Re: [Reuseware-users] problems with reuseware accepting code

[Jendrik J. <jen...@tu...>]

Hi Andreas,

please apologize for the late reply.

At first, the "*dummy();" thing works,
because according to the grammar "*dummy();" is already a complete
program and resueware parses only this part. If it finds one complete 
programs it attempts to parse the remaining part as well, but gives no 
error if it fails (this might actually be a bug).
	
This means the example can not be parsed regardlessly if you put 
"*dummy{};" in front of it or not. Consequently there must be mistakes 
in the grammar. The grammar is quiet big, so it is hard for me to get an 
overview.

One problem seems to be the return type of a method:

* main( int argc, char *args[] ) {

}

is parsed, while

int main( int argc, char *args[] ) {

}

is not.

Another problem is there with array declarations:

     int array;

is okay, but

     int array[100];

gives errors.

The reuseware tooling is not very advanced for grammar debugging. If you 
can not find the mistakes, one thing you could do is looking at the 
generated ANTLR file "c.g" in
"de.tudresden.reuseware.language.c.textsyntax.c"
and analyze it with some ANTLR tool. I think they have a good graphical 
tool now, but I have not tried it myself (www.antlr.org).

Hope that helps,
Jendrik


Andreas Leha wrote:
> Hi there,
> 
> I am trying to get reuseware to understand (at least partial) C code. 
> Unfortunatly I experience strange behaviour: reuseware accepts this:
>  *dummy{};
> 
>  int main( int argc, char *args[] ) {
>    int index;
>    int array[100];
> 
>    for(index = 0; index < 100; index++) {
>      array[index] = 0;
>    }
> 
>    return 0;
>  }
> But without the leading dummy line, the code raises errors. Strange is, 
> that just regenerating language plugins and text parsers leads to 
> different errors here. Anyway, all the errors are similar to this:
> state 0 (decision=75) no viable alt; token=[@5,10:12='int',<97>,1:10]
> or
> mismatched token: [@5,10:12='int',<97>,1:10]; expecting ')'
> And they indicate, that function definitions or array declarations... 
> are not recognized.
> 
> So my question is: Why does reuseware accept C code just after the 
> leading line "*dummy{};"?
> 
> I've attached the grammars (that follow closely the K&R, 2nd edition).
> 
> 
> Regards,
> 
> Andreas Leha
> 
> 
> ------------------------------------------------------------------------
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Reuseware-users mailing list
> Reu...@li...
> https://lists.sourceforge.net/lists/listinfo/reuseware-users


-- 
Dipl.-Medieninf. Jendrik Johannes
Research Assistant

Technische Universität Dresden
Department of Computer Science

Phone +49 351 463 42073
Fax   +49 351 463 38459
Email jen...@tu...