Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2006-08-03

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 timthelion joined perl6
00:10 timthelion left perl6
00:23 hikozaemon joined perl6
00:32 nekokak joined perl6
00:44 macroron joined perl6
00:54 cmarcelo seen audreyt
00:54 jabbot cmarcelo: audreyt was seen 10 hours 56 minutes 52 seconds ago
01:03 xdg joined perl6
01:20 unobe is now known as unobe_away
01:21 lambdabot joined perl6
01:25 dolmans joined perl6
01:42 weinig joined perl6
01:48 Vutral joined perl6
01:49 christopher joined perl6
01:55 rodi joined perl6
02:21 bradb joined perl6
02:26 esammer joined perl6
02:28 Captain_Fourier joined perl6
02:34 meppl gute nacht
02:34 mdiep joined perl6
02:52 bradb left perl6
02:56 * Khisanth wonders what is a lambdamoose
03:04 kattana_ joined perl6
03:16 svnbot6 r11896 | fglock++ | pX/Pugs-Compiler-* - implemented <%hash>; several fixes
03:20 cmarcelo audreyt: ping
03:25 amped3 joined perl6
03:39 perlchina joined perl6
03:49 amped3 left perl6
03:51 tup joined perl6
03:58 ajs_home joined perl6
04:09 mdiep STATUS says the Perl 5 backend is "Broken; once passed <50% of main tests (no ext/); inactive; will probably be rewritten around PIL2. [2006-01-30]"
04:10 mdiep is that different form the p6-to-p5 effort? or is this out of date?
04:11 obra I believe that's a whole different effort
04:22 amnesiac joined perl6
04:24 sahadev2 joined perl6
04:58 audreyt uhm, not exactly
04:58 audreyt the p5 backend was PIL-run, then pX, then lrep, then v6-alpha
04:58 audreyt there's some continuity involved (though code is all rewritten)
04:58 audreyt and now pugs -CPerl5 simply uses v6-alpha
04:58 audreyt so STATUS is outdated
04:59 obra was PIL-run an attempt to convert p6 to p5 or to do a lower-level bytecode and a bytecode interpreter?
05:00 audreyt it was an attempt to convert pil into perl5 and then eval"" it I think
05:00 audreyt but there was not sugar layer
05:00 audreyt and it's not as modularized, and does not attempt to bootstrap with a grammar implementation
05:01 obra *nod*
05:13 drrho joined perl6
05:23 scw joined perl6
05:29 pdcawley joined perl6
05:34 gaal hullo
05:34 obra hey gaal
05:34 gaal hi obra!
05:38 cmarcelo audreyt: hi. re: IArray, i'm defining type IArray = (Judy.Map2 ..., TVar ArrayIndex), the index would hold the size of array.. is it a bad idea? i'm doing this to avoid filling the structure with lots of undefs..
05:56 audreyt I think that's fine
06:04 xerox audreyt: yow :)
06:07 Captain_Fourier left perl6
06:17 masak joined perl6
06:36 * gaal mooses
06:37 * xerox is flying over to AngloHaskell!
06:41 gaal xerox: url?
06:42 xerox ?wiki AngloHaskell
06:42 lambdabot http://www.haskell.org/haskellwiki/AngloHaskell
06:43 gaal ha, haskell is the only language I know where the presence of a job opening shakes the community :)
06:46 gaal "Lemmih could give a short talk on breakpoints in GHC"
06:46 * gaal wants to hear that!
06:47 kanru joined perl6
06:47 Revision17 joined perl6
06:54 nirgle2 joined perl6
06:57 sofit joined perl6
07:08 Aankhen`` joined perl6
07:09 buetow joined perl6
07:19 elmex joined perl6
07:27 beppu_ joined perl6
07:28 iblechbot joined perl6
07:29 beppu_ I'd like to build pugs on OSX (x86).  Are there any resources that can help me along?
07:38 theorb joined perl6
07:43 beppu_ ghc has the most thorough ./configure script EVER.
07:43 dolmans since pugs in build on top of ghc, then the question is how to build ghc on osx. unoffical answer. :-)
07:48 StudiesInWords there is a 6.4.1 ghc for os x
07:51 beppu_ Indeed there is....  I think I'll use that.
07:55 beppu_ is the osx build of ghc on http://haskell.org/ghc/download_ghc_641.html popwerpc only?
07:55 lambdabot Title: GHC: Download version 6.4.1
07:57 StudiesInWords seems like powerpc only
07:58 beppu_ I guess I'll find out in a moment.
07:58 StudiesInWords i have only ppc
08:01 beppu_ It is a PPC binary package, but OSX for x86 manages to be able to run it, anyway.
08:01 StudiesInWords "undefined symbols: _rl_catch_sigwinch"
08:04 beppu_ doh.  ppc ghc won't build pugs on x86 osx.
08:05 beppu_ I get errors like: /tmp/ghc14275.s:24:no such 386 instruction: `addi'
08:05 beppu_ it's trying to use ppc assembly
08:07 tup joined perl6
08:09 StudiesInWords lots of stuff not found, but still compiling here
08:20 svnbot6 r11897 | dolmans++ | Builtin reduce operators on null or one elem list.
08:21 tup joined perl6
08:26 gaal beppu_: IRTS audreyt or nothingmuch mentioning you need arecentish 6.5.x for macs.
08:41 kane-xs joined perl6
08:50 Aankh|Clone joined perl6
09:39 ruoso joined perl6
09:57 Aankhen`` joined perl6
10:02 zeyata joined perl6
10:04 zeyata left perl6
10:08 rodi joined perl6
10:23 kanru joined perl6
10:26 hugues23 joined perl6
11:10 silug joined perl6
11:14 StudiesInWords "Failed 4/521 test scripts, 99.23% okay, 38/9471 subtests failed, 99.60% okay" (pugs 6.2.12 with ghc 6.4.1 on OS X)
11:16 chris2 joined perl6
11:21 ruz joined perl6
11:30 xdg joined perl6
11:41 marmic joined perl6
11:56 iblechbot joined perl6
12:00 tokuhirom is now known as toku_away
12:03 ivas joined perl6
12:25 Limbic_Region joined perl6
12:29 ivas joined perl6
12:37 Aankh|Clone joined perl6
12:44 Aankh|Clone is now known as Aankhen``
12:46 miyagawa audreyt: ping
12:50 lolage joined perl6
12:54 dakkar joined perl6
12:55 [particle] joined perl6
12:59 marmic joined perl6
13:02 dolmans joined perl6
13:07 Revision17 joined perl6
13:09 audreyt miyagawa: pong (though I'm about to sleep)
13:10 audreyt filterm++ # does what GNU screen cannot do
13:10 audreyt alias nethack="filterm UTF8-cp437 cp437-UTF8 nethack"
13:10 audreyt (display of line drawing characters)
13:10 miyagawa that's fine. have you read the JSON::Syck utf-8 complaint mail?
13:10 mauke_ joined perl6
13:11 audreyt uh no
13:11 miyagawa oh, ok. it's rt 20830
13:12 dakkar filterm? doesn't 'luit' do the same thing?
13:13 audreyt miyagawa: mlehmann contradicts himself
13:13 audreyt "If it is 1, it sometimes returns json objects with "bytes"
13:13 audreyt >255."
13:13 audreyt I don't understand that part.
13:13 weinig is now known as weinig|away
13:14 miyagawa maybe he's thinking latin-1 as Unicode
13:14 miyagawa ?
13:14 audreyt not sure
13:14 miyagawa alright. I'm glad you feel the same way i do
13:14 audreyt I think he means that when ImplicitUnicode is on
13:14 audreyt the Dump() should still yield byte buffers
13:15 audreyt not unicode strings
13:15 audreyt which does have a certain merit
13:15 mindbob joined perl6
13:15 audreyt but he can get that behaviour trivially using utf8::encode() or utf8::downgrade() or Encode::_utf8_off or Encode::encode_utf8
13:15 audreyt following the Dump
13:15 miyagawa probably he wants Pelr (Unicode) to JSON (utf-8)
13:15 miyagawa which JSON::Syck doesn't?
13:16 miyagawa yeah, it's really trivial ..
13:16 audreyt he wants the return value of Dump to by byte string
13:16 audreyt s/by/be/
13:16 audreyt can you reply to him to suggest that?
13:16 audreyt i.e. post process the Dump himself in perl land
13:16 miyagawa yep
13:16 audreyt I think it's generally more useful, our current behaviour
13:17 miyagawa yeah, we don't have any problem with the current
13:17 audreyt also
13:17 audreyt Perl only knows about binary strings (octet strings)
13:17 audreyt and text strings (unicode character strings), both of which can have the
13:17 audreyt UTF-8 flag set or cleared.
13:17 audreyt I think mlehmann is confused... there is no "utf8-flagged binary strings"
13:17 audreyt in perl5
13:18 audreyt but maybe it's just a terminology confusion
13:18 bpphillips joined perl6
13:19 audreyt anyway, you can always give him a commit bit and tell him to file bug reports in the form of regression tests :)
13:20 audreyt *wave* &
13:25 mauke_ is now known as mauke
13:40 frederico joined perl6
13:49 vel joined perl6
13:49 wazzuteke joined perl6
13:51 justatheory joined perl6
13:51 jiing_ is now known as jiing
13:57 justatheory joined perl6
14:00 fglock joined perl6
14:01 theorbtwo joined perl6
14:02 kanru joined perl6
14:07 svnbot6 r11898 | fglock++ | pX/Pugs-Compiler-Rule - TODO t/09-ratchet.t
14:12 penk joined perl6
14:13 _mago joined perl6
14:13 fglock mago: hi!
14:14 _mago yeah!
14:16 fglock did you get my last mail? I sent to e-via
14:16 _mago nop
14:17 _mago mago at rio dot pm dot org
14:17 fglock k
14:18 pochi left perl6
14:19 fglock re-sent :)
14:20 fglock I'm refactoring v6.pm - it was starting to take too much ram to load
14:20 [particle] fglock: i have some cycles to devote to get parrot's p6regex tests working for p6-on-p5
14:22 autark joined perl6
14:22 fglock [particle]: if the tests work for pugs they should work in v6 - or do you mean fixing v6 to get the tests running?
14:22 [particle] no, i mean writing a harness so t/p6regex/regex_tests can be run
14:22 [particle] it's just a data file right now
14:23 [particle] need a .t file that loads and runs the individual subtests
14:23 fglock sure, that would be very useful
14:23 [particle] it'll need to be written in perl6 i assume
14:23 fglock [particle]: yes
14:24 prefiks joined perl6
14:24 [particle] and should i try to only use syntax that passes 01-sanity?
14:24 hexmode joined perl6
14:24 [particle] or 02-test-pm, anyway
14:24 fglock [particle]: I'm rewriting some parts of Pugs-Compiler-Rule, it should be a bit easier to maintain
14:25 [particle] great
14:25 fglock [particle]: Test.pm works with v6.pm, but it would be good to have the basic tests not depending on it
14:26 [particle] well, these are p6regex tests, i don't know if you consider them basic or not
14:26 cjeris joined perl6
14:27 [particle] i'll need to open and iterate through a file, at minimum
14:27 [particle] i don't see open tests in 01-sanity
14:27 [particle] happily, split and substr are there :)
14:28 fglock I think using Test.pm is ok
14:28 [particle] great
14:28 [particle] brb
14:29 StudiesInWords the documentation Pugs::Doc::Hack says "cpan Bundle::Pugs::SmokeKit", but doing that gives "Bundle::Pugs::SmokeKit has been obsoleted by Task::Smoke."
14:31 justatheory joined perl6
14:35 gaal StudiesInWords: what version are you looking at?
14:35 StudiesInWords http://search.cpan.org/dist/Pe​rl6-Pugs/lib/Pugs/Doc/Hack.pod
14:35 lambdabot Title: Pugs::Doc::Hack - How to hack on Pugs - search.cpan.org
14:35 gaal look in HEAD...
14:35 gaal it no longer says that there :)
14:36 StudiesInWords right
14:37 StudiesInWords i was still checking out...
14:50 [particle] joined perl6
14:53 [particle] which Test.pm is 'use Test' actually loading? (there are several in the repo)
14:54 fglock it should use ext/Test/lib - or a copy of the same file in perl5/Pugs-Compiler-Perl6/t
14:55 gaal blib6/lib/Testl.pm.yml, which a precompiled verison of the Test.pm copied over to that dir from ext/Test/lib
15:01 lidden joined perl6
15:02 tup joined perl6
15:05 fglock any ideas to debug a message like 'Not a CODE reference at (eval 80) line 126.' in perl5 (in pX/v6.pm) ?
15:06 fglock I can't find the place
15:07 gaal fglock: install a __DIE__ handler that dumps Carp::longmess?
15:07 fglock gaal: I'll try - thanks!
15:08 fglock although this is functional code - I think it will just dump a long list of unnamed subs
15:09 gaal run under perl -d with trace on? (or one of the -D tracers)
15:10 gaal it'll be pretty slow and give lots of output
15:12 bpphillip1 joined perl6
15:19 rodi joined perl6
15:24 fglock gaal: It worked - I used '-d:Trace' and grepped for '(eval 80)'
15:25 fglock &
15:27 penk joined perl6
15:32 avar joined perl6
15:40 weinig|away is now known as weinig
15:48 ruoso joined perl6
15:49 vel joined perl6
15:49 vel 2Ac2rn1
15:53 cm joined perl6
15:57 weinig is now known as weinig_bbl
16:14 Psyche^ joined perl6
16:17 weinig_bbl is now known as weinig
16:23 typester joined perl6
16:27 Psyche^ is now known as StudiesInWords
16:34 rodi 1.5 ~~ 1^..^2  # this evaluates to False in pugs, and should evaluate to true, per S03 - "New Operators"
16:34 svnbot6 r11899 | cmarcelo++ | * Judy.Map2: Take* primitives and a missing file required for testing.
16:34 svnbot6 r11898 | fglock++ | pX/Pugs-Compiler-Rule - TODO t/09-ratchet.t
16:34 rodi Where should this test go?
16:34 rodi t/operators/?
16:35 rodi smartmatch.t?
16:35 fglock rodi: t/operators/range.t
16:35 rodi fglock: tx
16:38 frederico joined perl6
16:47 svnbot6 r11900 | rodi++ | test for lazy evaluation of range operator
16:48 buubot joined perl6
16:56 bernhard joined perl6
17:15 avar joined perl6
17:25 clkao fglock: are we nearly there yet? ;)
17:35 frederico joined perl6
17:38 fglock clkao: I'm debugging a failure in Term match - everything else works ...
17:39 clkao i was wondering. does A orz B being parsed as &infix:<orz>(A, B) only when the infix is defined?
17:40 fglock clkao: I think so - otherwise it would be A( orz( B )) or an error
17:40 clkao but when you have that compiled in foo.pm, and you define it somewhere else and the require foo.pm...
17:41 clkao i heard that one module is closed by itself, so it peeks for that during compilation or ?
17:42 Eimi joined perl6
17:43 ludan joined perl6
17:44 fglock I don't know - it would need to read that info from somewhere, or emit lazy code that gets fixed after 'use'
17:44 clkao TimToady: any idea about this?
17:49 fglock clkao: would you take a look at a bug?
17:50 clkao i am about to head out. but you can leave message on ir
17:50 clkao c
17:51 fglock ok - I'll commit what I have
17:52 awwaiid joined perl6
17:52 svnbot6 r11901 | fglock++ | pX/Pugs-Compiler-Perl6 - inserted debugging info for "Not a CODE
17:52 svnbot6 r11901 | fglock++ | reference at (eval 80) line 27124" compiling ':foo(42)'
17:52 fglock 'line 27124' means code chunk 27, line 124 - i made it print the code right before the error
17:53 fglock s/code/source code/
18:06 crem joined perl6
18:23 fglock ugh - I found the bug:
18:23 fglock ([_|\w] ) captures an array to $/[0]
18:23 fglock which is right, but I need the string it matched - how do I get that?
18:24 [particle] .match ?
18:24 fglock there is a plus sign between ] and ) - cgi-irc doesn't show that
18:24 [particle] $match = $/[0] ?
18:25 fglock [particle]: $/[0] is a bare array
18:25 fglock of matches
18:25 [particle] oh, it doesn't notice scalar context, and stringify the array?
18:25 [particle] i believe it should
18:26 [particle] in parrot, we do $S0 = match[0]
18:26 [particle] it sees it's assigning to a string register, and returns the matched string
18:26 [particle] however, this is probably a bit different in perl6
18:26 * [particle] checks S05
18:27 fglock maybe it should only be an array in array context
18:28 [particle] yes
18:28 [particle] Notionally, a match object contains (among other things) a boolean success value, a scalar result object, an array of ordered submatch objects, and a hash of named submatch objects. To provide convenient access to these various values, the match object evaluates differently in different contexts:
18:29 [particle] In string context it evaluates to the stringified value of its result object, which is usually the entire matched string:
18:29 [particle] $match = ~$/[0]
18:29 [particle] or $match ~= ...
18:29 avar joined perl6
18:30 fglock "the corresponding array element for the quantified capture will store a (nested) array rather than a single Match object" - maybe it means "store a submatch with positional captures?
18:30 fglock that would work
18:30 [particle] yep
18:32 fglock fixing this requires more coffee &
18:44 larsen joined perl6
18:48 buetow joined perl6
18:53 weinig is now known as weinig|away
18:57 fglock after some coffee - it may be an implementation bug, i'll check
19:04 svnbot6 r11902 | fglock++ | pX/Pugs-Compiler-Rule - added t/11-quantifier.t with a failing test
19:09 lollan joined perl6
19:09 svnbot6 r11903 | fglock++ | pX/Pugs-Compiler-Rule - no errors in t/11-quantifier.t
19:15 weinig|away is now known as weinig
19:19 justatheory joined perl6
19:20 salty-horse joined perl6
19:30 svnbot6 r11904 | fglock++ | pX/Pugs-Compiler-Rule - fixed the 'not a CODE' bug (localized static capture counts)
19:36 embedded_cm joined perl6
19:39 vel joined perl6
19:45 svnbot6 r11905 | fglock++ | pX/Pugs-Compiler-* - t/00-grammar/01-pcr.t passes
20:36 cognominal joined perl6
20:49 svnbot6 r11906 | fglock++ | pX/v6 - more parser tests
20:53 SageLT joined perl6
21:10 SageLT I've got a p5 regex question that I think somebody here might be able to help me with: in t/op/pat.t there's a match on the pattern / {([^}]*)}/. To me this looks illegal (the opening { seems to start a count, not be a de-meta'ed character). How does this work? By translator is braking on it, cause the way I've got it coded it thinks that's a count (which i isn't)
21:14 fglock SageLT: is this a pugs file? I don't see it. Maybe it was meant to read <(...)> ?
21:15 SageLT Sorry, it's a P5
21:15 SageLT my mistake
21:15 SageLT Sorry bout that, forgot where I eevn got it from. Dang file's been giving me problems all morning
21:19 Khisanth joined perl6
21:20 rindolf joined perl6
21:20 rindolf IO::All!
21:20 rindolf TimToady: here?
21:20 rindolf audreyt: here?
21:21 rindolf With all the hardware failures on the Perl NOC perhaps someone should shell them the money to get high-end hardware.
21:22 rindolf For {perl,cpan}.org that is.
21:24 avar joined perl6
21:30 SageLT Is ghc supported on intel macs? I just tried building some of my code for the first time on a Mac, and I get a slew of 'no such 386 instruction' errors, which would seem to be a problem targeting the processor. Is this a known bug? Is there a workaround?
21:30 cjeris left perl6
21:30 rindolf SageLT: also try #haskell
21:30 SageLT I was just looking for a good channel for haskell thanks
21:31 obra SageLT: there's a build. hang on.
21:31 SageLT Just figured with pugs being haskell based people might nkow here
21:31 obra (audrey pointed me at it ;)
21:31 SageLT obra: great!
21:31 rindolf audreyt is using Mac, right?
21:31 rindolf So does nothingmuch
21:32 rindolf TimToady seems to have an IBM laptop.
21:32 obra http://cvs.haskell.org/trac/ghc/wiki/X86OSXGhc
21:32 lambdabot Title: X86OSXGhc - GHC - Trac
21:32 obra That may be the right start
21:32 rindolf But Perl runs on anything remotely UNIXy.
21:32 SageLT obra: thanks
21:33 rindolf "All the world is a VAX, and all the actors are Unixes."
21:33 obra np.
21:33 rindolf This day was pretty quiet on Freenode.
21:34 rindolf Maybe it's Tisha B'av or something.
21:34 obra SageLT: how's the hacking been going, anyway? TimToady says you're doing good work :)
21:35 SageLT obra: Well, i hope I'm doing good work. I;m a but worried that I may not finish in time, but instead of worrying I'm working.
21:35 SageLT Worrying would only slow me down more, might as well just work
21:36 obra Indeed.
21:37 obra I _have_ to ask: have you tried feeding the p5 generated version of pugs Test.pm in?
21:37 SageLT With some code upgrades I'm about to push, I can at least partially translate about 98% of exising code, which a couple of annoying problems
21:38 SageLT I haven't tried feeding testInit.pm to pugs in some time, but I think it's about ready, I'll try it again sometime soon
21:39 obra  Thats' very cool
21:40 SageLT I've haven't re-translated it in a while. It made a good first target since it's short enough to read through the yaml parse by hand, now I've been testing with a simple script that runs that translator across everything in t/op/ of perl5, with fairly good results.
21:40 obra Very cool.
21:46 svnbot6 r11907 | fglock++ | pX/v6 - fixed 'Match' return object in some contexts
21:51 cognominal joined perl6
22:06 lolage joined perl6
22:14 mdiep fglock: ping
22:16 clkao greetings
22:18 mdiep greetings, clkao. you're working on v6-alpha, yes?
22:19 mdiep I'm looking for a bit of direction jumping in
22:19 mdiep t/01-sanity seems to pass, but the first test after that seemed to hang when I ran it
22:19 clkao i had been working on it. but not recently
22:19 mdiep ah, okay
22:19 clkao mdiep: which one?
22:20 rindolf Hi clkao
22:20 SageLT obra: I just tried installing the Mac haskell build you pointed me to, but when I try to run ghc or ghci it gives me an error about not finding a suitable version of /Library/Frameworks/Versions/A/GMP. Any clue what's up?
22:20 clkao iirc they all pass for me last i tried
22:20 rindolf clkao: how much time are you supposed to dedicate to svk?
22:20 clkao does it hang or is it just slow?
22:20 clkao rindolf: supposed? i decide that myself
22:20 rindolf clkao: but you're employed to work on it.
22:20 clkao mdiep: install cache::filecache
22:20 clkao if it's just slow
22:21 rindolf clkao: OK.
22:21 clkao rindolf: yes, and i can employ people to work on it as well
22:21 mdiep clkao: huh. it seems to be passing now. maybe it was just slow
22:21 clkao mdiep: ok. try having the cache
22:21 rindolf clkao: I'm now trying to get sponsorship, to do the random stuff I normally do.
22:21 clkao maybe we should suggest that *strongly* in makefile.pl ;)
22:21 clkao rindolf: like what? ;)
22:22 obra SageLT: yeah. audrey talked me through this. hm.
22:23 obra I don't actually remember how to fix it
22:23 SageLT obra: if it's too much trouble, don't worry, I can just run an older ghci for now.
22:23 rindolf clkao: http://www.shlomifish.org/
22:23 lambdabot Title: Shlomi Fish' Homepage
22:23 obra ok. likely audrey has a canonical doc somewhere. I'm sorry I don't have a good reference handy
22:23 SageLT obra: no prob
22:23 mdiep t/data_types/hash pretty much hangs my whole machine though
22:24 rindolf clkao: thing is the things I work in at the job, while interesting are of no relevance to what I like to do.
22:24 mdiep I'm really looking for tests that aren't passing atm but could be with a bit of work
22:24 clkao you better find another job or live with it then
22:24 clkao mdiep: oh ok.
22:24 clkao mdiep: actually, in v6 the tests are copied if they are in manifest
22:24 clkao but they are only in manifest if the test is passing.
22:24 rindolf clkao: did you read http://shlomif.livejournal.com/11004.html
22:24 mdiep ahh...
22:24 lambdabot Title: shlomif: Patronage and Self-Patronage
22:25 rashakil joined perl6
22:25 clkao so it's basically regression guardian
22:25 rindolf clkao: that's why I'm trying to get a sponsorship.
22:25 clkao mdiep: use util/prove6 for the pugs test suite itself
22:25 [particle] hrmm, i'm getting v6.pm missing in @INC with util/prove6
22:25 clkao imo, you don't _try_ to get sponsorship. they come around to you when people think it worths it
22:26 clkao in the worst case you can move to mountain and live with no cost and StFUaWSC
22:27 clkao sorry i might be cynical
22:27 rindolf clkao: are you feeling alright?
22:27 * [particle] wonders how those shirts came out
22:27 clkao cant be better
22:27 rindolf clkao: OK.
22:27 clkao [particle]: try make build_perl5 first
22:27 clkao [particle]: and/or perl util/src_to_blib
22:28 clkao see my use.perl for instruction, i can't remember them all
22:28 [particle] sure, i followed that
22:28 mdiep clkao: make build_perl6? from pugs/?
22:28 rindolf clkao: I also would like to publish two books.
22:28 mdiep err, build_perl5
22:28 [particle] i had it working before, then rebuilt
22:28 clkao build_perl5 ;) from toplevel pugs tree
22:28 [particle] now it's busted
22:28 rindolf One with essays and one with funny stories.
22:29 clkao [particle]: grr! i will take a look in a bit
22:29 [particle] take your time, clkao
22:29 mdiep but I can't use that makefile with install ghc :-)
22:29 clkao [particle]: there's no v6 in blib6/pugs/perl5 ?
22:29 clkao mdiep: someone on my journal make a script to avoid using the makefile requiring ghc
22:30 [particle] ./lib, it's there
22:30 [particle] yeah, the script is ./make_build_perl5 i think
22:30 [particle] not windows friendly, but it's just a / vs \ prob
22:31 clkao when you can easily fix provided you have a commit bit...
22:31 rindolf Bleh! I hate these Israelis Jews that think it's acceptable for Israel to be constitutionally discriminatory.
22:31 rindolf Or that it's the Land of the Jews.
22:31 rdice joined perl6
22:31 [particle] clkao: sure, when i have cycles. i'm committed already :)
22:32 [particle] currently working on a perl6 harness for sharing parrot's p6regex tests
22:32 clkao [particle]: i already done that
22:32 [particle] clkao: where?
22:32 clkao [particle]: v6/t/regex.t
22:32 clkao the harness is still in perl5. should be ported to perl6
22:32 [particle] that's what i'm doing
22:33 Khisanth joined perl6
22:33 clkao excellent
22:33 [particle] perl6 harness means written in perl 6
22:33 [particle] :)
22:33 [particle] i don't see a v6/t/regex.t
22:34 clkao Pugs-Compiler-Perl6/t/
22:34 clkao sorry my eyes skipped the 'perl6' ;)
22:34 [particle] np
22:34 [particle] pugs is a huge somewhat disorganized (to me) project, so it takes me a while to find what i'm looking for
22:35 clkao if only we have a source code index engine....
22:35 [particle] krugle? :)
22:35 weinig is now known as weinig|away
22:35 clkao Enter Kasago
22:35 mdiep lxr?
22:36 clkao Kasago is on CPAN
22:39 mdiep util/prove6 has "#!/bin/perl". ew.
22:40 clkao uhm i didnt do it. it started as a shell script just running perl with some env
22:46 svnbot6 r11908 | Sage++ | Fixed code that resulted in errors when head was called on an empty list, along with other sanity checks.
22:46 svnbot6 r11908 | Sage++ | Revisions to regex handling, both in the general case and count modifiers. Regex errors also now give more informative error messages.
22:46 svnbot6 r11908 | Sage++ | With these fixes, 95%+ of perl5 t/op is handeled, with the only remaining errors being in parsing: a couple of regex problems and a very infuriating general error that I haven't been able to track down.
22:53 SageLT audreyt: you around?
23:04 [particle] what the heck did $@ become in p6?
23:04 [particle] i can't find it in the synopses
23:05 clkao $!
23:05 clkao always $!
23:06 clkao $? $@ and $! are all $
23:06 clkao $!
23:06 [particle] great, i wasn't sure which one it always was
23:14 clkao $? for system $@ for eval and $! for builtins
23:14 clkao all confusing
23:14 [particle] ok( ! $! ?& $match, $description ); $! and diag $!; # :)
23:20 clkao excuse me, are you cursing?
23:21 [particle] perl6 is a vulgar language :)
23:22 _mago joined perl6
23:23 [particle] think this incantation will work?...
23:23 [particle] sub p6rule_is( $target, $pattern, $description?, :%todo ) {
23:23 [particle]    my $match = $target ~~ /$pattern/;
23:23 [particle]    ok( ! $! ?& $match, $description );
23:23 [particle]    $! and diag $!;
23:23 [particle] }
23:24 [particle] i should add %todo to the end of ok() there
23:24 [particle] heck, i guess i could use like() instead of ok
23:25 clkao might be, i am guess the current v6 should already suppor tthose.  not sure about ?&
23:25 [particle] like( $target, $pattern, $desc...)
23:25 [particle] ?& is easy to write, if it doesn't support it.
23:25 lambdabot Maybe you meant: . v
23:26 [particle] lambdabot: bite me
23:26 clkao yep. i was going to suggest like. but iirc there's something about checking the match as well.
23:26 [particle] well, i can leave it as is for now
23:27 clkao commit it somewhere and lets see if it works?
23:27 [particle] it's a good start, and once i get v6 working , i can test it
23:27 [particle] hrmm, ok
23:31 [particle] 11909: t/p6regex/p6regex.t
23:31 svnbot6 r11909 | particle++ | [t]: beginning of a perl6 test harness for parrot's p6regex tests
23:46 svnbot6 r11910 | particle++ | [t]: p6regex adds and changes, still not passing tests
23:52 clkao rx// is the same as // right?
23:52 [particle] i think so
23:53 [particle] i'm getting no such sub &p6rule_is ... with the haskell backend
23:54 [particle] perhaps it doesn't like the named / optional params
23:57 clkao (!)

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

Perl 6 | Reference Documentation | Rakudo