Roja Flex support #1705

Hofyyy · 2021-04-27T09:42:34Z

Supported Device:

Remote
Window Shutter

Extra feature:

Signal Cloner / Generator
Could be activated in source.
Take a real message, extract HOMEID and generate all possible command for all 15 channels:
This is helpful if you what use an own bridge and you need the commands.

Review Feedback:

Copyright line missing?

Decoder doc should be just Markdown (no leading // ), first line the device name.

data_make() should be wrapped with /* clang-format off/on */

done

those short functions are preferred to be inline code (nothing in the decoder is reuseable...)

done. Two big functions are still there. Which is think helps.

preamble is best shortened to match truncated start, i.e. 0xaa 0xaa 0xd3...

Not done, because this will change all offsets, and in my tests this works great.
I understand the idea behind this point. Maybe I can rewrite it in a second round with
changed offsets.

you would usually check for the required minimum (bit)length, then truncate -- we don't mind if there is garbage at the end.

Question: I only read 19 bytes (which is the message) what should I truncate?

Rojaflex support

zuckschwerdt · 2021-04-27T10:13:41Z

We usually want the fields "id" and "channel". Are "homeid", "radiochannel" these? Can we rename them?

this will change all offsets

usually we add the length of the preamble to start_pos so bitbuffer_extract_bytes() puts the first data byte a 0.

only read 19 bytes (which is the message) what should I truncate?

I thought msg_bitcount is used to treat the messages as variable length. I now see it's just for with/without CRC. Ok.

Try clang-format if you can. Otherwise at least: switch( command )-> switch (command) etc.

Hofyyy · 2021-04-27T10:44:45Z

We usually want the fields "id" and "channel". Are "homeid", "radiochannel" these? Can we rename them?

Yes.
If "channel" means a radio channel normally that fits. It can bis 1-15 or 0 for all.
Id seems to be ok too, if that means a static id which is used for one "remote" in one home installation.
It seems to be a second id, which I named token, which is different per message.

So long story short, I think the rename is fine.

Try clang-format if you can. Otherwise at least: switch( command )-> switch (command) etc.

I will try. but I have no clue how to update this magic push request.
I will do my best.

zuckschwerdt · 2021-04-27T11:20:38Z

No worries then, I'll just merge manually and do the style updates and id/channel myself.

Renamed HomeID in ID and RadioChannel to channel. The reason is a more generic naming in conjunction with other decoders in rtl_433.

used clang-format with default configuration

Hofyyy · 2021-04-27T15:24:27Z

So next round. Renamed and reformatted.

- Do not read the complete raw message for the decoder anymore. Direct jump to the real dataframe. So all offsets defines are changed to begin at the dataframe. That allows are better detection and "smaller" memory foodprint ...

Hofyyy · 2021-04-30T05:07:45Z

Passt alles so? :-)

Hofyyy · 2021-05-06T19:53:39Z

Ping :-)

zuckschwerdt · 2021-05-07T14:22:39Z

There are a few things to change still, I need to get around to merge and change this.

E.g. it should be "mic", "Integrity", DATA_COND, has_crc, DATA_STRING, "CRC",
Key order should be model, id, channel, ...
model needs to be a fixed string with no whitespace
if the same id can be multiple deviceType then that (model) info should be subtype, otherwise multiple model lines with DATA_COND

Can you run maintainer_update.py and commit those changes?

Hofyyy · 2021-05-08T08:55:42Z

Thanks for the feedback. I did run maintainer_update.py

run maintainer_update.py

zuckschwerdt · 2021-05-08T09:09:59Z

Thanks. Oops, put the device last (above /* Add new decoders here. */) in include/rtl_433_devices.h and rerun please.

Hofyyy · 2021-05-08T09:35:59Z

I did the move. Sorry I cloned one decoder, and used same position to know what I have to change. So I did not see the "right" pos. But the maintainer_update.py seems to not do any further change now.

zuckschwerdt · 2021-05-08T09:46:10Z

My bad. The order comes from the compiled rtl_433, recompile then rerun the update. Sorry for the mess.

Hofyyy · 2021-05-08T09:51:47Z

No problem. Looks good now.

Hofyyy · 2021-05-08T19:59:22Z

Hello, because of the reason that you have knowledge in the 433 area. Maybe you have a hint for us in the following topic. RFD-FHEM/RFFHEM#955. With the help of your tool and the new plugin I try to do the same for the FHEM server in receive and transmit direction. But until now I was not able to receive a message from the remote. Do you know the 1011 chip? Is it easy to transfer information from a rtl422 analyse into settings for the ti1011 chip?

merbanan · 2021-05-08T20:02:33Z

Not automatically but you can use the analyzer output values as a starting point for the CC1101 chip configuration.

Hofyyy · 2021-05-08T20:12:52Z

Not automatically but you can use the analyzer output values as a starting point for the CC1101 chip configuration.
I hope the questions are not to stupid, but my knowledge starts when I receive the hex values.

I wrote the decoder, so I know the frequency, the sync word (0xD391D391), Deviation 20KHz (433,90 -> 433,94),

I do not know the modulation (2FSK, GFSK, MSK)
I am not sure in case of bandwidth and Datarate (Assume, 110KHz and 10 KHZ)

merbanan · 2021-05-08T20:21:02Z

@Hofyyy well if you are trying to replicate the signal then just record a sample with the signal grabber (-S) then run it through the analyzer (-A). Then compare the output with your replicated signal. When they look the same you are done. The GFSK vs 2FSK parameter can be found by looking at the signal spectra of a recorded signal. If the frequency instantly goes from the low to the high frequency it is a 2FSK signal, if it smoothly transitions it is GFSK. My bet is on GFSK.

zuckschwerdt · 2021-05-08T20:23:58Z

With MSK the data rate is tied to the deviation for narrow bandwidth, so these parameters look good. https://en.wikipedia.org/wiki/Minimum-shift_keying
Perhaps record your CC1101 signal the same way you record the original, then compare the spectrum (https://triq.org/pdv/) and actual bits decoded with -w file.ook and also pdv.

Hofyyy · 2021-05-08T20:30:24Z

Thanks for the support. I will try to learn more about it. At the moment this is like a foreign language for me. Could that be easy answered from your side with an UHR capture in complex16s? :-P

zuckschwerdt · 2021-05-12T20:09:54Z

Well, the last one is just a sum, starting one byte after the sync words (0x31 in your examples).
The one before that could be a parity check or xor-sum, but the one row with the ea0102 obviously does not fit :/

Hofyyy · 2021-05-12T20:16:49Z

Well, the last one is just a sum, starting one byte after the sync words (0x31 in your examples).
The one before that could be a parity check or xor-sum, but the one row with the ea0102 obviously does not fit :/

Sorry I did not get it. Could you give me one more detailed example?
08 31 22 fd 20 0a 01 0a 85
How is the first stuff a sum which gives a 85?

The EA command is something special. Maybe they calculate it also in a different way.
How would you calculate the "0a" for the first examples. Excluding the 0xEA command?

zuckschwerdt · 2021-05-13T10:26:52Z

Just add the bytes, like this BitBench.

Generate internal checksum in an dynamic. Way!

merbanan · 2021-05-13T15:21:41Z

README.md

@@ -274,7 +275,7 @@ See [CONTRIBUTING.md](./docs/CONTRIBUTING.md).
  [-d <RTL-SDR USB device index>] (default: 0)
  [-d :<RTL-SDR USB device serial (can be set with rtl_eeprom -s)>]
 	To set gain for RTL-SDR use -g <gain> to set an overall gain in dB.
-	SoapySDR device driver is available.
+	SoapySDR device driver is not available.


Stray change from the maintainer update script.

merbanan · 2021-05-13T15:21:52Z

man/man1/rtl_433.1

@@ -151,7 +151,7 @@ RTL\-SDR device driver is available.
 To set gain for RTL\-SDR use \-g <gain> to set an overall gain in dB.
 .RE
 .RS
-SoapySDR device driver is available.
+SoapySDR device driver is not available.


It seems to be that the *.1 is a local generated file?
And its coming from #ifdef SOAPYSDR
"\tSoapySDR device driver is available.\n"
#else
"\tSoapySDR device driver is not available.\n"
#endif

I can correct the Readme.md or what should I do?

src/devices/rojaflex.c

- Added pramble desciption - Added syn word desciption - Added token part II desciption

- Added fprintf guards

Changed CRC output to Data_COND

Hofyyy · 2021-05-18T06:45:22Z

@merbanan Could you bring this through the door? I will not be able to provide much during this week.

zuckschwerdt · 2021-05-18T09:37:25Z

I'll be merging this, there are some final touches needed. Maybe on the weekend.

Hofyyy · 2021-06-06T17:54:40Z

Sorry for pushing, but the RTL software is great and I would like to have official Rojaflex support :-)

zuckschwerdt · 2021-06-06T20:41:22Z

My bad. I'll get on it.

zuckschwerdt · 2021-06-07T09:18:23Z

Merged, but note that there are some changes in the output to better represent the fields.

zuckschwerdt · 2021-06-07T09:43:00Z

@merbanan I've updated the code style and merged this, but I feel the model should be fixed to the protocol as "RojaFlex-Remote" with an additonal key of the decoded "device" as "Remote"/"Bridge"/"Shutter"/"Unknown". What do you think?

Hofyyy · 2021-06-07T09:55:53Z

Only for information, at the moment it can decode messages from the remote and from the shutter. and you can detect that from the message type. I do not know if that fits to your question.

zuckschwerdt · 2021-06-07T13:37:07Z

Yes, we do want a key to show "Remote"/"Bridge"/"Shutter"/"Unknown". But best case that should not be the "model" key, we use that more like a protocol-type information.

Hofyyy · 2021-06-07T14:14:42Z

Ok, I do not know the original Protocol specification, so I only can assume it.
It seems to be that the only key which shows it is the following:

e.g.
if ((msg[COMMAND_ID_OFFSET] & 0xF) == 0x5) {
deviceType = "RojaFlex-Shutter";
}

zuckschwerdt · 2021-06-07T15:43:07Z

With "keys" I'm referring to our common set of output fields, s.a. https://triq.org/rtl_433/DATA_FORMAT.html
We currently use e.g. "model", "RojaFlex-Shutter" but it should rather be

"model", "RojaFlex"
"subtype", "Shutter"

that's my thought anyway ;)

Hofyyy · 2021-06-07T15:50:34Z

Its fine for me :)

zuckschwerdt · 2021-06-07T16:24:42Z

Great! The convention for "model" values is "Xyz-Abc". We usually call advanced "remote control"-type protocols "Xy-Security". Some like the Nexa/Proove are generic "command to group-unit", so this would fit. Model values are not meant to convey a specific use-case anyway. But we could perhaps call this "RojaFlex-Automation", maybe. Not sure.

Hofyyy · 2021-06-07T17:58:27Z

I am not sure how to implement this in a quick way. I will check it next weekend. If you would like to change it, because you know how to do it. Feel free.

zuckschwerdt · 2021-06-07T18:16:12Z

No worries, I'll change that as needed. Just waiting for @merbanan to help out with:

split in "model" and "subtype"?
model of "RojaFlex-Security" (as usual) or "RojaFlex-Automation" or maybe just "RojaFlex"?

merbanan · 2021-06-07T21:24:37Z

The main aim in the output was always to make it descriptive and as unique as possible. Adding subtype makes sense. There might be other cases where this can be useful.

Hofyyy added 2 commits April 27, 2021 11:27

Rojaflex support

2f7f0fe

Rojaflex support

Removed one unneeded function

34ad2d2

Hofyyy added 2 commits April 27, 2021 17:11

Renamed homeid and radio channel

056fabc

Renamed HomeID in ID and RadioChannel to channel. The reason is a more generic naming in conjunction with other decoders in rtl_433.

Used clang-format

7f7883c

used clang-format with default configuration

Hofyyy added 2 commits April 27, 2021 18:23

Allow dynamic preamble size

f79997e

- Do not read the complete raw message for the decoder anymore. Direct jump to the real dataframe. So all offsets defines are changed to begin at the dataframe. That allows are better detection and "smaller" memory foodprint ...

Reformat with clang format again

e20f303

Hofyyy mentioned this pull request May 3, 2021

Rojaflex 433.92 Shuttercontroll support RFD-FHEM/RFFHEM#955

Closed

run maintainer_update.py

5d76552

run maintainer_update.py

Update rtl_433_devices.h

4b87e12

Hofyyy added 2 commits May 8, 2021 11:41

Update rtl_433_devices.h

13074f4

Renamed again to force reorder

49c0141

Reorder protocol number Rojaflex to 187

d9ec0fe

Update rojaflex.c

dfc7f79

Generate internal checksum in an dynamic. Way!

merbanan reviewed May 13, 2021

View reviewed changes

src/devices/rojaflex.c Outdated Show resolved Hide resolved

merbanan reviewed May 13, 2021

View reviewed changes

src/devices/rojaflex.c Outdated Show resolved Hide resolved

merbanan reviewed May 13, 2021

View reviewed changes

src/devices/rojaflex.c Outdated Show resolved Hide resolved

Hofyyy added 3 commits May 13, 2021 17:39

Updated signal documentation

8d92742

- Added pramble desciption - Added syn word desciption - Added token part II desciption

- Added fprintf guards

9b8aca9

- Added fprintf guards

Update rojaflex.c

7665ece

Changed CRC output to Data_COND

zuckschwerdt self-assigned this May 18, 2021

zuckschwerdt closed this in c6a821e Jun 7, 2021

Roja Flex support #1705

Roja Flex support #1705

Conversation

Hofyyy commented Apr 27, 2021

zuckschwerdt commented Apr 27, 2021

Hofyyy commented Apr 27, 2021

zuckschwerdt commented Apr 27, 2021

Hofyyy commented Apr 27, 2021

Hofyyy commented Apr 30, 2021

Hofyyy commented May 6, 2021

zuckschwerdt commented May 7, 2021

Hofyyy commented May 8, 2021 • edited Loading

zuckschwerdt commented May 8, 2021

Hofyyy commented May 8, 2021

zuckschwerdt commented May 8, 2021

Hofyyy commented May 8, 2021

Hofyyy commented May 8, 2021

merbanan commented May 8, 2021

Hofyyy commented May 8, 2021

merbanan commented May 8, 2021

zuckschwerdt commented May 8, 2021

Hofyyy commented May 8, 2021

zuckschwerdt commented May 12, 2021

Hofyyy commented May 12, 2021 • edited Loading

zuckschwerdt commented May 13, 2021

merbanan May 13, 2021

Choose a reason for hiding this comment

merbanan May 13, 2021

Choose a reason for hiding this comment

Hofyyy May 13, 2021

Choose a reason for hiding this comment

Hofyyy commented May 18, 2021

zuckschwerdt commented May 18, 2021

Hofyyy commented Jun 6, 2021

zuckschwerdt commented Jun 6, 2021

zuckschwerdt commented Jun 7, 2021

zuckschwerdt commented Jun 7, 2021

Hofyyy commented Jun 7, 2021

zuckschwerdt commented Jun 7, 2021

Hofyyy commented Jun 7, 2021

zuckschwerdt commented Jun 7, 2021

Hofyyy commented Jun 7, 2021

zuckschwerdt commented Jun 7, 2021

Hofyyy commented Jun 7, 2021

zuckschwerdt commented Jun 7, 2021

merbanan commented Jun 7, 2021

Hofyyy commented May 8, 2021 •

edited

Loading

Hofyyy commented May 12, 2021 •

edited

Loading