00:01 squashable6 joined 00:24 lucasb left 00:30 Kaiepi joined 00:31 Kaiepi left 01:00 Kaiepi joined 01:34 ZzZombo_ joined 01:36 squashable6 left, ZzZombo left, ZzZombo_ is now known as ZzZombo 01:38 squashable6 joined 02:33 wtwt5237 joined 02:54 ZzZombo_ joined 02:58 ZzZombo left, ZzZombo_ is now known as ZzZombo 03:05 wtwt5237 left 03:14 Geth joined 03:47 squashable6 left 03:48 squashable6 joined
Geth ¦ problem-solving: Kaiepi assigned to jnthn Issue IO::Socket::INET and IO::Socket::Async need a redesign in v6.e github.com/perl6/problem-solving/issues/125 04:48
05:09 evalable6 joined 06:49 sena_kun joined 07:03 ufobat joined 07:14 MasterDuke joined
MasterDuke in the changelog, have we been putting the hashes for the merge commits for PRs, or the individual commits? 07:45
08:13 sena_kun left
lizmat Files=1277, Tests=109545, 206 wallclock secs (28.11 usr 8.21 sys + 2900.75 cusr 267.28 csys = 3204.35 CPU) 08:38
09:54 pmurias joined
pmurias MasterDuke: I forgot to push a missing files commit sorry :( 09:58
tellable6 gist.github.com/065933c011d69e4d6f...ba3cb80c94
gist.github.com/0dab989be2899a9b0d...33b13af591
pmurias MasterDuke: but the way you implemented those ops is exactly how they should be 09:59
10:04 Guest13443 joined
MasterDuke pmurias: cool, wasn't terribly complicated. btw, istr you had some different ideas about how to do 6model stuff, i.e., the stuff in src/vm/jvm/runtime/org/perl6/nqp/truffle/nodes/sixmodel/* might not stick around? 10:12
pmurias hmm, we will need to add sprinkle some truffle magic over how we do method calls to make things fast 10:17
but the STables etc. will very likely stay
MasterDuke k 10:18
was trying to figure out to do some of the ops that make STable calls
pmurias the won't likely change, getStable() abstracts away stuff 10:19
MasterDuke k, good to know 10:20
pmurias generally the fixed sized "builtin" reprs inherit from FixedSizeObject
and P6opaque objects from DynamicObject (which comes from the builtin Truffle object model)
10:43 sena_kun joined
MasterDuke pmurias: btw, i can't even run an empty -e: `./nqp-j nqp-truffle.nqp -e ''java.lang.NoSuchMethodError: com.oracle.truffle.api.impl.TVMCI.setCallTarget(Lcom/oracle/truffle/api/nodes/RootNode;Lcom/oracle/truffle/api/RootCallTarget;)V` 10:43
am i missing something?
|Tux| Rakudo version 2019.07.1-439-gc745d6b10 - MoarVM version 2019.07.1-295-gbc93815a8
csv-ip5xs0.704 - 0.710
csv-ip5xs-206.298 - 7.141
csv-parser21.311 - 21.363
csv-test-xs-200.419 - 0.425
test6.927 - 7.147
test-t1.770 - 1.775
test-t --race0.760 - 0.793
test-t-2029.396 - 30.114
test-t-20 --race8.884 - 9.073
10:48
Geth nqp/truffle: 665cf33322 | (Paweł Murias)++ | 2 files
[truffle] Stub nqp::decodersetlineseps
10:52
nqp/truffle: ee4a2ebc98 | (Paweł Murias)++ | src/vm/jvm/bin/run_tests.pl
[truffle] Add passing test
pmurias MasterDuke: does running scripts as command line arguments work?
MasterDuke: -e works for me 10:53
MasterDuke: this seems running using a wrong jvm
MasterDuke java -versionopenjdk version "1.8.0_222"OpenJDK Runtime Environment (build 1.8.0_222-20190711120915.buildslave.jdk8u-src-tar--b08)OpenJDK 64-Bit GraalVM CE 19.2.0.1 (build 25.222-b08-jvmci-19.2-b02, mixed mode) 11:01
`which javac` gives `/usr/lib/jvm/java-8-graal/bin/javac` 11:02
pmurias: and nope, same error for scripts
pmurias MasterDuke: did it work previously? 11:09
MasterDuke i haven't tried in a while, but yeah 11:15
what does your `java -version` say?
pmurias GraalVM 1.0.0-rc3 (build 25.71-b01-internal-jvmci-0.45, mixed mode) 11:17
paste.debian.net/1109297/ 11:18
11:20 AlexDaniel left
pmurias MasterDuke: maybe I have a too old version 11:20
11:20 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
pmurias MasterDuke: the libs in 3rdparty/truffle need to be updated 11:22
I get them from mvnrepository.com/search?q=truffle-api 11:23
11:23 vrurg left
pmurias I'll go grab some food as I'm starving and I'll update my graal sdk (and the deps too) to the latest version 11:23
12:07 AlexDaniel left 12:10 lucasb joined 12:51 pmurias left 12:53 pmurias joined
pmurias MasterDuke: my crappy mobile internet seems to be interputting the the download of the new version of the graalvm sdk, might take a while till it gets through :( 13:05
13:10 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
Geth nqp/truffle: b8576758c9 | (Paweł Murias)++ | 5 files
[truffle] Update the version of truffle jars in repo to 19.2.0.1
13:14
pmurias MasterDuke: I have updated the deps, but will only test that when the SDK downloads :( 13:15
13:16 AlexDaniel left
Geth roast: tbrowder++ created pull request #591:
Add test for user-formatted pod leading declarator blocks
13:19
13:22 vrurg joined
MasterDuke pmurias: i did a make clean and then reconfigured and rebuilt. now `./nqp-j nqp-truffle.nqp -e ''` gives `java.lang.NoClassDefFoundError: com/oracle/truffle/api/frame/FrameDescriptor at org.perl6.nqp.truffle.adapters.Tru...r.java:57) at 13:28
org.perl6.nqp.truffle.adapters.Tru...n.java:6)`
Geth rakudo: tbrowder++ created pull request #3260:
Allow user to retain formatting of pod leading declarator blocks
13:29
15:10 pmurias left
Geth nqp/truffle: 8a9e57c00a | (Daniel Green)++ | 2 files
[truffle] Update truffle jars in the runners
15:21
MasterDuke pmurias: got it working with github.com/perl6/nqp/commit/8a9e57c00a 15:22
tellable6 MasterDuke, I'll pass your message to pmurias
MasterDuke pmurias: btw, why does it say `runtime NYI: unimplemented QAST::Op for` when for does work? 15:31
tellable6 MasterDuke, I'll pass your message to pmurias
15:44 Guest13443 left
Geth nqp/truffle: a4c8f68cd6 | (Daniel Green)++ | 5 files
[truffle] Implement nqp::getstdin
16:02
16:16 Altai-man_ joined 16:17 Altai-man_ left 16:18 Kaiepi left 16:19 sena_kun left 16:33 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
AlexDaniel vrurg: looks like that fixed it, yeah! Thanks! 16:34
nine So, all green on the release front? 16:39
lizmat Whee!
17:13 cognominal left
lizmat AlexDaniel: should I refrain from pushing atm, or is there a release branch? 17:14
AlexDaniel I just came back home today, I don't know if it's all good 17:17
it should be
kawaii: can you run blin again? 17:18
hopefully last time this release
but yeah, I don't know of any issue that could possibly block the release 17:39
Geth rakudo/die-early-on-duplicate-role: acd3bc6235 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/RoleToClassApplier.nqp
Die early on a duplicate role

When you apply the same role to a class more than once, you wind up with very confusing error about needing to resolve your methods because they are being provided by more than one role. This patch adds a check (by name) to see whether the same role has been specified more than once.
... (16 more lines)
17:50
rakudo: lizmat++ created pull request #3261:
Die early on a duplicate role
17:51
AlexDaniel sooo 18:09
6.d-errata tests are not passing, I think this needs to be cherry-picked github.com/perl6/roast/commit/9513...59c6e2a5d5
but I'm a bit surprised 18:10
I can totally see how people would use ‘localhost’,there's nothing obviously wrong with it 18:12
so it being a test in 6.d (6.c-errata seems to be OK wrt this change), should we add some backcompat? 18:13
nine AlexDaniel: yeah, makes me nervous as well
Geth rakudo/die-early-on-duplicate-role: ad96692087 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/RoleToClassApplier.nqp
Fix situation with same role but different parameterization

Only use the "same role" logic *if* we have found an issue in role composition.
18:18
nine This works just fine in Perl 5 with multiple addresses for localhost: perl -e 'use IO::Socket::INET; my $sock = IO::Socket::INET->new(Listen => 1, LocalAddr => 'localhost', LocalPort => 9000, Proto => 'tcp'); $sock->accept;'
18:25 Kaiepi joined
AlexDaniel Kaiepi: ↑ :) 18:26
vrurg If memory serves me right, the problem was about localhost resolving to IPv6 with IPv6 turned off on a system.
nine But that's just a misconfigured system 18:27
AlexDaniel but it used to work?
Kaiepi what's the problem?
AlexDaniel Kaiepi: github.com/rakudo/rakudo/issues/3262 18:28
vrurg: hey, maybe you can help too. What's up with this? 18:33
c: 2019.07,HEAD use Test; throws-like 'use v6.c; constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative';
committable6 AlexDaniel, gist.github.com/bb91110b5b82468982...fddeb087ed
AlexDaniel Bisected: (2019-07-11) github.com/rakudo/rakudo/commit/07...f610ca7dee
this is the 6.c-errata test that's failing: github.com/perl6/roast/blob/de5f71...tant.t#L39
Kaiepi AlexDaniel, those pass on my machine. what kind of error are you getting? 18:34
AlexDaniel Kaiepi: not ok 23 - Address already in use results in a quit 18:35
Kaiepi: and: not ok 1 - Error on trying to re-use port with UDP bind 18:36
Kaiepi: so in case of IO-Socket-Async-UDP.t there's no error where it expects one 18:37
let's see 18:38
e: my $hostname = ‘localhost’; my $port = 5001; my $sock = IO::Socket::Async.bind-udp($hostname, $port); IO::Socket::Async.bind-udp($hostname, $port); $sock.close()
evalable6 (exit code 1) An operation first awaited:
in block <unit> at /tmp/nz6b9XiIhr line 1

Died with the exception:
address already in use
in block <unit> at /tmp/nz6b9XiIhr line 1
AlexDaniel ok, on my system there's no error :)
m: my $hostname = ‘localhost’; my $port = 5001; my $sock = IO::Socket::Async.bind-udp($hostname, $port); IO::Socket::Async.bind-udp($hostname, $port); $sock.close() 18:39
camelia ( no output )
AlexDaniel and camelia agrees!
Kaiepi hm 18:40
m: slurp '/etc/hosts'
camelia ( no output )
Kaiepi m: say slurp '/etc/hosts'
camelia #
# hosts This file describes a number of hostname-to-address
# mappings for the TCP/IP subsystem. It is mostly
# used at boot time, when no name servers are running.
# On small systems, this …
vrurg AlexDaniel: that's weird bisection. But the test is wrong anyway. throws-like is using EVAL. 'use v6.c;' is not supposed to work in EVALs, we had this discussion with Jonathan once.
Kaiepi m: say slurp('/etc/hosts').lines.grep(* ~~ /localhost/) 18:41
camelia (127.0.0.1 localhost ::1 localhost ipv6-localhost ipv6-loopback)
AlexDaniel c: 075e60a56^,075e60a56 use Test; throws-like 'use v6.c; constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative';
vrurg AlexDaniel: but because it's same language version, it doesn't really matter.
committable6 AlexDaniel, gist.github.com/eeb27640ba927ef82c...0f72d93da5
AlexDaniel clearly bisected correctly ↑
Kaiepi m: say slurp('/etc/hosts').lines.grep(* ~~ /0.0.0.0/)
camelia ()
Kaiepi m: say slurp('/etc/hosts').lines.grep(* ~~ /127.0.0.1/)
camelia (127.0.0.1 localhost)
AlexDaniel c: 075e60a56^,075e60a56 use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative'; 18:42
committable6 AlexDaniel, gist.github.com/79e0b5cfa743d95472...9b5d13e6f2
AlexDaniel c: 075e60a56^,075e60a56 use v6.c; use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative'; 18:43
committable6 AlexDaniel, gist.github.com/c138cfcab7109211f6...41fbda4224
Kaiepi m: say slurp('/etc/hosts').lines.map(* ~~ /\d\d?\d?\.\d\d?\d?\.\d\d?\d?\.\d\d?\d?/)
camelia (Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil 「127.0.0.1」 Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil)
AlexDaniel vrurg: so I can remove `use v6.c` and the test will be failing anyway? :)
c: 2019.07.1,HEAD use v6.c; use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative'; 18:44
vrurg AlexDaniel: it should
committable6 AlexDaniel, gist.github.com/9a6f447872dce7318d...478d2b5db0
vrurg BTW, I would say that the new behaviour is more correct. It is consistent to my %h = "str"; 18:45
AlexDaniel vrurg: I'm still a bit confused. So to fix the test we remove v6.c and change the exception to X::Hash::Store::OddNumber, right?
`use v6.c` I mean, from the EVAL
vrurg We can leave 6.c. At some point I hope to implement unit semantics for EVAL, then throws-like can make use of it. But the error type – I would change it. 18:46
Kaiepi AlexDaniel, what's the output of ifconfig on your machine?
AlexDaniel Kaiepi: I don't have ifconfig 18:47
Kaiepi what os?
AlexDaniel debian
I can give you `ip addr show`
Kaiepi: gist.github.com/AlexDaniel/551e3a1...09fb213da7 18:48
vrurg: soo, I don't get it 18:49
vrurg: here's what I found in blame: github.com/perl6/roast/commit/ddaa...12903c93ed 18:50
vrurg: the commit itself argues that it should be reverted, once we implement the stuff, which you did
but, at the same time, it does check for `X::TypeCheck` 18:51
so why are we not getting X::TypeCheck ?
vrurg AlexDaniel: then let's get it reverted. As to why the exception type changed – beats me! No idea for now. 18:52
Kaiepi AlexDaniel, what happens when you change 0.0.0.0 to 127.0.0.1 in S32-io/IO-Socket-Async.t and rerun the tests? 18:53
AlexDaniel c: 46ef0ea08c^,46ef0ea08c use v6.c; use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative'; 18:55
committable6 AlexDaniel, gist.github.com/f87fc2671f545d9c85...f389941475
AlexDaniel c: 46ef0ea08c^,46ef0ea08c use v6.d.PREVIEW; use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative';
committable6 AlexDaniel, gist.github.com/0bde9fb7c9ca6f43dc...b69774a712
vrurg AlexDaniel: checked back to 2018.12, the construct always throws 'odd number of elements'. Why was it different under the old EVAL semantics – no idea and see no sense in finding this out.
AlexDaniel c: 46ef0ea08c^,46ef0ea08c use Test; throws-like 'use v6.c; constant %hash = "nothash"', X::TypeCheck, 'constant hash requires Associative'; 18:56
committable6 AlexDaniel, ¦46ef0ea08c^,46ef0ea: « 1..2␤ ok 1 - 'use v6.c; constant %hash = "nothash"' died␤ ok 2 - right exception type (X::TypeCheck)␤ok 1 - constant hash requires Associative␤»
AlexDaniel vrurg: have to go earlier, it changed in github.com/rakudo/rakudo/commit/46...96034408c8
I'm still confused about the whole thing
vrurg I see. One sec... 18:57
AlexDaniel Kaiepi: what do you mean? Change 0.0.0.0 where? It's using ‘localhost’
Kaiepi: remember, we're talking about the version in 6.c-errata
Kaiepi ohh
vrurg Spec changes. 6.c was reporting X::TypeCheck, 6.d turned it into correct exception type. Ok, that's the starting point. 18:58
AlexDaniel right, and we're supposed to have backcompat for 6.c for that change
that makes it clearer to me, at least
Kaiepi ohh, ok 18:59
those tests were changed because localhost can resolve to both 127.0.0.1 and ::1 now, but errors don't get thrown unless the last address fails to get used
which breaks the tests that depend on only the first address in localhost ever being used 19:00
vrurg And here we go. That's simple. constant.t doesn't specify the exact language version but uses just 'use v6' and defaults to d. Because EVAL doesn't support version change for now – we get 6.d semantics despite of 'use v6.c' in it.
AlexDaniel vrurg: what do you mean?? 19:03
use v6.c;
github.com/perl6/roast/blob/de5f71...stant.t#L1
vrurg AlexDaniel: so, the fix is to switch to implicit 'use v6.c'. Or use is-run
AlexDaniel: is it in -errata? Because master has v6 19:04
AlexDaniel vrurg: 6.c-errata, and it's failing on 6.c-errata
so use v6.c doesn't help
vrurg: here's a ticket for you too :) github.com/rakudo/rakudo/issues/3263
Kaiepi: right, so it used to resolve to 127.0.0.1 only, correct? 19:05
Kaiepi whatever was defined first for localhost in /etc/hosts, likely
AlexDaniel m: say slurp ‘/etc/hosts’ 19:07
camelia #
# hosts This file describes a number of hostname-to-address
# mappings for the TCP/IP subsystem. It is mostly
# used at boot time, when no name servers are running.
# On small systems, this …
AlexDaniel m: say slurp(‘/etc/hosts’).lines.map(*.words[0]) 19:08
camelia (# # # # # # # # # # Nil 127.0.0.1 Nil # ::1 Nil fe00::0 Nil ff00::0 ff02::1 ff02::2 ff02::3 Nil)
AlexDaniel e: say slurp(‘/etc/hosts’).lines.map(*.words[0])
evalable6 (# # 127.0.0.1 94.23.219.181 2001:41D0:2:5Eb5:: Nil # #(added Nil ::1 fe00::0 ff00::0 ff02::1 ff02::2 ff02::3)
AlexDaniel ok that didn't help at all 19:09
Kaiepi actually that does help, i just screwed up the regex i was trying to do that with earlier
AlexDaniel but they seem to be mostly identical 19:11
yet it works on one system and not the other
Kaiepi m: my IO::Socket::Async $ .= udp: '0.0.0.0', 8000; my IO::Socket::Async $ .= udp: '0.0.0.0', 8000
camelia Too many positionals passed; expected 1 argument but got 3
in block <unit> at <tmp> line 1
Kaiepi m: my IO::Socket::Async $ .= bind-udp: '0.0.0.0', 8000; my IO::Socket::Async $ .= bind-udp: '0.0.0.0', 8000
camelia An operation first awaited:
in block <unit> at <tmp> line 1

Died with the exception:
address already in use
in block <unit> at <tmp> line 1
Kaiepi e: my IO::Socket::Async $ .= bind-udp: '0.0.0.0', 8000; my IO::Socket::Async $ .= bind-udp: '0.0.0.0', 8000
evalable6 (exit code 1) An operation first awaited:
in block <unit> at /tmp/k3b4HTHySb line 1

Died with the exception:
address already in use
in block <unit> at /tmp/k3b4HTHySb line 1
Kaiepi e: my IO::Socket::Async $ .= bind-udp: '0.0.0.0', 8000
evalable6
Kaiepi ok, never mind then 19:12
e: slurp('/etc/hosts').lines.grep(* ~~ /localhost/) 19:14
evalable6
Kaiepi e: say slurp('/etc/hosts').lines.grep(* ~~ /localhost/)
evalable6 (127.0.0.1 localhost.localdomain localhost ::1 ip6-localhost ip6-loopback)
Kaiepi m: say slurp('/etc/hosts').lines.grep(* ~~ /localhost/)
camelia (127.0.0.1 localhost ::1 localhost ipv6-localhost ipv6-loopback)
Kaiepi m: say slurp('/etc/hosts').lines.grep(* ~~ /94.23.219.181/) 19:15
camelia ()
Kaiepi e: say slurp('/etc/hosts').lines.grep(* ~~ /94.23.219.181/)
evalable6 (94.23.219.181 ␉superserver␉superserver)
nine Kaiepi: I can understand trying all resolved addresses when connecting. Or trying them when getting a bind error because the resolved address is not available on any interface. But not when we get a bind error because the address is already in use 19:18
AlexDaniel nap & 19:36
Kaiepi nine, IO::Socket::INET needs to permit reusing addresses in order to maintain compatibility, according to a comment in src/io/syncsocket.c
it'd be weird to have IO::Socket::INET and IO::Socket::Async have the complete opposite behaviour for the same type of error 19:37
besides, the list of addresses returned aren't necessarily the same address, so that would prevent being able to do stuff like binding sockets for every address a host resolves to 19:39
19:43 ufobat_ joined
vrurg AlexDaniel: it was simple to find the cause. Not sure if the fix would be simple too. Basically, Grammar resets compiler language version for any unit – thus, we can have .c or .e core setting with .d compiler inside EVALs. ;) 19:44
19:46 ufobat left 19:47 pmurias joined 20:21 MasterDuke left
nine Kaiepi: I wonder. Why did we invent our own solution to the dual-stack problem instead of implementing Happy Eyeballs standard? 20:31
pmurias .tell MasterDuke nqp::for is not yet implemented, but the optimizer sometimes turns it into things that are 21:16
tellable6 pmurias, I'll pass your message to MasterDuke
Kaiepi i have zero excuse for this nine, i wasn't aware of it at the time of writing these changes. these changes were not something i took as seriously as those that were originally part of the grant because i heavily underestimated the scope of the problem 21:17
i still think the first couple changes included with them are reasonable, but handling when ipv6 hosts aren't reachable is not something i think we should be doing at all if we're not compliant with existing standards for it 21:21
Geth nqp/truffle: 7ba53f2dbf | (Paweł Murias)++ | 2 files
[truffle] Remove dead code needed for a long removed op
21:56
nqp/truffle: f8da3a1dd5 | (Paweł Murias)++ | src/vm/jvm/runtime/org/perl6/nqp/truffle/io/IIOSyncReadable.java
[truffle] Removed unneeded import
22:12 ggoebel left 22:17 Xliff joined
Xliff m: use NativeCall; class A is repr<CStruct> { has uint32 $.a; has CArray[Pointer[A] $.aa; }; 22:17
camelia ===SORRY!===
Cannot resolve caller ACCEPTS(Bool:U: A); none of these signatures match:
(Bool:D: Mu \topic, *%_)
(Bool:U: \topic, *%_)
Xliff m: use NativeCall; class A is repr<CStruct> { has uint32 $.a; has A $.aa; };
camelia ( no output )
Xliff m: use NativeCall; class A is repr<CStruct> { has uint32 $.a; has CArray[Pointer[A] $.aa; };
camelia ===SORRY!===
Cannot resolve caller ACCEPTS(Bool:U: A); none of these signatures match:
(Bool:D: Mu \topic, *%_)
(Bool:U: \topic, *%_)
Xliff ^^ This is a bug. 22:18
m: use NativeCall; class A is repr<CStruct> { has uint32 $.a; has CArray[Pointer[A]] $.aa; }; 22:20
camelia ===SORRY!===
Cannot resolve caller ACCEPTS(Bool:U: A); none of these signatures match:
(Bool:D: Mu \topic, *%_)
(Bool:U: \topic, *%_)
Xliff m: use NativeCall; class A is repr<CStruct> { has uint32 $.a; has CArray[A] $.aa; };
camelia 5===SORRY!5=== Error while compiling <tmp>
An exception occurred while parameterizing CArray
at <tmp>:1
Exception details:
Type check failed in binding to parameter '<anon>'; expected Any but got A (?)
in method parameterize at /ho…
Xliff Bugged as #3264 22:25
22:27 Xliff left 22:48 pmurias left
Geth rakudo: vrurg++ created pull request #3265:
Preserve compiler version in EVALs
23:24
23:29 ufobat_ left 23:39 ufobat joined
Geth roast/rakudo_3263: f9aca25958 | (Vadim Belman)++ | S29-context/eval.t
Add test for EVAL preserving caller's compiler version

Support for rakudo/rakudo#3265,
Related to rakudo/rakudo#3263
23:39
roast: vrurg++ created pull request #592:
Add test for EVAL preserving caller's compiler version
23:40
rakudo: 0eea939fdc | (Vadim Belman)++ | 2 files
Preserve compiler version in EVALs

Previously it was always reset to the default one.
Fixes #3263
rakudo: b45d99d45e | (Vadim Belman)++ (committed using GitHub Web editor) | 2 files
Merge pull request #3265 from vrurg/rakudo_3263

Preserve compiler version in EVALs
roast: f9aca25958 | (Vadim Belman)++ | S29-context/eval.t
Add test for EVAL preserving caller's compiler version

Support for rakudo/rakudo#3265,
Related to rakudo/rakudo#3263
roast: b4c7260717 | (Vadim Belman)++ (committed using GitHub Web editor) | S29-context/eval.t
Merge pull request #592 from perl6/rakudo_3263

Add test for EVAL preserving caller's compiler version