Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2015-05-31

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:26 andreoss joined #perl6
00:32 tadzik Hm, I would've used espeak
00:33 tadzik It works.pretty much like say, as long as QA hackathons go
00:34 tadzik Both allow you to have synthesized voice go “carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot” over the entire room :)
00:40 tinyblak joined #perl6
00:42 tinyblak joined #perl6
01:10 SevenWolf joined #perl6
01:21 yoleaux joined #perl6
01:23 timotimo send-more-money-loops has reached this value now: JIT-Compiled Frames
01:23 timotimo 95.03% (30138714)
01:24 timotimo didn't get much faster, though
01:24 BenGoldberg joined #perl6
01:25 timotimo i really wonder why neither &next nor &THROW-NIL gets inlined :\
01:25 timotimo neither by the static optimizer nor by spesh
01:25 skids joined #perl6
01:33 timotimo i'd be glad if people could build a rakudo wih the jit_throw_ops branch of MoarVM and check spec tests and some modules if things are really correct; i'm not 100% sure of the jit implementation of some of these new ops ...
01:33 timotimo with that, i'll probably go towards bed
01:46 yoleaux joined #perl6
01:56 Sqirrel joined #perl6
01:58 andreoss left #perl6
02:03 aborazmeh joined #perl6
02:04 aborazmeh joined #perl6
02:09 vendethiel joined #perl6
02:11 timotimo .tell tony-o_ in general i'm okay with being pm'd by anyone at any time :)
02:11 yoleaux timotimo: I'll pass your message to tony-o_.
02:11 timotimo .tell tony-o in general i'm okay with being pm'd by anyone at any time :)
02:11 yoleaux timotimo: I'll pass your message to tony-o.
02:11 atta joined #perl6
02:34 rmgk_ joined #perl6
02:42 noganex_ joined #perl6
02:49 bin_005 joined #perl6
03:21 zengargoyle hehe, on the `say done` thing, wondering if Coke has figured out that `... ; say done` is preferable to `... && say done`.
03:21 quester joined #perl6
03:24 * zengargoyle has a remote-notify-send command and a port forwarding daemon so that `ssh example.com 'long_job; remote-notify-send done'` works
03:39 zostay m: my $t = ($*SPEC.tmpdir ~ "test"); $t.IO.spurt("test"); my $h = $t.IO.open(:r); $h.read(2); say $h.tell; $h.seek(0, 0); say $h.tell;
03:39 camelia rakudo-moar c2a57e: OUTPUT«Failed to open file /tmptest: permission denied␤  in block <unit> at /tmp/MG8WLOyWTe:1␤␤»
03:40 zostay m: my $t = ($*SPEC.tmpdir ~ "/test"); $t.IO.spurt("test"); my $h = $t.IO.open(:r); $h.read(2); say $h.tell; $h.seek(0, 0); say $h.tell;
03:40 camelia rakudo-moar c2a57e: OUTPUT«2␤0␤»
03:43 zostay m: my $t = ($*SPEC.tmpdir ~ "/test"); $t.IO.spurt("test"); { my $h = $t.IO.open(:r); LEAVE { $h.seek(0, 0); say $h.tell }; $h.read(2); say $h.tell }
03:43 camelia rakudo-moar c2a57e: OUTPUT«2␤0␤»
03:45 zostay m: my $t = ($*SPEC.tmpdir ~ "/test"); $t.IO.spurt("test"); if True { my $h = $t.IO.open(:r); LEAVE { $h.seek(0, 0); say $h.tell }; $h.read(2); say $h.tell }
03:45 camelia rakudo-moar c2a57e: OUTPUT«2␤0␤»
03:49 vendethiel joined #perl6
03:50 zostay m: sub foo { sub bar($s) { $s x 2 } }; say foo("test")
03:50 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/56FKdpz7Dz␤Calling foo(str) will never work with declared signature ()␤at /tmp/56FKdpz7Dz:1␤------> 3sub foo { sub bar($s) { $s x 2 } }; say 7⏏5foo("test")␤»
03:50 zostay m: sub foo() { sub bar($s) { $s x 2 } }; say foo("test")
03:50 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/h8jjRdfhw4␤Calling foo(str) will never work with declared signature ()␤at /tmp/h8jjRdfhw4:1␤------> 3b foo() { sub bar($s) { $s x 2 } }; say 7⏏5foo("test")␤»
03:50 zostay m: sub foo() { sub bar($s) { $s x 2 } }; say foo()("test")
03:50 camelia rakudo-moar c2a57e: OUTPUT«testtest␤»
03:52 tinyblak_ joined #perl6
04:06 davido_ joined #perl6
04:32 ugexe i dont think theres much reason to use $*SPEC.tmpdir over $*TMPDIR is there?
04:58 zostay only because i didn't know there was a $*TMPDIR
04:59 zostay m: class X is Excpeption { }; die X.new;
04:59 camelia rakudo-moar c2a57e: OUTPUT«X<140648842773536>␤  in block <unit> at /tmp/y5uA2qAI7H:1␤␤»
04:59 zostay m: class X is Exception { }; die X.new;
04:59 camelia rakudo-moar c2a57e: OUTPUT«Unhandled exception: concatenate requires a concrete string, but got null␤   at <unknown>:1  (/home/camelia/rakudo-inst-2/share/perl6/runtime/​CORE.setting.moarvm:print_exception:4294967295)␤ from src/gen/m-CORE.setting:15058  (/home/camelia/rakudo-inst…»
05:00 zostay seems like there could be a nicer error like "Exception is missing method message()" or something rather than that mess
05:01 zostay or just provide a default that returns "Bad Stuff" ;)
05:25 kaare__ joined #perl6
05:28 mr-foobar joined #perl6
05:33 bin_005 joined #perl6
05:42 masak morning, #perl6
05:46 Psyche^ joined #perl6
05:46 kurahaupo joined #perl6
05:47 masak tadzik: thanks to you, I just had my laptop say 'carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot carrot' to the whole room ;)
05:47 masak though the room is empty apart from me... :)
05:48 masak oh, and a symlink later, now I have it as `say` :)
05:59 masak lizmat: `[:a]` triggers the reduce action at Actions.nqp:5811. `[:a,]` doesn't, nor `[(:a)]`.
06:01 rindolf joined #perl6
06:01 tadzik masak: there's more when it comes from :) Try hhhhhhhh, wwwwwwww similar :) The more the funnier
06:02 masak :D
06:09 masak lizmat: ah.
06:09 masak lizmat: the parser gets stuck in this adverb-handling logic: https://github.com/rakudo/rakudo/commit/52d708
06:10 masak it makes a sick, twisted kind of sense. operator adverbs, such as `3 + 4 :gently`, happen in "infix position" -- that is, when the parser has just seen a term and is ready to see another infix.
06:11 masak but in the case of `[:a]` (or any kind of reduce op), an adverb in that position doesn't make any sense. there's no previous operator to modify after the `[`
06:12 masak so the short-term fix could be to not consider that case if we're in a 'red'...
06:12 * masak tries that
06:14 masak yep, that fixes it.
06:15 masak -        | <colonpair> <fake_infix> { $*OPER := $<fake_infix> }
06:15 masak +        | <!{ $*IN_REDUCE }> <colonpair> <fake_infix> { $*OPER := $<fake_infix> }
06:15 * masak makes spectest
06:16 dalek roast: f7a1b9c | lizmat++ | S02-types/WHICH.t:
06:16 dalek roast: Obsolete is no more
06:16 dalek roast: review: https://github.com/perl6/roast/commit/f7a1b9c39c
06:16 dalek roast: 7d6a177 | lizmat++ | S02-types/deprecations.t:
06:16 dalek roast: Merge *OS/*OSVER test now they're compile time
06:16 dalek roast: review: https://github.com/perl6/roast/commit/7d6a177192
06:17 lizmat morning masak!  :-)
06:17 masak lizmat: morning! fixed that for you! :D
06:17 masak well, spectesting a fix now...
06:18 masak let's not get ahead of ourselves (like the parser) :P
06:19 masak basically, this was yet another "Leaky Unification"
06:19 masak amazingly, we've had this code path (and presumably the `[:a]` bug) since Nov 7, 2012.
06:20 itz2 joined #perl6
06:20 masak yeah, nothing in the spectests looks like my doing.
06:23 dalek rakudo/nom: 37e2c9c | (Carl Masak)++ | src/Perl6/Grammar.nqp:
06:23 dalek rakudo/nom: Don't consider op adverbs inside of reduce metaop
06:23 dalek rakudo/nom:
06:23 dalek rakudo/nom: This feels like a stopgap solution, in the sense of plugging the one gap that
06:23 dalek rakudo/nom: we know about in this case. A more unified approach would be welcome.
06:23 dalek rakudo/nom:
06:23 dalek rakudo/nom: This fixes RT #125289.
06:23 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125289
06:23 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/37e2c9c0cc
06:23 masak the thing that feels uneasy about the solution is that it breaks separation of concerns a little.
06:24 masak but that kind of breakage happens quite a bit in the Perl 6 grammar, sending signals all over the place with contextuals.
06:25 masak so maybe that's nothing to worry about.
06:25 masak but it could be that there's a cuter, more long-term solution.
06:27 masak breakfast &
06:32 bin_005 joined #perl6
06:32 lizmat ditto&
06:39 mr-foobar joined #perl6
07:15 vendethiel joined #perl6
07:24 gfldex joined #perl6
08:07 dalek roast: c2effba | lizmat++ | S03-metaops/reduce.t:
08:07 dalek roast: Add some tests for #125289, masak++
08:07 dalek roast: review: https://github.com/perl6/roast/commit/c2effba322
08:07 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125289
08:08 darutoko joined #perl6
08:15 espadrine joined #perl6
08:18 domidumont joined #perl6
08:22 domidumont joined #perl6
08:23 itz2 joined #perl6
08:28 masak lizmat++
08:38 dalek rakudo/nom: 2976e70 | lizmat++ | src/Perl6/Grammar.nqp:
08:38 dalek rakudo/nom: Remove whitespace at EOL
08:38 dalek rakudo/nom:
08:38 dalek rakudo/nom: git diff --ignore-space-at-eol is clean
08:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2976e70ec5
08:38 dalek rakudo/nom: e9265ae | lizmat++ | src/Perl6/Actions.nqp:
08:38 dalek rakudo/nom: Remove whitespace at EOL
08:38 dalek rakudo/nom:
08:38 dalek rakudo/nom: git diff --ignore-space-at-eol is clean
08:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e9265aea5d
08:38 dalek rakudo/nom: 3f510ca | lizmat++ | src/Perl6/World.nqp:
08:38 dalek rakudo/nom: Remove whitespace at EOL
08:38 dalek rakudo/nom:
08:38 dalek rakudo/nom: git diff --ignore-space-at-eol is clean
08:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3f510ca990
08:40 amurf joined #perl6
08:47 quester left #perl6
08:49 dalek rakudo/nom: 94eed8f | lizmat++ | src/core/ (6 files):
08:49 dalek rakudo/nom: Remove whitespace at EOL
08:49 dalek rakudo/nom:
08:49 dalek rakudo/nom: git diff --ignore-space-at-eol is clean
08:49 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/94eed8f413
08:57 tinyblak joined #perl6
09:00 brrt joined #perl6
09:01 bin_005 joined #perl6
09:05 RabidGravy joined #perl6
09:07 * moritz idly wonders why ssh-keygen can generate 4096 bit RSA keys basically in an instant, while gnupg takes several minutes, running out of entropy every so often
09:08 grondilu I have libreadline-dev installed and yet no history or other editing shortcuts in my REPL :(
09:09 itz2 joined #perl6
09:09 bin_005_l joined #perl6
09:11 moritz grondilu: did you have it installed when you configured MoarVM?
09:13 grondilu I think so.  I don't remember having installed it myself, so I believe it was installed from scratch.
09:15 masak I would say https://rt.perl.org/Ticket/Display.html?id=125216 comes down to this: is there a way in C to distinguish 0 and -0 (as n64 numbers)?
09:16 masak and if there is, (a) could someone point me to how, and (b) will that slow down the comparison for everyone?
09:19 dalek specs: df6900d | moritz++ | pod6-files:
09:19 dalek specs: Add a file which contains meta information about the pod6 files in this repo
09:19 dalek specs:
09:19 dalek specs: I hope to automatically generate HTML from pod6 for the website.
09:19 dalek specs: review: https://github.com/perl6/specs/commit/df6900d00c
09:20 moritz masak: fwiw I'm vary of breaking the symmetry between == and === on value-type numbers
09:21 masak ok, that is something I'm willing to stop and discuss :)
09:21 masak moritz: isn't that symmetry already broken (but in "the other way") for NaN?
09:21 moritz m: say NaN == NaN, NaN === NaN
09:21 camelia rakudo-moar c2a57e: OUTPUT«FalseTrue␤»
09:22 moritz masak: right, I didn't consider that
09:22 masak (which I really like)
09:22 moritz masak: consider my argument retracted
09:22 masak yay
09:22 * masak won, with LOGIC o/
09:22 masak boolean logic, even
09:23 * moritz hates people who can't change their minds
09:23 masak I used to think like that... :P
09:23 masak moritz: "strong opinions, weakly held" -- found that phrase recently. really like it.
09:26 masak ok, right now infix:<===> for Nums does exactly the same job as infix:<==>
09:26 masak but then NaN/NaN is special-cased.
09:27 masak I guess I could solve this with special cases for 0e0/-0e0 and -0e0/0e0
09:27 * masak tries that solution
09:29 lizmat afk for quite a while&
09:29 moritz masak: do we even generate -0e0 correctly?
09:29 moritz m: say -0e0
09:29 camelia rakudo-moar c2a57e: OUTPUT«-0␤»
09:30 moritz ah yes, we do
09:31 larion joined #perl6
09:39 dalek mu: 17d94d5 | moritz++ | util/update-design.perl6.org.sh:
09:39 dalek mu: Change util/update-design.perl6.org.sh to be run on hack.p6c.org
09:39 dalek mu:
09:39 dalek mu: right now it runs on www.p6c.org, but we do not have a perl6 + Pod::To::HTML
09:39 dalek mu: installation there, so we cannot easily compile the pod6 files to HTML there.
09:39 dalek mu: review: https://github.com/perl6/mu/commit/17d94d5965
09:43 larion joined #perl6
09:44 dalek mu: 558293a | moritz++ | util/update-design.perl6.org.sh:
09:44 dalek mu: Make DEST_DIR a temporary directory
09:44 dalek mu:
09:44 dalek mu: saves the problem of creating it and/or cleaning it out
09:44 dalek mu: review: https://github.com/perl6/mu/commit/558293a61d
09:47 dalek mu: e597d81 | moritz++ | util/update-design.perl6.org.sh:
09:47 dalek mu: Only create temp dir if necessary
09:47 dalek mu: review: https://github.com/perl6/mu/commit/e597d81b64
09:47 moritz (passwordless ssh)++
09:47 moritz if it didn't exist, it'd need to be invented immediately :-)
09:53 vendethiel joined #perl6
09:53 larion joined #perl6
09:53 dalek mu: a0609ab | moritz++ | util/update-design.perl6.org.sh:
09:53 dalek mu: Abort on smartlinks.pl error
09:53 dalek mu: review: https://github.com/perl6/mu/commit/a0609ab3fc
10:00 dalek mu: f80d1a3 | moritz++ | util/update-design.perl6.org.sh:
10:00 dalek mu: Make tempdir world-readable before rsyncing
10:00 dalek mu:
10:00 dalek mu: otherwise Apache will not be able to read the generated files
10:00 dalek mu: review: https://github.com/perl6/mu/commit/f80d1a30d4
10:01 moritz \o/ we now have a http://design.perl6.org/S15.html
10:02 moritz on the downside, one of the tables is rendered completely bonkers
10:02 telex joined #perl6
10:04 smls joined #perl6
10:04 moritz and now the pod6 files are converted to HTML from a cron job too
10:06 smls What's the Perl 6 way of having a "state" variable shared by two subroutines?
10:06 smls The P5 idiom won't work anymore because subs are lexical:
10:06 smls BEGIN { my $state = ...; sub a {...}; sub b {...} }
10:09 AlexDaniel joined #perl6
10:10 smls I suppose I'll have to replace the variable with a sub of the form    sub mystate($new?) { state $x; defined($new) ?? $x = $new !! $x }
10:10 timotimo moritz: awesome work!
10:15 FROGGS joined #perl6
10:15 smls m: sub a { return-rw (state $x) }; a() = 4; say $a
10:15 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/zs4DPdUXS4␤Variable '$a' is not declared. Did you mean '&a'?␤at /tmp/zs4DPdUXS4:1␤------> 3 { return-rw (state $x) }; a() = 4; say 7⏏5$a␤»
10:16 smls m: sub a { return-rw (state $x) }; a() = 4; say a()
10:16 camelia rakudo-moar c2a57e: OUTPUT«4␤»
10:16 smls nice
10:37 grondilu m: sub a is rw { state $ }; a() = 5; say a();
10:37 camelia rakudo-moar c2a57e: OUTPUT«5␤»
10:38 vendethiel m: sub a is rw { $ }; a() = 5; say a; # :P
10:38 camelia rakudo-moar c2a57e: OUTPUT«5␤»
10:38 smls :)
10:39 smls m: my $a = 5; say ($a += 6)
10:39 camelia rakudo-moar c2a57e: OUTPUT«11␤»
10:39 smls ^^ are += and similar operators guaranteed to return the new value *after* the assignment?
10:40 * grondilu expects them to
10:41 vendethiel smls: it seems so http://design.perl6.org/S03​.html#Assignment_operators
10:41 smls ok
11:10 nowan_ joined #perl6
11:30 diana_olhovik_ joined #perl6
11:51 DrForr They should get rewritten to $a.+=(6); #notionally, likely does not correspond to real Perl6 code.) so I'd assume so
11:51 DrForr Damn, now I need curry for dinner :)
12:08 koo6 joined #perl6
12:17 amurf joined #perl6
12:19 spider-mario joined #perl6
12:21 cognominal joined #perl6
12:31 yqt joined #perl6
12:32 newbie joined #perl6
12:33 itz2 joined #perl6
12:48 pecastro joined #perl6
12:52 tinyblak joined #perl6
13:12 BenGoldberg joined #perl6
13:13 aborazmeh joined #perl6
13:13 aborazmeh joined #perl6
13:13 larion joined #perl6
13:23 lolisa joined #perl6
13:36 orliesaurus joined #perl6
13:36 orliesaurus left #perl6
13:44 rindolf joined #perl6
13:56 beastd joined #perl6
14:07 amurf joined #perl6
14:34 Peter_R joined #perl6
15:15 llfourn how to do equiv of $obj->$method_in_var in p6?
15:16 timotimo you mean $method_in_var contains the method name or a method object?
15:16 llfourn name :)
15:16 timotimo ok
15:16 timotimo $object."$methodname"(...)
15:17 llfourn ahhh ofc thanks
15:17 timotimo you're welcome :)
15:17 llfourn btw I have got the -> thing working!
15:17 llfourn the slang -- thanks for the help yesterday
15:18 timotimo cool :)
15:25 FROGGS joined #perl6
15:27 moritz \o
15:28 moritz any objections to moving docs/feather/syn_index.html in the mu repo to specs/index.html?
15:28 skids joined #perl6
15:28 hoelzro [Coke]: is there a ticket open for that blocking-leading.t stuff?
15:31 kaare__ joined #perl6
15:33 dalek specs: 9c0d37f | moritz++ | index.html:
15:33 dalek specs: Import index.html from mu:docs/feather/syn_index.html
15:33 dalek specs: review: https://github.com/perl6/specs/commit/9c0d37fe9e
15:33 dalek specs: 4dda21f | moritz++ | html/ (4 files):
15:33 dalek specs: Import some more auxilliary files from mu:/docs/feather/
15:33 dalek specs: review: https://github.com/perl6/specs/commit/4dda21f83c
15:33 dalek mu: 2b903d2 | moritz++ | util/update-design.perl6.org.sh:
15:34 dalek mu: Adjust paths for synopses HTML generation
15:34 dalek mu: review: https://github.com/perl6/mu/commit/2b903d2382
15:37 salv0 joined #perl6
15:42 larion joined #perl6
15:47 domidumont joined #perl6
15:47 dalek specs: d736df3 | moritz++ | / (2 files):
15:47 dalek specs: index.html cleanup
15:47 dalek specs:
15:47 dalek specs: * delete accidentally duplicated index.html
15:47 dalek specs: * remove passages that aren't true anymore
15:47 dalek specs: * more "specification" -> "design doc" wording changes
15:47 dalek specs: review: https://github.com/perl6/specs/commit/d736df3dfd
16:03 zakharyas joined #perl6
16:07 dalek specs: 19928e3 | moritz++ | html/ (2 files):
16:07 dalek specs: index.html: get rid of a column
16:07 dalek specs:
16:07 dalek specs: instead of heaving a name/description, and a separate "Synopsis" column as a link,
16:07 dalek specs: make the name/description the link to the HTML version of the synopsis.
16:07 dalek specs:
16:07 dalek specs: I hope this makes it clearer that the synopsis is the most interesting
16:07 dalek specs: of these documents, and the "default" thing to go to
16:07 dalek specs: review: https://github.com/perl6/specs/commit/19928e38dc
16:10 moritz feedback on the new http://design.perl6.org/ front page welcome
16:11 RabidGravy anyone more familiar with python than I am explain why you might do "xdg_data_dirs = [x for x in xdg_data_dirs if x]"?
16:11 RabidGravy looks pointless to me
16:13 moritz RabidGravy: it removes false entries from xdg_data_dirs
16:13 moritz (None, empty list, 0)
16:14 RabidGravy ah okay, my brane was parsing it wrong
16:14 RabidGravy cheers
16:19 moritz yes, the placement of the "if" is weird
16:19 moritz it actually applies inside the loop
16:23 tinyblak joined #perl6
16:25 timotimo moritz: i welcome that change very much
16:37 tinyblak joined #perl6
16:55 ugexe seems odd it needs javascript enabled to view proper alignment
17:01 moritz ugexe: huh?
17:01 moritz without javascript, more text is shown by default
17:01 moritz with javascript, it's behind a [read more] link
17:01 moritz but I don't see any alignm difference
17:02 ugexe if i have javascript enabled, the links in the first column are center aligned in the column and the non-links are left aligned
17:02 ugexe er, if i have js disabled^
17:02 ugexe if i have js enabled, they are all left aligned
17:02 moritz ugexe: try a ctrl+shift+r to reload the CSS?
17:03 ugexe ah foolish me
17:03 moritz the synopsis links are now inside <th> elements, which are center-aligned by default, but the new CSS should left-align them
17:07 ugexe can a grammar read a stream of data? like the theoretical `~=` here: `while my $r = $sock.recv { $grammar.parse ~= $r }` such that the match would be froze when it reaches the end (assuming it has not failed yet) and wait for more input?
17:08 Ben_Goldberg joined #perl6
17:09 timotimo we want something like that at some point, yeah
17:10 ugexe ah. its nice i can emit data out of the grammar before its finished... just need to emit data into it now :)
17:14 mr-foobar joined #perl6
17:21 andreoss joined #perl6
17:28 FROGGS joined #perl6
17:40 TimToady m: BEGIN (my $state = 0; sub a {say ++$state}; sub b {say --$state}); a;a;a;b;b
17:40 yoleaux 29 May 2015 21:46Z <japhb> TimToady: Why doesn't '*@foo!' in a signature mean "slurp one or more"?  Was that a conscious decision or just not yet defined to have a meaning?
17:40 camelia rakudo-moar c2a57e: OUTPUT«1␤2␤3␤2␤1␤»
17:41 TimToady .tell smls see http://irclog.perlgeek.de/p​erl6/2015-05-31#i_10681349
17:41 yoleaux TimToady: I'll pass your message to smls.
17:41 TimToady japhb: I'd much rather see +@foo than *@foo!
17:45 amurf joined #perl6
17:58 xfix joined #perl6
18:09 rindolf Hi all. Sup?
18:09 * rindolf is psyched about perl-5.22.0.
18:10 * sjn is psyched about perl6-1.0.0.
18:10 sjn or something :)
18:20 Sqirrel joined #perl6
18:26 Sqirrel_ joined #perl6
18:26 dalek perl6-roast-data: 20624b8 | coke++ | / (9 files):
18:26 dalek perl6-roast-data: today (automated commit)
18:26 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/20624b85d6
18:29 spider-mario joined #perl6
18:35 Sqirrel joined #perl6
18:42 DrForr m: my %x; %x<a> = \%x; say %x
18:42 camelia rakudo-moar c2a57e: OUTPUT«(timeout)»
18:43 DrForr That bug's still there, eh.
18:44 dalek specs: 92fb08d | (Anthony Parsons)++ | html/perl.css:
18:44 dalek specs: Shuffle colours around for accessibility
18:44 dalek specs:
18:44 dalek specs: Some LHF - the page was assuming nobody would ever change their browser default
18:44 dalek specs: text colour from black, and most of these are redundant hard-coded declarations
18:44 dalek specs: of default-defaults anyway.
18:44 dalek specs: review: https://github.com/perl6/specs/commit/92fb08df87
18:45 flussence (I'm scared to do anything more with that file... the first line says it's for modern browsers and the second mentions netscape 4)
18:51 moritz flussence: feel free to completely redo it :-)
18:52 moritz flussence: afaict it was copied from perl.org before the last redesign
18:52 moritz and no, we don't need IE7 support
18:52 flussence heh :)
18:52 moritz I guess dropping IE8 support would be fine too (as long as it's not rendered completely bonkers with IE8)
18:53 smls joined #perl6
18:54 smls good evening
18:54 yoleaux 17:41Z <TimToady> smls: see http://irclog.perlgeek.de/p​erl6/2015-05-31#i_10681349
18:54 smls :P
18:55 yqt joined #perl6
18:57 smls TimToady: re "I'd much rather see +@foo than *@foo!", the problem with +@foo is that unlike *@foo it already has a commonly used meaning outside of signatures, which could make it confusing.
18:58 smls Why is "*@arg" using a prefix but "$arg!" a postfix, anyway?
18:59 flussence .oO( and why aren't the + and * postfix, like they are in regex? )
18:59 smls yeahm,   $foo?   @foo*   @foo+   could really push the analogy with regexes to the limit ;)
18:59 mr-foobar joined #perl6
19:00 flussence oh, I totally forgot about ?... and it's already used that way!
19:00 smls yeah
19:01 flussence +1 to that, as long as we don't start allowing things like @var+? in the middle of signatures :)
19:01 smls there's probably some reason for the way things are, though
19:03 smls probably something involving low-level parsing issues that only TimToady understands ;)
19:03 timotimo i finally got to see The Princess Bride
19:03 [Tux] joined #perl6
19:05 ugexe im sorry
19:06 timotimo hm?
19:07 ugexe could have seen the new mad max instead
19:11 smls m: my sub aaa (*$a) { say $a.perl }; aaa
19:11 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/YyUSkyi6OU␤Calling aaa() will never work with declared signature (Any $a)␤at /tmp/YyUSkyi6OU:1␤------> 3my sub aaa (*$a) { say $a.perl }; 7⏏5aaa␤»
19:11 smls ^^ Looks like the * is silently ignored there.
19:11 moritz yes
19:14 vendethiel I think we know we don't know what that's supposed to do, right?
19:14 vendethiel *g*
19:17 timotimo ugexe: i'd rather watch "Mario Kart: Fury Road"
19:18 ugexe ill watch if they put the speaker truck + flamethrower guitar shtick in it
19:19 DrForr King Koopa needs a soundtrack.
19:20 DrForr As long as he's voiced by Anthony Hopkins :)
19:30 [Coke] hoelzro: no ticket yet, sorry.
19:33 kurahaupo1 joined #perl6
19:38 SevenWolf joined #perl6
19:39 bin_005_l joined #perl6
19:52 koo6 joined #perl6
19:52 bin_005 joined #perl6
19:57 japhb timotimo: "You keep using that word.  I do not think it means what you think it means."
19:57 japhb "Iocaine powder.  I'd bet my life on it."
19:57 japhb Oh dangit, now I got started ....
20:01 gfldex joined #perl6
20:05 japhb Does anyone happen to know where jnthn put the event processing framework he wrote for the "objects intersect concurrency" talk?  He's got GitHub repos for OO::Monitors and OO::Actors, but I didn't see one for the last framework.
20:06 japhb I wonder if $jnthn.marriage has already happened ...
20:06 timotimo japhb: i found that line about the powder quite amusing myself
20:09 japhb timotimo: One of the things I love about that film is that if you stop to think about almost any line in the movie, you'll realize there's a joke in it.  "Now where is that secret knot?  It's impossible to find ...."  *immediately finds it*
20:12 dalek specs: aac6f72 | (Anthony Parsons)++ | html/ (2 files):
20:12 dalek specs: Give the index page a facelift
20:12 dalek specs:
20:12 dalek specs: - 1 fresh stylesheet, perl6.org-flavoured
20:12 dalek specs: - 1 heaped tablespoon of HTML5 semantic good stuff
20:12 dalek specs:
20:12 dalek specs: (Synopses are still using the original CSS for now)
20:12 dalek specs: review: https://github.com/perl6/specs/commit/aac6f724f6
20:13 flussence (now for the "get it to work in IE8" part...)
20:18 nbdsp joined #perl6
20:22 yqt joined #perl6
20:22 zakharyas joined #perl6
20:24 telex joined #perl6
20:26 timotimo why do we have "*" and "+" instead of a hoverable piece of text or even -gasp- a symbol?
20:27 moritz flussence++
20:28 moritz I also wonder if the DRAFT markers should be in a separate "Status" column
20:33 dalek specs: d04dd72 | (Anthony Parsons)++ | html/index.html:
20:33 dalek specs: Add HTML5 hack for IE <= 8 to display correctly
20:33 dalek specs: review: https://github.com/perl6/specs/commit/d04dd725d3
20:34 flussence I hope that works, it looks mostly okay for everything else on browsershots.org
20:36 timotimo https://developers.google.com/speed/pagespeed/in​sights/?utm_source=MFT&amp;utm_campaign=MFT&amp;​utm_medium=incoming-link&amp;priorityGroup=USABI​LITY&amp;url=http%3A%2F%2Fdesign.perl6.org%2F
20:36 timotimo :P
20:36 flussence timotimo++ # didn't know about that site, neat!
20:37 moritz http://domm.plix.at/perl/2015_05​_a_very_simple_perl6_script.html
20:42 moritz the error messages really shine
20:43 timotimo except the "please use :g", which could also mention that it has to go in front or inside
20:43 timotimo well, :g probably can't go inside
20:43 timotimo but the other adverbs can
20:45 moritz none of the match adverbs (:g, :x, :ov, :ex) can go inside
20:46 smls Neat, you can use 'when'/'default' directives directly in an 'if' block without wrapping them in a 'given':
20:46 smls if $foo -> $_ { when "a" {...}; when "b" {...}; default {...} }
20:46 timotimo aye
20:47 smls I already knew it forked with 'for' blocks; nice to know it's more general than that :P
20:47 smls *worked
20:49 timotimo works with any contextualizer
20:49 timotimo also does the "skipping" stuff
20:49 aindilis joined #perl6
20:50 smls m: {$_ = 5; when 5 {say "when"}; default {say "default"}}
20:50 camelia rakudo-moar c2a57e: OUTPUT«when␤»
20:50 smls it seems the definition of 'contextualizer' is very loose ;)
20:52 timotimo any block that sets a $_, or something like that
20:52 timotimo well, it gets interesting to see if and how it does the skipping thing
20:52 smls skips between all "sibling" when/default blocks I guess?
20:53 smls i.e. ones in the same parent block
20:53 timotimo it skips out of the block when one of the "when" blocks has not specifically "proceed"ed
20:53 timotimo m: {$_ = 5; 1.say; when 5 {say "when"}; 2.say; default {say "default"}; 3.say;}
20:53 camelia rakudo-moar c2a57e: OUTPUT«1␤when␤»
20:54 smls m: $_ = 5; when 5 { say "when" }; say "alive"
20:54 camelia rakudo-moar c2a57e: OUTPUT«when␤»
20:54 smls :P
20:55 timotimo ok, whatevs :P
21:04 bin_005_k joined #perl6
21:04 Ben_Goldberg joined #perl6
21:13 dalek specs: 5c624d1 | (Anthony Parsons)++ | html/style.css:
21:13 dalek specs: We don't *really* need an explicit font size
21:13 dalek specs: review: https://github.com/perl6/specs/commit/5c624d1d02
21:13 dalek specs: 60c6126 | (Anthony Parsons)++ | html/index.html:
21:13 dalek specs: Use HTML5 "hidden" attribute for «Read more» box
21:13 dalek specs:
21:13 dalek specs: Instead of loading all of jQuery to do it.
21:13 dalek specs: review: https://github.com/perl6/specs/commit/60c6126f06
21:13 dalek specs: 74dc6b0 | (Anthony Parsons)++ | html/ (2 files):
21:13 dalek specs: Make index slightly better on phones
21:13 dalek specs:
21:13 dalek specs: As suggested by Google's analysis tools, add a meta[name=viewport] tag and
21:13 dalek specs: adjust sizes so it doesn't render on phones using microscopic fonts and
21:13 dalek specs: horizontal scrollbars. The table still scrolls horizontally, but at least the
21:14 dalek specs: important bits fit on screen now.
21:14 dalek specs: review: https://github.com/perl6/specs/commit/74dc6b0016
21:14 dolmen joined #perl6
21:14 flussence that oughta do it for now
21:15 bin_005 joined #perl6
21:17 pierrot joined #perl6
21:19 smash joined #perl6
21:19 bin_005_k joined #perl6
21:20 Peter_R joined #perl6
21:22 amurf joined #perl6
21:24 dalek specs: 8bc51ea | (Anthony Parsons)++ | html/index.html:
21:24 dalek specs: Make "Read more" button work as intended
21:24 dalek specs: review: https://github.com/perl6/specs/commit/8bc51eaa2c
21:30 lizmat joined #perl6
21:32 lizmat japhb: $jnthn.marrage
21:32 lizmat argh
21:32 lizmat japhb: $jnthn.marriage ~~ blessed
21:33 timotimo yeah, throw out jquery
21:34 dalek specs: dbb3f62 | (Anthony Parsons)++ | html/style.css:
21:34 dalek specs: Need a tiny bit more padding to fit the image in
21:34 dalek specs: review: https://github.com/perl6/specs/commit/dbb3f629ef
21:34 flussence okay, *now* it's done :)
21:35 flussence .oO( time to do the same for the modules page? :)
21:40 RabidGravy joined #perl6
21:44 rararara joined #perl6
21:44 RabidGravy why would  e,g.  $foo.d be false and $foo.Str.IO.d be true where $foo is an IO::Path of a directory I just created?
21:45 RabidGravy very off
21:45 RabidGravy very odd
21:47 lizmat RabidGravy: that's indeed strange?  care to give an example?
21:53 lizmat good night, #perl6!
21:54 RabidGravy I'll split out in something concise I can test
22:08 dalek modules.perl6.org: bc179db | (Anthony Parsons)++ | web/ (2 files):
22:08 dalek modules.perl6.org: Move stylesheet into its own file
22:08 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/bc179db352
22:08 flussence ...I think I'll leave that one alone for now, it's harder to hack on than I thought.
22:09 andreoss joined #perl6
22:18 vendethiel joined #perl6
22:29 flussence oh crud, that didn't work how I thought it would...
22:29 Ben_Goldberg joined #perl6
22:30 dalek modules.perl6.org: 315c5b0 | (Anthony Parsons)++ | web/build-project-list.pl:
22:30 dalek modules.perl6.org: Put style.css in the output directory during build
22:30 dalek modules.perl6.org:
22:30 dalek modules.perl6.org: Oops.
22:30 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/315c5b0ba5
22:34 timotimo RabidGravy: we cache stat results, or something like that
22:37 timotimo you could maybe "strace" to figure out how often stat is being called?
23:09 dayangkun joined #perl6
23:13 amurf joined #perl6
23:14 Ben_Goldberg joined #perl6
23:16 Khisanth joined #perl6
23:20 vendethiel joined #perl6
23:34 BenGoldberg joined #perl6
23:40 SevenWolf joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo