Da
Show enters and exits. Hide enters and exits.
| 09:38:55 | Malediktus | anyone tested rubinius with llvm 2.7? |
| 10:05:08 | khaase | Malediktus: i remember someone saying it works fine or something. at least there was talk about that. |
| 10:05:25 | khaase | Malediktus: macruby has issues with it, afaik. |
| 10:27:35 | Malediktus | ah, ok |
| 16:26:03 | brixen | Malediktus: rbx will not work with 2.7 without some changes in the jit |
| 16:26:20 | brixen | I'm sure evan will work on that as soon as 1.0 is released |
| 16:27:19 | brixen | khaase: the redmine mailer failures also are related to the regexp issue |
| 16:27:40 | brixen | khaase: what I thought was a hang in the unit tests is just test_load taking 2 min to run |
| 16:27:51 | brixen | all the tests run through to finish |
| 16:28:02 | khaase | really? |
| 16:28:05 | brixen | yeah |
| 16:28:06 | brixen | http://gist.github.com/382883 |
| 16:28:22 | brixen | I need to talk to evan about that profile |
| 16:28:31 | brixen | mri runs in 30 sec, takes rbx 2 min |
| 16:28:47 | khaase | ouch |
| 16:28:48 | brixen | khaase: all the tests don't pass, but there are no hangs :) |
| 16:29:45 | brixen | one of the functional failures is the method-as-proc issue |
| 16:30:06 | brixen | all the rest appear to be regexp related |
| 16:33:27 | khaase | oh, ok. schmidtwisser startet making redmine work on jruby. he said he simply modified the redcloth regexp |
| 16:34:17 | khaase | http://github.com/finnlabs/redmine/commit/b2c0885a509a517b68370bea459e5af2542a15fe |
| 16:34:47 | khaase | (this change is likely to be applied to redmine trunk soon) |
| 16:41:23 | khaase | brixen: sorry, am currently at work, so i really cannot try that right now |
| 16:42:49 | khaase | also, feel free to do the splat patch if you're bored (only in case you're waiting for me) |
| 16:46:21 | brixen | khaase: ahh I see |
| 16:46:48 | brixen | khaase: I'll check with that patch, but I'm going to work on other stuff today |
| 16:47:38 | khaase | brixen: yeah, that's fine, no pressure. you spent a lot of time on it already (thanks btw). |
| 16:47:53 | brixen | n/p found a couple bugs anyway |
| 16:48:08 | brixen | besides, I don't know yet that onig shouldn't consider < punctuation |
| 16:48:19 | brixen | still waiting for a response on my MRI ticket |
| 20:43:44 | dbussink | brixen: still there? |
| 20:49:41 | brixen | dbussink: sup yo :) |
| 20:50:15 | dbussink | brixen: was wondering, any reason why rakelib/ext_helper.rb:229 also swallows stderr? |
| 20:50:38 | dbussink | brixen: since silencing it for stdout makes sense, but stderr usually only masks issues |
| 20:51:11 | brixen | you can always run with -v |
| 20:51:21 | brixen | I don't want any output, that's why ;) |
| 20:51:33 | dbussink | brixen: me neither, unless there are errors ;) |
| 20:51:39 | dbussink | which are outputted on stderr :P |
| 20:51:45 | brixen | not always |
| 20:51:52 | brixen | if everyone respected that, yes |
| 20:51:58 | brixen | like I said, you can run with -v |
| 20:52:07 | brixen | no error is swallowed, only the output |
| 20:52:15 | brixen | if it errors, run with -v |
| 20:52:24 | dbussink | well, i was just trying it on the latest ubuntu, had to go back and forth a few times because i couldn't see directly that for example headers where missing |
| 20:52:35 | brixen | and you ran with -v? |
| 20:53:00 | dbussink | i just ran rake |
| 20:53:05 | brixen | right |
| 20:53:10 | brixen | so, run with -v |
| 20:53:24 | brixen | I've debugged plenty a broken build |
| 20:53:45 | dbussink | thing is there's no difference that i could find output wise between not swallowing stderr and doing so |
| 20:54:24 | brixen | win 12 |
| 20:54:38 | brixen | that's much higher than win |
| 20:54:43 | brixen | that's 12 levels up |
| 20:54:56 | brixen | dbussink: well, you seem really really concerned about it... |
| 20:55:14 | brixen | so go ahead and remove stderr, but if I see shit I don't want to see, I'm adding it back |
| 20:55:25 | brixen | I've given you -v, but you want the world... |
| 20:55:27 | brixen | :( |
| 20:55:27 | dbussink | brixen: well, i'm concerned mostly with people dipping into rbx and being puzzled by what happens |
| 20:55:54 | dbussink | because they don't get the information and have to know that -v does give them more |
| 20:56:09 | brixen | we've had reports of this? |
| 20:56:40 | dbussink | basically http://github.com/evanphx/rubinius/issues#issue/90 |
| 20:57:27 | brixen | I think that predates ext_helper |
| 20:57:43 | dbussink | brixen: well, errors etc. are still cryptic sometimes |
| 20:58:06 | dbussink | and especially with extensions like readline, openssl etc. you don't get any output by default it something goes wrong |
| 21:02:50 | boyscout | Enable stderr output during extension compiling - e1a0688 - Dirkjan Bussink |
| 21:03:05 | dbussink | brixen: a clean build is just as silent for me with this change |
| 21:04:16 | brixen | yeah, just did the same |
| 21:09:37 | dbussink | but sleepy time for me |
| 21:09:47 | brixen | nite dbussink! |
| 21:09:49 | dbussink | brixen: do you think it's worth to add more prechecks to configure? |
| 21:10:02 | brixen | yes, we have some plans to do so |
| 21:10:06 | brixen | like for readline.h |
| 21:10:13 | brixen | well, I think it's a good idead |
| 21:10:16 | brixen | er -d |
| 21:10:28 | brixen | but last time it was brought up, evan had some objections |
| 21:10:33 | brixen | so, wait on it for now |
| 21:10:58 | boyscout | CI: rubinius: e1a0688 successful: 3448 files, 13480 examples, 41038 expectations, 0 failures, 0 errors |
| 21:12:20 | dbussink | ok, np |
| 21:12:22 | dbussink | nite then! |
| 21:12:25 | brixen | nite! |
| 21:12:39 | brixen | odd to be sleeping when it's so sunny out, but... |
| 21:12:42 | brixen | oh, nvm :) |