kadastr-17/read_mif_file.py

from dataclasses import dataclass, field
from pathlib import Path
from typing import List, Dict, Any, Optional, Tuple, Union


@dataclass
class MifFile:
    name: str
    path: str
    headers: Dict[str, Any] = field(default_factory=dict)
    features: List[Dict[str, Any]] = field(default_factory=list)


def _unquote(s: str) -> str:
	s = s.strip()
	if s.startswith('"') and s.endswith('"'):
		return s[1:-1]
	return s


def parse_mif(file_path: str) -> Union[MifFile, List[MifFile]]:
	"""
	Примитивный парсер MIF-файла. Обрабатывает заголовки (Version, Charset, Delimiter,
	CoordSys, Columns) и простые объекты в блоке Data: Region, Text, Center/Centre, Pen, Brush.

	Не претендует на полную спецификацию MapInfo MIF, но удобен для быстрого извлечения структур.
	"""
	p = Path(file_path)
	if not p.exists():
		raise FileNotFoundError(file_path)

	# If a directory is passed, parse all .mif files inside and return a list
	if p.is_dir():
		return load_mif_files(str(p))

	mif = MifFile(name=p.name, path=str(p.resolve()))
	lines = [ln.rstrip('\n') for ln in p.read_text(encoding='utf-8', errors='replace').splitlines()]

	i = 0
	n = len(lines)
	# parse header until "Data"
	while i < n:
		line = lines[i].strip()
		if not line:
			i += 1
			continue
		up = line.upper()
		if up.startswith('VERSION'):
			mif.headers['version'] = line.split(None, 1)[1].strip() if len(line.split()) > 1 else None
			i += 1
			continue
		if up.startswith('CHARSET'):
			# Charset "WindowsCyrillic"
			rest = line.split(None, 1)[1] if len(line.split()) > 1 else ''
			mif.headers['charset'] = _unquote(rest)
			i += 1
			continue
		if up.startswith('DELIMITER'):
			rest = line.split(None, 1)[1] if len(line.split()) > 1 else ''
			mif.headers['delimiter'] = _unquote(rest)
			i += 1
			continue
		if up.startswith('COORDSYS'):
			mif.headers['coordsys'] = line[len('CoordSys'):].strip()
			i += 1
			continue
		if up.startswith('COLUMNS'):
			# next N lines are column definitions
			parts = line.split()
			cnt = int(parts[1]) if len(parts) > 1 else 0
			i += 1
			cols: List[str] = []
			for _ in range(cnt):
				if i >= n:
					break
				cols.append(lines[i].strip())
				i += 1
			mif.headers['columns'] = cols
			continue
		if up == 'DATA':
			i += 1
			break
		i += 1

	# parse data features
	while i < n:
		raw = lines[i].lstrip()
		if not raw:
			i += 1
			continue
		parts = raw.split(None, 1)
		typ = parts[0]
		if typ.upper() == 'REGION':
			feat: Dict[str, Any] = {'type': 'Region', 'props': {}}
			# Region <N_polygons> — количество контуров (полигонов) в объекте
			try:
				n_polygons = int(parts[1]) if len(parts) > 1 else 1
			except (ValueError, IndexError):
				n_polygons = 1
			i += 1

			all_points: List[Tuple[float, float]] = []
			contours: List[List[Tuple[float, float]]] = []

			for _poly in range(n_polygons):
				# Первая строка каждого контура — количество точек
				if i >= n:
					break
				cnt_line = lines[i].strip()
				try:
					pts_cnt = int(cnt_line)
					i += 1
				except ValueError:
					pts_cnt = 0

				poly_pts: List[Tuple[float, float]] = []
				for _ in range(pts_cnt):
					if i >= n:
						break
					xy = lines[i].strip().split()
					if len(xy) >= 2:
						try:
							poly_pts.append((float(xy[0]), float(xy[1])))
						except ValueError:
							pass
					i += 1

				if poly_pts:
					contours.append(poly_pts)
					all_points.extend(poly_pts)

			# Если один контур — points == список точек (как было раньше)
			# Если несколько — points == точки всех контуров подряд, contours — отдельно
			feat['props']['points'] = all_points
			feat['props']['contours'] = contours  # List[List[(x,y)]]

			# Читаем атрибуты (Pen, Brush, Center) до начала следующего объекта
			while i < n:
				t = lines[i].lstrip()
				if not t:
					i += 1
					continue
				up2 = t.upper()
				if up2.startswith('PEN'):
					feat['props']['pen'] = t[len('Pen'):].strip()
					i += 1
					continue
				if up2.startswith('BRUSH'):
					feat['props']['brush'] = t[len('Brush'):].strip()
					i += 1
					continue
				if up2.startswith('CENTER') or up2.startswith('CENTRE'):
					rest = t.split(None, 1)[1] if len(t.split()) > 1 else ''
					coords = rest.split()
					try:
						cx = float(coords[0]); cy = float(coords[1])
						feat['props']['center'] = (cx, cy)
					except Exception:
						pass
					i += 1
					continue
				if up2.split()[0] in ('REGION', 'TEXT', 'POINT', 'PLINE', 'LINE'):
					break
				i += 1

			mif.features.append(feat)
			continue

		if typ.upper() == 'TEXT':
			feat = {'type': 'Text', 'props': {}}
			# next line: quoted string
			i += 1
			if i < n:
				txt_line = lines[i].strip()
				feat['props']['text'] = _unquote(txt_line)
				i += 1
			# next line: bounding coordinates (often 4 numbers)
			if i < n:
				coords = lines[i].strip().split()
				try:
					nums = [float(x) for x in coords]
					feat['props']['bbox'] = nums
				except Exception:
					pass
				i += 1
			# optional Font line
			if i < n and lines[i].lstrip().upper().startswith('FONT'):
				feat['props']['font'] = lines[i].lstrip()[len('Font'):].strip()
				i += 1

			mif.features.append(feat)
			continue

		# simple POINT parser
		if typ.upper() == 'POINT':
			feat = {'type': 'Point', 'props': {}}
			rest = parts[1] if len(parts) > 1 else ''
			coords = rest.split()
			try:
				feat['props']['xy'] = (float(coords[0]), float(coords[1]))
			except Exception:
				feat['props']['xy'] = None
			i += 1
			mif.features.append(feat)
			continue

		# unknown / unsupported object: skip one line
		i += 1

	return mif


def load_mif_files(dir_path: str) -> List[MifFile]:
	"""Найти все .mif файлы в директории (рекурсивно) и распарсить их в объекты MifFile."""
	root = Path(dir_path)
	if not root.exists():
		raise FileNotFoundError(dir_path)

	mif_files: List[MifFile] = []
	for p in root.rglob('*.mif'):
		try:
			mif_files.append(parse_mif(str(p)))
		except Exception:
			# при ошибке парсинга — пропустить файл
			continue
	return mif_files


__all__ = ["MifFile", "parse_mif", "load_mif_files"]