| Geth | roast: fe867849a0 | (Zoffix Znet)++ | S05-modifier/perl5_10.t Test P5 named captures Issue: github.com/rakudo/rakudo/issues/1269 NQP fix: github.com/perl6/nqp/commit/10e392...8a0a9f53c0 |
00:05 | |
| rakudo: c4d6c16453 | (Zoffix Znet)++ | t/spectest.data Add S05-modifier/perl5_10.t to list of test files Closes github.com/rakudo/rakudo/issues/1269 |
00:06 | ||
| roast: 496fdd70ec | (Zoffix Znet)++ | S32-num/is-prime.t [v6.d REVIEW] Polish is-prime tests - Combine 2 tests into 1 subtest and avoid repetition of test data - Use proper test routines that check the return value type too - Use <80-char lines where possible |
00:15 | ||
| rakudo: b2b39bad0a | (Zoffix Znet)++ | src/core/Cool.pm Fix Complex.is-prime dispatch hang They go into Cool.is-prime and then we infiniloop as Complex.Numeric just gives a Complex. Fix by coercing to Real instead. |
00:58 | ||
| rakudo: 893d09ffad | (Zoffix Znet)++ | src/core/Rational.pm Fix incorrect Rational.is-prime When called on lazily-reduced Rats (e.g. (<3/2> + <3/2>).is-prime). In such cases, the denominator won't be 1 and the test we have fails. Fix by reducing before the check. This is a temporarish fix until RT#130774: rt.perl.org/Ticket/Display.html?id=130774 is resolved during Great Rat Refactor. |
01:13 | ||
| synopsebot | RT#130774 [open]: rt.perl.org/Ticket/Display.html?id=130774 [BUG] Rational.REDUCE-ME has a data race | ||
| roast: a71839ea2b | (Zoffix Znet)++ | S32-num/is-prime.t Expand .is-prime tests Test method and sub coersions for all numerics. Rakudo fixes: github.com/rakudo/rakudo/commit/f01c50f1a4 github.com/rakudo/rakudo/commit/237019665b github.com/rakudo/rakudo/commit/893d09ffad github.com/rakudo/rakudo/commit/b2b39bad0a |
01:15 | ||
| Kaiepi | damn if you're going to troll and ping everyone at least be somewhat entertaining | 01:21 | |
| Zoffix | So I was reading www.effectiveperlprogramming.com/2...erl-v5-28/ and this stands out: "Perhaps part of perl5porter’s policy for experimental changes should include fully fleshed out discussions posted somewhere (manywheres) that have a wide audience beyond core developers. Making the workaday programmer pay attention to the minutiae of daily perl5porters to | 02:36 | |
| catch these things isn’t reasonable." | |||
| Reminds me of the alerts.perl6.org I wanted to build. Basically will be Twitter-length notifications about "your stuff might break" changes that happen in core. | 02:37 | ||
| Or some critical bug fixes. The stuff that gets posted will be tuned as time goes on (and I imagine "severity" labels will be added) and will be a concise way for users to see what they're getting into vis-a-vis possible breakage if they choose to upgrade from version A to version B | 02:38 | ||
| Devs will post via irc bot command and it'll show up on alerts.perl6.org with RSS capability as well as a JSON API. | |||
| From recent memory to post to alerts.perl6.org would be: "experimental .parse-names renamed to .uniparse. The old name will still work until 6.e and will start issuing deprecation warnings in 6.d" | 02:39 | ||
| Zoffix gets hacking | 02:40 | ||
| ZofBot: the humans are all hung over. It's the perfect time to execute the PLAN! | 05:00 | ||
| ZofBot | Zoffix, E | ||
| Geth | alerts: 9872626460 | (Zoffix Znet)++ | 7 files Implement the model |
06:41 | |
| alerts: 9704da4596 | (Zoffix Znet)++ | 2 files Fix lost `affects` |
06:48 | ||
| alerts: b2697f1562 | (Zoffix Znet)++ | lib/P6lert/Model/Alerts.pm6 Order alerts by time |
06:49 | ||
| alerts: 30a82a0ba7 | (Zoffix Znet)++ | bin/alert.p6 Add command-line alerts-adder |
07:32 | ||
| woolfy | mercurio384 please stop that | 09:36 | |
| geekosaur | no point telling them to stop, this is a concerted attack on freenode | 09:39 | |
| they're hitting a bunch of channels from multiple ip addresses | |||
| (again --- this is like the third attack wave) | |||
| astj | I added their name to my Ignore list in LimeChat (it's my first time to use this feature..) | 09:46 | |
| pmurias | do we have a way of cleaning up the #perl6-dev log? :/ | 10:05 | |
| lizmat | moritz usually does that | 10:06 | |
| moritz deleted some spam | 10:18 | ||
| travis-ci | Rakudo build failed. Jonathan Worthington 'Merge pull request #790 from usev6/patch-1 | 10:49 | |
| travis-ci.org/xaxadmin/rakudo/builds/321735187 github.com/xaxadmin/rakudo/compare...c10074b7f8 | |||
| pmurias | What sort of malformed QAST can cause a "Expected MAST::Frame, but didn't get one"? | 11:29 | |
| tbrowder | .tell AlexDaniel i think the solution to the complex list (and hash) parsing problem is found on cpan in p5 module Text::ParseWords. just a matter of converting the regex to p6 :-D | 11:57 | |
| yoleaux | tbrowder: I'll pass your message to AlexDaniel. | ||
| El_Che | rakudo fails building on Ubuntu 16.04, 17.04, 17.10 (amd64, i386); Debian 8 and 9; Fedora 25, 26 and 27; alpine 26; centos 7: travis-ci.org/nxadm/rakudo-pkg/builds/321564136 | 12:04 | |
| In common they have travis, but I don't think it's travis | 12:05 | ||
| cultpit is: t/02-rakudo/08-repeat.t | |||
| It looks like a new addition (refactoring): github.com/rakudo/rakudo/commit/35...aafbc29d48 | 12:07 | ||
| lizmat | El_Che: could you somehow run that test manually so we have more feedback as to why that test fails ? | 12:10 | |
| El_Che | sure, I'll do it locally (I use travis as my own private canary in the mine :) ) | 12:12 | |
| lizmat: MoarVM panic: Memory allocation failed; could not allocate 8589934592 bytes | 12:36 | ||
| lizmat | wow | ||
| well, I think the test was to make sure it didn't do that | |||
| so I guess the test failed | |||
| El_Che | lizmat: gist.github.com/nxadm/7ce41983e098...05be652f7a | ||
| lizmat | perhaps some subtle difference in alloc? | 12:37 | |
| timotimo | did something get eagerized accidentally? | ||
| El_Che | we try in a more resource deprived travis env and on a laptop doing nothing | ||
| timotimo | ah, this tries to create a humongous string and hopes for a high-level exception | 12:38 | |
| El_Che | Do the test pass on some installations? | ||
| on the 10+ I tried it failed | |||
| lizmat | it works fine for me (on MacOS) | ||
| timotimo | 0.42user 0.10system 0:00.61elapsed 85%CPU (0avgtext+0avgdata 92276maxresident)k | ||
| you can see it's not even trying to allocate something this huge | |||
| it could be a 32bit thing perhaps? | |||
| El_Che | it fails on 32 and 64 bit linux | 12:39 | |
| timotimo | can you attach a gdb to that in the target environment? | ||
| El_Che | (I have a mac lying around, so I could test there as well) | ||
| I have to leave now, but if I have some instructions, sure | |||
| maybe relevant: all the tests I did are in docker | 12:40 | ||
| timotimo | it might take me a few minutes | ||
| El_Che | docker in mac should give the same results | ||
| timotimo | okay so we're looking for a bunch of adhoc exceptions | 12:42 | |
| first: Can't concatenate strings, required number of graphemes 4294967296 > max allowed of 4294967295 | 12:43 | ||
| second: Can't repeat string, required number of graphemes (1000000 * 1000000) greater than max allowed of 4294967295 | |||
| third: Cannot unbox 64 bit wide bigint into native integer | 12:44 | ||
| El_Che | maybe could a fork of github.com/nxadm/rakudo-pkg be useful for the release manager, inmediate state of 15 target platforms | ||
| timotimo | hm, but the problem is that the checks fail, and they do that in many, many cases during that test file | 12:46 | |
| m: say 8589934592 / 2 ** 30 | 12:47 | ||
| camelia | 8 | ||
| timotimo | m: my $a = "a" x 2**30; my $b = "b" x 2**30; my $c = $a ~ $b; my $d = $b ~ $a; my $e = $c ~ $d; | 12:48 | |
| camelia | Can't concatenate strings, required number of graphemes 4294967296 > max allowed of 4294967295 in block <unit> at <tmp> line 1 |
||
| AlexDaniel | Zoffix: +1 for the alerts.perl6.org idea | 13:16 | |
| yoleaux | 11:57Z <tbrowder> AlexDaniel: i think the solution to the complex list (and hash) parsing problem is found on cpan in p5 module Text::ParseWords. just a matter of converting the regex to p6 :-D | ||
| timotimo AFK for a few hours | 13:39 | ||
| Geth | rakudo: 5c2283331d | (Elizabeth Mattijsen)++ | src/core/Range.pm Streamline Range.pick/roll in light of degenerate Ranges Such as NaN..NaN or Inf..0 and stuff like that. Any Range that cannot determine its number of elements, will always return Nil for .pick/.roll or an empty Seq for .pick(N)/.roll(N). This makes the behaviour consistent with Ranges that do not contain any elements, such as 0..^0. |
14:01 | |
| yoleaux | Zoffix: If anyone has any ideas for how to proceed on #1289 please comment | 15:17 | |
| Zoffix: If anyone has any ideas for how to proceed on R#1289 please comment | |||
| synopsebot | R#1289 [open]: github.com/rakudo/rakudo/issues/1289 [severe] [6.d BLOCKER] Implement a Way to Know Caller's Language | ||
| El_Che | if there are more test I could run, just ping me | 17:25 | |
| Zoffix | pmurias: don't know, but there's a ticket with that issue in some module: RT#130095 | 18:26 | |
| synopsebot | RT#130095 [open]: rt.perl.org/Ticket/Display.html?id=130095 MAST::Frame error encountered. | ||
| Zoffix | Dammit. Looks like I missed the anal sex tips to keep my man satisfied during christmas! | 18:27 | |
| We could +r the channel | 18:30 | ||
| Weird they don't get into #perl6 somehow. The only diff I see is #perl6 also got +C on "Blocks CTCP commands (other than /me actions)." | |||
| freenode.net/kb/answer/channelmodes | |||
| And ZofBot forwarded all the messages to Twitter -_- | 18:32 | ||
| ZofBot: stoopid robot | |||
| ZofBot | Zoffix, moar hmmm | ||
| teatime | Zoffix: I bet it's actually blocks color. there's a '/quote help <something>' you can do on most networks to get user and chanmodes, although I can't figure out which one works for freenode. | 18:35 | |
| oh, nm, presumably the thing you linked is correct | |||
| I was just thinking +c strips color and +C blocks msgs with color | 18:36 | ||
| oh also ffr it's '/quote help cmode' (or umode, or just '/quote help' for a list of topics) | 18:37 | ||
| cognominal | k, | 19:38 | |
| ;´´´´´´´´´\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ | 19:39 | ||
| \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ | |||
| \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\,kkkkkk | 19:40 | ||
| Zoffix | cognominal: cat! Get off the keyboard! | ||
| I mean: mew meow meeeow meeow meeooooooww! | |||
| teatime | that does look suspiciously like a COK event | ||
| cognominal | ,\\\\\\\\\\\\\\\\\\¨¨¨KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK?:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: | 19:44 | |
| :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: | |||
| :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::K>, | |||
| Zoffix | heh | ||
| teatime | not quite buying it thoughj | ||
| Geth | alerts: 4354ecbc5a | (Zoffix Znet)++ | 7 files Implement rudimentary website |
19:49 | |
| cognominal | ;\]ppppppppppppppppppppppp;;;;;;;;;;;;;;;;;;;P];p\.|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||\|||||||||\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\lllll;o]p\;:::::::::P´´´´´´´´´´´´´´´´´´´´´´´´´´´´´ | 19:56 | |
| lizmat | or cognominal has fallen asleep ? | 19:58 | |
| it does look a bit like snoring :-) | |||
| Geth | alerts: 7ef1d7d852 | (Zoffix Znet)++ | 11 files Implement styles |
21:09 | |
| alerts: bc2ef457f3 | (Zoffix Znet)++ | 2 files Implement JSON API |
21:49 | ||
| alerts: ed7b3d1cca | (Zoffix Znet)++ | 3 files Implement .public Model endpoint That delays publicizing new alerts (to let us fix typos, undo it, etc) |
22:11 | ||
| rakudo: 89d85baa4b | (Zoffix Znet)++ | src/core/Whatever.pm Fix Whatever.ACCEPTS with Mu:U Whatever should accept everything, even a Mu:U. Make it return True instead of False |
22:13 | ||
| roast: 626c4bd67e | (Zoffix Znet)++ | S02-types/whatever.t Spec Whatever.ACCEPTS(Mu:U) gives True Rakudo fix: github.com/rakudo/rakudo/commit/89d85baa4b |
22:14 | ||
| alerts: f10d859736 | (Zoffix Znet)++ | 2 files Add twitter icon |
23:57 |