-
Notifications
You must be signed in to change notification settings - Fork 522
Add initial support for Eria Adurosmart Wireless Dimming Switch #4211
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
I ran into a compiler error with this branch/commit:
This is fixed by adding and OR operator to the line above after eg
|
After successfully compiling I am still having the same issues as my previous comment in issue 1730 where:
With the code that was recently added for this switch, is there any specific brand/model that should be selected in Phoscon? I have tried:
Not sure if it would help, but for either of these brands/models, when I am in the pairing mode in Phoscon, I will get the something similar to the following output in deCONZ debugging:
I am very unfamiliar with this code so I don't have any idea what most of it does. Is there anything specific I should try? Including modifying any certain parts of the code? |
Thanks for checking. I meanwhile corrected the small error in the code. How does the switch look like in deconz GUI after you've paired it? Has a corresponding sensor resource been created in the REST API? If the device got paired correctly and everything worked out as expected, you should see some You'll not see the device in Phoscon unless explicitly supported there, which must be requested in the Phoscon beta repository. |
Well I somehow managed to bork my Ubuntu installation by installing a wrong driver so I currently cannot use any mouse or keyboard, and I never set up a means to remote into it. This means I'll have quite the trouble being of any help right now until I un-fudge my OS. At this point I'm a little hesitant to try messing around with this on my Raspberry Pi in case I mess up my setup, so I'll try my best from memory...
A couple things I remember:
How would I be able to tell is one was created? Regardless I would be going off memory at this point, but I can see if anything rings a bell.
Does this answer my previous question? Is this referring to the REST API? I don't remember anything in the CLI output specifically saying Would these
What would be the steps for this? Alternatively would it be easier to classify it as a Sorry I'm not much help at the moment, I'll try to see if I can get my Ubuntu working this weekend, and if not, I'll try debugging on my Pi. |
So I managed to get my Ubuntu up and running again and did a bit of investigating...
I think I can now confidently that no sensor has been created. While debugging with print statements I noticed whenever any buttons are pressed, the code ends up getting to I am also not getting the What would be involved in getting the sensor resource / sensor node set up? |
@mdolnik Sorry, I guess this one slipped through. The reason is that the manufacturer code and mac prefix are identical to the Xiaomi's and therefore, neither model ID nor manufacturer name are querried. The solution for this should be starting sensor search in Phoscon and then reading the basic cluster in deconz GUI while the device is awake (just press a button on the device and press the read button 1-2 seconds afterwards). |
Juts an FYI the merged branch above does not resolve this issue, which I'm sure is already known since nothing was changed since 65acc33 on Jan 19th, but I just wanted to confirm this is still an issue. I found the culprit as to why the sensorNode is not being created for this switch: deconz-rest-plugin/de_web_plugin.cpp Lines 5641 to 5643 in 08e1181
This code will ignore all nodes with One work-around that works for me is to change the logic in an if statement above: deconz-rest-plugin/de_web_plugin.cpp Lines 5626 to 5629 in 08e1181
To:
Which essentially whitelists this specific model (which there are no other existing ) before it is backlisted for its manufacturer code. After this change, I am now getting the following
But I'm not sure how to proceed or what to make of cluster |
Also regarding the manufacturer codes, is there a reason that this branch used deconz-rest-plugin/de_web_plugin.cpp Line 339 in 08e1181
When there is already a definition for deconz-rest-plugin/de_web_plugin_private.h Line 369 in 08e1181
Does the vendor have to match the mac prefix? |
Refer to the new Pull Request #4495 |
See #3383 and #1730