GetAllStatistics - Error 'Data length mismatch. Read 28 bytes of 24'

[Marcussjogren]

Hi,

When I am running the code on my Huqsvarna 440 I get the below error. Seems like the received data is larger than expected.

I commented line 207-209 and now it's working but I guess I'm missing the statistics.

I will try to resolve it - but I think this is out of my competence :)

Line 207 - 209:

#    statuses = await mower.command("GetAllStatistics")
#    for status, value in statuses.items():
#        print(status, value)

This was the error that I was receiving before.

No next start time
2024-12-16 19:31:20,940 protocol INFO: Writing: b'02fd1000b63b604701a100af761200000000ef03'
2024-12-16 19:31:21,024 protocol INFO: Received: b'02fd2900b63b6047017901af76120000001800e8'
2024-12-16 19:31:21,027 protocol INFO: Received: b'0ca6007ffe9d000b8e2700690e0800a8770000b7'
2024-12-16 19:31:21,032 protocol INFO: Received: b'0200006303'
2024-12-16 19:31:21,040 protocol INFO: Final response: b'02fd2900b63b6047017901af76120000001800e80ca6007ffe9d000b8e2700690e0800a8770000b70200006303'
Traceback (most recent call last):
  File "/home/pi/AutoMower-BLE/automower_ble/mower.py", line 299, in <module>
    asyncio.run(main(mower))
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 194, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 686, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/pi/AutoMower-BLE/automower_ble/mower.py", line 207, in main
    statuses = await mower.command("GetAllStatistics")
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/pi/AutoMower-BLE/automower_ble/mower.py", line 50, in command
    response_dict = command.parse_response(response)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/pi/AutoMower-BLE/automower_ble/protocol.py", line 208, in parse_response
    raise ValueError(
ValueError: Data length mismatch. Read 28 bytes of 24

[Marcussjogren]

By the way, is there any way to keep the mower connected while developing? I must open the lid before sending the commands.

[Marcussjogren]

I was not skilled enough to understand the code really, so I did what any mechanic would have done: I removed the dpos data lenght check :)

Removed line 208-211 in protocol.py:

#        if dpos != len(data):
#            raise ValueError(
#                "Data length mismatch. Read %d bytes of %d" % (dpos, len(data))
#            )

It's working now, but I guess there was a reason for the dpos data length-check, hence I know that it might screw something up further on.

This is the output data that caused the "Read 28 bytes of 24" error:

No next start time
2024-12-16 20:57:30,763 protocol INFO: Writing: b'02fd1000b63b604701a100af761200000000ef03'
2024-12-16 20:57:30,847 protocol INFO: Received: b'02fd2900b63b6047017901af76120000001800e8'
2024-12-16 20:57:30,857 protocol INFO: Received: b'0ca6007ffe9d000b8e2700690e0800a8770000b7'
2024-12-16 20:57:30,863 protocol INFO: Received: b'0200006303'
2024-12-16 20:57:30,866 protocol INFO: Final response: b'02fd2900b63b6047017901af76120000001800e80ca6007ffe9d000b8e2700690e0800a8770000b70200006303'
totalRunningTime 10882280
totalCuttingTime 10354303
totalChargingTime 2592267
totalSearchingTime 527977
numberOfCollisions 30632
numberOfChargingCycles 695
cuttingBladeUsageTime 0

[alistair23]

Sorry, I just got back to looking at this.

I'm assuming that the decoded data cuttingBladeUsageTime 0 is wrong, so just removing the check isn't enough to get the correct data.

What I'm assuming is happening is that your model doesn't support cuttingBladeUsageTime, as my 305 also has no way to know that information. So I'm guessing that the mower isn't sending the information, hence the 24 bytes. But we currently expect it

So we probably need to convert cuttingBladeUsageTime to being optional.

My other thought, does any mower support cuttingBladeUsageTime? Maybe it's just best to remove it

[alistair23]

@erikmol any thoughts on removing cuttingBladeUsageTime? Does it work for you?

[box2k2]

My other thought, does any mower support cuttingBladeUsageTime? Maybe it's just best to remove it

Doesn't appear to be available in the latest Android Flymo app for an Easilife Go 200

[Marcussjogren]

Sorry, I just got back to looking at this.

I'm assuming that the decoded data cuttingBladeUsageTime 0 is wrong, so just removing the check isn't enough to get the correct data.

What I'm assuming is happening is that your model doesn't support cuttingBladeUsageTime, as my 305 also has no way to know that information. So I'm guessing that the mower isn't sending the information, hence the 24 bytes. But we currently expect it

So we probably need to convert cuttingBladeUsageTime to being optional.

My other thought, does any mower support cuttingBladeUsageTime? Maybe it's just best to remove it

Yea I’m well aware that removing the check is not a good solution, but I’m using it like this on my own risk :-)

would be really nice to have it optional or remove it from the 440 model, because i have never seen anything regarding cutting blade time in the app nor in the mowers menu.

Thanks for looking in to it and really cool tool you’ve built!

[erikmol]

@erikmol any thoughts on removing cuttingBladeUsageTime? Does it work for you?

I've seem to kind of brick my BLE connection because my Automower can't be paired with any device so I can't do any testing currently.

When extending the protocol I've been opening a .apk in jadx and "Save as gradle project" and then used file search in VS Code to regex search for major/minor combinations. When looking at Flymo Bluetooth App_6.1.0_APKPure.apk I found that the libary code (e.g. sources\com\gardena\libraries\bluetooth_mower\bluetooth_mower\mowerconnection\commands\StatisticsCommands.java) is very readable. The following is a snippet from that file and what the GetAllStatistics command is based on:

        public GetAllStatisticsResponse() {
            super(4726, 0);
            getProtocolData().getParameters().add(new Parameter("totalRunningTime", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("totalCuttingTime", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("totalChargingTime", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("totalSearchingTime", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("numberOfCollisions", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("numberOfChargingCycles", DataTypes.UINT32));
            getProtocolData().getParameters().add(new Parameter("cuttingBladeUsageTime", DataTypes.UINT32));
        }

I don't think it's relevant to discuss if certain data is used/implemented for specific mowers as we should mimic the protocol. I don't think numberOfChargingCycles is visible in the app either?

Trying to decompile the Automower Connect_2024.7.1_APKPure.apk in the same fashion as above does not lead to the same readable and structured code, but I do recognize that there seems to be two implementations of this function:

/* compiled from: StatisticsCommands.java */
/* loaded from: classes3.dex */
public class C3 extends O7.b {
    public C3() {
        super(4726, 0);
        c().a().add(new O7.c("totalRunningTime", "uint32"));
        c().a().add(new O7.c("totalCuttingTime", "uint32"));
        c().a().add(new O7.c("totalChargingTime", "uint32"));
        c().a().add(new O7.c("totalSearchingTime", "uint32"));
        c().a().add(new O7.c("numberOfCollisions", "uint32"));
        c().a().add(new O7.c("numberOfChargingCycles", "uint32"));
        c().a().add(new O7.c("cuttingBladeUsageTime", "uint32"));
    }
}

and also

/* compiled from: StatisticsCommands.java */
/* renamed from: Oe.z3, reason: case insensitive filesystem */
/* loaded from: classes3.dex */
public class C1412z3 extends O7.e {
    public C1412z3() {
        super(4726, 0);
        e().a().add(new O7.c("totalRunningTime", "uint32"));
        e().a().add(new O7.c("totalCuttingTime", "uint32"));
        e().a().add(new O7.c("totalChargingTime", "uint32"));
        e().a().add(new O7.c("totalSearchingTime", "uint32"));
        e().a().add(new O7.c("numberOfCollisions", "uint32"));
        e().a().add(new O7.c("numberOfChargingCycles", "uint32"));
    }

So yes, there seems to be two definitions of the same command, the second one lacking the cuttingBladeUsageTime.

My suggestion would be to check if the following mappings (all uint32) works, and if so replace the GetAllStatistics function. These are also visible in the StatisticsCommands.java file referenced above:

4726,1 : "totalRunningTime"
4726,2 : "totalCuttingTime"
4726,3 : "totalChargingTime"
4726,4 : "totalSearchingTime"
4726,5 : "numberOfCollisions"
4726,6 : "numberOfChargingCycles"
4726,7 : "cuttingBladeUsageTime"

[erikmol]

A side note is that the command GetTask have an unknown uint16 appended which also does not match the .apk sources, so maybe some general forgiving processing is prefered for this library?