Hello,

I have the custom integration successfully added to HASS and configured the remote in the home assistant configuration.yaml, with my projector's ip address. This all works fine, I used the services tool, in the developer tools to turn on the projecter and change the profile to cinema. Super cool, and thank you for building this.

I'm now trying to add commands for switching between the Custom Mask settings (I have a scope screen and would like to mask the top and bottom of the screen when zoomed out, I'd like to turn off the mask when zoomed in for non scope content.)

I found these "IP codes" on AVS: (https://www.avsforum.com/threads/official-jvc-20ltd-rs640-x990-x9900-rs540-x790-x7900-owners-thread.2923938/post-60433741)

[Mask Off: 0x21, 0x89, 0x01, 0x49, 0x53, 0x4D, 0x41, 0x32, 0x0A
Mask Custom1: 0x21, 0x89, 0x01, 0x49, 0x53, 0x4D, 0x41, 0x30, 0x0A
Mask Custom2: 0x21, 0x89, 0x01, 0x49, 0x53, 0x4D, 0x41, 0x31, 0x0A
Mask Custom3: 0x21, 0x89, 0x01, 0x49, 0x53, 0x4D, 0x41, 0x33, 0x0A]

I realize you provided instructions on how to add commands, and I think I understand how to do that. However, I don't know WHERE and in what file to add those commands when using the hass_custom_components integration. The file located in the non HASS repo at https://github.com/bezmi/jvc_projector/blob/master/jvc_projector/__init__.py has lines like:

class Commands(Enum):
# power commands
power_on = b"\x21\x89\x01\x50\x57\x31\x0A"

I tried adding the following lines to /config/custom_components/jvcprojector/init.py and restarting home assistant, but then the projector remote does not show up in home assistant anymore, so obviously I'm doing something wrong.

class Commands(Enum):
# mask commands
mask_off = b"\x21\x89\x01\x49\x53\x4D\x41\x32\x0A"
mask_custom1 = b"\x21\x89\x01\x49\x53\x4D\x41\x30\x0A"
mask_custom2 = b"\x21\x89\x01\x49\x53\x4D\x41\x31\x0A"
mask_custom3 = b"\x21\x89\x01\x49\x53\x4D\x41\x33\x0A"

Could you please provide me a little guidance on how and where to add these additional commands for use in home assistant?

Thanks in advance!

Newer models support more saved lens memories. Just need to modify the command to include the extra ones.

Hi good sir,
Thanks for making this!
Any chance you could consider adding support for switching between Cinema and HDR (and that third button on the remote which I seldom use)?

Hi,

I am looking for the cursor commands to navigate in the menu via external commands, I do not see it in the documentation, anyone knows these?

LEFT
RIGHT
UP
DOWN
OK
BACK

Hi,

I tried your code with JVC DLA-X30B.

All worked accept the Power On command. After some debugging, I inserted 1 second delay after sending the power on command and I had to send 3 of these in a row from my script. WIth that, all finally worked. I have 4 separate scripts that work on the same code that you provide + my delay after the sendall command of 1 second. The 3 scripts of power off, status, and model inquiry work as is. The power on command works only if I send it 3 times in a row. I would hope to figure out what the timing issue is but may need your help.

In the HomeAssistant code I have to catch exceptions every location a command is called. Right now that is due to the calling of either command() or power_state().

They both trigger _send_command() but in a different way.

I wanted to refactor the command() so it can handle both commands that do return results or commands that don't return results:

A basic version was working, but it meant changing a lot of the basics and it felt like I was adding spaghetti-code and magic behaviour.

In essence I altered command() to check if the provided command_string was a ResultCommand or a Command. In case of a ResultCommand it needs an ack, which required knowing which ack to receive. For power state this is easy and with a minor change using the ACKs enum to get the ack based on the command string was possible. However imput changes are variable, but the ack the same. So that would mean introducing multiple ACKs with same command like so:

class ACKs(Enum):
    power_status = b"\x06\x89\x01\x50\x57\x0A"
    hdmi1 = b"\x06\x89\x01\x49\x50\x0A"
    hdmi2 = b"\x06\x89\x01\x49\x50\x0A"

It worked okay-ish. But after the comand was sent, the result must be handled and stored. I didn't want to alther the return True/False on command so extendend it to save the message, which could then be processed in a helper function for that power status. I ended in finding if a _translate_<command>() function-ish situation which did work. But as said, it felt spaghetti-code like.

The command looked something like this:

    def command(self, command_string, receive=False):
        print(command_string)

        if hasattr(Commands, command_string):
            self._send_command(Commands[command_string].value)
            return True

        if hasattr(ResultCommands, command_string):

            result_message = self._send_command(ResultCommands[command_string].value, ack=ACKs[command_string].value)
            if hasattr(self, '_translate_' + command_string) and callable(getattr(self, '_translate_' + command_string)):
                self.result_message = getattr(self, '_translate_' + command_string)(result_message)

            return True

        return False

Right now it feels like there is an extra layer of information is required which maps commands to results or a way of interpreting results. Maybe a grouping of certain commands and its generic handling (like that input command results should be handled the same) and that certain commands are ACK-ed the same.

If this is refactored, adding of the commands (which sroelens asked about in #4) is then possible.

Food for thought!

Playing around with a serial console and direct connection to the projector, it's apparent that multiple commands can be sent after the JVC handshake procedure. Currently, we close the socket after a command is sent and re-connect via the handshake for the next clmmand.

The code should be refactored to keep the connection alive. It will improve reliability.

To prevent future HA upgrades from breaking this great component, can you add the version as shown in the error below?

PS: Thanks for creating this. Absolutely fantastic!

No 'version' key in the manifest file for custom integration 'jvcprojector'. This will not be allowed in a future version of Home Assistant. Please report this to the maintainer of 'jvcprojector'
11:54:58 AM – loader.py (WARNING)

I've been trying to connect to my JVC LX-NZ30 projector via Home Assistant. I've tried the HA included integration and the custom integration that uses this in the backend and both fail to connect.

I traced down here from the custom integration and tried running this module direct from Python. The result is a connection failure based on a handshake timeout.

I've seen other mention of this handshake protocol on the web but the LX-NZ30 doesn't seem to require/support it. I've managed to successfully control my projector with as little as this:

import socket
import binascii

host = '192.168.1.19'
port = 4661

ctrl_power_on = b'\x06\x14\x00\x04\x00\x34\x11\x00\x00\x5D'

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.send(ctrl_power_on)

received_data = s.recv(1024)
print (binascii.hexlify(received_data))

I don't do any handshake, at least on the surface. Maybe socket is doing something generic under the hood. I'm wondering if there's a way to make the handshake conditional or base it off of the projector model number.

https://www.seanh.cc/2022/05/21/publishing-python-packages-from-github-actions/

The new JVC NZ line requires a password in addition to the IP address to execute commands.