pugscode.org | sial.org/pbot/perl6 | ?eval [~] <m oo se> | We do Haskell, too | > reverse (show (scanl (*) 1 [1..] !! 4))
Set by agentzh on 17 April 2007.
svnbot6 r16021 | Darren_Duncan++ | ext/QDRDBMS/ : split class QDRDBMS::AST::LitDefExpr into type-specific classes 00:07
chlli ? 02:02
svnbot6 r16022 | diakopter++ | new version of Perldoc from Damian. Someone with root on feather: please update it, if you please. 03:14
veritos Hello, I was thinking about reimplementing Perl6 in Python...sort of like Redsix, but...not. 04:07
Tene Sounds fun! 04:09
You have a commit bit yet?
veritos Tene: nah :( 04:10
it might help if i knew sv[nk] first
Tene m'eh, it's easy.
svn add foo
hack hack hack
svn commit
all done
veritos hmm...sounds like Fossil (a rev ctrl system i made a couple weeks ago) 04:11
offby1 you're writing your own rcs and you don't know about subversion? 04:13
revdiablo offby1: Market research is for losers
veritos offby1: i knew _about_ it, but not how to use it. i use Mercurial 04:14
revdiablo Ah, the basic mercurial commands are pretty much the same as svn's
Tene veritos: well, just give almost anyone here an email address and preferred username and they'll send you a commit bit.
geoffb bdfl? 04:16
veritos geoffb: benevolent dictator for life. bdfl(perl) = larry. bdfl(linux) = linus. etc.
in this case, bdfl(pugs) = audrey.
geoffb veritos: sorry, too tired for my ALUT to return results 04:17
veritos alut? now i get to ask ;)
geoffb Acronym Look Up Table. ;-)
veritos oh 04:18
geoffb and you don't need the bdfl to get your commitbit.
Most of us are project admins
veritos ok...nm.
geoffb Invite on its way ... 04:20
veritos k thanks 04:21
got it
geoffb It's traditional to add yourself to AUTHORS to test your setup
And it gets the nerve-wracking first commit out of the way. ;-)
veritos geoffb: let's download the tree in SVN first.
i'm in. 04:24
dduncan I have an opinion question ... if we're using the proper 3-part versions, is it better to release distros named Foo-1.2.3 or Foo-v1.2.3 ... which format is better ... both kinds seem to be in use. 04:25
geoffb <burns>excellent</burns>
I prefer the former, as a human
dduncan I also see that this is what Pugs itself uses 04:26
as does Perl 5.x itself
though more modules seem to have v 04:27
I currently intend to not use the v, unless doing so is "the best practice" of the future
veritos btw is that perl6-in-perl6-rules grammar complete?
geoffb It's still being tweaked, as there is not yet anything that can parse the whole thing correctly 04:28
(last I heard)
veritos as correctly as pugs at least? 04:29
geoffb Well, fglock has been programming up a storm the last week, but I don't know how far he has gotten with the STD grammar. 04:31
svnbot6 r16023 | karhu++ | The usual add-name-to-AUTHORS edit
r16024 | diakopter++ | inserting random typos... hopefully no one will notice. 04:40
zxagent hello everyone 05:00
agentzh Pugs no longer builds on feather? 05:24
pasteling "agentzh" at 221.131.74.148 pasted "Pugs build failure on feather" (63 lines, 3.6K) at sial.org/pbot/24338 05:25
agentzh i think i need to install the mtl lib for GHC.
irclogger-- 05:26
rindolf > reverse (show (scanl (*) 1 [1..] !! 4))
lambdabot "42"
rindolf Heh. 05:27
agentzh it has been fixed? 05:28
same error. 05:34
finally fixed 05:51
(root access)++
rindolf Any news regarding ingy? 05:57
zxagent agentzh: hello master 06:02
agentzh zxagent: hello, boy :) 06:14
zxagent: nice to see you here.
zxagent: want a commit bit? 06:15
oops, too late
TimToady maybe you scared him off 06:16
cj TimToady: what are you doing up at this hour!? 06:25
jrockway 11:30 pm?
cj rindolf: news regarding ingy? did something happen to him?
Tene cj: yeah, heard he got hurt pretty bad 06:26
jrockway /msg purl ingy hurt (on MagNET)
svnbot6 r16025 | Darren_Duncan++ | ext/QDRDBMS/ : added roles|classes QDRDBMS::AST::(Value|(List|Set|Seq|Bag)Sel) 06:28
agentzh TimToady: *nod* 06:31
cj Tene: this makes me sad
rindolf jrockway: purl hates me on MagNET. 06:32
jrockway: so that's not a solution.
jrockway: someone configured it to ignore me.
jrockway hahahaha
anyway, repeating what purl says
ingy needs wrist surgery 06:33
i am a total tertiary source though, so i have no idea
i saw him in here yesterday though, so at least he's alive
rindolf jrockway: that's good. 06:34
jrockway here's the exact purl text for completeness: 06:35
ingy hurt is 20070417 in motorcycle accident in .tw. Wrist
broken, pending operation. Gradually recovering, expected to
stabilize in a few days.
cj sounds like ingy will be okay 06:37
oh, you already did that
agentzh ingy was answering my questions here a few hours before the accident. :( 06:38
just can't believe it 06:39
yay! welcome back, ilogger2! 06:53
i miss you so ;)
svnbot6 r16026 | agentz++ | * run-smoke.pl now honours the HARNESS_PUGS environment which defaults to './pugs' 07:13
r16026 | agentz++ | * added util/limited_pugs for restricting './pugs' via the ulimit command in sh.
r16027 | agentz++ | limited_pugs now limited to a cputime of 120 sec
r16028 | agentz++ | * made prove6 *nix friendly.
agentzh my svk seems to have problems. 07:15
svnbot6 r16029 | agentz++ | * run-smoke.pl - some win32 love.
r16029 | agentz++ | * limited_pugs - added the 'nice -n5' and 'ulimit -v 800000' contraints.
agentzh these commits are very old.
i see a lot of conflicts. 07:19
and it seems i'm repeating existing commits made by others and myself over the last few days :(
maybe svk v2.0.1 is just unstable. 07:23
clkao huh? what are you trying to do ? 07:24
agentzh clkao: i'm trying 'svk push'
(as normal)
but a lot of conflicts between //mirror/pugs and //local/pugs occurred. 07:25
it seems svk is repeating old commits in the last few days.
very weird.
clkao is your uuid same as somebody else's? 07:26
agentzh no idea.
how to check that?
clkao cat db/uuid 07:28
is this depot a fresh mirror or you copied from somewhere else?
agentzh it's a fresh mirror. 07:29
i'm on my own machine, not feather.
6e2bbe1e-9bb6-457c-9519-9cf87a6f6883
sorry, but how can i revert svn commits using svk? (i know how on win32 with TortoiseSVN, but not linux) 07:31
clkao svk merge -c -N //path //path 07:33
agentzh clkao: ah, thanks. 07:34
clkao: is it possible to specify an SVN rev number? 07:40
clkao N@ 07:45
agentzh clkao: cool, thanks
$ svk merge -C -c -16029@ //local/pugs2 //mirror/pugs 07:46
Change spec -16029@ not recognized.
oh, i should exchange the paths. 07:47
clkao: it works now. thank you :) 07:48
i've started another local branch for the pugs mirror.
TimToady is still awake? wow 07:51
pasteling "evalbot_r16029" at 194.145.200.126 pasted "Pugs build failure" (314 lines, 17.1K) at sial.org/pbot/24340 07:59
svnbot6 r16030 | agentz++ | added shebang to update-syn
r16031 | agentz++ | updated util/t/01-smartlinks.t since gaal++ changed the template used by the HTML emitter. 08:50
r16032 | agentz++ | util/t/01-smartlinks.t - fixed a bug 08:53
dduncan ?eval my $s = 'hello'; $s =~ m/oops/; 09:09
evalbot_r16032 \\""
dduncan that should be failing at compile time, according to the specs, as =~ is a braino 09:10
but it accepts it instead
moritz write a test case ;)
dduncan what's with all this coming and going? 09:11
that is ... 09:12
I don't setup my client to not display them, but I personally ignore them unless its someone I want to talk to
svnbot6 r16033 | Darren_Duncan++ | ext/QDRDBMS/ : in AST.pm, implemented both ::EntityName possreps, renamed ::Value to ::Node 09:21
OwlEye looks like TimToady's connection is having fun :-( 09:25
moritz yay, smoke works again for me, agentzh++ 09:31
agentzh moritz: glad to hear that!
it's ingy++ who told me the solution :) 09:32
too bad he is now in the hospital :(
Aankhen`` patricklogan.blogspot.com/2007/02/m...elled.html # diatribe against STM... any comments?
lambdabot Title: Making it stick.: Misguided: The Road Not To Be Travelled, tinyurl.com/2n7gbd
svnbot6 r16034 | Darren_Duncan++ | added test t/operators/brainos.t which checks that a =~ invocation dies at compile time 09:42
dduncan I haven't read it yet, but the principle of STM is good, so an argument against that principle would be misguided ... will read now ...
Aankhen`` Thanks. I don't really understand the concepts and pitfalls involved well enough to figure out whether the arguments presented are valid, so I thought this would be a good place to find people who would. =) 09:45
s/who would/who could/
dduncan STM is just a mechanism where a coder says that a collection of actions must be atomic, and being able to do that is a good thing 09:48
Aankhen`` Right.
dduncan I don't see how that can be bad
the gist I got from the paper is that the writer says that STM can be misuded, and therefore its better not to include the feature, 09:49
moritz but perl never worked that way 09:50
dduncan which sounds like the antithesis of Perl
moritz perl always offers any possibility ;)
dduncan Perl gives you lots of tools and you choose what to use
Aankhen`` Heh.
moritz even goto *g*
Aankhen`` Indeed.
moritz: Hey don't be hatin' on the goto.
goto &foo; # très useful :-)
dduncan hey, goto is very powerful ... that's how we get coroutines and exceptions to work, the former particularly
moritz Aankhen``: I'm merely remarking upon the fact that most "modern" langauges don't have goto anymore 09:51
Aankhen`` moritz: Ah, right.
dduncan well, I mean the principle of goto is useful
that sometimes you don't have to complete your current block before executing another
Aankhen`` moritz: Sorry, it's 2:50 AM and I've spent the better part of the last few hours in a groggy state reading mostly abstract papers or articles on programming. :-P
dduncan its that way here too 09:52
2:52am here
moritz Aankhen``: no problem ;)
Aankhen`` ^_^
moritz 11:50 am here ;)
no, 10:50, sorry ;I
my irc client lies on a server in a different time zone
which is sometimes confusing ;)
Aankhen`` dduncan: Hmm, let me see. What I take away from the post, after repeated reading, is that we should be moving towards having no shared state. 09:53
Whereas he implies STM is about having everything shared.
dduncan I got that from it too
I see STM as dealing about things possibly being shared, and so be safe just in case ... its like being thread safe
Aankhen`` Ah, good, I'm not quite insane then. At least on that count.
Atomic operations cannot include I/O activity, right? 09:54
dduncan broadly speaking, yes they can
Aankhen`` (In Perl 6, at any rate.)
dduncan the caveat is that I/O operations can't be undone
Aankhen`` Yeah.
dduncan I mean, some can't be undone 09:55
if the I/O is with the filesystem, it should be undoable
that's what DBMSs do already
but if its with users or a network, probably not
Aankhen`` atomic { ...; $dbh->do('DELETE FROM `users`'); fail "Oops, didn't mean to do that"; } 09:56
Might be a tad hard to get out of that one.
dduncan that's why proper DBMSs have rollback
Aankhen`` Ergh, shame on me for mixing P5 and P6 there.
Aye. 09:57
But I guess from the language's point of view, it has no idea of knowing, so it has to err on the side of caution.
dduncan fyi, in my DBMS, you usually specify an atomic operation with, eg, "atomic { ... }" rather than begin/commit/rollback
Aankhen`` So throwing an exception within that means the block fails, falling off normally means it's committed? 09:58
dduncan though that's an option ... all statements in my DBMS are implicitly atomic anyway
yes
Aankhen`` Makes sense.
dduncan statements and routines both
the only time one would actually say atomic { ... } is if they want just a block within a routine to be separately atomic, and then that block may be called try { ...} rather 09:59
Aankhen`` Indeed.
dduncan since the STM is essentially tied to standard control flow or exceptions
but that's in my system, obviously Perl would have a separate atomic {}, since everything is not implicitly transactional 10:00
Aankhen`` Ergh, do you know which synopsis `contend` is in?
dduncan fyi, what I've written on the matter so far is in ext/QDRDBMS/lib/QDRDBMS/Language.pod
Aankhen`` Doesn't seem to be in S04 or S06.
Oh, excellent. I'll take a look. 10:01
Hmm, I should probably svn up Pugs too.
I've been pretty lax about that.
dduncan to be clear, my last comment was about transaction support in my DBMS, not 'contend'
Aankhen`` Heh, figured as much.
Er, did I just imagine `contend`? I can't find it in any synopsis. 10:02
dduncan as for 'contend', I would have expected it to be in S04
or either of those actually
looking ...
is 'async' anywhere?
I mean, 'atomic'
though both of those ought to appear together
Aankhen`` I see a mention of both contend and async in S04.
Under `Other do-like forms`. 10:03
dduncan Aankhen, actually, like DBMS transactions, I see one of the main benefits of STM not necessarily having anything to do with shared state ... its that it provides an easy mechanism to undo some partitially completed actions ... and this can be true in isolation 10:15
so then, STM is orthogonal to shared states, which the paper talked about 10:16
Aankhen`` Ah.
dduncan in my understanding, STM is just like a database transaction, but it just deals with a program's memory state 10:17
Aankhen`` That's what I thought too.
I think I must reluctantly go to bed now.
dduncan right then
Aankhen`` dduncan: Always nice to talk to you. :-)
G'night.
dduncan sure 10:18
moritz agentzh: good night
sorry, that should go to Aankhen`` ;)
agentzh it's too early for me to say "g'night" ;) 10:43
moritz same for me ;) 10:44
diakopter agentzh: can you run the smartlinks.pl again as soon as this next commit appears 11:05
svnbot6 r16035 | diakopter++ | committing Documentation.pod (S26) changes from upstream::Damian.
agentzh diakopter: i think feather's cron will do that automatically in the next hour. ;) 11:07
but anyway, i'll do that manually *now* :)
svnbot6 r16036 | Darren_Duncan++ | ext/QDRDBMS/ : minor tweaks 11:08
agentzh diakopter: done :) 11:09
diakopter: btw, you can always do that yourself via the command "~agentzh/update-syn" if you can't wait cron. :) 11:10
diakopter: do you have permission problems? 11:11
diakopter nope
agentzh cool 11:12
diakopter still waking up, here. wet memory problems.
agentzh hehe
diakopter agentzh: I don't know if anyone can currently create smartlinks against S26 - do you know? 11:13
agentzh diakopter: no, no one can at this moment. :( 11:14
because...because...
diakopter (Damian has included many inline examples of Perldoc/Pod that would be excellent initial tests)
agentzh smartlinks.pl can't parse Perl 6 POD yet.
are those tests in Perl 6?
diakopter oh. I thought it parsed the html. 11:15
agentzh diakopter: it parses the POD directly.
parsing HTML will be harder.
pasteling "evalbot_r16035" at 194.145.200.126 pasted "Pugs build failure" (9 lines, 429B) at sial.org/pbot/24344
agentzh (although not impossible)
diakopter oh. what makes it harder?
agentzh if parsing HTML, then we'll rely our code too heavily on Pod::Simple::HTML's emitter template. 11:16
(as well as Perl6::Perldoc's template) 11:17
they may change over releases.
HTML source is just too scattered and delicate. 11:18
moritz and you can loose semantics in the translation 11:19
agentzh moritz: indeed.
well, i'll try to find some time in the next few weeks to add support for Perl 6 POD to smartlinks.pl. 11:24
Damian's Perl6::Perldoc::Parser is ready anyway. that's the #1 prerequisite. 11:25
diakopter: feel free to add "broken" links for S26 to the test suite, which will be very helpful. :) 11:28
diakopter agentzh: i agree that would be helpful.
agentzh TDD++
diakopter: thank you for patching smartlinks.pl to emit HTML for S26 although there's some unnecessary code duplication. :) 11:31
diakopter agentzh: yw. hence my comment "agentzh will fix this sometime" :) 11:37
agentzh diakopter++ 11:40
but right now i'm just too lazy to fix it ;) 11:41
diakopter agentzh: yep. it was a half-tongue-in-cheek comment. 11:42
zxagent agentzh:I'm coming 12:12
moritz ?eval 1 13:41
evalbot_r16036 1
moritz not really what I want 17:10
17:11 veritos joined
svnbot6 r16043 | agentz++ | 01-smartlinks.t - sorted the .t files 17:12
r16044 | agentz++ | more tweaks in util/t/ 17:18
agentzh night, all & 14:01
14:09 TimToady joined 14:14 TimToady joined 14:15 vel joined 14:19 kanru joined, TimToady joined 14:34 TimToady joined 14:39 TimToady joined 14:44 TimToady joined 14:49 TimToady joined
audreyt @tell agentzh YAML::Syck back to service again with 0.85. Thanks! 14:50
lambdabot Consider it noted.
cmarcelo audreyt: hi 14:54
14:54 TimToady joined 14:57 agentzh joined
agentzh audreyt++ # JIT YAML::Syck bug fix 14:57
lambdabot agentzh: You have 1 new message. '/msg lambdabot @messages' to read it.
14:58 HiTech69 joined 14:59 TimToady joined 15:02 ilbot-test joined 15:04 TimToady joined 15:08 offby2 joined 15:09 TimToady joined 15:14 TimToady joined, ilbot-test joined 15:17 toshinori joined 15:19 TimToady joined 15:25 offby1` joined 15:32 TimToady joined
TimToady "this time fer shure!" 15:32
[particle] /kick TimToady ;) 15:33
moritz [particle]: he'll leave after a while anyway...
[particle] HA!
agentzh he has been kicked :) 15:34
moritz do we even have ops?
15:34 offby1 left
[particle] the folks with op generally only turn them on when needed 15:34
moritz ok 15:35
15:35 rlb3_work joined 15:36 offby1 joined 15:38 marmic joined 15:39 TimToady joined
TimToady I guess that trick never works... 15:41
offby1 Bullwinkle? 15:42
15:45 TimToady joined
[particle] TimToady: you should check your back for a "kick me" sign 15:45
15:49 kanru joined 15:50 TimToady joined, mj41 joined
agentzh hehe, TimToady is still struggling ;) 15:53
15:55 kanru joined 15:56 tzoa joined
svnbot6 r16037 | agentz++ | another attempt to make t/01-smartlinks.t more portable (hopefully it will now pass on feather) 16:03
moritz theorbtwo: that's what we want to avoid 17:04
as a workaround a deleted that one line
svnbot6 r16038 | audreyt++ | * yaml_harness.pl - YAML::Syck 0.85 is fine now, so prefer it anyway. 16:05
r16039 | agentz++ | added the util/t/gen-smartlinks-data script to update the regression data 16:14
16:18 justatheory joined 16:21 veritos joined
svnbot6 r16040 | agentz++ | gen-smartlinks-data now works for real 16:21
r16041 | agentz++ | smartlinks.pl - fixed the HTML syntax error in the emitter template
16:29 mr_ank joined 16:30 ilbot-test joined
svnbot6 r16042 | agentz++ | 01-smartlinks.t - another trial 16:33
veritos moritz: that's bad for you. 16:34
moritz veritos: bad for my CPU ;)
16:34 tzoa left
veritos moritz: stop then, it takes long enough to compile pugs anyway. 16:35
agentzh veritos: what if moritz had a strong machine? 16:36
moritz agentzh: I haven't - just a 3 years old laptop :(
veritos agentzh: silicon is known by the state of california to cause cancer. now go back to committing stuff.
agentzh oh...maybe you want to try util/limited_pugs then 16:37
veritos moritz: mine's a sempron. you do with what you have when you're a poor programmer ;)
moritz veritos: right. And when your maschine is not that strong you worry more about efficiency ;) 16:38
veritos moritz: `make fast` eq :)
moritz veritos: no, that's `make soon' now
agentzh soon is less confusing :) 16:39
in fact, the pugs produced by 'make fast' is slower.
moritz soon enough somebody will confuse it ;)
veritos agentzh: we know, but we can get it faster.
16:40 rindolf joined
rindolf Hi all. 16:40
moritz hi rindolf ;)
veritos morning
rindolf Hi moritz, veritos
What's up?
moritz has just tought his new irclogger how to log actions ;)
veritos installing FreeBSD. moritz: i read that as irc-clogger for a second. 16:41
moritz what's "clogging"?
I don't know that word
agentzh moritz: mind you, upgrade your YAML::Syck to 0.85 before you smoke :) 16:42
veritos moritz: what we're doing to this channel. filling it with so much stuff that it becomes hard to work.
moritz veritos: *lol*
moritz kills the smoke 16:43
rindolf Any news regarding ingy?
veritos rindolf: is he somehow unable to hack, perhaps due to illness?
rindolf veritos: not illness. 16:44
moritz veritos: in hospital, serious car accident
rindolf veritos: he's been in a motorbike accident.
veritos :(
veritos prays
16:47 edenc joined
rindolf veritos: no, I think he's more-or-less OK. 16:48
veritos: I don't think he's in danger of dying or something.
veritos: he's been here after the accident.
veritos rindolf: well we need him to help hack!
rindolf veritos: heh. 16:52
16:55 thorat joined, TimToady joined
TimToady suspects his Linksys was dropping lengthy TCP connections... 16:55
does IRC use TCP?
clkao yes 16:56
rindolf TimToady: yes, it does. 16:57
TimToady resetting it seems to have helped, anyway. we'll see... 16:58
(other evidence: yum couldn't actually download an entire rpm from any one repo)
moritz: is it possible to set up the logger table such that it doesn't exceed the width of the page and wraps the text instead? It's kindof irritating to have to scroll right to see the ends of many comments. 16:59
moritz TimToady: I'll do my best ;) 17:00
is there a simple html/css solution for auto-wrapping text in tables?
TimToady dunno, not an expert in css. I think bare html has percentage widths on table columns, but whether that can be percent of page width I don't know. 17:01
moritz ah, it normally does wrap, but in this case pastebot spammed the table with one long word...
and a single word is normally not wrapped automatically 17:02
TimToady well, banning pastbot isn't the answer. :)
theorbtwo overflow: scroll ? 17:03
TimToady hey, I passed my 5-minute-versary, maybe I'll stay up now.
could look for long words and break 'em, I suppose 17:04
moritz first I'll look if there is a HTMLish solution 17:05
TimToady that was kind of a weird paste tho 17:06
moritz spam
/dev/urandom | perl -e $randomScript spam ;)
franck__ moritz: may be www.thescripts.com/forum/thread96534.html ? 17:07
lambdabot Title: Prevent text wrap in a table cell - HTML
moritz franck__: thanks, I'll try 17:08
agentzh yay, all tests in 01-smartlinks.t passed on feather! 17:19
hopefully they also pass elsewhere. 17:20
TimToady moritz: are you now weeding out svnbot6? I think people would like to see those in the log, in general. 17:21
17:21 Psyche^ joined
TimToady or did your bot just happen to be down for a bit? 17:22
pasteling "evalbot_r16043" at 194.145.200.126 pasted "Pugs build failure" (332 lines, 19K) at sial.org/pbot/24362 17:23
moritz TimToady: let me check... (I don't filter anything excpet lines starting with [off]) 17:26
agentzh moritz: where's your log? url?
moritz agentzh: moritz.faui2k3.org/irclog/ 17:27
TimToady just go to irc.pugscode.org
lambdabot Title: IRC logs
agentzh moritz: thanks
oh, quite cool 17:29
moritz but why didn't it record the last two lines form svnbot6? 17:30
weird
17:30 thepler joined
TimToady not search for /[off]/ or some such, are you? 17:31
hmm, wouldn't explain the 2nd line...
moritz no, m/^\\[off\\]/i
TimToady it was logging svnbot earlier
moritz yes
17:33 Patterner is now known as Psyche^
TimToady I just blew away audreyt's dist/build; I think that's the solution to the pasted "can't link" problem, if I recall... 17:35
and if I don't recall, I'm sure that will also become evident... 17:36
b_jonas does anyone want to try my irc bot evaluating j phrases (j programming language => www.jsoftware.com/ )? 17:37
lambdabot Title: J Home
b_jonas it's called jeval, and evaluates any expression if you prefix it with a right square bracket
cmarcelo moritz: some ideas for your irc log => would be nice to make URLs "clickable". and if you wan't even more sugar, make things like "r(\\d*)" (revision numbers) link to dev.pugscode.org/changeset/\\1 17:38
svnbot6 r16045 | karhu++ | Added another food item. Real work will come later today.
17:39 fglock_ joined 17:40 fglock__ joined
agentzh enough commits for a day & 17:40
17:40 agentzh left
moritz cmarcelo: I planned that ;) 17:40
TimToady brainos.t++ 17:41
moritz cmarcelo: added to my (mental) todo list 17:42
cmarcelo moritz: thank you. :) a question: are we "dropping" colabti.de because of your solution or your solution came because colabti.de "dropped"? 17:44
moritz cmarcelo: second one 17:46
cmarcelo: we had no logger, so I wrote one ;)
17:46 edenc joined 17:57 ilbot-test joined
TimToady colabti also defaults to latin-1 rather than utf-8 17:57
and of course we pay no attention to the fact that it's written in Python. </wink> 17:58
Juerd I think an IRC logger should default to UTF-8, and then interpret invalid UTF-8 byte sequences as Windows-1252. 18:02
Irssi does this, and it really works out well in the real world.
moritz Juerd: mine currently uses Encode::Guess with utf-8 and latin1, and falls back to utf-8 18:03
Juerd How can you ever fall back to utf-8?
moritz Juerd: so far it semms to work 18:04
well, guess() returns either a decoding or undef
and in case of undef I use utf-8
Juerd Oh, you're using heuristics? :(
moritz Juerd: why not? it's rather easy to distinguish latin1 and utf-8 18:05
geoffb Don't knock heuristics. Heuristics made me the man I am today.
;-)
Juerd moritz: On a line by line basis, yes. 18:06
But you'll be considering entire lines to be latin1 if they contain a single utf8 error.
I guess that's not a big problem.
moritz Juerd: I decided not to think more about it until I see it failing 18:07
OwlEye TimToady, was somebody messing with your connection earlier?
Juerd I don't think it'll fail.
TimToady nah, flakey Linksys firewall, apparently 18:08
probably ran out of resources to keep track of TCP properly
so was dropping every TCP connection within a few minutes
OwlEye oh. i already feared there is bad people around
TimToady, dont you like the Freenode cloaks? 18:09
TimToady I'm paranoid, but I try to be realisticly paranoid. :D
rindolf Hi OwlEye
perl-begin.org/ # AndyArmstrong++
lambdabot Title: Perl Beginners' Site
OwlEye hey rindolf :)
rindolf OwlEye: hi.
Hi TimToady 18:10
OwlEye TimToady :D
TimToady *>>.hi
rindolf TimToady: what's up?
TimToady: still working on that p6 grammar?
TimToady random tweaks
someone needs to hack the new pod stuff in now.
moritz links should be clickable now 18:15
18:17 polettix joined
OwlEye moritz++ 18:17
moritz r16042 <-- just a test 18:19
18:27 geoffb joined
moritz \\br\\d+\\b now links to the corresponding changeset ;-) 18:29
jjore-w !seen adamk 18:30
moritz perl is so cool, fast successes every few minutes ;)
devbot6 jjore-w: Error: "seen" is not a valid command.
jjore-w boo.
moritz @seen adamk
lambdabot I haven't seen adamk.
moritz sigils do matter ;)
jjore-w ETOOMANYSYNTAXES.
gaal moose
cmarcelo moritz++
gaal back from a hike. why don't I do that more often? 18:31
18:31 Casan joined
jjore-w @seen alias 18:31
lambdabot I haven't seen alias.
jjore-w @seen alias_
lambdabot I saw alias_ leaving #perl6 5h 19m 5s ago, and .
moritz any more feature requests for the irclogs? 18:32
gaal moritz: a moose icon? 18:34
cmarcelo is it Perl 6?
gaal or moose in other aspects of it
moritz cmarcelo: perl5
gaal: where can I find one?
gaal @google moose 18:35
etc.
lambdabot en.wikipedia.org/wiki/Moose
Title: Moose - Wikipedia, the free encyclopedia
moritz gaal: I thought maybe there is *the* moose that is used in perl context... 18:36
gaal not really, but I can recommend an excellent book on the subject
forum2.org/moose/
lambdabot Title: The Forum 2 Moose Page
gaal bbiab...& 18:37
moritz *rofl* moose!
fglock_ hmm - .WHAT stringifies to short name, but the prototype object doesn't - means .WHAT is some kind of proxy? 18:39
rindolf Hi gaal 18:40
gaal: what's up?
18:42 veritos joined
fglock_ ah, it's a macro 18:43
TimToady fglock_: no, means the prototype isn't stringifying right
WHAT should merely return the prototype 18:44
fglock_ but aren't all objects prototypes (potentially)
offby1 The trouble with a kitten is THAT / Eventually it becomes a CAT -- Ogden Nash
TimToady there's a profound difference under tagmemics between "is a
between "is a" and "may be used as a"
any object may be used as a prototype 18:45
that doesn't mean that all objects are prototypes
just as all objects maybe used as a boolean
fglock_ ok
wolverian no wikipedia page for our moose? aww.
TimToady but not all objects are of type Bool
fglock_ where can I read more about prototypes x classes 18:47
TimToady moritz: does it distinuish revisions from other bots such as specbot?
18:47 buetow joined, jisom joined
TimToady fglock_: prototypes are discussed in S02 and S12 18:48
according to grep, which is my friend. :)
fglock_ TimToady: thanks, I'll reread
18:48 edenc joined
rindolf TimToady: do you know APL or J? 18:49
jjore-w had a very nice experience with Oz last night. What a nice little language for soving constraints. :-)
fglock_ I'm reimplementing kp6 object+container system
TimToady what makes you think I'd admit it if I did? :)
veritos TimToady: you write JAPHs. 18:50
moritz TimToady: no, it doesn't
rindolf TimToady: :-)
"I'm not an actor! I just play one on T.V."
moritz TimToady: where should I link specbot revisions to? 18:51
veritos although APL can be more clean:
TimToady APLs monadic vs dyadic distinction is not really that far off from what Perl 6 does with prefix: vs infix:
veritos perl: print <STDIN> + <STDIN>
apl: (box) + (box) 18:52
TimToady however, I think APL is badly Huffman coded
wolverian APL has monads?! :)
integral The other monadic :-)
fglock_ the prototype is just a special 'instance'?
TimToady some operators need to be longer than one character 18:53
it's an undefined instance
so it can be used in reasoning about the class as if it were an instance, as long as you don't actually try to ask it to behave as an object. 18:54
veritos en.wikipedia.org/wiki/Image:LifeInApl.gif
lambdabot Title: Image:LifeInApl.gif - Wikipedia, the free encyclopedia
TimToady Unfortunately it's a different meaning of "prototype" from how "prototype OO languages" define it. 18:57
but we never thought of a better term
but our prototype is actually rather more like a prototype than a parent 18:58
our Rocket class is made out of cardboard, and just looks like a Rocket
but if you try to launch it, it won't work.
(not that you can't make a real rocket out of cardboard...) 18:59
jjore-w thinks cardboard rockets sounds like a game.
fglock_ ok, I think I found out a possible implementation
TimToady many small rockets actually are cardboard.
jjore-w rolls to see if his rocket launches. 19:00
TimToady you have to worry about pitches and yaws too
hmm, more likely the Rocket is made out of Styrofoam® 19:01
veritos TimToady: we get it. you worked at JPL. great. /sarcasm 19:02
TimToady well, happens to be I did work there, but my interest in rockets goes back way before that... 19:04
It's hard to have been born in the 50's and not be a space junkie. 19:05
veritos TimToady: good point. you and ousterhout..shoulda figured. 19:06
offby1 suddenly realizes that `++', when appended to a nick, could quite reasonably be pronounced "double-plus good" 19:07
fglock_ hmm - it's easier to bootstrap the object system out of 'Method' instead of 'Object', because both Method and Object have methods
[particle] metamethod system? 19:13
19:18 rashakil joined
fglock_ yes (kp6) 19:19
19:39 ilogger2 joined 19:49 edenc_ joined 19:57 zgh joined 19:58 fglock_ joined, fglock__ joined 20:01 fglock_ left, fglock__ left 20:02 fglock joined 20:05 Aankhen`` joined 20:09 veritos joined
svnbot6 r16046 | fglock++ | kp6 - added alternate MOP implementation (at temp/) 20:12
pasteling "evalbot_r16045" at 194.145.200.126 pasted "Pugs build failure" (309 lines, 17K) at sial.org/pbot/24365 20:14
20:14 dduncan joined
TimToady that's a rather odd error, given that Setup looks quite executable 20:18
veritos Larry crashed...must be close to finished with a pugs compile. 20:21
20:21 TimToady joined
veritos seriously--i crashed twice on package 91 of 92. 20:21
moritz which OS?
20:26 TimToady joined
veritos moritz: FBSD... 20:27
ooh, bad connection eh?
20:31 TimToady joined
veritos TimToady: bad weather? 20:32
20:37 TimToady1 joined
veritos in p6 pod, is the following line a valid bulleted line: 20:37
svnbot6 r16047 | fglock++ | kp6 - (alternate mop) created metaclass
veritos * foo
or do i still need to do the =over 4 =item thing? 20:38
oh never mind
20:41 bonesss joined
fglock veritos: you have the option the use =kwid perlcabal.org/~audreyt/tmp/perlkwid.kwid 20:41
20:42 TimToady joined 20:49 edenc joined
avar Has anything been done in the 6PAN project in a while? 20:49
Or some project with the intent of getting Perl 6 modules to CPAN and having something install them 20:50
svnbot6 r16048 | karhu++ | Added framework for Pyrl6 - Perl 6 in Python.
20:53 sili joined
fglock ?eval Int.HOW 21:08
21:09 evalbot_r16048 is now known as evalbot_r16036
evalbot_r16048 \\Class.new(("is" => ("Rat",)), ("name" => "Int")) 21:09
fglock does Class.new create a class? 21:10
or a metaclass?
it looks like it creates a metaclass, 21:11
but in this case, how do you create a class? 21:12
[particle] what's the difference?
fglock a metaclass describes a class
[particle] a class is an instance of a metaclass 21:13
fglock and the class 'is' the class
[particle] ...so new should create a class given a metaclass.
at least, that's how i interpret it, but i may be way off 21:14
fglock I see I'm still stuck with a bad concept of 'prototype' (and now 'metaclass') 21:17
21:22 thorat joined
fglock & 21:24
svnbot6 r16049 | fglock++ | kp6 - (alternate mop) added value/container stubs 21:26
21:42 perlmonkey2 left 21:46 edenc joined 21:53 demq joined 22:00 veritos joined 22:06 lumi joined 22:11 rissy joined 22:18 ilbot-test joined
zgh â¢/bye 22:25
22:30 veritos joined 22:55 veritos joined
offby1 yikes 22:59
guess that wasn't UTF-8
veritos offby1: eh?
offby1 what zgh said 34 minutes ago 23:00
veritos oh.
moritz no, it wasn't
and it wasn't clean latin-1 or iso-8859-15 either
offby1 IRC channels ought to announce their preferred coding system in their topics. 23:01
TimToady the logger decided it meant â„¢/bye
offby1 and the language, too, if that's not obvious from the rest of the topic ...
moritz which is nonsense
TimToady it also decided I should be boldface red somehow...
offby1 does the logger try to interpret the bytes as characters?
moritz offby1: it tries to guess if a line is latin1 or utf8, and decodes accordingly 23:02
offby1 gosh, it's ecumenical
moritz TimToady: is there a color you'd prefer? 23:03
scott asked me to color-code nicks... 23:04
veritos i would assume that he prefers a toady green.
moritz and so I did ;)
TimToady and maybe audreyt would prefer red, given she's chinese...
veritos TimToady: that's TAIWANESE to you. 23:05
TimToady 'sides, after a while the two capital T's start looking like red thumtacks. 23:06
Tang is a perfectly good Chinese name.
has a lot of history
but yes, I do rather like green.
how are you assigning the other colors? 23:07
moritz currently hand-coded 23:10
TimToady actually, I also like purple 23:15
maybe you should make audreyt infrared and me ultraviolet. :) 23:16
veritos TimToady: what's the shade of blue (in rgb hex) for the camel v3 cover?
TimToady hmm, no clue
but I'll bet it's changing with Perl 6 in any case
veritos about a nice light orange? 23:17
TimToady gaal should probably be moose colored 23:20
another idea is to not repeat the name for subsequent messages from the same person. 23:21
moritz It ried color: moose; in the CSS file, but firefox didn't recognize it ;)
TimToady maybe you should have tried "chocolate moose"... :) 23:22
moritz my ppor keyboard ;)
s/po/oo/
23:26 Limbic_Region joined
TimToady unfortunately, O'Reilly already has a Moose Book, the Apache Cookbook. 23:27
Limbic_Region salutatoins all
er um
Tene Is there a Goose Book?
Limbic_Region well, you get the point
demq Caribou 23:28
TimToady maybe we just need a Camel strapped to a Rocket... 23:29
Tene We could have the Moose (plural) Book. 23:30
moritz or a moose folded around an onion
veritos Tene: meese??? 23:31
TimToady I don't think anyone has used the polar bear in the snowstorm yet... 23:32
23:32 ortmage_ joined
moritz Ayn Rand used that for "Atlas Shrugged" 23:33
but everbody thought it was just a plain white cover ;)
TimToady I don't suppose we could get rights to Rocky and Bullwinkle... 23:35
veritos hmm, i wonder if we could get some Microsoft clip art . . . (naughty look) 23:37
23:38 [particle] joined
Limbic_Region are we talking about possible O'Reilly animals for a book? 23:42
veritos Limbic_Region: i think
Limbic_Region I think the perfect one to use would be the spoof I am sure you have all seen
perl6.ru/img/perl6book-parody.gif 23:43
offby1 :-)
I must say I like that.
TimToady it hath its charms
veritos wasn't that in one of audrey's presentations?
moritz nice ;) 23:44
bed &
offby1 bed & breakfast
Limbic_Region veritos - possible but I don't know for sure
veritos Limbic_Region: it's on youtube...just search for audrey yapc. 23:45
TimToady yes, she used it in composition with some other things
Limbic_Region is too busy eating
TimToady I hope you are eating neither your bed nor your breakfast. 23:46
buu What's wrong with eating breakfast?
offby1 ... and too eating busy
Limbic_Region It is officially the weekend and hence breakfast, dinner, lunch, supper, brunch, etc hold little meaning 23:47
I eat when I am hungry
offby1 that's crazy talk 23:50
Limbic_Region so if I am hungry only once, what do you call that meal? Or better, if I am hungry 24 times, do you give each of those meals a different name?
veritos Limbic_Region: and what about second breakfast? 23:51
thorat Limbic_Region: single man, are you? (standard Jay Leno joke) 23:52
Limbic_Region veritos - second breakfast only comes up when I am picnicing in the brier 23:53
thorat - effectively single until Monday

Powered by Cro and the Raku® Programming Language.

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

Thank you!