Re: [R] R class accessor problem in 2.6 - FIXED

2008-02-12 Thread Joseph Wang
The reason R-swig was not working on 2.6 was that the C code was not setting the S4 flag. On 2.5, R would still treat it as an S4 class, but in 2.6, it is necessary to have this flag set for it to be treated as S4. I've checked in the fix to swig-trunk. There were no C coding changes, just a

[R] R class accessor problem in 2.6

2008-02-03 Thread Joseph Wang
I was able to track down the problem that causes R-Swig to fail in 2.6. In 2.5, I am able to use the set method against '$' and '$<-' to cause expressions like r$d to get dispatched to get and set functions. This no longer seems to work in 2.6, and I was wondering if anyone can give informatio