»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:17 ssotka left 00:22 vytas left 00:28 vytas joined, wbill joined 00:37 pierrot left 00:43 ssotka joined 00:48 tardisx joined 00:49 tardisx left 00:54 kid511 left, kid51 joined, tardisx joined 00:55 KillBill joined
raydiak .u — 00:57
yoleaux U+2014 EM DASH [Pd] (—)
00:58 tardisx left 01:16 molaf left 01:18 stevieb left 01:20 Actualeyes joined 01:28 matiaslina left 01:29 molaf joined 01:39 yqt left, bjz left 01:55 sortiz joined 01:56 Khisanth left
sortiz star-m: class Foo { submethod BUILD { fail "noway" } } ; dd Foo.new; 01:59
camelia star-m 2016.04: OUTPUT«noway␤ in submethod BUILD at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
sortiz m: class Foo { submethod BUILD { fail "noway" } } ; dd Foo.new; # fail broken?
camelia rakudo-moar 5d4949: OUTPUT«Foo.new␤»
geekosaur m: class Foo { submethod BUILD { fail "noway" } } ; say Foo.new; # fail broken? 02:00
camelia rakudo-moar 5d4949: OUTPUT«Foo.new␤»
geekosaur hm
would have expected a delayed exception
sortiz That new, and broken some modules! :-(
02:01 Khisanth joined
sortiz star-m: class Foo { submethod BUILD { fail "noway" } } ; with Foo.new { say "good" } else { .fail } 02:02
camelia star-m 2016.04: OUTPUT«noway␤ in submethod BUILD at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
sortiz m: class Foo { submethod BUILD { fail "noway" } } ; with Foo.new { say "good" } else { .fail }
camelia rakudo-moar 5d4949: OUTPUT«good␤»
sortiz bisect: class Foo { submethod BUILD { fail "noway" } } ; Foo.new ~~ Failure; 02:06
bisectable sortiz: exit code on a “good” revision is 1 (which is bad), bisecting with inverted logic
sortiz: (2016-06-11) github.com/rakudo/rakudo/commit/d5e09d9
02:10 agwind joined
sortiz .tell lizmat Seems that d5e09d9 broke Failure handling at BUILD, see: irclog.perlgeek.de/perl6/2016-06-13#i_12653483 02:10
yoleaux sortiz: I'll pass your message to lizmat.
02:18 sigstuff_ joined 02:20 sigstuff_ left, kid51 left 02:22 huggable left, huggable joined 02:27 mr-foobar left 02:29 mr-foobar joined, bjz joined 02:30 noganex_ joined 02:33 noganex left 02:35 bjz left 02:39 mr-foobar left 02:40 mr-foobar joined 02:41 BenGoldberg joined
sortiz rakudobuged at RT#128393 02:44
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128393
02:55 wbill left 02:59 bjz joined 03:02 cpage left 03:03 cpage__ joined
agwind When writing an action for a grammar, does any one know the best way to grab the sub token's from a grouping? gist.github.com/agwind/055b50c5202...est-pl-L12 $/ is a nested structure and I'm drawing a blank. 03:04
03:04 cpage_ left, cpage__ is now known as cpage_ 03:10 torbjorn left
Zoffix agwind, it'd be $<number> and $<period> if you weren't repeating the captures... Ummm. 03:11
m: '123456' ~~ /($<number>=\d) ** 0..*/; .<number>.say for $/[0] 03:13
camelia rakudo-moar 5d4949: OUTPUT«「1」␤「2」␤「3」␤「4」␤「5」␤「6」␤»
Zoffix m: gist.github.com/zoffixznet/7ad25eb...d00c205230 03:22
camelia rakudo-moar 5d4949: OUTPUT«1,2,3,2␤»
Zoffix agwind, ^ no idea if that's the best way, but it works.
Zoffix goes to bed
03:22 torbjorn joined
Zoffix P.S.: ** 0..* can be replaced with just * 03:23
agwind I like it! ty. I used an attribute in my action.. but I just thought there might be something better.
gist.github.com/agwind/deaad9b98b7...fc2049cb79 03:24
BenGoldberg m: '123456' ~~ /($<number>=\d)*/; .<number>.say for $/[0]; 03:26
camelia rakudo-moar 5d4949: OUTPUT«「1」␤「2」␤「3」␤「4」␤「5」␤「6」␤»
timeless wow, you can ask camelia to run a pastebin? 03:32
yoleaux 11 Jun 2016 22:22Z <AlexDaniel> timeless: This leaves us with no open pull requests from you :) Time to submit new ones, perhaps? Thank you for your efforts!
03:33 skids joined
geekosaur only form gist, I think 03:34
03:39 ggoebel116 left 03:49 CQ2 joined 03:52 CQ left, CQ2 is now known as CQ 03:54 bjz_ joined 03:56 bjz left 04:01 agwind left 04:02 KillBill left 04:14 Cabanossi left 04:16 Cabanossi joined
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; %tokenmap.keys.join; 04:24
camelia ( no output )
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; say %tokenmap.keys.join;
camelia rakudo-moar 5d4949: OUTPUT«?%.-#␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message = $message ~~ s/($tokens)/%tokenmap{$1}-/:g;$message = $message ~~ s/-$//;return $message;} say english-for-tokens("%") 04:25
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤You can't adverb ␤at <tmp>:1␤------> 3message ~~ s/($tokens)/%tokenmap{$1}-/:g7⏏5;$message = $message ~~ s/-$//;return $m␤ expecting any of:␤ pair value␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message = $message ~~ s:g/($tokens)/%tokenmap{$1}-/;$message = $message ~~ s/-$//;return $message;} say english-for-tokens("%") 04:26
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5===␤Unrecognized regex metacharacter - (must be quoted to match literally)␤at <tmp>:1␤------> 3tokenmap{$1}-/;$message = $message ~~ s/7⏏5-$//;return $message;} say english-for-t␤Unrecognized regex metacharacter $ (must be quot…»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message = $message ~~ s:g/($tokens)/%tokenmap{$1}-/;$message = $message ~~ s/\-$//;return $message;} say english-for-tokens("%")
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Strange text after block (missing semicolon or comma?)␤at <tmp>:1␤------> 3 = $message ~~ s/\-$//;return $message;}7⏏5 say english-for-tokens("%")␤ expecting any of:␤ infix␤ …»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message = $message ~~ s:g/($tokens)/%tokenmap{$1}-/;$message = $message ~~ s/\-$//;return $message;};say english-for-tokens("%") 04:27
camelia rakudo-moar 5d4949: OUTPUT«Cannot assign to a readonly variable or a value␤ in sub english-for-tokens at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timeless um, that's unhelpful
m: sub x($z){ return $z}; say x(2)+3; 04:29
camelia rakudo-moar 5d4949: OUTPUT«5␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;return $tokens;};say english-for-tokens("%")
camelia rakudo-moar 5d4949: OUTPUT«?%.-#␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message = $message ~~ s:g/($tokens)/%tokenmap{$1}-/;return $message;};say english-for-tokens("%")
camelia rakudo-moar 5d4949: OUTPUT«Cannot assign to a readonly variable or a value␤ in sub english-for-tokens at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:g/($tokens)/%tokenmap{$1}-/;return $message;};say english-for-tokens("%") 04:30
camelia rakudo-moar 5d4949: OUTPUT«%␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:g/($tokens)/%tokenmap{$1}-/;$message ~~ s/\-$//;return $message;};say english-for-tokens("%")
camelia rakudo-moar 5d4949: OUTPUT«%␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:g/($tokens)/%tokenmap{$1}-/;$message ~~ s/\-$//;return $tokens~" "~$message;};say english-for-tokens("%")
camelia rakudo-moar 5d4949: OUTPUT«?%.-# %␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:gx/([$tokens])/%tokenmap{$1}-/;$message ~~ s/\-$//;return $message;};say english-for-tokens("%") 04:32
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Adverb gx not allowed on substitution␤at <tmp>:1␤------> 3sage ~~ s:gx/([$tokens])/%tokenmap{$1}-/7⏏5;$message ~~ s/\-$//;return $message;};s␤»
timeless m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:ge/([$tokens])/%tokenmap{$1}-/;$message ~~ s/\-$//;return $message;};say english-for-tokens("%") 04:33
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Adverb ge not allowed on substitution␤at <tmp>:1␤------> 3sage ~~ s:ge/([$tokens])/%tokenmap{$1}-/7⏏5;$message ~~ s/\-$//;return $message;};s␤»
04:35 vibha left 04:39 sortiz left, BenGoldberg left 04:41 BenGoldberg joined
BenGoldberg m: my %tokenmap = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;$message ~~ s:g:e/([$tokens])/%tokenmap{$1}-/;$message ~~ s/\-$//;return $message;};say english-for-tokens("%") 04:45
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Adverb e not allowed on substitution␤at <tmp>:1␤------> 3age ~~ s:g:e/([$tokens])/%tokenmap{$1}-/7⏏5;$message ~~ s/\-$//;return $message;};s␤»
04:47 skids left 04:53 khw left 04:55 sortiz joined 05:02 mr-foobar left 05:07 sno left 05:25 mls joined 05:37 setty1 left 05:56 wamba joined 06:00 grondilu joined 06:02 CIAvash joined, BenGoldberg left 06:13 sno joined 06:16 bjz joined, bjz_ left 06:22 huggable left, huggable joined 06:23 rurban_ joined 06:30 domidumont joined 06:34 salv0 left 06:35 domidumont left, domidumont joined 06:37 firstdayonthejob joined
go|dfish 06:42
06:43 firstdayonthejob left 06:45 domidumont left
nine dogbert17: but it's used by htmlify.p6 :) 06:47
06:48 kurahaupo left
moritz jnthn++ # 6guts.wordpress.com/2016/06/12/gri...rovements/ 06:56
07:01 abraxxa joined 07:02 RabidGravy joined 07:03 domidumont joined, domidumont left 07:07 cognominal joined 07:08 wtw joined 07:10 abraxxa left 07:11 abraxxa joined 07:13 darutoko joined 07:17 zakharyas joined 07:20 ufobat joined, wamba left 07:23 wamba joined 07:26 zakharyas left, zakharyas joined 07:32 zakharyas left, _mg_ joined 07:33 zakharyas joined 07:36 ssotka left 07:38 pdcawley joined 07:39 cpage_ left 07:41 cpage_ joined 07:43 _mg_ left 07:48 dakkar joined 07:51 g4 joined, g4 left, g4 joined 08:01 vibha joined 08:08 hanekomu joined 08:10 hanekomu_ joined, hanekomu_ left 08:13 araujo joined, AlexDaniel joined 08:14 araujo left
AlexDaniel timeless: you can also use bitbucket snippets and gitlab snippets 08:14
08:15 araujo joined 08:16 araujo left 08:17 araujo joined
dalek c: 1827324 | (Wenzel P. P. Peppmeyer)++ | doc/Language/classtut.pod:
doc the also declarator
08:17
08:18 araujo left 08:19 araujo joined 08:26 ocbtec joined
dalek c: b6d194d | (Wenzel P. P. Peppmeyer)++ | doc/Language/classtut.pod:
add also to the index
08:31
08:32 hanekomu left, hanekomu joined, AlexDaniel left 08:45 iH2O joined
lizmat . 08:46
yoleaux 02:10Z <sortiz> lizmat: Seems that d5e09d9 broke Failure handling at BUILD, see: irclog.perlgeek.de/perl6/2016-06-13#i_12653483
lizmat looking into it
sortiz lizmat, Seems that 'cus nqp::while don't sink, if BUILD (BUILDALL case 0) returns a Failure, it is only stored. 08:49
lizmat yeah... :-)
08:49 girafe left
lizmat spectesting a fix now 08:50
sortiz: could you write a test for it ? 08:51
sortiz sure. 08:53
lizmat sortiz++
sortiz moreover, while digging the issue, nqp::while surprised me: 08:56
m: use nqp; my int $i = 0; nqp::while(nqp::islt_i($i,10),($i = nqp::add_i($i,1))).say
camelia rakudo-moar 5d4949: OUTPUT«(1 2 3 4 5 6 7 8 9 10)␤»
lizmat hmmm... indeed surprising
sortiz So, don't sink 'cus store all values!
lizmat wonder what jnthn would want to say about that 08:57
sortiz nqp::while used in p6 produces a Seq!
08:59 rindolf joined
sortiz Yep, wonder that too. 08:59
psch fwiw, it makes sense to me
nqp::while is a statement level while, the Perl 6 equivalent is "do while ... { ... }" 09:00
m: my $i = 0 ;say do while $i < 10 { ++$i }
camelia rakudo-moar 5d4949: OUTPUT«(1 2 3 4 5 6 7 8 9 10)␤»
psch although its not documented either way in nqp/doc/ops.markdown vOv 09:01
09:02 zakharyas left
lizmat sortiz: fix pushed: github.com/rakudo/rakudo/commit/b351d61006 09:03
09:03 zakharyas joined
sortiz lizmat++ 09:04
iH2O is this valid perl6 --var++ 09:07
DrForr m: my $x; --$x++; $x 09:09
camelia rakudo-moar 5d4949: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Only identical operators may be list associative; since '--' and '++' differ, they are non-associative and you need to clarify with parentheses␤at <tmp>:1␤------> 3my $x; --$x++7⏏5; $x␤ expect…»
gfldex m: my $i = 0; say (--$i)++;
camelia rakudo-moar 5d4949: OUTPUT«Cannot resolve caller postfix:<++>(Int); none of these signatures match:␤ (Mu:D $a is rw)␤ (Mu:U $a is rw)␤ (Int:D $a is rw)␤ (int $a is rw)␤ (Bool:U $a is rw)␤ (Bool:D $a is rw)␤ (Num:D $a is rw)␤ (Num:U $a is rw)…»
DrForr You can also run "perl6 -e'my $x; --$x++'" at your command prompt :) 09:10
gfldex m: my $i = 0; say $i.&prefix:<-->.&postfix:<++>;
camelia rakudo-moar 5d4949: OUTPUT«Cannot resolve caller postfix:<++>(Int); none of these signatures match:␤ (Mu:D $a is rw)␤ (Mu:U $a is rw)␤ (Int:D $a is rw)␤ (int $a is rw)␤ (Bool:U $a is rw)␤ (Bool:D $a is rw)␤ (Num:D $a is rw)␤ (Num:U $a is rw)…»
psch m: use nqp; my $x; say nqp::iscont(--$x)
camelia rakudo-moar b351d6: OUTPUT«0␤»
psch m: say --5 09:11
camelia rakudo-moar b351d6: OUTPUT«Cannot resolve caller prefix:<-->(Int); none of these signatures match:␤ (Mu:D $a is rw)␤ (Mu:U $a is rw)␤ (Int:D $a is rw)␤ (int $a is rw)␤ (Bool $a is rw)␤ (Num:D $a is rw)␤ (Num:U $a is rw)␤ (num $a is rw)␤ in …»
psch can't dispatch to "is rw" without a container
gfldex m: my $i = 0; say $i.VAR.WHERE; say $i.&prefix:<-->.VAR.WHERE;
camelia rakudo-moar b351d6: OUTPUT«140031204203056␤140031204203016␤»
psch m: my $i = 0; say $i.VAR.^name; say $i.&prefix:<-->.VAR.^name;
camelia rakudo-moar b351d6: OUTPUT«Scalar␤Int␤»
gfldex my guess is that .&prefix:<--> returns the value not the container
LTA in my eyes because chaining -- and ++ doesn't make sense to start with 09:12
sortiz lizmat: use Test; class Foo { submethod BUILD { fail "noway" }}; dies-ok { Foo.new }, "dies"; 09:13
m: use Test; class Foo { submethod BUILD { fail "noway" }}; dies-ok { Foo.new }, "dies"; 09:14
camelia rakudo-moar b351d6: OUTPUT«ok 1 - dies␤»
lizmat ok 1 - dies
gregf_ i'm unable to understand why one would need such an operation ;)
DrForr It's also undefined behavior in C, incorrectly attributed to "sequence points", so I'm sure it'll get tried and pointed out. 09:15
gregf_ m: my $a = 10; (($a+=1)-=1);print $a; # no op like a !!$a
camelia rakudo-moar b351d6: OUTPUT«10»
psch m: my $a = 10; say !!$a # noop?
camelia rakudo-moar b351d6: OUTPUT«True␤»
gregf_ m: my $a = 10; say !!$a
camelia rakudo-moar b351d6: OUTPUT«True␤»
DrForr Hrm, circumfix operator with '--' and '++' instead of '[' and ']' :) 09:16
masak circumfix `--` and `++` would clash with prefix `--` and `++`
DrForr Well aware of that, hence the smiley :) 09:17
psch m: sub postcircumfix:<-- ++> { $^a ** $^b }; say 2--2++ # /o\
camelia rakudo-moar b351d6: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in postcircumfix:sym<-- ++>; couldn't find final $stopper ␤at <tmp>:1␤------> 3<-- ++> { $^a ** $^b }; say 2--2++ # /o\7⏏5<EOL>␤ expecting any of:␤ statemen…»
psch so the "closing" ++ gets parsed at postfix:<++> i guess
gregf_ i was assuming strings were immutable in Perl6? so a ++ would be changing the string? 09:18
psch i thought that should work thanks to LTM, but i've been wrong about lots of things involving LTM
DrForr And of course in less than a minute someoen tries it :)
gfldex gregf_: it will return a new string
psch DrForr: i thought you might mean postcircumfix, yeah. circumfix:<[ ]> doesn't exist after all
sortiz lizmat, OTOH, Would be nice if, in that case, Foo.new ~~ Failure, i.e. the Failure propagates till new return value. Otherwise, inside BUILD, fail and die are the same.
psch err 09:19
lizmat sortiz: but that would be a change in behaviour, wouldn't it ?
psch obviously it does exist, as Array.new >_>
gfldex gregf_: you may want to learn about containers 09:20
sortiz Yes, but right now not tested in roast :-)
gfldex gregf_: the following may be helpful perl6advent.wordpress.com/2015/12/...or-2-bind/ 09:21
sortiz As this issue show us.
gregf_ gfldex: was wondering what a container was, Thanks ;) 09:23
the problem is i only know of containers in JEE, so kinda difficult to relate 09:24
psch containers are kinda sorta a bit like auto-dereferencing pointers, maybe..? :S
gfldex gregf_: also doc.perl6.org/language/containers
gregf_ pointers *shudders* #i must say there is nothing on the planet like C/C++ 09:26
but how much do C/C++ guys get paid nowadays? i can see that even js/php/UI-UX guys get paid much more 09:27
gfldex: cheers
09:27 edehont joined
gregf_ *will read those during my lunch break* 09:29
moritz doesn't pay depend much more on location, industry and negotiation than on programming language? 09:32
CQ moritz: if someone is desperate for a cobol expert and there aren't many, then no : ) 09:34
stmuk_ lizmat: can you plug the london.pm tech meeting perl 6 talk in the weekly plz? details on the calender entry at london.pm.org
lizmat stmuk_ : you mean "• Nigel Hamilton: SixFix - a Weekly Dose of Perl 6 Powered by Perl 5" 09:36
?
stmuk_ yes!
lizmat okidoke 09:37
sortiz: github.com/rakudo/rakudo/commit/4890cdc5da 09:40
09:43 araujo left 10:08 user9 left 10:13 finanalyst joined 10:14 ggoebel116 joined 10:15 rurban_ left
finanalyst m: my @x=<a b c>;@x.map({say $_ x 3 }) 10:15
camelia rakudo-moar 4890cd: OUTPUT«aaa␤bbb␤ccc␤»
10:16 user9 joined
finanalyst odd. On perl6 REPL same code gives "aaa bbb ccc (True True True)" (with newlines) 10:17
moritz yes, because say returns True 10:18
so one thing is the output, the other return value
I though the REPL suppressed the return value printing if there was output
but that seem to have changed or been broken
m: my @x=<a b c>; say @x.map({say $_ x 3 }) 10:19
camelia rakudo-moar 4890cd: OUTPUT«aaa␤bbb␤ccc␤(True True True)␤»
timotimo maybe it lazily evaluates the list after it checked "was there output?" 10:20
moritz ah, right 10:21
that's plausible
10:22 huggable left, huggable joined, iH2O left 10:23 pmurias joined 10:29 araujo joined 10:31 edehont left, samb1 left 10:32 kaare_ joined 10:43 labster left, samb1 joined 10:44 sortiz left 10:48 cpage_ left, cpage_ joined
Xliff When you .resume from a CATCH, is that supposed to restart execution from the point where the exception is thrown? 11:06
timotimo no, just after it was thrown
Xliff Right. How do you limit that to the current scope?
11:07 finanalyst left
psch Xliff: do you have an example? 11:07
Xliff OK, leme think.
psch as in, i don't quite understand the question, because the CATCH always binds to the scope you're throwing from :)
oh, unless you mean "don't CATCH in lower scopes" 11:08
Xliff Yes
psch hmm
Xliff my $c = C.new; try { CATCH { $err = 1; .resume } $c.this_method_throws_exception(); my $can_we_resume_here = 1 } 11:09
The .resume will continue from this_method_throws_exception()
11:10 kid51 joined
psch m: class C { method thower { CX::Next.new.throw; say "not this!" } }; my $err; my $c = C.new; { CATCH { $err = 1; .resume }; try $c.this_method_throws_exception(); say "this instead" } 11:11
camelia rakudo-moar 0b824b: OUTPUT«this instead␤»
psch m: class C { method thower { CX::Next.new.throw; say "not this!" } }; my $err; my $c = C.new; { CATCH { $err = 1; .resume }; $c.this_method_throws_exception(); say "this instead" }
camelia rakudo-moar 0b824b: OUTPUT«This exception is not resumable␤ in block at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
psch ...i'm not sure that helps :S 11:12
fwiw, CX::Next was just grasping for something i think should be resumable
i had X::AdHoc first, that also gives "not resumable"
timotimo m: { say "test"; die "oh no!"; say "test 2"; CATCH { default { .resume } } } 11:13
camelia rakudo-moar 0b824b: OUTPUT«test␤test 2␤»
psch m: sub f { die "oh no!"; say "test 2" }; say "test 1"; f; say "test 3"; CATCH { default { .resume } } 11:14
camelia rakudo-moar 0b824b: OUTPUT«test 1␤test 2␤test 3␤»
psch that "test 2" is wanted to not show up i gather
m: sub f { die "oh no!"; say "test 2" }; say "test 1"; try f; say "test 3"; CATCH { default { .resume } }
camelia rakudo-moar 0b824b: OUTPUT«test 1␤test 3␤»
psch Xliff: yeah, seems my idea was good enough, but the example kinda opaque :)
teatime hmm... I wonder if Google Calendar can handle this recursion specifier, "the fourth full weekend of June, beginning at 1800 UTC Saturday and running through 2059 UTC Sunday, recurring anually." I doubt it ;) 11:15
Xliff A real example would need to be gist, but I just don't have the time right now. 11:16
OK. Let's try the default. 11:17
CQ teatime: use www.roaringpenguin.com/wiki/index.php/Remind and hook it into google cal : )
Xliff Nope. That didn't work. 11:20
OK. Let me gist this.
RabidGravy teatime, if you decide you have to implement it yourself I could actually use a scheduling recurrence library, I started looking then broke my brane on rfc 5545 and set it aside 11:22
11:22 brrt joined 11:25 zakharyas left
teatime hehe... it's just for one event, I can schedule it manually every year... if I remember to, anyway :) 11:25
I think the biggest deal breaker for most existing systems is the "…fourth //full// weekend of…" part. 11:26
11:28 cognominal left
Xliff gist.github.com/Xliff/bc0a12bcb213...culties-md 11:30
psch Xliff: did you see my example? 11:32
Xliff: any statement that might throw but you don't want to resume, you can prefix with 'try', and it will resume after that statement
Xliff Yeah. gist should take that into acccount
11:32 Emeric joined
Xliff OK. Let me do it that way. 11:33
psch irclog.perlgeek.de/perl6/2016-06-13#i_12655360 these two snippets illustrate what i mean, for clarity 11:34
Xliff Right. 11:35
psch: Gist updated. If my coding needs adjustment, please comment there. 11:40
But I think I got the gist (ha ha!) of your second example. I may not have implemented it quite right, though.
11:41 cognominal joined
Xliff psch: Also note that the testing code no longer executes properly. 11:42
psch Xliff: i now notice NC is involved? might well make this all a lot more complicated than i can deal with right now... :S
Xliff LOL 11:43
Why would NC matter?
psch i don't know, but it *does* make stuff complicated, in general :P
Xliff Gotcha.
psch like, your second issue there probably..?
Xliff Well. If you get an insight, please let me know.
Well, yes. The second issue is NC-hell. 11:44
I was going to ignore it for the time being and circle back.
psch yeah, if i get more cycles than "i'm waiting for r-j to compile right now" i might take a closer look
Xliff However the exception issue ... .isn't.
psch *free cycles
Xliff kk
Thanks for the time.
psch fwiw, what you're doing now in the lower part looks like it's 1:1 my example, in less minimalistic 11:45
but if that doesn't work there's clearly something else going on, and i don't have the spare brain to dig for that right now... :)
11:53 kid51 left
Xliff kk 11:55
I will revert code to what was working and circle back later.
11:55 TEttinger left 12:03 stevieb joined
pmurias psch: what's the state of the java interop in rakudo-j?, ruoso was asking about that recently 12:07
[Coke] .
psch pmurias: well, considering rakudo-j itself doesn't install correctly currently... 12:08
pmurias: aside from that, it's not much further along than it was before r-j stopped working
pmurias: which means, some marshalling bits are still glr-induced broken, and nothing much new has happened
pmurias: the rakudo branch jvm-interop is probably the workiest bit we have at the moment, but i haven't spent any time with that for a few months i'd say 12:09
pmurias psch: ok, getting it running and installing seems higher priority than the interop
12:18 Sgeo left 12:19 jack_rabbit left 12:23 cognominal left 12:24 brrt left 12:28 brrt joined
Xliff Is negated =:= spelled !=:=? 12:31
ilmari m: say IterationEnd !=:= IterationEnd
camelia rakudo-moar 7f0f46: OUTPUT«False␤»
ilmari m: say IterationEnd =!= IterationEnd
camelia rakudo-moar 7f0f46: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Prefix ! requires an argument, but no valid term found␤at <tmp>:1␤------> 3say IterationEnd =!7⏏5= IterationEnd␤ expecting any of:␤ prefix␤» 12:32
masak m: my $a = 42; my $b := $a; say $a =:= $b; say $a !=:= $b
camelia rakudo-moar 7f0f46: OUTPUT«True␤False␤»
masak Xliff: yes.
Xliff LOL! Thanks!
masak LOL! you're welcome! 12:33
moritz
.oO( shouldn't that be =!=? )
masak moritz: that's what ilmari tried above 12:34
Xliff Rut roh! 12:35
perlpilot why wouldn't it be !:! ? :) 12:42
12:43 avalenn left 12:44 bjz_ joined, avalenn joined 12:46 bjz left
moritz !=:!= 12:46
ilmari =!:!= 12:49
12:51 tbrowder joined
dalek c: 7fe076c | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod:
doc trait handles
12:53
12:55 tbrowder left 13:04 stevieb left
lizmat :w 13:07
oops
lizmat got distracted by the monsoon like rain 13:08
masak lizmat: oh, I wonder if it's the upper end of that we have here in Malmö
lizmat: here it's more like a welcome little drizzle, at least so far 13:09
lizmat apart from the temperature and the duration, it reminded me very much of my last visit to India 13:10
13:13 Ven_ joined
gfldex .oO( if you poke the climate in the eye, expect lots of tears ) 13:13
13:13 Actualeyes left 13:18 araujo left, zakharyas joined
dalek c: 6a175f4 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod:
move word around for consistency
13:20
c: 14462dc | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod:
add TODOs for some traits
13:26 Actualeyes joined 13:30 cosimo left 13:34 domidumont joined
lizmat hmmmm... LinkedIn :-( 13:36
13:37 sena_kun joined
lizmat blogs.microsoft.com/firehose/2016/0...2cb5lbvlms 13:37
timotimo :\
"install your fucking windows 10 if you want to continue using linkedin you asshole"
sena_kun hi, #perl6. 13:38
13:38 Zero_Dogg left
timotimo oh hi! 13:38
13:38 wamba left, Zero_Dogg joined
RabidGravy lizmat, I'm glad someone else is getting that weather :) 13:38
Juerd Considering deleting my linkedin account...
13:38 Zero_Dogg left, Zero_Dogg joined
Juerd Repeated password leaks never convinced me, but this just might. 13:39
13:39 g4 left
sena_kun Yet another attempt of module porting... In perl5 we have s{something}{something}. Error says that: "in Perl 6 please use assignment syntax". But I can't find any example of how to do this exactly. Any help? 13:40
timotimo oh, well it'd be s{something} = something_else
sena_kun Aw, so that's how it is. Thanks. 13:41
timotimo but s/foo/bar/ also works
sena_kun About s/// in Perl 6 I knew, but there is exactly 'block' I need. 13:42
13:43 stevieb joined
timotimo OK 13:43
13:45 ZoffixW joined
ZoffixW sena_kun, keep in mind that s{foo} = {bar} executes bar, since it's a block. 13:46
sena_kun, personally, I find .subst method more Perl 6-ish
.subst-mutate rather
sena_kun ZoffixW, can I compute dynamic replacer for every iteration? 13:47
ZoffixW sena_kun, sure
masak wait, we have a .subst-mutate?
why in the world do we have a .subst-mutate? :)
sena_kun ZoffixW, aw, it changes the deal. Going to look up .subst docs now... 13:48
timotimo no need to put { } on the RHS of the =
13:48 cosimo joined
timotimo it'll thunk the thing for you already 13:48
ZoffixW masak, because it's awesome
m: $_ = 'foo bar'; my %repl = foo => 'FOOS!', bar => 'BAARS!'; s:g{$<wat>=\w+} = %repl{$<wat>}; .say 13:49
camelia rakudo-moar ea53ce: OUTPUT«FOOS! BAARS!␤»
13:49 cdg joined
ZoffixW I'm still fuzzy on the whole block thing, but ... whatev. 13:49
masak ZoffixW: oh, that syntax is indeed awesome.
ZoffixW Juerd, I deleted mine recently and never looked back. All I ever got are annoying messages from recruiters.
masak ZoffixW: is it connected to .subst-mutate somehow?
ZoffixW No idea 13:50
masak ZoffixW: if I didn't just TIL about .subst-mutate and someone asked me, I would answer "no, we don't need a .subst-mutate, 'cus we have .=subst"
but I suspect that's not the whole story, since TimToady added .subst-mutate, and he knows the above
anyway, design erosion and method proliferation :( those are our weaknesses 13:51
ZoffixW m: 'foobar'.subst-mutate: 'oo', 'AA'
camelia rakudo-moar ffba1f: OUTPUT«Cannot resolve caller subst-mutate(Str: Str, Str); none of these signatures match:␤ (Cool:D $self is rw: $matcher, $replacement, *%named)␤ (Str:D $self is rw: $matcher, $replacement, :ii(:$samecase), :ss(:$samespace), :mm(:$samemark), *%options)…»
ZoffixW :S
masak m: 'foobar'.subst-mutate: /oo/, 'AA'
ZoffixW m: my $x = 'foobar'; $x .= subst: 'oo', 'AA'; say $x
camelia rakudo-moar ffba1f: OUTPUT«Cannot resolve caller subst-mutate(Str: Regex, Str); none of these signatures match:␤ (Cool:D $self is rw: $matcher, $replacement, *%named)␤ (Str:D $self is rw: $matcher, $replacement, :ii(:$samecase), :ss(:$samespace), :mm(:$samemark), *%options…»
rakudo-moar ffba1f: OUTPUT«fAAbar␤»
psch m: say ($="foobar").subst-mutate: 'oo', 'ou'
camelia rakudo-moar ffba1f: OUTPUT«「oo」␤» 13:52
psch ZoffixW: "is rw" :)
ZoffixW masak++ the .= didn't even occur to me
13:52 skids joined
ZoffixW Shorter to type too 13:52
masak aye
psch m: $_ = "foobar"; say S/oo/ou/; say s/oo/ou/
camelia rakudo-moar ffba1f: OUTPUT«foubar␤「oo」␤»
ZoffixW m: my $x = 'foobar'; $x.subst-mutate('oo', 'AA').uc.say;
camelia rakudo-moar ffba1f: OUTPUT«OO␤»
masak psch: yeah, probably has something to do with return values
psch that's the analog. .subst-mutate returns what s/// returns, .subst returns what S/// returns 13:53
masak psch: ah, whereas .=subst returns the resulting string, not the matches
psch masak: right .=subst works, precisely because .subst doesn't act as s/// does 13:54
well, maybe "returns what s/// returns" instead vOv
13:57 rurban_ joined
ZoffixW m: $_ = 'foobar'; .substr-rw(1, 2) = 'AA'; .say 14:00
camelia rakudo-moar ffba1f: OUTPUT«fAAbar␤»
ZoffixW m: 'foobar'.substr-eq("bar", 3).say
camelia rakudo-moar ffba1f: OUTPUT«True␤»
ZoffixW
.oO( <masak> anyway, design erosion and method proliferation :( those are our weaknesses )
14:02 tbrowder joined
tbrowder hi, gang 14:02
14:02 wamba joined
ZoffixW \o 14:02
tbrowder the gui diff prog i use is not handling unicode well at all, any recs for a good one? 14:03
masak kinda likes `git diff` 14:05
psch yeah, that or vimdiff for me
timotimo vimdiff is one of my favs 14:06
psch i suppose if it has to be a gui diff prog i'd suggest gvimdiff instead :)
14:06 ZoffixW left
tbrowder i'm presently using smartsynchronize by syntevo, but i think git diff depends on your shell and i need too look closer at it on my box for the files in question 14:07
emacs i think also has a diff view but i haven't looked at it yet 14:09
thanks for the suggestions 14:10
14:10 AlexDaniel joined
AlexDaniel given how often I get spammy emails from linkedin, I'm not surprised! It feels very appropriate 14:11
14:14 ptolemarch joined, khw joined 14:16 tbrowder left
[Coke] I use beyond compare, but can't say that diffing unicode comes up frequently. 14:16
(and only for big stuff. git diff works for a page or two)
RabidGravy I can't remember the last time I actually used anything other than git diff 14:19
14:20 robinsmidsrod left 14:21 robinsmidsrod joined 14:22 Ven_ left, huggable left, huggable joined 14:26 sena_kun left
hoelzro o/ #perl6 14:30
14:31 jameslen_ left
hoelzro anyone opposed to merging this? github.com/rakudo/rakudo/pull/738 14:31
or should I wait until 2015.06 is out?
14:36 AlexDaniel left
lizmat looks 14:36
moritz would it lead to getting significantly more testing? 14:37
if not, I'd say merge now
ss/would it/would waiting/
14:37 kurahaupo joined 14:38 edehont joined
lizmat I would favour merging now 14:38
Xliff WTF does this mean? "Missing or wrong version of dependency" 14:39
Was working. Made some changes, and now I get that from rakudo.
timotimo it used to be more common when we didn't throw out precompiled files that were outdated 14:41
Xliff Lemme remove .precomp -- sigh
nine Xliff: how old is your rakudo? 14:42
timotimo every precompiled file has hard coded in it other precomp files it relies on. but that isn't a problem any more with recent rakudos where we can re-precompile on the fly if the precompilation is bogus in some way 14:43
14:45 dfcarpenterak joined, dfcarpenterak left
Xliff nine: Fairly recent. One sec. 14:48
This is Rakudo version 2016.05-68-g2bd4211 built on MoarVM version 2016.05-17-g6075599
implementing Perl 6.c.
nine Xliff: yesterday evening two fixes went in that should fix exactly your problem 14:50
hoelzro ok, thanks for the input moritz & lizmat
I'll merge after work 14:51
(unless someone feels like merging on my behalf)
14:51 dalek left
hoelzro lizmat++ # thanks! 14:52
lizmat yw
14:52 dalek joined, ChanServ sets mode: +v dalek
diakopter well, no questions worth mentioning/answering on TimToady's Slashdot AMA 14:53
timotimo i didn't even know about the existence of that AMA
but i imagine it's full of troll questions 14:54
diakopter a bunch of troll questions, a few sincere but inane ones
lizmat hoelzro: so who should I credit for this work on the P6W ? 14:55
timotimo "do you like hats?"
DrForr I bothered to look. They're mostly "Have you stopped beating your wife?".
diakopter I mean, I suppose the one titled "How to think in Perl 6" is well-thought-out
14:55 robinsmidsrod left
Xliff nine++ Thanks! 14:56
14:56 robinsmidsrod joined
hoelzro lizmat: awwaiid (Brock Wilcox) did all the work, so he deserves the credit! 14:56
15:00 kurahaupo left 15:03 mr-foobar joined 15:05 domidumont left 15:15 brrt left 15:16 kurahaupo joined
sexy-coder-girl Where's Larry's Slashdot AMA? My google-foo is not strong enough. 15:21
BinGOs developers.slashdot.org/story/16/0...a-question
sexy-coder-girl Thanks.
15:22 djbkd left
[Coke] can't tell if slashdot's interface sucked or if larry really didn't answer anything. 15:26
diakopter I think it's still accepting questions
15:26 ssotka joined, araujo joined, araujo left, araujo joined
sexy-coder-girl I wouldn't blame him if he won't. Those questions remind me of preschoolers loudly yelling for attention. 15:28
diakopter considering the audienc of slashdot is thousands of times smaller than Reddit... the internet moved on from slashdot long ago
masak it was significant at one point, though 15:37
I remember it being my favorive source of tech news ~15 years ago 15:38
diakopter too
[Coke] honestly, my search after seeing AMA was on reddit. i had to come back here to realize it was slashdot. 15:39
diakopter yes I misspoke, on slashdot it's called "ask slashdot" 15:40
which apparently means "ask troll bots and people who can't tell the others are troll bots"
15:41 TimToady left 15:43 TimToady joined
geekosaur well, that's pretty much all that's left on slashdot 15:49
15:59 rindolf left
masak m: use Test; skip("hm"); done-testing 16:00
camelia rakudo-moar 6b9359: OUTPUT«ok 1 - \# SKIP hm␤1..1␤»
masak what's with the `\` ?
psch isn't that for TAP parsing..? 16:01
masak seems to be because github.com/rakudo/rakudo/blob/nom/...#L543-L544
but I confess I do not understand that comment
my feeling is that there are good reasons, but I don't feel I have the full picture 16:03
lizmat: around?
16:04 zakharyas left, vibha left 16:05 _mg_ joined
Xliff masak: I worked at Slashdot 15 years ago. :/ 16:06
My how times change.
stevieb what I remember from the old days of Slashdot: "Frosty Piss!" 16:07
Xliff Wow! And "Ask Slashdot" was my section, back in the day.
nemo slashdot blackholed me for like 6 years for upvoting a post on questionable slashdot moderation practicies
*practices
masak Xliff: wow. 16:08
Xliff suffers a moment of nostalgia... then it passes.
16:12 hankache joined
hankache hello #perl6 16:14
masak m: for <ello ankache> { say "h$_" } 16:15
camelia rakudo-moar 6b9359: OUTPUT«hello␤hankache␤»
hankache masak :) 16:16
16:18 baest_ joined 16:20 baest left 16:21 sherwool joined 16:22 edehont left
masak .oO( it's called "h reduce" ) 16:23
masak .oO( it's called "getting the h out of there" ) :P
reuse*
hankache ;) 16:29
16:31 dakkar left 16:40 _mg_ left
dalek osystem: 728bfd5 | (Zoffix Znet)++ | META.list:
Add Config::From

Load configuration from file to variables via traits
16:41
16:41 cognominal joined, _mg_ joined
sexy-coder-girl github.com/zoffixznet/perl6-Config-From 16:41
16:41 _mg_ left 16:43 hankache left 16:44 rindolf joined 16:51 wamba left 16:54 wamba joined 16:56 pullphinger joined 17:05 tbrowder joined
gregf_ m: say (<h> X~ <ello ankache>).map: { my $y = $_; $y ~~ s:g/$<word>=\w+/{$<word>.tc}/;$y } 17:08
camelia rakudo-moar 6b9359: OUTPUT«(Hello Hankache)␤»
tbrowder ref gui diff progs: this works (Linux only tested) and handles my unicode tests okay: <sourceforge.net/projects/diffuse/> 17:09
17:10 setty1 joined 17:11 firstdayonthejob joined, sno left
DrForr I've used Meld frequently, but I've not extensively (okay, at all) tested Unicode support beyond the occasional \,t and \"o. 17:11
17:12 spider-mario joined 17:15 cognominal left
ilmari ţ is outside latin-1, so handling that is a good start 17:15
ł was the fun one when $ork[-1] launched a polish site 17:16
DrForr That was my first useful test case (thanks pmurias) for blogs.perl.org mk II. 17:17
[Coke] m: say "ţ" ~~ m:ignoremark/t/;
camelia rakudo-moar 6b9359: OUTPUT«「ţ」␤»
[Coke] oh, yay, that did get done.
stevieb just did a fresh MoarVM install on a new box, and configured it as a doc build system per CONTRIBUTING, then cloned/updated my fork but am getting an error on `perl6 htmlify.p6`: gist.github.com/stevieb9/ce8a940ee...b0d69f8e51 Could someone please take a look and let me know if I've done something wrong? I'm very new to perl6, so my troubleshooting skills are severely lacking as to where to look
ilmari m: say 'ł' ~~ m:ignoremark/l/ 17:18
camelia rakudo-moar 6b9359: OUTPUT«False␤»
ilmari m: say 'ł'.NFD
camelia rakudo-moar 6b9359: OUTPUT«NFD:0x<0142>␤»
ilmari m: say 'ţ'.NFD
camelia rakudo-moar 6b9359: OUTPUT«NFD:0x<0074 0327>␤»
ilmari m: say 'å' ~~ m:ignoremark/a/ 17:19
camelia rakudo-moar 6b9359: OUTPUT«「å」␤»
[Coke] stevieb: what does perl6 --version say? 17:20
dalek c: 49e72a2 | (Jan-Olof Hendig)++ | / (2 files):
Added docs for X::IO::Link
stevieb Coke: This is Rakudo version 2016.05-138-g6b93592 built on MoarVM version 2016.05-17-g6075599 17:21
implementing Perl 6.c.
tbrowder Zoffix, hi, please look at PR <github.com/perl6/doc/pull/592> 17:22
[Coke] and git rev-parse --short HEAD # in the perl6 doc build dir? 17:23
stevieb Coke: 0937200
[Coke] ok, that's fairly recent. 17:24
I just tried at 49e72a2 and did not get that error. 17:25
trying at 093... 17:26
sexy-coder-girl tbrowder: looks good. Thanks.
dogbert17 [Coke]: moritz made a fix to htmlify.p6 yesterday which fixed that problem 17:28
sexy-coder-girl stevieb: [Coke] that was found yesterday. I think FROGGZ++ was looking into it, but I'm unsure
[Coke] stevieb - there you go. a git pull should fix you, then.
stevieb awesome, thanks all!
[Coke] though I still did not get that failure at the revision you're getting it at. 17:29
sexy-coder-girl The failure was introduced in this commit: github.com/rakudo/rakudo/commit/49...9f132b2611 17:30
stevieb sorry for the noise... I had forgot to merge upstream/master. Looking much better now :) 17:32
geekosaur wasn't that .. .yes
introduced and fixed late yesterday
17:32 cognominal joined
dalek c: d080e93 | (Tom Browder)++ | doc/Language/testing.pod:
tidy grammar and punctuation; use consistent var names
17:33
c: 44370b0 | (Zoffix Znet)++ | doc/Language/testing.pod:
Merge pull request #592 from tbrowder/testing-doc

tidy grammar and punctuation; use consistent var names
17:40 domidumont joined 17:43 abraham_ joined 17:45 abraham_ left 17:46 Zero_Dogg left 17:47 Zero_Dogg joined 17:48 domidumont left 17:49 domidumont joined 17:56 Sound joined 18:01 Emeric left, Emeric joined
literal contributed a Perl 6 implementation to Make a Lisp: github.com/kanaka/mal/tree/master/perl6 18:03
18:03 dustinm` joined
[Coke] so, I know we have put this in and removed it in the past, but what are other people using to search once on the p6 doc site? If I'm scrolled 30 pages down, I have to scroll back up to get the input box to search. 18:04
18:05 girafe joined
sexy-coder-girl press "Home" 18:05
And when page loads, I just hit TAB twice to get to the search box
18:07 rurban_ left
nemo [Coke]: I found the site search rather disappointing so I stick to site: in my browser search ☺ 18:10
[Coke] nemo: that is not what i'm trying to do 18:11
is your browser search searching the whole site?
18:12 rurban_ joined
[Coke] (I mean, if it is, then yes, that is what I'm trying to do) 18:12
18:12 AlexDaniel joined
AlexDaniel yeah, 「Home」 18:13
as simple as that
[Coke] yah, that's a 3-key combo on this keyboard. and then still have to click in the search box. 18:14
AlexDaniel ah okay, I see a problem
there's no accesskey for the search box
sexy-coder-girl F5, TAB, TAB :P
18:15 nightfrog is now known as culb
[Coke] AlexDaniel: yes. I don't know if there's a way to do that is standard, but when we did it by hand, we had to back it out, because someone's browser was already using '/' for something. 18:15
AlexDaniel no-no, accesskeys work on all configurations 18:16
[Coke] ah, yes, we hardcoded it. (so it worked as it does, e.g. on facebook) 18:17
dalek c: d56de70 | (Aleks-Daniel Jakimenko-Aleksejev)++ | template/header.html:
Add accesskey for the search input
[Coke] why "f"
? 18:18
AlexDaniel because other websites use that
e.g. wikipedia
sexy-coder-girl Doesn't work on Chrome
AlexDaniel the reason is probably because ‘s’ is typically used for “Save”
that is true for wikis and forums
sexy-coder-girl Just opens the File menu
AlexDaniel sexy-coder-girl: not yet!
sexy-coder-girl AlexDaniel: not yet what? 18:19
AlexDaniel sexy-coder-girl: you mean, it does not work on wikipedia?
sexy-coder-girl: what are you pressing exactly?
[Coke] I had to use control-option-f testing just now on wikipedia (mac/chrome)
sexy-coder-girl Chrome's on Windows uses Alt+Key for access keys and that's an alt sequence
Unless Mozilla's dev docs are wrong and it's not Alt+Key
AlexDaniel alt+shift+key perhaps?
sexy-coder-girl Yeah, that works. 18:20
AlexDaniel anyway, you may say that accesskeys have poor default modifiers, but there's no reason not to have one
for those who have no troubles using accesskeys
[Coke] (the doc build I set to run to test earlier: stillllll running)
sexy-coder-girl Sure.
sexy-coder-girl was using developer.mozilla.org/en-US/docs/W.../accesskey 18:21
AlexDaniel personally I don't mind alt+shift because I use thumbs to hit both of these keys…
sexy-coder-girl tries to imagine what that looks like 18:22
AlexDaniel
.oO( YAY for japanese keyboards and custom keyboard layouts! )
18:22 huggable left, huggable joined
sexy-coder-girl Speaking of keyboards. I just got my second Corsair K95 RGB Perl 6 keyboard delivered at $work <3 Unicode all the way! 18:22
18:23 FROGGS joined
AlexDaniel does not take seriously any keyboard that does not have proper thumb keys 18:23
has no*
18:23 vendethiel joined
psch yeah, alt and mod4 are the worst on normal layouts :/ 18:24
FROGGS o/
AlexDaniel psch: the reason why I love Japan so much is because its existance is the reason why you can get a keyboard with short spacebar (and some extra keys for your thumbs) for almost *any* laptop! :) 18:27
nemo [Coke]: yes. thus the site: ☺
lizmat masak: am around now 18:28
nemo [Coke]: www.google.com/search?q=site%3Adoc....org+panda
panda is my default search test phrase
by contrast, the bar in the upper right of the site says "Sorry, no results found" which is what it does way way way too often
search.yahoo.com/yhs/search?p=site....org+panda works fine too 18:29
[Coke] nemo: as a commiter on the site, I'd rather use it and fix it. 18:30
AlexDaniel maybe instead of “sorry, no results were found” we should have a link that says something like “You can try duckduckgo to find it on this website”
[Coke] I'm glad you have something that is working for you, though.
dalek c: a97d94d | (Zoffix Znet)++ | doc/Language/modules.pod:
Add panda to search index
18:31
AlexDaniel alright that was quick
dalek c: c4ac6d4 | (Steve Bertrand)++ | CONTRIBUTING.md:
added blurb about python-dev in CONTRIBUTING
c: a43f339 | (Zoffix Znet)++ | CONTRIBUTING.md:
Merge pull request #593 from stevieb9/python-dev_CONTRIBUTING

added blurb about python-dev in CONTRIBUTING
18:32 sno joined
nemo heh. there was nothing particularly special about pandas you realise ☺ 18:32
18:32 Amnez777 left, Amnez777 joined
nemo apart from them being super cute when I test some search engine feature 18:32
AlexDaniel it was a super cute demonstration of live coding :D 18:33
nemo www.google.com/search?site=&tb...p;q=pandas case in point
kk 😃
dalek line-Perl5: 112f4b9 | niner++ | lib/Inline/Perl5.pm6:
Fix "Method 'DESTROY' not found" errors on Perl 5.24

Recent Perl 5 versions fixed a regression about AUTOLOAD not being called for a missing DESTROY: rt.perl.org//Public/Bug/Display.html?id=124387
Seems like Inline::Perl5 depended on this bug. With Perl 5.24 AUTOLOAD is now called and Perl 6 complains about not finding DESTROY. Worse, it even runs into an endless recursion ending in a segfault.
Fix by providing an explicit DESTROY method on Perl6::Object.
Fixes GH #63
Thanks to jdv++ for reporting!
nemo uh
18:33 sivoais joined
nemo why was he not rate limited? 18:34
sexy-coder-girl AlexDaniel: doesn't mean your point isn't valid. A link to use a regular search engine is not too bad, as "nothing found" can often be interpreted as "it's not there" and not "we have a shitty search feature"
nemo sexy-coder-girl: yeh. that's how I interpreted it at first until I switched to interpretation #2
18:37 labster joined
AlexDaniel ok, here's the place github.com/perl6/doc/blob/8ff5d14f...ate.js#L38 18:37
now what
18:37 CQ left 18:38 yqt joined, yqt left, yqt joined
sexy-coder-girl AlexDaniel: currently, the "not found" is made with pure CSS. That needs to be tossed and an element added on the search box that's hidden by default, but is shown when .not-found class is is present on #search 18:38
And its link would need to be modified in that JS code you linked to 18:39
AlexDaniel sexy-coder-girl: sure. Are you willing to work on it?
that'd be cool
sexy-coder-girl Yeah, in about 4 hours.
tbrowder sexy-coder-girl: another testing.pod PR 594 18:40
sexy-coder-girl tbrowder: just commit, no need to ask :) Feel free to change anything in that doc that you think should be changed.
tbrowder s-c-g: WILCO, thanks. 18:41
AlexDaniel tbrowder: I'd love to see such obvious changes to be committed directly, without a pull request 18:42
sure it works either way, but come on :)
[Coke] PRs are cheap 18:43
and I've broken enough stuff that "should never break anything".
... that said, that PR does look very inoffensive. :) 18:44
tbrowder Zoffix (or alias): I'm bothered by the new is-approx test with both rel and abs tols: it seems to me that the test ought to pass when either test passes, or at least make a version to give that option 18:47
18:48 cognominal left 18:56 Sound left
dalek c: d96c938 | (Tom Browder)++ | doc/Language/testing.pod:
another edit tweak
18:57
c: 14067be | (Tom Browder)++ | doc/Language/testing.pod:
Merge pull request #594 from tbrowder/tdoc

another edit tweak
lizmat tbrowder: looking forward to seeing you at www.yapcna.org/yn2016/wiki?node=PRC 18:59
18:59 CIAvash left
nemo sexy-coder-girl: BTW Zoffix... what's w/ this account anyway? Fighting FOSS stereotypes or something? 18:59
tbrowder lizmat: ditto!
sexy-coder-girl tbrowder: What's the usecase for that? Just curious 19:00
19:01 cognominal joined
nemo s/account/nick/ 19:01
sexy-coder-girl nemo: well, I registered it 7 years ago to troll people in #css and I needed a non-Zoffix nick so I picked it, but sure, your reason why sounds better :P
nemo heh 19:02
I have a vague recollection of it appearing in #css yeah
but not why
sexy-coder-girl I think the real reason was everyone was directing questions to me instead of the channel and I got fed up with it.
So I changed it so people wouldn't recognize a Zoffix :P 19:03
19:03 ssotka left
nine lizmat: looking at travis-ci.org/niner/Inline-Perl5/b.../137321222 it seems like it's impossible to provide an Inline::Perl5 that works with rakudos before and after your removal of BUILDALL's positional :( 19:06
lizmat hmmm... perhaps if add another candidate *with* positionals ? 19:07
19:07 labster left, darutoko left
Xliff <sexy-coder-girl> Speaking of keyboards. I just got my second Corsair K95 RGB Perl 6 keyboard delivered at $work <3 Unicode all the way! 19:08
What is this mysterious keyboard referred to above? 19:09
tbrowder s-c-g: hm, I was going to ask the same question of you but thought that might be a bit impertinent, but I think I came upon one in cpan's Geo::Ellpsoid whicj I'm trying to port to Perl 6; I'll try to get details and report back
Xliff is discovering that porting XS to Nativecall is harder than he thought. 19:10
Particularly the pointer bits. 19:11
sexy-coder-girl Xliff: it got 18 extra keys that you can map to Perl 6 Unicode ops. And there are 3 modes to, so you get 18*3 keys to map: www.corsair.com/en-ca/corsair-gamin...rry-mx-red
It also got shiny lights
Xliff Ooo!
Xliff was drooling until he saw the price tag.... 19:13
Which should have been expected, really.
$300 for a keyboard.... *faints*
sexy-coder-girl tbrowder: all technical questions are pertinent :) The old implementation let you use rel and abs but always together. My new version lets you use them separately. I've used "both pass" instead of "either pass" because I figured it's just a combination of two tests. 19:14
19:14 ufobat left
sexy-coder-girl Xliff: but it got lights!! 19:15
:)
geekosaur looked at it, but first checked os support and dropped it without following further
FROGGS Xliff: there is also this: www.getdigital.co.uk/Optimus-Maxim...1465845276 19:16
sexy-coder-girl geekosaur: but it got shiny lights! :P 19:18
geekosaur I'd probably want to turn them off anyway
sexy-coder-girl What's funny is the hand-rolled driver lets you configure all sorts of lights but assigning custom values to the G keys is NIY :) Priorities!
19:18 hankache joined
geekosaur is recovering from ~ 3 weeks of his brain refusing to filter. blinkenlights were verboten that whole time... 19:19
"oh look, it can configure light shows. is there an off switch for them?"
but, no evidence it does anything at all outside of bios mode on linux, so. 19:20
stevieb I know it's pricey honey, but I can start Microsoft Word with a single keyboard key click!
19:21 labster joined
sexy-coder-girl The one thing I notice right away about this keyboard is it's about 5 times louder than my old non-mechanical one. 19:24
DrForr How would I determine if a constant exists? Other than eval'ing, that is... 19:25
FROGGS DrForr: say so ::('FOO')
DrForr Thanks.
FROGGS or better: say ::('FOO') !~~ Failure
Xliff sexy-coder-girl: I drool over the blinkenlights. I really do. However I have better things to do with $300. 19:29
My disposable cash reserves are low.
Xliff looks at the link from FROGGS and *double faints*
stevieb m: say ::('FOO') !~~ Failure 19:30
camelia rakudo-moar f7c8dc: OUTPUT«False␤»
Xliff ( ಠ益ಠ) 19:31
19:31 brrt joined
Xliff FROGGS: Who da f*%^# cares about OLEDs on each key? 19:31
(which is actually kind kewl... but not at that price!)
( ಠ益ಠ)
sexy-coder-girl Xliff: my keyboard got that too, I think
Xliff $300 is beginning to sound reasonable!!! 19:32
19:32 domidumont left
sexy-coder-girl Each key can be one of the 16 mil colors and you can configure them individually or make them wave and stuff 19:32
Xliff And I can't just believe I said that.
sexy-coder-girl But the reason I like this keyboard is thanks to its extra keys, not lights :)
Xliff sexy-coder-girl: Oh no. FROGGS' keyboard has OLED displays for each key glyph. And you can assign them. 19:33
Yes, extra keys good. Especially if they are extra-clacky.
19:33 AlexDaniel left
Xliff However my current keyboard thrives in a hostile environment. Both of those keyboards would turn their noses up and flee! 19:33
sexy-coder-girl looks at FROGGS's link 19:34
:o $2574 Canukistan dollars!
19:36 hankache left
FROGGS sadly I dont own such a thing :o) 19:36
19:37 jjido_ joined
sexy-coder-girl I don't think I'd want to :) Well, unless I were making enough money to feel the same about spending $2500 on a keyboard as I do about spending $300 right now :P 19:37
stevieb $2500 CDN... that's only like what, $25 USD at our CDN exchange rate ;) 19:39
konobi optimus?
stevieb at least that's what it's like when I buy USD before a trip down south lol
sexy-coder-girl :) 19:40
nemo Xliff: I do a surprising amount of actual coding these days on a $20 BT folding kbd
my phone has replaced my laptop
sexy-coder-girl 0.o
nemo www.amazon.com/dp/B018P6BSPQ one of these 19:41
stevieb I hate using my phone for stuff. Rarely do I use it for anything but phone, text and some rare web browsing when I really need to
nemo run a linux chroot on my note 4 - use the stylus as a mouse
(the stylus button does right click \o/ ) 19:42
well... not that particular vendor - that kbd is on like a couple of dozen vendors on amazon
I paid like $22 to be precise
price seems to have bumped up a bit. maybe was popular?
has a little magnetic stand/keyboard case that some sell it with 19:43
www.amazon.com/dp/B019PIXO78 this one is selling it w/ the stand. again way too expensive tho 19:44
Xliff nemo: Heh! Good on you, mate! 19:47
I have one of those, however I they never felt quite right cos.... FOLD!
The keys adjacent to the fold just weren't.... right!
Of course, that is more me than the keyboard. PEBCAK!! 19:48
19:48 jjido_ left 19:50 jjido_ joined
dalek line-Perl5: 628a066 | niner++ | lib/Inline/Perl5.pm6:
Revert "Mu.BUILDALL doesn't take positionals anymore"

This reverts commit d854a8384cf96a8c871e1ad93367dce4152acb8c.
19:50
line-Perl5: e4add10 | niner++ | / (3 files):
Make @*ARGS available as @ARGS in Perl 5

Fixes GH #56
line-Perl5: 8b1a59b | niner++ | META.info:
Version 0.7

Changes:
  * Fix "Method 'DESTROY' not found" errors on Perl 5.24
  * Make @*ARGS available as @ARGS in Perl 5
  * Update perlbrew instructions to build latest stable perl
nemo Xliff: so... the X key was "mushy" on mine, and I finally popped it open and figured out some jackass had installed a ribbon cable w/ a crimp in it, and in the wrong direction
Xliff nemo: Hmm! Thanks! I may just have to do that... if I can find it. 19:52
s/do/check/
nemo Xliff: doing a half fold/unfold usually fixes that
19:52 sherwool left
nemo I use mine tons tho 19:52
basically any time I'm anywhere and have an urge to have a computer
sooo much better than chatting w/ those awful virtual kbds
Xliff I had a 10.1" tablet to use that keyboard with. I fiddled and dabbled, but found out that the best way to have an SSH session was to wait until I got to a real computer. 19:53
nemo haha
Xliff Of course, this also could have been due to the quality of the SSH apps at the time.
nemo Xliff: well. connectbot is crap w/ this kbd I've discovered
Xliff Ooh..... connectbot. I thought I deconnected that thing.
nemo also I had to disable the irritating kbd profile that was trying to give me french accents
I use connectbot when I'm in a hurry, but usually use a linux chroot and XSDL for a nice responsive session 19:54
dalek line-Perl5: c0fd19a | niner++ | / (2 files):
Remove bogus build dependency on panda

Fixes GH #61
nemo Xliff: apart from that works great
Xliff nemo: at any rate, I am glad you are having some success. I may encourage me to try to set things up again.
nemo
Xliff So, FROGGS... here is part of the issues I'm currently having gist.github.com/Xliff/bc0a12bcb213...4fde829193 19:56
Problem #2 is what I wanted to talk to you about.
19:59 abruanese left
Xliff FROGGS: gist.github.com/Xliff/fa743c54a516...a4d4c46390 20:04
20:06 kaare_ left
FROGGS hmmm 20:06
20:06 dha joined
Xliff Of course, I foolishly thought to start with t/06elements.t as my starting point! =) 20:07
FROGGS I'm not sure if the last test in the second gist actually should pass...
Xliff It's been interesting. And I have a long way to go.
FROGGS but if the P5 version does that magic, we maybe should too... 20:08
Xliff Right. If I do another "$attr = getAttributeNode($attname1)" before that test... it does. Which is expected.
And yes, the test applies that it works through P5. BUT! P5 had XS.
And I'm wondering if, in some of these operations, there is an implicit copy done. 20:09
stevieb m: (1..5).put.say
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2016/06/13/...ng-to-tpc/
camelia rakudo-moar ed510c: OUTPUT«1 2 3 4 5␤True␤»
stevieb why does that print True? is it literally saying "yes, it printed"?
FROGGS Xliff: unlikely 20:10
Xliff kk
That's something to cross of the list, then.
20:10 jjido_ left
FROGGS Xliff: though kina 20:10
kinda*
Xliff heh! =) 20:11
FROGGS the $attr object needs updating (nqp::refresh or so)
commt the first test of $attr.value, I bet then the tests will pass 20:12
Xliff FROGGS: Can you add the details as a comment to the gist and I will try that tonight. 20:16
I have not yet delved into nqp.
FROGGS will do in a bit
Xliff No worries. Whenever you have time is OK. I appreciate it! =)
20:17 ZoffixMobile joined
ZoffixMobile lizmat++ good weekly and awesome speed improvements! 20:18
masak hey, channel: is it reasonable to expect `END` phasers to run on a ^C ?
psch wouldn't think so
FROGGS lizmat++ 20:19
[Coke] m: 1.say.say
camelia rakudo-moar ed510c: OUTPUT«1␤True␤»
[Coke] ^^ same thing; .say returns True.
(if it worked)
as does put 20:20
lizmat masak: only if you activate a Supply on the signal
psch irclog.perlgeek.de/perl6/2016-04-08#i_12308594
i knew we had had that before :)
ZoffixMobile stevieb, IIRC in Perl 5 it returns false if print failed, so returning True is in the spirit of that
m: $*OUT.close; ^5 .put.say 20:21
camelia rakudo-moar ed510c: OUTPUT«write string requires an object with REPR MVMOSHandle␤ in block <unit> at <tmp> line 1␤␤»
stevieb thanks for the clarifications. Makes sense
psch m: $*ERR.close; say note "foo"
camelia ( no output )
ZoffixMobile it dies
psch yes, and LTA at that
ZoffixMobile nods 20:22
masak lizmat: oh, that's nice -- any prior art I can look at? 20:23
lizmat masak: did you see the link that psch posted just now ?
jdv79 i also thought that sub and -> were equivalent ways to create an anon sub. 20:24
lizmat jdv79: a sub can have a name, a Block cannot
masak looks
lizmat m: signal(SIGTERM).act: { exit 1 }; Nil while 1; END say "this is the end, my friend" # masak
jdv79 m: my $foo = sub ($a,$b) {say $a-$b};$foo(1,2);
m: my $foo = -> ($a,$b) {say $a-$b};$foo(1,2); 20:25
camelia rakudo-moar ed510c: OUTPUT«(timeout)this is the end, my friend␤»
rakudo-moar ed510c: OUTPUT«-1␤»
rakudo-moar ed510c: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤ in block <unit> at <tmp> line 1␤␤»
jdv79 guess the bot is dead?
oh
dalek c: 9a4bc10 | coke++ | STYLEGUIDE.md:
add a note about using p5 to explain things.
lizmat jdv79: it was timing out with my signal tap example
psch m: my $foo = -> ($a,$b) {say $a-$b};$foo([1,2]);
camelia rakudo-moar ed510c: OUTPUT«-1␤»
psch jdv79: if you specify destructuring in your Block Signature you have to give something that can be destructured 20:26
jdv79 oh, the parens on the ->. thanks.
ZoffixMobile jdv79, you can also return from a sub but not block
jdv79 hasn't p6'ed in a while:(
20:27 buharin joined
buharin hi 20:27
I get perl one liners book
do you think it is good practical source
to learn perl?
Xliff BBL, #perl6 -- hopefully with SOUND!
ZoffixMobile buharin, it is but Peteris Krumins?
Xliff /quit
20:27 Xliff left, acrussell joined
ZoffixMobile *by 20:28
buharin yep
Peteris
[Coke] buharin: never read it, but it looks more like a cookbook than a learning guide.
ZoffixMobile buharin, sure, he's a good dude. But it's a Perl 5 book :)
buharin oh ok 20:29
;d
so maybe just take only recipes which I really need
ZoffixMobile whoa! nine++ # @ARGV in Inline::Perl5 20:30
20:31 buharin left 20:32 frobisher joined, dha left, frobisher is now known as dha
tbrowder Zoffix: I remove my concern...can't find a good example, but discussion has helped me understand the Geo::Ellipsoid tests...thanks! 20:34
vendethiel Damn, slashdot is terrible 20:35
ZoffixMobile cool
vendethiel they're not even criticizing perl(6). they're just bad questions
20:35 ZoffixMobile left
dha People still read slashdot? 20:37
Juerd dha: One guy in our local hackerspace does.
tadzik I occasionally do
Juerd dha: He keeps complaining about it, but apparently can't kick the addiction :)
dha Should I actually mention it as a resource in my YAPC talk? Or are these people outliers? 20:38
I'm now also wondering if people still use PerlMonks.
perlpilot I looked it today to see the drivel that people were asking Larry Wall
20:39 jjido_ joined
dha Yeah, I don't think I want to point people at drivel. :-) 20:39
20:39 jjido_ left
masak wow, to-json and from-json have moved out of core? 20:40
FROGGS O.o
masak didn't notice when that happened
can't say I disapprove, though
moritz they were never meant to be in there
nine last week or so
masak moritz: there is that 20:41
FROGGS was still handy though
nine ok, more like a month ago
masak goes with a custom simple format for his data file instead
moritz Date: Sun May 8 17:34:16 2016 +0200
move {from,to}-json to Rakudo::Internals::JSON
... and add deprecations for the old locations.
nine boy how time flies
masak like an arrow
boy how fruit flies like a banana
:P
FROGGS *g* 20:42
gnight #perl6 20:43
20:43 FROGGS left 20:45 brrt left
masak 'night, FR<tab><tab><tab> 20:45
20:47 vendethiel left 20:58 tbrowder_ joined 21:02 tbrowder_ left 21:08 kid51 joined, wamba left 21:11 skids left 21:12 Amnez777 left
Zoffix htmlify.p6 needs a "I'm just trying to make a few tweaks to the website" build mode :/ 21:15
dalek c: 8935ae1 | (Zoffix Znet)++ | CONTRIBUTING.md:
Add graphviz install command in build docs
21:19
21:19 agentzh joined
dalek c: f3eec8c | (Zoffix Znet)++ | CONTRIBUTING.md:
Provide faster and less quiet way to install Mojolicious

The official website install instructions omit tests and they're generally lengthy and thorough and not really needed.
21:20
21:21 SHODAN left
timeless Zoffix: yes 21:24
i'm trying to get make to run faster
Zoffix timeless++
timeless Zoffix: `less quiet`? ?
21:26 kerframil joined, pullphinger left 21:27 SHODAN joined
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:gx/(<[$tokens]>)/-%tm{$1}-/;$m ~~ s:g/(^|\s)-/$1/;$m ~~ s:g/-(\s|$)/$1/;$m ~~ s:g/--/-/;return $m;}; say english-for-tokens('%'); 21:27
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Adverb gx not allowed on substitution␤at <tmp>:1␤------> 3join;$m ~~ s:gx/(<[$tokens]>)/-%tm{$1}-/7⏏5;$m ~~ s:g/(^|\s)-/$1/;$m ~~ s:g/-(\s|$)␤»
timeless right, so, what's the correct magic for `i want to run some perl in the second part of an s/// ?
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:gx/(<[$tokens]>)/"-"~%tm{$1}~"-"/;$m ~~ s:g/(^|\s)-/$1/;$m ~~ s:g/-(\s|$)/$1/;$m ~~ s:g/--/-/;return $m;}; say english-for-tokens('%'); 21:28
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Adverb gx not allowed on substitution␤at <tmp>:1␤------> 3m ~~ s:gx/(<[$tokens]>)/"-"~%tm{$1}~"-"/7⏏5;$m ~~ s:g/(^|\s)-/$1/;$m ~~ s:g/-(\s|$)␤»
jnthn Stick it in a block
s/.../{...}/
21:29 kurahaupo left
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:g/(<[$tokens]>)/{"-"~%tm{$1}~"-"}/;$m ~~ s:g/(^|\s)-/$1/;$m ~~ s:g/-(\s|$)/$1/;$m ~~ s:g/--/-/;return $m;}; say english-for-tokens('%'); 21:29
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5===␤Unrecognized regex metacharacter - (must be quoted to match literally)␤at <tmp>:1␤------> 3s]>)/{"-"~%tm{$1}~"-"}/;$m ~~ s:g/(^|\s)7⏏5-/$1/;$m ~~ s:g/-(\s|$)/$1/;$m ~~ s:g/--␤Unrecognized regex metacharacter - (must be quot…»
timeless ok, i think that's progress
someone will have to explain to me what `regex metacharacters` are, and why i can't just do that... 21:30
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:g/(<[$tokens]>)/{"-"~%tm{$1}~"-"}/;$m ~~ s:g/(^|\s)\-/$1/;$m ~~ s:g/\-(\s|$)/$1/;$m ~~ s:g/\-\-/-/;return $m;}; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«%␤»
timeless well, gee, that didn't work
m: my %tm = <% percent # hash - dash . dot ? question>; my $tokens = %tm.keys.join;say "<[$tokens]>" 21:31
camelia rakudo-moar 865bf3: OUTPUT«<[?%.-#]>␤»
timeless m: my %tm = <% percent # hash - dash . dot ? question>; say %tm{'%'}
camelia rakudo-moar 865bf3: OUTPUT«percent␤»
jnthn Just try s:g/@(%tm.keys)/.../ or so 21:32
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:g/(@(%tm.keys))/{"-"~%tm{$1}~"-"}/;$m ~~ s:g/(^|\s)\-/$1/;$m ~~ s:g/\-(\s|$)/$1/;$m ~~ s:g/\-\-/-/;return $m;}; say english-for-tokens('%'); 21:33
camelia rakudo-moar 865bf3: OUTPUT«Use of Nil in string context in block at <tmp> line 1␤Use of uninitialized value %tm of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block at <tmp> line 1␤Cannot assign to a readonl…»
21:33 AlexDaniel joined
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; say "-".subst(/@(%tm.keys)/, { %tm{$_} }, :g) 21:34
camelia rakudo-moar 865bf3: OUTPUT«dash␤»
Zoffix timeless, that looks horrifying
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m ~~ s:g/@(%tm.keys)/{"-"~%tm{$1}~"-"}/;$m ~~ s:g/(^|\s)\-/$1/;$m ~~ s:g/\-(\s|$)/$1/;$m ~~ s:g/\-\-/-/;return $m;}; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«Use of Nil in string context in block at <tmp> line 1␤Use of uninitialized value %tm of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block at <tmp> line 1␤Cannot assign to a readonl…»
timeless Zoffix: keep in mind that i'm mostly a JS dev who speaks perl5 21:35
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; say /@(%tm.keys)/
camelia rakudo-moar 865bf3: OUTPUT«/@(%tm.keys)/␤»
Zoffix jnthn, what sorcery is that? :o
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m = $m.subst(%tm.keys,{"-"~%tm{$1}~"-"});$m ~~ s:g/(^|\s)\-/$1/;$m ~~ s:g/\-(\s|$)/$1/;$m ~~ s:g/\-\-/-/;return $m;}; say english-for-tokens('%'); 21:36
camelia rakudo-moar 865bf3: OUTPUT«Cannot assign to a readonly variable or a value␤ in sub english-for-tokens at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timeless AlexDaniel or someone else, can rakudo please report better than `line 1`?
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; my $x = 'a-b'; $x ~~ s:g/@(%tm.keys)/-%tm{$/}-/; say $x;
camelia rakudo-moar 865bf3: OUTPUT«a-dash-b␤»
21:36 bjorkintosh joined, bjorkintosh left
jnthn timeless: If you run it locally and spread your code out, it can ;) 21:37
timeless jnthn: the fact that you can use `$/` inside a s/// is horrifying.
jnthn Zoffix: /@a/ means /@a[0] | @a[1] | .../
timeless: Why?
AlexDaniel doesn't it make sense?
timeless jnthn: ... it just requires a different kind of parser
masak timeless: I find it rather natural
jnthn Well, yeah :)
Zoffix jnthn, wow, neat
jnthn Zoffix: And then @(...) just lets you put any expression that can evaluate to or coerce to a list in there :) 21:38
Zoffix :)
jnthn Zoffix: Pun on the @(....) coercer in the main language.
timeless the /@a/ notation is quite awesome
AlexDaniel timeless: just don't use / then
Zoffix timeless, it can if you stick ␤ to separate lines.
m: ␤␤␤␤␤␤; lfcsadjfdasdbs
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ lfcsadjfdasdbs used at line 1␤␤»
21:38 setty1 left
Zoffix :/ 21:39
timeless =b
AlexDaniel why does it need newlines in order to produce a proper message?
Zoffix m: say 42+42; ␤␤␤␤␤␤; lfcsadjfdasdbs
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ lfcsadjfdasdbs used at line 7␤␤»
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; my $x = 'a-b'; $x ~~ s:g/@(%tm.keys)/ = "-%tm{$/}-"; say $x;
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed replacement part; couldn't find final /␤at <tmp>:1␤------> 3 s:g/@(%tm.keys)/ = "-%tm{$/}-"; say $x;7⏏5<EOL>␤ expecting any of:␤ postfix␤»
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; my $x = 'a-b'; $x ~~ s:g{@(%tm.keys)} = "-%tm{$/}-"; say $x;
camelia rakudo-moar 865bf3: OUTPUT«a-dash-b␤»
21:39 Emeric left
jnthn You can also write the assignment form of substitution, which may be easier on the eye 21:39
timeless easier on the eye, maybe, definitely not on my brain
this is not the perl5 i thought i knew :) 21:40
Zoffix AlexDaniel, what's "proper message"? I think the Q was how it can report the line number of the error and since everything is on one line, of course it'll report the error on line 1
timeless Zoffix: it can report a cursor position w/in the line
which it does in some examples
m: my %tm = <% percent # hash - dash . dot ? question>; my $x = 'a-b'; $x ~~ s:g/@(%tm.keys)/ = "-%tm{$/}-"; say $x;
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed replacement part; couldn't find final /␤at <tmp>:1␤------> 3 s:g/@(%tm.keys)/ = "-%tm{$/}-"; say $x;7⏏5<EOL>␤ expecting any of:␤ postfix␤»
Zoffix Ah
AlexDaniel yes, that's how I understood it – where's a proper position?
Zoffix m: say 42+42; ␤␤␤␤␤␤; say 5+6; lfcsadjfdasdbs 21:41
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ lfcsadjfdasdbs used at line 7␤␤»
AlexDaniel saying “line 1” is basically like saying “file foo.p6”
Zoffix *shrug*
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) {my $tokens = %tm.keys.join;$m = $m.subst(%tm.keys,{"-"~%tm{$1}~"-"});$m ~~ s:g/(^|\s)\-/$1/;$m ~~ s:g/\-(\s|$)/$1/;$m ~~ s:g/\-\-/-/;return $m;}; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«Cannot assign to a readonly variable or a value␤ in sub english-for-tokens at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timeless clearly there's a `readonly` variable somewhere
and an assignment somewhere
AlexDaniel timeless: basically, in such cases we just say that it is less than awesome
timeless to me, the pointer should be near the assignment
tony-o you need is rw on the english-for-tokens sig
AlexDaniel timeless: and submit a LTA bug report! :) 21:42
tony-o sub english-for-tokens($m is rw) ...
timeless tony-o: oh, i'm trying to use copy value
tony-o ah, then use 'is copy'
timeless is there sub english-for-tokens($m is copy) ?
Zoffix yes
timeless wow 21:43
that's the perl i like :)
things just as you'd expect them to be :)
Zoffix timeless, perl6.party/post/Perl-6-There-Are-T...t-1#iscopy
timeless AlexDaniel: i don't have time to do the TLA
AlexDaniel yea, timeless
tony-o Zoffix: good link
AlexDaniel timeless: TLA?
timeless LTA
AlexDaniel timeless: well, if you wont submit it then I'll do in an hour or so :) 21:44
timeless thanks
timeless goes back to try to find the assignment form :/
jnthn timeless: Note that if you're doing $m = $m.subst...; return $m, you could do away wiht the `is copy` and just so return $m.subst...; :-) 21:45
timeless jnthn: ok, what does this actually look like?
(my scrollback and brain aren't handling this stream well at all) 21:46
(also, i have ~30mins and i want to get something working, not learn all of perl6 or boil an ocean)
21:47 stevieb9 joined
jnthn my %tm = <% percent # hash - dash . dot ? question>; english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('%'); 21:47
m: my %tm = <% percent # hash - dash . dot ? question>; english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$m' is not declared␤at <tmp>:1␤------> 3h . dot ? question>; english-for-tokens(7⏏5$m) { return $m.subst(/@(%tm.keys)/, {"-␤»
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«-percent-␤»
jnthn That maybe? :) 21:48
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { "-%tm{$^a}-" }; say english-for-tokens('%');
camelia rakudo-moar 865bf3: OUTPUT«-percent-␤»
Zoffix doesn't really get what that sub is meant to be doing 21:49
timeless : my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('THIS%%THAT');
my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('THIS%%THAT');
jnthn m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('text with % and -');
camelia rakudo-moar 865bf3: OUTPUT«text with -percent- and -dash-␤»
jnthn That I guess :)
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g); }; say english-for-tokens('THIS%%THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent--percent-THAT␤»
jnthn Enough for me today...'night o/
timeless hrm...
almost right, except i wanted it to collapse doubled dashes 21:50
tony-o that's p neato
timeless but i can fix tha
m:my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($message) {my $tokens = %tokenmap.keys.join;return $message.subst(/@(%tm.keys)/, {"-"~%tokenmap{$_}~"-"}/.subst(/(^|\s)\-/, "$1").subst(/\-(\s|$)/, "$1").subst(/\-\-/, '-');}; say english-for-tokens('%%%') 21:51
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_}:exists ?? "-%tm{$_}-" !! $_}).join }; say english-for-tokens('THIS%%THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent--percent-THAT␤»
timeless ok, is that more perlish? 21:53
21:53 stevieb9 left
timeless not that i still really want to strip `-`s that hit spaces and start/end 21:53
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_}:exists ?? "-%tm{$_}-" !! $_}).join.subst(/(^|\s)\-/, "$1").subst(/\-(\s|$)/, "$1").subst(/\-\-/, '-'); }; say english-for-tokens('THIS%%THAT');say english-for-tokens('##'); say english-for-tokens("$ $") 21:54
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Non-variable $ must be backslashed␤at <tmp>:1␤------> 3r-tokens('##'); say english-for-tokens("7⏏5$ $")␤ expecting any of:␤ argument list␤ double quotes␤ prefix␤ …»
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_}:exists ?? "-%tm{$_}-" !! $_}).join.subst(/(^|\s)\-/, "$1").subst(/\-(\s|$)/, "$1").subst(/\-\-/, '-'); }; say english-for-tokens('THIS%%THAT');say english-for-tokens('##'); say english-for-tokens('$ $') 21:55
camelia rakudo-moar 865bf3: OUTPUT«Use of Nil in string context in sub english-for-tokens at <tmp> line 1␤Use of Nil in string context in sub english-for-tokens at <tmp> line 1␤THIS-percent-percent-THAT␤Use of Nil in string context in sub english-for-tokens at <tmp> line 1␤Use of…»
timeless AlexDaniel: LTA ^ 21:56
say "-a-".subst(/(^|\s)\-/, "$1")
m:say "-a-".subst(/(^|\s)\-/, "$1")
m: say "-a-".subst(/(^|\s)\-/, "$1") 21:57
camelia rakudo-moar 865bf3: OUTPUT«Use of Nil in string context in block <unit> at <tmp> line 1␤a-␤»
timeless m: say "-a-".subst(/(^|\s)\-/, "$_")
camelia rakudo-moar 865bf3: OUTPUT«Use of uninitialized value $_ of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at <tmp> line 1␤a-␤»
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { my $out; my $is-t = False; for $^a.comb { if %tm{$_} { $out ~= ($is-t ?? '' !! '-' ) ~ "%tm{$_}-"; $is-t = True; next; }; $out ~= $_; $is-t = False; }; $out }; say english-for-tokens('THIS%%THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent-percent-THAT␤»
Zoffix timeless, don't know about "more perlish" but to my eye more readable (granted, a large part of that is 'cause I just wrote it :P) 21:58
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { my $out; my $is-t = False; for $^a.comb { if %tm{$_} { $out ~= ($is-t ?? '' !! '-' ) ~ "%tm{$_}-"; $is-t = True; next; }; $out ~= $_; $is-t = False; }; $out }; say english-for-tokens('THIS% %THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent- -percent-THAT␤»
timeless Zoffix: i'm hoping to get `THIS-percent percent-THAT`
timeless hopes that makes sense 21:59
Zoffix timeless, what's the condition there? Whitespace? Space?
timeless in my mind, \s
21:59 cognominal left
masak 'night, #perl6 21:59
Zoffix night
timeless night
Zoffix: basically, if a token runs up against some other "word", i want a "-" to split it from the word 22:00
but if it's running up against ^, \s, or $, then it doesn't need something, because there's naturally something
a token running up against another token should have one dash, not two
hopefully that sorta makes sense 22:01
we're trying to translate $a to dollar-a, "$a #b" to "dollar-a hash-b"
we=me, but hopefully i can convince everyone that this is a good and human-understandable way to read things :) 22:02
22:03 ptolemarch left
timeless the easiest way i can think of it is `subst token for -word-` + remove extraneous (^/ / $) dashes + collapse double -- 22:04
yes, you can create a state machine to decide whether you need a dash or not, but it seems more complicated than just doing multiple passes 22:05
ideally strings are fairly short (<10 chars of input?)
22:05 acrussell left
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; $_ = $_ but role MyToken {} for %tm.values; sub english-for-tokens { my $out; my $last-t = False; for $^a.comb.map: { %tm{$_} // $_ } { unless $_ ~~ MyToken { $out ~= '-' if $last-t; $out ~= $_; $last-t = False; next }; $last-t = True; if $out ~~ /\w$/ { $out ~= '-' }; $out ~= $_ }; $out }; say english-for-tokens('THIS%%THAT'); 22:06
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent-percent-THAT␤»
Zoffix Yey for overly complex, convoluted code :) 22:07
timeless m: my %tm = <% percent # hash - dash . dot ? question>; $_ = $_ but role MyToken {} for %tm.values; sub english-for-tokens { my $out; my $last-t = False; for $^a.comb.map: { %tm{$_} // $_ } { unless $_ ~~ MyToken { $out ~= '-' if $last-t; $out ~= $_; $last-t = False; next }; $last-t = True; if $out ~~ /\w$/ { $out ~= '-' }; $out ~= $_ }; $out }; say
english-for-tokens('% #THAT$');
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5===␤Argument to "say" seems to be malformed␤at <tmp>:1␤------> 3out ~= '-' }; $out ~= $_ }; $out }; say7⏏5<EOL>␤Other potential difficulties:␤ Unsupported use of bare "say"; in Perl 6 please use .say if you meant $_, or u…»
Zoffix timeless, if that's for what I think it is (fixing the issue with punctuation URLs on docs), then just trip double '--' or not worry about them at all
22:07 rindolf left
Zoffix s/trip/strip/; 22:07
timeless right issue
but, i really don't want someone to have to search for `-dash` 22:08
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; $_ = $_ but role MyToken {} for %tm.values; sub english-for-tokens { my $out; my $last-t = False; for $^a.comb.map: { %tm{$_} // $_ } { unless $_ ~~ MyToken { $out ~= '-' if $last-t; $out ~= $_; $last-t = False; next }; $last-t = True; if $out ~~ /\w$/ { $out ~= '-' }; $out ~= $_ }; $out }; say english-for-tokens('THIS% %THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent- percent-THAT␤»
Zoffix doesn't even work right heh
timeless that's just asking for a search engine to hate me
the problem is that `-` is the right general joiner
Zoffix timeless, what do you mean "search for dash"?
timeless but it's a horrible lead joiner
google: {term from perl docs}
if we escape "%%" as "-percent-percent-" or similar
and you google "-percent-percent" 22:09
(or w/ the trailing -)
google is going to see that as a search for {minus} {percent-percent....}
and exclude everything you want
which is really really not what we want :/
Zoffix why the f would you google for "-percent-percent"? :S
timeless well, if we encode %% as -percent-percent- ...
you might try googling that way 22:10
Zoffix I wouldn't use any dashes
In fact, I'd likely google for %%
But I wouldn't. 'cause I'd use the doc search.
timeless some of these don't google well
`//` especially doesn't iirc
half the reason for this exercise is to make searching work better
(part of the reason is to make urls work better for the server, but, that's only part) 22:11
most people don't use local search engines
Zoffix Well, my suggestion would be to forget about this. Right now the problem is those pages don't generate and don't load. It's pointless trying to ascertain how well those pages are googlable when there aren't any pages to begin with :)
timeless i'd rather fix it right once
than get them loading w/ crappy urls
Zoffix Didn't you just have ~30mins?
timeless yep 22:12
(it's 6:15 now, i need to be somewhere by 7:15, and it takes ~20mins, plus i need to put on clothes for outside...)
22:13 rurban_ left
timeless so, is your a.comb.map really better than using an s/// or subst? 22:13
perl: the language where you have half a dozens ways to do whatever you want to do 22:14
Zoffix Right, TimToady is the philosophy, so it's a bit silly to ask if it's better
timeless hasn't read TimToady 22:15
22:15 stevieb9 joined
Zoffix TimToady = TIMTOWTDI = There Is More Than One Way To Do It 22:15
22:18 cognominal joined
timeless doc.perl6.org/routine/subst 22:19
> There is an in-place syntactic variant of subst spelled s/matcher/replacement.
that last bit is missing a trailing `/`, right?
22:20 Sgeo joined
dalek c: b2904a1 | (Zoffix Znet)++ | doc/Type/Str.pod:
Fix typo timeless++
22:20
timeless hrm, captures start at 0? 22:21
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst: /^\-|'--'|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('THIS% %THAT');
camelia rakudo-moar 865bf3: OUTPUT«THIS-percent percent-THAT␤»
Zoffix timeless, yes
timeless is that consistent w/ perl5? 22:22
Zoffix No, in Perl 5 they start with 1
22:22 huggable left, huggable joined
timeless doc.perl6.org/language/5to6-nutshell didn't mention that... 22:22
dalek c: 6e3a139 | (Zoffix Znet)++ | doc/Language/5to6-nutshell.pod:
Mentions captures start with $0
22:24
Zoffix timeless++
22:24 TEttinger joined
timeless + /(.+)/ and print $1; # Perl 5 22:24
+ /(.+)/ and print $0; # Perl 5
...
so close
dalek c: 32c14cd | (Zoffix Znet)++ | doc/Language/5to6-nutshell.pod:
Fix typo
22:25
22:25 baest joined
timeless better :) 22:25
m: say "c a".subst(/(.)a/, "b$0") 22:26
camelia rakudo-moar 865bf3: OUTPUT«Use of Nil in string context in block <unit> at <tmp> line 1␤cb␤»
timeless so... i still don't understand wha ti'm doing wrong here
22:27 TEttinger left, baest_ left
Juerd $0 in Perl 5 has nothing to do with regexes even 22:27
Zoffix timeless, that doesn't work 'cause you're interpolating the $0 from your current scope, not one when subst is doing its thang 22:28
dha perl -e 'print "Just another p$0-r-l hacker\n"
'
Zoffix m: say "c a".subst(/(.)a/, -> $/ { "b$0"} )
camelia rakudo-moar 865bf3: OUTPUT«cb ␤»
timeless is there any more concise way to do that? 22:29
timeless will use that otherwise
Zoffix m: <improve-search-on-docs-site play-Assassins-Creed>.pick.say 22:30
camelia rakudo-moar 865bf3: OUTPUT«improve-search-on-docs-site␤»
timeless =b
thanks camelia
Juerd Speaking of docs, any thoughts on b2gills's comment at gist.github.com/Juerd/b6531810922d...0be4915ca?
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}; }; say english-for-tokens('text with % and -');
camelia rakudo-moar 865bf3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in argument list; couldn't find final ')' ␤at <tmp>:1␤------> 3}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}; 7⏏5}; say english-for-tokens('text with % a␤ expecting any of:…»
Juerd I know b2gills's opinion about .WHAT vs .^name, but I'd like to sollicit more responses 22:31
timeless i don't see a comment 22:35
just 118 lines of text, all of which are readable
Juerd I see a comment at the bottom of that
timeless has an actual book w/ 2 errors to report on tonight
Juerd juerd.nl/i/14f77b90306d38b7d5c871371d693431.png 22:36
timeless :o -- irccloud didn't show me that
22:36 dha left 22:37 bjz_ left
timeless i can't comment on b2gills 's comment 22:37
because i don't understand how .^name and .WHAT differ
i sure hope they're explained somewhere
Juerd .WHAT returns a type object, which is undefined 22:38
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}); }; say english-for-tokens('text with % and -');
camelia rakudo-moar 865bf3: OUTPUT«text with percent- and -dash-␤»
Juerd .^name returns a string
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}).subst(/($|\s)\-/, -> $/ { "$0"}); }; say english-for-tokens('text with % and -');
camelia rakudo-moar 865bf3: OUTPUT«text with percent- and dash-␤»
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}, :g).subst(/($|\s)\-/, -> $/ { "$0"}, :g); }; say english-for-tokens('text with % and -');
camelia rakudo-moar 865bf3: OUTPUT«text with percent- and dash-␤»
22:38 cdg left
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}, :g).subst(/($|\s)\-/, -> $/ { "$0"}, :g).subst(/\-\-/, "-", :g); }; say english-for-tokens('text with %, ##, @ /, and -'); 22:40
camelia rakudo-moar 865bf3: OUTPUT«text with percent-, hash-hash-, @ /, and dash-␤»
dalek c: 6c38120 | (Zoffix Znet)++ | html/css/style.css:
Remove too-large margins on mobiles
timeless ok, that /almost/ works for me
Zoffix timeless, what wasn't working in my last version?
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst: /^\-|'--'|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('text with %, ##, @ /, and -'); 22:41
camelia rakudo-moar 865bf3: OUTPUT«text with percent, hashhash, @ /, and dash␤»
timeless might not have noticed it
Zoffix hm
timeless `hashhash`
Zoffix never mind
timeless mine isn't working either
and my example sucks
Zoffix Ah
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens($m) { return $m.subst(/@(%tm.keys)/, {"-%tm{$_}-"}, :g).subst(/(^|\s)\-/, -> $/ { "$0"}, :g).subst(/($|\s)\-/, -> $/ { "$0"}, :g).subst(/\-\-/, "-", :g); }; say english-for-tokens('text with %, ##, . ?, and -');
camelia rakudo-moar 865bf3: OUTPUT«text with percent-, hash-hash-, dot- question-, and dash-␤»
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst(/'--'/, '-', :g).subst: /^\-|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('text with %, ##, @ /, and -'); 22:42
camelia rakudo-moar 865bf3: OUTPUT«text with percent, hash-hash, @ /, and dash␤»
timeless expected is «text with percent-, hash-hash-, dot question-, and dash␤»
Zoffix :S
timeless m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst(/'--'/, '-', :g).subst: /^\-|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('text with %, ##, . ?, and -');
camelia rakudo-moar 865bf3: OUTPUT«text with percent, hash-hash, dot question, and dash␤»
timeless so `hash-hash,` is questionable 22:43
if it doesn't mess anything else up, then yours is a keeper
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst(/'--'/, '-', :g).subst: /^\-|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('% .x, #?#, . ?, and -'); 22:44
camelia rakudo-moar 865bf3: OUTPUT«percent dot-x, hash-question-hash, dot question, and dash␤»
timeless ok. i'll take yours
timeless isn't interested in figuring out why `,` ended up special, but hey, it's not bad :)
Zoffix Because it's not \w 22:45
timeless ah, `\w` vs. `\s`
ok
m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst(/'--'/, '-', :g).subst: /^\-|\-<before \W>|<after \W>\-|\-$/, '', :g }; say english-for-tokens('%" .x, #?#, . ?, and -');
camelia rakudo-moar 865bf3: OUTPUT«percent" dot-x, hash-question-hash, dot question, and dash␤»
timeless nope
well 22:46
hrm
yeah, i don't think `percent"` is particularly good
Zoffix m: my %tm = <% percent # hash - dash . dot ? question>; sub english-for-tokens { $^a.comb.map({ %tm{$_} ?? "-%tm{$_}-" !! $_ }).join.subst(/'--'/, '-', :g).subst: /^\-|\-<before \s>|<after \s>\-|\-$/, '', :g }; say english-for-tokens('%" .x, #?#, . ?, and -');
camelia rakudo-moar 865bf3: OUTPUT«percent-" dot-x, hash-question-hash-, dot question-, and dash␤»
timeless yeah, i'll take that
although i have absolutely no idea where to put NLs in :/ 22:47
(and i have to run shortly)
22:47 pmurias left
Zoffix See! And we could've had working pages by now :P 22:47
timeless does this work www.irccloud.com/pastebin/Q69mkVqM/ 22:48
Zoffix sure
timeless this is what i'm thinking www.irccloud.com/pastebin/apCalljE/ 22:49
although, it occurs to me that i'll need to make sure we only use `href_escape` for internal documents and not wikipedia 22:50
Zoffix Will that update the generated search index too?
You're also missing slash
(for a token)
timeless ?
22:51 cooper_ joined, cooper_ left, cooper_ joined
Zoffix ¿ 22:51
timeless please assume that i'm really not good at spotting errors
(which is ironic, but hey)
Zoffix Your %tm doesn't have / slash in it
timeless sub write-search-file() { 22:52
has a sub escape(Str $s) {
which would need to be fixed
i think that's /probably/ all that needs to be patched
Zoffix And I don't know how it's generated but there's a search index that picks things out of X<> in pod (and I'm guessing other things)... will that get automatically updated to link to new URLs?
timeless assuming i'm half understanding what's going on
hrm, we probably need to add bs to the list of tokens 22:54
m: my %tm = <% percent # hash - dash . dot ? question / slash \ blackslash>; say %tm{'\\'}
camelia rakudo-moar 865bf3: OUTPUT«blackslash␤»
timeless is that what you meant by me missing a `/`? 22:55
Zoffix yes
timeless sorry, my code-interpolate is weak 22:56
and now i'm late
something like this www.irccloud.com/pastebin/yN7My85e/ 22:57
timeless will be back in +4h or so
AlexDaniel uh, actually gonna get some sleep
will submit stuff tomorrow :)
Zoffix ? 22:59
23:04 spider-mario left 23:05 AlexDaniel left
b2gills Juerd: If .WHAT vs .^name is all that's stopping you, just do it whichever way you want. It's just a minor nit. 23:06
dalek c: adbcb5c | (Zoffix Znet)++ | / (3 files):
Improve search when no results found

Changed "Nothing found" to "Not in index," to suggest the searched item may be on the site, but not indexed.
Added a "try web search" link that googles for the searched term on the doc site. Chose Google instead of DuckDuckGo, because some stuff wasn't giving any results on DDG, but Google did find some pages.
Juerd b2gills: It's not stopping me, but if it's going te be changed I might want to change the text and maybe even the ordering as well. 23:12
b2gills: What's stopping me most, is lack of time, and that I lost my handwritten notes :(
23:12 cognominal left
Juerd I looked up how symbol tables, packages, namespace and type names interacted. But I lost my notes and I forgot the specifics. 23:14
b2gills .oO( What usually stops me is being unable to read my handwritten notes )
Juerd Oh, I have sufficiently legible handwriting :)
23:16 lushkava joined 23:18 tbrowder_ joined 23:19 kerframil left
Juerd b2gills: The primary reason why I'm hesitant to change it to .^name is that I find it less obvious code. It being a meta-method requires explaining, which distracts from the point of this document. 23:19
It's not the name of the LHS
23:19 fhorck joined
b2gills .WHAT isn't a normal method either 23:19
Juerd No, but at least it looks like one.
b2gills .?method is a method call, and looks just as much like a method call as .^method 23:20
Juerd Yes, but these examples should remain as simple as possible, imho
The only explanation necessary should be relevant to the document's topic 23:21
So it shouldn't raise questions outside that scope
23:21 cognominal joined
Juerd I went off that path for the junction example, and it was caught by two people that I asked to proofread this. 23:22
23:22 tbrowder_ left
Juerd They're not Perl 6 programmers and they could follow everything in this text except for the part about junctions, and that weird use of what they thought was the bitwise AND operator... 23:22
b2gills I just don't like it when introductory materials give a false impressions 23:23
Juerd Which they didn't realise was a junction.
I think you may be over-interpreting the impression that it gives
b2gills Well a junction doc should start with all() any() one() none() then proceed to the infix operators
Juerd But really, I already know your opinion and I think you make a good point. But I need more people's opinions before I decide my own :) 23:24
23:24 RabidGravy left
Juerd Well, junctions are kind of relevant to a document about types, because they're the sole distinction between Mu and Any. 23:24
23:24 yqt left
Juerd So the word "junction" should link to the relevant page, and the "Hello" & "World" example should have a comment that explains that this uses a junction 23:25
That way I can discuss the relevant parts without having to explain what junctions are 23:26
b2gills in some cases I mean a false impression of how limited a feature is. most new Perl 6 programmers don't have a clue how powerful signatures are for example 23:27
Juerd Perhaps, but you can't use complex examples all the time because that makes learning the language too hard.
b2gills I think that a page about Mu should use &any not &infix:<|>, and just punt the problem 23:28
Juerd So where signatures are not the topic of discussion, they should be kept simple and self-explanatory
I'm not writing a page about Mu, but about types :)
Of which Mu is an important one
b2gills I'm thinking just a note pointing to a page on signatures while mentioning that they are far more capable 23:29
Juerd Perhaps.
But I don't think I will. This is not a table of contents for all the powerful features that Perl 6 has to offer. It's a leaf node. 23:30
b2gills &infix:<|> should only be on advanced pages or in a page on junctions, just for the fact it looks like a C operator
Juerd Yes, I should probably use all() instead of &.
23:31 Calibellus joined
b2gills That's also why I tried to persuade TimToady to change the name of 「with」 to something else, because other languages have a with statement that is more similar to our given statement 23:33
Juerd I dislike with and without. 23:34
I don't interpret those as conditionals. 23:35
b2gills I actually think that given sounds more like a conditional than with does, but it isn't going to change now 23:36
23:36 ocbtec left, TEttinger joined
Juerd But then, I haven't seen any reason for using with/without anyway... 23:36
23:39 Timbus left 23:51 Timbus joined 23:54 cognominal left
b2gills m: without (Str,Int,5).pick -> $_ { .say } 23:56
camelia ( no output )
b2gills m: without (Str,Int,5).pick -> $_ { .say }
camelia rakudo-moar 865bf3: OUTPUT«(Str)␤»
Juerd If I really wanted a topicalizing if defined, I'd probably call that if-defined 23:59
b2gills That wouldn't work the same way with 「if defined(...) -> $var {...}」 or 「unless defined(...) -> $var {...}」
Juerd I don't see much reason to topicalize the not defined case, so "if not defined" and "unless defined" are good enough for me :)