Hi @AntonJansen, I feel that the issue is to do with the learning, specifically around RF. There doesn’t seem to be many complaints from folks with IR devices… In relation to your test scenario, whereby I throw an IR command its way, I’ll see if I can work it through later.
I’ve no problem interacting/learning/sending via python-broadlink, so I think there’s likely just a small RF related tweak or something, to do with the learning aspect in the addon… anyway…
In relation to python-broadlink,
Python 3.9.2 (default, Mar 12 2021, 04:06:34)
[GCC 10.2.1 20210110] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import broadlink
>>> devices = broadlink.discover()
>>> print(devices)
[broadlink.remote.rmpro(('192.168.1.124', 80), mac=b'x\x0fw\xebI\x8a', devtype=10153, timeout=10, name='智能遥控', model='RM pro+', manufacturer='Broadlink', is_locked=False)]
Started working on replicating the issue on my own side. I already hit a wall, as my RM Pro 4 was not recognized. Fixed some bugs to make it work and I seem to be able to learn RF commands using the broadlink python program.
Investigating the python program, the conclusion is that we need to add an additional part for RF, as it uses a different command and learning cycle. I can test the learning part, but currently not verify the sending part.
First step in RF learning is working; finding the correct frequency. I do run into issues with confirming the learned command, so it still needs more work.
I have managed to learn and send a RF command on a RM4 Pro. Next steps are to clean up the code and see if it works with a RM pro+ device as well. I have also changed the way commands are learned. It will now be completely possible through the web UI, without the need for the logging information.
These changes might come after the initial version has been accepted into openhab, we will see. Once I have a build ready for testing, I will post it here @Vini, so you can test it and let me know whether it works for you.
An alpha version of the binding with RF support is now available here. I have tested this with a RM Pro 4, please see if it also works with a classic RM Pro.
Please find the instructions on learning commands here, as I have made it more convenient to learn new commands.
Please let me know what you experience with this new alpha version!
@Bernd_Ritter this is about the PR @hmerk was talking about in the other topic. Based on the PR, I have created a new branch to extend the functionality of the binding. It now comes with RF support for the Pro models and allow for GUI based learning of the IR/RF commands. It is still fully backwards compatible with the PR, as it uses the same file and format for storing learned IR commands.
Thank you! I have an RM2 Pro and an RM4 Pro.
Both already work with IR- and RF- using the old version of the binding, I think.
But I use only the RM2 Pro for RF.
It might be worth putting this in the marketplace, starting a separate (new) discussion for it, then pointing to it from these old threads. I think that would reduce confusion stemming from the past discussions. There’s a lot of baggage that can be left behind.
Thanks for the update. I am using a rm2 to send RF to some old plugs. It is working fine for me and i was able to reuse the already learned codes from the broadlink.map file.
Hi, Apologies for being late to the party. @AntonJansen , any chance you can sync your openhab-addons fork? I started trying to commit the storage changes to your repo, but i think some of the requirements for openhab changed between yesterday and today, and the brinding does not compile anymore.
Branch broadlink-binding-4.2.x is now synched with the latest of openhab-addons. Coming week I intend to spend some time to clean-up and merge the rm4pros branch that will bring in RF handling and a better UI way of creating new IR/RF commands. Hence, you might want to use the rm4pros branch as a starting point.
My thinking is that it’ll simplify the installation process in the short term. When people move to 4.2 (which many won’t do until it’s released), they’ll just have to uninstall the marketplace binding, install the release version, and carry on.
Is the rm4pros branch also based on 4.2.x and also synced? My commits also have some UI changes for IR Learning, hopefully you will be able to use them.
