Da
Show enters and exits. Hide enters and exits.
| 00:01:26 | evan | :( |
| 00:01:49 | evan | why not call it where you are |
| 00:02:40 | brixen | I have a few more |
| 00:05:03 | evan | ok |
| 00:30:50 | evan | finally getting the thread state nicely untagled. |
| 00:30:53 | evan | untangled. |
| 00:34:19 | brixen | sweet |
| 00:34:40 | evan | and we pass the long-return-doesn't-kill-$! spec |
| 00:34:49 | brixen | awesome! |
| 00:34:52 | evan | which is really not yet a spec |
| 00:34:57 | evan | but rather a script i need to make into a spec |
| 00:35:20 | brixen | yep, often have to do that |
| 01:28:49 | alexhq | The note on rubinius homepage "Rubinius is not without its flaws. Namely, ObjectSpace#_id2ref, ObjectSpace#each_object(limited). At least, Not yet." is outdated and can be removed. Right? |
| 01:31:20 | evan | yeah |
| 01:31:22 | evan | i need to update that |
| 01:31:23 | evan | my mistake. |
| 01:40:38 | rue | Previously not without those flaws, now with :) |
| 04:25:51 | manveru | evan: should i open some tickets for the broken links on rubini.us? |
| 09:01:26 | matthewd | Well, that's fun.... Kernel.rand(n) is broken for n > 46408 |
| 09:02:23 | matthewd | (for values of "broken" that are equal to "can return n") |
| 09:05:10 | matthewd | Oh, actually only for some n > 46408, such as: 46409, 46567, 46653 |
| 09:06:20 | matthewd | I don't look forward to consistently demonstrating that in a spec :) |
| 09:49:02 | matthewd | Erm, looking closer, I think it's broken for any value of n |
| 09:51:27 | matthewd | At the extreme, Math.rand(1) has a 1 in 2147483648 chance of returning 1 (when FFI::Platform::POSIX.rand returns RAND_MAX) |