Re: Compilers For Typeless Languages

In article <1989Sep21.051232.1868@esegue.segue.boston.ma.us> Will@cup.portal.com writes:
    >Do any of you who have written interpreters or compilers for
    >typeless languages have any thoughts on whether it is worthwhile
    >to even bother writing a compiler for the language?
    >
    >Let's say I have a language, X, where every variable is treated by
    >the user as a character string, and the interpreter knows that in
    >certain contexts a string implies a non-character type of data and
    >operation (e.g., "2/3", "3.343434 * 23423.23423"). The user
    >doesn't ever declare a variable to be of any type; datatypes are
    >implied by the use of a variable. Since in many situations there
    >is no a-priori way to know whether a variable is going to be used
    >exclusively as a character type, integer, or floating point type,
    >a compiler for such a language would need to include run-time
    >support to determine the way in which variables were being used as
    >statements are executed.
    >...
    > Will
    >[These issues have been hashed over in the Lisp community for
    >30 years, and most Lisp systems include compilers. Anybody have some
    >good references on compilation issues? -John]
    >[From Will@cup.portal.com]


The difference between Will's scenario and Lisp (modern Lisps, anyway)
is that Lisp's data is typed. For example, you can't add two numbers
together and then concatenate a string to the result. Unless you do
something like


                (string-append (number->string (+ 1 2)) " more characters")


which explicitly creates a new string before the concatenation.


The "command language" here at Mentor treats variables and data in a
way similar to what Will described. To me it is very confusing with
little or no benefit, especially if procedures, like "number->string"
above, exist to perform conversions where they make sense. Then there
is more benefit from a compiler and one can tap into the lore of Lisp
and similar languages. In fact, one might choose Lisp or to build the
language on top of Lisp.


See Byte, 2/88 for more on creating languages within Lisp, and
        Kranz's PhD dissertation at Yale for more on compiling Lisp-like
        languages.


See also articles on SELF for interesting new work on compiling
        languages with type-less variables and typed data (like Lisp,
        Smalltalk, and SELF). They even have application to C++, if
        you're interested in that sort of thing.


        These articles have appeared in OOPSLA '87 and SIGPLAN Notices,
        July 1989. Another will be presented next week at OOPSLA '89.


--
Patrick Logan | ...!{decwrl,sequent,tessi}!mntgfx!plogan
Mentor Graphics Corporation | plogan@pdx.MENTOR.COM
Beaverton, Oregon | "My other computer is a Lisp machine."

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