Christoph Otto wrote:
I'd appreciate comments or a quick code review as to whether I should apply the patch as-is (sans randomization) once the failing OrderedHash test passes. It's admittedly not a complete solution, but it does hide Parrot's hash seed from any PARROT_EXPORT functions and makes testing and fixing the other bugs easier.
After some testing, it turns out that the OrderedHash bug is orthogonal to this patch and can easily be reproduced by changing the value in src/hash.c:816 in trunk (e.g. to 3794).
Since I now know that the patch doesn't break anything new, I'd appreciate a +1 to apply it (with a fixed seed of 3793, matching the current behavior). I'll assume the patch [1] is ok to commit if I don't hear a negative responses by the 16th.
I'll also be working on the OrderedHash bug and anything else that's keeping randomized seeds from working.
[1] http://rt.perl.org/rt3/Ticket/Attachment/498952/234164/noseed_1.patch
