Re: GScheme. (Bug fix re. booleans.)

[Stefan Urbanek]

Hi,

Ok, I've unpacked the archive and did 
        > make debug=yes
        > debugapp GScheme.app
...
The backtrace after segfault is the same:
(gdb) bt
#0  0x8061ef8 in yy_get_next_buffer () at lex.yy.c:1279
#1  0x8061bc5 in yylex () at lex.yy.c:1114
#2  0x8066fab in main ()
#3  0x405e314f in __libc_start_main () from /lib/libc.so.6
(gdb) 

...and there is nothing more. Strange is, that yylex() is called from main(),
but in the main.m there is no such code. And there is no lex.yy.c.

It seems that some another main() is used.

I have no idea, what is going on here.


Marko Riedel wrote:
> 
> Hi there,
> 
> it's hard to tell but I noticed a couple of things:
> 
> - you should compile with the -g flag so that gdb has more data about
> where the segmentation fault occured
> - your backtrace is missing a couple of frames, e.g. yylex is called
> from yyparse which is called from VScheme, which doesn't show up in
> your trace
> - you shouldn't be compiling lex.yy.c at all; you should run the
> following commands:
> 
> flex scheme.flex
> mv lex.yy.c scheme.lex.m
> 
> The file scheme.lex.m is listed in the original GNU makefile.
> 
> Best regards,
> 
> --
> +------------------------------------------------------------+
> | Marko Riedel, EDV Neue Arbeit gGmbH, mriedel@neuearbeit.de |
> | http://www.geocities.com/markoriedelde/index.html          |
> +------------------------------------------------------------+

-- 
Home page: http://urbanek.host.sk