    rasterio_logger.addFilter(warning_filter)
    try:
        with rasterio.open(source) as dataset:
            crs = dataset.crs
            crs_metadata = normalize_crs_metadata(crs)
            _log_crs_status(source, crs_metadata)
            bounds = dataset.bounds
            return {
                "file": source.name,
                "width": int(dataset.width),
                "height": int(dataset.height),
                "count": int(dataset.count),
                "dtypes": list(dataset.dtypes),
                "nodata": float(dataset.nodata) if dataset.nodata is not None else None,
                **crs_metadata,
                "bounds": {
                    "left": float(bounds.left),
                    "bottom": float(bounds.bottom),
                    "right": float(bounds.right),
                    "top": float(bounds.top),
                },
                "transform": [float(value) for value in list(dataset.transform)[:6]],
                "resolution": [float(value) for value in dataset.res],
                "likely_type": "dem" if dataset.count == 1 else "image",
            }
    finally:
        rasterio_logger.removeFilter(warning_filter)


_PLANNED_ROUTE_COLUMNS = ("route_name", "sequence", "x", "y", "z", "crs_epsg")


def _parse_route_integer(value: str, *, field: str, row_number: int) -> int:
    normalized = value.strip()
    if not re.fullmatch(r"[0-9]+", normalized):
        raise ValueError(f"CSV {row_number}행의 {field} 값은 양의 정수여야 합니다.")
    parsed = int(normalized)
    if parsed <= 0:
        raise ValueError(f"CSV {row_number}행의 {field} 값은 양의 정수여야 합니다.")
    return parsed


def _parse_route_coordinate(value: str, *, field: str, row_number: int) -> float:
    try:
        parsed = float(value.strip())
    except (AttributeError, ValueError) as exc:
        raise ValueError(f"CSV {row_number}행의 {field} 값은 숫자여야 합니다.") from exc
    if not math.isfinite(parsed):
        raise ValueError(f"CSV {row_number}행의 {field} 값은 유한한 숫자여야 합니다.")
    return parsed


def analyze_planned_route_csv(path: str | Path) -> dict[str, Any]:
    """원청 계획노선 CSV를 검증하고 경로 메타데이터를 반환한다."""
    source = Path(path)
    with source.open("r", encoding="utf-8-sig", newline="") as csv_file:
        reader = csv.DictReader(csv_file)
        if reader.fieldnames is None:
            raise ValueError("계획노선 CSV 헤더를 찾을 수 없습니다.")

        normalized_headers = [header.strip() for header in reader.fieldnames]
        if len(set(normalized_headers)) != len(normalized_headers):
            raise ValueError("계획노선 CSV 헤더에 중복된 열이 있습니다.")
        header_map = dict(zip(normalized_headers, reader.fieldnames, strict=True))
        missing = [column for column in _PLANNED_ROUTE_COLUMNS if column not in header_map]
        if missing:
            raise ValueError(f"계획노선 CSV 필수 열이 없습니다: {', '.join(missing)}")

        route_name: str | None = None
        crs_epsg: int | None = None
        points: list[tuple[float, float, float]] = []
        for expected_sequence, row in enumerate(reader, start=1):
            row_number = expected_sequence + 1
            current_name = (row.get(header_map["route_name"]) or "").strip()
            if not current_name:
                raise ValueError(f"CSV {row_number}행의 route_name 값이 비어 있습니다.")
            if route_name is None:
                route_name = current_name
            elif current_name != route_name:
                raise ValueError("계획노선 CSV에는 하나의 route_name만 사용할 수 있습니다.")

            sequence = _parse_route_integer(
                row.get(header_map["sequence"]) or "",
                field="sequence",
                row_number=row_number,
            )
            if sequence != expected_sequence:
                raise ValueError(
                    f"CSV {row_number}행의 sequence는 {expected_sequence}이어야 합니다."
                )

            current_epsg = _parse_route_integer(
                row.get(header_map["crs_epsg"]) or "",
                field="crs_epsg",
                row_number=row_number,
            )
            if crs_epsg is None:
                crs_epsg = current_epsg
            elif current_epsg != crs_epsg:
                raise ValueError("계획노선 CSV의 crs_epsg는 모든 행에서 같아야 합니다.")

            points.append(
                tuple(
                    _parse_route_coordinate(
                        row.get(header_map[field]) or "",
                        field=field,
                        row_number=row_number,
                    )
                    for field in ("x", "y", "z")
                )
            )

    if len(points) < 2:
        raise ValueError("계획노선 CSV에는 좌표가 2개 이상 있어야 합니다.")

    xs, ys, zs = zip(*points, strict=True)
    return {
        "file": source.name,
        "extension": "csv",
        "size_bytes": source.stat().st_size,
        "purpose": "planned_route",
        "route_name": route_name,
        "point_count": len(points),
        "epsg": crs_epsg,
        "columns": list(_PLANNED_ROUTE_COLUMNS),
        "bounds": {
            "x_min": min(xs),
            "x_max": max(xs),
            "y_min": min(ys),
            "y_max": max(ys),
            "z_min": min(zs),
            "z_max": max(zs),
        },
        "start_point": list(points[0]),
        "end_point": list(points[-1]),
    }


def analyze_input_metadata(path: str | Path) -> dict[str, Any]:
    """입력 파일 확장자에 맞는 B03 메타데이터 분석 함수를 호출한다."""
    source = Path(path)
    extension = source.suffix.lower()
    if extension == ".csv":
        return analyze_planned_route_csv(source)
    if extension in {".las", ".laz"}:
        return analyze_las_metadata(source)
    if extension == ".prj":
        return analyze_prj_metadata(source)
    if extension == ".tfw":
        return analyze_tfw_metadata(source)
    if extension in {".tif", ".tiff"}:
        return analyze_tif_metadata(source)
    return {
        "file": source.name,
        "extension": extension.lstrip("."),
        "size_bytes": source.stat().st_size,
    }