valetudo-map-parser 0.1.7__py3-none-any.whl → 0.1.9__py3-none-any.whl

valetudo_map_parser/config/optimized_element_map.py

@@ -0,0 +1,406 @@
+"""
+Optimized Element Map Generator.
+Uses scipy for efficient element map generation and processing.
+Version: 0.1.9
+"""
+
+from __future__ import annotations
+
+import logging
+import numpy as np
+from scipy import ndimage
+
+from .drawable_elements import DrawableElement, DrawingConfig
+from .types import LOGGER
+
+
+class OptimizedElementMapGenerator:
+    """Class for generating 2D element maps from JSON data with optimized performance.
+
+    This class creates a 2D array where each cell contains an integer code
+    representing the element at that position (floor, wall, room, etc.).
+    It uses scipy for efficient processing and supports sparse matrices for memory efficiency.
+    """
+
+    def __init__(self, drawing_config: DrawingConfig = None, shared_data=None):
+        """Initialize the optimized element map generator.
+
+        Args:
+            drawing_config: Optional drawing configuration for element properties
+            shared_data: Shared data object for accessing common resources
+        """
+        self.drawing_config = drawing_config or DrawingConfig()
+        self.shared = shared_data
+        self.element_map = None
+        self.element_map_shape = None
+        self.scale_info = None
+        self.file_name = (
+            getattr(shared_data, "file_name", "ElementMap")
+            if shared_data
+            else "ElementMap"
+        )
+
+    async def async_generate_from_json(self, json_data, existing_element_map=None):
+        """Generate a 2D element map from JSON data with optimized performance.
+
+        Args:
+            json_data: The JSON data from the vacuum
+            existing_element_map: Optional pre-created element map to populate
+
+        Returns:
+            numpy.ndarray: The 2D element map
+        """
+        if not self.shared:
+            LOGGER.warning("Shared data not provided, some features may not work.")
+            return None
+
+        # Use existing element map if provided
+        if existing_element_map is not None:
+            self.element_map = existing_element_map
+            return existing_element_map
+
+        # Detect JSON format
+        is_valetudo = "layers" in json_data and "pixelSize" in json_data
+        is_rand256 = "map_data" in json_data
+
+        if not (is_valetudo or is_rand256):
+            LOGGER.error("Unknown JSON format, cannot generate element map")
+            return None
+
+        if is_valetudo:
+            return await self._generate_valetudo_element_map(json_data)
+        elif is_rand256:
+            return await self._generate_rand256_element_map(json_data)
+
+    async def _generate_valetudo_element_map(self, json_data):
+        """Generate an element map from Valetudo format JSON data."""
+        # Get map dimensions from the JSON data
+        size_x = json_data["size"]["x"]
+        size_y = json_data["size"]["y"]
+        pixel_size = json_data["pixelSize"]
+
+        # Calculate downscale factor based on pixel size
+        # Standard pixel size is 5mm, so adjust accordingly
+        downscale_factor = max(1, pixel_size // 5 * 2)  # More aggressive downscaling
+
+        # Calculate dimensions for the downscaled map
+        map_width = max(100, size_x // (pixel_size * downscale_factor))
+        map_height = max(100, size_y // (pixel_size * downscale_factor))
+
+        LOGGER.info(
+            "%s: Creating optimized element map with dimensions: %dx%d (downscale factor: %d)",
+            self.file_name,
+            map_width,
+            map_height,
+            downscale_factor,
+        )
+
+        # Create the element map at the reduced size
+        element_map = np.zeros((map_height, map_width), dtype=np.int32)
+        element_map[:] = DrawableElement.FLOOR
+
+        # Store scaling information for coordinate conversion
+        self.scale_info = {
+            "original_size": (size_x, size_y),
+            "map_size": (map_width, map_height),
+            "scale_factor": downscale_factor * pixel_size,
+            "pixel_size": pixel_size,
+        }
+
+        # Process layers at the reduced resolution
+        for layer in json_data.get("layers", []):
+            layer_type = layer.get("type")
+
+            # Process rooms (segments)
+            if layer_type == "segment":
+                # Get room ID
+                meta_data = layer.get("metaData", {})
+                segment_id = meta_data.get("segmentId")
+
+                if segment_id is not None:
+                    # Convert segment_id to int if it's a string
+                    segment_id_int = (
+                        int(segment_id) if isinstance(segment_id, str) else segment_id
+                    )
+                    if 1 <= segment_id_int <= 15:
+                        room_element = getattr(
+                            DrawableElement, f"ROOM_{segment_id_int}", None
+                        )
+
+                        # Skip if room is disabled
+                        if room_element is None or not self.drawing_config.is_enabled(
+                            room_element
+                        ):
+                            continue
+
+                        # Create a temporary high-resolution mask for this room
+                        temp_mask = np.zeros(
+                            (size_y // pixel_size, size_x // pixel_size), dtype=np.uint8
+                        )
+
+                        # Process pixels for this room
+                        compressed_pixels = layer.get("compressedPixels", [])
+                        if compressed_pixels:
+                            # Process in chunks of 3 (x, y, count)
+                            for i in range(0, len(compressed_pixels), 3):
+                                if i + 2 < len(compressed_pixels):
+                                    x = compressed_pixels[i]
+                                    y = compressed_pixels[i + 1]
+                                    count = compressed_pixels[i + 2]
+
+                                    # Set pixels in the high-resolution mask
+                                    for j in range(count):
+                                        px = x + j
+                                        if (
+                                            0 <= y < temp_mask.shape[0]
+                                            and 0 <= px < temp_mask.shape[1]
+                                        ):
+                                            temp_mask[y, px] = 1
+
+                        # Use scipy to downsample the mask efficiently
+                        # This preserves the room shape better than simple decimation
+                        downsampled_mask = ndimage.zoom(
+                            temp_mask,
+                            (
+                                map_height / temp_mask.shape[0],
+                                map_width / temp_mask.shape[1],
+                            ),
+                            order=0,  # Nearest neighbor interpolation
+                        )
+
+                        # Apply the downsampled mask to the element map
+                        element_map[downsampled_mask > 0] = room_element
+
+                        # Clean up
+                        del temp_mask, downsampled_mask
+
+            # Process walls similarly
+            elif layer_type == "wall" and self.drawing_config.is_enabled(
+                DrawableElement.WALL
+            ):
+                # Create a temporary high-resolution mask for walls
+                temp_mask = np.zeros(
+                    (size_y // pixel_size, size_x // pixel_size), dtype=np.uint8
+                )
+
+                # Process compressed pixels for walls
+                compressed_pixels = layer.get("compressedPixels", [])
+                if compressed_pixels:
+                    # Process in chunks of 3 (x, y, count)
+                    for i in range(0, len(compressed_pixels), 3):
+                        if i + 2 < len(compressed_pixels):
+                            x = compressed_pixels[i]
+                            y = compressed_pixels[i + 1]
+                            count = compressed_pixels[i + 2]
+
+                            # Set pixels in the high-resolution mask
+                            for j in range(count):
+                                px = x + j
+                                if (
+                                    0 <= y < temp_mask.shape[0]
+                                    and 0 <= px < temp_mask.shape[1]
+                                ):
+                                    temp_mask[y, px] = 1
+
+                # Use scipy to downsample the mask efficiently
+                downsampled_mask = ndimage.zoom(
+                    temp_mask,
+                    (map_height / temp_mask.shape[0], map_width / temp_mask.shape[1]),
+                    order=0,
+                )
+
+                # Apply the downsampled mask to the element map
+                # Only overwrite floor pixels, not room pixels
+                wall_mask = (downsampled_mask > 0) & (
+                    element_map == DrawableElement.FLOOR
+                )
+                element_map[wall_mask] = DrawableElement.WALL
+
+                # Clean up
+                del temp_mask, downsampled_mask
+
+        # Store the element map
+        self.element_map = element_map
+        self.element_map_shape = element_map.shape
+
+        LOGGER.info(
+            "%s: Element map generation complete with shape: %s",
+            self.file_name,
+            element_map.shape,
+        )
+        return element_map
+
+    async def _generate_rand256_element_map(self, json_data):
+        """Generate an element map from Rand256 format JSON data."""
+        # Get map dimensions from the Rand256 JSON data
+        map_data = json_data["map_data"]
+        size_x = map_data["dimensions"]["width"]
+        size_y = map_data["dimensions"]["height"]
+
+        # Calculate downscale factor
+        downscale_factor = max(
+            1, min(size_x, size_y) // 500
+        )  # Target ~500px in smallest dimension
+
+        # Calculate dimensions for the downscaled map
+        map_width = max(100, size_x // downscale_factor)
+        map_height = max(100, size_y // downscale_factor)
+
+        LOGGER.info(
+            "%s: Creating optimized Rand256 element map with dimensions: %dx%d (downscale factor: %d)",
+            self.file_name,
+            map_width,
+            map_height,
+            downscale_factor,
+        )
+
+        # Create the element map at the reduced size
+        element_map = np.zeros((map_height, map_width), dtype=np.int32)
+        element_map[:] = DrawableElement.FLOOR
+
+        # Store scaling information for coordinate conversion
+        self.scale_info = {
+            "original_size": (size_x, size_y),
+            "map_size": (map_width, map_height),
+            "scale_factor": downscale_factor,
+            "pixel_size": 1,  # Rand256 uses 1:1 pixel mapping
+        }
+
+        # Process rooms
+        if "rooms" in map_data and map_data["rooms"]:
+            for room in map_data["rooms"]:
+                # Get room ID and check if it's enabled
+                room_id_int = room["id"]
+
+                # Get room element code (ROOM_1, ROOM_2, etc.)
+                room_element = None
+                if 0 < room_id_int <= 15:
+                    room_element = getattr(DrawableElement, f"ROOM_{room_id_int}", None)
+
+                # Skip if room is disabled
+                if room_element is None or not self.drawing_config.is_enabled(
+                    room_element
+                ):
+                    continue
+
+                if "coordinates" in room:
+                    # Create a high-resolution mask for this room
+                    temp_mask = np.zeros((size_y, size_x), dtype=np.uint8)
+
+                    # Fill the mask with room coordinates
+                    for coord in room["coordinates"]:
+                        x, y = coord
+                        if 0 <= y < size_y and 0 <= x < size_x:
+                            temp_mask[y, x] = 1
+
+                    # Use scipy to downsample the mask efficiently
+                    downsampled_mask = ndimage.zoom(
+                        temp_mask,
+                        (map_height / size_y, map_width / size_x),
+                        order=0,  # Nearest neighbor interpolation
+                    )
+
+                    # Apply the downsampled mask to the element map
+                    element_map[downsampled_mask > 0] = room_element
+
+                    # Clean up
+                    del temp_mask, downsampled_mask
+
+        # Process walls
+        if (
+            "walls" in map_data
+            and map_data["walls"]
+            and self.drawing_config.is_enabled(DrawableElement.WALL)
+        ):
+            # Create a high-resolution mask for walls
+            temp_mask = np.zeros((size_y, size_x), dtype=np.uint8)
+
+            # Fill the mask with wall coordinates
+            for coord in map_data["walls"]:
+                x, y = coord
+                if 0 <= y < size_y and 0 <= x < size_x:
+                    temp_mask[y, x] = 1
+
+            # Use scipy to downsample the mask efficiently
+            downsampled_mask = ndimage.zoom(
+                temp_mask, (map_height / size_y, map_width / size_x), order=0
+            )
+
+            # Apply the downsampled mask to the element map
+            # Only overwrite floor pixels, not room pixels
+            wall_mask = (downsampled_mask > 0) & (element_map == DrawableElement.FLOOR)
+            element_map[wall_mask] = DrawableElement.WALL
+
+            # Clean up
+            del temp_mask, downsampled_mask
+
+        # Store the element map
+        self.element_map = element_map
+        self.element_map_shape = element_map.shape
+
+        LOGGER.info(
+            "%s: Rand256 element map generation complete with shape: %s",
+            self.file_name,
+            element_map.shape,
+        )
+        return element_map
+
+    def map_to_element_coordinates(self, x, y):
+        """Convert map coordinates to element map coordinates."""
+        if not hasattr(self, "scale_info"):
+            return x, y
+
+        scale = self.scale_info["scale_factor"]
+        return int(x / scale), int(y / scale)
+
+    def element_to_map_coordinates(self, x, y):
+        """Convert element map coordinates to map coordinates."""
+        if not hasattr(self, "scale_info"):
+            return x, y
+
+        scale = self.scale_info["scale_factor"]
+        return int(x * scale), int(y * scale)
+
+    def get_element_at_position(self, x, y):
+        """Get the element at the specified position."""
+        if not hasattr(self, "element_map") or self.element_map is None:
+            return None
+
+        if not (
+            0 <= y < self.element_map.shape[0] and 0 <= x < self.element_map.shape[1]
+        ):
+            return None
+
+        return self.element_map[y, x]
+
+    def get_room_at_position(self, x, y):
+        """Get the room ID at a specific position, or None if not a room."""
+        element_code = self.get_element_at_position(x, y)
+        if element_code is None:
+            return None
+
+        # Check if it's a room (codes 101-115)
+        if 101 <= element_code <= 115:
+            return element_code
+        return None
+
+    def get_element_name(self, element_code):
+        """Get the name of the element from its code."""
+        if element_code is None:
+            return "NONE"
+
+        # Check if it's a room
+        if element_code >= 100:
+            room_number = element_code - 100
+            return f"ROOM_{room_number}"
+
+        # Check standard elements
+        for name, code in vars(DrawableElement).items():
+            if (
+                not name.startswith("_")
+                and isinstance(code, int)
+                and code == element_code
+            ):
+                return name
+
+        return f"UNKNOWN_{element_code}"