From mboxrd@z Thu Jan  1 00:00:00 1970
From: dave@horsfall.org (Dave Horsfall)
Date: Sat, 13 Sep 2014 06:02:26 +1000 (EST)
Subject: [TUHS] Etymology of bc(1)
In-Reply-To: <1410529036.3893715.166731793.44CF06BA@webmail.messagingengine.com>
References: <877g19g8a7.fsf@gmail.com>
 <7A7060B7-0229-425F-BCF4-23C3669E167C@bsdimp.com>
 <alpine.BSF.2.00.1409120800490.63870@aneurin.horsfall.org>
 <20140911221742.GJ23708@mercury.ccil.org>
 <alpine.BSF.2.00.1409121055300.63870@aneurin.horsfall.org>
 <1410529036.3893715.166731793.44CF06BA@webmail.messagingengine.com>
Message-ID: <alpine.BSF.2.00.1409130539280.63870@aneurin.horsfall.org>

On Fri, 12 Sep 2014, random832 at fastmail.us wrote:

> What language does the undocumented option (assuming it is supported at 
> all) "bc -c" generate on FreeBSD and OSX? "Standard" bc, which actually 
> pipes to dc, generates (obviously) dc when run in this way, but GNU bc 
> generates a completely different and as far as I know undocumented 
> language, which is handled by the execute function (in execute.c) but as 
> far as I know there is no way to make it accept it on standard input.

According to my trusty Mac:

DIFFERENCES
    This version of bc was implemented from the POSIX P1003.2/D11 draft and
    contains several differences and extensions relative to the  draft  and
    traditional  implementations.  It is not implemented in the traditional
    way using dc(1).  This version is a single  process  which  parses  and
    runs  a  byte  code  translation  of the program.  There is an "undocu-
    mented" option (-c) that causes the program to output the byte code  to
    the  standard  output  instead  of  running it.  It was mainly used for
    debugging the parser and preparing the math library.

My trusty FreeBSD box says the same thing.  Well, they're both Gnu 1.06, 
after all, and it's some sort of byte-code.  I can sort see how it works; 
single digit integers, unless wrapped by "Knnn:" etc..

> Standard bc:
> $ echo '2+2' | bc -c
>  2 2+ps.
> q$
> 
> GNU bc:
> $ echo '2+2' | bc -c
> @iK2:K2:+W at r
> @i
> $

Oops; a "1" stands for itself, but other ints are bracketed:

davehorsmacbook:~ dave$ bc -c
@ibc 1.06
Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'. 
1+1
11+W at r
@i2+2
K2:K2:+W at r
@i

Amusing that it treats "1" and "not 1" differently.  Oh, and "0" stands 
for itself.  One more test:

00+W at r
@i-1+1
1n1+W at r

So, that's something like "1 negate 1 add" - gasp, that's RPN!  I'd check 
the sources if I had them (none for Mac, and I didn't bother with FBSD).

-- Dave