lizmat Files=1231, Tests=75909, 313 wallclock secs (14.48 usr 5.12 sys + 2170.64 cusr 211.60 csys = 2401.84 CPU) 08:43
afk until later today& 09:41
Geth roast: tbrowder++ created pull request #366:
adds test for fix for RT #126740, populates table caption property
13:03
synopsebot RT#126740 [new]: rt.perl.org/Ticket/Display.html?id=126740 [POD] Pod::Block::Table node caption property is not populated properly 13:04
SPEC#366 [open]: github.com/perl6/roast/pull/366 adds test for fix for RT #126740, populates table caption property
Geth rakudo: tbrowder++ created pull request #1303:
fix for RT #126740, populates table caption property, if any; tests a…
13:08
synopsebot RAKUDO#1303 [open]: github.com/rakudo/rakudo/pull/1303 fix for RT #126740, populates table caption property, if any; tests a…
tbrowder .tell AlexDaniel please review PR #1303 (or suggest someone to review it) 13:10
yoleaux tbrowder: I'll pass your message to AlexDaniel.
AlexDaniel . 13:19
yoleaux 13:10Z <tbrowder> AlexDaniel: please review PR #1303 (or suggest someone to review it)
AlexDaniel GH#1303
synopsebot GH#1303 [open]: github.com/rakudo/rakudo/pull/1303 fix for RT #126740, populates table caption property, if any; tests a…
AlexDaniel that looks reasonable 13:22
AlexDaniel looks closer
tbrowder: heh I wonder, what happens if you do :caption(0) ? :) 13:23
tbrowder good question! i’m guessing no caption at all. but this is just for the <> case really. the next PR will be for all the quoting and extraneous quote issues. but 13:39
i will look at the caption(0) which should be an exception imho 13:40
i wanted to get the caption property issue separated from the quoting issue—two clearly separate areas. 13:42
AlexDaniel tbrowder: so should we merge this one for now? 13:53
tbrowder if you please, but first let me check the caption(0) results. what do you think would be a reasonable reasonable response? 14:30
AlexDaniel same as caption(1) ? 14:32
well, similar
tbrowder yes, that’s correct, i think that is part of the other quote issue since it affects all 14:37
the %config values in a table. 14:38
so i recommend merging the PR and test as is
the next PR will have extensive tests for most every e 14:39
*%config value we can think of, and it will warn or fail gracefully otherwise 14:40
AlexDaniel can't we have a TODO test now? :) 14:44
Geth rakudo: 2cd266fe08 | (Tom Browder)++ | src/Perl6/Pod.nqp
fix for RT #126740, populates table caption property, if any; tests are in roast PR
14:45
rakudo: f1be29c0dd | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | src/Perl6/Pod.nqp
Merge pull request #1303 from tbrowder/fix-table-caption

fix for RT#126740, populates table caption property, if any
synopsebot RT#126740 [new]: rt.perl.org/Ticket/Display.html?id=126740 [POD] Pod::Block::Table node caption property is not populated properly
RAKUDO#1303 [closed]: github.com/rakudo/rakudo/pull/1303 fix for RT #126740, populates table caption property, if any; tests a…
tbrowder AlexDaniel: thanks! 14:46
don’t forget the roast PR 14:47
um.
Geth roast: 3ae2640538 | (Tom Browder)++ | S26-documentation/07-tables.t
adds test for fix for RT #126740, populates table caption property
roast: 5bd7083960 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | S26-documentation/07-tables.t
Merge pull request #366 from tbrowder/fix-table-caption

adds test for fix for RT #126740, populates table caption property
synopsebot SPEC#366 [closed]: github.com/perl6/roast/pull/366 adds test for fix for RT #126740, populates table caption property
AlexDaniel tbrowder: 👍 14:48
tbrowder TODO test? in what form
i can start adding skipped tests 14:49
AlexDaniel well I was thinking of a thing for =begin table :caption(0)
tbrowder: todo is better than skip because once it's implemented it'll show up as a warning 14:50
tbrowder before i work on the PR, actually not a bad idea, test-driven development, right?
AlexDaniel yea
tbrowder well, i can also add the test for caption(0) now since the results should not change! 14:51
DrForr Good thing I read that, I was going to file a bug on that this evening. 14:54
tbrowder DrForr: what did you do and what did you expect? 14:56
Geth roast: bc48eed684 | (Tom Browder)++ (committed using GitHub Web editor) | S26-documentation/07-tables.t
add another test for table %config key/value pairs
15:05
tbrowder .ask DrForr ref captions: what did you enter and what did you expect? captions as properties have not been available until PR #1303. 15:12
yoleaux tbrowder: I'll pass your message to DrForr.
DrForr tbrowder: I think I noticed them in the spec and wondered why they weren't done yet is all 15:36
yoleaux 15:12Z <tbrowder> DrForr: ref captions: what did you enter and what did you expect? captions as properties have not been available until PR #1303.
tbrowder Ok, great! User testing would be cool! 15:46
DrForr I was appropriating the spectests so I could have some proper test coverage for Pod::To::BlogspotHTML (which probably wil have Pod::To::HTMLFragment as a side benefit) and noticed the omission. 15:56
Zoffix R#1299 could use more comments 16:45
synopsebot R#1299 [open]: github.com/rakudo/rakudo/pull/1299 Issue 1268: Fix lazy slicing stopping at the first deleted
Zoffix IMO we should revert 456358e and maintain "stops at first hole" behaviour for the foreseeable future. Half-fixing it just muddles things up and completely-fixing it would likely waste some perf for an edge-case problem. 16:47
Unless we could have more than one thing that gives True for nqp::null 16:48
Unless we could have more than one thing that gives True for nqp::isnull
(and have one indicate no thing and another indicate a hole) 16:49
Last night's skip fix actually fixes RT#132184 16:50
\o/ 16:51
synopsebot RT#132184 [new]: rt.perl.org/Ticket/Display.html?id=132184 [6.d] .skip semantics are inconsistent
Geth roast: 49956451de | (Zoffix Znet)++ | S32-list/skip.t
Test Seq.skip

RT#132184: rt.perl.org/Public/Bug/Display.html?id=132184 Rakudo fix: github.com/rakudo/rakudo/commit/854c10c27e
17:00
Zoffix m: my %days := SetHash.new: Date.today … Date.new: '2030-04-02'; %days ∖= %days.grep: *.key.day-of-week > 5; say %days.elems 17:04
camelia 1
Zoffix That used to be an error in RT#132353 17:05
synopsebot RT#132353 [open]: rt.perl.org/Ticket/Display.html?id=132353 [LTA] error with using meta assign ops with bound SetHash
samcv .till pmurias <:space> is a property name alias for White_Space, though if there is <:ascii> then it should be <:InASCII> if it's meant as alias for InBasicLatin
.tell pmurias <:space> is a property name alias for White_Space, though if there is <:ascii> then it should be <:InASCII> if it's meant as alias for InBasicLatin
yoleaux samcv: I'll pass your message to pmurias.
Zoffix But if there's no error, I would've expected it to DWIM and alter the SetHash with values rather than sticking 1 set into it
Geth rakudo: 20d67a3d4d | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test autogenerated methods show up in .^methods

RT#132283: rt.perl.org/Ticket/Display.html?id=132283
17:14
synopsebot RT#132283 [open]: rt.perl.org/Ticket/Display.html?id=132283 [REGRESSION] BUILDALL is listed as one of the methods, maybe that's not right (say $foo.^methods)
Geth roast: 786472ae43 | (Zoffix Znet)++ | S10-packages/require-and-use.t
Test `use lib ''` throws useful error

RT#132249: rt.perl.org/Ticket/Display.html?id=132249
17:17
synopsebot RT#132249 [new]: rt.perl.org/Ticket/Display.html?id=132249 [LTA] use lib ‘’ spills guts (use lib ‘’)
Geth roast: e7962eb855 | (Zoffix Znet)++ | S12-methods/calling_syntax.t
Test `$.` syntax shorthand works with meta-methods

RT#131478: rt.perl.org/Ticket/Display.html?id=131478
17:22
synopsebot RT#131478 [open]: rt.perl.org/Ticket/Display.html?id=131478 Warning about $. when using metamethod
Geth rakudo: 831dab1422 | (Zoffix Znet)++ | t/04-nativecall/05-arrays.t
Test CArray.^shortname shows sane value

RT#131406: rt.perl.org/Ticket/Display.html?id=131406
17:25
synopsebot RT#131406 [open]: rt.perl.org/Ticket/Display.html?id=131406 Bug in Perl6::Metamodel::Naming method set_name. `CArray[Pointer].^shortname` returns 'Pointer]'; should be 'CArray[Pointer]'
Geth rakudo: a563544669 | (Zoffix Znet)++ | t/04-nativecall/05-arrays.t
Restructure test code a bit for nicer look

No functional changes
17:27
roast: b610337bd5 | (Zoffix Znet)++ | S32-num/rat.t
Test Rat literals with numerators over 32 bits work

RT#126103: rt.perl.org/Ticket/Display.html?id=126103 Rakudo fix: github.com/rakudo/rakudo/commit/27...5026fe58fe
17:32
synopsebot RT#126103 [open]: rt.perl.org/Ticket/Display.html?id=126103 [BUG] Rat literal fails for moderately large numerator
AlexDaniel reportable6: list 17:54
reportable6 AlexDaniel, gist.github.com/f4d5f7885e585a3b91...86c155926d 17:55
AlexDaniel reportable6: 2017-12-04T06:07:00Z 2017-12-11T00:00:00Z
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel, gist.github.com/7d691db38e28192c2c...c174f36aa6 17:56
AlexDaniel .tell lizmat irclog.perlgeek.de/perl6-dev/2017-...i_15562338 17:57
yoleaux AlexDaniel: I'll pass your message to lizmat.
Geth roast: 8362ba817a | (Zoffix Znet)++ | S02-types/list.t
Test List.new does not create unwanted containers

Closes github.com/rakudo/rakudo/issues/1260
18:22
roast: 0580a62a80 | (Zoffix Znet)++ | S02-types/WHICH.t
Mixins to Code:U objects do not cause crash on .WHICH

RT#126099: rt.perl.org/Ticket/Display.html?id=126099
18:30
synopsebot RT#126099 [open]: rt.perl.org/Ticket/Display.html?id=126099 [BUG] .WHICH fails for Block but True
Geth rakudo: 1b0be0f631 | (Zoffix Znet)++ | t/05-messages/11-overflow.t
Test overflow in .roll()

RT#125820: rt.perl.org/Ticket/Display.html?id=125820
18:39
synopsebot RT#125820 [new]: rt.perl.org/Ticket/Display.html?id=125820 [9999] .roll(-9999999999999999999999999999999999999999999999999)
Geth roast: fb9e904279 | (Tom Browder)++ (committed using GitHub Web editor) | S26-documentation/07-tables.t
correct test, add test for a number in parens
18:41
rakudo: 062c27f343 | (Zoffix Znet)++ | t/05-messages/11-overflow.t
Test overflow behaviour in .indent()

RT#125816: rt.perl.org/Ticket/Display.html?id=125816
18:44
synopsebot RT#125816 [new]: rt.perl.org/Ticket/Display.html?id=125816 [9999] "x".indent(9999999999999999999999999999999999999999999999999) 18:45
Geth roast: 0d871001ff | (Zoffix Znet)++ | S02-types/native.t
Test meta-assign op with native nums

RT#124083: rt.perl.org/Ticket/Display.html?id=124083
18:56
synopsebot RT#124083 [open]: rt.perl.org/Ticket/Display.html?id=124083 Problem with metaops on num32
Geth roast: 8cbbf89c9e | (Zoffix Znet)++ | S06-operator-overloading/prefix.t
Coverage for crashes in certain custom operator setups

RT#123216: rt.perl.org/Ticket/Display.html?id=123216
19:04
synopsebot RT#123216 [open]: rt.perl.org/Ticket/Display.html?id=123216 [BUG] QRPA: Can't pop from an empty array! 19:05
Geth roast: 08ed79bfbb | (Zoffix Znet)++ | S04-declarations/my.t
Test throwage on conflicting type constraints in `my`

RT#126123: rt.perl.org/Ticket/Display.html?id=126123
19:38
synopsebot RT#126123 [open]: rt.perl.org/Ticket/Display.html?id=126123 [BUG] clashing type constraints accepted
Zoffix AlexDaniel: what dis meen? gist.github.com/19c77dc6fc0fea67b6...8d3dda45f0 19:42
c: af5b39ac7460e49979,af5b39ac7460e49979~1 use lib <sandbox/roast/packages/>; use Test; use Test::Util; chdir $*TMPDIR; is_run 「/a/; print "pass"」, {:out<pass>, :err(''), :0status}
committable6 Zoffix, gist.github.com/a4e3332415bd4642c8...38be8cfdb2
Zoffix c: af5b39ac7460e49979,af5b39ac7460e49979~1 use lib <sandbox/roast/packages/>; use Test; use Test::Util; chdir sandbox/; is_run 「/a/; print "pass"」, {:out<pass>, :err(''), :0status}
committable6 Zoffix, gist.github.com/1815f80ae894a43e39...0e07088ad8
Zoffix AlexDaniel: nm; I guess it's just old rakudo doing stuff. It's like 2015 rakudo 19:46
Geth roast: ca6de727f9 | (Zoffix Znet)++ | S05-metasyntax/regex.t
/a/ in sink context in in uppermost scope does not explode

RT#126972: rt.perl.org/Ticket/Display.html?id=126972
19:49
synopsebot RT#126972 [open]: rt.perl.org/Ticket/Display.html?id=126972 [LTA] error mesage about getlexrelcaller (/a/)
AlexDaniel yeah indeed, looks like old rakudo was expecting $HOME to be writable 20:01
that changed in (2015-12-28) github.com/rakudo/rakudo/commit/4c...fd05fcfe17 and it's kinda sad 20:04
(sad because it seems like you can't bisect stuff with precomp that is older than that)
travis-ci Rakudo build failed. Zoffix Znet 'Test autogenerated methods show up in .^methods 20:07
travis-ci.org/rakudo/rakudo/builds/314889408 github.com/rakudo/rakudo/compare/f...d67a3d4d40
buggable [travis build above] ☠ All failures are due to: failed make test (1 failure). Across all jobs, only t/02-rakudo/repl.t test file failed.
Geth roast: 5cbef9d682 | (Zoffix Znet)++ | S03-metaops/not.t
Test chaining of !before/!after

RT#125575: rt.perl.org/Ticket/Display.html?id=125575
Includes fudges for mixed combinations, pending resolution of R#1304: github.com/rakudo/rakudo/issues/1304
synopsebot RT#125575 [open]: rt.perl.org/Ticket/Display.html?id=125575 [BUG] negation of before/after cannot be chained
R#1304 [open]: github.com/rakudo/rakudo/issues/1304 [LTA] Chainable ops silently fail to be chained when negated
Geth roast: 65336f6eb1 | (Zoffix Znet)++ | S04-exceptions/fail.t
Test Failure.new(Exception.new) does not segfault

RT#124434: rt.perl.org/Ticket/Display.html?id=124434
20:35
synopsebot RT#124434 [open]: rt.perl.org/Ticket/Display.html?id=124434 [BUG] Passing an Exception object to Failure.new segfaults (SEGV) in sink context in Rakudo
Geth rakudo: e57c771cd7 | (Zoffix Znet)++ | 3 files
Rename Test::Helper's is-run `$status` to `$exitcode`

Since that's what we actually are interested in testing.
20:38
rakudo: 1101fea183 | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test Failure.new(Exception.new) does not segfault

RT#124434: rt.perl.org/Ticket/Display.html?id=124434
Less-specific test also included in roast:
  github.com/rakudo/rakudo/commit/e57c771cd7
AlexDaniel wow, Zoffathon :) 20:50
Geth rakudo: fed56be258 | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test List.ASSIGN-POS gives PDG error

RT#125964: rt.perl.org/Ticket/Display.html?id=125964
synopsebot RT#125964 [open]: rt.perl.org/Ticket/Display.html?id=125964 [LTA] assigning to AOB List elem does not hint right
AlexDaniel buggable: tag testneeded
buggable AlexDaniel, There are 48 tickets tagged with TESTNEEDED; See fail.rakudo.party/t/TESTNEEDED for details
[Tux] Rakudo version 2017.11-90-gfed56be25 - MoarVM version 2017.11-28-gba1fb9542
csv-ip5xs1.132 - 1.158
csv-ip5xs-2013.152 - 13.299
csv-parser12.270 - 12.637
csv-test-xs-200.493 - 0.500
test11.720 - 12.369
test-t2.988 - 3.011
test-t --race1.234 - 1.472
test-t-2053.449 - 53.867
test-t-20 --race18.755 - 19.214
21:52
timotimo uh oh 22:03
lizmat ? 22:06
ah, above 3 again?
and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2017/12/11/...injectile/ 22:30
Geth nqp/master: 6 commits pushed by pmurias++ 22:38
jnthn lizmat++ # weekly 22:39
Zoffix .ask jnthn how come `$!signals-by-Str-setup` and `$!signals-setup` are set to `True` *outside* the lock? Isn't there a race that as one thread just enters the lock another one just exited it but hasn't yet set those to True, so the other thread would check it and find it false and duplicate the work? github.com/rakudo/rakudo/commit/1f...bebea8374d 23:22
no bot -_-
lizmat++ # good weekly
Was gonna poke at Signal enum issues since it's a failing spectest on Windows, but no idea how come we got nqp constants and a Signal enum, but then on top of that do Kernel.signal stuff that has different values (e.g. on Windows, SIGHUP is 1 via former and 3 via latter). Are they meant to be different? Are they interchangeable? 23:31
timotimo i think we can't use nqp constants because different systems have different values, right? and we might run an nqp/rakudo that was compiled on a slightly different system maybe? 23:33
Zoffix So all of those should be removed? 23:34
geekosaur the signal values thing is an open bug 23:35
and I think so far it's even eluded jnthn++
Zoffix Failures in t/spec/S02-magicals/args.t and anything else that uses is_run() and passes args RT#132258 23:36
synopsebot RT#132258 [new]: rt.perl.org/Ticket/Display.html?id=132258 [SECURITY][WINDOWS] `run "perl6" ...` can be made to execute shell commands
Zoffix We need to make perl6 a proper executable (at least on Windows), not a cmd.exe script
*are due to the same problem as in that RT, I meant 23:37
travis-ci Rakudo build passed. Zoffix Znet 'Restructure test code a bit for nicer look 23:39
travis-ci.org/rakudo/rakudo/builds/314896354 github.com/rakudo/rakudo/compare/8...6354466908
Geth roast: d315f50251 | (Zoffix Znet)++ | S02-types/num.t
Increase alllowed tolerance in tan()/sec() tests

Part of addressing #320 #232 #197
23:49
synopsebot SPEC#320 [open]: github.com/perl6/roast/issues/320 Windows roast failures just before 2017.09 release
Geth synopsebot: 9f5cc43a28 | (Zoffix Znet)++ (committed using GitHub Web editor) | lib/SB/Plugin/GitHub.pm6
Handle all mentioned Issues on a single line
23:52
Zoffix ZOFFLOP: t/spec/S11-modules/nested.t 23:54
Geth rakudo: 8df8ce09ce | (Zoffix Znet)++ | src/core/Kernel.pm
Move attribute settage inside the lock

As far as I understand it, if we keep settage outside the lock, there's a chance for a race where one thread would just get out of the lock, but not yet set the attribute, while another would go into the lock test the attribute and re-do the stuff it wasn't meant to.
Moving the attribute assignment inside the locks fixes that.
23:55
Zoffix .tell jnthn I moved those inside the locks. If that's incorrect let me know: github.com/rakudo/rakudo/commit/8df8ce09ce
yoleaux Zoffix: I'll pass your message to jnthn.
Zoffix .tell jnthn that was RE this (that bot didn't pass): `jnthn how come `$!signals-by-Str-setup` and `$!signals-setup` are set to `True` *outside* the lock? Isn't there a race that as one thread just enters the lock another one just exited it but hasn't yet set those to True, so the other thread would check it and find it false and duplicate the work?
yoleaux Zoffix: I'll pass your message to jnthn.
Zoffix github.com/rakudo/rakudo/commit/1f...bebea8374d `
.tell jnthn github.com/rakudo/rakudo/commit/1f...bebea8374d
yoleaux Zoffix: I'll pass your message to jnthn.