Re: Algol W

rpw3@rpw3.org (Rob Warnock)
Fri, 24 Apr 2009 05:25:06 -0500

          From comp.compilers

Related articles
Algol 68 Genie Mark 16 - An Algol 68 interpreter algol68g@xs4all.nl (Marcel van der Veer) (2009-03-24)
Re: Algol W, was Algol 68 Genie Mark 16 - An Algol 68 interpreter gah@ugcs.caltech.edu (glen herrmannsfeldt) (2009-03-25)
Re: Algol W tk@ic.unicamp.br (Tomasz Kowaltowski) (2009-03-27)
Re: Algol W kym@svalbard.freeshell.org (russell kym horsell) (2009-03-31)
Re: Algol W rpw3@rpw3.org (2009-04-24)
| List of all articles for this month |

From: rpw3@rpw3.org (Rob Warnock)
Newsgroups: comp.compilers
Date: Fri, 24 Apr 2009 05:25:06 -0500
Organization: Rob Warnock, Consulting Systems Architect
References: 09-03-091 09-03-100 09-03-104 09-03-122
Keywords: algol60, history
Posted-Date: 24 Apr 2009 06:50:27 EDT
Originator: rpw3@rpw3.org (Rob Warnock)

[Sorry to be so late replying; "comp.compilers" got messed up in my ~/.newsrc]


russell kym horsell <kym@svalbard.freeshell.org> wrote:
+---------------
| Tomasz Kowaltowski <tk@ic.unicamp.br> wrote:
| } compilers-owner@lists.iecc.com wrote:
| } > [Algol W was somewhere between Algol 60 and Pascal, with records and
| } > defined I/O, and no call by name. It was different enough from its
| } > predecessors to be called a language. -John]
| } I have the impression that Algol W was designed so that it could be
| } parsed using simple precedence grammars defined by Wirth and Weber as a
| } generalization of operator precedence grammars.
|
| In the 70s there was an implementation in META2 for the DECsystem10.
| Boy, was META2 a can of worms. :)
+---------------


Was that Michael Green's implementation of Schorre's META II? The one
that was distributed with Green's ALGOL-W compiler? I appreciate why you
might say that [building it from scratch was... "interesting"], but on
the other hand, it was quite useful for some tasks. E.g., I successfully
used that implementation in 1970 to write a (toy) BLISS compiler in only
a weekend. [It compiled a usable subset of BLISS to MACRO-10 source.]


In 1991, Henry Baker did a very loose re-interpretation (or compilation)
of the META style into CL functions, macros, & readmacros to make META-
based parsers very concisely embeddable within a CL propram:


        http://home.pipeline.com/~hbaker1/Prag-Parse.html
        "Pragmatic Parsing in Common Lisp"


Jochen Schmidt typed in and reorganised Baker's code, and made it
available here:


        http://cclan.cvs.sourceforge.net/cclan/cclan/meta/


Another version by Fare Rideau is linked from here:


        http://www.cliki.net/meta


And, finally, yet a different version of Baker's code, assembled
by Henrik Motakef, is also available as a FreeBSD "port" under
the name "cl-meta".


Anyway, the last time I looked, Michael Green's versions of META II
and ALGOL-W were still available from the DECUS repository:


        http://pdp-10.trailing-edge.com/decus/10-186.html
        http://pdp-10.trailing-edge.com/decuslib10-02/01/43,50272/
        DECUS Library Program 10-186, "META2"


        http://pdp-10.trailing-edge.com/decus/10-133.html
        DECUS Library Program 10-133, "ALGOLW"


though the "trailing-edge.com" domain doesn't seem to be responding
at the moment, so you might have to use Achive.org's WayBackMachine...




-Rob


-----
Rob Warnock <rpw3@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607


Post a followup to this message

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