protocols.ini question

jon_armstrong · Post by **jon_armstrong** » Mon Jul 19, 2004 5:05 pm

I was trying to fix the "remote address" that is 1 to 32 for Dish that corresponds to what we call sub-device 0 to 31.

I'm not getting the adjustment parameter to work correctly in the device or command translators. It is in bold. This is an ugly case where the Unit code/remote address is split between fixed and variable data.

I did get it to work on the other variants of the protocol where all the bits are contiguous and in fixed data.

I'm just trying to see if I missed something obvious.

[Dish Network]
PID=00 02
DevParms=Remote Address (1 to 32) :5=1,Device Code:5=0
DeviceTranslator=Translator(lsb,0,3,0,2,-1) Translator(lsb,1,5,3)
CmdParms=OBC:6=0
CmdTranslator=Translator(0,6) TranslatorFromDev(lsb,0,2,6,0,-1)
Notes=Standard SAT/0775 is Unit 0 / Device 0 \
\n\n\
Device, Unit and OBC are all incompatible with KM (version 7.34) \
They are compatible with DecodeIR.dll as Device.Unit:OBC \
EFC values are compatible with KM within the same model remote. \
EFC values are compatible across models of remote for the default unit \
number and in some cases for other unit numbers.
FixedData=00
Code.S3C80=2B 5C 11 8B 12 B5 05 08 08 00 C5 03 32 00 C5 05 \
73 00 00 00 C5 0B F2 8D 01 46
Code.6805-C9=06 0E 21 20 12 B8 72 17 05 06 03 18 04 99 02 A3 \
17 08 E9 00 04 01 29 07 5B 08 B6 75 B7 6E B6 76 \
B7 6F 33 5A 3C 59 CC 01 83
Code.6805-RC16/18=0B 18 11 20 10 B5 45 07 08 01 62 A3 02 62 C4 02 \
C4 00 06 62 03 CC 01 AF
Code.740=06 0D 11 80 0E B0 72 17 08 07 03 19 04 9A 02 46 \
17 08 EC 44 5D 4C 00 FF

gfb107 · Post by **gfb107** » Tue Jul 20, 2004 7:46 am

I think the problem is that the adjust parameter, in this case -1 is applied after extraction of the bit field, rather than before.

So, if Remote Address is 30 ($1E), then the 1st device translator extracts lsb bits 2-5, which is 7, which is adjusted to 6.
The 2nd CmdTranslator ,the TranslatorFromDev, extract bits 0-1, which is 2, which is adjusted to 1.
This is not the behavior you want here.

I think adjusting the parm value before extracting the bits might solve the problem here.

I'll have a fix available today, and will include your protocols.ini update as well. I haven't had a chance to work on the RC-5/RC-5x combo stuff, so that won't be included.

jon_armstrong · Post by **jon_armstrong** » Tue Jul 20, 2004 9:52 am

gfb107 wrote:I think the problem is that the adjust parameter, in this case -1 is applied after extraction of the bit field, rather than before...
I think adjusting the parm value before extracting the bits might solve the problem here.

That makes a lot of sense to me.

I'll have a fix available today, and will include your protocols.ini update as well. I haven't had a chance to work on the RC-5/RC-5x combo stuff, so that won't be included.

I posted a protocols.ini entry for Rob's new DishNet combo protocol in this thread. I also have two other corrected variants of $0002 so the Remote Address will be consistent with KM and DecodeIR.dll. You want me to post them too? Is there a better way to get them to you?

Also, until you do the elegant fix to RC5/5x you might include my version. It does work and I put some comments in the notes that describe how to use it and to ignore the variables that change in what ever protocol they are not using.

gfb107 · Post by **gfb107** » Tue Jul 20, 2004 12:07 pm

A PM or email might be a good way to make sure I see these.

Alternatively, you can check your changes directly into SourceForge, and then I can rebuild and repackage as needed.

I'll include RC5/5x in the next release whether or not I get a chance to implement the "elegant UI".