Re: AT&T lex port

toor@y1.jdyson.net (John S. Dyson)
22 Oct 2000 01:10:24 -0400

          From comp.compilers

Related articles
AT&T lex port jbrink@chat.carleton.ca (2000-10-19)
Re: AT&T lex port toor@y1.jdyson.net (2000-10-22)
Re: AT&T lex port broeker@physik.rwth-aachen.de (Hans-Bernhard Broeker) (2000-10-22)
Re: AT&T lex port paule@martex.gen.oh.us (Paul Evans) (2000-10-23)
Re: AT&T lex port paule@martex.gen.oh.us (Paul Evans) (2000-10-26)
Re: AT&T lex port g.douglas@readytext.co.uk (Graham Douglas) (2001-01-09)
| List of all articles for this month |
From: toor@y1.jdyson.net (John S. Dyson)
Newsgroups: comp.compilers
Date: 22 Oct 2000 01:10:24 -0400
Organization: http://extra.newsguy.com
References: 00-10-145
Keywords: lex

jbrink@chat.carleton.ca (Jason Brink) writes:
> Does anyone know of a port of AT&T lex to Windows NT/2000 or Linux? I
> know flex is better than AT&T lex, but I'm currently doing a co-op
> term and working on a large project, which runs on Solaris, HP-UX,
> AIX, etc. All those systems have AT&T lex, and so that is what the
> lexers in the project were written for. Now we would like to port it
> over to Windows 2000 and Linux. I've tried using flex on some of our
> lexers, and then compiling the generated code, and there are lots of
> problems. Any help would be appreciated!
>
> Jason Brink
> jbrink@chat.carleton.ca
> [I don't think that AT&T lex has ever been freed from the onerous
> System V source license. Having seen the source code, I can say that
> it's some of the ugliest C code I've ever seen. -John]


I *strongly* suggest porting FLEX or an equivalent clone, rather than
using AT&T LEX. Even if one has to port FLEX to all of the target
platforms, I think that the effort of porting will be nil.


This is just FWIW, and only a personal and kindly meant opinion.


John
[I entirely agree. AT&T lex is slow and buggy. An entirely reasonable
approach is to run the lexer through flex once, then port the generated
C code, since flex writes rather good portable code. -John]





Post a followup to this message

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