ZiGate icon indicating copy to clipboard operation
ZiGate copied to clipboard

Device Short Address not provided in Group & Scene 0x8xxx messages family

Open KiwiHC16 opened this issue 5 years ago • 10 comments

For example In message 0x80A6, the address of the device is not provided by the zigate. Could we add this info in the message ? And in all messages for Group and Scene.

Mandatory/Urgent: View Scene Response / 0x80A0 Scene Membership Response / 0x80A6

Optional/Could wait: Add Scene Response / 0x80A1 Remove Scene Response / 0x80A2 Remove All Scene Response / 0x80A3 Store Scene Response / 0x80A4

KiwiHC16 avatar Jan 16 '19 11:01 KiwiHC16

Very similar need as https://github.com/fairecasoimeme/ZiGate/pull/90/files

KiwiHC16 avatar Jan 16 '19 11:01 KiwiHC16

OK but it's not so easy because the location where I have to add the feature concern a lot of command. If I add src address between endpoint and clusterid (for example), we'll lost a lot of compatibility. To save retro compatibility, we have to add srcAddress at the end of response. I don't know if it's clean. what do you think ?

fairecasoimeme avatar Jan 16 '19 15:01 fairecasoimeme

As soon as this is documented, we don't really care where it is.

In the heading it would be easier to read, to decode.

As some answers have variable length, it would complexify the decoding if it's at the end.

Or it could be at the end of the heading and before the variable part.

As this is a very important field and we are still in the ramp up phase of the solution I would say lets go for the hard now, it will ease a lot our live later.

Perhaps we should review all answer messages and secure the fact that we do it for all.

I think we should proceed if you know how to do it.

Perhaps @pipiche and @doudz could comment.

PS: Perhaps need to be carefull with potential impact on crc.

KiwiHC16 avatar Jan 16 '19 15:01 KiwiHC16

From my end, I think at the end is the best place, as we can differentiate from one version to the other based on the length of the payload.

pipiche38 avatar Jan 16 '19 15:01 pipiche38

Make sense, let's go for this solution.

KiwiHC16 avatar Jan 16 '19 16:01 KiwiHC16

In aim to manage versioning, it's better at the end. The inconvenient is the reading of the code. so, I'll add at the end Thanks

fairecasoimeme avatar Jan 16 '19 16:01 fairecasoimeme

Following issues are also somewhat related to this #116 #122 #39

ISO-B avatar Jan 18 '19 05:01 ISO-B

There's a mistake in doc for response 0x8063, two src addr ;)

doudz avatar Feb 18 '19 12:02 doudz

This is right. In fact It would be easier to document the data structure when in firmware 3.0d or in firmware 3.0f

pipiche38 avatar Feb 18 '19 12:02 pipiche38

Another mistake in response 0x80A7, src addr is at the end

doudz avatar Feb 18 '19 14:02 doudz