Tester wanted: pugs.blogs.com/pre1/Perl6-Pugs-6.2.13.tar.gz | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com
Set by audreyt on 15 October 2006.
TreyHarris TimToady: that precedence is really, really unintuitive in the extreme. 00:00
lanny ?eval pick 33
evalbot_r14375 Error: pick not defined: VInt 33
TimToady it's only off by one. :)
lanny ?eval pick (33)
evalbot_r14375 Error: pick not defined: VInt 33
lanny oh, right
?eval pick (33,)
evalbot_r14375 33 00:01
lanny ?eval pick (33,)..77
evalbot_r14375 7
TreyHarris i would expect range ops to bind very tight
lanny :)
TimToady they're non chaining because you can't do $a .. $b .. $c
lanny There's your answer, TreyHarris. 7
jojo ?eval pick (1..1e10) 00:02
evalbot_r14375 pugs: out of memory (requested 1048576 bytes)
TimToady pugs isn't range object savvy yet... 00:03
TreyHarris TimToady: it's a chaining issue, not a prec issue? i mean, to me, .. "feels" like **... it should bind tightly enough to where i can treat a range as a single thing like i treat an exponentiated number as a single thing... 00:04
jojo well, at least it gave a sensible error
TimToady what should abs $a <=> abs $b give you?
abs $a .. abs $b is the same 00:05
if .. were as tight as ** you'd have to be saying ($a-1)..($b+1) all the time.
TreyHarris i'd expect to do that, as otherwise i'd think i'd get $a - (1..$b) + 1, which is silly. 00:06
lanny ?eval say pick (5,)..10
evalbot_r14375 OUTPUT[3ā¤] Bool::True
TreyHarris i need to ask audreyt what font she's using for IRC... there is no font installed on the Mac that has that NL character evalbot uses 00:07
TimToady she just gets "NL" herself, is my understanding.
TreyHarris (Macs let you search by codepoint for fonts with that character, so i know none of mine have it) 00:08
lanny ?eval say 5..10
evalbot_r14375 OUTPUT[5678910ā¤] Bool::True
lanny Ok. Why doesn't 'pick (5,)..10' give me 5..10?
TimToady beats me 00:09
lanny :) A test!
TimToady ?eval say (5,)..10
evalbot_r14375 OUTPUT[12345678910ā¤] Bool::True
jojo ?eval print 1
evalbot_r14375 OUTPUT[1] Bool::True
TimToady I'd say it's taking the length of (5,)
TreyHarris lanny: (5,) in scalar context is... yes. ==TimToady 00:10
jojo ?eval say 1
evalbot_r14375 OUTPUT[1ā¤] Bool::True
TreyHarris so you're saying 1..10
lanny But I'm pick-ing it.
TimToady no you're not
lanny ?eval say pick (5,)..10
evalbot_r14375 OUTPUT[6ā¤] Bool::True
TimToady ?eval say pick(5,)..10
evalbot_r14375 Error: pick not defined: VInt 5
lanny aha. Ok. I see it now 00:11
TimToady P6 is *very* picky about whitespace before postfixes.
lanny I don't understand it. But I see it.
?eval say pick(33,) 00:12
evalbot_r14375 Error: pick not defined: VInt 33
jojo is "say" perl6 or bot-specific?
lanny ?eval say pick (33,)
evalbot_r14375 OUTPUT[33ā¤] Bool::True
TreyHarris is there a POE irc channel? i've gummed up this app with debug logging, and I *know* my event generator's getting called several times, but while my event handler gets called the right *number* of times, it's getting the first set of data each time... 00:13
TimToady pick with whitespace is always a list operator. 00:14
lanny It surprised me that pick (33,) gives 33 but pick (33,)..77 really gives pick (1..77) after the conversation you and TreyHarris were having on ..
I see why now. 00:15
?eval pick (pick (33,))..77
evalbot_r14375 72
lanny tada
jojo: perl6. check out syn.pugscode.org 00:17
lambdabot Title: Official Perl 6 Documentation
jojo thanks lanny . i see say all over it. i don't immediately find the justification 00:21
but i will never object to taking two letters off print
is $\ obsolete?
TimToady jojo, yes. 00:23
that sort of thing should now be done by the filehandles. 00:24
infact, default input filehandle autochomps now.
jojo arguably, that would always have been correct 00:25
00:25 nekokak joined
TimToady wouldn't have worked in earlier Perl 00:25
because an empty newline would have been false. 00:26
now that we use for instead of while, it works much better.
but for "for" to work well, lists had to become lazy.
jojo good point. i always use defined ($line = <>)
actually, does perl5 -ln have the same problem? 00:27
TimToady no, because perl 5 hacks in a hidden defined() for you. 00:29
but it's a kludge
while in p6 is not special that way.
jojo that is nasty, especially since p5 man perlrun shows the loop without defined 00:31
TimToady there are many, many nasty things in p5...
jojo and p5 man version of the loop is correct in terms of braces and that nasty golf hack
pasteling "bsb" at 203.214.67.82 pasted "more $_ defaulting prints and says" (14 lines, 350B) at sial.org/pbot/20496 00:34
TimToady the S04 example is not actually incorrect. 00:36
(it's using a feed ==> to supply print's arguments)
bsb I wasn't sure about that one 00:38
I still mistakenly think of ==> like a map sometimes 00:39
maybe that's not completely wrong from a certain angle... 00:40
00:47 gugod joined
TimToady decommute & 00:51
00:53 Pomin joined 00:55 gugod joined 01:55 gnuvince joined 02:05 diakopter joined, diakopter left 02:10 nipotaway is now known as nipotan 02:22 justatheory joined 02:41 weinig is now known as weinigZzZ
svnbot6 r14376 | audreyt++ | * Another pass at ChangeLog... 02:41
02:43 lanny joined 02:53 putter joined
putter For Boston area folks, there is a Sage sage.soe.ucsc.edu/ talk on Wednesday. lists.ccs.neu.edu/pipermail/pl-sem...00356.html . Fyi. 02:55
lambdabot Title: Sage: A Programming Language with Hybrid Type-Checking
lanny Very odd. My win32 pugs build runs fine except if I just type ./pugs when it hangs and never gives output of any kind. 03:00
audreyt surely you mean .\pugs ?
lambdabot audreyt: You have 2 new messages. '/msg lambdabot @messages' to read them.
audreyt @messages
lambdabot Limbic_Region said 1d 7h 46m 17s ago: r14363 smokes 100% ok for me on Win32
lanny said 5h 5m 24s ago: Pugs::Doc::Hack refers to grabbing Bundle::Pugs::SmokeKit rather than Task::Smoke
lanny mingw. mean ./pugs
audreyt lanny: please fix Pugs::Doc::Hack and commit 03:01
ho hum. not sure what's going on
lanny audreyt: gladly. where do I find it? (or I would have)
audreyt docs/Pugs/Doc/Hack.pod
lanny Ok. Do I feel silly. I figured that was .hs documentation.
audreyt :) 03:02
lanny Ok. Someone's already fixed it. But CPAN still has it as Bundle::Pugs::Hack and pugscode.org points to that 03:03
audreyt that's okay today CPAN will be populated with a new pugs
lanny Also I found a workaround to nmake1.5. I hand-edited the Makefile and removed the Makefile run of docs/perl6. Been trying to tracedown how to break up the size of the stanza in docs/perl6/Makefile but ExtUtils::MakeMaker is scary 03:05
audreyt yeah, you can maybe write a Makefile postprocessor? 03:06
I'd like to see that fixed
lanny I'm pretty sure I could hack up a postprocessor. Was hoping to do it correctly but will get it working and then will get it right. 03:07
audreyt first get it working and then refactor
in that order :) 03:08
clkao audreyt: hey
audreyt hey clkao
clkao in US already?
lanny Smoke in 19 minutes. :)
clkao audreyt: flickr.com/photos/clkao/271801444/ 03:09
lambdabot Title: "Why Subversion hasn't taken over the world" on Flickr - Photo Sharing!
audreyt clkao: no, plane takes off tomorrow
clkao: lol # "We've taken over the white board" 03:15
"take over the white board, one bullet at a time"
svnbot6 r14377 | lanny++ | * Pugs::Doc::Hack.pod - mention smoke.yml as param to smokeserv-client.pl 03:20
lanny 'make smoke' shouldn't find a failure that succeeds with prove6 or './pugs -Iblib6/lib' I take it? 03:29
audreyt maybe it's todo
but yes 03:30
lanny Not a :todo. Re-running smoke and seeing what it says
svnbot6 r14378 | audreyt++ | * ChangeLog: 47 lines left... 03:39
03:40 SubStack joined
lanny It passed this time. Odd. Cosmic ray I guess. 03:47
svnbot6 r14379 | audreyt++ | * ChangeLog: 39 lines left...
gnuvince audreyt: is that 39 lines until release? 03:52
03:54 reZo_ joined
audreyt something like that yes 03:55
will pre2, then after some smoke succeses, then release
gnuvince Cool
You rock :)
audreyt I try :) 03:56
I'd rather hack Perl/Haskell, even C, than English , though...
audreyt hacks English in ChangeLog some more
gnuvince I can imagine.
Although the medium is not terribly interesting.
I mean, a ChangeLog? It's more fun to write a release message 03:57
audreyt that I already did, in a sketch form, on the journal 03:58
gnuvince Yeah, read that.
TreyHarris audreyt: if you'll be finished during my waketime (+3 hours from nowish) I'll be happy to hack the English if you just outline your thoughts in your commit... 04:04
audreyt: though i just realized you're probably speaking more metaphorically as "hack the english" meaning figuring out your thoughts, the medium being English. (around here when people say "hack english" they usually mean deal with grammar/spelling/writing style/etc.) 04:07
04:08 cmarcelo joined
audreyt TreyHarris: ok, a sec... 04:09
pasteling "gnuvince" at 24.37.81.6 pasted "failure to compile pugs (with ghc-6.6)" (14 lines, 720B) at sial.org/pbot/20498
04:10 Eidolos joined
svnbot6 r14380 | lwall++ | [runpugs.html] typo 04:14
cmarcelo maybe interesting for some of you (and maybe old...) GHC hackaton videos => hackage.haskell.org/trac/ghc/wiki/AboutVideos 04:15
lambdabot Title: AboutVideos - GHC - Trac, tinyurl.com/w2uvy
audreyt gnuvince: did you install GHC from source? have you extracted the -extralibs package?
this may be worth probing on Makefile.PL I think 04:16
will think about it
svnbot6 r14381 | audreyt++ | * Pugs.Prim: New "printf" builtin. 04:23
r14382 | audreyt++ | * Prelude.pm: Avoid using Perl 6 regex for &sprintf. Instead
r14382 | audreyt++ | use rx:P5// for greater compatibility.
audreyt TreyHarris: please speelcheck, grammarcheck, and sanitycheck ChangeLog as you see fit 04:26
svnbot6 r14383 | audreyt++ | * ChangeLog: Finished the "Feature Changes" section.
TreyHarris audreyt: ok :)
audreyt this is almost twice the size of the 6.2.12 changelog 04:27
04:27 SubStack joined
audreyt same time duration 04:27
twice as more commits as well
lambdabot: how's the Makefile hacking doing? 04:29
TreyHarris that's a smart bot! 04:30
lambdabot does Makefile haking?
audreyt it's meant to be "lanny"...
lunch, bbiab. I'll hack in "mtl" detection 04:32
other than that I think pre2 is in order
and then release
TimToady I note that runpugs says the server isn't running. is that because it's currently recompiling?
audreyt I have no idea... 04:35
@tell andara: does webterminal server actually spawn children? it doesn't look like it
lambdabot Consider it noted.
audreyt TimToady: recompilation produces ./pugs.new 04:36
the previous ./pugs continues to work
TimToady k
audreyt lunch, bbiab for real
&
TreyHarris question: "POD". Damian used "Pod", which I assumed meant that he was distinguishing Perl 5's POD from Perl 6's Pod. Was I incorrectly reading semantics into an orthographic error? 04:43
TimToady I think you're seeing a de-emphasis of the acronym. Maybe it's a new shibboleth. :) 04:46
TreyHarris TimToady: I pronounce that "sibboleth"
TimToady I guess I'll have to kill you then... 04:47
TreyHarris ?eval 1 04:50
04:50 evalbot_r14375 is now known as evalbot_r14383
evalbot_r14383 1 04:50
TreyHarris ?eval role A does B { }; role B does A { }; class Foo does A { }; my $obj = Foo.new() 04:51
04:51 baest joined
TreyHarris ?eval 1 04:51
uh-oh.
evalbot_r14383 (no output) 04:52
1
TreyHarris ah, ok. i was afraid it went into an infinite loop.
?eval role A does B { }; role B does A { }; class Foo does A { }; my $obj = Foo.new(); say "Hi!"
TimToady probably did
evalbot_r14383 (no output) 04:53
TreyHarris is that catchable? should be... i noticed the changelog said, "Circular mixin is no longer allowed" and was wondering if that meant support had been dropped or if it actually caught that 04:54
04:54 baest joined
TimToady maybe it means run-time mixin, not role composition? 04:55
TreyHarris ah, perhaps. i don't know how to write a test for "doesn't go into an infinite loop" though. Paging Drs. Turing and Godel, code blue.... 04:56
dduncan I have a quick Perl 5 question ... does a weak reference pointing to an object become undefined before or after that object's DESTROY() method is called? 04:57
TimToady haven't the foggiest
svnbot6 r14384 | audreyt++ | * ChangeLog: clarify self-mixin
TreyHarris dduncan: try it and see? i'd think it would have to be "before", but it would be easy to try 04:58
dduncan yes, but I thought it might be faster to ask someone
no matter
TimToady might be wronger too. :)
dduncan actually, I think it is a non-issue for my problem anyway 04:59
svnbot6 r14385 | trey++ | A few grammars, speeling, and usificationages fixes, more coming. 05:01
r14386 | trey++ | Standardize on International English spellings, as there was a mixture 05:07
r14386 | trey++ | in the new Changelog and IE spellings seems slightly more common in
r14386 | trey++ | text so far than American English.
TreyHarris yes, the past two logs have been tongue-in-cheek....
TimToady what's this International English crap? 05:09
TreyHarris TimToady: i can switch it all to American English easily enough if you'd like. There were slightly more "-ise" than "-ize" in the changelog thus far, so I just made the rest match 05:10
TimToady you can spell it however you like, just don't call it international unless all nations agree on it. :)
TreyHarris non-American English :) 05:11
cmarcelo ?seen nothingmuch 05:14
lambdabot nothingmuch is in #darcs, #haskell and #perl6. I last heard nothingmuch speak 10h 8m 59s ago.
05:15 BooK_ joined 05:18 nothingmuch joined
cmarcelo nothingmuch: ping 05:19
svnbot6 r14387 | lwall++ | more noun/verb agreement
r14388 | trey++ | More random ChangeLog fixes 05:22
TreyHarris hmm, re "Typed constraints on autovivification": why shouldn't this: 05:23
?eval my $x; $x<moose> = "foo"; $x[5] = 32; 05:24
05:24 evalbot_r14383 is now known as evalbot_r14387
evalbot_r14387 Error: Can't modify constant item: VUndef 05:24
TreyHarris cause $x to turn into a Capture?
(not thinking that it *should*, just wondering why not)
TimToady it would be possible to make it work that way. I suppose the question is how often such a thing would indicate an error rather than an intention. 05:28
if it's likely to be an error, then perhaps the intention should be more cleary expressed via some declaration before letting it autocapturize 05:29
TreyHarris yes, that's a good answer. i just wanted to know what to say to the question i'm sure will come up from a student at some point :)
TimToady it's a good waffle. I don't know if it's a good answer. :)
TreyHarris oh-hoh! 05:30
?eval my $x; $x<moose> = "foo"; $x[0]
05:30 evalbot_r14387 is now known as evalbot_r14388
evalbot_r14388 ("moose" => "foo") 05:30
TreyHarris I didn't know you could do that
audreyt this is "everything autoflatten to 1-seq" 05:31
svnbot6 r14389 | audreyt++ | * Makefile.PL: Detect the case of missing mtl library caused
r14389 | audreyt++ | by installing only the core GHC-6.6 libraries, instead of extra-lib.
audreyt which may or may not be correct
this is because $x[0] is counterpart to both perl5's
($x)[0] 05:32
and
($x)->[0]
fwiw, the 1[0] == 1 behaviour agrees with rub 05:33
TimToady seems to that at one point I was considering that the first has to be written [$x][0]
audreyt namely that 1.to_a there is [1]
instead of an error
spec is unclear about this 05:34
TimToady I guess the question is whether that would mask a significant number of programming mistakes. 05:37
audreyt the same can be asked about list-context autoflattening...
TimToady ?eval 1[0][0][0][0][0][0][0][0][0] 05:38
05:38 evalbot_r14388 is now known as evalbot_r14389
evalbot_r14389 \1 05:38
audreyt ?eval my $x = 1; $$$$$$$x
evalbot_r14389 \1
05:38 cmarcelo left
TreyHarris ?eval my $x = 42; $$$$$$$$x[0][0][0][0][0] 05:39
evalbot_r14389 \42
TimToady ?eval my $x; $x<moose> = "foo"; $x<squirrel> = "bar"; $x[0]
evalbot_r14389 ("moose" => "foo")
TimToady ?eval my $x; $x<moose> = "foo"; $x<squirrel> = "bar"; $x[1] 05:40
evalbot_r14389 ("squirrel" => "bar")
audreyt this is hash-flattens-to-list-of-pairs
TimToady yes, and I wonder how long till people start depending on the hash ordering. :)
audreyt it's unordered :) 05:41
(Judy StrMap got excised)
compare:
irb(main):014:0> {9 => 2, 3 => 4}.to_a[0]
=> [3, 4]
TreyHarris well, it will make algorithms where once you see something you can take anything you've seen before except the one you're looking at now easier... but I don't know how many such algorithms there are... :D
audreyt pugs> {9 => 2, 3 => 4}[0]
("3", 4)
TimToady why isn't that a pair? 05:42
audreyt pugs> WHAT {9 => 2, 3 => 4}[0]
::Pair
just bad prettyprinting notation.
TimToady Bad prettyprinter!! Prettyprinter no cookie!!! 05:43
audreyt oy
TimToady ah well, we can fix it in Perl 7...
audreyt I wonder if the concept of Christmas will still be around by then... 05:44
audreyt tries to not think too much about eschatology and goes back to releng 05:45
svnbot6 r14390 | trey++ | More changelog foo 05:49
audreyt TreyHarris++ # changelog-fu 05:50
05:54 Alias_ joined
svnbot6 r14391 | audreyt++ | * ChangeLog: More -fu. 05:54
Alias_ audreyt: ping? 05:55
audreyt, are you ok if I spam various mongers groups here about sponsoring you over here?
audreyt I'd be honoured 05:56
Alias_ righto
mugwump is spamming NZ too, about getting you an AU and NZ ticket
audreyt indeed without sponsorship of ticket/accomodation I prolly can't get there myself
ooh NZ will rock. 05:57
Alias_ I can offer floorspace in Sydney, with air beds now
So I figure that's covered
mugwump er, 'spamming' isn't quite right, it will be more of an internal discussion here... as the Wellington.pm leader works here
Alias_ ah
06:02 gusdorf joined
svnbot6 r14392 | trey++ | More changelog fixes; making tenses present rather than past 06:06
TreyHarris okay, i think i'm done hacking at the changelog for now. i'll take another swag if there are changes before release and i'm awake 06:07
audreyt good moose 06:09
TreyHarris: 06:20
similar to Perl 5's `goto &sub` 06:21
it's not really similar
as .call returns
bsb Alias_, audreyt: I'm sure something can be sorted for melbourne too
06:22 marmic joined
audreyt cool 06:23
bsb when would this be? (Strategic Data can probably contribute $ again) 06:24
partial funding at least (talking to AC now) 06:25
svnbot6 r14393 | kudra++ | Placeholder, containing the correction 06:27
TreyHarris is it time for another erlangwanderung? 06:30
svnbot6 r14394 | trey++ | Clarified that .call returns, thanks to audreyt++
06:33 dakrone joined 06:44 buetow joined 07:07 iblechbot joined 07:21 zakharyas joined
Alias_ Have we nailed down the semantics for "is" yet in what will be the primary testing lib? 08:11
Any change of making it is_deeply by default is there? :/
08:29 dduncan left
ingy audreyt: ping 08:30
08:31 DHGE joined
svnbot6 r14395 | audreyt++ | * chglog fu! 08:43
r14396 | audreyt++ | * Releng: Metadata adjustments.
08:44 nothingmuch joined
svnbot6 r14397 | audreyt++ | * Pugs.Val: Squash a warning. 08:52
08:52 andara joined
audreyt ingy: pong 08:52
(on taxi now; be back in 7mins) 08:53
ingy audreyt: ok
audreyt: wondered how much you luv .p -> .pm for M::C 08:54
08:54 ruoso joined
audreyt ingy: how does the .p get loaded at first place? 08:58
*in the first place 08:59
ingy back 09:00
09:00 elmex joined
ingy audreyt: you set a PERL5OPT incantation in the author env 09:00
which puts a coderef n @INC
simple eh? 09:01
btw, ingy.net 09:02
lambdabot Title: Ingy dƶt Net
audreyt how do I control .pl and .t? 09:04
a draggable name car? 09:05
*card
ingy it's not fo those
PERL5OPT=-MModule::Compile::Options=p2pm
audreyt it all makes sense.
I'd like to configure extension though.
PERL5OPT=-MModule::Compile::Options=.p,.pm
ingy sure
audreyt or some such 09:06
ingy waves his hands
audreyt how's libyaml btw?
I mean YAML.pm: The Next Generation
ingy it's coming
I've been just a little stalled 09:07
but i got out a new M::C
with use_means_now :)
audreyt cool!
ingy so i can write M::C::Template more easily 09:08
which I'll use to write Compile::Spiffy
audreyt t/compile........NOK 1# Failed test 'use_means_now 09:09
# @@ -1,9 +1,7 @@
# package Foo;
# \# start
# -\# Do it now use Compile::Now 'c', 'd';
# \# middle
# use Data::Dumper;
ingy which i always want to use, and won't be afraid to after M::C::Options ;)
audreyt # \# earth
# -\# Do it now use Compile::Now 'a', 'b';
# \# done
# 1;
# '
# in t/compile.t at line 8.
# Looks like you failed 1 test of 2.
I think you nee to get 0.20 out
ingy !
?
audreyt and then I'll up v6.pm to depend on it (we got burned by heredocs)
ingy I fixed <<= 09:10
audreyt nod. not sure what caused the compile.t failure above though
ingy yeah it passes here 09:11
I think...
audreyt fails for 3cpan testers
search.cpan.org/dist/ already lists FAIL(3)
so I'm hardly alone...
ingy All tests successful.
Files=8, Tests=25, 2 wallclock secs ( 0.67 cusr + 0.15 csys = 0.82 CPU)
suck
audreyt fails even after i install 09:12
ingy I just ran it on another box and it works... 09:13
huh 09:14
audreyt fails on svn checkout too
ingy unset PERL5LIB makes it fail for me
audreyt do you have some Compile::Now module that I don't have?
ingy cool
audreyt uploads pre2
ingy Compile::Now is inline
in the test 09:15
audreyt nod
ingy found it
use XXX
sad
audreyt sigh 09:16
ingy I suck
XXX is so useful though 09:17
maybe I should have M::I put XXX in inc :D 09:18
audreyt sure, as long as it passes tests... 09:19
you can include XXX in inc/
it will get used by tests
using the includes() directive
ingy I was mostly kidding 09:20
M-C-0.20 on cpan 09:21
and svn
audreyt: I was thinking that .pxc should be used for P6 stuff by convention, and .pm generation could be used mre for P5 stuff. but it's not hard and fast by any means 09:24
there are just some modules where using .pmc is going to piss people off, and using .pm is not. 09:25
audreyt: anyway, update v6 please. I have to sleep. 09:28
audreyt okie. thanks!
Please test preflight #2, which will be the real 6.2.13 if there's nothing wrong: 09:34
pugs.blogs.com/pre2/Perl6-Pugs-6.2.13.tar.gz
lambdabot tinyurl.com/ybzmsg
audreyt pugs.blogs.com/pre2/Perl6-Pugs-6.2.13.tar.gz | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com 09:34
audreyt Please test: pugs.blogs.com/pre2/Perl6-Pugs-6.2.13.tar.gz | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com 09:57
audreyt @tell fglock Pugs::Emitter::Perl6::Perl5 depends on Pugs::Runtime::Perl6 which is in PCP6 which depens on PEP5 -- circularity! 10:01
lambdabot Consider it noted.
svnbot6 r14398 | audreyt++ | * Pugs::Compiler::Perl6 - Update Makefile.PL dependency to Module::Compile 0.20. 10:04
r14399 | audreyt++ | * Pugs::Runtime::Perl6AST: Add two "use v5" chunks to prevent miscompilation. 10:13
Patterner is testing preflight #2 now 10:22
audreyt thanks! 10:24
Patterner getting lots of "Expected: Foo" vs. "Actual: Bar" is normal as long it's "ok", right? 10:31
audreyt that's fine 10:33
Patterner i need a faster computer... 10:44
10:50 anatolyv joined 11:00 nothingmuch joined
audreyt win32 smoke all pass. osx/intel all pass. bbiab... 11:02
andara hi all! Thanks all of you for testing runpugs, especially those who managed to crash it :-D 11:04
I'm monitoring now to see what makes it crash, feel free ... 11:05
11:05 chris2 joined 11:08 sanwall joined 11:09 reZo joined 11:14 sanwall left 11:28 weinigZzZ is now known as weinig
audreyt Patterner: how's it going? 11:28
Patterner still testing 11:31
11:32 nipotan is now known as nipotan_DNBK, nipotan_DNBK is now known as nipotan
Patterner @ ext/Perl6-Container-Array now 11:32
audreyt goes writing release notes
11:34 buetow joined
svnbot6 r14400 | fglock++ | v6/emitter(2) - fixed array slice 11:38
11:39 fglock joined 11:46 anatolyv left
Patterner All tests successful, 569 subtests skipped. Files=578, Tests=15849. Gentoo Linux x86_64 11:48
audreyt yay!
Patterner ./pugs -V:pugs_revision returns revision 'unknown' <- normal too? 11:49
"./pugs -V:pugs_revision" => " pugs_revision: 0" 11:50
audreyt that is normal
it's only set on repo versions
11:50 elmex joined 11:58 Yappo joined
audreyt please sanity check html: perlcabal.org/~audreyt/tmp/6_2_13.html 11:58
12:02 Limbic_Region joined, eae joined
masak line 5 column 1 - Warning: <meta> element not empty or not closed 12:06
according to tidy
maybe <meta ... />
would be appropriate in xml
audreyt fixed; reload?
masak now passes 12:07
& # chinese class
audreyt have fun
12:13 iblechbot joined
audreyt anyone with GHC 6.4.1 handy who can give the preflight a try? 12:16
svnbot6 r14401 | fglock++ | v6/emitter(1,2) - fixed Pugs::Runtime::Perl6 circularity 12:20
fglock hi
audreyt hi 12:21
Limbic_Region audreyt - just a sec, I might have kept it around
audreyt kzm on #haskell is helping out
Limbic_Region sorry - got rid of it to avoid confusion
audreyt that's fine 12:22
12:23 Psyche^ joined
audreyt waiting for kzm's build report on 6.4.1 before flying... 12:24
Limbic_Region who is responsible for vanilla/strawberry perl? 12:26
audreyt xdg I think, and irc.perl.org #win32 in general
kolibrie audreyt: I have 6.4.2
Limbic_Region err, I knew xdg but not his real name
kolibrie do you need that?
Limbic_Region Dan Golden apparently
audreyt kolibrie: sure, a 6.4.2 testflight on pre2 will be useful 12:27
just "make fast" and make sure it produces something
kolibrie ok
Psyche^ i'm installing 6.4.1 now, but that could take a few minutes... 12:32
audreyt Psyche^: kzm confirmed it works with 6.4.1
on linux
Psyche^ 64bit? 12:33
audreyt not 64bit. so will help
to have 64bit tested.
Psyche^ using the ghc-bin version, so it is minutes, not hours :)
audreyt :) and "make fast" will help as well 12:34
12:34 Psyche^ is now known as Patterner, ofer1 joined
fglock what is the signature for a subroutine without an argument list? sub {...} - I'm wondering if it depends on the context, eg. inside a map() 12:34
audreyt no
it's always *@_
a bare block however is :($_ is rw) 12:35
fglock does it slurp the entire mapped list?
ok
audreyt but implicit params $^a etc changes that
fglock so sub{...} would slurp the map() list 12:36
12:36 HiTech69 joined
audreyt that depends on how map dispatches 12:36
currently pugs's &map asks for mandatory arity
and use that (min 1) for looping 12:37
so sub{} still only gets one per run
kolibrie audreyt: pre2 compiles to something that runs on linux ghc 6.4.2
fglock right - so the decision is left to map() implementation 12:38
audreyt yup
ok. I think we're ready to fly.
rgs woot
svnbot6 r14402 | audreyt++ | * This be Pugs, version 6.2.13.
r14401 | fglock++ | v6/emitter(1,2) - fixed Pugs::Runtime::Perl6 circularity
audreyt uploading to CPAN... 12:39
Limbic_Region wooooooooot 12:40
grats audreyt++
kolibrie good work everyone
fglock ?eval (sub(*@_){...}).arity # is this supposed to return Inf or Whatever? 12:41
12:41 evalbot_r14389 is now known as evalbot_r14401
evalbot_r14401 1 12:41
fglock audreyt++
audreyt :) 12:42
thanks everyone :)
fglock meeting & 12:44
audreyt pugs.blogs.com/dist/Perl6-Pugs-6.2.13.tar.gz released! | run.pugscode.org | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com 12:46
Patterner "make fast" for 6.4.1 successful 12:51
audreyt cool! 12:56
Patterner tests are successful so far 12:58
audreyt announced on pugs.blogs.com 13:00
Limbic_Region audreyt - did you want to link to the complete change log? 13:01
audreyt it's pugs.blogs.com/pugs/2006/10/pugs_62....html#more
lambdabot tinyurl.com/ylb2zb 13:02
Limbic_Region it isn't even Monday 13:03
thanks
audreyt it's tuesday 9pm for me :)
Limbic_Region audreyt - at what point will 6.6 become mandatory (for those use pugs in between releases) 13:08
audreyt Limbic_Region: I don't know :) likely during my stay at Brazil, may be sooner if gaal feels like it 13:09
so in 3 weeks or so?
rafl Are there any known issues on Debian with ghc 6.6? I just upgraded from 6.4 and the release doesn't build anymore because of a missing package: mtl.
audreyt rafl: you need extralibs 13:10
not sure if it's separately packaged iwth debian or not
the linux bindist includes both
Limbic_Region ok - looking forward to it. I have really been avoiding perl 6 objects but since my recent foray into Java land - I am looking forward to it 13:11
audreyt smiles
announced on p6[acu]
er, I mean p6[aclu]
rafl audreyt: That's what the error message told me as well, bug I could not find any package containing that.
audreyt rafl: then talk to Igloo on #haskell / # Ghc
#ghc
Limbic_Region wanders off 13:12
rafl audreyt: I already did so. He didn't reply so far, so I decided to ask here as well :-)
audreyt and I may have to sleep soon as well
whew :) my meatspace flight is tomorrow...
and I havn't packed my, uhm, tarball.
audreyt goes packing and sleeping, not neccessary in that order... 13:13
see y'all tomorrow :) *wave* & 13:14
13:14 [particle] joined
audreyt rafl: btw we now depend on libperl-dev, if it's not already a dep 13:14
rafl audreyt: It already is.
audreyt cool! 13:15
rafl audreyt: Sleep well.
audreyt thanks :) *thunk* 13:17
13:28 kanru joined 13:31 weinig is now known as weinig|bbl 13:54 chris2 joined
nothingmuch mooo 14:03
14:06 vel joined 14:09 hexmode joined, cjeris joined
svnbot6 r14403 | andara++ | [runpugs] 14:18
r14403 | andara++ | -now limits generated output (to 256 lines)
r14403 | andara++ | -runpugs is now a semi-self(re)starter
r14403 | andara++ | -added some minimal logging
14:29 andara_ joined, ZeroCool joined 14:35 Pomin joined 14:41 jojo joined 15:01 BooK joined 15:05 andara left 15:08 ZeroCool left, andara_ joined 15:10 baest joined 15:11 andara_ joined 15:12 andara_ joined
Juerd audreyt et al: congrats with the new release! 15:14
15:15 andara_ joined 15:22 andara_ joined 15:28 andara_ joined 15:29 andara_ left
rafl Juerd: >>> sudoers file: syntax error, line 24 <<< 15:33
TimToady Juerd: me too. :) 15:47
15:57 amnesiac joined
Juerd Fixed 15:57
15:57 buetow joined
rafl Juerd: Thanks. 16:04
16:05 weinig|bbl is now known as weinig
rafl Juerd: Do you regularly do updates on feather? 16:05
16:13 asz joined
Juerd rafl: Semi 16:24
svnbot6 r14404 | lanny++ | [docs/Perl6/Makefile.PL]
r14404 | lanny++ | * Work around a bug in ExtUtils::MM_Any::manifypod_target()
r14404 | lanny++ | * Hopefully this will fix nmake-1.5 issues on Win32.
r14404 | lanny++ | * Need to report this back to ExtUtils.
Juerd rafl: I do it manually, approx. once per month, I think 16:25
rafl Juerd: OK. FYI, I just updated the available list, upgraded pbuilder and installed cowdancer.
Juerd Thank you 16:26
What, er, is a cowdancer?
rafl Juerd: A weird Copy-on-write directory tree utility. Used by cowbuilder, which is something like pbuilder, but it uses COW directories instead of a huge base tarball which needs to be unpacked every time. 16:28
Juerd: Also, you don't want to upgrade to ghc6 (6.6) in sid. It doesn't include the extralibs stuff yet, so you won't be able to build pugs as it needs mtl (and probably more) from extralibs. 16:29
Oh, there's another ghc installation in /usr/local. Don't mind then.. 16:31
svnbot6 r14405 | fglock++ | v6/emitter(2) - "auto-unbox" where possible 16:43
16:48 TimToady_ joined 16:49 weinig is now known as weinig|bbl 16:55 andara joined 17:00 Daveman joined 17:01 |OUF|HiTech69 joined 17:08 ofer1 joined 17:13 kanru joined, andara left 17:17 TimToady_ is now known as TimToady, snowstalker joined
TimToady Juerd: may I please be a sudoer? 17:28
17:39 Captain_Fourier joined
Captain_Fourier hey 17:39
runpugs seems down 17:40
Limbic_Region TimToady - I have found Juerd to have more timely responses to emails than to channel chatter
?seen andara 17:42
lambdabot I saw andara leaving #perl6 28m 20s ago, and .
Limbic_Region Captain_Fourier - I believe andara is responsible for runpugs
perlbot andara
and apparently there isn't a factoid for further info
TimToady andara's connection has been flapping in the breeze.
Captain_Fourier well I see there is source code for runpugs 17:43
hmm lets see what I can do
TimToady there is, but of course by the time you set it up to run yourself you can just run pugs in a regular terminal. :)
might be worth having a bevy of runpugses scattered around the world though. 17:44
Captain_Fourier i have run pugs in terminal
but not runpugs!
17:51 kanru joined 17:52 elmex joined 18:02 justatheory joined 18:07 b_jonas joined 18:13 babusri joined 18:22 babusri joined, HiTech69 joined 18:24 babusri joined
svnbot6 r14406 | fglock++ | v6/emitter(2) - fixed some scalar dereferencing problems 18:24
18:25 babusri joined 18:29 eae left 18:33 andara joined
andara TimToady: that seems like a good idea, as it's usually the shell ulimits that kill runpugs now. I don't think feather could run more than 50 sessions anyway. 18:37
18:44 lisppaste3 joined
svnbot6 r14407 | andara++ | [runpugs] 18:44
r14407 | andara++ | -increased listen() queue
r14407 | andara++ | -turned a few die's into non-fatal errors
18:48 penk joined, araujo joined 18:53 reZo joined 19:00 babusri joined 19:03 ubuntu_ joined 19:08 babusri joined 19:11 buetow joined 19:15 babusri joined 19:21 babusri joined 19:26 theorbtwo joined 19:35 polettix joined 19:42 weinig|bbl is now known as weinig
svnbot6 r14408 | lanny++ | [docs/Perl6/Makefile.PL] 19:42
r14408 | lanny++ | * Back out previous attempt fix.
r14408 | lanny++ | * Put in a real fix (tested) for nmake-1.5 issues on Win32.
r14408 | lanny++ | * nmake version 1.5 should be able to run the Makefile produced now
19:42 babusri joined 19:52 lanny joined
lanny On Win32 during 'make smoke' I get that 'pugs -V' isn't reporting the revision. Any ideas on how to fix? 19:53
Must run so if you know please @tell me. 19:54
[particle] @tell lanny i think it's a subversion 1.4.0 problem... is that the version you're running? 19:55
lambdabot Consider it noted.
ingy seen rindolf 19:56
jabbot ingy: rindolf was seen 3 hours 1 minutes 46 seconds ago
19:56 babusri joined 19:58 andara left 20:00 DaGo joined 20:06 nothingmuch joined 20:12 amnesiac joined
Captain_Fourier how do I pull from the repository 20:13
20:14 penk joined 20:32 lanny joined
lanny [particle]: Yes that is the subversion I'm using. Do you know if svk will fix it? I have that on my box but didn't want to set up a mirror and the push/pull. 20:33
lambdabot lanny: You have 1 new message. '/msg lambdabot @messages' to read it.
lanny @messages
lambdabot [particle] said 38m 3s ago: i think it's a subversion 1.4.0 problem... is that the version you're running?
clkao lanny: ? 20:34
[particle] i dont use svk, so i don't know
clkao: can't get pugs revision info with svn 1.4.0
lanny I think he was gawking at my fear of push/pull. :)
[particle] it could be a problem with the perl code that parses it, i suppose 20:35
lanny Now 'svn info' shows 14408 as expected so I don't know why Pugs can't figure it out 20:36
[particle] ok, the problem is that pugs scans for .svn/entries, and the working copy format changed for 1.4.0 20:40
lanny [particle]++
[particle] i guess instead it should run C<svn --version> and parse output 20:41
SamB ReactOS had the same problem...
[particle] i don't have an older svn version to test with, only 1.4.0
SamB if you do that it should work fine with older versions...
lanny [particle]: That would be saner 20:42
svnbot6 r14409 | fglock++ | v6/emitter(2) - range elements are Typed
[particle] well, i don't know if the svn --version output is a different format 20:43
SamB the people on #svn gave me the impression it would be fine
and that only the internal files you weren't supposed to be reading anyway had changed format
lanny Well it should probably use 'svn info' to figure out the revision
SamB well, whatever "svn foo" it is, that is 20:44
[particle] right, duh, info, not --version
SamB should work fine ;-) 20:45
lanny looks like info is 'key: value' and I doubt that would change soon
SamB they definately gave me the impression that it had not changed recently and was not going to change anytime soon, either ;-) 20:46
lanny Bah. Developers. Can't trust em like that. They'll bite you ever time!
[particle] updates his pugs working copy
SamB well, another option would be to call the SVN API... 20:48
lanny Too funny. The code is there in util/version_h.pl to read from 'svn info' but it prefers to read from $base/.svn/entries.
[particle] samb: does that require any client modulesL
?
SamB lanny: and it doesn't deal with the unreadability thereof?
[particle] well, it'll read svk info, but not svn 20:49
SamB you need the SVN library...
20:49 fglock left
lanny Working it. It looks like if it can read .svn/entries that is preferred 'elsif ( -r $svn_entries ) { ...' 20:49
SamB which the "svn" command needs too
it should also check if it can *parse* that
lanny oh. Good catch, [particle]. I'm starting to see what I want instead of what is there.
SamB heh 20:50
[particle] cruft--
lanny++ 20:51
svnbot6 r14410 | lanny++ | * util/version_h.pl - prefer `svn info` to .svn/entries (which is broke for svn 1.4) 21:12
21:12 mauke joined 21:14 explorer joined
lanny karma audreyt 21:15
jabbot lanny: audreyt has karma of 2550
lanny karma lanny
jabbot lanny: lanny has karma of 92
lanny Oh. Hmm. That's not nearly as funny as I thought it was gonna be. I was planning on making a joke about being a centi-audreyt. 21:16
oh, good. Still works. :) 3.6 centi-audreyt 21:17
Patterner Hat.Hat.Hat. 21:19
lanny Must run. Thanks again, [particle], for the version troubleshooting.
[particle] np
21:25 weinig is now known as weinig|away 21:34 Limbic_Region joined
nothingmuch croaks 21:36
good mangos as big as house cats are dangerous things
Patterner Killer Mangos From Outer Space?
nothingmuch something like that 21:37
They Came From The Supermarket
So I Married a Fruit
Juerd TimToady: Yes, you may 21:38
nothingmuch @moosages
lambdabot Limbic_Region said 1d 2h 34m 7s ago: no, my memory has to be added in pairs but I really do appreciate it
nothingmuch lambdabot:++
21:38 MacVince joined
Juerd TimToady: You now have sudo access. Please use debian packages where possible, and keep me informed of what you do. 21:38
Limbic_Region: My response, chat versus mail, depends heavily on when and where I read things. Hard to predict :) 21:39
TimToady thanks! just wanted to fix a typo this time. 21:41
21:42 larsen joined
Juerd Maybe next time you need it for something more intrusive. If so, I'd like to know :) 21:42
TimToady: Oh, sorry, put you in the file as timtoady, not larry A:) 21:59
s/A//
TimToady: Try again
Juerd shouldn't be using root shells while almost falling asleep
Juerd yawns again. 22:02
Good night
TimToady night, and thanks. 22:07
22:14 cognominal joined 22:51 weinig|away is now known as weinig 22:54 buubot joined
TimToady is pugs known not to compile under cygwin? 23:00
(specifically, using cygwin perl5) 23:02
when configuring I get: runcompiler.exe: /home/lwall/pugs/third-party/installed/packages.conf: openFile: does not exist (No such file or directory) 23:08
23:09 dakrone joined 23:23 mg joined 23:24 Alias_ joined 23:26 cognominal joined
mg hi all. with the release of 6.12.13, i thought i might try to help out a bit (a very little bit). i have fixed some broken smartlinks and updated some from the /regex/ style. 23:28
should i submit a patch? or can i get a svn commit bit? 23:29
TimToady you may have an svn commit bit if you /msg your email addr to me (or put it in disguised form here so the spammers don't get it) 23:32
Alias_ So YAPC::AU now has Damian, Randal, cog and (hopefully) audreyt 23:33
All we need now is a language author and we'll have the complete set
Alias_ stares at TimeToady
s/Time/Tim/
23:34 weinig is now known as weinig|food
TimToady mg: have an email addr? 23:38
mg mgrimes at cpan org 23:43
TimToady an invite is on its way. 23:44
mg thanks
TimToady after you get it, it is customary to add yourself to AUTHORS as your first checkin. Anything after that is up to you. :P)
:) rather 23:45
23:45 mdiep joined
TimToady :P) looks like a bugger hanging from my nose... 23:45
welcome aboard, and have fun! 23:46
mg anything to get p6 here soon!
TimToady anything? 23:47
mg :-) well... anything that fits my limited skills and free time... 23:49
Alias_ What are you guys managing the repository with these days?
TimToady as in openfoundry kind of "what"?
or as in svn kind of "what"?
and some of us are girls. 23:50
Alias_ I was using the royal "you guys"
But yes, I was refering to openfoundry
TimToady you are, in fact, an admin on the project. :) 23:54
avar TimToady: Whatever the final "official" Perl 6 parser will be please make it as open a project as pugs:) 23:55
Alias_ TimToady, apparently Audrey was even more generous handing out the bits in the early days of the project :)
avar Alias_: I have a commit bit because I joined the channel one day and said hi or something:) 23:56
TimToady there is no "official" parser, only the official test suite. the first parser that passes the test suite is likely to come from the pugs project.
Alias_ avar: I have an admin bit because I donated some Perl 5 modules for conversion to Perl 6 some times back in the dark mist of the first few months 23:57
(apparently)
TimToady you're an admin because we've met you and Know Who You Are in case you misbehave...
Alias_ true :) 23:58
TimToady, whatever this thing is that gets released for the current Christmas, are you planning on actually slapping some sort of "Perl 6 alpha" or "Perl 6 developer preview" on it?
(please add all the usual caveats to that question
(
)
TimToady you can already say "use v6-alpha;" in your P5 code. :) 23:59

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!