Related articles |
---|
Non-sequential compilation. KODIS@delphi.com (1993-09-18) |
Re: Non-sequential compilation. macrakis@osf.org (1993-09-21) |
Re: Non-sequential compilation. cliffc@rice.edu (1993-09-21) |
Re: Non-sequential compilation. pcg@aber.ac.uk (1993-09-21) |
Re: Non-sequential compilation. conway@mundil.cs.mu.OZ.AU (1993-09-22) |
Re: Non-sequential compilation. gafter@mri.com (1993-09-27) |
Re: Non-sequential compilation. rbe@yrloc.ipsa.reuter.COM (1993-10-04) |
Newsgroups: | comp.compilers |
From: | conway@mundil.cs.mu.OZ.AU (Thomas Charles CONWAY) |
Keywords: | design, parallel |
Organization: | Computer Science, University of Melbourne, Australia |
References: | 93-09-068 93-09-076 |
Date: | Wed, 22 Sep 1993 06:33:45 GMT |
macrakis@osf.org (Stavros Macrakis) writes:
>I'm not sure it's terribly useful to read the whole _source_ into memory
>at once; after all, you only want to do tokenization and parsing once. On
>the other hand, it may well be useful to keep large intermediate
>representations around for global optimization.
It is entirely possible that reading the whole source in could be useful
in languages where the lexical and syntactic rules are not static - for
instance where new types may be declared - or new operators with arbitary
precidence; especially if you want to allow them to be used before they
are declared.
Thomas
--
| Thomas Conway
| Computer Science
| Melbourne University
| conway@mundil.cs.mu.oz.au
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.