Comments for What's up at Control-Q

Comment on Very light weight scripting … in Forth(?). by admin

admin — Thu, 19 May 2011 22:07:48 +0000

Hmmm … I’d have to look at the ANS spec to be sure, but look for it in 1.0.7.

Thanks.
Rob.

Comment on Very light weight scripting … in Forth(?). by ur2c

ur2c — Wed, 18 May 2011 09:29:10 +0000

Re: 00-01-06

Shouldn’t comment “(” be immediate?

Comment on Very light weight scripting … in Forth(?). by admin

admin — Thu, 12 May 2011 02:51:37 +0000

Actually, the depth thing is an order of operations issue, and differs between clang and gcc. Any number of definitions might have been simplified, and written as one liners, but to do so, would break clang code. In keeping, I’ve uglified a bunch of the entries for portability. Thanks for the input, though, I have applied the requested changes, and removed the clang warnings.

Comment on Very light weight scripting … in Forth(?). by crc

crc — Thu, 12 May 2011 00:00:55 +0000

Ok. First the issue with depth. You have:

void depth(){
push( tos – (Cell_t *) StartOf( stack ) – 1 ) ;
}

Which returns -1 if the stack is empty, 0 if one item is on the stack and so on. Removing the – 1 at the end will fix this.

Second, where you have:

#ifdef linux

Needs to be replaced with:

#if defined (linux) || defined (__APPLE__)

To allow building under OS X.

Third, the compiler warnings can be viewed at http://sprunge.us/HNbS

The unused variables are tolerable, but there’s a few missing casts that should be fixed.

Overall though, this is a nice little thing which I’ll be watching in the future. Thanks for sharing it.

Comment on Very light weight scripting … in Forth(?). by crc

crc — Wed, 11 May 2011 01:03:52 +0000

I found a bug in the depth() function, which was returning one less than the number of items on the stack. Also, the default source fails to build under MacOS X.

I put up a version at http://sprunge.us/FUae which fixes both of these, and also corrects all compiler warnings reported by clang.

Comment on Very light weight scripting … in Forth(?). by admin

admin — Tue, 10 May 2011 20:51:53 +0000

Yes, sorry Ed. If you download again the glitches for the mingw32 compile have been repaired. This single file runs primarily on Linux and Unix, and I test Windows *MUCH* less often. Thanks for the input!

Comment on Very light weight scripting … in Forth(?). by Ed Davis

Ed Davis — Tue, 10 May 2011 17:36:46 +0000

I tried to compile this on Windows with MingW. I get:

miniforth.c:201: error: conflicting types for ‘eof’
c:/bin/mingw/bin/../lib/gcc/i686-pc-mingw32/3.4.2/../../../../include/io.h:276:
error: previous declaration of ‘eof’ was here
miniforth.c:201: error: conflicting types for ‘eof’
c:/bin/mingw/bin/../lib/gcc/i686-pc-mingw32/3.4.2/../../../../include/io.h:276:
error: previous declaration of ‘eof’ was here
miniforth.c: In function `outfile’:
miniforth.c:1796: error: `S_IRGRP’ undeclared (first use in this function)
miniforth.c:1796: error: (Each undeclared identifier is reported only once
miniforth.c:1796: error: for each function it appears in.)
miniforth.c:1796: error: `S_IWGRP’ undeclared (first use in this function)
miniforth.c:1796: error: `S_IROTH’ undeclared (first use in this function)

Thanks for any tips!