flex++ bison++

Antonio Bleile <toni@rbg.informatik.tu-darmstadt.de>
6 Aug 2001 04:05:56 -0400

          From comp.compilers
Related articles
flex++ bison++ toni@rbg.informatik.tu-darmstadt.de (Antonio Bleile) (2001-08-06)
Re: flex++ bison++ toni@rbg.informatik.tu-darmstadt.de (Antonio Bleile) (2001-08-08)
| List of all articles for this month |
From: Antonio Bleile <toni@rbg.informatik.tu-darmstadt.de>
Newsgroups: comp.compilers
Date: 6 Aug 2001 04:05:56 -0400
Organization: ask - Innovative =?iso-8859-1?Q?Visualisierungsl=F6sungen?= GmbH http: //www.askvisual.de & Technische Universitaet Darmstadt
Keywords: lex, yacc, C++, question
Posted-Date: 06 Aug 2001 04:05:55 EDT

Hello all,


I'm a new subscriber to this newsgroup, so sorry if this might be off
topic (maybe you can suggest more appropriate groups).


I always use bison and flex for my C programs and I don't want to miss
these tools for C++ development. So I downloaded flex++ and bison++.
I found that it behaves very strange (especially under IRIX !!!). I
get segmentation faults and when I debug the program it tells me that
it stopped in a vector->push_back(...) operation (that's code in the
.y file, in the middle of a rule). But that's impossible, because the
program does NOT crash if i remove a certain line of code that is
processed AFTER this whole parsing, thus AFTER the push_back call.
The only explanation to me is a f..... up memory. I'm doing something
wrong with new and delete I guess.


Has somebody experienced these problems? What am I doing wrong? Please
note that this happens on an IRIX machine (O2, newest OS version...).


Best regards and thank you for reading,


Toni

Post a followup to this message

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