Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2006-05-01

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:09 mako132_ joined perl6
00:10 justatheory joined perl6
00:30 rbysamppi Can you method-call a multisub? multi f (Int $x: Int $z) { ... }; f 2: 3
00:33 audreyt yes, but that's a fallback
00:33 audreyt if method dispatch fails
00:33 audreyt it pretends you did
00:33 audreyt f(2, 3)
00:33 audreyt instead
00:33 audreyt i.e. if 2.f() can't be found
00:33 audreyt (bbiab)
00:34 Arathorn hum
00:34 Arathorn prelude compiling seems to be broken - on the range operator
00:34 Arathorn in push @rv, (~ $start)..(~ $end);
00:35 Arathorn (line 412 of src/perl6/Prelude.pm)
00:35 * Arathorn investigates as best he can
00:41 macroron__ joined perl6
00:45 dvorak joined perl6
00:49 Arathorn whee, this is rather fun
00:50 Arathorn losing the char '.' >> notFollowedBy (char '.') in #10167 is the kicker...
01:09 hexmode joined perl6
01:13 FurnaceBoy_ joined perl6
01:17 Arathorn bah, i don't understand parsec well enough
01:17 Arathorn (assuming this is parsec)
01:38 aufrank Arathorn++ just for chasing it down :)
01:44 ruporama joined perl6
01:47 oylenshpeegul left perl6
01:48 dvorak joined perl6
01:52 schneiker joined perl6
01:53 weinig|away is now known as weinig|sleep
01:56 Arathorn Arathorn-- for not having a clue what he's doing, though
01:57 * Arathorn defers in favour of playing with Judy, which has the advantage of being written in a language i speak :)
01:57 aufrank have you looked at the "Build a Scheme in 48 hours" tutorial?
01:57 aufrank it has a good parsec crash course
01:58 * Arathorn will - although tomorrow, at this rate
02:03 dvorak joined perl6
02:05 Khisanth joined perl6
02:11 ruporama left perl6
02:28 shachaf joined perl6
02:35 Arathorn is now known as Arasleeps
02:41 schneiker I'd like to help with Perl6, but don't have much spare time.
02:41 schneiker In the course of searching for scattered Perl6 info,
02:41 schneiker I thought about starting a "perl.perl6.using"
02:41 schneiker group on Google Groups, with the following focus:
02:41 schneiker "Information exchange for using Perl 6.
02:41 schneiker The most useful info and resource links are added to a
02:41 schneiker weekly-updated "Using Perl 6 FAQ" posting. Topics include:
02:41 schneiker getting started with Perl 6, incremental migration from Perl 5,
02:41 schneiker developing Perl 6 tools, what early adapters are doing, etc.
02:41 schneiker Apparently I cannot do this directly through Google Groups.
02:41 schneiker I get this error message: "This group name is not suitable.
02:41 schneiker You cannot start a group name to be similar to a Usenet group."
02:41 schneiker However my newserver doesn’t show "perl.perl6.*" as part of
02:41 schneiker the Usenet newsgroup hierarchy (versus just in nttp.perl.org)
02:41 schneiker My search efforts haven't turned up any helpful info.
02:41 schneiker Information, feedback welcome.
02:49 sanxiyn joined perl6
02:49 FurnaceBoy_ 'similar' though
02:49 FurnaceBoy_ who knows what their actual rules are. Try not using  a '.'
02:49 FurnaceBoy_ just "Perl6 Users"
02:49 FurnaceBoy_ should be okay?
02:50 sanxiyn Interesting to see Pugs relicensing. I use this license myself:
02:50 sanxiyn http://sparcs.kaist.ac.kr/~tinuviel/fepy/COPYING
02:50 FurnaceBoy_ a.k.a. BSD
02:50 FurnaceBoy_ heh
02:50 * FurnaceBoy_ is only joking
02:50 FurnaceBoy_ you could just say public domain
02:51 sanxiyn FurnaceBoy_: It's more fun this way.
02:51 FurnaceBoy_ more swearing involved, if that's your  thing
02:51 sanxiyn Official homepage of WTFPL is here: http://sam.zoy.org/wtfpl/
02:51 FurnaceBoy_ I came across this one recently:
02:52 FurnaceBoy_ * --------------------------------------​--------------------------------------
02:52 FurnaceBoy_ * "THE BEER-WARE LICENSE" (Revision 42):
02:52 FurnaceBoy_ * <phk@login.dknet.dk> wrote this file.  As long as you retain this notice you
02:52 FurnaceBoy_ * can do whatever you want with this stuff. If we meet some day, and you think
02:52 FurnaceBoy_ * this stuff is worth it, you can buy me a beer in return.   Poul-Henning Kamp
02:52 FurnaceBoy_ * --------------------------------------​--------------------------------------
02:52 schneiker "Perl6 Users" should be OK, but I'd like to see it grouped with other "perl.perl6.*"
02:52 FurnaceBoy_ impossible schneiker.
02:52 FurnaceBoy_ you could, however, propose an addition to Usenet.
02:52 FurnaceBoy_ that would be about 100x more difficult however.
02:52 FurnaceBoy_ Usenet is a tree owned by root and not world writable;-)
02:53 FurnaceBoy_ and I think the rules are there to prevent exactly what you propose
02:53 FurnaceBoy_ if you want to augment Usenet, go through that procedure, but it will take longer
02:53 sanxiyn FurnaceBoy_: alt.* hierarchy is easier to get in.
02:53 FurnaceBoy_ true.
02:54 FurnaceBoy_ but that wouldn't have the ring of officiality that schneiker seems to be pursuing.
02:54 sanxiyn FurnaceBoy_: Big 8 needs voting.
02:54 * FurnaceBoy_ notes 'officiality' isn't a word. sorry.
02:54 FurnaceBoy_ yes, I know.
02:54 FurnaceBoy_ depends on whether he wants quick'n'easy, or Usenet :)
02:55 sanxiyn One can read http://www.eyrie.org/~eagle/faqs/big-eight.html and start the process.
02:55 schneiker I got Usenet to add comp.lang.ruby, years ago, but it took several months.
02:55 schneiker However perl.perl6.* does not appear to be part of actual Usenet hierarchy.
02:55 schneiker That's why I was puzzled.
02:55 FurnaceBoy_ whatever way you cut it, a random Google Group is easier. which is why there are a zillion of them.
02:55 sanxiyn File RFD to news.announce.newsgroups...
02:55 FurnaceBoy_ find its owners schneiker
02:55 FurnaceBoy_ but I warn you, this way will take work and commitment
02:56 FurnaceBoy_ compared to opening a random GG
02:56 FurnaceBoy_ the result may be worth it ...
02:56 FurnaceBoy_ are there no perl6 groups in perl.* ?
02:56 * FurnaceBoy_ goes looking
02:56 sanxiyn I still prefer Yahoo! Groups to Google Groups actually.
02:56 * FurnaceBoy_ only uses Usenet
02:57 sanxiyn FurnaceBoy_: I also use Gmane.
02:57 FurnaceBoy_ ck
02:57 FurnaceBoy_ worst web interface ever.
02:57 FurnaceBoy_ :_)
02:57 sanxiyn FurnaceBoy_: Well, it has NNTP too.
02:57 * FurnaceBoy_ thanks Allah
02:57 FurnaceBoy_ I can;t believe nobody has called them on it
02:57 FurnaceBoy_ in all these years
02:57 FurnaceBoy_ maybe all the sane people just use NNTP
02:58 Quell joined perl6
02:58 sanxiyn FurnaceBoy_: I only use search.gmane.org from the web interface.
02:58 FurnaceBoy_ schneiker, yeah, there seems an opening for a 'general' group under perl.perl6.*
02:58 sanxiyn (Which I found to work better than Google for some cases.)
02:58 FurnaceBoy_ you could probably find a way to propose one.
02:59 aufrank I'm pretty sure Larry mentioned something about being to able to help in getting that process off the ground
02:59 FurnaceBoy_ perl.perl6.using doesn't seem to have any precedent though... I mean 'using'
02:59 FurnaceBoy_ there must be a better word
02:59 aufrank but I could be imagining things
02:59 FurnaceBoy_ yes, TimToady could surely help
02:59 FurnaceBoy_ if anyone can
02:59 FurnaceBoy_ schneiker, pounce on TimToady about this
02:59 FurnaceBoy_ but I suggest a better word than 'using'
02:59 FurnaceBoy_ general has precedents in Usenet
03:01 schneiker FurnaceBoy: OK. I like general better.
03:05 schneiker FurnaceBoy: Recommended means to "pounce on TimToady about this"?
03:05 schneiker (I'm brand new to IRC, although I have been skimming logs for a while.)
03:07 hcarty joined perl6
03:08 hcarty left perl6
03:09 FurnaceBoy_ schneiker, you can ping him now, if he's asleep, wait till he's around here. ?
03:16 schneiker Thanks.
03:16 schneiker /quit
03:23 aufrank ok, I'm off for the night now too
03:23 aufrank bye all
03:23 q[uri] joined perl6
03:35 wilx`` joined perl6
03:54 coumbes joined perl6
04:34 spinclad Arathorn: i've been pondering looking under the hood of JudyHS, seeing as it's described as implemented through Judy1 and JudyL
04:36 spinclad it seems the author doesn't see a natural or useful (or stable?) order to the keys, so doesn't provide iterators
04:38 spinclad but if one needs to iterate, in an arbitrary order, a hash order should work
04:38 spinclad so perhaps we may have something to contribute back to him...
04:40 spinclad otoh, Judy1 + JudyHS may be all we need, i haven't seen through this to the other side yet
04:51 H5N1 joined perl6
04:58 innit joined perl6
04:58 innit left perl6
05:20 gaal joined perl6
06:08 iblechbot joined perl6
06:33 cmarcelo joined perl6
06:54 Aankhen`` joined perl6
06:59 ko1_away is now known as ko1
07:10 ko1 is now known as ko1_away
07:36 q[uri]_ joined perl6
07:50 xinming joined perl6
08:12 elmex joined perl6
08:22 chris2 joined perl6
08:30 FurnaceBoy_ joined perl6
08:30 bernhard joined perl6
08:30 larsen joined perl6
08:33 svnbot6 r10170 | audreyt++ | * Massive parsing cleanup.
08:33 svnbot6 r10170 | audreyt++ | * Newly implemented parses:
08:33 svnbot6 r10170 | audreyt++ |     @<foo> @123
08:33 svnbot6 r10170 | audreyt++ |     %<foo> %123
08:33 svnbot6 r10170 | audreyt++ |     map {...}, 1, 2, 3
08:33 svnbot6 r10170 | audreyt++ |     map {...} 1, 2, 3   # is this form still kosher?
08:33 svnbot6 r10170 | audreyt++ |     "$var\  .moose()"
08:33 svnbot6 r10170 | audreyt++ | * Fixed parsefails:
08:33 svnbot6 r10170 | audreyt++ |     1..2
08:33 svnbot6 r10170 | audreyt++ |     "$var."
08:33 svnbot6 r10170 | audreyt++ |     "$var.moose"
08:33 svnbot6 r10170 | audreyt++ |     rule foo { ... \{ ... }
08:35 gaal what's @<foo> @123  ?
08:35 audreyt short for @($<foo>) @($123)
08:37 gaal what's that mean?
08:37 gaal I got as far as @($<foo>)
08:38 audreyt @($<foo>) means @($/.{'foo'})
08:38 audreyt @($123) means @($/.[123])
08:38 audreyt i.e. the 123rd capture
08:39 gaal okay, so what do the two mean juxtaposed?
08:40 audreyt they mean that I was too lazy to type \n
08:41 audreyt TimToady: is this form still kosher?
08:41 audreyt    map {$_+3} 1,2,3,4
08:41 gaal thanks :)
08:41 audreyt np, sorry :)
08:41 audreyt TimToady: the only mention is in S03
08:41 audreyt      @result = map { floor($^x / 2) }
08:41 audreyt                  grep { /^ \d+ $/ }
08:41 audreyt                    @data;
08:42 audreyt if it's not a typo, then it needs to be clearly documented that whitespace-follows-block-literal means comma
08:42 audreyt because you've specced that non-comma-follows-block-literal will no longer be arglist
08:42 audreyt (S12, "it will be considered to be the final argument")
08:42 audreyt so
08:43 audreyt map {moose}.blah # map({moose}).blah
08:43 audreyt map {moose} .blah # map({moose}, $_.blah)
08:43 audreyt map {moose}, .blah # map({moose}, $_.blah)
08:43 audreyt is the current pugs behaviour
08:43 audreyt and I wonder how correct it is
08:44 audreyt but I need to go take a nap. been in $hospital.bureaucracy for an entire day...
08:44 audreyt bbiab &
09:07 drbean joined perl6
09:09 dakkar joined perl6
09:11 dakkar is it possible that there are still problems re: Cabal and fps?
09:12 dakkar here (ghc 6.4.1, Cabal 1.1.3) I can't compile pugs (fresh checkout, svn head)
09:13 dakkar build_pugs.pl calls (I think) Setup, who then says «Setup: cannot satisfy dependency fps-any
09:13 dakkar Setup: error reading ./.setup-config; run "setup configure" command?»
09:13 dakkar «ghc-pkg list» shows the fps library...
09:53 KingDiamond joined perl6
10:05 Juerd audreyt: I am glad we discussed the long dot on p6l, \. is much better than . .
10:06 theorbtwo It's even somewhat better then .:.
10:07 theorbtwo So it's now \\(<ws>|_)+\. ?
10:07 theorbtwo I'm not sure now if the _ idea was accepted or not.
10:10 dakkar am I the only one unable to compile pugs at the moment? (see above)
10:12 kanru joined perl6
10:20 Arasleeps is now known as Arathorn
10:20 Arathorn morning folks
10:21 dakkar Arathorn: what was the error you got while building pugs, yesterday?
10:21 dakkar just «can't find Data.ByteString», or «can't read .setup-config»?
10:24 Arathorn i had three - the .. parsefail which audreyt corrected in the night, FPS not being handled by Cabal correctly(?) (which audrey fixed at the time), and a "No instance for (Typeable1 Tree)" at src/Pugs/AST/Internals/Instances.hs:138, which i fixed by swiching to ghc 6.4.1 rather than ghc-cvs in debianland
10:24 cmarcelo left perl6
10:25 dakkar yesterday I got «can't find Data.ByteString», today I get «Setup: error reading ./.setup-config; run "setup configure" command?»
10:25 dakkar of course there is no .setup-config file...
10:27 dakkar fixed
10:27 dakkar argh
10:27 dakkar ./Setup configure --user
10:30 Arathorn you shouldn't need to do that, though...
10:31 trampel joined perl6
10:33 dakkar Arathorn: you tell it ;-) it's compiling now
10:34 Arathorn hehe
10:35 dakkar I just can't figure what I should change in the Makefile.PL to make it work without running Setup manually...
10:38 dakkar ok, now that it is compiled, I'll check the parsefails due to whitespace changes and fix the tests
11:15 dakkar ?eval my $path="abc";$path ~~ s:perl5[^([a-z]:)][{uc$0}];$path;
11:15 evalbot_10165 is now known as evalbot_10170
11:15 evalbot_10170 Error:  unexpected end of input expecting block, "\\", "$/", "$!" or ":"
11:16 dakkar parsefail?
11:16 dakkar (the code is from File::Spec::Win32)
11:19 coumbes left perl6
11:20 dakkar ?eval my $path="abc";$path ~~ s:perl5[^a][A];$path;
11:20 evalbot_10170 Error: No such sub: "&a"
11:20 dakkar what?
11:20 dakkar ?eval my $path="abc";$path ~~ s:perl5/^a/A/;$path;
11:20 evalbot_10170 \"Abc"
11:21 dakkar ?eval my $path="abc";$path ~~ s:perl5 [^a][A];$path;
11:21 evalbot_10170 \"Abc"
11:21 dakkar oh, now I get it
11:26 iblechbot joined perl6
11:27 dakkar ?eval my $path="abc";$path ~~ s:perl5 [^([a-z]:)][{uc$0}];$path;
11:27 evalbot_10170 Error:  unexpected end of input expecting block, "\\", "$/", "$!" or ":"
11:27 dakkar ?eval my $path="abc";$path ~~ s:perl5 [^([a-z]:)][{uc $0}];$path;
11:27 evalbot_10170 Error:  unexpected end of input expecting block, "\\", "$/", "$!" or ":"
11:27 dakkar what's going on?
11:36 mako132_ joined perl6
11:37 dakkar there's something very wrong in the pugs parser to parse regex...
11:38 clkao dakkar!
11:39 dakkar clkao!
11:39 dakkar I'm trying to fix the tests, but I found some strangeness in the rx parser (see above)
11:40 dakkar did the syntax of 'require' change recently?
11:41 dakkar 'Junction of Something' does not parse («unexpected "o", expecting comment or variable name»)
11:42 dakkar I should write down these problems...
11:52 clkao \. is ugly!
11:52 Arathorn just as ugly as .:, imo
11:53 Arathorn especially as the : looked attached to the term it prefixed, rather than part of a longdot
11:53 dakkar this whole whitespace-sensitivity thing doesn't feel very DWIM to me...
11:53 audreyt dakkar: fixing
11:53 dakkar audreyt: which one?
11:54 audreyt s[][]
11:54 audreyt s!!! works
11:54 audreyt as does s////
11:54 audreyt but s[][] not
11:54 audreyt a sec
11:54 audreyt sorry for breaking it
11:54 dakkar oh, thanks
11:54 Arathorn the white-space sensitivity stuff feels amazingly unDWIM to me too :(
11:55 dakkar audreyt: s{}{} has the same problem, btw
11:55 audreyt sure
11:55 Arathorn we sacrifice compulsory brackets at the expense of our immortal souls...
11:55 dvorak joined perl6
12:03 dakkar is there a more recent smoke table than the one audreyt linked to?
12:07 audreyt yes
12:07 audreyt see smoke.pugscode.org
12:08 audreyt http://m19s28.vlinux.de/iblech/stuff/pugs-sm​okes/pugs-smoke-6.2.11-r10148-freebsd-normal​--1146485254-3339--10735-10523-212-1348-1610​-4--5f15fff81963ebf5af2fcbad72ca4201.html
12:09 dakkar got it
12:11 audreyt s[][] fixed
12:11 audreyt committing
12:11 szbalint joined perl6
12:14 kolibrie joined perl6
12:19 svnbot6 r10171 | audreyt++ | * Parser: Allow nested begin/end pairs in regex body again:
12:19 svnbot6 r10171 | audreyt++ |     rx[ [a-z] ]
12:19 svnbot6 r10171 | audreyt++ |   however, the parser cannot currently distinguish between:
12:19 svnbot6 r10171 | audreyt++ |     rx[ \[a-z]      # should be valid
12:19 svnbot6 r10171 | audreyt++ |     rx[ \\[a-z] ]   # should be valid as well
12:19 svnbot6 r10171 | audreyt++ |   as the escape codes are passed thru to the rules engine
12:19 svnbot6 r10171 | audreyt++ |   without further manipulation.  Thinking about how to solve this...
12:21 xerox_ is now known as xerox
12:21 r0nny joined perl6
12:21 Juerd [mac mini]--  # sometimes refuses to eject, then requires reboot.
12:21 Juerd Sometimes being twice in 12 cd's alreday.
12:21 Juerd already.
12:22 Arathorn hope you have enough paperclips handy
12:22 Juerd Arathorn: No such emergency eject thingy.
12:23 Juerd audreyt: Well, \w characters that are escaped are special to the engine, while \W characters are special to the engine when NOT escaped.
12:23 Juerd audreyt: Also, \w characters are no longer valid delimiters
12:23 audreyt ah. point.
12:25 Juerd But I think there's, ultimately, no other solution than to have the full rule grammar embedded
12:25 Juerd 'cause there's also rx<"...]...">]
12:25 Juerd s/</<[/
12:25 Juerd (That was a p5 regex ;))
12:26 dakkar pugs's build system is fragile... but I don't know enough to change it...
12:26 dakkar «Wasn't able to find 'Syck_stub.o', aborting»
12:27 audreyt oh no, not that again
12:27 audreyt which ghc version? is Syck_stub anywhere in dist/?
12:27 audreyt (it's really Cabal 1.0 that is very fragile.)
12:27 dakkar ghc 6.4.1, Cabal 1.1.3
12:28 audreyt yay, another different point version :/
12:28 audreyt find dist |grep Syck_
12:28 dakkar and no, no stub anywhere in sight
12:28 audreyt ok. I think I can fix it.
12:28 dakkar it's still in src/
12:28 audreyt a sec
12:29 dakkar should I copy it?
12:29 audreyt well, to get it working, sure
12:29 audreyt I'll have a better (and saner) fix in a bit
12:29 audreyt cp the stub to dist/build/src/Data/Yaml/Syck_stub.o
12:29 dakkar oops
12:30 dakkar it's Syck_stub.c, no .o
12:30 audreyt I'm implementing the saner fix (move Data.Yaml.Syck to third-party)
12:30 audreyt which is good, as it allows D.Y.S. be used separately
12:30 dakkar ok, thanks again
12:31 audreyt np :) thanks for reporting
12:31 * dakkar is great at finding obscure bugs
12:32 dakkar audreyt: while you are there, do you have any idea why I have to manually run «./Setup configure --user» to make it find fps?
12:36 audreyt I believe it's a Cabal 1.1.3-specific thing :/
12:36 trampel left perl6
12:37 audreyt try it again
12:37 dakkar ok, I'll do it by hand
12:37 nnunley audreyt: Could it be because the Pugs.cabal file doesn't refer to third-party?
12:37 audreyt nope, try 10172
12:38 audreyt nnunley: no... it's not that
12:38 audreyt dakkar: r10172 should fix that
12:38 audreyt (essentially by adding --user for you)
12:38 dakkar ok
12:38 dakkar I'm fixing a lot of whitespace-related parsefails, btw
12:39 svnbot6 r10172 | audreyt++ | * build_pugs: add --user.
12:39 audreyt yay
12:39 Juerd audreyt: I think the whitespace rules for . and .= should be the same, for consistency: no whitespace around .=, if it's also disallowed around .
12:39 audreyt dakkar++ dakkar++
12:39 Juerd Uglier, but easier to predict.
12:39 audreyt er, uhm, hm, I thought the other way around.
12:39 audreyt I see .= as infix
12:39 audreyt not postfix-method
12:39 Juerd Also, with .foo meaning $_.foo, I'd expect .=foo to mean $_.=foo; when it does not, it should probably not mean anything, and die with an appropriate error message.
12:40 Juerd audreyt: .= is based on .
12:40 Juerd Not on +=
12:40 Juerd The RHS of .= is not a normal expression
12:40 audreyt I think it is
12:40 Juerd It is, like .'s RHS, a method name.
12:40 Juerd i.e. bareword, or simple variable.
12:40 audreyt when the pugs parser sees
12:41 audreyt $_ .= ......
12:41 audreyt it transforms it as
12:41 Juerd I don't reason from a parser's point of view...
12:41 audreyt $_ .= $_......
12:41 dakkar audreyt: r10172 fixed the '--user'. now I only have the Syck_stub problem
12:41 audreyt Juerd: I see your point
12:42 audreyt my Dog $fido.=new;
12:42 audreyt is what you mean
12:42 Juerd That, and other things.
12:42 audreyt I think it's a bit too draconian; we can easily fix the whitespace problem.
12:42 Juerd Mostly, I'm worried about confusion if they different whitespace rules exist.
12:42 Juerd I also think of it the other way around: if whitespace is allowed around .=, it should be allowed around .
12:42 audreyt it is a normal infix macro... I think
12:42 Juerd It's not a macro, I think.
12:42 audreyt whilst as . is strictly postfix
12:43 audreyt ok. we Think Differently
12:43 audreyt that's fine :)
12:43 Juerd It isn't just that $foo.=bar always translates to $foo = $foo.bar
12:43 svnbot6 r10173 | audreyt++ | * Beginning to implement HsSyck.
12:43 audreyt it'd make $foo.=bar in the same cate as $foo.+bar
12:43 audreyt instead of $foo += bar
12:43 dakkar in enum.t: enum <<:Alpha<A> Bravo Charlie Delta Echo>>
12:43 Juerd It could have to call the inline-optimized version of bar
12:44 Juerd Of which I forgot the syntax :)
12:44 dakkar gives a parsefail at the <A>
12:44 Juerd audreyt: I think that would be a good solution
12:44 dakkar changing it to enum <<:Alpha<<A>> Bravo Charlie Delta Echo>> works
12:44 audreyt dakkar: a bug. fixing
12:44 Juerd Gotta go; afk
12:44 dakkar ok, so I won't touch the test
12:45 audreyt fixed
12:45 audreyt committed
12:46 svnbot6 r10174 | audreyt++ | * Lexer: In
12:46 svnbot6 r10174 | audreyt++ |     <<  < x > >>
12:46 svnbot6 r10174 | audreyt++ |   do not commit on '<<' when we see inner '<'.
12:46 svnbot6 r10174 | audreyt++ |   Reported by dakkar.
12:46 dakkar is 1..* still valid syntax?
12:47 audreyt it becomes valid syntax last week
12:47 audreyt havn't implemented term * yet
12:48 dakkar so I  won't change the lazy_list test
12:48 hexmode joined perl6
12:48 audreyt k
12:51 dakkar I'm changing a lot of lives_ok{...} to lives_ok {...}
12:51 dakkar is that needed, or am I manking some other problem?
12:51 dakkar masking
12:52 audreyt mm
12:52 audreyt I think it's always needed now
12:52 audreyt postfix {} is always hash deref
12:53 dakkar ok, so I fixed right
12:53 * audreyt misses the old backtracking dwim days
12:53 svnbot6 r10175 | audreyt++ | * HsSyck.
12:53 audreyt dakkar++
12:53 dakkar is q:to/SOMETHING/ implemented?
12:53 clkao new commits keep coming while i am compiling
12:55 Arathorn that's a good thing, no?
12:56 dakkar there's something screwy in q:to... either the implementation, or the usage
12:59 Coke joined perl6
13:00 audreyt both.
13:01 dakkar what type is the parameter to the :to adverb? a regex? a string?
13:02 clkao is pugs startup slower again
13:02 dakkar clkao: seems so
13:03 bernhard joined perl6
13:03 svnbot6 r10176 | dakkar++ |  r7940@nautilus:  dakkar | 2005-11-12 17:54:58 +0100
13:03 svnbot6 r10176 | dakkar++ |  local branch
13:03 svnbot6 r10176 | dakkar++ |  r10192@nautilus:  dakkar | 2006-05-01 12:59:37 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10193@nautilus:  dakkar | 2006-05-01 13:18:05 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10194@nautilus:  dakkar | 2006-05-01 13:50:32 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10195@nautilus:  dakkar | 2006-05-01 14:04:11 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10196@nautilus:  dakkar | 2006-05-01 14:12:00 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10197@nautilus:  dakkar | 2006-05-01 14:20:29 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10200@nautilus:  dakkar | 2006-05-01 14:39:12 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10203@nautilus:  dakkar | 2006-05-01 14:43:18 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10204@nautilus:  dakkar | 2006-05-01 14:45:00 +0200
13:03 svnbot6 r10176 | dakkar++ |  wrong fix, revert
13:03 svnbot6 r10176 | dakkar++ |  r10207@nautilus:  dakkar | 2006-05-01 14:50:09 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 svnbot6 r10176 | dakkar++ |  r10208@nautilus:  dakkar | 2006-05-01 14:57:42 +0200
13:03 svnbot6 r10176 | dakkar++ |  long dot & whitespace fixes in tests
13:03 dakkar oops...
13:04 Coke GAH.
13:04 * clkao grins
13:04 dakkar svk ci -l ...
13:04 dakkar i.e. svk push -l
13:04 Coke (worse for my client, since each of those commit lines turned into 3 lines. whoops. Need to fix that.)
13:04 xinming --verbatim
13:04 audreyt pugs startup should be fast enough
13:04 audreyt if you are playing with tests without "make install"
13:04 audreyt try
13:04 audreyt make ; ./pugs -Iblib6/lib t/foo/bar.t
13:04 dakkar xinming: yeah, my fault
13:04 clkao i was just doing ./pugs -e1
13:05 clkao hurrah, you fixed js backend?
13:05 audreyt I fixed PIL2JS
13:05 clkao still got fatal to run the generated js
13:05 audreyt havn't really investigated codegen
13:05 audreyt yes
13:05 audreyt take a look?
13:06 dakkar ok, with that badly-described commit, all the parsefails in the test suite that were directly caused by whitespace-rulings should be fixed
13:06 svnbot6 r10177 | audreyt++ | * Sanity fix and warning removal for HsSyck API.
13:08 audreyt dakkar++
13:08 dakkar now, make distclean;svk up;make
13:09 clkao how to tell javascript to carp::confess?
13:10 audreyt alert()?
13:10 * audreyt has noidea
13:10 audreyt oh, you mean stack trace
13:10 audreyt doubly no idea.
13:10 dakkar clkao: you mean, die with a stacktrace?
13:10 clkao yes
13:10 clkao i thought throw would do
13:10 * dakkar looks in his JavaScript book
13:10 Qiang joined perl6
13:11 xinming src/Pugs/Lexer.hs:232:24: Not in scope: `startRule'
13:11 xinming Build failed: 256 at util/build_pugs.pl line 96.
13:11 xinming make: *** [pugs] Error 2
13:11 xinming audreyt: I got this while compiling
13:11 dakkar xinming: me too, trying after a make distclean
13:12 audreyt fixed
13:12 audreyt the Syck_stub nightmare is fixed too
13:13 audreyt there's no need to rm Syck every build, yay
13:13 gaal audreyt++
13:13 dakkar audreyt: you're faster than my compiler!
13:13 audreyt considering it's GHC, is't not very hard...
13:13 svnbot6 r10178 | audreyt++ | * Lexer: oops, typo. xinming++
13:13 audreyt gaal: MIT license okay with you for HsSyck?
13:14 * audreyt notes Why The Lucky Stuff attached the MIT license for syck but called it the BSD license.
13:14 Arathorn clkao: Opera prints pretty stack traces by default - otherwise in Spidermonkey, if you're using exceptions properly, you can do, uhm...
13:14 Limbic_Region joined perl6
13:14 dakkar clkao: I think that the side-effects of 'throw' are implementation-defined...
13:14 gaal audreyt: sure, as is sqlite non-
13:14 audreyt yay. gaal++
13:14 gaal also beer-license
13:14 clkao the problem is jsbackend things the JS::inline()-returned function requires one arg. how does it know?
13:15 gaal or wine-license
13:15 clkao s/function/sclosure
13:15 gaal or I'll let you crash at my place when you're at a conf or just touring-license
13:15 audreyt but I'm not fine with Syck's Death and Repudiation license
13:15 dakkar clkao: js functions have a quasi-signature attribute
13:16 audreyt "This software may not be used directly by any living being... f you are found to be a ghost or angel, you will be punished to the fullest extent of the law..."
13:16 clkao ya, i know, but the pil2js backend shouldn't know, because that's something not evaled yet
13:16 dakkar ???
13:16 audreyt fortunately it's dual-licensed.
13:16 * gaal knows of no jusdictions that have laws against ghosts using software...
13:17 gaal there used to be a zine back in the BBS days, Scream I think it was called
13:17 gaal the author did some interesting experiments - one time he asked that people not read his stuff at a certain hour
13:17 gaal he wanted a rest from being accessed
13:17 gaal another issue was "torn"
13:18 gaal with ascii-art ommisions\
13:18 gaal worked well in 80-char days :)
13:19 gaal ooh I know, we can put in a moose license: "to use this library in your code, you must form a mental image of a moose. there, that wasn't so bad, was it?"
13:19 dakkar lol
13:20 Limbic_Region gaal - that would be very bad
13:21 Limbic_Region apparently merlyn would not be able to use the library
13:21 gaal no moose no use!
13:21 Limbic_Region according to him - he can't picture things in his head
13:21 gaal hey, I never said he should form it in his head. if he drew a picture of a moose that would be cool also
13:22 Limbic_Region mental image isn't in one's head?
13:22 Limbic_Region </PITA>
13:23 gaal (Who said license discussions had to be a drag? All you need to do to liven them up is to invent your own dra^H^H^H glorious license)
13:23 gaal no, mental can be a Californian adjective
13:25 clkao audreyt: maybe as_js in PIL/PApp?
13:26 audreyt clkao: sorry, I'm still in syckland
13:26 audreyt maybe it's the END block causing problems... not sure
13:26 audreyt I wonder if PIL2JS can load the YAML representation instead
13:27 audreyt that can fix things
13:27 clkao *nod*
13:28 gaal bbl &
13:28 * dakkar is ironing shirts
13:33 * audreyt proceeds to (potentially) break everybody's build by committing the HsSyck "fix"
13:33 dakkar ok, rm -rf again ;-)
13:33 audreyt nono
13:33 audreyt no need to do that
13:33 audreyt just run perl Makefile.PL :)
13:33 dakkar really? great!
13:33 audreyt (but heh, if you feel like rm-rf ...)
13:34 lumi ghc is evil
13:35 svnbot6 r10179 | audreyt++ | * Move Data.Yaml.Syck to HsSyck in third-party.
13:36 dakkar src/DrIFT/YAML.hs:45:24: Not in scope: `nodeElem'
13:36 svnbot6 r10180 | audreyt++ | * Pugs.Run: more YAML fix.
13:37 audreyt dakkar: aww
13:37 audreyt dakkar: have you rerun makefile.pl?
13:39 xinming /tmp/ghc13730.hc:8:29: ../../syck/syck.h: No such file or directory
13:39 xinming Build failed: 256 at util/build_pugs.pl line 96.
13:39 xinming make: *** [pugs] Error 2
13:39 xinming already run 'perl Makefile.PL'
13:39 dakkar audreyt: yes, I did
13:39 dakkar should I try in a fresh working copy?
13:40 audreyt rm -rf dist/build/Data
13:40 audreyt xinming: try that?
13:41 audreyt dist/build/src/DAta rather
13:41 xinming still fails...
13:42 xinming Compiling Data.Yaml.Syck   ( src/Data/Yaml/Syck.hs, dist/build/src/Data/Yaml/Syck.o )
13:42 xinming /tmp/ghc14393.hc:8:29: ../../syck/syck.h: No such file or directory
13:42 xinming Build failed: 256 at util/build_pugs.pl line 96.
13:42 xinming make: *** [pugs] Error 2
13:42 audreyt there really shouldn't be a src/Data/Yaml/Syck.hs anymor
13:43 aufrank joined perl6
13:43 xinming make clean; perl Makefile.PL; make
13:43 audreyt rm -rf src/Data/Yaml/*
13:43 audreyt and then update again from svn?
13:44 xinming if this time fails, I think I'll checkout a fresh copy, as in svk, co doesn't need the connection. :-)
13:44 svnbot6 r10181 | audreyt++ | * rm -rf src/Data/Yaml/.
13:44 aufrank audreyt: can I bug you with an svk question?
13:45 dakkar aufrank: ask clkao, he wrote it
13:45 aufrank heh, he just showed up in the other channel.  that's what I get for being impatient
13:45 aufrank sorry for the noise =\
13:45 audreyt that's fine :)
13:45 dakkar ok, after rm src/Data/Yaml/* it seems to compile
13:46 audreyt whew.
13:47 dakkar 'seems' only...
13:47 dakkar src/Pugs/PIL1/Instances.hs:1:0:
13:47 dakkar    conflict: module `Data.Yaml.Syck' belongs to the current program/library
13:47 dakkar    and also to package HsSyck-0.1
13:47 Coke is http://svk.elixus.org/ the best place to go for svk info?
13:47 dakkar Coke: yes, or the IRC channel
13:48 aufrank that and #svk
13:48 aufrank not #perl6, as I just proved ;)
13:48 Coke #svk on this server?
13:48 clkao yes
13:48 xinming Coke: yes, and IMO, you'd better have experience with svn. they are really simililar except svk rocks... :-)
13:49 audreyt dakkar:  ./Setup unregister
13:49 svnbot6 r10182 | audreyt++ | * link in HsSyck.
13:50 Coke /me tries to build pugs again, and wonders if it'll work this time.
13:50 dakkar audreyt: where?
13:50 audreyt dakkar: toplevel pugs dir
13:50 audreyt this is if you had "make install" before
13:50 dakkar never did install...
13:50 audreyt aw.
13:51 audreyt hm.
13:51 audreyt I wonder why.
13:51 audreyt is there still any Syck related stuff in dist or src for you?
13:52 xinming collect2: ld returned 1 exit status
13:52 xinming Build failed: 256 at util/build_pugs.pl line 193.
13:52 xinming make: *** [pugs] Error 2
13:52 audreyt xinming: pull and try again
13:52 xinming bbl, do a fresh install
13:52 audreyt no need to do that
13:52 audreyt :)
13:53 audreyt just pull and "make" again should work
13:53 dakkar ./dist/lib/HsSyck-0.1/ghc-6.4.1/Data/Yaml/Syck.hi
13:53 dakkar ./dist/lib/HsSyck-0.1/ghc-6.4.1/libHSHsSyck-0.1.a
13:53 dakkar ./dist/lib/HsSyck-0.1/ghc-6.4.1/HSHsSyck-0.1.o
13:53 dakkar ./dist/build/src/syck/syck.o
13:53 dakkar ./dist/build/src/syck/syck_st.o
13:53 dakkar ./dist/build/Data/Yaml/Syck.hi
13:53 dakkar which should I kill?
13:53 audreyt ./dist/build/Data/Yaml/Syck.hi
13:53 audreyt actually all three indist/build
13:55 dakkar same result after a 'make'
13:56 audreyt duplicated on feather. sec
13:57 xinming Could not find module `Data.Yaml.Syck':
13:57 xinming  use -v to see a list of the files searched for
13:57 xinming  (imported from src/Pugs/AST/Internals/Instances.hs)
13:57 xinming Build failed: 256 at util/build_pugs.pl line 96.
13:57 xinming make: *** [pugs] Error 2
13:57 svnbot6 r10183 | audreyt++ | * GHC-specific for now for HsSyck
13:58 xinming fresh copy...
13:58 audreyt xinming: that's fixed in 10184
13:58 audreyt run Makefile.PL again :)
14:00 jserv-- joined perl6
14:00 svnbot6 r10184 | audreyt++ | * try again...
14:01 dakkar trying
14:02 * xinming is considering if he should C-c and restart....
14:02 audreyt dakkar:  find dist |grep Instance | xargs rm
14:02 audreyt xinming: not if you were on 10184
14:02 hexmode joined perl6
14:02 audreyt (which I think you were)
14:03 xinming yes, 10184
14:04 dakkar audreyt: 10184 works on a fresh checkout
14:04 dakkar apart from «Generating precompiled Prelude, Test... user error (no parse: YamlNil)»
14:04 dakkar which doesn't stop the process
14:05 dakkar on the non-fresh copy, even after romeving all Instance, still conflicts
14:07 audreyt weird.
14:10 theorbtwo joined perl6
14:11 vel joined perl6
14:22 marmic joined perl6
14:22 xerox is now known as xerox__
14:23 dakkar ok, I'm stopping my work on pugs here, today.
14:24 svnbot6 r10185 | dakkar++ | long dot & whitespace fixes in tests
14:24 audreyt dakkar++ dakkar++ dakkar++
14:24 vel joined perl6
14:25 xerox__ Why so?
14:25 dakkar there are still parsefails, but I can't figure them out: they don't seem to depend on whitespace
14:33 audreyt xinming: I guess dakkar means "for today"
14:33 audreyt not, like, he'll be gone tomorrow
14:33 audreyt er sorry
14:33 audreyt s/xinming/xerox/
14:34 dvorak joined perl6
14:35 dakkar yes, you'll hear from me again ;-)
14:35 Arathorn hehe
14:37 justatheory joined perl6
14:42 FurnaceBoy_ joined perl6
14:48 svnbot6 r10186 | audreyt++ | * do not print "YamlNil" error message when there's no prelude.
14:52 audreyt ?eval 1+1
14:52 evalbot_10170 is now known as evalbot_10184
14:52 evalbot_10184 2
14:59 schneiker joined perl6
15:00 FurnaceBoy_ is now known as FurnaceBoy
15:03 schneiker TimToady:  What do you think about establishing Usenet comp.lang.perl6.general?
15:03 schneiker (I did the Usenet registration process for comp.lang.ruby years ago.)
15:03 schneiker I'd like to use it to accumulate material for a Perl6 Eary Adapter FAQ to be posted
15:03 schneiker there weekly, with info and useful links, etc.
15:04 audreyt people still use usenet? </troll>
15:05 Arathorn google killed usenet :(
15:05 audreyt (I think it's a good idea -- I did registraion for tw.bbs.religion.perl on, oh, 1996)
15:05 audreyt it's been nearly 10 years, wow.
15:06 schneiker Got to run to $work.
15:06 schneiker Will check back evening PDT (actually AZ MST).
15:06 schneiker audreyt: old habit, true, but also often handy way to focus searches.
15:06 schneiker Arathorn: Perl NGs still get quite a bit of traffic.
15:06 Arathorn schneiker: for sure. still, usenet could be much more alive if its archiving hadn't been butchered by google groups
15:10 FurnaceBoy actually I don't think that's entirely true
15:11 FurnaceBoy Google Groups would have given Usenet usage a large shot in the arm,
15:11 FurnaceBoy but you can argue the wrong type of users (more eternal September)
15:12 FurnaceBoy I see a lot of activity in the groups I hang in, anyway
15:13 audreyt the last time I followed usenet actively
15:13 audreyt was during the saga of Ellora at rec.game.roguelike.nethack
15:13 mkhl joined perl6
15:13 audreyt that was a brilliant work.
15:14 avar Arathorn: I thought AOL killed Usenet when it started the September of forever?
15:14 Arathorn well, i'd argue that GG kills both markets - its UI was/is unusable for powerusers, and too daunting for AOLers
15:14 mkhl left perl6
15:14 avar Arathorn: Why is it unusable?
15:14 avar Arathorn: I've used it twice or thrice with good results.
15:15 avar I managed to read & post to a popular newsgroup, google groups helped me to get from A to B => GG is good
15:15 Arathorn broken threading, frames, broken posting, broken message-list syncing, non-proportional fonts, obfuscated message headers, etc.
15:15 Arathorn the latter 2 points got fixed over the years, of course
15:15 avar you might argue not great, but certanly not unusable.
15:15 Arathorn and *terrible* performance for searching
15:15 avar Arathorn: What do you suggest instead?
15:15 Arathorn which is hilarious, given how it's run by google :|
15:16 * Arathorn built an alternative setup during the deja/GG hiatus called siliconarchive.net
15:16 Arathorn which is very down atm, and never made it out of beta
15:16 Arathorn but it was a damn sight better, imo ;)
15:17 macintsh joined perl6
15:18 macintsh story of the day to make us all feel old
15:18 macintsh my brother was having some regex problems and I was helping him
15:18 macintsh at one point he mentioned that he was using sed
15:18 macintsh I said, "Oh, I thought this was perl"
15:18 macintsh his answer: "i thought sed just called perl"
15:19 Arathorn heh
15:19 audreyt maybe his system maps sed to psed
15:19 audreyt in which case that'd be true
15:19 Arathorn in other news, does anyone with Judy-fu have any idea why http://rafb.net/paste/results/y5uyuv89.html melts?
15:20 js0000 joined perl6
15:21 audreyt $ ./a.out
15:21 audreyt Got PValue for 'a' as 0xffffffff
15:21 eric256 joined perl6
15:23 Arathorn uhm
15:24 avar This would be a funny name if "PV" stood for "Perl Value"
15:24 Arathorn how about http://rafb.net/paste/results/9K870438.html ?
15:24 avar Since you could read PValue as (Perl Value)alue or (Perl)Value, or ...
15:25 * avar hides
15:25 * Arathorn hopes that this isn't dear ol' debian being crap
15:25 audreyt Arathorn: bar=foo is bad
15:26 chris2 joined perl6
15:26 cognominal does use of  $!foo short-circuit accessors?
15:26 audreyt JLI is a macro that has & before bar
15:26 audreyt cognominal: yes, that's the idea, though not the implementation
15:26 audreyt Arathorn: if you s/bar/foo/g all works
15:26 cognominal audreyt: thx
15:26 Arathorn right
15:26 * Arathorn assumed that Pvoid_t's were, well, pointers to void types
15:26 Arathorn and could be assigned around as such
15:27 Arathorn thanks for checking it
15:27 audreyt np :)
15:28 audreyt    JLI(PValue, foo, 'a');
15:28 audreyt    *PValue = 123;
15:28 audreyt and finally
15:28 audreyt    fprintf(stderr, "Got PValue for 'a' as %i\n", *PValue);
15:28 audreyt is the usual usage
15:29 cognominal can we have both %.foo and @%.foo on a given class?
15:29 cognominal oops
15:29 cognominal I meant: can we have both %.foo and @.foo on a given class?
15:30 audreyt no.
15:30 cognominal that would preclude calling %.foo with array context?
15:30 audreyt actually
15:30 audreyt has %.foo;
15:30 audreyt is just
15:30 audreyt has $.foo is Hash;
15:30 audreyt i.e. it concerns the storage
15:31 audreyt the public-facing interface remains constant
15:31 audreyt if you have declared
15:31 audreyt has $.foo;
15:31 audreyt you can use %.foo freely in your method body
15:31 audreyt and vise versa
15:32 cognominal so that is for module name space, not instance name space that sigils are part of the name.
15:32 audreyt $.foo is really just short for self.foo :)
15:32 audreyt (it used to be ./foo)
15:33 cognominal reading meijer, I think that would be nice to support something like  my @?a (array with 0 or 0 element), @+a (array with more than one element) and so on
15:34 audreyt my @a is (Array where .elems < 1)
15:34 audreyt my @a is (Array where .elems < 2)
15:34 cognominal that would slightly complexify the current type system but would give support for sp,e feaire of xml schemas
15:35 audreyt I think constraint types can already do that
15:35 cognominal ho, so we alrady got that semantically
15:35 audreyt albeit checked at runtime
15:35 audreyt not compile time
15:35 audreyt for compile time checking, a great deal more thought is needed
15:35 cognominal Meijer system does not seem complecated as type system go.
15:36 audreyt it's not. it's at least not dependent typing.
15:36 cognominal It seems to be a slight complexification of perl that know only about scalar, hash and arrays
15:37 Limbic_Region and globs and formats
15:37 cognominal no I don't thinks it really as complex as dependant typing
15:37 audreyt I do wish to work out more 6.2831 (typesystem) details
15:37 audreyt with Erik during the Redmond hackathon
15:37 audreyt but for now my brain is still on 6.28 :)
15:37 cognominal you mean Eriki meijer?
15:37 audreyt yeah
15:37 cognominal cool
15:37 audreyt going to visit him June 19~24
15:37 audreyt the week before YAPC::NA
15:38 audreyt and give a talk there
15:38 audreyt maybe also visit more lambdafolks nearby, e.g. @ galois.com
15:38 cognominal he seems both knowledgeable and practical about type systems.
15:38 cognominal and Perl is all about practicalyity
15:39 audreyt Erik's design has many things in common with perl6 design :)
15:39 rafl Juerd: Could you please send my PBP to me?
15:39 cognominal I noticed...
15:40 * eric256 wishes he could manage all the travling audreyt does. ;)  ...or even half
15:41 audreyt actually, June 15~16 I'll be @ Oslo
15:41 audreyt so it's like, three time zones
15:41 audreyt in 72 hours
15:41 audreyt tw -> eu -> tw -> us
15:42 audreyt never tried that before; may or may not work :)
15:42 TimToady audreyt: foo {...} 1,2,3 is wrong, needs a comma.
15:42 audreyt then S04 needs updating. I'll do that.
15:42 TimToady k
15:42 audreyt er I mean S03.
15:42 TimToady k2
15:43 audreyt so what about
15:43 audreyt foo {...}   ,1,2,3
15:43 audreyt that's fine right?
15:43 TimToady I'd be okay with not making it fine, but if you're okay with making it fine, that also works...
15:44 audreyt I think making it fine is slightly less surprising :)
15:45 audreyt foo({...}.moose)
15:45 audreyt is still fine, right?
15:45 audreyt it's just
15:45 audreyt foo {...}.moose
15:45 audreyt is parsed differently
15:45 particle_ will 'my $foo=  $bar;' do what it does in perl5, or am i stuck with using ' = ' for assignment in p6?
15:45 TimToady { } terminates list if next thing is . yes
15:46 audreyt particle_: until someone defines postfix =, you should be fine
15:46 wolverian output as lines? :)
15:46 jsiracusa joined perl6
15:46 avar Can one add stuff like "s[]{}g" to perl 6? I.e. define a function that looks like that sort of op?
15:47 audreyt sure, it's called a macro
15:47 avar (replacement, or y///)
15:47 wilx`` joined perl6
15:47 avar is the macro aware of what sort of quote delimiters you used?
15:47 TimToady the macro gets control of the parse right after the macro name.
15:48 particle_ audrey, does my example extend to all assignment ops?
15:48 audreyt particle_: it extends to all infix-that-may-clash-with-postfix.
15:48 audreyt including but not limited to assignment
15:48 TimToady S02 discusses this.
15:48 avar TimToady: And I presume quote delimiters are not parts of the name?;)
15:48 particle_ ah, i'll have a look there.
15:49 TimToady not unless you've defined the name to include them.
15:49 TimToady macro regognition is not predicated on ident recognition.
15:49 TimToady it's just "longest token".
15:49 svnbot6 r10187 | audreyt++ | * Parser: The Perl5ish form
15:49 svnbot6 r10187 | audreyt++ |     map { $_ + 1 } 1, 2, 3, 4;
15:49 svnbot6 r10187 | audreyt++ |   is now history.  Always write:
15:49 svnbot6 r10187 | audreyt++ |     map { $_ + 1 }, 1, 2, 3, 4;
15:49 svnbot6 r10187 | audreyt++ |   expect major smoke breakage.
15:50 * eric256 just got used to leaving the quote off in p5 ;)
15:52 audreyt hm. is rx:P5:{} legal?
15:52 TimToady basically map {...} 1,2,3 was the last vestige of P5 indirect object syntax.
15:52 audreyt the "rx:P5 {}" form is a bit visually nondistinct.
15:53 TimToady I think using space consistently to indicate "this is not an adverb arg" is better than space-or-colon.
15:53 audreyt okie.
15:53 TimToady frees up :{} to mean something else too
15:54 * audreyt yields to TimToady's judgement
15:55 * eric256 supppose most the time in p6 you are going to say @a.map {  } or @a ==> map { } any way   or even map { } <<== @a  ....unless i missed the demise of the pipes ;)
15:56 audreyt the pipes is alive and well and I even understand what they are doing now :)
15:56 * wolverian ponders about >>=
15:56 gilimanjaro joined perl6
15:56 * audreyt ponders about using that to redefine macro infix<;>.
15:56 TimToady dakkar: in q:to/foo/ the /foo/ is *not* an argument to :to
15:56 audreyt infix:<;>, even.
15:57 eric256 are pipes implemented in pugs yet?
15:57 audreyt no. I'll have to implement Capture first
15:58 eric256 will pipes only work with functions that expect a list/array as their last arg?
15:58 audreyt not so; it can work with anything :)
15:59 audreyt say(int <== 1);
15:59 eric256 so it just starts filling in the functions arguments starting where the user left off?
16:00 audreyt normally, yes.
16:00 audreyt the function can choose to distinguish between "pipe batches" using @@foo
16:00 audreyt in their signature
16:00 eric256 ahhh. cool.
16:00 audreyt but otherwise, yes, it's asif the function args just gets extended
16:01 * eric256 thinks p6 is giong to have so many new features its going to take years to figure it all out ;)
16:01 TimToady "pipe batches"++
16:01 audreyt that's okay, same for p5
16:01 eric256 not a problem at all...just exciting ;)
16:01 audreyt TimToady: better than "parallel arguments"? :)
16:01 Juerd rafl: email me your address again please
16:02 rafl Juerd: Aye
16:02 TimToady I think of them as "multidimensional slices", but pipe batches is a user-friendlier term.
16:02 cognominal $object = $class.bless: {k1 => $v1, k2 => $v2 }; # is this correct?
16:02 rafl Juerd: Done
16:03 TimToady and indicates the laziness of them better.
16:03 audreyt cognominal: looks fine
16:03 cognominal that would be a third way to initialize
16:03 beppu joined perl6
16:04 weinig|sleep is now known as weinig
16:04 Arathorn gah - if foo = bar doesn't work in judyland, how on earth /do/ you assign a JudyL around?  does anyone know anywhere else more appropriate where i can ask judy q's?
16:05 audreyt Arathorn: you assign JudyL around by assigning &foo around
16:05 TimToady cognominal: as long as you realize that $class probably isn't one...
16:05 audreyt I think
16:06 cognominal TimToady: I may be dense, but I don't get what you mean
16:06 TimToady $class is just a class indicator--something that can be used to find the real class.
16:06 audreyt cognominal: Dog.bless(...) works
16:06 audreyt but also does
16:06 audreyt $fido.bless(...)
16:06 cognominal ok
16:06 TimToady Dog is not a class, but an undefined prototype object.
16:06 * Arathorn puts &foo in a PPvoid_t, then, and tries that...
16:06 TimToady but it still has .meta
16:06 rw-rw-r-- <bad-joke>why, did fido sneeze?</bad-joke>
16:07 audreyt (where "undefined prototype object" means simply that it has .defined set to false, and throws exception on attribute slot access)
16:07 TimToady at least for Dog attributes...
16:07 audreyt right.
16:08 TimToady .defined(Mammal) might be true at the same time.
16:08 audreyt you mean .Mammal::defined?
16:09 audreyt or defined(x, y) is another form now?
16:09 audreyt (x: y), even
16:09 TimToady I was thinking of the latter, actually.
16:09 audreyt 'k.
16:09 TimToady more like an isa.
16:09 audreyt gotcha.
16:09 cognominal no one followed up my p6l post about interroperability with js. I thought it raised interesting questions beyond  inteterroperability  was it unreadable or what?
16:12 audreyt cognominal: parrot distinguishes between Null and Undef
16:12 Arathorn cognominal: i suspect it got lost in the ensuing "A shorter long dot" epic
16:12 audreyt where Null means your "uninitialized"
16:12 szbalint There is no direct way to use C libraries with Pugs atm, apart from that Haskell workaround mentioned in the docs, right?
16:13 audreyt szbalint: right, sadly. can you submit it as a Summer of Code ticket?
16:13 eric256 and undefined means initialized but no value? doesn't that sound backwards?  /me isn't advocating change, just amused
16:13 TimToady P6 distinguishes undef that is "interesting" from undef that is "uninteresting".  I'd say uninitialized is uninteresting.
16:13 audreyt szbalint: http://hackage.haskell.org/t​rac/summer-of-code/newticket
16:14 cognominal and how does it translate at the Perl6 level?
16:14 TimToady if an undef contains an exception, it's not merely "uninitialized"
16:14 audreyt szbalint: use the subject "Pugs - Foreign Function Interface"
16:14 audreyt the "interesting" part is called "Failure" atm
16:15 audreyt    Undef       Undefined (can serve as a prototype object of any class)
16:15 audreyt    Failure     Failure (throws an exception if not handled properly)
16:15 audreyt cognominal: see S04 for "Failure"
16:15 cognominal ok
16:15 TimToady so basically we already make the distinction
16:16 szbalint audreyt: ok, give me a sec
16:16 TimToady but people can think of the two kinds of undef as one until they need to make the distinction
16:16 audreyt actually, Whatever belongs there too
16:16 audreyt as it also has a false .defined
16:16 TimToady that's because it's just a prototype
16:17 TimToady the extra "value" of Whatever is that you can't instantiate it.
16:17 cognominal I suppose this will be usefultoo  for  Perl6 macros for support of SQL within Perl6
16:18 * audreyt adds Whatever to S06.
16:18 TimToady don't let dduncan hear you mentioning SQL in the same breath as P6... :)
16:18 cognominal I meant a cleaner SQL :)
16:19 TimToady I believe dduncan thinks the cleanest SQL is a dead SQL.
16:19 nnunley cognominal: Something like LINQ?
16:20 cognominal yea, all this Meijer stuff sounds vey good.
16:21 audreyt sometimes, not having a huge IDE to maintain can be an advantage :)
16:21 nnunley So, basicly stronger set/iterator manipulation as syntactic sugar.
16:21 audreyt (Erik ranted on LtU that a lot of effort went into ensuring that Visual Studio can still work)
16:23 cognominal I expect that some day the reference representation of a program will be its syntax tree,  programmer will just happen to work on a convenient deparsing view.
16:23 cognominal But that means a redesign of the whole ecosystem.
16:23 cognominal s/syntax tree/AST/
16:24 cognominal so Larry will be free to invent new syntax without fear of breaking backard compatibility
16:24 audreyt that is already the case with macro-enabled languages
16:25 audreyt Perl6 is not unique; CodeDOM, TH, Scheme, etc all do that
16:25 TimToady It could almost be said that P6 was designed with that as the primary goal
16:25 audreyt though Perl6 is somewhat unique in doing a scannerless macro system
16:26 cognominal renaming a variable would be a simple operation.
16:26 audreyt with unlimited prec levels
16:26 cognominal that my litmus test.
16:27 audreyt everywhere (mkT renameVar) p6ast
16:27 audreyt    where
16:27 audreyt    renameVar (Var "$x") = Var "$y"
16:27 audreyt    renameVar others = others
16:27 audreyt (that's how currently it works in Pugs internals)
16:27 audreyt need to expose that to p6 level.
16:28 TimToady does that hide the old name?
16:28 cognominal nice to see that my brain is already partly in resonance with yours :)
16:28 audreyt TimToady: "hide"?
16:28 audreyt it replaces all leaves in the tree
16:28 * eric256 toggles a few switches on the resonance generator and laughs evily
16:29 TimToady so it has a different name at a different level, not a duplicate name.  That's all I was asking.
16:30 TimToady not the same as aliasing then
16:30 audreyt no, not the same
16:30 TimToady fine
16:30 audreyt it's as if you did a semantic-sensitive s:g///
16:31 TimToady so I wonder if term:<$foo> works everywhere it ought to.
16:31 xinming my $s = "Test"; require "$s";    <-- will this be valid?
16:31 TimToady that is, macro term:<$foo>
16:31 audreyt xinming: yes. see S11
16:31 audreyt require "/home/non/Sense.pm" <common @horse>;
16:33 TimToady it's more like a macro ast:<$foo>
16:33 TimToady term implies syntactic matching, not semantic, I think
16:34 audreyt I think it's probably a generic programming operation on the AST object.
16:34 TimToady it's one of those "is deeply" things...
16:34 TimToady but how do you declare it in P6?
16:34 audreyt we don't have a "generic deep-map" operation equivalent of the "everywhere" above
16:34 audreyt it can be simulated easily with a visitor pattern
16:35 wolverian isn't that everywhere just fmap? I don't quite understand the generics library..
16:35 audreyt wolverian: no... fmap goes on level
16:35 TimToady I guess at worst it's just a .wrap on something somewhere...
16:35 audreyt one level
16:36 wolverian audreyt, oh? I thought you're supposed to define, e.g. fmap xs = map fmap xs for lists.
16:36 TimToady except a .wrap is not likely to be reversible...
16:36 audreyt multi replaceVar (Var :(:name<$foo>)) { Var.new(name => '$bar') }
16:36 audreyt multi replaceVar ($x) { $x }
16:37 audreyt wolverian: it works on any data type
16:37 audreyt wolverian: your fmap can only work for [[[[a]]]]
16:37 audreyt or similar
16:37 audreyt wolverian: but "everywhere" works even on [([a],b)]
16:37 audreyt wolverian: it skips the leaves where type doesn't match the replacement function
16:37 wolverian and fmap works only on Functors?
16:38 wolverian hrm, right. case :)
16:38 audreyt :)
16:39 audreyt $p6ast.everywhere(&replaceVar)
16:39 audreyt now, the definition of .everywhere is very handwavy at this moment :)
16:39 TimToady .whatever (ducks)
16:39 audreyt *.*(*)
16:40 TimToady that'd do it...
16:40 wolverian $program.whenever({ $x == 3 } => { say "debug" })
16:40 * audreyt ponders "pugs -e '*'"
16:40 audreyt maybe it dies with "Oh yeah, whatever, never mind"
16:40 wolverian I guess that would print debugs on every clock tick when $x == 3
16:41 wolverian which might be a bit flood-ish
16:41 audreyt maybe it flipflops
16:41 TimToady last message repeated 2784937289734893 times...
16:42 eric256 by flipflops you mean its only fired when $x becomes 3 ?
16:42 audreyt yeah
16:44 svnbot6 r10188 | audreyt++ | * Makefile.PL: Print some more messages to explain that
16:44 svnbot6 r10188 | audreyt++ |   we are building dependencies.  This dep-building may be better
16:44 svnbot6 r10188 | audreyt++ |   factored out into "make" time.  Maybe...
16:46 flounder99 joined perl6
16:47 TimToady must go primp &
16:47 szbalint audreyt: how much manpower/time do you think a FFI for C, etc. would be to implement?
16:48 flounder99 ?eval my @a=(1,2);my @b=(2,3);(@a,@b)=(@b,@a);@a.say;@b.say
16:48 evalbot_10184 OUTPUT[2312 23 ] bool::true
16:48 audreyt szbalint: if you remove the "etc" and focus on just C
16:48 audreyt based on hs-plugins
16:48 audreyt I think 1month?
16:48 audreyt 2people*2week
16:48 szbalint k
16:48 szbalint ty
16:48 audreyt np
16:49 szbalint submitted.
16:49 flounder99 ?eval my @a=(1,2);my @b=(2,3);(@a,@b):=(@b,@a);@a.s​ay;@b.say;;@a.say;@b.say?eval
16:49 evalbot_10184 Error:  unexpected "?" expecting term postfix, comment, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
16:49 audreyt flounder99: that's a bug you found
16:50 audreyt write a test?
16:50 flounder99 ?eval my @a=(1,2);my @b=(2,3);(@a,@b):=(@b,@a);@a.say;@b.say;
16:50 evalbot_10184 OUTPUT[23 12 ] bool::true
16:50 flounder99 bind works
16:50 audreyt under t/pugsbugs/ or t/data_types/array.t maybe
16:50 audreyt you have a commit bit?
16:50 flounder99 nope
16:50 audreyt k. email?
16:51 flounder99 i'm new to irc, how do I send it privately
16:51 szbalint /msg audreyt <text here>
16:52 ingy hola
16:54 audreyt uh, this is freenode
16:54 audreyt /msg won't work without registering your nick
16:54 audreyt mangle it somehow?
16:54 audreyt foo at bar dot org
16:54 audreyt etc
16:54 audreyt or mail me at audreyt at audreyt dot org
16:54 szbalint ouch, no private msges from unregs, I forgot sorry.
16:54 audreyt ingy: yo. how's Chupei hackathon?
16:56 ingy audreyt: great. gugod and I started implementing a new ODBMS
16:56 ingy it's really awesome
16:56 obra oooh
16:56 obra ingy: details?
16:57 ingy obra: I'll have something on cpan soon
16:57 obra nice
16:57 obra what's the searching like?
16:58 ingy like?
16:58 audreyt sample syntax?
16:58 ingy gugod is familiar with jifty, which came up a few times today
16:58 ingy audreyt: I'll have something on cpan soon :p
16:59 audreyt cool :)
16:59 ingy actually the query language is one thing we haven't sttled yet
16:59 obra an object backend for jifty would be very hot
16:59 audreyt clearly you should use J or APL as the query language
16:59 ingy obra: yes it would
17:00 ingy audreyt: ok
17:00 ingy oren of yaml is helping out too
17:00 audreyt for example, the "avg" function can be defined in J as
17:00 audreyt . +/ % #
17:01 nnunley ingy: You've looked at kdb?
17:01 ingy nnunley: no
17:01 * ingy looks
17:02 nnunley kx.com
17:02 nnunley It's also an APLish thing.
17:02 ingy http://www.kdb.com/
17:02 nnunley But there's some good papers written by the author about how it works.
17:02 ingy mmm classical
17:04 ingy nnunley: url?
17:04 ingy sorry
17:05 audreyt flounder99: welcome aboard :)
17:05 audreyt please add yourself to AUTHORS, and write a test for the list-assignment bug
17:05 eric256 .
17:05 eric256 . + / % # looks like some kind of joke not a syntax for AVG() ;)
17:05 nnunley ingy: http://www.kx.com/ is the corporate web site.
17:05 szbalint J is even more compressed than Perl
17:06 audreyt eric256: QuickSort in J:
17:06 Limbic_Region eric256 - J is obfu by design
17:06 nnunley Vector languages.
17:06 audreyt ]`(($:@:((}.<:{.)#}.)),{.,​($:@:((}.>{.)#}.)))@.(*@#)
17:07 * nnunley shivers.
17:07 gugod now it obfu recursively
17:07 audreyt the golfability is amazing.
17:07 Limbic_Region eric256 - http://en.wikipedia.org/wi​ki/J_programming_language
17:07 eric256 like BF for people who wanted more pain?
17:08 audreyt unlike BF, this is actually 1)fast 2)short
17:09 Limbic_Region yeah, I should have said J is golfed by design with obfu as a side-effect
17:09 svnbot6 r10189 | audreyt++ | * HsSyck.cabal: suppress warnings.
17:09 nnunley Heh.  Pagerank in K: http://www.langreiter.com/k/pagerank.k
17:12 svnbot6 r10190 | audreyt++ | * examples/nested_loops/functional.p6: Fix
17:12 svnbot6 r10190 | audreyt++ |     map {...} ...
17:12 svnbot6 r10190 | audreyt++ |   to
17:12 svnbot6 r10190 | audreyt++ |     map {...}, ...
17:12 svnbot6 r10190 | audreyt++ |   Helps welcome to grep through all examples/tests/etc and update
17:12 svnbot6 r10190 | audreyt++ |   this misidiom...
17:14 * eric256 attempts to update his svk directory on feather to assist ;)
17:15 audreyt TimToady: can we retire lvalue undef?
17:15 audreyt (undef, $x) = (1,2);
17:15 audreyt and write
17:15 audreyt (*, $x) = (1,2);
17:15 audreyt instead
17:15 eric256 map { }, 1,2,3 works already? or are we breaking stuff until it does?
17:15 audreyt lvalue undef never really made sense to me...
17:15 audreyt eric256: it works already
17:16 audreyt eric256: also, you can copy ~autrijus/pugs/pugs :)
17:16 audreyt it's optimized and a recent build
17:17 eric256 excellent ;)
17:17 eric256 thanks.
17:17 svnbot6 r10191 | audreyt++ | * Types.Array: flounder99 reported this bug:
17:17 svnbot6 r10191 | audreyt++ |     (@x, @y) = (1,2,3);
17:17 svnbot6 r10191 | audreyt++ |   @y should get cleared, but was not.  Same applies for:
17:17 svnbot6 r10191 | audreyt++ |     (undef, $x) = (1);
17:17 svnbot6 r10191 | audreyt++ |   the $x here was not assigned undef either.
17:20 audreyt np :)
17:22 eric256 shouldn't  grep -r map *.*  look at all files under the current directory and any of its sub directories?
17:22 audreyt grep -r map .
17:22 audreyt pcregrep -r 'map \{.*\} ' .
17:22 audreyt perhaps
17:22 eric256 ahh much better.../me kept reading the man page
17:25 eric256 does grep need it too then?
17:25 eric256 and i'm a little confused what something like
17:25 eric256 push @spelfiles, map { "$d/$_" } grep { $_ ~~ /.spel$/ } readdir $d;
17:26 eric256 should become....if we add the commas in it starts looking a little confusing. ;)
17:26 eric256 push @spelfiles, map { "$d/$_" }, grep { $_ ~~ /.spel$/ }, readdir $d;
17:26 eric256 those look like  they are al arguments to push now. ;) /me watches audreyt try to figure out how to ask me to NOT help ;)
17:27 audreyt grep needs it too
17:27 audreyt listop is naturally right assoc
17:27 audreyt so that's just fine
17:28 audreyt otoh, you can write it like
17:28 TimToady yes, we can retire undef assignment.
17:28 audreyt @spelfiles.push(readdir($d).g​rep:{/.spel$/}.map{"$d/$_"});
17:28 audreyt TimToady: yay
17:29 TimToady 'Course, assignment to * should pick a random variable and assign to that.  :P
17:31 audreyt riight. :p
17:32 audreyt committed to S03.
17:34 eric256 erri i think i need to just completly redo my svk setup....or learn how to use it right ;)
17:34 svnbot6 r10192 | audreyt++ | * Pugs.AST.Internals: Move the VCode Perl5-instance-for-JavaScript
17:34 svnbot6 r10192 | audreyt++ |   kluge to Internals.hs itself, so it no longer needs hand tweaking.
17:34 svnbot6 r10193 | eric256++ | Added rules example directory, and a short example of using rules
17:34 * eric256 fears a giant broken merge from his machine is coming.....
17:35 eric256 hmmm i don't even know when i did that commit /me begins to prey to the pugs gods for forgiveness ;) lmao
17:35 svnbot6 r10194 | eric256++ | Added a realy basic BASIC parser, that doesn't realy work yet.
17:36 TimToady gee, only 451 files with map/grep breakage...
17:36 audreyt eek.
17:36 svnbot6 r10195 | eric256++ | basic.p6 BASIC implementation in P6....well it wil be eventualy.
17:36 svnbot6 r10196 | audreyt++ | * Pugs.AST.Internals.Instances: regen DrIFT instances.
17:37 audreyt 451 files...
17:37 audreyt color me shocked. :)
17:38 TimToady probably a few false positives there, but it's still gonna be >400
17:38 svnbot6 r10197 | eric256++ | Adding the comma after the block in all map/grep statments.
17:38 TimToady are blib6 files derived?
17:39 audreyt yeah
17:39 audreyt from ext/
17:39 audreyt so that cuts the number in half
17:39 TimToady then take away 70 files
17:39 TimToady down to 381
17:39 svnbot6 r10198 | fglock++ | docs/notes/laziness.txt - added some notes on map() internals
17:40 TimToady 380, can chuck Changelog :)
17:40 audreyt heh
17:40 audreyt ok, I'll leave the massive regrepping and remapping to #perl6
17:40 audreyt see you tomorrow :)
17:40 audreyt zzz &
17:41 TimToady are .pl files always P5?
17:41 * eric256 sees that while his commits where numerous they actualy seemed to work
17:42 TimToady are .hs files always Haskell?  :)
17:42 flounder99 ?eval
17:43 flounder99 ?eval my @a=(1,2);my @b=(2,3);(@a,@b):=(@b,@a);@a.say;@b.say;
17:43 * particle_ wonders if a haskell compiler has been implemented in perl6 yet
17:43 evalbot_10184 OUTPUT[23 12 ] bool::true
17:43 flounder99 is that right?
17:44 TimToady Seems right.
17:45 TimToady As Damian would say, you're just swapping the nameplates on @a and @b.
17:46 flounder99 if it were = not := @a would slurp up everthing and @b would be an empty array
17:47 TimToady right
17:47 flounder99 I guess slurping and binding don't go together.  That makes sense
17:48 TimToady slurping must be explicit under binding.
17:48 clkao aufrank: got any rule stuff done/
17:48 TimToady need to commute to work, then I'll work on fixing map/grep breakage.
17:49 TimToady got it down to 289 files... :)
17:49 TimToady I figure Makefiles are all P5...
17:50 TimToady self ==> my $job &
17:55 aufrank clkao: no, I just finished running my subject
17:55 aufrank was cramming on that project all morning
17:55 clkao ok, i will let you work ion that and i will fix interactive commit in svk
17:56 aufrank deal :)
17:57 aufrank clkao: where in the tree do you think I should be working?
17:57 clkao pugs-grammer-rules ? and then make use of that in pugs-capture-something?
17:59 aufrank ok, I'll try to commit something in p-g-r today
18:02 ghenry joined perl6
18:11 bpphillips joined perl6
18:15 meppl joined perl6
18:49 elmex joined perl6
18:58 sahadev joined perl6
19:06 meppl joined perl6
19:10 eric256 did anyone figure out a good grep for the map|grep fix?  .... i can't seem to convince regular grep i'm sending it a perl regex
19:14 spinclad build failure: i've svn up'd to r10198, and, even after C<make clean> and C<perl Makefile.PL>, C<make> gives me:
19:14 spinclad Compiling Data.Yaml.Syck   ( src/Data/Yaml/Syck.hs, dist/build/src/Data/Yaml/Syck.o )
19:14 spinclad /tmp/ghc27057.hc:9:29: error: ../../syck/syck.h: No such file or directory
19:14 spinclad Build failed: 256 at util/build_pugs.pl line 96.
19:15 szbalint wait a sec spinclad
19:19 szbalint audreyt broke it earlier today but it should have been fixed by 10184
19:19 spinclad i still have these files around (plus dist/lib/HsSyck* and third-party/HsSyck which should be fine):
19:19 spinclad ./src/Data/Yaml/Syck.hs
19:19 spinclad ./src/Data/Yaml/Syck_stub.h
19:19 spinclad ./src/Data/Yaml/Syck_stub.c
19:19 spinclad ./dist/build/src/Data/Yaml/Syck.hi
19:19 spinclad ./dist/build/src/src/Data/Yaml/Syck_stub.o
19:20 spinclad szbalint: yes, i thought so too.  i can try expunging these and try again...
19:21 spinclad (but will read them first)
19:21 szbalint 15:53:34 < dakkar> which should I kill?
19:21 szbalint 15:53:38 < audreyt> ./dist/build/Data/Yaml/Syck.hi
19:21 szbalint 15:53:43 < audreyt> actually all three indist/build
19:21 szbalint maybe this helps?
19:22 spinclad i did, these came back, i think... will try again after reading
19:23 Toaster joined perl6
19:25 spinclad (that is, the ones in dist/build came back)
19:29 spinclad after removing all five, i get:make: *** No rule to make target `src/Data/Yaml/Syck.hs', needed by `pugs_requirements'.  Stop.
19:29 szbalint oh
19:29 spinclad trying make clean, perl Makefile.PL, make again
19:30 szbalint < audreyt> rm -rf src/Data/Yaml/*
19:30 szbalint < audreyt> and then update again from svn?
19:31 svnbot6 r10199 | eric256++ |  r19475@feather:  eric256 | 2006-05-01 21:31:58 +0200
19:31 svnbot6 r10199 | eric256++ |  Adding comma to map
19:31 spinclad trying
19:31 eric256 argi forgot my --verbose flag....anyone know if there is a way to make svk default to --verbose on push?
19:33 xinming eric256: It's --verbatim   :-)
19:33 eric256 err thats what i meant...any way to deafult that? ;)
19:34 xinming don't know, But In my bash,  I use `svkpush="svk push --verbatim"`  :-)
19:34 xinming hmm,
19:34 xinming alias svkpush="svk push --verbatim"    :-)
19:35 eric256 thanks.  /me is still a unix newb after 10 years of working *around* them ;)
19:36 xinming so is /me  :-/
19:37 spinclad szbalint: better now, it's gotten past that point.  i expect it will now skip trying to compile Data.Yaml.Syck as well :)
19:37 spinclad thanks
19:39 szbalint yw
19:40 xinming ?eval .say
19:40 evalbot_10184 OUTPUT[perl6 ] bool::true
19:42 eric256 ?eval for (5) { .say };
19:42 evalbot_10184 OUTPUT[5 ] undef
19:42 eric256 ?eval $_.perl
19:43 evalbot_10184 "\\\"perl6\""
19:43 eric256 hehe why does $_ start with "perl6" in it?   bug in evalbot or feature?
19:44 xinming don't know, but In pugs, It seems, $_ will keep the last return value instead of empty.
19:45 xinming hmm, I mean in interactive shell.
19:46 sahadev joined perl6
19:47 eric256 hmm in interactive it definitly seems broken
19:52 spinclad it looks to me like the syck transition will require a make realclean
19:52 svnbot6 r10200 | lwall++ | First whack at map and grep commification.
19:53 js0000 left perl6
19:53 * eric256 waits to see if its commits and larrys conflict
19:54 TimToady Already had to deal with a bunch of conflicts.
19:54 eric256 jeez...i remember him saying he was leaving that to #perl6....and now off you go missen me up
19:54 TimToady and then I said I was working on it.
19:55 eric256 ;) fine.... /me didn't want to help anyway
19:55 spinclad isn't TimToady part of #perl6?
19:56 eric256 no!
19:56 svnbot6 r10201 | eric256++ |  r19478@feather:  eric256 | 2006-05-01 21:54:08 +0200
19:56 svnbot6 r10201 | eric256++ |  added comma to t/syntax/parsing/spacehip.t
19:56 svnbot6 r10202 | eric256++ |  r19479@feather:  eric256 | 2006-05-01 21:55:25 +0200
19:56 svnbot6 r10202 | eric256++ |  added comma for map to t/statments/*
19:56 eric256 ahh now look what i did....didn't use my svkpush alias.  /me goes to get more caffine
19:59 * eric256 checks lwalls commit and sees that lwall works much much faster at that. ;)
19:59 eric256 i got like 5 files and you got the other 100 ;)
20:00 TimToady probably missed some sorts though..
20:01 TimToady the hard part was *not* fixing the P5 files... :)
20:01 eric256 yea i had to undo that a couple of times.
20:01 mschmidt joined perl6
20:03 eric256 i need to write a program to search thinkgeek and get me the most cool stuff for $100 ;)
20:05 clkao w/win 22
20:05 Limbic_Region the difficulty in that eric256 is the definition of "cool"
20:06 mschmidt left perl6
20:06 TimToady eric256: you took out the parens I put around the maps in t/statements/map-and-sort-in-for.t
20:07 TimToady those shouldn't parse without parens these days
20:07 TimToady or teh {...} is taken as the for's block
20:09 larsen joined perl6
20:10 eric256 i took parens out?  i didn't think so
20:10 TimToady well, didn't put parens in, then overrode my change...
20:11 eric256 i didn't think my version had parns... i certianly didn't mean to remove them if i di.
20:11 TimToady don't sweat it--I'll fix...
20:11 * eric256 has spent most the day fighting with svk instead of working with it lol
20:14 svnbot6 r10203 | lwall++ | bare {...} in a for list needs parens now
20:15 TimToady hmm, looks like ext/HTML-Entities/t/entities.t is looping...
20:19 davidfetter joined perl6
20:19 mschmidt joined perl6
20:21 TimToady lunch.fetch; biab &
20:23 eric256 can for be postifxed now? hehe
20:24 eric256 @a.sort.map:{1}.for { print $_ };
20:30 * davidfetter 's brain melts
20:33 davidfetter left perl6
20:34 bpphillips left perl6
20:35 cognominal joined perl6
20:50 TimToady ext/URI/t/escape.t also loops
20:53 TimToady eric256: that'd have to be @a.sort.map:{1}.for({print $_}) or something equivalent.
20:53 TimToady methods don't take arguments unless followed by : or (
20:54 eric256 so that should probably be      for @a.sort.map:{1} { ... }    anyway
20:54 eric256 or would that still need to be in ( )?
20:55 TimToady depends on wehter there's really a :{} adverb.
20:55 TimToady if there is, it doesn't need parens.
20:56 TimToady but if that's really .map: {1}, where the colon is just introducing an unparenthesized list, then it would need parens.
20:56 aufrank joined perl6
20:57 TimToady at the moment I'm leaning towards saying that :{} exists as an adverbial form, so it doesn't imply any following args.
20:57 TimToady and you wouldn't need parens
20:57 eric256 makes sense
20:58 TimToady that is, it parses as .map :{1} so you have an adverb modifying an argumntless .map
20:58 particle_ is that the goatee op?
20:59 eric256 i thought it was a cookie monster
20:59 TimToady the .map:{1} form parses the :{ as a token
20:59 particle_ :-{0}
21:00 TimToady otherwise it would have to parse as .map : {1} which would be the argument form of .map
21:00 * eric256 needs to go read about adverbs
21:01 spinclad 8^{[=]}
21:02 TimToady t/builtins/sprintf_and_as.t also loopos
21:02 TimToady "loopos"++
21:02 spinclad #{[ENOTANADVERB]}
21:03 TimToady E No Tan Adverb
21:04 spinclad adverbs must stay completely covered when out in the sun
21:08 TimToady t/examples/examples.t also loopos
21:08 particle_ r10204 | particle++ | apply sunblock to all adverbs
21:13 * eric256 tries to convices a co-worker that dewed is actualy a word ;)
21:14 trym joined perl6
21:14 particle_ playing scrabble at work?
21:14 spinclad the dewed rose, or the dewed spouse
21:14 particle_ mine 'glistens'
21:15 eric256 the ground was wet this morning and i said "i wonder if it rained last night, or just dewed heavily"
21:15 TimToady kewel, dewed!
21:19 spinclad lol (really!)  so that's three meanings!
21:20 * eric256 wishes there was a lottery or something to when to go to the YAPC::NA
21:24 buu Mmm, parrot.
21:29 TimToady would appear that t/syntax/list_quote_whitespace.t is the 5th loopo
21:30 eric256 loopo?
21:30 eric256 infinite loop?
21:30 buu #     Failed test (t/op/calling.t at line 2327)
21:30 buu =[
21:31 buu Can I ignore it or do I need to bash it?
21:43 coumbes joined perl6
21:43 Ymmv joined perl6
21:46 TimToady it seems that sprintf is hanging
21:48 particle_ buu: that's a known parrot test failure
21:48 TimToady yes, this hangs: say sprintf '%d', 123;
21:55 kanru joined perl6
22:00 svnbot6 r10204 | lwall++ | any<a b c> is not the same as any <a b c>
22:01 eric256 are we getting just a little crazy now?  /me fears perl6 will be way too much like python with all these space issues.
22:10 * avar wouldn't care if it was like python in that regard
22:10 TimToady the postfix vs non-postfix difference is crucial.
22:11 Juerd eric256: I think we are, but I don't see any good solution to the problem.
22:12 Juerd eric256: Not requiring parens is a problem if every other character is both an infix and a prefix operator :)
22:12 Juerd And sometimes postfix too.
22:12 eric256 in these cases make the postfix and non-postfix forms act the same?  of course then it might just increase the confusion
22:12 Juerd eric256: How would you do that?
22:12 eric256 what would any<a b c> be?
22:13 TimToady any.{'a','b','c'}
22:14 eric256 and any <a b c> is any("a","b","c") ....i don't know what any.{ ... } would mean though. must be time to reread the S's again
22:14 svnbot6 r10205 | lwall++ | any<a b c> is not the same as any <a b c>, continued
22:14 TimToady Doesn't mean anything, which is why you don't do that...
22:15 eric256 so why not make .{ a postfix op on any that does what we expect?  i dunno, guess i'm just stuck in the but it doesn't DWIM phase.
22:15 jsiracusa joined perl6
22:16 TimToady It should blow up and say that you can't treat a Junction like a Hash.
22:17 eric256 oh yea...forgot < > was hash access also.  i feel soo lost some days
22:18 eric256 if it were to DWIM  i would assume that < > just replaced ( ) when used in a function arg context and auto split/quoted liked < > does
22:18 rgs joined perl6
22:18 eric256 then  any<a b c> ~~ any <a b c> ~~ any('a', 'b', 'c') ~~ any(<a b c>)
22:18 cognominal salut rgs
22:19 eric256 but i wouldn't presume to have any idea what side effects that would cause
22:25 * eric256 stuns everyone into silence with his brilliance or ignorance ;)
22:26 avar Whitespace may cause cognative overload;)
22:29 hexmode joined perl6
22:56 eric256 hehe i thought at least someone would tell me i'm nuts. .... oh well i'll just assume i'm nuts and go home for the day
23:04 buu t/pmc/objects.t    2   512    74    2   2.70%  73-74
23:04 buu Are those known failures as well?
23:06 TimToady pugs now smokes without looping, yay!  11,000 tests in <43 minutes on my machine.
23:08 szbalint looposing?
23:08 buu Yay!
23:08 TimToady buu: those failed when I ran parrot this morning, so I suppose they're "known"...
23:08 buu TimToady: Heh, ok.
23:10 buu Hrm. Where does parrot look for .so files, LD_LIBRARY_PATH?
23:12 buu Excellent.
23:18 pasteling "spinclad" at 209.94.133.172 pasted "Prelude compile failure (debian sarge+/i386, ghc6 6.4.1-2.1, pugs r10199)" (14 lines, 1.9K) at http://sial.org/pbot/17000
23:26 ayrnieu joined perl6
23:27 _SamB_ joined perl6
23:36 drbean I make fun of the Bugs Manifesto at http://use.perl.org/~mr_bean/journal/29477, repeating the Scientology identification at http://perlmonks.org/?node_id=540655
23:38 _SamB_ is now known as SamB
23:50 meppl gute nacht

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

Perl 6 | Reference Documentation | Rakudo