IRScope decoding - can any of this be automated?

[The Robman]

I have just loaded a zip file here:
https://www.hifi-remote.com/forums/dload ... e_id=11393

This file shows all the manual steps that I have to follow in order to decode learns when they are posted in an ICT file rather than an IR file.

File 1: Ortek Mouse.ict
This is a sample ICT file (originally posted here, if you're interested). There is a bug in IRScope where it doesn't recognize a new learned signal if the leadout time of the previous learned signal is not long enough, so the first thing you need to do with an ICT file is open it up using Notepad and examine all the leadout times to make sure they're about 500k.

File 2: Ortek Mouse2.ict
This is the sample file after I've fixed the leadout times, so this one will open OK using IRScope.

File 3: ExportedUEI.txt
Next, you need to manually open the ExportedUEI.txt file (in the IRScope folder) if it exists and remove its contents. Then, in IRScope, select Export > Export Mode > UEI Learned. Then, select all the learns (CNTL+A works for this). Then select Export > Export Selected (CNTL+E also works).

File 4: IRScope-decoder.xls
Next, open my IRScope-decoder.xls spreadsheet (a copy is included in the zip). Then, open the ExportedUEI.txt and cut & paste its contents into the "Input" tab of the spreadsheet. Now go to the "Output" tab and cut & paste the first column into a Notepad file and save it as an IR file.

File 5: Ortek Mouse2.ir
This is the results of the previous step. Open this file using IR.exe (or the version of RMIR that has Vyrolan's changes) and go to the Learned Signals tab. Select an appropriate rounding value, and (if necessary) a Bi-Phase value, and click the Times Summary button. Then cut & paste the contents of the pop-up screen into another Notepad session.

File 6: manual-edits.txt
In this case, 500 is the correct rounding and EVEN is the correct Bi-Phase option. In the Notepad session, you would replace "+500 -500; " with "1", "-500 +500; " with "0", then the replace the remaining "+500 " with "1" also. This will give you the binary of the signal.


So, can any of these steps be automated? I don't really care which program it all ends up in (ie, IR, RMIR, IRScope, IRMaster, etc) but I would really appreciate not having to do all this manual work whenever someone posts learns that contain a new protocol in an ICT file rather than an IR file.

[cauer29]

The obvious thing (to me anyway) is to skip all of the intermediate programs. Just do it all in Excel. VBA can open the ICT file do all the fixups needed, put whatever is needed into the spreadsheet and export the result as needed.

A.A.

[The Robman]

Wanna have a crack at it?

[Vyrolan]

File 5: Ortek Mouse2.ir
This is the results of the previous step. Open this file using IR.exe (or the version of RMIR that has Vyrolan's changes) and go to the Learned Signals tab. Select an appropriate rounding value, and (if necessary) a Bi-Phase value, and click the Times Summary button. Then cut & paste the contents of the pop-up screen into another Notepad session.

Somewhat pleased that the auto analysis got most of those correct. It looks like it missed on a few...I'll have to look into why.

So, can any of these steps be automated? I don't really care which program it all ends up in (ie, IR, RMIR, IRScope, IRMaster, etc) but I would really appreciate not having to do all this manual work whenever someone posts learns that contain a new protocol in an ICT file rather than an IR file.

I still feel like RMIR should be able to import learns from an ICT file directly. So you would just click an "Import" button the Learned Signal tab and then pick your ICT file and have them all show up...then just click the Timing Summary.

[mdavej]

Keep in mind these particular signals aren't very clean, as far as I know. The original remote has a round pad that sends one of 16 possible commands depending on where it's touched. If you don't press it just right, you'll get a mix of 2 or more OBCs. I don't know if you're actually seeing this case in some of the ICT files, but it's a possibility. Just wanted to give you guys a heads up since I did the captures and am not very proud of them.