Think twice before running "make install" for Pugs | moritz.faui2k3.org/irclog/ | pugscode.org | sial.org/pbot/perl6 | ?eval [~] <m oo se> | We do Haskell, too | > reverse (show (scanl (*) 1 [1..] !! 4)) | "Perl 6 Today" video from YAPC::Asia: xrl.us/v6op
Set by agentzh on 6 May 2007.
00:08 bsb joined 00:13 autin joined 00:54 nipotaway is now known as nipotan 00:56 REPLeffect joined
svnbot6 r16210 | rhr++ | [Date.pm] use named args to BUILD 01:11
rhr when I run ext/DateTime/t/basic.t there's now a 50/50 chance that all test pass, which is an improvement! 01:12
01:15 araujo joined 01:19 lambdabot joined 01:28 sreeram_ joined 01:38 rindolf joined 01:56 TimToady_ joined 02:13 TimToady_ is now known as TimToady
svnbot6 r16211 | lwall++ | nick change message was backwards 02:22
02:25 knewt joined 02:38 jisom joined 02:48 autin joined 03:03 mako132_ joined, ProperNoun joined 03:25 mithraic_ joined 03:27 agentzh joined, kanru joined 03:43 dduncan joined
svnbot6 r16212 | agentz++ | PCR - added some more tests and a smartlink for sigspace 03:52
04:04 justatheory joined 04:07 justatheory joined
agentzh <'...'> and <"..."> are no longer valid according to S05? 04:38
lambdabot agentzh: You have 1 new message. '/msg lambdabot @messages' to read it.
agentzh @messages
lambdabot unobe said 9h 52m 57s ago: the current.build.yml should now show good paths to install pugs into.
agentzh unobe: nice
04:41 penk joined
TimToady agentzh: yes, they're now just '...' and "..." 04:42
agentzh TimToady: thanks :)
PCR needs some updates.
TimToady though PCR already fixed those, though maybe <> forms were going through a deprecation cycle...
*thought 04:43
agentzh oh
TimToady ?eval $_ = '$x'; / '$x' /
04:44 evalbot_r16189 is now known as evalbot_r16212
evalbot_r16212 Match.new(ā¤ ok => Bool::True, ā¤ from => 0, ā¤ to => 2, ā¤ str => "$x", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤) 04:44
TimToady ?eval $_ = '$x'; / <'$x'> /
evalbot_r16212 *** Cannot parse regex: <'$x'> ā¤*** Error: ā¤Match.new(ā¤ ok => Bool::False, ā¤ from => 0, ā¤ to => 0, ā¤ str => "", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤)
agentzh sorry, i was looking a very old version of PCR. ;)
*looking at
TimToady yeah, I think it's fixed
agentzh cool :)
rindolf Hi agentzh 05:10
agentzh rindolf: hi 05:11
fglock: i wonder if you would like to give Smart::Comments a short (for debugging) :) 05:12
rindolf agentzh: what's up?
agentzh rindolf: adding smartlinks to PCR's test suite.
(and also polishing the tests themselves)
rindolf agentzh: what is PCR?
agentzh Pugs::Compiler::Rule
rindolf agentzh: oh.
agentzh under perl5/
rindolf agentzh: what is the connectivity at your place? 05:13
agentzh web connectivity? 05:14
rindolf agentzh: no, Inet connectivity. 05:15
agentzh: upstream/downstream
REPLeffect must be about 100baud :-) 05:16
agentzh 100 MB, not sure. 05:17
rindolf agentzh: 100 MB per-second? 05:18
Mega-bytes, Mega-bits?
agentzh bits 05:19
rindolf agentzh: OK. 05:31
agentzh: and how much upstream do you have?
agentzh no idea 05:32
svnbot6 r16213 | agentz++ | PCR - more smartlinks for 09-ratchet.t (we now have 28 working links :)) 05:38
dduncan agentz, about your r16200 05:39
agentzh ?
dduncan agentzh, [smartlinks.pl]
* used Perl6::Perldoc::To::Xhtml's full-doc feature
* now we don't set UTF-8 for S26 since it's in Latin-1
why don't we just resave S26 as UTF-8 instead, which is what all files should be standardized as? 05:40
agentzh well, S26 is already in UTF-8.
dduncan so what did your comment mean?
agentzh but Perl6::Perldoc converts that to Latin-1, it seems.
i'm not sure.
that comment is incorrect.
dduncan if so, then Perl6::Perldoc needs fixing 05:41
agentzh dduncan: agreed
05:41 sreeram_ is now known as sreeram, sreeram left
agentzh there is a bug entry in Perl6::Perldoc's POD: "The parser does not assume a default encoding of UTF-8 (as per the specification in Synopsis 26)." 05:42
dduncan okay 05:43
agentzh ?eval $_ = 'a'; / $<cap> := (<alpha>) / 05:54
05:54 evalbot_r16212 is now known as evalbot_r16213
evalbot_r16213 *** Cannot parse regex: $<cap> := (<alpha>) ā¤*** Error: ā¤Match.new(ā¤ ok => Bool::False, ā¤ from => 0, ā¤ to => 0, ā¤ str => "", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤) 05:54
agentzh ?eval $_ = ' '; / $<cap> := (<ws>) / 05:55
evalbot_r16213 *** Cannot parse regex: $<cap> := (<ws>) ā¤*** Error: ā¤Match.new(ā¤ ok => Bool::False, ā¤ from => 0, ā¤ to => 0, ā¤ str => "", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤)
agentzh weird. it works on my machine. 05:56
is there any difference between /$<a> := <b>/ and /$<a> := (<b>)/ ? 06:03
06:03 lisppaste3 joined
agentzh i'd expect $/<a><b> to return something in the latter case. 06:05
but PCR doesn't work that way. 06:07
06:13 SubStack joined 06:17 marmic joined
svnbot6 r16214 | agentz++ | PCR - 09-ratchet.t - more tests for capture aliases and we have 34 smartlinks there now 06:29
06:37 b_jonas joined
agentzh fglock: it seems S05 does not state that <ident> and <alpha> are builtin subrules although S05 makes heavy use of them. 06:53
07:04 iblechbot joined
agentzh ?eval 'abc' ~~ /ab|abc/ 07:16
07:17 evalbot_r16213 is now known as evalbot_r16214
evalbot_r16214 Match.new(ā¤ ok => Bool::True, ā¤ from => 0, ā¤ to => 2, ā¤ str => "ab", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤) 07:17
agentzh bug?
| is with longest-token semantics, no?
moritz agentzh: I'd suppose so... 07:20
agentzh :)
moritz from => 0, to => 2 looks as if 'abc' is matched, isn't it? 07:21
lumi Isn't (from => 0, to => 2) 3 long?
moritz lumi: exactly
agentzh interesting
lumi What engine is that?
agentzh PCR
moritz ?eval 'a' ~~ m/a/
evalbot_r16214 Match.new(ā¤ ok => Bool::True, ā¤ from => 0, ā¤ to => 1, ā¤ str => "a", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤)
lumi I guess "to" means "just before" 07:22
agentzh *nod* 07:23
lumi So longest-token | is not implemented
?eval 'abc' ~~ /ab||abc/ # should be the other variant, right? 07:24
evalbot_r16214 Match.new(ā¤ ok => Bool::True, ā¤ from => 0, ā¤ to => 2, ā¤ str => "ab", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤)
agentzh yup
svnbot6 r16215 | agentz++ | PCR - more tests and more smartlinks (42 links now) 07:27
moritz ?eval ('abc' ~~ /abc | ab /).str
evalbot_r16214 Error: No such method in class Match: &str
agentzh fglock: failing tests are all marked as TODOs so that it's easier to do regression. ;) 07:28
moritz ?eval ~ ('abc' ~~ /abc | ab /)
evalbot_r16214 "abc"
lumi ?eval my %foo = ('ab' => 1, 'abc' => 2); 'abc' ~~ /%foo/
evalbot_r16214 *** Cannot parse regex: %fooā¤*** Error: ā¤Match.new(ā¤ ok => Bool::False, ā¤ from => 0, ā¤ to => 0, ā¤ str => "", ā¤ sub_pos => (), ā¤ sub_named => {}ā¤)
moritz agentzh++ # test cases and smartlinks
lumi Not like that, I guess
agentzh hmm, evalbot is using PGE? 07:29
%foo works in PCR
lumi I have no idea how to check
07:30 BooK joined
agentzh perlcabal.org/agent/S05 # linked with PCR's test suite 07:33
lambdabot Title: S05
agentzh apparently PCR need more test cases and more links. :)
*needs
fglock: maybe you'll find it interesting or even helpful :) 07:34
i'd reuse pugs' smoking mechanism to generate smoke.yml from PCR's test suite and then use it to mark the snippets in S05. 07:35
Hmm, 'make smoke' and 'make smartlink' can even be integrated into Module::Install as development tools (author-side) :) 07:36
not sure if audreyt will like the idea.
gotta rush back to the campus & 07:40
07:40 agentzh left 07:44 demq joined 07:57 buetow joined 08:01 DarkWolf84 joined 08:33 andara joined 09:16 theorbtwo joined, VanilleBert joined 09:21 aukjan|gone joined, aukjan|gone is now known as aukjan 09:48 ruoso joined 09:58 Grrrr joined 10:18 aukjan is now known as aukjan|gone 10:26 dduncan left 10:42 buetow joined 10:50 scw_ is now known as scw
avar @tell dvergin in S28draft.pod you say "These variables are all read-only" about $0, $1, $2. This needs to mention :rw apperently 10:56
lambdabot Consider it noted.
11:01 chris2 joined 11:22 ferreira joined 11:23 iblechbot joined 11:34 Kattana joined 11:38 fglock joined 11:43 nipotan is now known as nipotaway
svnbot6 r16216 | fglock++ | [PCR] - added non-greedy quantifier to "ratchet" emitter 11:43
fglock agentzh++ # more tests 11:49
agentzh: re %foo - I think it's a problem in the haskel/perl5 bridge; this works: 11:57
perl -Ilib -e 'use v6-alpha' - ' my %foo = (ab => 1, abc => 1); "abc" ~~ /%foo/; say $/.perl ' 11:58
haskell
12:03 prism joined 12:05 prism joined
fglock hmm - PCR hash-value semantics needs an update 12:13
12:29 lichtkind joined 12:51 IllvilJa joined 13:16 fglock joined 13:30 CindyLinz joined 13:36 andara joined
fglock TimToady: re STD - do you mean { ($<start>,$<stop>) = $.findbrack() } instead of <?{ ($<start>,$<stop>) = $.findbrack() }> ? 13:37
<{...}> would be used if it returned a regex, right? 13:38
13:51 vel joined 13:55 VanilleBert joined 14:05 [particle] joined
svnbot6 r16217 | fglock++ | [v6.pm] - added preprocess-STD.pl 14:17
moritz fglock: is preprocess-STD.pl the v6.pm equivalent to cheat.pl? 14:19
14:24 elmex joined
svnbot6 r16218 | andara++ | [runpugs] 14:26
r16218 | andara++ | - Session.pm: now with Moose
r16218 | andara++ | - Session startup handling now more robust
r16218 | andara++ | - Repl.pm: a very simple mini-REPL for testing
fglock moritz: yes; it replaces the {*} blocks with the actual code 14:28
PerlJam fglock: in which language? 14:29
moritz p5
fglock hmm - it could be written in p6 14:30
PerlJam It seems like it's taking forever to start using p6 to parse p6
fglock PerlJam: why? v6.pm has a lot of p6 code
PerlJam fglock: I know, it just seems like we're stuck in the hell of having most of the pieces needed for a full-fledged perl6, but not enough momentum to actually produce and release a full-fledged perl6 (i.e., something that one could announce to the world) 14:32
of course, my grousing about it isn't helping anything either 14:33
14:34 vel joined
avar building a cpan-installable command-line utility from the kp6/mp6 stuff would help IMO, even if the subset it parses/runs is very minimal 14:38
fglock avar: how about a v6.pm based one? 14:39
avar would work too 14:40
14:42 kanru joined
PerlJam fglock: Did you apply for a TPF grant yet? :-) 14:47
fglock re 'produce and release a full-fledged perl6', I agree it's a momentum problem, rather than a purely technical problem
PerlJam: no
back in ~2h & 14:48
PerlJam fglock: Well, consider that I at least think you should get paid for the cool work you do (have done) and TPF seems a likely source of funds.
avar the issue with p6 is not purely a lack of runtime problem, there are no p6 modules on cpan yet for instance, which sucks;/ 14:49
14:55 rashakil_ joined 14:56 veritos joined
veritos hey...new idea. use yacc or something to generate perl code from c. 15:00
15:00 amnesiac joined 15:07 mithraic joined 15:10 ruz joined 15:28 theorbtwo joined 15:31 b_jonas joined 15:33 jhorwitz joined
cj_ what is this "golf" that I hear so much about? 15:41
avar ?eval [~] <g o l f> 15:42
15:42 evalbot_r16214 is now known as evalbot_r16218
evalbot_r16218 "golf" 15:42
svnbot6 r16219 | andara++ | [runpugs] 15:44
r16219 | andara++ | -Preparing to make new version go live
r16219 | andara++ | No changes to front-end, more stable back-end
r16219 | andara++ | -Goal for next iteration is to port back-end to Perl 6
r16219 | andara++ | Help welcome!
avar andara: back end for what? 15:45
andara avar: runpugs consists of a server "back-end" and a client "front-end" 15:46
wolverian cj_, en.wikipedia.org/wiki/Perl#Perl_golf
cj_ wolverian: I play that game! 15:47
15:49 Loro_ joined 15:53 buetow joined 16:02 IllvilJa joined 16:09 weinig_ joined 16:15 frederico joined 16:16 chris2 joined 16:19 veritos joined
frederico Hi, I tried to install v6-alpha via cpan but I got recursive dependence with Pugs::Emitter::Perl6::Perl5 , but in source dir I did make install and everything worked. 16:19
16:40 [particle] joined 16:43 riffraff joined, cjeris joined
riffraff hi 16:43
16:43 cjeris left
veritos hmm...this is fun. making a man page for myself. 16:45
TimToady do you need to be documented? :) 16:47
myself, I just try to have an intuitive interface. :) 16:48
veritos TimToady: ok. am i section 1 or what?
feather.perl6.nl/~karhu/ for more info.
lambdabot Title: NAME
TimToady do you call yourself from shell scripts?
veritos TimToady: no, D\ Koenig is too hard to type. 16:49
[particle] veritos: don't forget to document your appendix
veritos [particle]: heh.
TimToady I don't want your colon
riffraff question: I wont to do an iteration like: for foo($x) { for foo($x) {bar($_} }
[particle] has a semicolon 16:50
riffraff is there a shortcut? I can't use two statement modifiers one after the other, sadly
16:50 mithraic_ joined
veritos riffraff: try closing the paren in bar($_ 16:50
riffraff lol 16:51
[particle] for foo($x) {.bar for foo($x) }
TimToady ?eval for (1..2) X ('a','b') -> $a, $b { say $a,$b }
16:51 evalbot_r16218 is now known as evalbot_r16219
evalbot_r16219 OUTPUT[1a1bā¤2a2bā¤] undef 16:51
TimToady well, it's not quite right yet, should autoflatten 16:52
riffraff [particle], yep, thought of that, still looks perfectible :) 16:53
TimToady so when you use $_, are you expecting to get only the inner loop var? 16:54
riffraff wow, .bar for foo($x) X foo($_) seem to work 16:56
TimToady, not really, I was just convinced I could not use $_ in the cross product in the beginning of the loop 16:57
TimToady well, X is there to do cross products, except it's supposed to flatten by default. but maybe the default is wrong in this case... 16:58
[particle] for foo($x) X foo($_) ?? not $x?
PerlJam [particle]: that looks like a syntax error to me.
[particle] meetoo
well, at least a bug. 16:59
oh, wait.
i mean, should the $_ be $x
?? not $x? is surely a syntax error, if it's perl :) 17:00
does X set $_ for you?
PerlJam That would be surprising to me.
riffraff interesting,. if I change my code from the for to the cross operator, it sucks so much CPU that it gets killed 17:01
avar veritos: your ~ page is funny:)
17:01 cj_ is now known as da_man
riffraff and my keybioard layout changed, wow 17:02
17:02 fglock joined, da_man is now known as cj
fglock oops - v6.pm Makefile.pl -> requires('Pugs::Emitter::Perl6::Perl5' => '0'); ??? 17:04
this will take some time to fix; v6.pm depends on a new release of Sub::Multi 17:05
17:06 zafoeta joined
zafoeta hi guys 17:06
PerlJam hello
zafoeta i am new in this group
just read some stuff about pugs
i am programming myself in perl5 currently 17:07
and wanted to get involved in pugs
fglock re v6.pm Makefile.PL - it was already fixed; needs a release 17:08
17:09 frederico joined
zafoeta anybody there? 17:09
TimToady nobody but us chickens... 17:10
PerlJam zafoeta: we're here ... ish
TimToady what you you like to do with pugs?
and what's your background?
zafoeta i am computer science student and working in a finance company 17:11
having been doing some perl in the last 3 years, but i would not call myself a huge perl hacker
i know how to program with it, though
TimToady would you like a commit bit to pugs? 17:12
zafoeta also, i like functional languages, and i am working at the moment in a project which involves functional languages, but have onle basics in haskell
yes
that is my intention
TimToady msg me your email and I'll get one sent
zafoeta helping in writing some code, and learning more haskell
i subscribed to the mailing list 17:13
TimToady most of the real work happens here
PerlJam zafoeta: do you know how to use subversion?
zafoeta yes 17:14
using it at work
riffraff sorry, could someone check if he can do "my %s = {'a'=>1}" in interactive pugs ?
it seem it parser correctly in a file but I get a sttrange error in the repl
zafoeta i know some good basics, will learn more if necessary
PerlJam great! Once you've got a commit bit, you can start hacking on pugs one way or another.
zafoeta :)
TimToady what's your emali address
*email
zafoeta sounds quite good and straight forward 17:15
i think i sent it to u privately
TimToady didn't get here, are you registered?
zafoeta nope
im actually never in irc
TimToady freenode requires reg to /mst
zafoeta ok, wait then
TimToady to /msg rather
or just mail to [email@hidden.address]
zafoeta wow, i am now taling to larry wall?? 17:16
:)
TimToady I'll never tell...
zafoeta *honoured*
PerlJam zafoeta: no, it's Mr. Incog Nito 17:17
TimToady I usually just answer to "Hey, you!"
zafoeta good
anyway, sent u the address 17:18
17:18 [M]erk joined
TimToady commitbit is thinking about you, but being slow... 17:20
okay, you should have email now 17:21
veritos Hey, you! Perl's supposed to be intuitive, and it has a manpage
TimToady rofl 17:22
obviously Perl 6 won't need a manpage. :) 17:23
veritos of course, it has p6doc
which is oddly enough written in perl 5
fglock TimToady: I've got a problem with STD - moritz.faui2k3.org/irclog/out.pl?ch...07#id_l183 17:24
lambdabot Title: IRC log for #perl6, tinyurl.com/2yx3e8
TimToady they've got drugs for that...
fglock :) 17:25
veritos ?eval my %s = a => 1; print %s
evalbot_r16219 OUTPUT[a1] Bool::True
TimToady fglock: no, that's intentional. That's an assertion being used as a bare block to be invisible to longest-token processing 17:26
ordinarily bare {...} limits your longest token
zafoeta k, im back
just checking out the code
TimToady it's customary to add yourself to AUTHORS as your first checkin to make sure it works right. 17:27
welcome aboard!
zafoeta ok.
will to that.
do you normally do check ins on the whole directory or only on single files?
TimToady either way is fine 17:28
zafoeta good
riffraff mah
(onomatopeic sound of not understanding life, universe and everything)
veritos zafoeta: we have few rules here. just don't destroy it beyond repair.
forgiveness > permission
TimToady if you've done a bunch of changes across directories, just checkin at the top
zafoeta ok
:) sounds all good. 17:29
fglock TimToady: oops - I confused it with a non-capturing <{...}>
TimToady yeah, the ? is not entirely orthogonal in this case
bugs me a little
but even if you've only changed one file, svn at the top is still pretty fast in my experience 17:30
fglock how about <??{..}> - as in ?? !!
TimToady so I mostly only do specific checkins when I know there's other stuff out there that I don't want to checkin yet (or are not sure)
veritos fglock: that's nice. and then <??{foo} bar !! baz> 17:31
TimToady it's starting to look a little too much like perl 5...
veritos TimToady: with (?(foo)bar|baz)? at least the ?? !! one wasn't picked at random 17:32
TimToady but <?{...}> isn't a conditional, it's just an assertion
fglock <true {...}>
zafoeta so, is this actuallly a group for perl6 in general or more for pugs? 17:33
PerlJam zafoeta: noth
er, both
zafoeta ok
TimToady we always answer "yes", except when we answer "no"
zafoeta and parrot as well, i suppose
PerlJam no, #parrot is for that. 17:34
TimToady parrot happens more on the mailing list
and #parrot
and #parrotsketch
zafoeta ok
cj TimToady: I hear that can get you in deep trouble with the security people :)
[particle] although i'm usually here for random parrot questions
TimToady but if you like functional programming, you probably want to hang out here more
cj TimToady: I think they prefer you to say no except where you say yes
PerlJam TimToady: Was that meant to be punny? :)
TimToady yes and no 17:35
zafoeta i wanna learn about haskell and perl in this project. that is why i chose pugs
veritos PerlJam: is 'hang' supposed to be a functional thing?
foo = 1:foo; main = do { putStrLn foo }
TimToady hang is a word that expresses dependency 17:36
PerlJam veritos: sure! It's solving the halting problem and that takes a while.
TimToady and in fact japanese basically uses the same word for the two concepts
(though spelled with different kanji)
zafoeta: what kind of computer are you running on? 17:37
zafoeta mac os x 17:38
what do u prefer?
have a linux as well
TimToady okay, we have some of those folks here to help
17:38 b_jonas_ joined
zafoeta good 17:38
thanks:)
17:38 b_jonas_ is now known as b_jonas
zafoeta just installed ghc yesterday btw 17:39
anything else i need?
PerlJam zafoeta: the newer the ghc the better 17:40
TimToady 6.6 is the minimum currently
zafoeta than i need to update ... 17:41
have 6.4.
TimToady 6.6 will be considerably faster too
offby1 _alarmingly_ faster.
cj gnu haskell compiler? 17:42
PerlJam offby1: you were alarmed?
TimToady zafoeta: how much memory do you have?
gnuvince cj: Glaskgow
PerlJam cj: glasgow
zafoeta 2 gb
TimToady that will suffice :)
zafoeta glasgow ghc, yes
TimToady any other mac folks on now? 17:43
cj ah
I don't think anybody else with names that start with 'g' should be able to create software. It muddies the water :)
PerlJam cj: and in an interesting turn of events, GHC was bootstrapped using perl5 :-) 17:44
zafoeta so, maybe people should think staring about a merger, in the world of globalization ...
TimToady (audreyt uses mac, but is being heavily distracted by real life at the moment.)
cj PerlJam: ha. everything is bootstrapped using perl5 - autotools require perl5 :)
[particle] sans is creating a perl security test: www.adtmag.com/article.aspx?id=20594 17:45
b_jonas cj: you mean like gnuplot?
lambdabot Title: Application Development Trends - Putting Coders' Security Chops to the Test
cj b_jonas: not following you
PerlJam TimToady: What's with all of you cabal people and your "real life" distractions? :-)
b_jonas cj: gnuplot has nothing to do with the GNU project, it's just named that accidentally
cj b_jonas: oh noes! 17:46
zafoeta so, just commited revision 16220
PerlJam b_jonas: though, had they taken a little time and explored the net a bit, they would have surely not named it that.
zafoeta seems to work fine
svnbot6 r16220 | walter++ | Walter Pallestrong joined.
zafoeta yep
TimToady there you are
17:47 fred__ joined
cj "oh noes" invokes images of Andy's LJ icon... 17:47
www.livejournal.com/allpics.bml?user=petdance
lambdabot Title: Userpics
zafoeta so ... where do i start of?
PerlJam zafoeta: what do you want to do?
zafoeta: write tests? Write docs? Write perl? write haskell?
zafoeta i would say the latter 2
PerlJam (almost anything involves writing at some point)
TimToady some start by reading the Haskell, some start by reading the test files 17:48
or the examples files
if you haven't read the synopses, that would be a good place to start
PerlJam TimToady: as big as the Synopses have gotten, I occasionally wish for a cliff notes version :) 17:49
b_jonas cj, perljam: www.gnuplot.info/faq/faq.html#SECTI...0000000000
lambdabot Title: Gnuplot FAQ
PerlJam b_jonas: man were they foolish! :) 17:50
17:51 ofer joined, _bernhard joined
zafoeta i am actually having a c++ exam tomorrow, but this intrigues me:) 17:53
cj b_jonas: cute :)
zafoeta: and well it should!
TimToady that's called "creative procrastination", and some of us are quite good at it
[particle] it's a continuation-based approach to learning 17:54
PerlJam heh
cj TimToady: I heard a bit on NPR that reminded me of you. There was a guy who wrote a book about creative procrastination 17:55
b_jonas cj: is that not Paul Graham? 17:56
cj I forget the name of the program, but it's a quiz show, and the host wrote a book called "Innuendo and out the other"
b_jonas: he was English, so his name might have been Graham...
TimToady and ouch the other is more like
b_jonas cj: www.paulgraham.com/procrastination.html 17:57
svnbot6 r16221 | fglock++ | [PCR] - added a 'modifier' attribute to the 'closure' node
lambdabot Title: Good and Bad Procrastination
cj there we are... "Whad'Ya Know"
17:58 Psyche^ joined
PerlJam Michael Feldman 17:58
17:58 Psyche^ is now known as Patterner
cj PerlJam: yeah, but he was the host, not the author 17:59
PerlJam www.amazon.com/Something-I-Said-Inn...0299202704 18:00
lambdabot tinyurl.com/2g4szo
PerlJam not it?
cj PerlJam: that's Michael's book, the one on procrastination was a different author's 18:01
offby1 PerlJam: I sort of _was_ alarmed at how much faster ghc 6.6 was compared to 6.4 -- I'm not used to that kind of speedup, and initially I thought something was wrong -- perhaps instead of compiling all my files, it was rejecting them!
cj oh, TimToady... you feel like being a guinnea pig?
PerlJam Oh, I was just taking my cue from the title mentioned.
offby1: Well, at least it was a good alarm :) 18:02
TimToady nobody who has spent two months in a hospital ever feels like being a guinea pig again...
cj www.amazon.com/connect 18:03
lambdabot Title: Sign In
cj TimToady: it's the blog software I get paid to write. I hear you've written a book or two and you don't use a mac ( which we don't support yet :( )
TimToady heh, it thinks I'm Heidi...
cj TimToady: you might be! :)
TimToady that would explain why she just came down with shingles... 18:04
18:04 SubStack joined
cj there's a "I'm not Heidi!" link at / 18:04
PerlJam As far as we know, you are Heidi
cj TimToady: TMI!!!
TimToady it seems anatomically unlikely to me
18:04 rindolf joined
PerlJam In fact, I was going to comment earlier that I occasionally wonder if you're really lwall and not some other family member or even a bot 18:04
:-)
TimToady what makes you think my family members aren't bots? 18:05
cj PerlJam: TimToady: the public face of the Wall Nuthouse
PerlJam heh
rindolf Hi PerlJam
Hi TimToady
Hi everyone.
TimToady howdy
PerlJam Hello rindolf.
cj hi rindolf!
PerlJam rindolf: how's unemployment been treating you? 18:06
rindolf TimToady: what have you daughter (now married) study in college.
PerlJam: it treated me very well.
PerlJam rindolf: or should I say "self-employment"?
rindolf PerlJam: heh.
I'm not self-employed yet.
PerlJam well, thus the scare-quotes
rindolf I contacted a Ma'ariv journalist about writing some articles for Ma'ariv.
Could be a nice source of income.
b_jonas PerlJam: he is a bot: lxer.com/module/newswire/view/29032/index.html
lambdabot Title: LXer: Paul Ferris: Pundit for a Day, 2005
rindolf I need to write them in Hebrew though. 18:07
18:07 drupek12 joined
rindolf If I can make a living from being an essayist, that would be awesome. 18:07
PerlJam rindolf: What is Ma'ariv ?
never mind. 18:08
wikipedia tells me enough
rindolf PerlJam: ok.
PerlJam: it's a newspaper.
It's also a prayer.
Or a time for a prayer.
PerlJam well, wikipedia uses the adjective "tabloid" to describe it.
rindolf PerlJam: it's not really a tabloid. 18:09
PerlJam: well at least not officially.
It's better than Yedioth Aharonoth and worse than Ha'aretz.
18:10 REPLeffect joined
PerlJam rindolf: what would be the subject matter of your essays? 18:10
rindolf PerlJam: lots of stuff.
PerlJam rindolf: perl?
:)_ 18:11
rindolf PerlJam: some about computers and FOSS.
PerlJam: Perl too, probably.
TimToady UFOs, mutant children...
rindolf Dynamic languages in general most likely.
Some about philosophy.
I still have to write the samples in Hebrew.
PerlJam: have you seen my "The End of IT Slavery"?
PerlJam: how are you kids doing? 18:12
PerlJam yes, I've seen it.
rindolf PerlJam: oh good.
PerlJam My kids are fine. The twins are slightly sick, but they've gotten over it except for the runny noses.
rindolf PerlJam: that's good.
TimToady: have you ever had depressions? 18:13
TimToady had a tropical depression once
rindolf I had depressions, anxieties, hypomanias, manias - the works.
I think I'm Manic-Depressive.
TimToady was pretty depressed in my early 20's
rindolf TimToady: what's a tropical depression?
TimToady: oh.
PerlJam rindolf: They make medication for that. A friend of mine started taking meds and he was night-and-day better
rindolf PerlJam: I'm taking medication. 18:14
REPLeffect Better than depressions from people touching you with 10 foot poles :-)
PerlJam TimToady: were those pre-Gloria days?
rindolf PerlJam: but it does not completely prevent it.
PerlJam: cogntivie psychology has helped me more.
PerlJam: have you read "Feeling Good"?
perlbot: feeling good
perlbot "Feeling Good" by David A. Burns - a field guide to Cognitive Psychology. xrl.us/vurb
lambdabot Title: Shlomi Fish' Book Recommendations
PerlJam no I've not.
rindolf PerlJam: you should. 18:15
PerlJam: it explains how to handle depressions very well.
PerlJam: and also useful as a preventative measure.
PerlJam I'll take it under advisement :)
18:15 jisom joined
PerlJam Though I've only had external experiences with depression. 18:16
rindolf PerlJam: you might be able to find an ebook online.
PerlJam: yes.
But people at any age can develop depressions.
PerlJam I'm too stubborn to get depressed about anything in particular.
TimToady PerlJam: yes, pre-Gloria 18:17
rindolf TimToady: how did you meet your wife?
18:17 macroron joined
TimToady but it's probably more like SAD, and California is a good place to get plenty of light 18:17
rindolf: her brother was my best buddy at college before she showed up 18:18
rindolf TimToady: oh nice.
TimToady: connections.
cj is pretty cheery most of the time... I might explode if left in sunlight too long...
PerlJam Maybe that's it. If I lived further north, perhaps I'd get depressed like other people.
rindolf "If you have connections, you don't need protexia"
PerlJam: I live in Israel and I get depressed. 18:19
cj sounds like a brand of condom
PerlJam rindolf: clearly you need to move closer to the equator then! :)
rindolf PerlJam: heh.
REPLeffect PerlJam: I think it depends on the person.
I could stay in a basement coding every day for a week easily. 18:20
I don't need the windows -- some folks do.
cj REPLeffect: me, too!
REPLeffect In fact. In my old house, that's exactly what I did :-)
(have no basement now).
cj they put us in this friggin' corner office with tons of windows - we pull the blinds when the sun comes out
18:21 amnesiac joined
PerlJam tries to imagine TimToady in college and doesn't notice much difference 18:21
TimToady cj: in Seattle?
PerlJam maybe the mustache was bigger? Sideburns longer?
cj TimToady: yar
TimToady: it's not often...
[particle] watching the folks walk in and out of uwajimaya i suppose
TimToady well, they really don't understand that big yellow light in the sky up there...
REPLeffect "Is that a UFO?" :-) 18:22
TimToady no, it's a mutant children
PerlJam REPLeffect: more like "Ahhhh! It burns!"
cj [particle]: the front door is obscured by some building... but otherwise I would be :)
REPLeffect "my eyes, my eyes!!!"
PerlJam exactly
cj [particle]: they put me in the big black building, not the ones by union station
[particle] the sun definitely has that effect here. 18:23
PerlJam Suddenly all programmers have been compared to Gollum.
TimToady the buildings are too shiny from all the rain
[particle] yesterday, ~60 and cloudy. today, the sun came out, so it'll be ~70.
cj: i see
TimToady if that were here, it would have swung between 50 and 80
cj [particle]: I get to watch people drive out of the express ramp 18:24
[particle] heh
TimToady that would be down at Seneca or James or so? 18:26
or Mercer?
[particle] wonders when his favorite open sores developer returns to seattle
REPLeffect Ooh. An "open sores" developer --- sounds painful. 18:27
TimToady Perl 6 is definitely a sort point with some folks...
[particle] yeah, ingy dot net has a framework on his left wrist
TimToady *sore
REPLeffect darn muscle memory.
TimToady gah, fingers are overtrained
cj TimToady: 5th & Columbia
rindolf PerlJam: I'm trying to find the time to take a look at Punie. 18:28
[particle] TimToady: that's below jackson
rindolf PerlJam: then hopefully, I'll start adapting it into rindolf.
[particle] err, no.
i'm still thinking about union station
REPLeffect trying to figure out how Perl 6 could actually be a "sort point" with some folks :-)
perl -e " .... sort ... " 18:29
s/"/'/g
TimToady yeah, between Seneca and James
offby1 jerks awake
those are streets in Seattle.
[particle] offby1: you're so five minutes ago 18:30
offby1 I am indeed.
cj: you're at Amazon?
REPLeffect "jerks awake": That sounds like a bad slogan of some kind.
TimToady I'm just showing off my knowledge of geography.
or maybe ancient history...
offby1 REPLeffect: ok, then feel free to substitute "wakes with a start". 18:31
TimToady for me, Seattle is like, *so* 30 years ago...
rindolf Hi offby1 , [particle]
offby1 waves
PerlJam TimToady: As long as you have the appropriate amount of fun (and we get a perl6 compiler out of the deal somehow ;-)
TimToady the former is more certain than the latter. :)
REPLeffect offby1: Sorry, I just can't resist seeing alternate meanings (a family trait). 18:32
cj offby1: yar
zafoeta can it be that perl6 has a lot of functional influences?
REPLeffect and disfunctional influences.
PerlJam zafoeta: of course.
REPLeffect (there I go again)
[particle] zafoeta: yes, and a lot of disfunctional interfaces 18:33
REPLeffect: beat me to it ;)
cj TimToady: I figured you were using google maps :)
REPLeffect I was just hoping to beat TimToady to it :-)
TimToady you did, barely
but I'd've spelled it right 18:34
reaching for the y key cost me the prize
REPLeffect hehe.
insert random rant about English here.
(or not-so-random-rant)
TimToady cj: I admit that I double checked with google there at the end.
[particle] it's greek to me
TimToady greek is a nice language 18:35
and great training in case grammars
cj when people say "functional programming" what do they mean?
REPLeffect I'll let it pass for the next guy this time :-) 18:36
TimToady generally they mean programming without side effects, to some extent or another
cj can you give me an example of a side-effect?
TimToady $a = 42 18:37
zafoeta reading from the synopsis i see that perl6 will have proper types ... :)
offby1 cj: How many times a day do you hear someone say "Only connect! -- EM Forster" ?
TimToady zafoeta: yes, primarily motivated by the desire to do multimethod dispatch an the like 18:38
cj TimToady: is that it? I must be missing some context...
ferreira but (let (a 42)) does not produce a side-effect.
TimToady Lisp is not a pure functional language
zafoeta so, it might also be more interesting for numerical programming purposes
ferreira but the let construction looks like a pure functional construct 18:39
TimToady it's interesting for problems that are solved by lining up a lot of definitions
zafoeta people in my industry always complain that perl is not good enough for what they are doing, and prefer python ...
TimToady not so good for modeling the real world, where OO reigns
zafoeta reading the synopsis, i like it a lot:) 18:40
TimToady we've been working on it a long time...
so most of the rough edges are worn off already 18:41
ferreira I dare to say that to many people the problem with Perl is that it gives them too many choices. And they don't want that.
offby1 there's something to that.
that's the attraction of scheme, in my humble opinion: There's Only One Way To Do It
TimToady yeah...that's part of why we're trying to give at least one obvious way to do it right. 18:42
offby1 s/the/part of the/
18:42 japhb joined
offby1 choice can indeed be overwhelming. 18:42
choice without guidance, anyway.
rindolf Bleh! I feel bad that my Wikipedia article was deleted.
TimToady we think we can make the right way easy to do by accident, and still provide flexibility
offby1 that of course is the ideal.
rindolf About the Haifa Linux Club.
Maybe I should join the wikipedia mailing list and discuss it there. 18:43
Discussing stuff on wikipedia using threading makes me frustrated.
ferreira Blessed hashes as objects is the easy way for OO in perl 5. And even so, inside-out objects are amazing.
zafoeta also, the problem of choice is that bad programmers will normally do a bad choice
TimToady ferreira: inside-out is amazing, but also kinda wrong. :) 18:45
all the p5 approaches force the programmer to know too much about the class.
well, except maybe Moose... 18:46
and that can be blamed on p6 :)
ferreira What do you mean? The opaqueness that inside-out objects provide by saving the attributes elsewhere is the cunning trick?
TimToady at the price of making it harder to write metacode like serializers 18:47
even DESTROY can be problematic, depending
p5 really falls down on expressing the relationships between classes 18:48
and the confusion between functions and methods just keeps biting people over and over 18:49
ferreira I see. 18:50
cj isn't a function a method!? :) 18:51
TimToady p5 basically proves it's possible to make your OO design too orthogonal with the rest of the language.
ferreira I learned a lot from reading "Perl Best Practices". The name of the book emphasizes the positive side of the things. On the hand, there are the hundreds of ways not to do it. That makes Perl a kind of dangerous language: don't try it if you don't know what you're doing. And that's only true if you start to do advanced programming. 18:54
zafoeta my Hash of Array of Recipe %book; is the same as my Hash of Hash of Array of Recipe $book; ?? 18:55
TimToady pretty close, except for the variable itself 18:56
certainly you could subscript them the same once you're past the var name 18:57
but if you had both in the same namespace, they wouldn't conflict
zafoeta i see, so, with the above declaration, could i say %book = $book ? 18:58
ferreira cj: nope, a method is a function. But there's no easy way to tell when a function is not a method.
TimToady %book := $book
zafoeta ok,thx
18:59 mithraic joined
rhr I'd like to start writing some tests, any suggestions on what needs more testing? 19:06
19:10 REPLeffect_ joined 19:11 fglock joined
TimToady the sections in the synopses that are barren of smartlinks, maybe? 19:12
rhr OK, I'll have a look 19:18
19:19 cognominal joined
[particle] looks like pugs needs a coverage tool too 19:20
19:26 ozo joined 19:33 ferreira left
svnbot6 r16222 | fglock++ | [PCR] - simplified perl5-ratchet-closure emitter 19:45
19:55 jvoorhis joined 19:58 zafoeta joined 20:02 DarkWolf84 joined 20:07 veritos joined 20:08 dduncan joined 20:11 dduncan left, dduncan joined
veritos heylo 20:14
20:15 justatheory joined
masak veritos: good evening 20:15
veritos afternoon.
20:16 Aankhen`` joined
veritos TimToady: i think we should have all people write their own man pages. 20:18
makes everyone feel like one big (dys)?functional family.
20:20 obvio171 joined 20:21 justatheory joined 20:33 iblechbot_ joined 20:37 riffraff joined
riffraff hi 20:38
svnbot6 r16223 | fglock++ | [PCR] perl5-ratchet - implemented <?{...}> and <!{...}> 20:40
avar fglock: go go go:) 20:42
fglock: I know PCR has emitting features, could it potentially be made less slow my caching the compiled P5 code?
fglock avar: :)
avar: it does, already - it uses Cache::Cache 20:43
avar mm, cool 20:44
fglock it's about 30% slower with an empty cache (perl Makefile.PL empties the cache) 20:45
20:46 polettix joined
avar Pugs::Emitter::Rule::Perl5::Regex looks cool:) 20:46
fglock yes, I plan to replace parts of the perl5-ratchet emitter with it 20:48
the non-ratchet engine is harder to replace; backtracking into subrules is hard in p5regex-land 20:49
avar how so? 20:50
fglock when you call a subrule like /.*/, it needs to return all possible combinations 20:51
it may be possible with inlining
avar you could always make it (*FAIL) if I understand you correctly 20:52
fglock where / a <subrule> b / becomes / a (.*) b / 20:53
avar would compiling it to something like / (?: a <subrule> | a (.*) b (*FAIL) ) .. /x work? 20:55
fglock avar: not sure I understand it 20:56
avar not sure I do:) 20:57
fglock I mean: rule subrule { .* }; / a <subrule> b /;
can perl5.10 call sub-regexes like this? 20:59
home & 21:01
21:07 prism joined 21:17 prism joined 21:23 mithraic joined 21:30 IllvilJa joined 21:33 polettix is now known as poletti 21:35 poletti is now known as polettix 21:41 stef__ joined 21:43 SubStack joined 21:45 jjore-w joined 21:52 araujo joined 22:01 obvio171 joined 22:03 zafoeta left 22:06 obvio171 joined 22:14 pbuetow joined 22:34 larsen_ joined 22:39 goban joined 22:48 SubStack joined 23:17 obvio171 joined 23:20 obvio171 joined 23:22 obvio171 joined 23:59 DarkWolf84 joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!