02:11 sourceable6 left, bisectable6 left, benchable6 left, coverable6 left, tellable6 left, evalable6 left, nativecallable6 left, quotable6 left, committable6 left, notable6 left, linkable6 left, bloatable6 left, squashable6 left, shareable6 left, releasable6 left, statisfiable6 left, unicodable6 left, greppable6 left 02:12 bloatable6 joined, notable6 joined, quotable6 joined, greppable6 joined, unicodable6 joined, nativecallable6 joined 02:13 bisectable6 joined, tellable6 joined, evalable6 joined, committable6 joined, linkable6 joined, coverable6 joined, sourceable6 joined, statisfiable6 joined 02:14 releasable6 joined, squashable6 joined, benchable6 joined, shareable6 joined 04:41 leont left 06:34 sena_kun joined 09:07 Altai-man joined 09:09 sena_kun left
lizmat Files=1346, Tests=117151, 225 wallclock secs (29.93 usr 8.23 sys + 3130.81 cusr 279.92 csys = 3448.89 CPU) 09:50
tellable6 2020-12-11T18:06:34Z #raku-dev <melezhik> lizmat here is the recent rakudo issues covered by RICH twitter.com/melezhik2/status/13374...2026587137 so if there is no unit tests for them they are still might be checked by RICH testsa
10:23 patrickb joined 10:47 domidumont joined 11:39 MasterDuke joined 12:41 stoned75 joined
stoned75 hi, I'm considering adding some missing bits to EVAL documentation, and I'm wondering if its :filename paramater default value should be documented. I admit I do not quite understand it. 12:45
I do not quite understand the following commit :} github.com/rakudo/rakudo/commit/1e...16864a39b1
nine stoned75: my guess is that this prevents a warning because of an Any being coerced to Str 12:47
12:47 Kaiepi left
stoned75 nine: hum. I guess I'll have to experiment to grasp it 12:48
Altai-man m: sub foo(:$a, :$b = Str) { say "$a and $b" }; foo 12:49
camelia Use of uninitialized value element of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something meaningful.
in sub foo at <tmp> line 1
Use of uninitialized value element of type Str in s…
Altai-man m: sub foo(:$a = Str, :$b = Str) { say "$a and $b" }; foo
camelia Use of uninitialized value element of type Str in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something meaningful.
in sub foo at <tmp> line 1
Use of uninitialized value element of type Str in s…
Altai-man hmm
This is a bad example, because it wants Str:D. 12:50
Geth rakudo: Altai-man assigned to vrurg Issue Blin 2020.12, round 3 github.com/rakudo/rakudo/issues/4108
2d5d3bf4a3 | (Elizabeth Mattijsen)++ | src/core.c/array_slice.pm6

By saving 1 Scalar allocation per call.
Altai-man Consistency is hard.
12:56 Kaiepi joined 13:08 sena_kun joined 13:10 Altai-man left
[TuxCM] Rakudo v2020.11-126-g79b55ac3c (v6.d) on MoarVM 2020.11-85-g82a34e25a
csv-ip5xs0.807 - 0.820
csv-ip5xs-207.934 - 8.198
csv-parser23.883 - 25.531
csv-test-xs-200.385 - 0.390
test7.643 - 9.096
test-t1.916 - 1.928
test-t --race0.841 - 0.887
test-t-2032.835 - 33.230
test-t-20 --race8.922 - 10.116
lizmat afk for a few hours& 13:40
Geth nqp: 994107012d | (Christian Bartolomäus)++ | 2 files
[JVM] Align behaviour for NQP hash to MoarVM

The JVM backend now throws for certain operations on hashes in the same way as MoarVM:
  * Iterating past the end of the iterator dies with "Iteration past
   end of iterator" instead of throwing a NoSuchElementException.
  * Before starting to iterate and after reaching the end of the
   iterator, trying to access key or value gives a sane error message.
nqp: 021853592c | (Jonathan Worthington)++ (committed using GitHub Web editor) | 2 files
Merge pull request #683 from usev6/jvm_vmhash_exceptions

  [JVM] Align behaviour for NQP hash to MoarVM
13:50 b2gills left 14:09 leont joined 14:49 b2gills joined
nwc10 cool, thanks. That was somewhere on my list of "things to investigate and try to implement", but it didn't reach the top yet 15:43
16:20 domidumont left 16:23 stoned75 left 16:27 domidumont joined 16:28 stoned75 joined
Geth nqp: eb4173a365 | (Elizabeth Mattijsen)++ | tools/templates/MOAR_REVISION
Bump MoarVM to get the latest fixes
nwc10 nine: as someone who appears to be an even more dilligent tank watcher than I am... 16:53
on www.youtube.com/watch?v=n5ozYnVQahE the crane is back. Apparently slowly, overnight.
*I* am confused, and I can't find it referenced anywhere
it went from the production site (where it is again now) to the lauch site some weeks ago, on the transporter, slowly, with the long pause. 16:54
It stacked the nosecone of SN8
about a week later it came back
then somehow it warped itself back (back) to the launch site
and I find no reference to that third move.
and I'm confused.
you can't sneak a crane like that around.
(even if it has 2 (or more) names)
Geth rakudo: c63f078a2f | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get latest NQP/MoarVM goodies
17:07 Altai-man joined
Geth nqp: d35e5061bc | Coke++ | docs/ops.markdown
Document setdebugtypename op

[Coke] nwc10: it just respawned, don't worry about it 17:09
nwc10 :-)
[Coke] finds a TODO from over a week ago with a note from nine about how an opcode works
17:10 sena_kun left
Geth nqp: a3e5e0baa5 | Coke++ | docs/ops.markdown
fix typo.

stoned75 I'm trying to understand the relationship between $?FILES and $?FILE, if there is any, can someone point to a relevant piece of code ? so far I've been lost 17:15
*point me 17:16
nine nwc10: SN9 toppled: www.youtube.com/watch?v=aIoEQwCVEtE&t=335s
nwc10 yes, that I know. (Sorry, wasn't clear) 17:20
that's why the crane is back (trip 4)
it's trip 3 I can't find reference to. And I watched a bit of trip 2 live, so I know that trip 2 happened. 17:21
meanwhile, NFS is fun.
stoned75 ah! found it in Perl6::World.current_file 17:28
Geth nqp/atime-with-atime: f7be5c5932 | (Nicholas Clark)++ | t/nqp/019-file-ops.t
floating point atime should be compared with integer atime, not mtime.

Similarly ctime with ctime, not mtime.
This appears to be a copy-paste thinko that dates back to when the test was originally added in 2016 by commit 37862cd52d63eb61:
   Add subsecond file time ops stat_time and lstat_time
nqp: nwc10++ created pull request #688:
floating point atime should be compared with integer atime, not mtime.
17:56 morayj joined
nwc10 gah, the github interface won't let me assign the original culprit as a reviewer :-) 17:57
nine: fun! excitement! crane jib inside highbay! (one out of three is correct) 18:09
18:23 morayj left
nine If it continues like this, I'll have to find another - calmer - channel :D 18:34
nwc10 the live stream is about as exciting as watching paint dry. That should be calm enough for you? 18:35
18:35 domidumont left
nwc10 otherwise, try compiling the setting running under valgrind, and watch that? 18:38
nine Or run spectests with GC_DEBUG = 3. Takes more than a week 18:39
nwc10 oh gee, yes.
IIRC setting took over 24 hours with that
nine I've never committed the patch that makes it _really_ slow though :D 18:40
The one where instead of 2 you get 15 nurseries
nwc10 that might actually be useful as smoething to conditionally enable 18:41
I know that I hacked in something like that 4 or 5 years ago
but I can't remember exactly what I did
nine It catches cases where with GC_DEBUG=3 something would get promoted to gen2 before someone even makes the unprotected copy of a pointer 18:42
github.com/MoarVM/MoarVM/commit/9d...4d9f5b579c 18:44
It's....super pretty 18:45
18:47 leont left 18:52 Altai-man left 18:54 sena_kun joined
Geth roast: ad6b636c41 | (Christian Bartolomäus)++ | S09-typed-arrays/native-num.t
[JVM] Fudge tests that expect specific error

The expected error message has been changed recently (1a84eefa47). The new (more specific) error message stems from dedicated multi candidates for 'sub postcircumfix:<[ ]>'.
On the JVM backend these multi candidates are not used, because some of them cause spectest failures, compare
  * github.com/rakudo/rakudo/commit/60fa48ec8f
  * colabti.org/irclogger/irclogger_lo...-12-06#l67
21:07 Altai-man joined 21:10 sena_kun left 21:14 stoned75 left 21:25 stoned75 joined 21:49 Geth left, Geth joined 22:03 Altai-man left 22:07 patrickb left 22:56 stoned75 left 23:20 finsternis left