Nick Roberts wrote:
>  
> Looks good.  Maybe varobj_update could use an enum:
>
> enum varobj_update_values
>   {
>     SCOPE_FALSE = -1,
>     TYPE_CHANGED,
>     SCOPE_INVALID
>   }
>   
Yes, it's better and more readable.
As you'll see in my patch, the varobj_update function could return more 
than these case of error, it's the caller that decides how to deal with 
these errors.
>
> (It looks like we could remove the return value of varobj_update_one as it
> doesn't seem to be used.)
>   
You're right, I removed it as well.

Attached is the new implementation plus the new exp file.

Denis