Markus Fischer wrote: > So, what do others thinkg about this? Is it really better to > stay with the current behaviour and document it that way?
Obviously, I would like this to be fixed :) One little thing is how LVALUE's hash position should be initilized. My common sense says bit by bit copy is the way to go. (i.e. Keep the same hash position as RLVALUE does) However, I'm fine with init position to 0 as far as object and array behaves the same way. -- Yasuo Ohgaki
