RMIR v2.12 available

greeno · Post by **greeno** » Wed Jun 16, 2021 7:39 am

thanks barf, i'll make a mac specific thread. i dl'd irscrutinizer from the link you provided and it launches and runs just fine natively. yea i agree its a java issue, but i'm over my head with that.

best,
jeff

Barf · Post by **Barf** » Sat Jun 19, 2021 8:46 am

Just for the fun of it:

While looking at greeno's thread, it occurs to me that the class DynamicURLClassLoader appears to be a solution in search of a problem. The only thing it is (presently) used for is to locate icons (RemoteMaster.createIcon(String)), and there are much easier ways to do that than to deploy a custom ClassLoader. (Tried imageURL = RemoteMaster.class.getResource( "/" + imgLocation ) -- works.)

According to SVN it was checked in by user yawor. Marcin, are you there?

Post by **mathdon** » Sat Jun 19, 2021 11:52 am

Barf wrote:While looking at greeno's thread, it occurs to me that the class DynamicURLClassLoader appears to be a solution in search of a problem.

If you look up revision 1595 in which Marcin introduced this, he describes it as "Attempt at fixing Java 9 and 10 compatibility issue". It apparently replaced an earlier module ClassPathAdder.java that I haven't looked up. So there was an issue to be resolved, though I have no idea if it was a sensible way of doing it. These Java issues are way beyond my understanding.

Barf · Post by **Barf** » Tue Jun 29, 2021 7:40 am

Graham, it appears that in the recent check-in, you created a new file JP11USB.java, but forgot to check it in.

Post by **mathdon** » Tue Jun 29, 2021 9:48 am

Barf wrote:Graham, it appears that in the recent check-in, you created a new file JP11USB.java, but forgot to check it in.

Sorry about that, it is there now.

Barf · Post by **Barf** » Thu Jan 20, 2022 9:25 am

I just tried to export this file as Girr. The file uses the NEC2 protocol, Device=50, Sub Device defaulted. Turned out that the export was erroneous; it had S (subdevice) = 0. Correct would have been to leave it out (or to have S=255-D=155).

Post by **mathdon** » Thu Jan 20, 2022 12:40 pm

@Barf
Your post is in the wrong thread, as the current RMIR thread is that for RMIR v2.13. The bug causing your error appears to be a very longstanding error in protocols.ini. If you look at the entry for the NEC 4DEV Combo you will see

Code: Select all

DevParms=Device 1,Sub Device 1=[-0],...

with similar entries for the other three Device / Sub Device pairs. The Sub Device default [-0] means that the default is the complement of the device parameter with index 0, which is Device 1. A number of other NEC protocols have similar entries, but those that use PID=00 5A do not. This is not normally significant as the NEC005ATranslator handles the default for this executor, but it matters for the export to a Girr file. So the Sub Device default for protocol.ini entries with PID=00 5A needs to be changed from 0 to [-0]. If you try this, you should find that it fixes the problem.

There are a few NEC entries that use other executors and have a Sub Device default of 0. Unless I have missed something, these are NEC-34 and NEC1-rnc, both of which I suspect are little used. Unless you think otherwise, I will change these too to [-0] for the next release of RMIR.

Barf · Post by **Barf** » Thu Jan 20, 2022 2:55 pm

mathdon wrote:@Barf
Your post is in the wrong thread,

The bug causing your error appears to be a very longstanding error in protocols.ini. If you look at the entry for the NEC 4DEV Combo you will see
Code: Select all
DevParms=Device 1,Sub Device 1=[-0],...
with similar entries for the other three Device / Sub Device pairs. The Sub Device default [-0] means that the default is the complement of the device parameter with index 0, which is Device 1. ... Unless you think otherwise, I will change these too to [-0] for the next release of RMIR.

Sounds good!