AlexDaniel m: my uint64 $x; $x = 2⁶⁴-1; say $x 01:27
camelia -1
AlexDaniel :S
lizmat Files=1295, Tests=109679, 210 wallclock secs (28.36 usr 8.33 sys + 2937.42 cusr 275.30 csys = 3249.41 CPU) 08:49
Geth rakudo: 8cfa659017 | (Elizabeth Mattijsen)++ | src/core.c/StrDistance.pm6
Remove unneeded BUILD method
 12:30
Geth rakudo/is-built: e76f8e4d3e | (Elizabeth Mattijsen)++ | 7 files
Implementation of the "is built" attribute trait

has $!a is built; # allow setting of private attribute in .new
   has $.b is built(False); # do *not* allow setting of public attribute in new		 14:03
lizmat .tell <uzl[m]> I've left some comments with the gist 14:25
tellable6 lizmat, I cannot recognize this command. See wiki for some examples: github.com/perl6/whateverable/wiki/Tellable
lizmat .tell uzl[m] I've left some comments with the gist
tellable6 lizmat, I'll pass your message to uzl[m]
Geth rakudo/is-built: 98abd6a730 | (Elizabeth Mattijsen)++ | 7 files
Rename is_settable attribute to is_built
 14:50
rakudo: lizmat++ created pull request #3423:
Is built		 14:52
lizmat argh, that was too quick
vrurg lizmat: not sure renaming the method would avoid the problem with NQPAttr 15:07
lizmat: I think the problem is in .say being called on HOW which is a NQP class. NQPAttribute has has_accessor method as a stub – thus everything just worked with has_accessor. 15:13
lizmat: you'd either need to add 'is_settable' on NQPAttribute in NQP or check for the attribute type. 15:14
Kaiepi m: sub foo(|bar) { }; say &foo.signature.params[0].perl 15:30
camelia |bar is raw
Kaiepi shouldn't this be |bar instead?
lizmat vrurg: already tried adding a stub to NQPAttribute 15:31
vrurg: BTW renaming was not intended to fix the problem, just to get more consistency
vrurg lizmat: I realized it a bit later. :) Just was passing by and saw the commit. 15:32
Geth nqp: 6361e0965a | (Elizabeth Mattijsen)++ | src/how/NQPAttribute.nqp
Add NQPAttribute.is_built method

This appears to be the reason seeing failures in the 'is-built' Rakudo branch.		 15:50
lizmat ok, turns out that adding the method to NQPAttribute *does* fix it 15:54
Geth rakudo: e81b0b686d | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get NQPAttribute.is_built method stub
 15:56
rakudo: e76f8e4d3e | (Elizabeth Mattijsen)++ | 7 files
Implementation of the "is built" attribute trait

has $!a is built; # allow setting of private attribute in .new
   has $.b is built(False); # do *not* allow setting of public attribute in new
rakudo: 98abd6a730 | (Elizabeth Mattijsen)++ | 7 files
Rename is_settable attribute to is_built
rakudo: da9929c89c | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | 7 files
Merge pull request #3423 from rakudo/is-built

Implementation of the "is built" attribute trait
ab5tract flapping NativeCall tests hurt my brain :( 17:07
tellable6 2019-12-10T11:42:25Z #raku <tbrowder> ab6tract when i actually started designing a check-file sub it was not as simple as i first imagined, but, for my use cases, it's becoming clearer
2019-12-11T12:09:04Z #raku <tbrowder> ab6tract i have a usable version of CheckFile that suits most of my use cases, but it may not suit other users
Xliff ab5tract: Mine too. I deal with them constantly with p6-GtkPlus... 17:19
ab5tract Xliff: so I shouldn't be taking it personally against my implementation I guess ... 17:21
not sure if that is good or bad news :/
Xliff nine / jnthn: OK. So what you are telling me is that it's safe to leave .lock files littering the .precomp dir. I *think* I am finally understanding.
nine / jnthn: Are there any facilities for such in nqp, or is it only a Perl6 thing? 17:22
lizmat Perl6 ? 17:23
:-)
Xliff /o\ 17:24
Keep whacking me with that nerfbat, lizmat++.... I will get there eventually.
:)
lizmat :-)
Xliff So who is going to Texas this summer? 17:25
lizmat: I take it you are skipping? :(
lizmat yes, I had 3 reasons not to go 17:26
Xliff You've intimated a few of those. I think.
lizmat 1 is solved (the name), 3 is being solved (how to handle CoC violations)
2 is not solvable for the foreseeable future, in any case not as long as there's a cheeto in the white house 17:27
Xliff Yup! That's the one. I understand, completely. You are singing to the choir. 17:35
Hence the reason I think it might be better to get out of the States for a P-R conference.
Even though I have to jump through a lot more hoops. The main one being: a valid Passport.
lizmat yeah, that can be troublesome in the US 17:40
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = -2.31' 17:41
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3re Str | uint32; my StrOrUint $x = -2.317⏏5'
expecting any of:
infix
infix stopper
postfix
statement end
…
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = -2.31;'
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in single quotes; couldn't find final "'" (corresponding starter was at line 1)
at <tmp>:1
------> 3 Str | uint32; my StrOrUint $x = -2.31;'7⏏5<EOL>
expecting …
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = -2.31;
camelia Type check failed in assignment to $x; expected StrOrUint but got Rat (-2.31)
in block <unit> at <tmp> line 1
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = '2.31';
camelia ( no output )
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = -231;
camelia Type check failed in assignment to $x; expected StrOrUint but got Int (-231)
in block <unit> at <tmp> line 1
Xliff m: use NativeCall; subset StrOrUint where Str | uint32; my StrOrUint $x = uint32(-231); 17:43
camelia Cannot invoke this object (REPR: P6int; uint32)
in block <unit> at <tmp> line 1
Xliff Hmmm...
NativeCall literals for uint/int*??
m: use NativeCall; subset StrOrUint where Str | uint32; my uint32 $ni = -231; my StrOrUint $x = $a; 17:46
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$a' is not declared
at <tmp>:1
------> 3my uint32 $ni = -231; my StrOrUint $x = 7⏏5$a;
Xliff : use NativeCall; subset StrOrUint where Str | uint32; my uint32 $ni = -231; my StrOrUint $x = $ni;
m: use NativeCall; subset StrOrUint where Str | uint32; my uint32 $ni = -231; my StrOrUint $x = $ni;
camelia Type check failed in assignment to $x; expected StrOrUint but got Int (4294967065)
in block <unit> at <tmp> line 1
Xliff ?!??!?
bisectable: use NativeCall; subset StrOrUint where Str | uint32; my uint32 $ni = -231; my StrOrUint $x = $ni;
bisectable6 Xliff, Bisecting by output (old=2015.12 new=da9929c) because on both starting points the exit code is 1
Xliff jnthn: Might this be a problem? ^^ 17:47
bisectable6 Xliff, bisect log: gist.github.com/f7ee65db703c389b1e...5cd1b71e90 17:48
Xliff, (2016-02-18) github.com/rakudo/rakudo/commit/99...4283b48d0a
Geth roast: 6a14ff87f9 | (Elizabeth Mattijsen)++ | S12-attributes/class.t
Add tests for Attribute "is built" trait
 18:20
rakudo: 826c3ead82 | (Elizabeth Mattijsen)++ | src/core.c/Exception.pm6
Make searching for method name suggestions a bit more resilient

Some bugs, especially when dealing with the setting and additions, can cause strange types of values to occur. Make sure that StrDistance only gets strings, and that the suggestion is stored as a string. This should have made searching for the "is_built" problem a *lot* easier.		 18:23
rakudo/master: 5 commits pushed by (Elizabeth Mattijsen)++ 19:09
rakudo/master: 5 commits pushed by (Elizabeth Mattijsen)++ 19:19
Geth rakudo/default-degree: 9fd29cd2c3 | (Elizabeth Mattijsen)++ | src/core.c/Iterable.pm6
Make default degree for hyper/batch make more sense

Currently, the default :degree for .hyper / .race is set to 4. This may be too high for small single core CPU's, and to small for bigger beasts (such as my MBP with 2x8 cores). Set the default to the number of CPU cores available, and introduce the $*DEGREE dynamic variable for setting the default for a given scope.		 19:36
rakudo: lizmat++ created pull request #3424:
Make default degree for hyper/batch make more sense
Geth rakudo/default-degree: 5def05c9a8 | (Elizabeth Mattijsen)++ | src/core.c/Iterable.pm6
Set default degree to cpucores - 1 (with a minumum of 1)
 20:12
nine Xliff: yes, leaving behind .lock files is OK. But maybe we can do without as we may be able to take the lock on any of the involved files. But if in doubt, use a dedicated lock file. We can always try to get rid of it once we got things working 20:44
Xliff: btw. there's actually no good reason for making the lock files hidden. That the repo's one is is just a historical artefact (which is fancy for "I didn't think all that much about it when I did it") 20:48
Geth rakudo: 81311a209c | (Ben Davies)++ | src/core.c/Parameter.pm6
Fix error when calling Parameter.usage-name on unnamed parameters

This now returns Nil like Parameter.name does in this case.		 21:24
Xliff nine: What about the possibility of using a randomly generated temporary name for the --output phase? This way much of the locking can be avoided? 21:28
By output phase, I mean here: github.com/rakudo/rakudo/blob/mast...y.pm6#L289 21:30
Locking would have to be performed, here: github.com/rakudo/rakudo/blob/mast...y.pm6#L351
Geth rakudo: f08cd00fcd | (Ben Davies)++ | src/core.c/Parameter.pm6
Remove redundant "is raw" for capture parameters in Parameter.raku
 21:32
Xliff nine: Temporary name generator implementtion is here: gist.github.com/Xliff/ca96e9b018fa...acf4471502 21:33
...and now... I nap! 21:34
lizmat Kaiepi++ 21:38
Geth rakudo: 9b29bcc4bb | (Elizabeth Mattijsen)++ | src/core.c/Supply-factories.pm6
Implememt fully functional Supply.split

With the same API as Str.split. Remove previously prematurely optimized version and replace it with a wrapper around Str.split.		 21:39
roast: f9ce5c675d | (Elizabeth Mattijsen)++ | S17-supply/split.t
Expand Supply.split testing with regexen
rakudo: 68808c54e3 | (Ben Davies)++ | src/core.c/Parameter.pm6
Fix signature of Parameter.BUILD
 21:42
rakudo/default-degree: 6314277844 | (Elizabeth Mattijsen)++ | src/core.c/Iterable.pm6
Drop the idea of $*DEGREE for now
 21:47
rakudo/master: 4 commits pushed by (Elizabeth Mattijsen)++
rakudo: 3f51728de1 | (Elizabeth Mattijsen)++ | src/core.c/Iterable.pm6
Oops, it should be the max, not the min!
 21:50
Geth rakudo: 348d54dfa7 | (Elizabeth Mattijsen)++ | src/core.c/Supply-factories.pm6
Implement Supply.comb()
 22:19
roast: d99082d565 | (Elizabeth Mattijsen)++ | S17-supply/comb.t
Add tests for Supply.comb()
 22:20
roast: b1d25cfdb7 | (Elizabeth Mattijsen)++ | spectest.data
Make sure we run tyhe Supply.comb tests
 22:23
lizmat and that concludes my hacking for today& 22:25

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!