From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,3fc79ad704f81a40 X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 1995-03-15 07:19:27 PST Path: bga.com!news.sprintlink.net!cs.utexas.edu!swrinde!hookup!news.mathworks.com!mvb.saic.com!buckie.ucha!tali.hsc.colorado.edu!boulder!news.coop.net!news.den.mmc.com!iplmail.orl.mmc.com!usenet Newsgroups: comp.lang.ada Subject: Re: Writing a Mac Ada compiler Message-ID: <3k70hv$g30@theopolis.orl.mmc.com> From: Theodore Dennison Date: 15 Mar 1995 15:19:27 GMT References: <3jg167$a3l@nic.umass.edu> <3joj6a$plb@felix.seas.gwu.edu> <3k0j74$6fj@zoom.jtec.com.au> Organization: IPL InterNetNews site NNTP-Posting-Host: payday.orl.mmc.com Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 7bit X-URL: news:3k0j74$6fj@zoom.jtec.com.au Date: 1995-03-15T15:19:27+00:00 List-Id: Fraser Wilson wrote: > I nearly cried when mfeldman@seas.gwu.edu (Michael Feldman) explained: > > >Ada compilers are not easy. Best to reuse something, especially if it > >is free. > > It's true they're not easy. But writing an Ada compiler is a great > way to learn how compilers work. > .. > > And even if it's ultimately a useless endeavour, it was fun. I don't think we meant to imply that it was "useless". Personally, I'm a firm believer that learning is never a waste. Its just that trying to craft a complete implementation of Ada single-handedly in one's spare time is practicly a hopeless task. The image that springs to mind is that of the old man in Indiana in the seventies who was building his own turnpike (by himself). Sure, you'll learn a lot, you'll just never finish it. I would like to suggest that, for learning purposes, a smaller language would be more suitable. This way, you are more likely to get around to building a code generator (to go with your parser and lexical analyser). Pascal and C would probably be good choices. Oberon would also be a good choice, as a book (Written by Nicklaus Wirth) is avaiable on building compilers that uses Oberon as an example. The complete source to Oberon is gone over in detail during the course of the text. Of course, if its really fame and glory you are after, porting GNAT would be the better choice. :-) T.E.D.