Fix room detection and extract no-go zones, virtual walls and obstacles for xiaomi.vacuum.ov21gl#748
Open
luchusnet wants to merge 3 commits into
Conversation
…es for xiaomi.vacuum.ov21gl Some Xiaomi models (e.g. xiaomi.vacuum.ov21gl / Xiaomi Robot Vacuum 5 Pro) have a firmware quirk where every entry in map_room_info has room_id=0. The upstream vacuum_map_parser_xiaomi library uses room_id as a dict key, which causes all rooms to collapse into a single Room object. Fix: in decode_and_parse(), intercept the JSON payload before passing it to parse(). If all room_ids are identical, replace each room_id with its grid_id (which is unique and matches the id field used in room_attrs). This restores correct room detection (14 rooms instead of 1). Additionally, the library does not parse fb_regions (no-go zones), fb_walls (virtual walls), or ai_obj (AI-detected obstacles). These are now extracted manually in a new _extract_map_objects() static method and injected into the MapData after the library parse, only if the library left those fields empty.
The library's parse() calls draw_map() internally, so zones injected
after parse() were never rendered. Fix by normalizing the firmware's
fb_regions/fb_walls format to the library's expected format before
calling parse(), so all elements are rendered in the camera image.
Firmware format:
fb_regions: [{"fb_point": [x0,y0,x1,y1,x2,y2,x3,y3]}]
fb_walls: [{"wall_points": [x0,y0,x1,y1]}]
Library expected format:
fb_regions: [{"type": "no_go"|"wall", "points": [{"x":..,"y":..},...]}]
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When the vacuum is actively cleaning its LAN may be unresponsive, causing should_update_map's MiOT property read to throw a DeviceException. The previous code returned False on any non-token DeviceException, which stopped all cloud map downloads for the rest of the session. Return True instead so the cloud map continues to be fetched even when the local LAN status check fails. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Author
|
@PiotrMachowski Friendly bump 🙂 This PR has been running on my OV21GL (Xiaomi Robot Vacuum 5 Pro) for the past week without issues — room detection, no-go zones, virtual walls and obstacles all extract correctly. Happy to address any feedback or make adjustments if you have time to take a look. Diff is small (+125/-3), focused on OV21GL specifics. Thanks for maintaining this great project! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
On some Xiaomi models (confirmed:
xiaomi.vacuum.ov21gl/ Xiaomi Robot Vacuum 5 Pro), every entry inmap_room_infohasroom_id=0due to a firmware quirk. Thevacuum_map_parser_xiaomilibrary usesroom_idas a dict key, so all rooms overwrite each other and collapse into a singleRoomobject — resulting in 1 room instead of the actual 14.Additionally, the library does not parse
fb_regions(no-go zones),fb_walls(virtual walls), orai_obj(AI-detected obstacles) from the JSON payload.Finally, when the vacuum's LAN is temporarily unreachable (e.g. during active cleaning), the
should_update_mapMiOT status check throws aDeviceException. The previous code returnedFalseon any non-token exception, stopping all cloud map downloads for the rest of the session.Fix
Room detection (
_fix_map_room_info)In
decode_and_parse(), the JSON payload is intercepted before being passed toparse(). If allroom_idvalues are identical, each is replaced with itsgrid_id(which is unique and matches theidfield used inroom_attrs). This restores correct room detection.No-go zones & virtual walls (
_normalize_firmware_map_objects)Normalizes the firmware-specific
fb_regions/fb_wallsformat to the library's expected format before callingparse(), so the library renders them in the camera image.AI obstacles (
_extract_obstacles)Extracts
ai_objentries post-parse and injects them intoMapData.obstacles(non-destructive, only if the library left it empty).Map not updating during cleaning (
should_update_map)When the LAN status check fails with a non-token
DeviceException, now returnsTrueinstead ofFalse, so cloud map downloads continue even when LAN is temporarily unreachable.Tested on
xiaomi.vacuum.ov21gl(Xiaomi Robot Vacuum 5 Pro)