Re: Compiler/Language eXperiment (Rick Hohensee)
30 Jan 2002 20:42:03 -0500

          From comp.compilers

Related articles
Compiler/Language eXperiment (Axel Kittenberger) (2001-11-29)
Re: Compiler/Language eXperiment (Tomek Zielonka) (2002-01-24)
Re: Compiler/Language eXperiment (2002-01-28)
Re: Compiler/Language eXperiment (Toon Moene) (2002-01-28)
Re: Compiler/Language eXperiment (Toon Moene) (2002-01-30)
Re: Compiler/Language eXperiment (2002-01-30)
| List of all articles for this month |

From: (Rick Hohensee)
Newsgroups: comp.compilers
Date: 30 Jan 2002 20:42:03 -0500
References: 01-11-135 02-01-103 02-01-148
Keywords: design, syntax
Posted-Date: 30 Jan 2002 20:42:03 EST

Toon Moene <> wrote in message news:02-01-148...
> Tomek Zielonka wrote:
> [ ... snip ... ]
> > [Back in the 1970s, languages with extensible syntax led to write-only
> > code with no two programs using the same syntax. Have they solved that
> > problem? -John]
> I thought Perl was the solution to that problem ?
> "There's more than one way to do it" (TM).

Forth's syntax is basically no syntax. Words can be executed as soon
as parsed, unless the particular word does some parsing. You can at
any time extend the parser. Forth also has no lexer. Tokens are
whitespace-delimited, period. I adhere to that in my stuff. To put it
another way, punctuation is words. ( is the same class of thing as

Forth is postfix. What that really means is that numbers are verbs;
put yourself on the stack. Seen that way, that's no syntax. That's
"words occur when parsed".

Rick Hohensee

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.