Related articles |
---|
lex/flex question joe.macdonald@acm.org (1999-10-06) |
Re: lex/flex question rkrayhawk@aol.com (1999-10-11) |
From: | rkrayhawk@aol.com (RKRayhawk) |
Newsgroups: | comp.compilers |
Date: | 11 Oct 1999 02:30:58 -0400 |
Organization: | AOL http://www.aol.com |
References: | 99-10-028 |
Keywords: | lex, code |
Joe MacDonald, joe.macdonald@acm.org
<<
I'd rather not spam the group with my lex code unless someone wants to see it.
>>
Serious code is not spam. Perhaps you could let us see some of your
code. If this is a bug, many of us can gain from observing the process
of isolating it.
Further, if it is a bug in flex then there may be a way around it. If
you do consider posting some code, change some names to protect your
ideas, and just post a portion (or a few small portions). From what
you have said so far, it seems like the weekness in the utility (flex)
is beig ferreted out by the code which is in front of the code that
sneaks into your .cc file.
Some of the 'gurus' may be able to tell you how to get your results
another way, while awaiting a fix. So if you choose, tell us a little
about what the specific lines of code are trying to do.
You probably only need to post a few lines _precceding_ the
symptom(s), with maybe some comments about how things are declared
(especially the start states).
Also tell us how much junk is getting pumped into your .cc file.
You said your compiler barfs on things like
<SC>[^"] { ... }
which is not surprising, but does the lexer also pump a portion of the
preceding lex line into the .cc file? The point is not to guess at the
problem but to emphasize that we might better be able to help you if
we know the exact symptom in the .cc file.
Best Wishes,
Robert Rayhawk
RKRayhawk@aol.com
[If possible, I prefer that people put their code on a web or FTP server
and post a URL for interested parties. -John]
Return to the
comp.compilers page.
Search the
comp.compilers archives again.