Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2017-10-13

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:11 Geth_ joined #perl6-dev
00:12 Zoffix for crying out loud... "No such method 'slurp-rest' for invocant of type 'Any'   in sub  at /home/zoffix/services/geth/lib/Geth/GitHub/Hooks.pm6"
00:12 Zoffix Why did I restart it :()
00:13 Geth_ joined #perl6-dev
00:13 Geth_ joined #perl6-dev
00:14 Geth_ joined #perl6-dev
00:16 Geth_ joined #perl6-dev
00:16 Geth_ ¦ roast: bcaaad4d42 | (Zoffix Znet)++ | 2 files
00:16 Geth_ ¦ roast: Consolidate bug cover tests
00:16 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/bcaaad4d42
00:16 Zoffix I see. If you precompile modules with FileSystem CUR and then uninstall a module with zef, then the precomp files still remain untouched, using whatever version was installed
00:17 Geth_ ¦ rakudo/nom: f72be0f130 | (Zoffix Znet)++ | t/spectest.data
00:17 Geth_ ¦ rakudo/nom: Remove dir-stress test file
00:17 Geth_ ¦ rakudo/nom:
00:17 Geth_ ¦ rakudo/nom: It was consolidated with other lone bug cover stresstests in
00:17 Geth_ ¦ rakudo/nom: https://github.com/perl6/roast/commit/bcaaad4d42
00:17 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f72be0f130
00:45 Zoffix m: use Test; for (NaN..NaN) => NaN, (Inf..0) => Inf, (-Inf..0) => -Inf -> (:key($range), :value($expected)) { .map: { next unless $++ > 1050; is-deeply $_, $expected, "$_ range keeps producing $expected ad infinitum"; last } }
00:45 camelia rakudo-moar f72be0: ( no output )
00:45 Zoffix dahell?
00:46 Zoffix Ah, missing $range
00:51 Geth_ ¦ roast: bfc143e0ac | (Zoffix Znet)++ | S02-types/range.t
00:51 Geth_ ¦ roast: no floating point drifts in degenerate Ranges
00:51 Geth_ ¦ roast:
00:51 Geth_ ¦ roast: RT#125336: https://rt.perl.org/Ticket/Display.html?id=125336
00:51 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/bfc143e0ac
00:51 synopsebot RT#125336 [new]: https://rt.perl.org/Ticket/Display.html?id=125336 Weird infinite ranges: Inf..0, -Inf..0 and NaN..0
00:53 llfourn joined #perl6-dev
00:58 astj joined #perl6-dev
00:59 AlexDaniel- Zoffix++
01:56 ilbot3 joined #perl6-dev
01:56 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
02:37 AlexDaniel- \o/
02:38 AlexDaniel- OK I think I got regressionable working
04:18 awwaiid joined #perl6-dev
04:51 AlexDaniel- bisect: ("foo" => 10000000000000000000).BagHash.grab(1)
04:51 bisectable6 AlexDaniel-, Bisecting by exit code (old=2015.12 new=f72be0f). Old exit code: 0
04:51 AlexDaniel- I think somebody mentioned this a few days ago
04:51 bisectable6 AlexDaniel-, bisect log: https://gist.github.com/bf8346742ef61f4d599b870077131430
04:51 bisectable6 AlexDaniel-, (2017-05-11) https://github.com/rakudo/rakudo/commit/87a95fc1355e01972670895b55b02bc382797fc9
04:52 AlexDaniel- aw yeah, that's exactly how I found out about it
06:00 lizmat joined #perl6-dev
06:11 AlexDaniel- ha… almost submitted a ticket asking to unfix a bug
06:14 AlexDaniel- .tell Zoffix should this fail like it does now? 「-> (*%_) {dd %_}(Failure.new)」 https://gist.github.com/41a100146ec62ece32d91833060faf00
06:14 yoleaux AlexDaniel-: I'll pass your message to Zoffix.
06:23 AlexDaniel- .tell lizmat this does not look right, but the commit seems to be innocent. What do you think? Let me know if I have to rakudobug it https://gist.github.com/AlexDaniel/de2c5f1fe66b1606589041ca10c82f50
06:23 yoleaux AlexDaniel-: I'll pass your message to lizmat.
06:24 AlexDaniel- c: releases Blob[num32].new(2e0)
06:24 committable6 AlexDaniel-, https://gist.github.com/3a8ab6ac823d9b25491388a99934471d
06:25 AlexDaniel- bisect: old=2017.06 Blob[num32].new(2e0)
06:25 bisectable6 AlexDaniel-, Bisecting by output (old=2017.06 new=f72be0f) because on both starting points the exit code is 1
06:25 bisectable6 AlexDaniel-, bisect log: https://gist.github.com/41e9da04c8065b7888ef65534a3e98fb
06:25 bisectable6 AlexDaniel-, (2017-06-26) https://github.com/rakudo/rakudo/commit/823011281de7fcb4ae5b542ff41c313ed693f9cb
06:27 AlexDaniel- c: 823011281^,823011281 Blob[num32].new(2e0)
06:27 committable6 AlexDaniel-, https://gist.github.com/72544ca121a3ff6c7f0bac78b8885689
06:28 AlexDaniel- bisect: old=823011281 Blob[num32].new(2e0)
06:28 bisectable6 AlexDaniel-, Bisecting by output (old=8230112 new=f72be0f) because on both starting points the exit code is 1
06:28 bisectable6 AlexDaniel-, bisect log: https://gist.github.com/9dcb237e77570f9dccd1328fa6b963ec
06:28 bisectable6 AlexDaniel-, (2017-07-08) https://github.com/rakudo/rakudo/commit/1a4d94930c581e915a2351049e8ee12246c4c26c
06:28 AlexDaniel- c: 1a4d94930^,1a4d94930 Blob[num32].new(2e0)
06:28 committable6 AlexDaniel-, https://gist.github.com/70e94426796265d6213581e0c1df36b1
06:28 AlexDaniel- actually I thought that I'm on #whateverable
06:28 AlexDaniel- so nevermind this spam
06:31 AlexDaniel- .tell MasterDuke fwiw RT #132285
06:31 yoleaux AlexDaniel-: I'll pass your message to MasterDuke.
06:31 synopsebot RT#132285 [new]: https://rt.perl.org/Ticket/Display.html?id=132285 [REGRESSION] Error message talks about null strings when trying to create a num32 Blob (Blob[num32].new(2e0))
06:36 AlexDaniel- by the way, why the numbers suddenly became bigger?
06:36 AlexDaniel- mch: my $x; my $y := $x; $x = (42,$y); say $x
06:36 committable6 AlexDaniel-, ¦2015.12: «(\List_82832240 = (42 List_82832240))» ¦HEAD(f72be0f): «(\List_93951630447504 = (42 List_93951630447504))»
06:37 moritz Zoffix: I think I can connect to the box on which geth runs... now what?
06:38 sjn joined #perl6-dev
06:41 lizmat joined #perl6-dev
06:44 AlexDaniel- .tell Zoffix also maybe see https://irclog.perlgeek.de/perl6/2017-10-03#i_15250701 which does not work anymore
06:44 yoleaux AlexDaniel-: I'll pass your message to Zoffix.
06:58 Ven joined #perl6-dev
07:06 [TuxCM] joined #perl6-dev
08:24 lizmat joined #perl6-dev
08:54 lizmat joined #perl6-dev
08:58 jnthn AlexDaniel: Those numbers are just memory addresses (or, likely, future memory addresses), which tend to be unpredictable by design.
09:12 Zoffix moritz: Dunno. I can connect too, but thought maybe you know more about it. For some reason connections to it are blocked. `telnet geth.niner.name 8888` fails for example
09:12 yoleaux 06:14Z <AlexDaniel-> Zoffix: should this fail like it does now? 「-> (*%_) {dd %_}(Failure.new)」 https://gist.github.com/41a100146ec62ece32d91833060faf00
09:12 yoleaux 06:44Z <AlexDaniel-> Zoffix: also maybe see https://irclog.perlgeek.de/perl6/2017-10-03#i_15250701 which does not work anymore
09:13 Zoffix AlexDaniel: yes and yes
09:19 Zoffix First one documented in https://docs.perl6.org/type/Failure#method_Capture second on in https://docs.perl6.org/type/Str#method_Capture
09:19 Zoffix m: try 「sub c(:&funk! (:(:$param))) { &funk(:parrrrrrrrrrrrrrrrrrrrrrrrrrrrram('finally. Zoffix++')) }; c( funk => {say $:parrrrrrrrrrrrrrrrrrrrrrrrrrrrram} );」.EVAL; say $!.^name
09:19 camelia rakudo-moar f72be0: OUTPUT: «X::Cannot::Capture␤»
09:20 Zoffix yup
09:22 Zoffix AlexDaniel: heh: https://irclog.perlgeek.de/perl6/2017-10-13#i_15297137
09:22 Zoffix That's one of the bot evals
09:30 Geth_ ¦ roast: 11fb4415f0 | (Zoffix Znet)++ | S32-io/open.t
09:30 Geth_ ¦ roast: .open error does not incorrectly complain about malformed UTF-8
09:30 Geth_ ¦ roast:
09:30 Geth_ ¦ roast: RT#125813: https://rt.perl.org/Ticket/Display.html?id=125813
09:30 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/11fb4415f0
09:30 synopsebot RT#125813 [open]: https://rt.perl.org/Ticket/Display.html?id=125813 [UNI] Malformed UTF-8 (string out of bounds) with “say ('a' x 10000).IO.open”
09:34 Ven joined #perl6-dev
09:35 Geth_ ¦ roast: c8ebdbbf89 | (Zoffix Znet)++ | S03-smartmatch/any-num.t
09:35 Geth_ ¦ roast: can smartmatch num64 against Num
09:35 Geth_ ¦ roast:
09:35 Geth_ ¦ roast: RT#126927: https://rt.perl.org/Ticket/Display.html?id=126927
09:35 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/c8ebdbbf89
09:35 synopsebot RT#126927 [open]: https://rt.perl.org/Ticket/Display.html?id=126927 [BUG] Spurious warning when trying to smartmatch num64 against Num in Rakudo
09:39 Geth_ ¦ roast: 809abcd5a7 | (Zoffix Znet)++ | S06-advanced/return.t
09:39 Geth_ ¦ roast: Can return from parameter defaults
09:39 Geth_ ¦ roast:
09:39 Geth_ ¦ roast: RT#122345: https://rt.perl.org/Ticket/Display.html?id=122345
09:39 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/809abcd5a7
09:39 synopsebot RT#122345 [open]: https://rt.perl.org/Ticket/Display.html?id=122345 [TESTNEEDED] Returning from a parameter list default counts as returning 'outside of any Routine' in Rakudo on Moar
09:43 cog_ joined #perl6-dev
09:44 lizmat joined #perl6-dev
09:45 Geth_ ¦ roast: e0d3c550f1 | (Zoffix Znet)++ | S32-io/open.t
09:45 Geth_ ¦ roast: Make test more specific
09:45 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/e0d3c550f1
09:50 ZofBot joined #perl6-dev
09:50 [TuxCM] This is Rakudo version 2017.09-383-gf72be0f13 built on MoarVM version 2017.09.1-594-gb9d3f6da
09:50 [TuxCM] csv-ip5xs        1.154 -  1.182
09:50 [TuxCM] test             9.424 -  9.481
09:50 [TuxCM] test-t           3.094 -  3.156
09:50 [TuxCM] csv-parser       0.837 -  0.862
10:03 Geth_ ¦ rakudo/nom: 969853c673 | (Zoffix Znet)++ | t/05-messages/01-errors.t
10:03 Geth_ ¦ rakudo/nom: Use actual test plan
10:03 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/969853c673
10:05 Zoffix Actually, startup boost likely had a much better improvement to spectest/stresstest, since we have a ton of `is_run` tests that fire up an $*EXECUTABLE
10:10 timotimo oh, neat :)
10:14 lizmat joined #perl6-dev
10:15 Geth_ ¦ rakudo/nom: fb7abf064e | (Zoffix Znet)++ | t/05-messages/01-errors.t
10:15 Geth_ ¦ rakudo/nom: Test sane error when missing variables with my and where
10:15 Geth_ ¦ rakudo/nom:
10:15 Geth_ ¦ rakudo/nom: RT#125902: https://rt.perl.org/Public/Bug/Display.html?id=125902
10:15 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fb7abf064e
10:15 synopsebot RT#125902 [open]: https://rt.perl.org/Ticket/Display.html?id=125902 [LTA] error message “Type 'Str' is not declared” (my Str where 'foo' $test;)
10:31 lizmat joined #perl6-dev
11:17 lizmat_ joined #perl6-dev
11:19 lizmat__ joined #perl6-dev
11:24 lizmat joined #perl6-dev
11:25 Geth_ ¦ rakudo/nom: 975fcf6cfd | (Elizabeth Mattijsen)++ | src/core/Rakudo/QuantHash.pm
11:25 Geth_ ¦ rakudo/nom: Fix for RT #132279
11:25 Geth_ ¦ rakudo/nom:
11:25 Geth_ ¦ rakudo/nom: Cannot be sure we have int64 range on BagHash
11:25 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/975fcf6cfd
11:25 synopsebot RT#132279 [new]: https://rt.perl.org/Ticket/Display.html?id=132279 [REGRESSION][9999] BagHash.grab with huge values (("foo" => 10000000000000000000).BagHash.grab(1))
11:36 AlexDaniel Zoffix: ok
11:37 AlexDaniel and yeah, I knew this would've happened (bot evals)
11:39 AlexDaniel maybe I should've put some irc keywords into the blacklist, hmm…
11:40 AlexDaniel “13721 snippets tested, 3983 are known to be unstable”
11:40 AlexDaniel ZofBot: when are you going to be smart enough to do this work?
11:40 ZofBot AlexDaniel, IO::Spec subclasses The following methods should be provided by the "IO::Spec" subclasses, or may be inherited from another class
11:44 AlexDaniel oh… right. it printed “Hi, blub” 5 times because it tests every snippet on 5 revisions…
11:44 Zoffix imma move Geth to hack tonight
11:44 Zoffix And update all the hooks in perl6 org to use that
11:44 AlexDaniel and it does it in random order so we're going to see more of these soon I think
11:46 Ven joined #perl6-dev
11:58 Zoffix "NotZoffix joined #perl6"
11:58 * Zoffix squints
11:58 Zoffix ZofBot: EXTERMINATE!
11:58 ZofBot Zoffix, html?id=132128 Link: https://rt
12:00 ZofBot joined #perl6-dev
12:06 buggable joined #perl6-dev
12:09 buggable joined #perl6-dev
12:10 buggable joined #perl6-dev
12:10 Zoffix buggable: christmas
12:10 buggable Zoffix, Christmas is in 10 weeks, 2 days, 11 hours, 49 minutes, and 22 seconds
12:10 Zoffix \o/
12:15 llfourn joined #perl6-dev
12:17 Geth_ ¦ roast: bitrauser++ created pull request #347: Add test for RT#131490 | Crash in Junction:D.BUILDALL `This type (Scalar) does not support elems`
12:17 Geth_ ¦ roast: review: https://github.com/perl6/roast/pull/347
12:17 synopsebot RT#131490 [open]: https://rt.perl.org/Ticket/Display.html?id=131490 [TESTNEEDED] Crash in Junction:D.BUILDALL `This type (Scalar) does not support elems`
12:19 Geth_ ¦ roast: 5ebf6f9303 | Joel++ (committed by Zoffix Znet) | S03-junctions/misc.t
12:19 Geth_ ¦ roast: Add test for RT#131490 (#347)
12:19 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/5ebf6f9303
12:21 Geth_ ¦ roast: 1b0d85a39c | (Zoffix Znet)++ (committed using GitHub Web editor) | S03-junctions/misc.t
12:21 Geth_ ¦ roast: Fix up test description
12:21 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/1b0d85a39c
12:21 Zoffix Hm, it was fixed in https://github.com/rakudo/rakudo/commit/61ecfd51172b0e3cf20dc23a87dba9307fa94da6 so presumably test already existed :/
12:23 Zoffix ah nope. Tests were added but they don't catch the bug in the ticket: https://github.com/perl6/roast/blob/master/S03-junctions/misc.t#L526-L530
12:24 Geth_ ¦ roast: c2788cce70 | (Zoffix Znet)++ (committed using GitHub Web editor) | S03-junctions/misc.t
12:24 Geth_ ¦ roast: Consolidate related tests
12:24 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/c2788cce70
12:33 lizmat joined #perl6-dev
12:53 MasterDuke arg, how do i get the message out of a BOOTException?
12:54 Zoffix nqp::getmessage or nqp::getpayload maybe? Some ops are listed here: https://github.com/perl6/nqp/blob/master/docs/ops.markdown#-exceptional-opcodes
12:55 MasterDuke getmessage doesn't work, but maybe i haven't tried getpayload
13:27 Ven joined #perl6-dev
13:28 Ven_ joined #perl6-dev
13:30 Geth_ ¦ rakudo: MasterDuke17++ created pull request #1189: Correctly explain why couldn't instantiate a role
13:30 Geth_ ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/1189
13:31 Geth_ ¦ rakudo/nom: b3bb8c406a | MasterDuke17++ | src/Perl6/Metamodel/ParametricRoleHOW.nqp
13:31 Geth_ ¦ rakudo/nom: Correctly explain why couldn't instantiate a role
13:31 Geth_ ¦ rakudo/nom:
13:31 Geth_ ¦ rakudo/nom: This broke when Exceptions were changed to not create their message
13:31 Geth_ ¦ rakudo/nom: unless .gist'ed.
13:31 Geth_ ¦ rakudo/nom:
13:31 Geth_ ¦ rakudo/nom: Fixes RT #132285.
13:31 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b3bb8c406a
13:31 synopsebot RT#132285 [new]: https://rt.perl.org/Ticket/Display.html?id=132285 [REGRESSION] Error message talks about null strings when trying to create a num32 Blob (Blob[num32].new(2e0))
13:31 Geth_ ¦ rakudo/nom: f4f36915f5 | (Zoffix Znet)++ (committed using GitHub Web editor) | src/Perl6/Metamodel/ParametricRoleHOW.nqp
13:31 Geth_ ¦ rakudo/nom: Merge pull request #1189 from MasterDuke17/RT132285
13:31 Geth_ ¦ rakudo/nom:
13:32 Geth_ ¦ rakudo/nom: Correctly explain why couldn't instantiate a role
13:32 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f4f36915f5
13:45 * Zoffix grumbles about using RT for dick measuring contests
13:46 Zoffix .ask AlexDaniel what do you think of this ticket? https://rt.perl.org/Ticket/Display.html?id=131626#txn-1493785 The problem is swept under the rug and it the cause appears to be just that we shuttle a capture around from Unicode ops to ASCII ones, so that issue will be resolved on its own. So should the ticket be closed? What about tests?
13:46 yoleaux Zoffix: I'll pass your message to AlexDaniel.
13:46 Zoffix *people using
13:47 MasterDuke i'd almost think we should have spectest, stresstest, and perftest
13:48 MasterDuke and perftest would get things like that
13:48 Zoffix Yeah. Definitely
13:48 Zoffix .ask AlexDaniel also this one can prolly be closed entirely? https://rt.perl.org/Ticket/Display.html?id=131515#ticket-history
13:48 yoleaux Zoffix: I'll pass your message to AlexDaniel.
13:48 MasterDuke are those moarvm benchmarks still generated daily?
13:48 * Zoffix shrugs
13:57 Ven joined #perl6-dev
14:04 Zoffix oh oops. Modules.perl6.org hasn't updated since oct 9
14:06 yoleaux Zoffix: make separate t/spectest.data for each language version, bruh
14:06 Zoffix yoleaux: dun it long time ago, bruh
14:07 Zoffix modules.perl6.org@www:/tmp$ ls -1 | wc -l
14:07 Zoffix 3339
14:07 Zoffix yu
14:07 Zoffix Looks like File::Temp doesn't delete the temp dirs :/
14:12 Zoffix oh hahaha
14:13 Zoffix RE: https://irclog.perlgeek.de/perl6/2017-10-09#i_15276445
14:13 Zoffix I have the answer: it'll break the entire build :P
14:15 Zoffix .tell ugexe this may have been a misnamed upload for zef? Dunno if fixed version was uploaded already. modules.perl6.org will ignore it due to lack of name in the filename itself that it uses to figure this stuff out: https://irclog.perlgeek.de/perl6/2017-10-09#i_15276439
14:15 yoleaux Zoffix: I'll pass your message to ugexe.
14:17 Zoffix And yeah, theoretically it could figure it out from the .meta file, I guess, but then this kinda makes the build process depend on processing all the dists of the user before knowing which dists are what.
14:17 Zoffix Didn't consider this during making of original version of the CPAN dist source
14:18 Zoffix (this = that filenames could be different than what the dist is)
14:22 skids joined #perl6-dev
14:22 ugexe it kind has to eventually, because dists can have dashes and version like strings in their name
14:22 yoleaux 14:15Z <Zoffix> ugexe: this may have been a misnamed upload for zef? Dunno if fixed version was uploaded already. modules.perl6.org will ignore it due to lack of name in the filename itself that it uses to figure this stuff out: https://irclog.perlgeek.de/perl6/2017-10-09#i_15276439
14:24 ugexe Eg cant reliably decode module name + version from a file name alone
14:24 Zoffix Right now it treats `-([^-]+)$` as version part (sans file extension). So basically everything after last dash
14:25 Zoffix m: say Version.new: "42-x"
14:25 camelia rakudo-moar f4f369: OUTPUT: «v42.x␤»
14:26 * Zoffix scratches head
14:26 MasterDuke Zoffix: whoops, completely missed that you commented on 132285
14:26 Zoffix m: say Version.new: "-"
14:26 camelia rakudo-moar f4f369: OUTPUT: «v␤»
14:26 Zoffix m: Version.new("-").perl.say
14:26 camelia rakudo-moar f4f369: OUTPUT: «Version.new␤»
14:30 ugexe App::ecogen handles it ok I believe
14:30 Zoffix MasterDuke: "Not sure where tests should go, roast or rakudo?" It'd go to rakudo
14:31 Zoffix MasterDuke: if the new version threw some useful typed exception, then it could go to roast, but I see both buggy and non-buggy versions throw X::AdHoc and the non-buggy says it's NYI, so there's nothing really we can spec to be language behaviour, so the tests to cover the content of error would go to t/05-messages/errors.t instead of the roast
14:32 MasterDuke Zoffix: and then we're fine removing those tests if/when the NYI behavior is implemented?
14:33 Zoffix Right, t/* tests can be modified whenever we need to.
14:33 MasterDuke or should the test be something like: if exception was thrown, it has "NYI" and "num32" in it?
14:33 Zoffix Yeah, prolly
14:33 Ven_ joined #perl6-dev
14:35 Zoffix c: 2017.09,HEAD use Test; throws-like 「Blob[num32].new: 2e0」, Exception, :message{.contains: 「not yet implemented」 & 「num32」 and not .contains: 「got null」}, 'sane error for num32 Blob'
14:35 committable6 Zoffix, https://gist.github.com/80672fba1cba63a1cbf789ff053488b4
14:37 * Zoffix &
14:39 Ven_ joined #perl6-dev
14:41 Geth_ ¦ rakudo: MasterDuke17++ created pull request #1190: Test RT132285, correct error for Blob with num32
14:41 Geth_ ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/1190
14:41 synopsebot RT#132285 [open]: https://rt.perl.org/Ticket/Display.html?id=132285 [REGRESSION] Error message talks about null strings when trying to create a num32 Blob (Blob[num32].new(2e0))
14:45 Geth_ ¦ rakudo/nom: e611978f54 | MasterDuke17++ | t/05-messages/01-errors.t
14:45 Geth_ ¦ rakudo/nom: Test RT132285, correct error for Blob with num32
14:45 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e611978f54
14:45 Geth_ ¦ rakudo/nom: e475ae8e05 | (Zoffix Znet)++ (committed using GitHub Web editor) | t/05-messages/01-errors.t
14:45 Geth_ ¦ rakudo/nom: Merge pull request #1190 from MasterDuke17/test_for_RT132285
14:45 Geth_ ¦ rakudo/nom:
14:45 Geth_ ¦ rakudo/nom: Test RT132285, correct error for Blob with num32
14:45 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e475ae8e05
14:53 MasterDuke Zoffix++
15:42 Zoffix .tell tyil the "my App::MPD::AutoQueue still appears twice, even though I removed it from the META.list" was now fixed. If CPAN modules are missing from mp6o while there generated PAUSE .meta files for them, let me know. If PAUSE isn't generating .meta for some dists. Try to track down PAUSE people and ask them why
15:42 yoleaux Zoffix: I'll pass your message to tyil.
15:59 * Zoffix received a sad private message about RT
16:00 Zoffix Would be nice if new comers were
16:01 Zoffix Would be nice if new comers were not ridiculed/attacked in ticket responses. You don't have to be Einstein to file a ticket and if there's some misconception about user's ticket, just clarify it, if the clarification is needed to resolve the ticket, or just leave it alone. Writing miles of text arguing who is most right ain't helping—especially ain't helping the person who'll be fixing the ticket later
16:01 Zoffix on and has to read through miles of pointless dialogue
16:02 Zoffix (I'm not an angel myself; but we can all aim towards some ideal)
16:04 Zoffix ZofBot: so when will the robots kill all humans?
16:04 ZofBot Zoffix, ^name}"; } } X::Range::InvalidArg Thrown when an argument to "Range
16:05 Zoffix :( I need to reset your brain.
16:10 ZofBot joined #perl6-dev
16:10 Zoffix ZofBot: aho matey
16:10 ZofBot Zoffix, Think you, but that I know our state secure, I would be so triumphant as I am? STANLEY
16:10 Zoffix Ah, I love robots.
16:19 ZofBot joined #perl6-dev
16:29 Zoffix Seems we need some `make extratest` that runs extra rakudo-only tests that aren't spec. Like tests for this ticket would need installation fo the debugger module, for example: https://rt.perl.org/Ticket/Display.html?id=131481#ticket-history
16:29 Zoffix Also, I imagine the bug-cover error message tests isn't really something end users need to run, so that can be shoved to make extratest too. Make it run on travis
16:30 Zoffix also need `make dangeroustest` that will run dangerous roast tests.
16:30 Zoffix s:g/'make extratest'/make xtest/;
16:32 Zoffix And stuff like RT#127671 would go to dangeroustest
16:32 synopsebot RT#127671 [open]: https://rt.perl.org/Ticket/Display.html?id=127671 [EXOTICTEST] 「dir」 dies if weird unicode sequences are encountered (dir;)
16:32 Zoffix That'd be run only by people using some destructable VM and stuff
16:47 camelia joined #perl6-dev
16:53 Geth_ ¦ rakudo/nom: 4473591b0a | (Zoffix Znet)++ | t/05-messages/01-errors.t
16:53 Geth_ ¦ rakudo/nom: Add NYI to test description
16:53 Geth_ ¦ rakudo/nom:
16:53 Geth_ ¦ rakudo/nom: So that when it is implemented the failure of the test would be
16:53 Geth_ ¦ rakudo/nom: more informative as to why it's failing.
16:53 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4473591b0a
16:55 Geth_ ¦ rakudo/nom: 80cee36223 | (Zoffix Znet)++ | t/05-messages/01-errors.t
16:55 Geth_ ¦ rakudo/nom: Test callframe.my throws sane NYI error message
16:55 Geth_ ¦ rakudo/nom:
16:55 Geth_ ¦ rakudo/nom: RT#77754: https://rt.perl.org/Ticket/Display.html?id=77754
16:55 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/80cee36223
16:55 synopsebot RT#77754 [open]: https://rt.perl.org/Ticket/Display.html?id=77754 [BUG] Null PMC access when calling .perl on callframe.my in Rakudo
17:10 buggable New CPAN upload: Propius-0.1.0.tar.gz by ATROXAPER https://cpan.metacpan.org/authors/id/A/AT/ATROXAPER/Perl6/Propius-0.1.0.tar.gz
17:10 buggable New CPAN upload: TimeUnit-0.1.2.tar.gz by ATROXAPER https://cpan.metacpan.org/authors/id/A/AT/ATROXAPER/Perl6/TimeUnit-0.1.2.tar.gz
17:14 pmurias joined #perl6-dev
17:15 Geth_ ¦ roast: 522e3da803 | (Zoffix Znet)++ | S02-types/mix.t
17:15 Geth_ ¦ roast: Test Mix.roll with fractional weights does not crash
17:15 Geth_ ¦ roast:
17:15 Geth_ ¦ roast: RT#131855: https://rt.perl.org/Ticket/Display.html?id=131855
17:15 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/522e3da803
17:15 synopsebot RT#131855 [open]: https://rt.perl.org/Ticket/Display.html?id=131855 [REGRESSION] Mix.roll doesn't work with fractional weights
17:28 travis-ci joined #perl6-dev
17:28 travis-ci Rakudo build failed. Zoffix Znet 'Merge pull request #1190 from MasterDuke17/test_for_RT132285
17:28 travis-ci https://travis-ci.org/rakudo/rakudo/builds/287594754 https://github.com/rakudo/rakudo/compare/f4f36915f564...e475ae8e0542
17:28 travis-ci left #perl6-dev
17:28 buggable [travis build above] ✓ All failures are due to: GitHub connectivity (1 failure).
17:30 Geth_ ¦ roast: 2a45053f2b | (Zoffix Znet)++ | S17-procasync/kill.t
17:30 Geth_ ¦ roast: Test Proc::Async.kill doesn't hang
17:30 Geth_ ¦ roast:
17:30 Geth_ ¦ roast: If killing immediately after .start
17:30 Geth_ ¦ roast:
17:30 Geth_ ¦ roast: RT#125653: https://rt.perl.org/Ticket/Display.html?id=125653
17:30 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/2a45053f2b
17:30 synopsebot RT#125653 [new]: https://rt.perl.org/Ticket/Display.html?id=125653 [CONC] A promise returned by Proc::Async.start will never be kept or broken if the process is killed by the current process
17:37 Geth_ ¦ roast: 7f52f7a4e4 | (Zoffix Znet)++ | S17-procasync/kill.t
17:37 Geth_ ¦ roast: Simplify tests
17:37 Geth_ ¦ roast:
17:37 Geth_ ¦ roast: - Use a temp file routine instead of rolling the feature manually
17:37 Geth_ ¦ roast: - Use a heredoc instead of poorly idented string
17:37 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/7f52f7a4e4
17:59 Zoffix RT#132258 is making me cry. Not the security part, but the more I research the issue the more it seems there's no sane way to pass args on Windows if you don't know what you're passing them to.
17:59 synopsebot RT#132258 [new]: https://rt.perl.org/Ticket/Display.html?id=132258 [SECURITY][WINDOWS] `run "perl6" ...` can be made to execute shell commands
18:00 Zoffix And we have a bunch of Proc::Async.new: $*EXECUTABLE in roast, which basically will remain LTA, as even with some :cmd-windows-quoting arg to Proc::Async, the tests will work only for implementatinos that use a batch file for perl6 like we do and will break for those that don't :|
18:10 AlexDaniel- joined #perl6-dev
18:19 Geth_ ¦ rakudo/nom: 3acde358c1 | (Samantha McVey)++ | src/core/Int.pm
18:19 Geth_ ¦ rakudo/nom: Show hex value as well as dec in char() error message
18:19 Geth_ ¦ rakudo/nom:
18:19 Geth_ ¦ rakudo/nom: When char() is requested to too large a value, also show the hex value
18:19 Geth_ ¦ rakudo/nom: and not just the decimal one.
18:19 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3acde358c1
18:19 Geth_ ¦ rakudo/nom: 7bea3a2ddd | (Samantha McVey)++ | src/core/IO/Path.pm
18:19 Geth_ ¦ rakudo/nom: Add .parent(Int) for getting up more than one level
18:19 Geth_ ¦ rakudo/nom:
18:19 Geth_ ¦ rakudo/nom: .parent only gets the path one level up. Now .parent(Int) will
18:19 Geth_ ¦ rakudo/nom: get the nth parent of the requested directory. This is different
18:19 Geth_ ¦ rakudo/nom: than adding ../ to the path as when there are links it acts
18:19 Geth_ ¦ rakudo/nom: differently.
18:19 Geth_ ¦ rakudo/nom: In addition:
18:19 Geth_ ¦ rakudo/nom: `Path::IO.add('../').absolute ne Path::IO.parent.absolute`
18:19 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/7bea3a2ddd
18:30 Zoffix FWIW we have UInt type for non-negative Ints
18:35 ugexe almost feels like parent should do cleanup
18:40 [Coke] Zoffix++ #UInt
18:40 [Coke] ugexe: isn't that add?
18:41 ugexe add() adds a path part to a path, parent(1) returns the path with 1 path part removed
18:42 samcv should I change it to UInt?
18:43 ugexe e.g. "foo/../bar".IO.parent does the programmer usually want "foo/" or "foo/../"
18:43 ugexe foo/../ or "" whatever
18:44 samcv ugexe, depends on if they want to remove part of the path or not. ../ will follow links
18:45 samcv most of the time people also don't want to see /foo/bar/baz/raz/this/../../../../../
18:45 samcv but if you want to follow links on the way one level up, then you should use ../
18:46 Zoffix samcv: IMO yeah. It'd unify the error with the rest of our errors rather than having an Ad::Hoc with a custom message (that can't really be specced properly) and moves the logic defining acceptable params into signature, so it can be inspected, for example with .cando()
18:46 samcv ok
18:46 samcv don't need a full error message. just need a more generic error than negative .parent's make no sense :P
18:47 Zoffix m: sub (Int) {}.cando(\(-42)).so.say; sub (UInt) {}.cando(\(-42)).so.say
18:47 camelia rakudo-moar 7bea3a2dd: OUTPUT: «True␤False␤»
18:47 samcv cool :)
18:47 Zoffix Wonder why Block don't got a .cando
18:47 * Zoffix &
20:12 Geth_ ¦ rakudo: dwarring++ created pull request #1191: optimize recently added AT-POS for typed pointers
20:12 Geth_ ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/1191
21:41 Geth_ ¦ rakudo/nom: bc5fbfcbf5 | (David Warring)++ | lib/NativeCall/Types.pm6
21:41 Geth_ ¦ rakudo/nom: optimize recently added AT-POS for typed pointers
21:41 Geth_ ¦ rakudo/nom:
21:41 Geth_ ¦ rakudo/nom: inline the add method and subsequent calls to nativecast and Int and add
21:41 Geth_ ¦ rakudo/nom: methods. Improves performance by 4-5x.
21:41 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/bc5fbfcbf5
21:41 Geth_ ¦ rakudo/nom: a54da11aa6 | (Zoffix Znet)++ (committed using GitHub Web editor) | lib/NativeCall/Types.pm6
21:41 Geth_ ¦ rakudo/nom: Merge pull request #1191 from dwarring/pointer-optimize
21:41 Geth_ ¦ rakudo/nom:
21:41 Geth_ ¦ rakudo/nom: optimize recently added AT-POS for typed pointers
21:41 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a54da11aa6
22:00 tyil Zoffix: alright, thanks
22:00 yoleaux 15:42Z <Zoffix> tyil: the "my App::MPD::AutoQueue still appears twice, even though I removed it from the META.list" was now fixed. If CPAN modules are missing from mp6o while there generated PAUSE .meta files for them, let me know. If PAUSE isn't generating .meta for some dists. Try to track down PAUSE people and ask them why
22:01 Zoffix :)
22:06 Geth_ ¦ rakudo/nom: 012c80f336 | (Nick Logan)++ (committed using GitHub Web editor) | src/core/ThreadPoolScheduler.pm
22:06 Geth_ ¦ rakudo/nom: Remove unused variable
22:06 Geth_ ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/012c80f336
22:13 tec joined #perl6-dev
22:13 tec left #perl6-dev
22:27 timotimo my disappearance (of json::fast) from the ecosystem made JSON::Tiny get the #1 spot on the module citation site :S
22:35 travis-ci joined #perl6-dev
22:35 travis-ci Rakudo build passed. Samantha McVey 'Add .parent(Int) for getting up more than one level
22:36 travis-ci https://travis-ci.org/rakudo/rakudo/builds/287676452 https://github.com/rakudo/rakudo/compare/80cee3622363...7bea3a2dddcf
22:36 travis-ci left #perl6-dev
22:53 Geth_ ¦ roast: 2800632eb0 | (Zoffix Znet)++ | S02-types/baghash.t
22:53 Geth_ ¦ roast: Can .grab() a BagHash key with weight larger than 64 bits
22:53 Geth_ ¦ roast:
22:53 Geth_ ¦ roast: RT#132279: https://rt.perl.org/Ticket/Display.html?id=132279
22:53 Geth_ ¦ roast: review: https://github.com/perl6/roast/commit/2800632eb0
22:53 synopsebot RT#132279 [open]: https://rt.perl.org/Ticket/Display.html?id=132279 [REGRESSION][9999] BagHash.grab with huge values (("foo" => 10000000000000000000).BagHash.grab(1))
22:57 weabot joined #perl6-dev
23:07 ggoebel joined #perl6-dev
23:12 weabot joined #perl6-dev
23:20 BenGoldberg joined #perl6-dev
23:31 nine joined #perl6-dev
23:32 TimToady joined #perl6-dev
23:38 Ben_Goldberg joined #perl6-dev
23:42 Geth_ ¦ rakudo: MasterDuke17++ created pull request #1192: Convert INTERPOLATE into multis
23:42 Geth_ ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/1192
23:53 timotimo joined #perl6-dev

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary