03:25 leont left 05:10 squashable6 left 05:13 squashable6 joined 05:24 squashable6 left 05:25 squashable6 joined 06:21 Altai-man joined 06:47 sivoais_ joined, sivoais left 06:51 frost-lab joined 07:22 sivoais_ left, sivoais joined 07:52 sivoais left 07:57 sena_kun joined, patrickb joined 07:59 Altai-man left 08:03 sivoais joined
Geth rakudo/rt-6588: 336a21701b | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | src/core.c/IO/Path.pm6
Changes message string

Also DRY by putting it into a constant. Closes Raku/old-issue-tracker#6588 when accepted.
08:52
rakudo: JJ++ created pull request #4146:
Changes message string
08:53
10:16 leont joined
Geth rakudo: 4384fc5c65 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | 5 files
Implement "is implementation-detail" for classes (#4135)

Implement "is implementation-detail" for classes
This implements the "is implementation-detail" trait on classes and roles. It also implements an "is-implementation-detail" method on classes (and roles), that will either return 0 if the class **is not** an implementation detail, and 1 if the class *is* an implementation detail.
10:30
linkable6 RAKUDO#4135 [closed]: github.com/rakudo/rakudo/pull/4135 Implement "is implementation-detail" for classes
rakudo: ed6e2c8b0e | (Elizabeth Mattijsen)++ | src/core.c/Compiler.pm6
Fix problem in Compiler.verbose-config

Spotted by Will Coleda++
10:32
[Tux] Rakudo v2020.12-25-ged6e2c8b0 (v6.d) on MoarVM 2020.12-14-g78f4f1d9e
csv-ip5xs0.816 - 0.851
csv-ip5xs-208.032 - 8.059
csv-parser26.504 - 26.613
csv-test-xs-200.377 - 0.392
test7.725 - 7.909
test-t1.969 - 1.988
test-t --race0.858 - 0.884
test-t-2032.876 - 33.146
test-t-20 --race9.504 - 9.745
10:51
Geth roast/solaris: 1919f16632 | (Nicholas Clark)++ | S16-io/eof.t
/proc/$*PID/status is a binary file on Solaris, so skip this test there.

The test is for reading line-by-line, so trying to change it to handle binary files defeats the intent.
10:52
roast/solaris: 1aa335ddd9 | (Nicholas Clark)++ | S29-os/system.t
Tweak the subtest "run and shell's :cwd" so that it also passes on Solaris.

It turns out that `echo $PWD` isn't *quite* portable. The test assumes that the shell will set the environment variable PWD to the current directory as part of its startup. This is needed, because the implementation detail being tested here is that the command is run with a *different* current directory from the Raku code - ie there is a `chdir` happening as part of the command ... (24 more lines)
roast: nwc10++ created pull request #713:
Two more spectests that need tweaking to work on Solaris
10:53
nwc10 lizmat: I'm not in a positiong to confirm that spectests (still) pass on Win32. I don't think that I goofed my edit 11:01
but "I don't think I goofed my edit" is true even when it turns out that I did
lizmat he 11:08
nwc10 m: -0.0 ** -3.0 11:09
camelia WARNINGS for <tmp>:
Useless use of "-" in expression "-0.0 **" in sink context (line 1)
nwc10 m: say -0.0 ** -3.0
camelia -Inf
nwc10 m: say -0.0 ** -2.5
camelia -Inf
nwc10 m: say -0.0 ** -2.0
camelia -Inf
nwc10 m: say 0.0 ** -2.0
camelia Inf
nwc10 m: say 0.0 ** -3.0
camelia Inf
nwc10 "today I learned" but I think I need to check one of those
yes, I think that -0.0 ** -2.5 should be +inf 11:11
Geth rakudo/Seq.slice: c53d080c3d | (Elizabeth Mattijsen)++ | 2 files
Introducing Seq.slice(indices)

It appears that a "slice" method on Seq has been discussed extensively in the longer ago past. So I thought I'd have a go at it today.
The .slice method takes any number of indices that are monotonocally increasing (that is, any later index value *must* be greater than any ... (17 more lines)
11:14
rakudo/Seq.slice: 9e61a71be7 | (Elizabeth Mattijsen)++ | src/core.c/Any-iterable-methods.pm6
Add return type to Any.slice
nwc10 oh, and -0.0 ** 2.0 should also be +Inf 11:18
Geth rakudo: 63be83d321 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | 2 files
Introducing Seq.slice(indices) (#3983)

Introducing Seq.slice(indices)
It appears that a "slice" method on Seq has been discussed extensively in the longer ago past. So I thought I'd have a go at it today.
... (19 more lines)
11:33
linkable6 RAKUDO#3983 [closed]: github.com/rakudo/rakudo/pull/3983 Introducing Seq.slice(indices)
11:56 Altai-man joined 11:59 sena_kun left 12:03 patrickb left
Geth rakudo/faster-slice-access: e20c14d21c | (Elizabeth Mattijsen)++ | 2 files
Streamline handling of iterable.elems

By making a private method for it. We only need this when a Callable or a Whatever is encountered. And it should inline quickly enough when needed.
12:11
12:19 frost-lab left
lizmat m: my @a = <a b c>; dd @a[1]:delete = 42; dd @a # feels to me this should either fail or assign to @a[1] 12:27
camelia Int @a = 42
Array @a = ["a", Any, "c"]
12:29 squashable6 left 12:31 squashable6 joined 12:41 patrickb joined
lizmat m: my @a = <a b c>;my $b := @a[1].delete; dd $b = 42; dd @a 12:43
camelia No such method 'delete' for invocant of type 'Str'
in block <unit> at <tmp> line 1
lizmat m: my @a = <a b c>;my $b := @a[1]:delete; dd $b = 42; dd @a
camelia Int @a = 42
Array @a = ["a", Any, "c"]
dogbert17 lizmat: what should your slice branch return here? 12:48
m: my @a = ("a", "b", "c", "d", "e", "f"); @a[3, 30, 5]:delete; say @a
camelia [a b c (Any) e]
lizmat the same ? 12:50
there's some debate about DELETE-POS on a non-existent element returning Nil rather than Any, but yeah 12:51
that's another discussion
dogbert17 it does return the same. I naively thought it would return [a b c e] 12:53
lizmat :delete, does not delete elements, just nulls them
there's a PR for introducing an :eject adverb
that *would* eject the element from the arrauy 12:54
*array
didn't get much traction, probably severely bitrotted now
dogbert17 ok, cool, so why does the (Any) show up at index 3 when the delete was at index 30? I'm obviously missing something obvious here. 12:55
dogbert17 is trying to improve the slice test PR 12:58
lizmat yeah, you're looking at the array *after* the deletion: what you see is the result of the deletion of element at 3
the deletion of element at 5 caused the array to shorten as it was the last element 12:59
m: my @a = ("a", "b", "c", "d", "e", "f"); dd @a[3, 30, 5]:delete; dd @a
camelia ("d", Any, "f")
Array @a = ["a", "b", "c", Any, "e"]
lizmat note that in the *result* of the deletion, the Any appears at index 1 13:00
and the question is: should that be an Any or a Nil (in another discussion :-)
dogbert17 thx for the explanation. there will be more questions I sure :-) 13:01
nwc10 m: say -0e0 ** -2.5 13:22
camelia -Inf
nwc10 m: say -0,0 ** -2.5
camelia 0Inf
nwc10 m: say -0.0 ** -2.5
camelia -Inf
nwc10 m: say -0.0 ** -2.0 13:23
camelia -Inf
nwc10 m: say -0e0 ** -2.0
camelia -Inf
nwc10 m: use nqp; say nqp::pow_n(-0e0, -2e0 13:25
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in argument list; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3use nqp; say nqp::pow_n(-0e0, -2e07⏏5<EOL>
nwc10 m: use nqp; say nqp::pow_n(-0e0, -2e0) 13:26
camelia Inf
nwc10 there. +Inf
so NQP isn't "wrong". That's good.
13:26 domidumont joined
nine lizmat: there's nqp::hllboolfor($value, 'Raku') which makes it easy to return proper Bools from NQP code 13:44
lizmat ah? TIL
ok, if that's the case, then I will make them Bools :-)
13:47 domidumont left
Geth rakudo/class-implementation-detail: 4 commits pushed by (Elizabeth Mattijsen)++ 13:50
patrickb I'd value some feedback on github.com/rakudo/rakudo/pull/4138 14:39
dogbert17 lizmat: does the updated github.com/Raku/roast/pull/712 look better? 15:44
lizmat looks pretty good to me :-) 15:45
ah, in the :delete cases, maybe also test the state of the array ?
and also test the :!delete case? 15:46
dogbert17 ok, will fix 15:49
lizmat dogbert17++
15:57 sena_kun joined 15:59 Altai-man left 17:01 domidumont joined
Geth rakudo: nwc10++ created pull request #4148:
Implement `sec`, `asec` and `sech` in terms of other NQP trigonometric functions.
17:16
18:35 domidumont left 19:56 Altai-man joined 19:59 sena_kun left 20:05 b2gills left 20:06 MasterDuke left 20:16 lucasb joined 20:23 MasterDuke joined 20:27 b2gills joined 21:25 Xliff joined
Geth nqp: a53e8a9d94 | (Patrick Böker)++ | 2 files
JVM, JS: Change `procspawnasync` to explicitly take the program name

As the value passed as program name and arg0 need to be different in some cases (e.g. a space in the program name on Windows) we need to retrieve both separately. Rakudo can then dictate what each value should be. Part of a fix for rakudo/rakudo#4117
21:55
nqp: 118062e71c | (Patrick Böker)++ | tools/templates/MOAR_REVISION
Bump MoarVM

Pulls in the changed `nqp::procspawnasync` semantics
linkable6 RAKUDO#4117 [open]: github.com/rakudo/rakudo/issues/4117 routine run bad works in windows 10
nqp: d65a8c1092 | (Patrick Böker)++ (committed using GitHub Web editor) | 3 files
Merge pull request #689 from patrickbkr/fix-win-proc-with-space

JVM, JS: Change `procspawnasync` to explicitly take the program name
rakudo: 570372f1f2 | (Patrick Böker)++ | src/core.c/Proc/Async.pm6
Fix calling programs with spaces in their path on Windows

  `procspawnasync` will now explicitly take the program name as a
separate argument. This allows us to pass a different value for program name and arg0. That's necessary in some cases (e.g. a space in the program name on Windows).
Fixes rakudo/rakudo#4117
22:02
linkable6 RAKUDO#4117 [closed]: github.com/rakudo/rakudo/issues/4117 routine run bad works in windows 10
Geth rakudo: b3367e95ea | (Patrick Böker)++ | tools/templates/NQP_REVISION
Bump NQP

Pulls in the changed semantics of `nqp::procspawnasync`.
rakudo: 86fd7f67ba | (Patrick Böker)++ (committed using GitHub Web editor) | 2 files
Merge pull request #4138 from patrickbkr/fix-win-proc-with-space

Fix calling programs with spaces in their path on Windows
22:06 patrickb left 22:14 Altai-man left 22:19 travis-ci joined
travis-ci NQP build errored. Patrick Böker 'Merge pull request #689 from patrickbkr/fix-win-proc-with-space 22:19
travis-ci.com/Raku/nqp/builds/211361262 github.com/Raku/nqp/compare/de93c6...5a8c109273
22:19 travis-ci left 22:23 patrickb joined 22:29 travis-ci joined
travis-ci NQP build errored. Patrick Böker 'Merge pull request #689 from patrickbkr/fix-win-proc-with-space 22:29
travis-ci.org/Raku/nqp/builds/752192447 github.com/Raku/nqp/compare/de93c6...5a8c109273
22:29 travis-ci left
Geth nqp: patrickbkr++ created pull request #690:
Adapt tests to new `nqp::spawnprocasync` semantics
22:55
22:57 patrickb left 23:01 patrickb joined