[antlr-interest] 0xcdcdcdcd myth with antlr base tree pointer u

[Jim Idle]

Also, I am wondering if you are taking the 'right' approach with your  
code. If you send the source I'll take a look.

Jim

On Jun 26, 2009, at 1:24 PM, "Xie, Linlin" <linlin.xie at siemens.com>  
wrote:

> Hi Jim,
>
>
>
> I’ve noticed that that empty pointers are all (am I right?) initiali 
> zed to be 0xcdcdcdcd by default. Looks like it’s a debugging value c 
> oming from the C runtime library. I wonder if you have your reason f 
> or this to work this way, or better to initialize them to be 0? it’d 
>  be helpful for us that we don’t have to check against this value wh 
> ich is perhaps not platform independent.
>
>
>
> Also, I made some small changes in the antlr 3.1.3 to fix the  
> previous two bugs I mentioned, and a change to make the @after  
> action work for our purpose. I wonder if I should submit to the  
> antlr repository to get checked if they are good fixs, and also I  
> wonder we can push you a little bit to get @after supported soon J?
>
>
>
> Thanks a lot.
>
> Regards,
>
> Linlin