| [Coke] wonders what the commit message on bf7063d1ae is. | 00:08 | ||
| sorry, what the *point* of that commit message is. | 00:09 | ||
| timotimo | to warn potential imitators? | ||
| [Coke] | (it's very long, and has a bunch of backstory that isn't relevant to the point of the commit: | ||
| ./win 2 | 02:05 | ||
| I am apparently not eligible for the lottery. :| | |||
| Zoffix | can't use the dot :) | 02:07 | |
| [Coke] | added TEST_THREADS support to a few files in the doc test; is not perfect plan, but allows for someone to have individual test use multiple cores. | 02:12 | |
| AlexDaniel | whew! That's a lot of commits in these 2.5 weeks. Need to get some good sleep before committing this changelog monstrosity… :) | 03:20 | |
| Zoffix | NeuralAnomaly: status | 03:24 | |
| NeuralAnomaly | Zoffix, [?] Next release will be in 1 week and 5 days. Since last release, there are 34 new still-open tickets (7 unreviewed and 0 blockers) and 135 unreviewed commits. See perl6.fail/release/stats for details | ||
| Zoffix thinks "wow, not a lot of commits" | 03:25 | ||
| :) | |||
| AlexDaniel | Zoffix: by the way, I didn't really find this list useful. For example, now I have to click the button 135 times. Maybe I'm missing something? | 03:29 | |
| Zoffix | AlexDaniel: "this list" being what? | 03:30 | |
| geekosaur | release/stats, presumably | ||
| AlexDaniel | Zoffix: the one on perl6.fail/release/stats (commits) | ||
| Zoffix | AlexDaniel: how do you plan to log those commits to ChangeLog? | ||
| (the list provides partially-processed things you can copy-paste most of the time) | 03:31 | ||
| Or, how do you plan to track which commits you added to ChangeLog and which you still need to add? | |||
| (that's where the button comes in; click it once you added commit to ChangeLog) | |||
| AlexDaniel | Zoffix: well, I simply used git log with some custom formatting. With some macro trickery it was probably significantly more efficient than fiddling with the browser | 03:32 | |
| Zoffix | OK, then don't use the browser | ||
| nine | A reason for setting up NativeCall subs at first call may be because at mixin time, we don't have access to all information we'd need. E.g. a return type may be added later by a returns trait. | 05:47 | |
| Even the signature is only available later | 05:48 | ||
| [Tux] | This is Rakudo version 2017.07-136-gda4a0f50a built on MoarVM version 2017.07-320-g600c2e9c | 06:56 | |
| csv-ip5xs 2.285 | |||
| test 12.432 | |||
| test-t 3.713 - 3.726 | |||
| csv-parser 14.057 | |||
| lizmat | whee! | 07:00 | |
| nine | How evil and/or wrong is this? gist.github.com/niner/0de8371393bf...7fdacb5e6b | 09:22 | |
| jnthn | Dangerous | 09:24 | |
| I'm also surprised it works | |||
| invocation_attr_name is menat to be an attribute name o.O | 09:25 | ||
| But also | |||
| An invoke spec lives on an STable | |||
| The assumption is that it won't change | |||
| nine | jnthn: but it does change when we mixin the role containing the CALL-ME? | 09:26 | |
| jnthn | True | ||
| Roles are interned on their parameters | 09:27 | ||
| Though I guess they should be unique always in this case | |||
| So I guess it's safe | |||
| nine | It is a bit faster than before. Though what I'd really like is re-using the signature of the original sub. | 09:29 | |
| Or rather replacing the original sub's body by "nqp::nativecall($!rettype, self, $args);" | |||
| Geth | rakudo/nom: 46ef1b5b48 | (Stefan Seifert)++ | lib/NativeCall.pm6 Speed up NativeCall by replacing the subroutine code in setup Instead of overriding the subroutine's code via a CALL-ME method, we replace the subroutine's code object ($!do) with the native calling code in !setup. This saves the check for $!setup on each call and probably makes the call cheaper in general. |
10:10 | |
| AlexDaniel | did we ever have perl6 syntax highlighting in github comments? | 11:43 | |
| modules.perl6.org looks awesome, by the way | 11:46 | ||
| well, a little bit low on the contrast in some places, but still awesome | 11:47 | ||
| timotimo | nine: got a number for how much faster? | 13:18 | |
| nine | timotimo: saved about 6 % in csv-ip5xs on 100000 rows (dropped from 16 to 15 seconds) | 13:20 | |
| samcv | grant status update 3 is up cry.nu/perl6/grant-status-update-3/ | 13:26 | |
| Zoffix | AlexDaniel: some areas are deliberately faded out until you hover | 14:28 | |
| samcv++ | 14:30 | ||
| .tell moritz FWIW, modules.perl6.org rejects any logotypes that aren't 32x32 px in size. I see a couple of your dists get reported with that problem | 14:36 | ||
| yoleaux | Zoffix: I'll pass your message to moritz. | ||
| AlexDaniel | Zoffix: I was mainly concerned with navbar buttons. At least here, #333 is noticeably easier to read, and also passes WCAG recommendations for all font sizes | 18:09 | |
| Zoffix: but everything else is so beautiful that I feel bad complaining about this :) | 18:10 | ||
| samcv | we don't currently have a routine to strip all text so it only has base characters do we? | 18:13 | |
| TimToady waves briefly from home after driving 8400 klicks across the USA in 10 days, now packing for Amsterdam... | 18:14 | ||
| Zoffix | samcv: .samemark | ||
| samcv | well i guess you can use that as well | ||
| timotimo | Zoffix: i wonder how that works with ZWJ and such on emoji | ||
| samcv | samemark? | 18:15 | |
| timotimo | yes | ||
| samcv | ah i think it works | ||
| it goes based on the base character | |||
| Zoffix | m: "\c[RAINBOW FLAG]".samemark("z").uninames.say | ||
| camelia | (WAVING WHITE FLAG)? | ||
| Zoffix | hah | ||
| timotimo | damn you, mark. | ||
| samcv | it's broken with prepend characters though | 18:16 | |
| AlexDaniel | m: dd “\c[woman gesturing OK]”.samemark(‘a’).uninames | ||
| camelia | ("FACE WITH OK GESTURE",).Seq? | ||
| samcv | m: dd "\c[RAINBOW FLAG]".uninames | ||
| camelia | ("WAVING WHITE FLAG", "VARIATION SELECTOR-16", "ZERO WIDTH JOINER", "RAINBOW").Seq? | ||
| Zoffix | AlexDaniel: fixed in github.com/perl6/modules.perl6.org...3a48f9ff99 | 18:17 | |
| AlexDaniel | Zoffix: Thank you! | ||
| samcv | ah i see why it's broken | 18:18 | |
| it just uses nqp to do it and doesn't use the proper functions | |||
| it shifts things instead of using baseordat or whatever | |||
| tries to do it manually | |||
| hmm maybe i should just write a MVM function that does something similar? | 18:22 | ||
| that changes the base character at least | |||
| AlexDaniel, dd “\c[woman gesturing OK]”.samemark(‘a’).uninames # this is working fine. it just removes the marks | 18:30 | ||
| and the base character is FACE WITH GESTURE | 18:31 | ||
| m: dd “\c[arabic number sign]a”.samemark('b').uninames | |||
| camelia | ("ARABIC NUMBER SIGN",).Seq? | ||
| samcv | this doesn't work though | ||
| m: dd 'a'.samemark(“\c[woman gesturing OK]”).uninames | 18:32 | ||
| camelia | ("LATIN SMALL LETTER A", "ZERO WIDTH JOINER", "FEMALE SIGN", "VARIATION SELECTOR-16").Seq? | ||
| samcv | m: dd 'a'.samemark(“\c[woman gesturing OK]”) | ||
| camelia | "a???"? | ||
| b2gills | bisect: old=2017.04.3 new=2017.05 say Mix('a' => 0.3).roll | 18:41 | |
| bisectable6 | b2gills, Bisecting by exit code (old=2017.04.3 new=2017.05). Old exit code: 0 | ||
| b2gills, bisect log: gist.github.com/c664055209ebd18eca...2eef8630bc | |||
| b2gills, (2017-04-30) github.com/rakudo/rakudo/commit/a2...8cb79b4bed | |||
| AlexDaniel | c: a2602b9c^,a2602b9c say Mix('a' => 0.3).roll | ||
| committable6 | AlexDaniel, ¦a2602b9c^: «a» ¦a2602b9: «Type check failed in binding; expected Int but got Rat (0.3)? in block <unit> at /tmp/TsKgDPJwiK line 1? «exit code = 1»» | ||
| b2gills | Continuation from irclog.perlgeek.de/perl6/2017-08-01#i_14950241 | 18:42 | |
| Rakudobugged: RT: 131855 | 19:00 | ||
| Geth | nqp: 9b94aae27b | (Samantha McVey)++ | 2 files Add eqatim and indexim ops. Fix a bug when using ignoremark There was a bug with ignoremark if the first two codepoints in the haystack matched, it would act as if the entire needle had matched. This was only with ignoremark by itself, as ignorecase+ignoremark had already been fixed. This bug was less noticable than the previously fixed bug in ... (6 more lines) |
19:29 | |
| nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...-g127fa2ce 07e5c80d19 | (Samantha McVey)++ | tools/build/MOAR_REVISION Changes brought in by this bump: 127fa2ce Add indexim_s op and improve/fix bugs in eqatim_s 0fe5cd60 More precise CFGs for control exception handlers. 39dbe1bd Allow for GOTO handler to become unreachable. 35f82730 Factor our and re-use dead BB elimination. ... (21 more lines) |
|||
| ¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...-g127fa2ce | |||
| rakudo/nom: a91ad2da85 | (Elizabeth Mattijsen)++ | src/core/Rakudo/QuantHash.pm Fix for RT #131855 Stupid copy-pasto. |
20:09 | ||
| synopsebot6 | Link: rt.perl.org/rt3/Public/Bug/Display...?id=131855 | ||
| AlexDaniel | planned maintenance :) ? | 21:07 | |
| :| broken | |||
| .seen perlpilot | 21:08 | ||
| yoleaux | I saw perlpilot 2 Aug 2017 17:52Z in #perl6: <perlpilot> Altreus: gist works | ||
| AlexDaniel | .tell perlpilot github.com/perlpilot/p6-sake/pull/6 plz | ||
| yoleaux | AlexDaniel: I'll pass your message to perlpilot. | ||
| AlexDaniel | there we go | 21:09 | |
| lizmat_ | and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2017/08/07/...eekly-101/ | 22:09 | |
| Zoffix | Opened a @LARRY ticket: rt.perl.org/Ticket/Display.html?id=131856 | 22:37 | |
| m: sub foo (*@a) { @a.join: '|' }; say foo <h H>.any, 'amadryas' | |||
| camelia | any("h", "H")|amadryas? | ||
| Zoffix | junctions with slurpies are kinda LTA, because what you get in reponse from core subs and ops depends on whether they have a slurpy candidate or not. For exaple, you get different things from `+` vs `~` op (the latter stringifies the junction itslef) | ||
| AlexDaniel | :| | 23:27 | |
| github.com/rakudo/rakudo/blob/nom/...oc.pm#L223 | |||
| github.com/rakudo/rakudo/blob/nom/...nc.pm#L272 | |||
| (it's ENV vs env, and to make things worse, proc.start simply eats env without complaining) | 23:31 | ||
| ugexe | yep, there is also cwd/CWD | 23:39 | |
| yoleaux | 6 Aug 2017 09:31Z <Zoffix> ugexe: how is this file generated? I see source URLs are still pointing to GitHub; how come all of them aren't pointing to CPAN? github.com/ugexe/Perl6-ecosystems/.../cpan.json | ||
| ugexe | Zoffix: because those came before the automatic source-url transformer was implemented in PAUSE | 23:40 | |
| its generated by github.com/ugexe/Perl6-App--ecogen (linked in the main repo's github description thing) | 23:42 | ||
| AlexDaniel | ugexe: cwd/CWD ? | 23:50 |