 | | Re: NQC 2.1 b2 in beta test
|
|
(...) You are wrong. The precedent, naturally, is C, coz NQC, isn't quite, err, C. Backslash _is_ the escaping character in C, even on Windoze. So, to write \a\b\c as a correct C string (yes, even a VC++ one), you have to write "\\a\\b\\c", and if (...) (25 years ago, 16-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) Unfortunately, I don't know "official" rules nor did I find a place where to look them up. Nevertheless, I can only think that this whole escaping stuff is not compliant with normal, expected Win32 behavior. The backslash is the standard (...) (25 years ago, 16-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: offtopicy sort of thing
|
|
According to that elementary physics, you are converting power into heat as part of this process. Usually, when downgearing an engine, the intention is to have more power at the slower speed (for better climbing, lifting, etc.). This will not be the (...) (25 years ago, 14-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) the exec (...) The answer isn't pretty.... As near as I can tell (sorry, not much of a Windows expert), the command line gets passed in its entirety to the executable, which then is responsible for parsing it into separate arguments. For the (...) (25 years ago, 15-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC default output file
|
|
(...) I just added the error check to nqc. I haven't changed the directory yet, though. Using the source directory was strictly from laziness. Assuming there aren't massive objections, I'll switch to using the current directory. Dave Baum (25 years ago, 15-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) The same problem was already present with the b1 (I had asked about it in this newsgroup but got no reply). Also, I don't think it is related to the parameters. I have now checked the exec call of RcxCC with a debugger and found this is the (...) (25 years ago, 14-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) Hmmm - the only reason I can think of for this would be if RcxCC is using one of the deprecated options that were removed in 2.1 (-o, -e, and -s which were replaced by -O, -E, and -S). If Mark can confirm which (if any) of these options are (...) (25 years ago, 14-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: offtopicy sort of thing
|
|
(...) It's offtopic in .nqc, but not in lugnet.robotics in general. (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: offtopicy sort of thing
|
|
I think that's a really good idea. I tried it, and found out that the motor recieving power spins slower than the motor you spin. This is about 4:3. This could be a really good speed reduction technique that doesn't use gears, or you could power (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC default output file
|
|
(...) I don't want the outfile to go where the executable is -- I want it to go to the current working directory. So nqc goes in /usr/local/bin, and I can run it from "binaries" directory on source files that might be located elsewhere (the test.nqc (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC default output file
|
|
I don't think it's nonintuitive. I think it makes sense to have the output go to the same directory as the input. That way you can have a directory set up for the executable with only read/execute permissions, and yet have all of your source & (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | NQC default output file
|
|
I was checking to see if NQC already does have an option to spit bytecodes to stdout (-L would work, if rcxcomm would understand the format) and I came across something that seems nonintuitive to me. The default output file is "named the same as the (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) If this is done, the compiler should have an option to spit out compiled bytecode on stdout, and the rcxcomm program should have a matching option to accept bytecodes on stdin and send them to the rcx. And then there should be a wrapper that (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) I think NQC should be strictly a compiler. It seems like it would be more manageable for both developers and users if the RCX communication pieces were in a separate executable. This change should make NQC completely portable (if there is such (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) The beta does not seem to work with RcxCC. All it does is display the parameters for NQC and then exits. No compile. --- DonC donc@cccd.edu (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 Linux binaries + rpm (was Re: NQC 2.1 b2 in beta test)
|
|
(...) Huh. Which brings me to an interesting conceptual correctness problem with RPM -- you don't have the documentation in .tar.gz format, which is how RPM wants it. So do I cheat and make my own tar.gz containing them, or do I cheat the other way (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) Yes, this is why I'd like someone else to implement this. I just want to use it. *grin* (...) RCX messages is what I'm interested in. It might be reasonable to have several modes. (...) Good question. I don't think this feature is out of line (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) I remember discussing this briefly, but I don't think we ever came up with a final "spec" for the feature. Here are the issues I forsee: 1) The IR tower times out. NQC should probably do something to keep it alive - and that "something" should (...) (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 Linux binaries + rpm (was Re: NQC 2.1 b2 in beta test)
|
|
(...) Go right ahead. Dave Baum (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
| |
| | Re: NQC 2.1 b2 in beta test
|
|
(...) that to (...) I don't see any reason why you wouldn't be able to copy the beta version of NQC on top of the one RcxCC normally invokes. I suspect Mark will update RcxCC shortly after 2.1 becomes final. Dave (25 years ago, 13-Jan-00, to lugnet.robotics.rcx.nqc)
|
