Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2007-02-09

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 Aankhen`` joined perl6
00:05 svnbot6 r15221 | putter++ | misc/pX/Common/yet_another_regex_engine - some non-demonstrable progress.
00:08 Psyche^ joined perl6
00:10 Psyche^ is now known as Patterner
00:15 jferrero joined perl6
00:35 gnuvince joined perl6
00:39 svnbot6 r15222 | lwall++ | pugs parseability tweaks
00:48 svnbot6 r15223 | lwall++ | better workaround
00:52 nipotaway is now known as nipotan
01:03 dduncan joined perl6
01:03 MikeJS joined perl6
01:21 allbery_b amusingly that last typo fix missed a typo in the next line:  "triger"
01:23 rashakil_ is now known as rashakil
01:32 dmq joined perl6
01:57 svnbot6 r15224 | putter++ | misc/pX/Common/yet_another_regex_engine - minor progress.  --repl is currently speaking a P5WithSubrules dialect.
02:13 zev_ joined perl6
02:23 Aankhen`` joined perl6
02:26 thepler joined perl6
02:26 masak_ joined perl6
02:27 Debolaz joined perl6
02:27 Debolaz http://en.wikipedia.org/w/index.php?title=Pugs&curid=1532361&diff=106736305&oldid=106069367 heh, I wish they'd be more creative at least.
02:27 lambdabot http://tinyurl.com/298n2s
02:36 svnbot6 r15225 | putter++ | misc/pX/Common/yet_another_regex_engine/ - Now, to play with a code fork, just copy a single file.  You can use the same tests.  Thus it is easy to check in and share your explorations.  See README.
02:39 svnbot6 r15226 | putter++ | cleanup after r15225 - svn didn't like this directory removal being part of the revision.
02:45 [particle] joined perl6
02:53 mjk joined perl6
03:00 rashakil_ joined perl6
03:03 svnbot6 r15227 | audreyt++ | * Pugs.Types.Array: @array.delete(1) should remove the
03:03 svnbot6 r15227 | audreyt++ |   binding (i.e. same as @array[1] := ::Scalar()), not
03:03 svnbot6 r15227 | audreyt++ |   the same as assigning undef into @array[1].
03:15 [particle] joined perl6
03:15 bucky joined perl6
03:20 mrossetti joined perl6
03:23 bsb joined perl6
03:43 Psyche^ joined perl6
03:43 Psyche^ is now known as Patterner
03:48 putter joined perl6
03:51 putter Anyone know what the story is regards ':' in S05?   Is : as a sequence commit operator gone?  Having it as a quantifier commit is nice, but doesn't seem to subsume its former role.
03:52 TimToady what do you mean by a sequence commit operator?
03:53 TimToady it has always applied only to the previous token.
03:55 TimToady previous atom, I should say
03:56 putter (?>...)...  => ... : ...
03:56 putter no?
03:56 TimToady => [...]:
03:56 putter np
03:58 putter what was the motivating reason for attaching it to the atom, vs being standalone like :: and ::: ?
04:00 svnbot6 r15228 | audreyt++ | * log.t: unTODO
04:01 TimToady well, that's a smaller scope than [...], and often handy
04:01 TimToady and the same scope as quantifiers
04:03 putter pity it makes incremental commit brackety.
04:04 putter [[[ ... ]: ... ]: ... ]: ...   vs ... : ... : ... : ...
04:04 TimToady :: works for that
04:05 TimToady 'sides, we have :ratchet
04:05 audreyt TimToady: the nofat rule.
04:05 TimToady ?
04:05 TimToady I noticed pugs flunked "use =>"
04:05 svnbot6 r15229 | audreyt++ | * arrays.t: Fix bogus test.
04:05 audreyt it has \h
04:06 audreyt er, it no longer
04:06 putter re ::, ?    [...]: ...  || ...  !=  ... :: ... || ...
04:06 audreyt as of r15230
04:06 justatheory joined perl6
04:06 putter (assuming I'm correctly remembering || is the p5 | )
04:06 nipra joined perl6
04:06 svnbot6 r15230 | audreyt++ | * Pugs.Lexer: Implement the <nofat> regex so:
04:06 svnbot6 r15230 | audreyt++ |     use => 1
04:06 svnbot6 r15230 | audreyt++ |   etc can work.
04:07 TimToady good, though I worked around it
04:07 TimToady ?eval "if => 1"
04:07 evalbot_r15228 "if => 1"
04:07 TimToady ?eval if => 1
04:07 audreyt pugs> if => 1
04:07 audreyt ("if", 1)
04:07 evalbot_r15228 Error: ␤Unexpected ">"
04:08 TimToady 228
04:08 putter re :ratchet,   ... : ...  !=   . : . : . : . : . : .     ;)
04:08 audreyt anyway, so I'm wondering, isn't that essentially reparse?
04:08 audreyt as the \h* is different as used in the normal \s* after the "if"
04:08 audreyt <ws>, even
04:08 audreyt which means the spaces in "if =>1" needs to be parsed twice
04:09 TimToady yes, I'm assuming that branch will usually fail completely
04:09 audreyt ok...
04:09 TimToady putter, if everything that can be backtracked is :, then its the same as backing all the way out anyway
04:09 audreyt in Pugs.Lexer I merged with <ws> handling to avoid reparse
04:10 audreyt and so "if\n=>1" would be allowed
04:10 audreyt I wonder what's the downside of that?
04:10 audreyt (aside from not being faithful to perl5)
04:10 putter no doubt I'm missing something obvious, but...
04:10 TimToady eh, p5 doesn't care
04:11 TimToady it's just a comma
04:11 audreyt $ perl
04:11 audreyt if
04:11 audreyt => 1
04:11 audreyt syntax error at - line 2, near "=>"
04:11 audreyt $ perl
04:11 audreyt if => 1
04:11 audreyt $
04:11 TimToady interesting
04:12 audreyt ditto "use"
04:12 audreyt so p5 does care
04:12 putter Given  <a> <b> : <c> || <d>   it seems the only correct transform is the first mentioned   [ <a> <b> ]: <c> || <d>
04:12 putter all the other arguments seem to be "yes, incremental commit is brackety, but given this other
04:12 putter stuff, it won't come up that often".
04:13 putter what am I missing?
04:13 TimToady :: already does what you want
04:13 TimToady : is aimed more as possessive quantification
04:14 putter :: commits the surrounding alternation, no?    In the example, a failing <c> should not preclude <d> being tried.
04:17 TimToady hmm, yeah, but it still seems to me that :ratchet takes away most of your use case
04:18 putter could be.  I haven't pondered.  always just assumed : was like :: and :::.
04:18 TimToady and there's certainly no guarantee that just because you didn't want to backtrack <b> that you don't want to backtrack <a>
04:18 audreyt was like that initially
04:19 TimToady .*? \d+: for instance
04:19 TimToady well, maybe that's a bad example
04:19 TimToady .*? foo \d+:
04:19 putter re no guarantee... oo, good point.  that could be worth trading bracketyness for
04:20 TimToady and I got my fill of brackets with (?>...)   :)
04:21 TimToady 'course, now that we actually put in possessive +, we could write
04:21 TimToady .*? foo \d++
04:22 TimToady but that doesn't work for brackets
04:23 TimToady [...]+ means something else
04:23 putter re "got fill", err, we still have [[[ ... ]: ... ]: ... ]: ... ugliness.  we're just saying it comes up less frequently in p6, and it's worth the pain in order to do   ... [...]: ...  which otherwise would be painfully like ... [...] : ...
04:23 TimToady you'd have to write things like [...:]
04:23 TimToady which just looks wrong
04:24 TimToady and is terribly ambiguous if there's | inside too
04:24 putter yeah :(
04:24 mako132_ joined perl6
04:25 TimToady In the whole P6 grammar I've not wanted : once, as far as I recall.
04:25 TimToady but then I'm relying on :ratchet and longest-token-ish-ness
04:25 putter the whole p6 grammar is designed to be non-backtracking...
04:25 TimToady well, there is that.. :D
04:26 putter lol
04:26 TimToady though I still haven't tackled a non-backtracking [email@hidden.address] matcher
04:27 TimToady since you can't decide whether it's interpolating till you see the ()
04:27 TimToady one could hold things in two states "@obj" is array or string...
04:28 TimToady the heredoc whitespace trimmer has to do things after the fact as well.
04:29 TimToady and as audreyt was pointing out <ident> \h* => is also a bit of a problem.
04:30 audreyt I wonder if hash composer can be moved from circum {} to circum :{}
04:30 audreyt and leave circum {} to code.
04:31 audreyt (that is also a postprocess point, and disambig diagnostics surrounding it)
04:32 audreyt eg the one STD.pm used to trigger: "Closing hash curly may not terminate a line;\nplease add a comma or a semicolon to disambiguate"
04:32 TimToady well, %[...] might work as well
04:33 TimToady we talked about this some the other day, but I haven't come to any conclusion yet.
04:33 audreyt sure, but was looking at notational space of null-keyed pairs in p6 terms
04:33 audreyt since they have proved so useful as rx terms :)
04:33 TimToady heh
04:33 TimToady except we've got :{...} adverbial blocks, I think
04:33 * putter likes a policy of "given two equally pretty alternatives, only one non-backtracking, choose that.  but don't make the language uglier just to make the grammar non-backtracking.  nothing much is gained, as the parser will have to deal with user modifications inducing backtracking anyway."  no?
04:34 TimToady yeah, on the ident => thing, it actually only backtracks if it sees the =>, because it has to uncommit the keyword in that case.
04:35 Psyche^ joined perl6
04:35 Psyche^ is now known as Patterner
04:36 audreyt ah right. because \h* is subset of <ws> not the other way around
04:36 audreyt ok, fine then
04:37 * audreyt goes back to Fast MO Dispatch
04:37 jiing joined perl6
04:38 audreyt but maybe I should get trac set up first.
04:39 TimToady and maybe mappings should just autopromote to hashes with sufficient provocation.
04:39 audreyt like assigning into a scalar?
04:40 audreyt (analogous to List/Array)
04:40 TimToady dunno, but definitely assigning to anything that doesn't care about the order
04:40 audreyt so Mapping composer becomes %() and Hash composer becomes %[]
04:41 TimToady lemme stew on it for a while more
04:42 audreyt sure, same for the .:delete
04:49 amnesiac joined perl6
04:50 jdv79 joined perl6
05:14 drbean joined perl6
05:27 nipra joined perl6
05:37 bsb joined perl6
05:39 amnesiac joined perl6
05:51 nipra joined perl6
05:55 svnbot6 r15231 | putter++ | project_planning/TASKS - Added a new task: Snarf new parrot regex tests into pugs t/regex/p6regex/.
05:55 svnbot6 r15231 | putter++ | project_planning/TASKS_DONE - created.
05:55 svnbot6 r15231 | putter++ | Moved a done task from TASKS to TASKS_DONE file. moritz++
06:03 cnhackTNT joined perl6
06:13 BooK joined perl6
06:23 jisom joined perl6
06:24 justatheory joined perl6
06:35 buetow joined perl6
06:36 bsb left perl6
06:51 jisom_ joined perl6
06:54 rfordinal joined perl6
07:21 leed joined perl6
07:58 sarg joined perl6
07:58 sarg left perl6
08:03 iblechbot joined perl6
08:04 audreyt help wanted!
08:05 audreyt does dev.pugscode.org resolve/display currently (DNS just setup)?
08:06 tene resolves for me
08:06 allbery_b works for me
08:06 tene "Pugs - Trac"
08:07 gaal works here too.
08:08 audreyt try login with commitbit
08:08 audreyt click "Login"
08:08 audreyt and see if admin pages like http://dev.pugscode.org/admin/ticket/components display correctly
08:10 pdcawley joined perl6
08:10 audreyt @tell putter is it okay if we port project_planning/ into dev.pugscode.org? it's a bit more structured there, and it'd be great if we populate sufficient tickets/milestone/version/component metadata (use the "Admin" link on the left) so people can see a coherent "Roadmap" in http://dev.pugscode.org/roadmap
08:10 lambdabot Consider it noted.
08:11 allbery_b dislikes my password
08:11 gaal i can't login either.
08:12 gaal iirc commitbit username ~~ email, unlike openfoundry
08:12 gaal but that neither works.
08:12 audreyt hm.
08:15 audreyt fixing
08:18 audreyt aha, bad PAM interaction
08:19 audreyt try again?
08:19 audreyt (should work this time)
08:19 gaal yes
08:19 gaal userid, not email
08:19 audreyt userid is correct
08:20 gaal *nod*
08:20 audreyt maybe edit homepage to say so? :)
08:20 allbery_b yep
08:20 gaal btw is there a web svn diff?
08:20 gaal will edit
08:21 gaal "logging in with your Pugs committer id" <- already says so :)
08:21 gaal the confusion's only possible because commitbit uses the email.
08:22 audreyt yes, web svn diff
08:22 gaal oh found it, it's nice
08:22 audreyt going to allow snapshot tgz download. sec
08:23 allbery_b when I went to your admin url I get a page saying "internal error / no admin pages available
08:23 gaal oh, it's *excellent* even, with its sleek multifile diff and option for unidiff
08:25 gaal i don't even see an admin link. :)
08:25 allbery_b <audreyt> and see if admin pages like http://dev.pugscode.org/admin/ticket/components display correctly
08:25 allbery_b I saw no admin link either
08:26 allbery_b of course that may just mean I'm not an admin (which would make sense...)
08:26 audreyt huh.
08:26 audreyt oh well, instead of separating admin vs committer
08:26 audreyt I might as well ignore commitbit settings and make all authors admins
08:26 audreyt anarchy++
08:27 audreyt done
08:27 audreyt try again
08:27 clkao audreyt!
08:27 audreyt a http://dev.pugscode.org/admin should magickally appear
08:27 gaal i like how some components have "somebody" as their owner and some don't got nobody.
08:27 rgs joined perl6
08:27 audreyt clkao: heya! I've got this GoodEnough(tm) Trac 0.10 set up!
08:27 audreyt as compared to, e.g., a customized hiveminder
08:28 clkao audreyt: what do you mean? also would you like to visit me this sunday
08:28 audreyt clkao: http://dev.pugscode.org/ (login with your pugs id)
08:28 lambdabot Title: Pugs - Trac
08:28 audreyt clkao: no, got family gathering sorry
08:29 audreyt @tell putter the Admin link should work now - when I get timeslice I'll be happy to chat about org, but for now I think porting tickets over would make (some) sense? also add new components / milestones / etc as needed
08:29 lambdabot Consider it noted.
08:30 Aankh|Clone joined perl6
08:34 audreyt hm, automagic RSS feeds. could be useful
08:34 audreyt (the one at bottom of http://dev.pugscode.org/timeline)
08:34 lambdabot Title: Timeline - Pugs - Trac
08:37 gaal putting up putter++'s christmas graph there could be nice
08:39 lumi It works, I can login, and the admin pages show
08:40 audreyt cool. it might be useful to shop around http://trac-hacks.org/ and install useful things in http://dev.pugscode.org/admin/general/plugin
08:40 lambdabot Title: Trac Hacks - Plugins Macros etc. - Trac
08:40 audreyt but mostly it's about content now
08:41 audreyt I really need to find food (and maybe work on Monday's slides) -- but hopefully this can get us a bit more transparency
08:41 audreyt GHC was impossible to hack, but ever since http://cvs.haskell.org/trac/ghc/wiki/Commentary is up
08:41 lambdabot Title: Commentary - GHC - Trac
08:42 gaal whee cool: http://trac-hacks.org/wiki/GraphvizPlugin
08:42 lambdabot Title: GraphvizPlugin - Trac Hacks - Plugins Macros etc. - Trac
08:42 audreyt hacking suddenly became much much easier
08:42 audreyt lifting the GHC commentary's outline, or something like it, to dev.pugscode may be fun too
08:43 audreyt gaal: you can install plugins remotely by uploading the PAR file
08:43 audreyt er, I mean EGG file.
08:43 gaal cool. uh, i forget where putter's dag is
08:43 audreyt conveniently it's linked to from the wiki home
08:43 audreyt http://dev.pugscode.org/browser/project_planning/a_dependency_graph.png
08:43 lambdabot Title: /project_planning/a_dependency_graph.png - Pugs - Trac, http://tinyurl.com/yu9fvz
08:43 audreyt http://dev.pugscode.org/browser/project_planning/a_dependency_graph.dot
08:44 lambdabot Title: /project_planning/a_dependency_graph.dot - Pugs - Trac, http://tinyurl.com/24btwq
08:44 gaal excellent!
08:44 gaal conveniently it's in .dot :)
08:45 iblechbot_ joined perl6
08:46 drrho joined perl6
08:48 NewCoke_ joined perl6
08:49 gaal ok I know zero python. to build the gv plugin, I need "setuptools" installed. how do I do that on debian?
08:49 audreyt it's already.
08:49 audreyt but shouldn't you just download the .egg to your laptop
08:49 audreyt and then upload it again in the plugins page?
08:49 moritz aptitude install python-setuptools
08:49 gaal ImportError: No module named setuptools
08:49 moritz I'd guess ;)
08:50 gaal is there a binary egg somewhere?
08:50 gaal don't see it
08:51 audreyt oh I see, you need to make an egg first
08:51 gaal yeah okay found alink w/ explanations.
08:53 audreyt platform-neutral bytecode really is goodthing
08:53 gaal yay, that was really easy.
08:53 gaal for reference:
08:53 gaal wget http://peak.telecommunity.com/dist/ez_setup.py
08:53 gaal sudo python ez_setup.py
08:53 gaal python setup.py bdist_egg
08:55 audreyt and find the egg under dist/
08:55 gaal yeah. thoguh now the upload fails, with no meaningful error message.
08:55 gaal but yeah, i'll moose it from here.
08:56 audreyt the PAR equivalent is simply "parl -p"
08:56 audreyt but PAR::Repository is not really so well-adopted or supported
08:57 audreyt let me check errorlog
08:58 gaal was looking for it. where's trac instlaled?
08:58 gaal on feather that is
08:58 gaal brb
08:58 audreyt heh, bad typo on part of trac
08:58 audreyt it's in /usr/share/trac
09:00 dduncan with this talk about GHC 6.8, any guesses around when it'll be out?
09:01 dduncan or should I say, does the current 6.7 do that stuff?
09:01 dduncan about GPUs
09:02 dduncan or never mind
09:03 audreyt gaal: works now.
09:04 gaal thanks.
09:04 audreyt dduncan: http://haskell.org/haskellwiki/GHC/Data_Parallel_Haskell
09:04 lambdabot Title: GHC/Data Parallel Haskell - HaskellWiki
09:04 dduncan ok
09:04 audreyt in short: not really GPUs, but true SMP linear-scale parallelism; also SSE instructions is in
09:05 audreyt the ndp library is not yet core, but if you install it as third-party then you can already use those data structures
09:05 pfarmer joined perl6
09:05 audreyt the plan is to make it core and then map the native syntax support for GHC parallel-arrays, so our usercode don't need to change
09:06 audreyt it just automagically gets parallelism
09:06 gaal audreyt: where does it put the plugin/conf file? it's not in /usr/share/trac/plugins etc.
09:06 dduncan ok
09:06 audreyt I'd say another month or two before that's generally stable
09:06 audreyt gaal: /data/svn/trac
09:06 gaal thanks
09:06 audreyt np
09:06 audreyt bbiab
09:07 audreyt gaal: the bug was, very sillily: http://trac.edgewall.org/attachment/ticket/3168/plugin_upload.patch
09:07 lambdabot Title: #3168: plugin_upload.patch - The Trac Project - Trac, http://tinyurl.com/2735tq
09:07 audreyt but trac++ to make it instantly findable
09:09 marmic joined perl6
09:14 dduncan fyi, I would expect that this low-level support for parallelism should give my work a large speed boost, which deals heavily with unordered parallel collections
09:15 gaal heh.
09:15 gaal three mo ago? hmm
09:15 audreyt dduncan: as long as you use hyperops and junctions the win should be automatic
09:15 dduncan yes
09:16 gaal how do i restart trac's apache?
09:16 audreyt gaal: you tell me to
09:16 gaal (I needed to edit trac.ini)
09:16 gaal audreyt: pls restart trac apache k thx
09:16 audreyt # sudo /etc/init.d/apache restart
09:16 devogon joined perl6
09:16 audreyt already did
09:16 gaal thanks :)
09:16 audreyt :)
09:17 audreyt oh hey I see a graphviz plugin
09:17 audreyt that's very moose
09:17 gaal uh how do i create a new page on the wiki?
09:17 dduncan I don't know if junctions specifically are very applicable to what I'm doing, but working with sets is very relevant
09:18 audreyt gaal: you link to it?
09:18 dduncan I also plan to be writing in functional notation as much as possible
09:18 audreyt eg http://dev.pugscode.org/wiki/ThisPageIsNotHere
09:18 allbery_b typically you just try to go to it
09:18 dduncan or maybe I will be using junctions a lot, or at least junction syntax
09:18 audreyt also part of GHC 6.8 is the integration of an event model
09:19 audreyt instead of like everybody else where event-queue and thread-scheduler are two subsystems, they are now unified
09:20 audreyt so we can have a "view" of arbitrarily low-level async events, and a "view" of convenient thread-STM abstractions, both describing the same things
09:20 audreyt it's like Coro::Event but reentrant and nonblocking and SMP :)
09:20 dduncan in any event, if things go to plan, I should have a Perl 5 prototype of what I want to do within a few weeks, and then we can figure out the best Perl 6 syntax to represent the same work
09:21 audreyt dduncan: excellent.
09:21 Caelum so there'll be awesome POE thingie for perl6?
09:21 dduncan the latter should be a lot more concise
09:21 audreyt Caelum: that seems to follow naturally, yes
09:21 audreyt need to figure out the event interface's userland ergonomics
09:21 Caelum cool :)
09:21 audreyt but that's @Larry's forte
09:22 tene audreyt: you have a convenient link describing the event model stuff?
09:22 audreyt (the STM vocabluary we ended up (contend/maybe/defer) is lovely)
09:22 audreyt tene: sure, http://www.seas.upenn.edu/~lipeng/homepage/unify.html
09:22 lambdabot Title: Unifying events and threads
09:22 tene audreyt: thanks
09:22 tene audreyt++
09:23 dduncan good night
09:23 dduncan left perl6
09:25 gaal http://dev.pugscode.org/wiki/ProjectDependencyGraph
09:26 lambdabot Title: ProjectDependencyGraph - Pugs - Trac
09:26 gaal but we have to find a way to this automatically
09:27 gaal also, if there's a nicer way to display dot files in browsers than huge pngs, that'd rock too.
09:32 audreyt ZGRViewer surely?
09:33 Aankhen`` joined perl6
09:36 buetow joined perl6
09:37 gaal cute!
09:39 * smash good morning
09:40 buu Explosive!
09:41 gaal this looks promising: http://zvtm.sourceforge.net/zgrviewer/applet/
09:41 lambdabot Title: claribole.net - ZGRViewer Applet
09:41 gaal but the svg must be on the same server as the applet
09:42 gaal I <3 how they avoid requiring graphviz clientside by parsing svg :)
09:43 gaal bbiab&
09:46 buubot joined perl6
09:57 elmex joined perl6
10:02 pdcawley joined perl6
10:03 Grub_Now joined perl6
10:03 Grub_Now left perl6
10:15 GabrielVieira2 joined perl6
10:26 Aankhen`` That "Unifying events and threads" paper made my head explode.
10:46 ruoso joined perl6
10:46 jferrero joined perl6
11:17 foo\ joined perl6
11:20 devogon joined perl6
11:36 ofer0 joined perl6
11:39 jferrero joined perl6
11:39 alexg joined perl6
11:58 devogon joined perl6
12:01 TimToady joined perl6
12:08 VanilleBert joined perl6
12:12 kanru joined perl6
12:14 chris2 joined perl6
12:31 orafu joined perl6
12:35 rashakil joined perl6
12:46 cjeris joined perl6
12:46 iblechbot joined perl6
13:08 Schwern joined perl6
13:09 nipotan is now known as nipotaway
13:20 Limbic_Region joined perl6
13:37 bonesss joined perl6
13:37 bonesss joined perl6
13:46 VanilleBert left perl6
13:58 trym__ joined perl6
13:59 jferrero joined perl6
14:13 neonse joined perl6
14:27 vel joined perl6
14:39 polettix joined perl6
14:39 neonse joined perl6
14:39 Schwern joined perl6
14:39 cjeris joined perl6
14:39 rashakil joined perl6
14:39 devogon joined perl6
14:39 ruoso joined perl6
14:39 GabrielVieira2 joined perl6
14:39 rgs joined perl6
14:39 rfordinal joined perl6
14:39 BooK joined perl6
14:39 drbean joined perl6
14:39 jdv79 joined perl6
14:39 Patterner joined perl6
14:39 [particle] joined perl6
14:39 thepler joined perl6
14:39 sri_ joined perl6
14:39 stevan__ joined perl6
14:39 Caelum joined perl6
14:39 kolibrie joined perl6
14:39 lumi joined perl6
14:39 drupek12167 joined perl6
14:39 Southen joined perl6
14:39 araujo joined perl6
14:39 jcrigby joined perl6
14:39 pnu joined perl6
14:39 nekokak_ joined perl6
14:39 spinclad joined perl6
14:39 baest joined perl6
14:39 spo0nman joined perl6
14:39 stef_ joined perl6
14:39 mj41 joined perl6
14:39 tokuhirom joined perl6
14:39 beppu_ joined perl6
14:39 allbery_b joined perl6
14:39 pasteling joined perl6
14:39 Khisanth joined perl6
14:39 qmole_ joined perl6
14:39 mugwump joined perl6
14:39 integral joined perl6
14:39 f0rth joined perl6
14:39 SamB joined perl6
14:39 takanori joined perl6
14:39 Daveman joined perl6
14:39 diotalevi joined perl6
14:39 rafl joined perl6
14:39 TreyHarris joined perl6
14:39 xern joined perl6
14:39 Odin- joined perl6
14:39 SCalimlim joined perl6
14:39 idiotmax_ joined perl6
14:39 awwaiid joined perl6
14:39 dvorak joined perl6
14:39 tene joined perl6
14:39 ingy joined perl6
14:39 evalbot_r15228 joined perl6
14:39 yves joined perl6
14:39 takesako__ joined perl6
14:39 nnunley joined perl6
14:39 broquaint joined perl6
14:39 LCamel_ joined perl6
14:39 Gothmog_ joined perl6
14:39 jamessan joined perl6
14:39 gugod joined perl6
14:39 arguile joined perl6
14:39 mtve joined perl6
14:39 buu joined perl6
14:39 GeJ joined perl6
14:39 clkao joined perl6
14:39 obra joined perl6
14:39 nothingmuch joined perl6
14:39 Shabble joined perl6
14:39 trym__ joined perl6
14:39 cls_bsd joined perl6
14:39 spoop joined perl6
14:39 lisppaste3 joined perl6
14:39 beppu__ joined perl6
14:39 Grrrr joined perl6
14:39 merlyn joined perl6
14:39 avar joined perl6
14:39 DebolazX joined perl6
14:39 knewt joined perl6
14:39 Yappo joined perl6
14:39 perlbot joined perl6
14:39 jql joined perl6
14:39 felipe joined perl6
14:39 cj joined perl6
14:44 mdiep joined perl6
14:49 iblechbot joined perl6
14:59 cnhackTN1 joined perl6
15:08 mdiep joined perl6
15:10 kkosmo joined perl6
15:12 elmex joined perl6
15:24 orafu joined perl6
15:32 woremacx joined perl6
15:44 pjcj joined perl6
15:51 hexmode joined perl6
15:54 hcchien joined perl6
16:10 iblechbot joined perl6
16:10 svnbot6 r15232 | moritz++ | Added testcases for print with explicit filehandle $*DEFOUT
16:17 mrossetti joined perl6
16:33 _jcrigby joined perl6
16:53 justatheory joined perl6
16:53 moritz ?eval (1, 2, 3).isa("Array")
16:53 evalbot_r15228 is now known as evalbot_r15232
16:53 evalbot_r15232 Bool::True
16:53 moritz hey, isn't that a list but not an array?
16:53 rindolf joined perl6
16:54 rindolf Hi all!
16:54 moritz hi rindolf ;)
16:54 rindolf How do I evaluate perl6 code here?
16:54 rindolf Hi moritz
16:54 rindolf moritz: what's up?
16:54 moritz ?eval say "Hello, reindolf"
16:54 evalbot_r15232 OUTPUT[Hello, reindolf␤] Bool::True
16:54 rindolf moritz: thanks!
16:54 moritz rindolf: np ;)
16:54 [particle] if you want to do a lot of playing around, use run.pugscode.org
16:55 moritz [particle]: is (1, 2) a list, an array or both?
16:55 [particle] that way you don't flood the channel with evals
16:55 [particle] moritz: i think the answer is yes due to autopromotion
16:56 moritz damned, I shouldn't ask "or"-questions ;)
16:56 moritz ok
16:58 rindolf [particle]: hmmm... I get weird formatting errors there on Konqi 3.5.6. Let me try in Firefox.
17:00 stevan_ joined perl6
17:01 rindolf print $s.uc();
17:01 rindolf pugs>  Sorry, that took too long! Aborted.
17:07 svnbot6 r15233 | moritz++ | lc/uc test cases with German Umlauts
17:10 TimToady yes, a list promotes to array in scalar context, and HERE.foo is a scalar context
17:11 moritz TimToady: so .isa("List") and .isa("Array") are always the same?
17:12 rindolf Hmmm... print $s; reports that it takes too long in http://feather.perl6.nl:8080/runpugs/?
17:12 lambdabot Title: Run Perl 6 Now -- in your browser!
17:13 TimToady well, at the moment they seem to be the same, but it's certainly something that can be tweaked as we go along.
17:13 TimToady ?eval (1,2,3).WHAT
17:13 evalbot_r15232 is now known as evalbot_r15233
17:13 evalbot_r15233 ::Array
17:13 TimToady ?eval WHAT(1,2,3)
17:13 evalbot_r15233 Error: No compatible subroutine found: "&WHAT"
17:13 TimToady heh
17:14 rindolf ?eval WHAT((1,2,3))
17:14 evalbot_r15233 ::Array
17:14 TimToady ?eval WHAT (1,2,3)
17:14 evalbot_r15233 ::Array
17:14 TimToady ?eval WHAT \(1,2,3)
17:14 evalbot_r15233 ::Val
17:15 TimToady hmm, should be Capture
17:15 TimToady WHAT :($a,$b,$c)
17:16 TimToady ?eval WHAT :($a,$b,$c)
17:16 TimToady dur
17:16 evalbot_r15233 ::Sig
17:16 [particle] you are in a twisty maze of disgarded sigils, all alike
17:16 Ricardo joined perl6
17:17 TimToady how significant
17:18 Ricardo Hi everyone i have a question about unpack. I have read the documentation, i know how it works within PERL. But i got to get unpack working in a C# environment.
17:18 Ricardo Does anyone know where i might be able to find more information on the internal workings of unpack
17:19 moritz Ricardo: the source is publicly available ;)
17:19 TimToady joined perl6
17:19 Ricardo :) im not a low level programmer
17:19 devogon joined perl6
17:22 svnbot6 r15234 | moritz++ | Mentioned wantarray() in Perl6::Perl5::Differences
17:28 justatheory_ joined perl6
17:30 prism joined perl6
17:31 ruz joined perl6
17:32 bernhard joined perl6
17:33 amnesiac joined perl6
17:37 [particle] Ricardo: perhaps this link and others like it can get you on your way: http://www.thescripts.com/forum/thread127484.html
17:37 lambdabot Title: C++ IO (or equivalency of perl's pack/unpack?) - C++
17:40 Dr_Pi joined perl6
17:43 rindolf How do I get an account on feather?
17:45 [particle] http://feather.perl6.nl/ -- "see the original announcement for details"
17:45 lambdabot Title: Perl6 Community Development Server
17:49 fglock joined perl6
17:56 svnbot6 r15235 | fglock++ | kp6 - added t/01-sanity, edited to fit the kp6 grammar
17:56 svnbot6 r15235 | fglock++ | - added the sanity errors to TODO list
17:56 svnbot6 r15235 | fglock++ | - fixed control block pads, logical ops
17:56 svnbot6 r15235 | fglock++ | - added 'coro', 'hypotheticals' prototypes to docs/
17:56 svnbot6 r15235 | fglock++ | - roles
17:56 svnbot6 r15235 | fglock++ | - added 'does' and 'is' class traits
17:56 svnbot6 r15235 | fglock++ | - no longer uses Class::MOP
17:57 xinming joined perl6
18:05 nipra joined perl6
18:15 wilx` joined perl6
18:25 GabrielVieira joined perl6
18:31 wilx` is now known as wilx
18:38 trym_ joined perl6
18:55 jisom joined perl6
18:57 jisom joined perl6
19:01 NewCoke_ (hiveminder) yet another site blocked by work. :|
19:01 NewCoke_ while looking at event models for perl6, be sure to peek at tcl's.
19:03 NewCoke_ (two reasons: if perl6 has something similar, I can be sure partcl eventually gets it, and I've always been happy with tcl's event handling. it JFW.)
19:04 justatheory joined perl6
19:21 iblechbot joined perl6
19:21 NewCoke_ hurm. the synopses don't really introduce "say", do they. just jump right in there.
19:27 renormalist joined perl6
19:32 buetow joined perl6
19:38 renormalist Hi all! Can someone re-assign me my commit bit. I had one before the svn change. I'm "renormalist" on svn.
19:38 renormalist please, of course :-)
19:40 TimToady looking
19:43 rindolf joined perl6
19:43 TimToady sent, assuming email address is the same...
19:43 TimToady at webit
19:43 renormalist k, looking
19:46 renormalist nice, it works again, thank you
19:46 svnbot6 r15236 | renormalist++ | - just commit bit test, sorry for svn log spamming
19:47 rindolf OK, finished checking out r15235 - now to compile pugs.
19:48 TimToady you should check out r15236 now.  :)
19:48 rindolf TimToady: OK.
19:49 rindolf :-)
19:49 rindolf r15236 | renormalist++ | - just commit bit test, sorry for svn log spamming
19:50 bonesss joined perl6
19:51 renormalist Anyone using svk here on pugs?
19:51 svnbot6 r15237 | renormalist++ | - cleanup comments that mark perl6 related changes
19:53 * moritz not
19:56 rindolf In this presentation - http://www.szabgab.com/perl6.html - all the Perl 6 code samples don't have "use v6;" at the top. Is it a good idea to omit it?
19:56 lambdabot Title: Perl6
19:56 Juerd rindolf: Check thy mail
19:56 * Juerd thinks it's a good idea to omit "use v6;"
19:57 Juerd But not a bad idea to include it.
19:57 TimToady for a presentation it's fine to omit it just to save room
19:57 rindolf Juerd: OK, thanks.
20:02 rindolf Juerd: ok, I logged in and changed my password.
20:05 Juerd Note to feather users: feather will be subjected to password cracking attempts, next week.
20:05 Juerd (This info is also in the motd)
20:20 moritz during `make smoke-js` I get messages like this:
20:20 moritz # runjs.pl  t/builtins/math/rounders.t
20:20 moritz open2: exec of js failed at /mnt/ex/moritz/pugs/perl5/PIL2JS/lib/PIL2JS.pm line 167
20:20 shay hello folks
20:21 shay hey Juerd :)
20:21 shay rindolf, shlomi
20:21 moritz what's wrong? any ideas?
20:21 moritz hey shay ;)
20:21 rindolf Hi shay
20:21 shay moritz, hey there, what's up :)
20:21 offby1 joined perl6
20:22 offby1 is there an easy way to dump both the name of a variable, and its value?  I find myself doing      say "Foo:"; say Foo.yaml;   a lot.
20:22 moritz shay: not much ;)
20:22 rindolf offby1: using eval?
20:23 rindolf Many of the pugs' tests fail on my system after I ran make.
20:23 moritz rindolf: how many?
20:24 rindolf moritz: I don't know, I haven't finished running all of them.
20:24 moritz I just smoked and had 97.9% success
20:25 rindolf moritz: I'm on Mandriva cooker.
20:25 nwc10 joined perl6
20:25 TimToady ?eval my $foo = 123; say (:$foo.yaml)
20:26 * moritz is on Debian etch
20:26 evalbot_r15233 is now known as evalbot_r15237
20:26 evalbot_r15237 OUTPUT[--- !hs/Pair ␤- ␤  - !hs/VStr ␤    - foo␤  - !hs/VRef ␤    - !hs/IScalar ␤      - !hs/VInt ␤        - 123␤␤] Bool::True
20:27 TimToady ?eval my $foo = 123; say (:$foo.perl)
20:27 evalbot_r15237 OUTPUT[("foo" => \123)␤] Bool::True
20:27 moritz should sqrt() work on negative and complex numbers?
20:28 TimToady ?eval my $foo = 123; say (:$foo)
20:28 evalbot_r15237 OUTPUT[foo123␤] Bool::True
20:28 moritz and if so, which of the complex roots should come out? 0 <= arg(sqrt($x)) <= pi?
20:29 TimToady ?eval sqrt(-1)
20:29 evalbot_r15237 NaN
20:29 moritz TimToady: that's why I'm asking ;)
20:29 TimToady ?eval sqrt(-1+0i)
20:29 evalbot_r15237 0.0 + 1.0i
20:30 moritz ;)
20:30 TimToady sqrt keeps its type the same, I think
20:30 moritz ?eval my $i is Int = 3; sqrt($i)
20:30 nwc10 ?eval exp(2i * atan2(1,0))
20:31 evalbot_r15237 1.7320508075688772
20:31 evalbot_r15237 Error: No compatible subroutine found: "&atan2"
20:31 nwc10 oh. where are all the maths functions? :-(
20:31 Squern joined perl6
20:31 moritz nwc10: atan2 wants two arguemnts
20:31 moritz no, i'm wrong, sorry
20:31 nwc10 I thought that I had given it two
20:32 nwc10 ?eval exp(2i * atan2(1.0,0.0))
20:32 evalbot_r15237 Error: No compatible subroutine found: "&atan2"
20:34 TimToady sqrt(Int) autopromotes to sqrt(Num) by the usual Int->Num rules
20:34 TimToady we haven't got autopromotion to Complex
20:35 TimToady and I believe luqui argued against it for sqrt()
20:35 moritz most users would be surprised if sqrt(-1) worked, yea
20:42 moritz ?eval sqrt(-1i)
20:42 evalbot_r15237 0.7071067811865476 + -0.7071067811865475i
20:44 TimToady ?eval my $x = -1; sqrt($x.i)
20:44 evalbot_r15237 Error: No compatible subroutine found: "&i"
20:44 TimToady hah
20:44 moritz TimToady: should that work?
20:44 TimToady yes
20:44 TimToady ?eval my $x = -1; sqrt(($x)i)
20:44 TimToady that shouldn't
20:44 evalbot_r15237 0.7071067811865476 + -0.7071067811865475i
20:45 TimToady er, should
20:45 moritz I'm writing a few testcases for sqrt, so I can include that one
20:45 TimToady ?eval my $x = 1; $x.++; say $x
20:45 evalbot_r15237 OUTPUT[2␤] Bool::True
20:46 TimToady ?eval my $x = 1; $x.i; say $x
20:46 evalbot_r15237 Error: No compatible subroutine found: "&i"
20:46 TimToady strange
20:47 TimToady ?eval my $x = 1; $x.'i'; say $x
20:47 evalbot_r15237 Error: ␤Unexpected "'"␤expecting ".", "\187", ">>", "=", operator name, qualified identifier, variable name, "...", "--", "++", "i", array subscript, hash subscript or code subscript
20:48 TimToady ?eval my $x = 1; $x>>i;
20:48 evalbot_r15237 Error: Hyper OP only works on lists
20:50 offby1 left perl6
20:50 moritz TimToady: should sqrt(-1i) return (1-1i)/sqrt(2) or (-1+i)/sqrt(2)?
20:51 moritz TimToady: both are valid solutions
20:52 moritz TimToady: or should it just return any valid solution?
20:53 TimToady I have no opinion on the subject, but maybe it returns both in list context, and then you can write any(sqrt(-1i))
20:54 moritz hm
20:55 TimToady or maybe that should be sqrts
20:55 TimToady just to avoid surprising people in accidental list context
20:56 moritz sounds like a good idea
20:56 TimToady generalizes to any root/roots kind of call too.
20:56 rindolf My test results are:
20:56 rindolf Failed 41/670 test scripts, 93.88% okay. 195/16168 subtests failed, 98.79% okay.
20:57 TimToady the plural form returns the list, and the user coerces to any, all, Set, whatever
20:57 moritz TimToady: mathematicians will love you/us for that one ;)
20:59 TimToady I've long since learned that my intuitions don't map well onto the intuitions of standard mathematicians... :/
20:59 svnbot6 r15238 | moritz++ | Added complex text cases to sqrt()
20:59 TimToady feel free to update Functions with the plural version
21:01 putter joined perl6
21:01 putter
21:01 lambdabot putter: You have 2 new messages. '/msg lambdabot @messages' to read them.
21:01 moritz roots(x) gives the two roots of x, and roots(x, n) all n-th roots x?
21:01 TimToady roots(n) always gives all of them
21:02 moritz right
21:02 TimToady er, roots(n,x) rather
21:03 TimToady I wonder if the 2.5th root of 2 returns 2.5 values...
21:03 moritz and n mandatory?
21:03 TimToady yes
21:03 TimToady maybe the other way around
21:04 TimToady $x.roots(3) is cube roots
21:04 coke joined perl6
21:04 coke .
21:04 TimToady 3.roots($x) is weird
21:04 TimToady so put x first
21:04 * coke is finding that irssi with multiple server connections doesn't deal well with network loss. :|
21:04 putter @tell audreyt re "is it okay if we port project_planning/ into dev.pugscode.org?", certainly.
21:04 lambdabot Consider it noted.
21:04 coke is now known as NewCoke
21:05 NewCoke any comment on my earlier "say" sends that I missed?
21:05 TimToady someone should change the channel headline to point to project planning
21:05 TimToady say is mentioned in IO
21:05 TimToady and even refers to a slightly more correct print entry now.  :)
21:06 putter audreyt: Could you outline your vision for the trac?
21:06 moritz should say accept a filehandle just like print?
21:06 TimToady yes
21:06 mrossetti joined perl6
21:06 NewCoke TimToady: be nice if it was at least given a forward reference earlier... or is that something more for merlyn to worry about?
21:07 TimToady it was forward referenced in one of the early apocalypses.  :)
21:07 NewCoke (IO) er... which synopsis is that?
21:07 NewCoke you need a horde of documentation daemons. =-)
21:08 mrossetti left perl6
21:08 TimToady or make that one of the late apocalypses...
21:09 TimToady dunno, it's called IO.pod in pugs
21:09 TimToady S16, looks like
21:09 moritz TimToady: so what should I write as  roots()'s signature? is there something like List<Num> ?
21:10 ludan joined perl6
21:10 NewCoke S16? http://dev.perl.org/perl6/doc/synopsis.html shows no such beast. =-)
21:10 lambdabot Title: Synopses - perl6
21:10 * NewCoke checks to see if he has write access to that website...
21:11 TimToady (in Num) method roots (Num $x: Int $n --> List of Num) {...}
21:11 TimToady or some such
21:12 TimToady oh, with an "is export" in there to get the roots($x,$n) form as well.
21:12 NewCoke TimToady: Hurm. seems like the website has the .pod in there. Any idea if those are links to another repository? to me, they look like snapshots...
21:12 TimToady otherwise people would always have to write roots($x: $n)
21:12 putter audreyt: Having a wiki again would be nice.  I wonder if trac is quite the right one.  Vs say wikimedia.  Trac has gotchas, like irreversible "delete page", can't rename pages, ah well.  Has some nice things too.  I'm just not sure how usable the wiki side is.
21:13 TimToady or how 'bout leave out sqrts and just have roots default $n to 2?
21:13 TimToady maybe not.
21:13 moritz TimToady: that's not "least surprise" ;)
21:14 TimToady yeah
21:14 NewCoke wiki: there's the wiki that parrot's using. I'm sure andy'd let you use that one.
21:15 TimToady someone should write a wiki in Perl 6 named "viki"
21:15 svnbot6 r15239 | moritz++ | Added spec for roots($x, $n)
21:16 NewCoke that's werry funny.
21:16 diotalevi Because it'd have less 'v'?
21:16 diotalevi vviki -> wiki - v.
21:16 TimToady and then write a sane vi syntax and call in 'vik'
21:16 diotalevi er.
21:16 TimToady then we could all be vikings
21:17 NewCoke ah. the update script in the doc directory you mentioned doesn't take care of IO. =-)
21:17 TimToady perl vi
21:17 nwc10 it's discrimination against emacs users, I tell you.
21:17 nwc10 there will be a perl emacs fork if we're not careful :-)
21:17 putter audreyt: re encoding the graph in tickets, the graph really ended up merely an illustrative sketch.  Even with more powerful graphing tools than graphviz, I couldn't show real dependencies and tasks.  Just don't have the tool set for the dynamic richness of anarchistic flooding development. ;)
21:17 TimToady perl evacs
21:17 TimToady or eviks
21:17 moritz nwc10: emacs users have lisp ;)
21:18 TimToady Perl 6 is just another dialect of Lisp
21:18 diotalevi evacs - e is a heating and ventilation system?
21:18 TimToady someone needs to write an elisp-to-plisp translator.  :)
21:18 moritz TimToady: that's worth a `fortune'-Line ;)
21:18 diotalevi plisp?
21:19 TimToady yeah, with a file extension of .pl
21:20 NewCoke so, anyone want to make sure all the SYN end up on the dev.perl.org site? Or should I open a ticket and force Robrt to do it? =-)
21:20 * diotalevi complains that elisp doesn't have closures or lexical variables (unless you also load the cl common lisp package).
21:20 Aankhen`` joined perl6
21:20 moritz NewCoke: where are they now?
21:20 moritz NewCoke: and where in the repository is dev.perl.org?
21:21 NewCoke I don't know where they all live now.
21:21 TimToady a more up-to-date location is http://perlcabal.org/syn/?
21:21 lambdabot Title: Official Perl 6 Documentation
21:21 NewCoke I was thinking dev.perl.org was canonical, which it ain't. =-)
21:21 TimToady maybe dev.perl.org should just point to something that easier to update
21:22 NewCoke TimToady: do you see an issue with redirecting dev.perl.org/ users there?
21:22 Khisanth I thought Perl was good for automating stuff :P
21:22 NewCoke ... ok. that I think I have rights to do.
21:23 TimToady only potential turf issues, which are always so easy to solve...  :)
21:25 TimToady bbl &
21:25 putter TimToady: What is the state of the STD regex section?
21:25 TimToady um, untested
21:26 putter extend of coverage?
21:26 putter (work in progress, thought to be fairly complete, etc)
21:26 TimToady mostly complete syntactically, I think, but depends on things that aren't implemented in pugs
21:26 putter np
21:26 TimToady like multi rules
21:26 putter thanks
21:26 dduncan joined perl6
21:26 putter oh, hmm, multi rules
21:27 putter ok, thanks
21:27 TimToady &
21:27 putter ohhh, those multi rules, np
21:27 putter bye
21:28 putter &
21:32 renormalist I haven't seen an entry on pugs.blogs.com from audrey for a long time. Is she ok? I could only loosely follow pugs so I'm not best informed about the progress and read pugs.blogs.com for first time in weeks.
21:35 Khisanth she was chatting about stuff this morning :)
21:35 NewCoke she's only got half her free time allocated to pugs atm, as I recall.
21:36 NewCoke ... but that was up from 0.
21:41 shay|p6 folks, is someone working on testing Perl6 on NetBSD and/or sparc64?
21:41 shay|p6 I got a Sun Ultra10 and I'd like to test and report bugs from it, but I want to make sure that I don't reinvent the wheel :)
21:42 [particle] check http://smoke.pugscode.org
21:43 shay|p6 I will
21:48 cjeris joined perl6
21:58 tene Hmm... the BSD::Resource stuff in evalbot doesn't seem to protect against "fork while fork'
22:00 TimToady that part is enforced by Dutch Goons.
22:01 Khisanth in wooden shoes?
22:02 TimToady the concrete shoes are for you.
22:05 shay|p6 can you guys give me pointers to stuff I should try other than running the Pugs tests/smoke? other testing resources? maybe BSD/sparc specific?
22:06 shay|p6 I'm sorry that I'm bugging with questions, but I'm not experienced with programming languages internals and I'm seeking for a start, I think that a rare architecure/system mix can be a good one
22:08 [particle] shay: you want to be a platform porter for pugs on your shiny new box, and don't know where to start? is that right?
22:09 shay|p6 [particle]: that might be a way for saying part of the idea, yes :)
22:09 [particle] you'll need to get ghc to compile before you can compile pugs
22:09 Psyche^ joined perl6
22:10 shay|p6 I guess that it's the Haskell compiler, right?
22:10 [particle] yes
22:10 shay|p6 I'll check documentation for building it
22:10 [particle] i'd love it if you could try building parrot, too
22:11 * shay|p6 writes down on his todo list
22:11 shay|p6 I hope that I'll be able to give you results before Sunday
22:12 [particle] there's a wider audience on the mailing lists than on irc.
22:12 shay|p6 I'm registred to..
22:12 shay|p6 sec
22:12 shay|p6 perl6-announce, perl6-language, perl6-users and perl6-compiler
22:13 shay|p6 any other important list that I should be registered on?
22:13 [particle] perl6-internals is the parrot list
22:13 shay|p6 subscribin
22:14 shay|p6 g
22:14 [particle] perl6-compiler has both parrot and pugs related info
22:14 [particle] shay|p6++ for enthusiasm and initiative
22:16 TimToady indeed
22:16 mr_ank_ joined perl6
22:16 TimToady have you got a pugs commit bit yet?
22:17 shay|p6 me?
22:18 TimToady well, I know [particle] has one...
22:18 shay|p6 no, I don't
22:18 shay|p6 didn't had nothing to commit yet
22:18 TimToady if you send an email address to me, I can send you a commit bit
22:18 TimToady portability fixes are certainly welcome
22:19 TimToady and you'll probably want to make your first commit when everyone here is off in a hotsprings or some such...
22:19 shay|p6 sent you an e-mail address in /msg
22:19 TimToady what do you want for your nickname?
22:19 shay|p6 shay would be ok :)
22:20 * [particle] remembers the pohoiki warm springs fondly
22:20 TimToady sent
22:20 TimToady it's customary to add your name to AUTHORS as your first commit to see if it works.
22:21 TimToady so you *do* have something to commit.  :)
22:21 shay|p6 awesome, thanks a lot :)
22:22 TimToady np
22:22 Psyche^ is now known as Patterner
22:38 reZo_ joined perl6
22:39 Aankh|Clone joined perl6
22:40 audreyt @tell putter clicking on "Admin" displays a "wikirename" page; I've blocked irreversible deletes
22:40 lambdabot Consider it noted.
22:45 nothingmuch @tell audreyt moose!
22:45 svnbot6 r15240 | moritz++ | Added test file for roots() (complex roots)
22:45 lambdabot Consider it noted.
22:46 * nothingmuch is in a silly oh-dear-god-i-really-should-be-packing mood
22:46 audreyt moose!
22:46 lambdabot audreyt: You have 11 new messages. '/msg lambdabot @messages' to read them.
22:49 cjeris left perl6
22:50 gaal nothingmuch: moose!
22:50 nothingmuch yo yo yo homie yo
22:51 gaal what hour's your flight?
22:51 nothingmuch 11:30
22:51 gaal am?
22:51 nothingmuch si
22:51 nothingmuch pm would be23
22:51 nothingmuch the situation is not so grim though
22:52 gaal thou needs must pack
22:52 nothingmuch i'm almost done finding the stuff i should be packing ;-)
22:52 nothingmuch that usually takes me the longesy
22:52 nothingmuch *longest
22:53 pdcawley joined perl6
22:53 gaal don't forget reservation #s, passport, and credit card. the rest is recommended. :)
22:54 gaal have fun and don't die! moose.
22:54 gaal z&
22:54 gaal nothingmuch: try { all that stuff }
22:54 * nothingmuch will try extra hard
22:54 nothingmuch i already have the most important thing
22:55 nothingmuch (1kg of turkish coffee)
22:55 gaal haha. ++
22:55 gaal re-z&
22:55 nothingmuch ciao ciao
22:55 nothingmuch btw
22:55 shay|p6 layla tov gaal
22:55 nothingmuch if you have friends looking to offload furniture please drop amir a line
22:55 nothingmuch furniture and/or appliances
22:56 gaal will do. shay|p6: layla tov
22:56 gaal re-re-z&
22:56 nothingmuch oh and one last thing, gaal
22:56 nothingmuch (/me wants to get a re-re-re-z, to see if it becomes re^3-z
22:56 shay|p6 yeah, why should we let him sleep
22:57 nothingmuch he's being serious about sleep =(
22:57 TimToady 'sides, it's already tomorrow for the important people like audreyt
22:57 * nothingmuch feels important, since it's 00:57
22:58 shay|p6 well, here is Feb 02 already if that counts
22:58 shay|p6 nm, are you from Israel too?
22:58 nothingmuch si si
22:58 nothingmuch currently residing near beer sheva
22:58 shay|p6 y hablas español también? ;)
22:58 nothingmuch but me and my BFFL just signed a contract on a shiny old flat in T.A.
22:58 nothingmuch no i don't parle vouz spanish at all ;-)
22:58 Limbic_Region joined perl6
22:59 shay|p6 mazal tov
22:59 nothingmuch i can say esteban
22:59 nothingmuch (my boss's name)
22:59 shay|p6 I'm from jerusalem
22:59 nothingmuch and I can type things like señora
22:59 nothingmuch jerusalem is awesome
22:59 shay|p6 jerusalem is teh placer
22:59 shay|p6 -r
22:59 * nothingmuch is sort of settling for T.A., but as long as it isn't beer sheva i'm happy ;-
22:59 nothingmuch )
22:59 nothingmuch friends are more important than cities though
22:59 shay|p6 yeah, I don't like BS too much neither
23:00 shay|p6 see, most of my service was in the south
23:00 shay|p6 so, no matter what, you must go over BS without any good memories behind it :)
23:02 nothingmuch anymoose, i have to sleep
23:02 nothingmuch err, pack
23:02 nothingmuch flight tomorrow morning =)
23:03 shay|p6 g'nite, where are you traveling to?
23:03 nothingmuch step 1 is to work at my boss's house
23:03 nothingmuch step 2 is to die (err, not die, sorry gaal) in the snow
23:04 gaal if yer so smart why don't you reverse the order huh? that way you don't have to work
23:05 gaal re-etc. &
23:05 shay|p6 I knew he was spying.
23:05 gaal no privacy these days.
23:05 shay|p6 you bet
23:08 gaal Oh, what sad times are these when passing ruffians can say "moose" at will to old ladies.
23:09 shay|p6 do you have insomnia or something?
23:09 gaal no, but this chair's so warm and cozy.
23:09 Aankh|Clone is now known as Aankhen``
23:10 gaal however! we shall overcome. zzZZZ(!) &
23:10 shay|p6 lets start the bets on how many time will it take for gaal to type again
23:15 lumi 10 credits on 6 hours
23:16 shay|p6 no way, he's still spying
23:16 Limbic_Region @tell audreyt see http://use.perl.org/~jjore/journal/32362 regarding STM and some *apparent* negative press
23:16 lambdabot Consider it noted.
23:16 lumi 10 credits on 6 minutes, then
23:16 Limbic_Region oh, gaal - you might be interested in that too
23:16 shay|p6 5 credits on less than 30 minutes
23:21 Aankhen`` joined perl6
23:21 larsen_ joined perl6
23:27 svnbot6 r15241 | shay++ | Added myself to try the SVN :-)
23:28 audreyt Limbic_Region: heya :)
23:28 lambdabot audreyt: You have 1 new message. '/msg lambdabot @messages' to read it.
23:28 audreyt lambdabot: @moosages
23:28 lambdabot Limbic_Region said 11m 47s ago: see http://use.perl.org/~jjore/journal/32362 regarding STM and some *apparent* negative press
23:29 * allbery_b rolls his eyes
23:29 audreyt Limbic_Region: well, the point is valid, that share-everything has drawbacks
23:30 audreyt STM fixes the drawbacks to a degree but still if you can design the code to share nothing, the better
23:30 Limbic_Region audreyt - didn't read it, just saw that diotalevi was asking TimToady about it at PerlMonks and then saw his use.perl entry
23:30 Limbic_Region though I would pass it on
23:30 allbery_b he's right, share-nothing is ideal.  so's world peace.  both are about equally likely...
23:30 audreyt not really sensational :)
23:30 Limbic_Region out of curiosity, anyone have Solaris 9 or 10 handy?
23:31 allbery_b both
23:31 allbery_b (9/sparc 10/x86)
23:31 allbery_b ...but pugs on neither, nor likely
23:31 Limbic_Region allbery_b - can you try: use Fcntl ':flock'; open(my $fh, '<', $0) or die $!; flock($fh, LOCK_EX | LOCK_NB) or die $!; print 'made it';
23:31 Limbic_Region this is perl 5 question
23:32 * Limbic_Region isn't sure if it is a platform or a perl issue (I had the default perl which is 5.6.1 I believe)
23:32 Limbic_Region audreyt - are you up early or haven't been to sleep yet?
23:34 allbery_b sol9 prints "made it"
23:34 Limbic_Region interesting - I got bad file descriptor
23:35 allbery_b sol10 x86 "Bad fle descriptor"
23:35 allbery_b er, Bad file number
23:35 Limbic_Region right
23:35 Limbic_Region ok - so what is the perl -v of each?
23:35 allbery_b 5.8.3 on sol9, 5.8..4 on sol10
23:36 stephang joined perl6
23:38 Limbic_Region ok, so perhaps it is a platform thing cause I think I had 10 when it failed
23:38 allbery_b at a guess, flock may be emulated with fcntl on sol10 --- although the error seems a little odd in that case, but maybe it's correct too
23:38 allbery_b (fcntl you can
23:39 allbery_b t lock exclusive if you open for read)
23:39 diotalevi allbery_b: I don't have a dog in this horse race. I'm pondering how to use STM badly and how to use it well and what new kinds of problems having it around means. I'm pretty ignorant on the subject right now.
23:39 audreyt diotalevi: replied
23:39 Limbic_Region allbery_b - it is a MJD trick to make sure you only have one copy of a program running at a time (exclusive lock on the program itself)
23:40 Limbic_Region allbery_b - it works fine if you flock something other than $0
23:40 Limbic_Region allbery - in any case, I worked around it
23:40 allbery_b ah
23:40 Limbic_Region 3333......
23:40 allbery_b hm
23:40 * allbery_b wonders what $0 is
23:40 Limbic_Region my daughter says hello
23:40 allbery_b /dev/fds/mumble?
23:41 diotalevi saw that, audreyt. I'm looking to get a copy of CTM now since that has PR saying it covers lots of this stuff.
23:41 audreyt diotalevi: "just because your RDBMS has row-level locking, it doesn't mean you want to put all data into one single table"
23:41 audreyt that's essentialy patricklogan's argument
23:41 audreyt which is valid.
23:41 allbery_b nope
23:41 allbery_b interesting
23:41 audreyt but rowlevellocking is useful.
23:41 audreyt bbiab
23:42 audreyt ("row-level locking encapsulated in SQL transactions" is analagous to STM.)
23:43 stephang is now known as sgt
23:43 sgt is now known as stephang
23:44 diotalevi So, fair enough. It seems fairly uncontroversial as far as databases as big external global variables go. I'd recently learned some Mozart/Oz to get at their constraint solving libraries and found it had builtin data flow concurrency which I'm told is similar to whatver it is that Erlang does.
23:48 TimToady stephang: if you use #cbstream you can keep an eye on the chatterbox at the same time.  :)
23:49 stephang ok thanks
23:49 polettix joined perl6
23:50 ashelyb_ joined perl6
23:51 Limbic_Region ++
23:51 Limbic_Region +
23:51 diotalevi Also, Mozart/Oz was *really* only of interest because it has constraints as a feature that's very close to the standard syntax so the intention is that it's trivial to say "ok, now this next bit I want the constraint solver to deal with" instead of having to actually go code all the foreach/prune code that'd be normal in perl or whatever.
23:51 Limbic_Region Jasmine wanted to karma someone apparently
23:51 Limbic_Region ;-)

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

Perl 6 | Reference Documentation | Rakudo