On Mon Mar 21 20:40:19 2011, sprout wrote: > > On Mar 20, 2011, at 4:30 PM, Jesse via RT wrote: > > I'd sort of been waiting to see the testing of edge > > cases get sorted out. > > I’m afraid I had forgotten about the edge cases. Now that I think > about it though, we have a chicken-and-egg problem. There is no > problem with push, splice, etc., as the edge cases are all > currently errors and can be resolved later without backward > compatibility concerns. The edges cases for keys, each and values, > however, entirely depend upon your decision with regard to array > refs and objects. Attached is a first go at a test script. Edit the JC constant to choose between the different behaviours tested for. BTW, I noticed that ck_push is duplicating some logic from ck_fun. That was why in my first patch to this ticket I deleted ck_push and added some code to the array section of ck_fun.Thread Previous