Fix missing powerview shade data when initial refresh fails (home-ass…

…istant#113033)

homeassistant/components/hunterdouglas_powerview/cover.py

@@ -71,7 +71,6 @@ async def _async_initial_refresh() -> None:
 
     entities: list[ShadeEntity] = []
     for shade in pv_entry.shade_data.values():
-        coordinator.data.update_shade_position(shade.id, shade.current_position)
         room_name = getattr(pv_entry.room_data.get(shade.room_id), ATTR_NAME, "")
         entities.extend(
             create_powerview_shade_entity(

homeassistant/components/hunterdouglas_powerview/number.py

@@ -41,7 +41,7 @@ def store_velocity(
     value: float | None,
 ) -> None:
     """Store the desired shade velocity in the coordinator."""
-    coordinator.data.update_shade_velocity(shade_id, ShadePosition(velocity=value))
+    coordinator.data.update_shade_position(shade_id, ShadePosition(velocity=value))
 
 
 NUMBERS: Final = (

homeassistant/components/hunterdouglas_powerview/shade_data.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+from dataclasses import fields
 import logging
 from typing import Any
 
@@ -12,74 +13,66 @@
 
 _LOGGER = logging.getLogger(__name__)
 
+POSITION_FIELDS = fields(ShadePosition)
+
+
+def copy_position_data(source: ShadePosition, target: ShadePosition) -> ShadePosition:
+    """Copy position data from source to target for None values only."""
+    # the hub will always return a velocity of 0 on initial connect,
+    # separate definition to store consistent value in HA
+    # this value is purely driven from HA
+    for field in POSITION_FIELDS:
+        if (value := getattr(source, field.name)) is not None:
+            setattr(target, field.name, value)
+
 
 class PowerviewShadeData:
     """Coordinate shade data between multiple api calls."""
 
     def __init__(self) -> None:
         """Init the shade data."""
-        self._group_data_by_id: dict[int, dict[str | int, Any]] = {}
-        self._shade_data_by_id: dict[int, BaseShade] = {}
+        self._raw_data_by_id: dict[int, dict[str | int, Any]] = {}
+        self._shade_group_data_by_id: dict[int, BaseShade] = {}
         self.positions: dict[int, ShadePosition] = {}
 
     def get_raw_data(self, shade_id: int) -> dict[str | int, Any]:
         """Get data for the shade."""
-        return self._group_data_by_id[shade_id]
+        return self._raw_data_by_id[shade_id]
 
     def get_all_raw_data(self) -> dict[int, dict[str | int, Any]]:
         """Get data for all shades."""
-        return self._group_data_by_id
+        return self._raw_data_by_id
 
     def get_shade(self, shade_id: int) -> BaseShade:
         """Get specific shade from the coordinator."""
-        return self._shade_data_by_id[shade_id]
+        return self._shade_group_data_by_id[shade_id]
 
     def get_shade_position(self, shade_id: int) -> ShadePosition:
         """Get positions for a shade."""
         if shade_id not in self.positions:
-            self.positions[shade_id] = ShadePosition()
+            shade_position = ShadePosition()
+            # If we have the group data, use it to populate the initial position
+            if shade := self._shade_group_data_by_id.get(shade_id):
+                copy_position_data(shade.current_position, shade_position)
+            self.positions[shade_id] = shade_position
         return self.positions[shade_id]
 
     def update_from_group_data(self, shade_id: int) -> None:
         """Process an update from the group data."""
-        self.update_shade_positions(self._shade_data_by_id[shade_id])
+        data = self._shade_group_data_by_id[shade_id]
+        copy_position_data(data.current_position, self.get_shade_position(data.id))
 
     def store_group_data(self, shade_data: PowerviewData) -> None:
         """Store data from the all shades endpoint.
 
-        This does not update the shades or positions
+        This does not update the shades or positions (self.positions)
         as the data may be stale. update_from_group_data
         with a shade_id will update a specific shade
         from the group data.
         """
-        self._shade_data_by_id = shade_data.processed
-        self._group_data_by_id = async_map_data_by_id(shade_data.raw)
+        self._shade_group_data_by_id = shade_data.processed
+        self._raw_data_by_id = async_map_data_by_id(shade_data.raw)
 
-    def update_shade_position(self, shade_id: int, shade_data: ShadePosition) -> None:
+    def update_shade_position(self, shade_id: int, new_position: ShadePosition) -> None:
         """Update a single shades position."""
-        if shade_id not in self.positions:
-            self.positions[shade_id] = ShadePosition()
-
-        # ShadePosition will return None if the value is not set
-        if shade_data.primary is not None:
-            self.positions[shade_id].primary = shade_data.primary
-        if shade_data.secondary is not None:
-            self.positions[shade_id].secondary = shade_data.secondary
-        if shade_data.tilt is not None:
-            self.positions[shade_id].tilt = shade_data.tilt
-
-    def update_shade_velocity(self, shade_id: int, shade_data: ShadePosition) -> None:
-        """Update a single shades velocity."""
-        if shade_id not in self.positions:
-            self.positions[shade_id] = ShadePosition()
-
-        # the hub will always return a velocity of 0 on initial connect,
-        # separate definition to store consistent value in HA
-        # this value is purely driven from HA
-        if shade_data.velocity is not None:
-            self.positions[shade_id].velocity = shade_data.velocity
-
-    def update_shade_positions(self, data: BaseShade) -> None:
-        """Update a shades from data dict."""
-        _LOGGER.debug("Raw data update: %s", data.raw_data)
-        self.update_shade_position(data.id, data.current_position)
+        copy_position_data(new_position, self.get_shade_position(shade_id))