Search the Community

I have been trying to get my FLIRC working with an MCE remote to interface to an older (non-MCE) app and have found the following. I go into the GUI, clear the configuration and uncheck all of the Build in Profiles (Side note: After exiting and restarting the GUI I sometimes find some or all of the Built in Profiles are now checked. Multiple retries will finally get them all unchecked). I then confirm using Notepad that none of the numeric keys on the remote are sending any commands. I return to the GUI and using the Full Keyboard controller, I map the "a" key to the "1" button, "b" key to the "2" button ...... "j" key to the "0" button. I return to Notepad to confirm operation and find I must press a numeric button twice to get a single letter. I simply want to send an "a" each time I press the "1" button and so on. How do I get that to work? Thanks, Bill Windows 10 Home Version 1909 (OS Build 18363.836) FLIRC GUI Version 3.25.3 Firmware v4.9.3

A forum member asked how I converted a CSV file I had created originally into a JSON file and I figured I would share with everyone. I pulled my codes from here https://www.remotecentral.com/cgi-bin/codes/yamaha/rx-v1000/ Here is how I got my json file. I am familiar with the Linux command line so I performed the following steps. Before I show these let me say that an easier way could be to open a spreadsheet program and copy the button name into one column and the code for the button into another column. The second column needs to have the spaces between the numeric codes replaced with commas. You should then be able to save to the CSV file. It is possible you spreadsheet program may be able to save the file to JSON. If not, you may be able to use an online tool to do that basic work. I followed the following steps to dump the html page into a bit more of a manageable file. The steps below have been modified from the blog post. lynx -dump 'https://www.remotecentral.com/cgi-bin/codes/yamaha/rx-v1000/' | sed -e 's/^ *//' | tr -s ' ' ',' | sed 's/$/,/' > dumped_page.txt I then ran the following commands to strip out more of what I did not need. sed -i '/\[filler\.gif\]/d' dumped_page.txt sed -i '/\(Copy,to,Clipboard\)/d' dumped_page.txt sed -i '/\[filler\.gif\]/d' dumped_page.txt I then opened the dumped_page.txt file into a text editor and removed all the text I did not need. How I found that was by looking for these couple of lines of text Remote Model: RAV222 [filler.gif] Power ([23]Copy to Clipboard) Power is where my first button started so I removed every line above Power in the file. I then looked for the equivalent of the bottom the button codes. [filler.gif] [ < Back | Page: 1 [43]2 | [44]Next > ] [filler.gif] I removed everything starting from the first filler.gif line to the bottom of the file. I then combined all the codes for each button onto one line. BEFORE Stadium, 0000,006b,0022,0002,0156,00ad,0015,0015,0015,0041,0015,0015,0015,0041, 0015,0041,0015,0041,0015,0041,0015,0015,0015,0041,0015,0015,0015,0041, 0015,0015,0015,0015,0015,0015,0015,0015,0015,0041,0015,0041,0015,0015, 0015,0041,0015,0041,0015,0015,0015,0015,0015,0015,0015,0041,0015,0015, 0015,0041,0015,0015,0015,0015,0015,0041,0015,0041,0015,0041,0015,0015, 0015,05f8,0156,0057,0015,0e57, AFTER Stadium,0000,006b,0022,0002,0156,00ad,0015,0015,0015,0041,0015,0015,0015,0041,0015,0041,0015,0041,0015,0041,0015,0015,0015,0041,0015,0015,0015,0041,0015,0015,0015,0015,0015,0015,0015,0015,0015,0041,0015,0041,0015,0015,0015,0041,0015,0041,0015,0015,0015,0015,0015,0015,0015,0041,0015,0015,0015,0041,0015,0015,0015,0015,0015,0041,0015,0041,0015,0041,0015,0015,0015,05f8,0156,0057,0015,0e57 Since I know I am working with a CSV file I removed the last comma from each line since I do not need it and it could cause problems when creating the JSON file I saved that file which now gives me 20 lines for my remote and then i ran another script to convert the CSV into a JSON file. Notice that I filled in the "types", "brand", and "model" with what was relevant to me. I redirected the output of the script to a file. #!/usr/bin/env bash cat << EOF { "header": { "version": "2.0" }, "types": [ "devices.audio" ], "brand": "Yamaha", "model": "RX V-1000", "signals": [ EOF while read -r line; do _label=$(echo "${line}" | cut -d ',' -f 1) _code=$(echo "${line}" | cut -d ',' -f 2- | tr '[:lower:]' '[:upper:]') cat << EOF { "label": "${_label}", "code": "${_code}", "protocol": "PRONTO" }, EOF done < dumped_page.txt cat << EOF ] } EOF One thing to keep on mind is that JSON is picky about commas and where then end up and the end of blocks. I will show what I mean with the last two entries of my JSON file. { "label": "RO Concert", "code": "0000,006B,0022,0002,0156,00AD,0015,0015,0015,0041,0015,0015,0015,0041,0015,0041,0015,0041,0015,0041,0015,0015,0015,0041,0015,0015,0015,0041,0015,0015,0015,0015,0015,0015,0015,0015,001 5,0041,0015,0015,0015,0015,0015,0041,0015,0041,0015,0015,0015,0015,0015,0015,0015,0041,0015,0041,0015,0041,0015,0015,0015,0015,0015,0041,0015,0041,0015,0041,0015,0015,0015,05F8,0156,0057,0015,0E57", "protocol": "PRONTO" }, <----- NOTICE THIS COMMA { "label": "Stadium", "code": "0000,006B,0022,0002,0156,00AD,0015,0015,0015,0041,0015,0015,0015,0041,0015,0041,0015,0041,0015,0041,0015,0015,0015,0041,0015,0015,0015,0041,0015,0015,0015,0015,0015,0015,0015,0015,001 5,0041,0015,0041,0015,0015,0015,0041,0015,0041,0015,0015,0015,0015,0015,0015,0015,0041,0015,0015,0015,0041,0015,0015,0015,0015,0015,0041,0015,0041,0015,0041,0015,0015,0015,05F8,0156,0057,0015,0E57", "protocol": "PRONTO" } <----- IT CANNOT EXIST AT THE END ] }

Most devices come with a toggle button for power. It tells the device to go to the opposite state that it is currently in. So if the device is in an on state, the toggle command turns it off. If it's off, the toggle command turns it on. The problem is when you have multiple devices in your entertainment center and they are told to toggle. If one isn't in the same state as the others, the devices won't end up in the same state. This can happen if something interferes with the IR signal to the device. Discrete Power, on the other hand, has separate commands for On and Off. If the device is off, it will turn on, but if the device is already on, sending the Power On doesn't do anything (well sometimes it does). Same with off. As for CEC, I avoid it at all costs. It's a pain to use for anything other then a basic setup. It might be appropriate for your configuration, but as you noted it can take time. You could try like I show in the image below, just remove the existing toggle and then drag the power off commands to the macro.

So the reason I have the TV power toggle option set on "power down" is because I have CEC enabled so the AVR is turned off when the TV is powered down. I set up my "power on" options to have the TV and AVR come on at the same time because CEC would take up to 15 seconds to tell the AVR to power on as well. Could you explain what you mean by the discrete power off codes? Do you just mean the power off command that's associated with the TV and AVR config? I think you're right in going the route of just using one of the color buttons to say "send the power off command regardless of current power state."

I came back to the TV tonight after the remote was sitting for about 24 hours. I was able to turn the TV on, switch activities, etc. But then when trying to turn on and then off my B activity, the remote wouldn't respond. I then just kept trying to power B on and off and on and off (I'd wait about 20 seconds between each power on/off attempt). It would sometimes respond, but more often than not, I wouldn't get a response (either on or off). Very inconsistent. I'm not sure how helpful this log is going to be. I don't see much here aside from power button presses. The remote logs that power is pressed and then released, but that doesn't always translate to the TV turning on or off. I don't know if it's a weak signal coming from the remote or not. The IR lights when viewing through a cell phone camera turn on, so I can see that a command is actually sent. The original remotes for each device work fine, so I know it's not related to that or environment. It'd be helpful in the logging to see when A, B, or C is pressed, followed by if the power button is turning on or off an activity, followed by the devices that the remote is sending an IR command for and any delay or repeats that the remote is doing. flirc-remote-control20230504.log

I've installed it in two distros. PCLinuxOS: It's in their repositories, but the rules file I mentioned above isn't there. So you'll have to add it physically as above. MX Linux: This will install from the "curl" command mentioned above, but you'll have to locate libreadline.so.8 and create a symlink to libreadline.so.6 — I just found it again, it's in /lib/x86_64-linux-gnu/ As your distro is Ubuntu-based, I don't see why the curl command won't work, but Ubuntu has moved away from Debian so maybe that's why. If you're new to Linux, the two I mentioned are easy for beginners. I could never get on with Ubuntu and gave it up as a bad job years ago. Never could understand the fuss, apart from the fact that the distro was headed by someone with a huge amount of money who could buy publicity. But believe me, MX and PCLOS are good!

I'm assuming that you ran Flirc (capital F) and flirc_util (small F) from the command line? This may give an indication of missing dependencies. Without a .desktop file, Flirc won't appear in your menus. Without the rules file, Flirc will run, but won't find your flirc and will report that it's disconnected. I had another problem when I installed it on MX Linux, where you can use the "curl" command. It complained that libreadline.so.6 was not present. You'd have to check a number of folders, starting from /lib, /usr/lib or /usr/lib64 (there could be others in your distro, or further subfolders). I found libreadline.so.8, and made a symlink to libreadline.so.6 — this worked for me. Other than that, I have no idea whatsoever.

The B activity light does light up, but in some cases if I hit the power button multiple times, sometimes the B button light is delayed in responding. The remote will queue up the commands and the B light will light up even if I am not currently hitting the power button. In some cases the TV will turn on or off (depending on whatever command actually made it to the TV), but often the remote is just in a bad state.

I'm finding the software side of things a little sloppy. In the archive for non-Debian users, there are only two files: Flirc and flirc_util. The reason that the flirc wasn't detected on my other partition running PCLinuxOS is that they forgot to include the Udev Rules file /etc/udev/rules.d/99-flirc.rules. That had been installed by the curl command, so I copied the file from my MXLinux installation to my PCLinuxOS installation and now it's all working! It's a great invention but they need to keep the software up to date!

Ok, yeah. I did see those instructions as well. That's going back a little bit for me, but I believe I ran into a different error altogether with that one. I use Zoran OS which is Ubuntu based. I can't remember if it was the curl command itself or something else. But for your effort, I will definitely try it when I get home from work so I can let you know what it tells me. Maybe you can help me with that when I post it. I'm glad you were able to get something of it working on your and. Hopefully I'm not far behind you either. There are a couple of functions on the remote that still aren't working for me. Like, stop. I can play a video but I can't stop it. Little things like that. So I would very much like to get the flirc utility working so I can remap the buttons to work properly lol I'll update after work today. Thanks!

I just found the page you tried to work from, which mentioned repositories. Wow. Glad that's not the one I found else I'd still be struggling with it. I found a different page which gave the following command line you have to put in, in a terminal to install Flirc: curl apt.flirc.tv/install.sh | sudo bash After that, you should check whether your distro has libreadline.so.6 available. This is because whoever programmed flirc_util marked it to require libreadline.so.6 "only" instead of "or later". If it's not there, in my case in the folder /lib/x86_64-linux-gnu but your distro may be different. What I had to do, was to go to the folder: cd /lib/x86_64-linux-gnu/ (or wherever you find libreadline.so.<number> Assuming it's 8 ... ) sudo ln -sf libreadline.so.8 libreadline.so.6 This worked on MX Linux. I'm actually struggling to get it to work on PCLinuxOS at the moment.

I've got it working on MX-Linux-KDE version, but I had to symlink libreadline.so.6 to libreadline.so.8 in order to use the command line flirc_utils. MX is directly Debian based (not via Ubuntu).

pi@raspberrypi:~ $ curl apt.flirc.tv/install.sh | sudo bash % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 8266 100 8266 0 0 15450 0 --:--:-- --:--:-- --:--:-- 15479 `..-:://++ooossyyyyhhhhhhhhhhhhhhhhyyysso++/:-.` `.:/+oosyhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhyo:` -/oyhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhy /hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh- hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh: yhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh: ohhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh- -hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh` yhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhs /hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh/ yhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhho:-:+hhhhhhhhhhhh. /hhhhhhhhhhhs- -shhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh. `hhhhhhhhhhs yhhhhhhhhhh` .hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh yhhhhhhhhh- :hhhhhhhhhho` `ohhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhs:` `-shhhhhhhhho shhhhhhhhhhhssshhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh` .hhhhhhhhhhhhhhysso++///:::-----........-----:://+oshhhhhhhhhhh: :hhhhhhhhhhhs/:---......```````````.......---:::/+ohhhhhhhhhho +hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhs ohhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhy` +hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhs` -shhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhys: `.-://++oosssyyyyyhhhhhhhhhhhhhyyyssoo++//:--.` Distribution: debian Checking for curl... Detected curl... Installing flirc deb-repo... Running apt-get update... done. Installing apt-transport-https... done. Installing /etc/apt/sources.list.d/flirc_fury.list...done. Running apt-get update... done. Do you want to install the flirc utilities? [Y/n] Installing flirc utilities...done. pi@raspberrypi:~ $ flirc_util sendir --raw="+9092 -4403 +638 -456 +639 -456 +638 -456 +639 -456 +638 -456 +635 -456 +638 -456 +639 -456 +638 -1589 +638 -1585 +638 -1589 +638 -1584 +643 -1585 +638 -1584 +617 -1610 +639 -1584 +642 -452 +643 -1584 +638 -456 +639 -1584 +642 -1585 +638 -456 +613 -1610 +643 -451 +643 -1584 +638 -456 +639 -1588 +639 -451 +643 -452 +617 -1610 +639 -456 +638 -1584 +643" --repeat=1 [W] lib/libcmds/cmds.c handle_longopt(182): `sendir' doesn not take '--raw' option

Hey folks! I'm still struggling to connect Skip 1s remote to my desktop. I ran a Powershell command to watch the log file in real-time and what I'm seeing looks like the app is constantly trying to find a remote to connect. The same entry in the log appears over and over again: 2023-03-02 18:47:14 2023-03-02 18:54:38 RN C:\buildkite-agent\builds\Flirc-Windows-2\flirc-inc\remote-application\desktop\src\helpers\FlircHidApi.js:44 JB75TT8QCJJYGLXK INFO Searched|for|available|devices,|found|5|available|devices|to|connect. 0.9.66.5681 Microsoft|Windows|NT|10.0.19044.0|x64 #hidapi I'm starting to think maybe one of my USB devices is causing this conflict with the app. Anyway, I've attached the .log file for review as well. Any assistance would be appreciated! flirc20230302.log

Hi, That would be great! No support (for comma delimited sendir) required since I can roll back the GUI software to 3.26.0 Our use case for FLIRC is programatically sending IR codes. These IR codes are a large variety of TV codes which depends on which TV(s) the customer has installed. On a lot of TVs that we dont have in our database, or for some reasone they dont work, we have to capture the IR codes, assemble them into a group/database, and then we are able to use these programatically with flirc_util -sendir as needed. We need to improve the current capture process, opening the FLIRC GUI Device Log IR Debugging, pressing the remote, copying the captured ir code, pasting into a database/file, verifying operation by replay, and adding meta data .... it is not automated at the customer end currently. As a result we cannot scale product sales due to tech support hours required (not all tech support hours are for this process but it would signifcantly help) Is it possible to extend flirc_util record (or record_api) to produce the captured codes either in a file or as console output? Or there is a better way that we are just not seeing?

Hi, I have a short term problem. Our software we are using only sends the --pattern= in the exported command line. So now that I have upgraded the firmware (there is no going back right? possible to flirc_util upgrade oldfirmware x?) I dont (currently) have a path to convert the --raw= current captured codes to the --pattern= codes (b/c our current software database and execution is dependent on / hard coded --dont shoot me pls! We will fix just takes time!) Any options like: GUI secret setting? command line capture to specific format? With kind regards, Rob

This was long overdue, and should be fixed in an update I just pushed: Now supports pronto codes or the raw format that is printed directly from the GUI. Examples below. Just put the code in a string. Just a note, interkey delay is built into pronto codes. You can specify it, but it will be ignored. Also pronto codes should work with comma's or spaces. Sorry, it was batshit stupid before, especially with the leading zero needed. This is much easier to comprehend. sendir --pronto="0000,006D,0022,0002,0154,00A9,0014,0014,0014,0014,0014,0040,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0040,0014,0040,0014,0014,0014,0040,0014,0040,0014,0040,0014,0040,0014,0040,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0014,0040,0014,0014,0014,0040,0014,0040,0014,0040,0014,0040,0014,0040,0014,0040,0014,0014,0014,0040,0014,0608,0154,0056,0014,04F" --repeat=4 sendir --raw="+8840 -4394 +520 -520 +520 -520 +520 -1664 +520 -520 +520 -520 +520 -520 +520 -520 +520 -520 +520 -1664 +520 -1664 +520 -520 +520 -1664 +520 -1664 +520 -1664 +520 -1664 +520 -1664 +520 -520 +520 -520 +520 -520 +520 -520 +520 -520 +520 -520 +520 -1664 +520 -520 +520 -1664 +520 -1664 +520 -1664 +520 -1664 +520 -1664 +520 -1664 +520 -520 +520 -1664 +520" --ik=23000 --repeat=10

As I understand it, we have to click a button, what opens a field, where we drag the command from the left side into it. Would it be possible, to drag the command directly to the button, so we could save one step? Clicking on the button would have still the same function as now.

I wrote my own decoders, analyzers, and convertors. It's a command line app now and I would publish it but it's not a part of my other repo that automatically cross compiles and signs stuff. I'll get to it soon.

Can you please add the different sources to the json of Cambridge Audio EDGE A and find a solution for the Toggle Mute command and: Not fully working: Input Next / Input Previous. Both commands select the next / previous input, but it works only once. The input selector is a row, so "Input next" should select the next input clockwise and "Input previous" should select the next input anticlockwise, both with no limitation.

Sorry, I expected it to find under Misc, because in an other post you wrote, that all devices are there. I found it in Audio. Thank you for adding it. Just tested it: Many do not work. Open/Close works here :-) Above I wrote, that NAD "All Models (CD)" works almost perfect. Proposition: Wouldn't it make sense to copy the needed commands from the json for "All models" with copying the Open/Close command from C 542 into it - see attached .xlsx Reverse / Previous and Forward / Next: On the original remote these are separat commands on separate buttons. I attached a picture from the original remote - we won't talk about their design ;-) Maybe these days it was not possible to do it in the way of today: Pressing shortly the Previous or Next button lets the player skip to the previous / next track of a CD, whereas holding the same buttons for longer the reverse or forward action is done. Is that possible? Wouldn't it make sense to program the jsons for all devices like this, to make it uniform and thus easier for the user? NAD C542 json.xlsx

Thank you for integrating it! Toggle Off works correct. Toggle On doesn't work. All other commands work fundamentally: They select correct command. But they are toggle commands, flipping back and forth. That's not yet working Volume Up works: It seems, that it sends the command twice, so volume goes up in steps of 2. Volume down doesn't work: it mutes the device.

Some further info: yesterday I powered on activity A and remembered I wanted to modify the remote config, so less than 30 seconds after powering on I went to power it off by pressing power again. In that short span, it forgot the power state and tried to power on again. It will only power off activity A if I press the power button and then immediately after (no more than a few seconds) press it again so it queues a power down command after the power on. Otherwise, it forgets it's powered on and just tries to power on again. I can swear that activity B didn't behave the same because I did successfully power down without pressing the button twice a few times. However, there's maybe some kind of idle timeout involved too because I've also had it completely forget that I am in activity B and when I went to power down it just reverted to activity A and tried to power it up instead. It's very confusing what's going on with power states. I'd appreciate some kind of feedback about it. I can run tests for you or something but I think this is a big bug (or more than one) that really needs to get addressed.

Hi Just enabeled the admin mode on macOS 12. The terminal command does only work, if you create a folder named ".config" in the activ user folder. Inside this folder you create the folder "SkipApp". After that it works fine.

I have a FlircSE and I am trying to use it with a special board requiring for the power on/ power off that the power command is held for 2s (power button has to be pressed for 2s - this works well with pass through). How can I define the power on power off via the FlircSE (IR command) to be 2s?