home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gatech!pitt!willett!ForthNet
- From: ForthNet@willett.pgh.pa.us (ForthNet articles from GEnie)
- Newsgroups: comp.lang.forth
- Subject: ANS TC Magnet for ROMability
- Message-ID: <4038.UUL1.3#5129@willett.pgh.pa.us>
- Date: 4 Sep 92 11:20:45 GMT
- Organization: EIEI-U
- Lines: 66
-
- Category 10, Topic 22
- Message 84 Thu Sep 03, 1992
- ELLIOTT.C at 12:07 EDT
-
- -----via CRS Premium Bulletin Board
-
- Date: 08-29-92 (03:09)
- To: B.RODRIGUEZ2 [BRAD]
- From: MARCEL HENDRIX
- Subj: Cross-CompilingOR ROMABIL
-
- Brad Rodriguez writes about ANS Forth Translators:
-
- br> definitions. As a minimum, the words
- br> : ALLOT CONSTANT CREATE IMMEDIATE VARIABLE
- br> from the Core word set must be available
- interpretively
- br> to create Forth programs. The arithmetic and stack
- br> operators, and such words as
- br> ' , C,
-
- The idea is very good, I particularly like the master stroke to
- differentiate between cross-compiling and translating.
-
- I'm not sure if you're proposal includes meta-compilation, that
- is, generating a complete working Forth interpreter for cpu a on
- platform b, using a xx Forth program running on cpu t on
- platform u.
-
- The process of meta compilation needs the ability to manipulate
- headers of both host and target words. Your proposals do not
- mention headers anywhere, so maybe a ``cross-compiler'' just
- generates a executable program for any hardware? (but note that
- this might include an outer interpreter and thus could need
- headers).
-
- I'm positively terrified that names like C, : ; become
- officially accepted as *meta* compilation word names. A meta
- ``:'' is completely different in concept from the host Forth
- ``:'' and the programmer should never (want) to forget this
- (He'll need both in a truly universal system). In t/iForth we
- use TC, T: and T; (among others), but at least two other meta
- compilers adopt a similar convention (F83, F-PC).
-
- An ordinary user will expect a Translator to let her build a
- completely new Forth system of her own design (why not?) but as
- soon as she starts defining : ; for the new system, she'll get
- dreadful conceptual problems (most users *I* know anyway).
-
- The problem can be resolved if it is officially stated that an
- ANS Forth System is not required to be able to completely
- regenerate itself. The translator need then not support the
- above features. Of course the last three lines should be a joke.
-
- Marcel Hendrix,
- Dutch Transputer Forth Workshop
-
- PCRelay:OLEF -> #181 RelayNet (tm)
- 4.11 OLEF BBS London England. +(44)-81-882-7353
-
-
- -----
- This message came from GEnie via willett. You *cannot* reply to the author
- using e-mail. Please post a follow-up article, or use any instructions
- the author may have included (USMail addresses, telephone #, etc.).
- Report problems to: dwp@willett.pgh.pa.us
-