home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!oasys!roth
- From: roth@oasys.dt.navy.mil (Pete Roth)
- Newsgroups: comp.lang.pascal
- Subject: Re: Comments on Turbo (Borland) Pascal 7.0?
- Message-ID: <27741@oasys.dt.navy.mil>
- Date: 20 Nov 92 20:01:26 GMT
- References: <1992Nov18.131557.7438@odin.diku.dk> <1992Nov18.174335.2758@exu.ericsson.se>
- Reply-To: roth@oasys.dt.navy.mil (Pete Roth)
- Organization: The David Taylor Model Basin
- Lines: 24
-
- In comp.lang.pascal, exuhag@exu.ericsson.se (James Hague) writes:
- [...]
- >And on a vaguely related subject, here's a neat idea: since the
- >editor already has to go through the trouble of doing lexical
- >scanning (to determing proper colors), maybe it should be taken
- >one step further. Keep the files, both internally and on disk,
- >in a tokenized format. This would allow better memory usage (smaller
- >files) and would speed up the compiler. Many assemblers and a
- >few compilers on 8-bit computers used to use such a scheme.
-
- naah, not so neat.
-
- Okay, maybe the new IDE has a bunch of good stuff plugged in - I think
- color coded code is cool..., and, alliterative.
- But tokenized files should be an option that is not important.
- Because frankly, the editor in the IDE is pretty puny.
- Most of the time I put my programs together outside of the IDE with
- decent tools. And most of us have reasonably quick machines now anyway,
- nicht wahr? So where's the beef in a tokenized file?
-
- regards, pete
- - - - - - - - - - - - - - - - - - - - - - - - - - -
- Peter N Roth roth@oasys.dt.navy.mil
- You can take a horse to water, but a pencil must be lead.
-