pugscode.org <<Overview Journal>> | pugs.kwiki.org | logged: xrl.us/e98m | FBSD_4.11:r1284(118/3242) win2k:r1284(118/3242) Linux:r1197(105/3116) MacOSX:r1197(105/3116)
Set by kungfuftr on 28 March 2005.
kungfuftr (perl medic)++ # good business reasons as well as technical 00:05
stevan ingy? you still there? 00:58
Limbic_Region seen ingy 01:02
jabbot Limbic_Region: ingy was seen on Tue Mar 29 07:54:42 2005
stevan damn the rotation of the earth 01:03
they sleep while we are awake, we sleep while they are awake
bah
thank L~R 01:05
Limbic_Region walang anuman stevan
stevan gahzuntight :P 01:06
Limbic_Region the solution btw is to work on 26 hour days - you shift in an out of alignment with all 24 timezones ;-)
walang anuman = you're welcome in Tagalog (Philippines)
stevan we need a perl6 jet which just circles the earth chasing the sun 01:07
Limbic_Region what we really need is for me to win the lottery so I can personally fund the pugs/p6/parrot/ponie hackers around the world full-time in a central location 01:08
stevan Limbic_Region: from your secret world headquarters high in the himalayas 01:09
where you can plan to take over the world one shell script at a time :P 01:10
(sorry, been a long day, the caffine is wearing off)
Limbic_Region was voted most likely to sit on a mountain top philosophizing (really) in highschool 01:13
stevan s/philosophizing/plotting world domination/ :)
sounds cooler
Odin-LAP You mean those aren't synonymous?
stevan Odin-LAP: good point 01:14
ingy stevan: back 01:25
stevan hey
so I was wondering how your Makefile battle went?
did you fix the Test.pm problem?
ingy good I think
yes.
stevan excellent
ingy yes, sorry about that 01:26
stevan do you think it is safe to turn on the File::Spec Makefile?
ingy did it overwrite or shadow the Perl5 Test.pm
stevan not sure I never use Test.pm :)
ingy stevan: I can do it
I'll make sure it is safe 01:27
stevan ingy: your a sweetheart :)
ingy ^^ *you're*
stevan yeah that
stevan has had a long day, and should really just go watch TV and vegetate, but is somehow drawn to Pugs 01:28
ingy Installing /Users/ingy/local/lib/perl6/File/Spec.pm 01:30
Installing /Users/ingy/local/lib/perl6/File/Spec/Unix.pm
Installing /Users/ingy/local/lib/perl6/File/Spec/Win32.pm
:)
stevan excellent
thanks much
ingy++
jabbot pugs - 1292 - We can install File::Spec now that MakeM 01:37
pugs - 1294 - more Pod::Stream::Parser work; added bet 02:07
pugs - 1293 - adding Rule test to ref.t; adding variab
pugs - 1295 - added File::Spec usage in the test; adde 02:17
kungfuftr chip: yullo? 02:31
crysflame File::Spec? 03:30
in perl6?
ingy++
04:42 cnhackTNT|away is now known as cnhackTNT
joejiang wait a minute 04:43
jabbot pugs - 1296 - add some tests on **,Inf,NaN 04:57
05:09 cnhackTNT is now known as cnhackTNT|away
theorbtwo wonders if anybody is up. 05:36
simcop2387 no
Braccae tuae aperiuntur
theorbtwo wonders if there is English involved somewhere there.
simcop2387 its latin for "Your fly is open" 05:37
theorbtwo Ah.
simcop2387 Quid agis, medice? i wish i could say that with the right accent
"What's up, doc?"
gaal www.kimbly.com/blog/000402.html <- this, thankfully, isn't happenning to us 06:49
theorbtwo Aaah, people are beginning to show up!
gaal morning, theorbtwo :)
theorbtwo ...that we know of, gaal. 06:50
gaal for one thing, we aren't billing.
theorbtwo True enough.
gaal if par makes it to the core (it had better!) then deployment will be ever easier
most of the things in the top row are fixed by the open dynamics of the project 06:51
there was no mention of quality, but i think we're not skimping on that one 06:52
well, the recent talk on p6l makes me think of the system analyst pane :) 06:53
theorbtwo Sadly, I missed it.
I suppose I could have read the IRC log... in fact, IIRC, I read some of it...
gaal deploying p5 in production environments is a *pain*. 06:54
to2, read p6-lang, thre's some talk about very weird new operators for units. they'll be hella powerful, but weird. 06:55
mugwump yes, that's why there are so few production environments running p5 06:56
jabbot pugs - 1297 - Locale::KeyedText update - completed syn 06:57
theorbtwo Interesting, gaal. 07:03
sdtr443w So how do I actually run PUGS after I install it? 07:04
gaal the same way you run PERL 07:05
sdtr443w Well ... uhh ... is there supposed to be a binary or something? 07:06
A script?
scw binary
gaal there's a binary, sure. did you "make install"?
it's called pugs
sdtr443w yes
Well...
gaal pugs -v
sdtr443w I don't see it
I'll try it one more time
gaal how did you install, from sourec? binary package? 07:07
jabbot pugs - 1298 - SQL::Routine update - this work was done
sdtr443w Wow it's doing a lot now for whatever reason
I gotta look at my history and make sure I really did run it twice before already 07:08
(make install, that is)
It was skipping a bunch of stuff so I have this feeling it got halfway through it and quit before 07:15
gaal before you make install, please make test 07:16
mugwump Anyone know who steven little @ iinteractive is on IRC? 07:19
gaal stevan?
mugwump oops, ye s:)
theorbtwo was looking through Parser.hs earlier... 07:20
mugwump oh, duh, there's the AUTHORs file
theorbtwo grins.
sdtr443w It seems to terminate while trying to compile AST in the make phase 07:22
gaal error message?
sdtr443w It had died earlier in Rule.Token
nothing
just "make: *** [pugs] Terminated"
gaal welrd. do you have enough free memory / disk? sounds like a system problem to me. 07:23
what version of ghc are you using?
+ OS
sdtr443w It may be a system problem -- the computer sucks 07:24
128MB of RAM being the main constraint. I have many GB of hard disk space free.
I installed GHC 6.4; I'm running all this on 32-bit SuSE 9.2. 07:25
gaal i'd try attaching strace to ghc and see if it dies from some signal
start make in one window; then in another, ps aux|grep ghc to get its pid 07:26
then do strace -o somefile -p that_pid
when make dies, tail -30 somefile 07:27
sdtr443w Would you be interested in this now? I wasn't expecting so much trouble and I intended to go to bed soon.
gaal that might clue you in.
sdtr443w But I'll be trying this tomorrow evening.
gaal it's you who should be interested :) we're just here to help.
sdtr443w Well, if it's interesting some people would likely want to know
Though I dunno if this is the right hour for this kind of thing on this channel 07:28
gaal this channel has no hours, or rather, has all hours
it's morning for me now, but if you want to pick this up later, feel free to come again later.
there are sure to be people here to help you then. 07:29
nothingmuch sdtr443w: i think it might be worth to ask on #haskell about ghc issues like this 07:30
i think i might be able to help
but only with constant ping-ponging
so you'd better bring this up again when you have more time
sdtr443w Yeah well it may be a GHC thing since the machine I'm doing this on seems to sieze up when compiling AST 07:31
I couldn't even get it to run ps
Juerd food morning!
eh
good morning!
nothingmuch there's a param to GHC in the Makefile
-H200M i think
it tells it to use 200M of memory to do some stuff
Juerd Psychological typo - I do want food :)
gaal "freudean sleep" :)) 07:32
sdtr443w Oh that would be relevant
Juerd gaal: heh
sdtr443w nothingmuch: I changed -H200 to -H100 wherever I found it in the Makefile. It seemed to survive this time 07:34
If it's assumed I will have 256MB of RAM, does this mean PUGS is going to crawl for me? 07:35
Juerd Are you discussing Hyundai vans? :)
(H100, H200)
sdtr443w They make vans?
Juerd Yes, the H100 and the H200 :)
Perhaps more.
sdtr443w Well I must have the "lite" version then.
theorbtwo Anyway, I was wondering if there was a good reason that identifier `sepBy1` (try $ string "::") is spread all over the place, rather then being in some central location.
Juerd My dad owned two H100 and has his second H200 now
sdtr443w He must like them 07:36
Juerd A comfortable turbo diesel, that current one
sdtr443w I wish I had a turbo on mine
Juerd Airco (uncommon here), and for a van, silent too.
sdtr443w I have a short bus and it crawls sometimes haha
Juerd My dad easily wins from my 1.6 Hyundai Lantra 07:37
sdtr443w Oh grand -- the make command didn't find readline
Juerd It's really shameful to see a van accellerate quicker
With you flooring
gaal readline is optional, sdt443w. 07:38
Juerd Wow, gaal, do you type nicknames fully?
sdtr443w well it caused ld to return 1 and then I get "make: *** [pugs] Error 1
Now gaal didn't type it right 07:39
Here's how you do it:
Juerd heh.
sdtr443w 1st -- type Rocko and realize all variants of it are in use. Then run your left fist all over the left side of the keyboard, and hit enter.
And then you get an enigmatic handle that everybody talks about.
Juerd The s, d, t, r and w are all on the right side of good keyboards. 07:40
sdtr443w Oh God here we go with the Dvorak
Juerd I'm not God though
I don't think God is particularly interested in Perl 6 development 07:41
sdtr443w If you can get PUGs working for me, I'll worship you like one.
Juerd As it'll eventually cause His unemployment
:)
theorbtwo Well, I'm a god, and I am.
Juerd theorbtwo: Which one? 07:42
gaal s.*, when you started make, did you get a message about readline not being found?
sdtr443w This installer certainly treats readline like it's mandatory. I do have it installed anyhow.
lemme see
Juerd theorbtwo: One of those deprecated Roman gods, or the all new and shiny ucfirst God? 07:43
theorbtwo is a perlmonks god of the semiactive sort.
sdtr443w Well when I run it now I don't see a warning about it missing until the linker keels over.
Juerd sacrifices a piece of toblerone to theorbtwo 07:44
sdtr443w I've had to run make a few times and I don't have the first run anymore, if they did anything.
theorbtwo tell sdtr443w about pastebot
pastebot?
Juerd pasteling?
Or just use tnx.nl/scribble/
theorbtwo perlbot, tell sdtr443w about pastebot
perlbot, nopaste?
perlbot Paste your code here and #<channel> will be able to view it: sial.org/pbot/<channel>
theorbtwo Ahh, there it is. 07:45
Anyway, now that we've introduced you to about ten million different services, use any of them to show us what you see when you do a make.
Juerd bot/<channel> 07:46
gaal s.*, the readline note comes only during configuration, ie at the first make after a make clean.
Juerd argh.
gaal brb
Juerd DRAG AND DROP COPY AND PASTE? WHO THE HELL INVENTED THIS MADNESS?
theorbtwo Microsoft, I think.
Juerd Apple, but close 07:47
jabbot pugs - 1299 - A possible Perldoc implementation strate
Juerd But in a terminal
Rather inconvenient.
q[acme] i wish jabbot would give a few more characters
Juerd q[acme]: gy 07:48
sdtr443w OK I have a save of the original run when I configured it
brentdax Does pugs just operate at :u0, or does it have some sort of Unicode support?
sdtr443w It claims it can't find libreadline.so.4: No such file or directory. That's interesting.
Juerd q[acme]: :)
sdtr443w Do you want a paste still?
I did a locate and I see it in /lib
Juerd brentdax: IIRC, everything's unicodeish
brentdax: With no level specification
u4 that is iirc 07:49
But these *numbers* are terrible to remember.
brentdax Hmm, okay, guess I'll leave it stubbed ot...
sdtr443w But that must be GHC's problem
theorbtwo Hmm...
brentdax is working on porting URI::Escape.
theorbtwo perlbot, seen autrijus
jabbot theorbtwo: autrijus was seen on Tue Mar 29 03:00:41 2005
theorbtwo jabbot, what timezone?
jabbot theorbtwo: Have you asked such questions before?
Juerd brentdax: I still think all encoding/escaping should be abstracted and unified 07:50
obra autrijus is on a train
theorbtwo perlbot, seen theorbtwo
jabbot theorbtwo: theorbtwo was seen on Tue Mar 29 15:49:52 2005
sdtr443w Oh God scratch that. That was from my original botched attempt before installing GHC. Afterwards, I don't see any mention of readline
hcchien obra: on a plane. :)
sdtr443w I think I need a nap
obra hcchien: oh!
theorbtwo Aaah. Know when he'll be on-channel again? 07:51
Juerd hands sdtr443w a nap
sdtr443w Yeah I'm going to put this off until tomorrow. 07:52
brentdax Wow, I'm slightly less than clever. It's already been ported, although it's a straight port, not an idiomatic one. 07:53
hcchien theorbtwo: more 1.5hrs, I guess.
theorbtwo Hmm. I want to hack on the parser a bit under his tootlage, but if I stay up another hour and a half, I'll be too asleep to do so. 07:55
clkao seen autrijus
jabbot clkao: autrijus was seen on Tue Mar 29 03:00:41 2005
obra clkao: when should I pull to test that your fix fixes it all?
er. mischan 07:56
theorbtwo That's GMT+7, BTW.
scw +0800, infact 08:03
nothingmuch has a go at prototyping that GC stuff 08:04
gaal hey, vim users, add stuff to util/perl6.vim - i 08:28
''ve been doing it as i go along
kungfuftr morning all 08:33
jabbot pugs - 1300 - add some more builtin types 08:37
pugs - 1302 - 08:57
pugs - 1301 - Fixing perl6.vim: 0o123 and :key<val>
gaal please put messages on your svn ci's :) 08:58
scw That one is generated by svk and contains no changes.. 09:05
(except svk:merge property..)
kungfuftr pugscode.org <<Overview Journal>> | pugs.kwiki.org | logged: xrl.us/e98m | FBSD_4.11:r1302(183/3383) win2k:r1284(118/3242) Linux:r1197(105/3116) MacOSX:r1197(105/3116) 09:30
brentdax Can Pugs built-ins be implemented in Perl? 09:51
nothingmuch not all
or do you mean perl6?
in that case not yet, perhaps
brentdax Yeah, Perl 6.
nothingmuch there is one boundry that's hard to cross
system calls
that has to be builtin, as some kind of interface
brentdax I'd like to write part (which I backported to Perl 5), but I don't know Haskell. 09:52
nothingmuch anything else is MMD dispatched
the builtins are pretty simple, actually
could you list a few? I might be able to help you start
they are mostly wrappers for haskell functions 09:53
except that they do a little castin
g
as for writing them in p6: we don't really have all the stuff in pugs yet
brentdax nods.
nothingmuch but when perl 6 will be written in perl 6, it's only logical that they are implemented as perl 6
brentdax Could probably be done by having a sort of default 'use', though..
nothingmuch :w
oops, wrong window
brentdax vi user? 09:54
nothingmuch yeah, but we don't have many required facilities
nothingmuch confesses
no BEGIN { }, no type based MMD lookups
from specific to generic
no objects to represent many builtins (fhs) 09:55
(or arrays, hashes, etc, for that matter)
brentdax Can you construct an anonymous array, take and execute a coderef, and perform a for() loop? 09:56
nothingmuch you mean, implement map, etc? 09:57
i think that should be possible technically
brentdax Yes. part is not much different from map.
search.cpan.org/src/BRENTDAX/List-P...03/Part.pm --My Perl 5 version
nothingmuch good module, i use it occasionally
since for basically takes pointy subs
i wonder if gather/take could be streched to make them be able to do part
brentdax++ 09:58
nothingmuch doesn't need it that often though
if i go that far I usually need set operations, or nothing at all
brentdax Neither do I, but when I need it, there's nothing better.
nothingmuch you could write a test
that tries to see if for can take code refs 09:59
brb
brentdax k
nothingmuch ok, back 10:03
but it's lunch time
do you have commit access?
hola lightstep
i'm prototyping various GC schemes, would you like to see later?
i'm starting to think GC zones actually do make sense 10:04
brentdax Me? Not on Pugs, I've not done anything to it yet.
(I assume autovivification is a pipe dream at this point?)
lightstep yes 10:05
but how can you implement gc in a gc'ed interpreter?
nothingmuch lightstep: an object called 'memory'
which is really a set of objects
lightstep oh
nothingmuch and an object called 'memory object'
and some event generation
lightstep but in an interpreter, memory blocks in the object language are usually memory blocks in the meta language? 10:06
nothingmuch (object has ref taken of it, object is released, object refers other, object removes reference to other)
lightstep (like pugs works now)
nothingmuch i don't understand the terms
autrijus good morning folks.
nothingmuch (object is born)
obra hi autrijus
nothingmuch hola autrijus
ack, i can't find my lunch slips
lightstep object language = pugs, meta language = haskell
nothingmuch ah 10:07
autrijus "host language" too.
nothingmuch i'm doing it in p5
jabbot pugs - 1303 - More playing with the DOM
nothingmuch and either way, it doesn't really matter
lightstep the memory of the unref'ed objects won't be freed anyway until haskell gets to it
oh
nothingmuch since deletion is just moving the 'object' object to the 'deleted' space
ah
ok 10:08
well, again, if it were freed i couldn't check that freeing was correct
i plan to use full mark & sweep as the reference GC
=)
brentdax: can i have your email please?
lightstep you can also create a noop gc module to test the prototype
brentdax [email@hidden.address]
nothingmuch autrijus's rule is that you have to comit tests first, though 10:09
so do the for/cref thing
and then try to hack it so it works
or, test for 'part'
and try to implement it in p6, and then in haskell
invitation sent
brentdax I assume the Chinese(?) legalese doesn't require me to give them my first-born child or something. 10:11
rgs no, only the second one 10:12
brentdax Oh, okay. My family has a history of...interesting middle children anyway.
autrijus # wagner.elixus.org/~hcchien/termtouse.html 10:13
sf.net boilerplate
brentdax Thanks. 10:14
I guess I can just port over the List::Part tests... 10:18
autrijus yup.
obra pugscode.org <<Overview Journal>> | pugs.kwiki.org | logged: xrl.us/e98m | FBSD_4.11:r1302(183/3383) win2k:r1284(118/3242) Linux:r1302(193/3383) MacOSX:r1197(105/3116) 10:19
brentdax We don't have is_deeply, do we? 10:21
autrijus we don't but you can trivially implement it by comparing .perl
brentdax Ah, cute...
ihb perhaps a stupid question, but why are modules being ported to Perl 6? aren't you supposed to be able to use Perl 5 modules in Perl 6? (i can see some reasons, but i'd much rather have a good answer :-)) 10:46
obra ihb: eventually, sure. but there's no machine translation framework.
it's currently possible to do some things using an embedded perl5 interpreter.
kungfuftr autrijus: DBI::PurePerl has way too many Perl5 idiosyncraticies 10:47
jabbot pugs - 1304 - test syntax about @=DATA etc.
obra kungfuftr: that's a pity
kungfuftr yup, is a pain 10:49
fayland_ jabbot: ya,it's from me. I think maybe I should add the "todo_fail("FIXME parsefail");"
jabbot fayland_: That's quite interesting.
fayland_ jabbot: it runs wrong on my OS
jabbot fayland_: Your OS?
fayland_ jabbot: yes. can u test it? 10:50
jabbot fayland_: Why don't you tell me a little more about this.
obra jabot++ # passes the turing test
gugod be proud 10:51
obra grins at gugod
fayland_ jabbot: did u test it? 10:52
jabbot fayland_: That is interesting. Please continue.
gugod fayland_: jabbot is a real robot, fyi
fayland_ o
god save me
jabbot pugs - 1307 - * cleanup on AUTHORS and ChangeLog 11:17
pugs - 1306 - * support for variables in ~~ RHS
pugs - 1305 - fix a small test file from me.
pugs - 1308 - something new about POD Data object to P 11:27
Jonathan Hmmm....is build broken (on Win32) for anyone else? 11:28
Fails for me at Spec.pm line 1, saying that Perl 6 is required. 11:29
hlafarge hi everybody - i'm getting an 'impossible' error from GHC when compiling Main.hs on debian (GHC 6.4), any clue ? 11:47
pjcj works fine for me - debian testing, self compiled ghc-6.4 11:49
autrijus hlafarge: try check out again 11:51
Jonathan: oh? hrm
hlafarge pjcj: the exact error message is: ghc-6.4: panic! (the `impossible' happened, GHC version 6.4): \nMaybe.fromJust: Nothing
autrijus Jonathan: check for your perl5 File::Spec
perl -MFile::Spec
see if it's overwritten
hlafarge: yeah, I ran into that yesterday
a fresh checkout worked
hlafarge autrijus: you mean get rid of my old checkout and get a new one ?
autrijus hlafarge: just
rm -rf * ; svn up
or "svk revert -R ." 11:52
hlafarge autrijus: thanks, trying that right now ...
autrijus: wonderful, it works, thanks! 11:58
autrijus :)
nothingmuch is having trouble conceptualizing 11:59
without increasing usage complexity
i hate this dilema
hlafarge autrijus: btw, is the mirror at svn.perl.org/perl6/pugs still supposed to be kept up to date ? 12:01
autrijus: i'm mirroring that via svk and got no updates for a few days
autrijus er, the syncer dies, sorry. 12:02
restarted.
nothingmuch hates sysadmins who think their job is more important than the jobs of the users they support 12:04
hlafarge autrijus: made my day ;-) you're my hero (already were anyway) autrijus++
autrijus :) 12:05
nothingmuch hlafarge: you can just use svn.openfoundry.org/pugs
autrijus nothingmuch: sysadmins usually hate back :)
nothingmuch hates even more so, hacking into the US sites boxes because the sysadmin is sleeping and I need something done
i don't hate him
i hate his process
he's soooo stubborn
there's a firewall zone for nearly every machine over there 12:06
hlafarge nothingmuch: that's what i'm using with subversion, but i also had an svk mirror, just in case ;-)
nothingmuch so each time someone wants to copy a file i have to say 'ah, you have to pull, you can't push because Aaron is too paranoid'
autrijus bbiab :)
nothingmuch the reason i'm in ruining his security policy in the first place is that he placed some stupid cap on the mail message sizes
so i couldn't send 30MB by email
jabbot pugs - 1309 - * rename IMCC to IMC 12:07
pjcj How do people mail word docs and excel files containing screen captures?! 12:08
obra heh
nothingmuch pjcj: are you asking how in the techincal sense, or the "dear god" sense? 12:09
pjcj the latter, unfortunately
nothingmuch oh crap 12:10
now i have to play with his LVM settings
osfameron Annoyingly, it seems to be the easiest way to mail a screen capture from windows
nothingmuch can't you just paste it into outlook/thunderbird? 12:11
osfameron Don't think you can paste any attachments into thunderbird. 12:12
Can't remember re outlook, think no though.
nothingmuch oh well
nothingmuch uses mutt or apple mail
i expect that behavior from the latter
and normally enjoy the lack of it in the former 12:13
osfameron www.joelonsoftware.com/items/2005/03/28.html has a comment on user-friendliness of print-screen 12:14
shapr This place is growing fast. 12:30
autrijus verily. 12:31
Odin- Ya figure? 12:34
shapr I reckon
Is that close enough?
autrijus :) 12:36
bbiab. 12:37
autrijus needs to write PA02 in a couple days.
Odin- What'll that be about? 12:38
autrijus technical architecture
"how to start hacking pugs"
Odin- Ah. :p
autrijus explaining key concepts etc. 12:39
&
jabbot pugs - 1310 - * Add a couple of common keywords 12:47
rgs continued to hack perl6.vim
Jonathan Seems that the build process is choosing to use the copy of File::Spec in the Pugs blib folder rather than the one installed with Perl itself. 12:51
Perl 5, that is. 12:52
rgs that's probably the Right Thing. 12:53
:q 12:54
unless you're speaking of the p6 port of File::Spec of course
Jonathan Not really; that version of File::Spec requires Perl 6 which, until you've built pugs, you don't have.
Yeah, that's right.
rgs but it should go in blib6 then
rgs doesn't see a File::Spec in blib* 12:56
Jonathan I have the latest version, have done a make clean, run make and end up with a module starting "use v6;" in blib\lib\File. 12:58
rgs \ ? you're on windows ?
Jonathan Yes.
rgs maybe a platform specific bug then. trying to reproduce. 12:59
kungfuftr the IMC renaming seems to have broken building 13:00
kungfuftr has a peek 13:01
rgs that's right 13:02
Jonathan It does "cp lib/File/Spec.pm F:\Hacking\Pugs\blib\lib\File/Spec.pm"
rgs here (on linux) it goes to blib6/lib/File/Spec.pm
which is right.
kungfuftr anyone know when stevan usually gets onto irc? 13:03
nothingmuch his afternoon to evening
kungfuftr TZ?
shapr stupid question, can I create new operators with limited semantics in Perl? Is there some way to generate always correct SQL from made-up database operators in Perl? 13:05
nothingmuch shapr: you could use the given operators, and overload them, in p5
or declare sub infix:<xxx> in p6 13:06
see s13
shapr and then it's an infix op?
pasteling "Jonathan" at 82.133.67.25 pasted "My build problem" (62 lines, 3.4K) at sial.org/pbot/8714
nothingmuch uhuh
shapr nifty, thanks.
Jonathan That's what I see on build.
nothingmuch you can also to declare precedence
metaperl I pointed shapr to Rosetta 13:07
and all the SQL generators such as Class::DBI, Rose::DB, etc
nothingmuch Jonathan: looks like the manifest for File::Spec is broken
can you paste whatever was before that?
Jonathan I just have the running of Makefile.PL 13:08
nothingmuch i'd like to see it all... 'perl Makefile.PL; make'
Jonathan I do get "Warning: prerequisite ExtUtils::MakeMaker 6.15 not found. We have 6.03."
nothingmuch ah
kungfuftr hhhmmm... distclean and make seems to have worked
Jonathan Is that going to be it?
nothingmuch that's one good reason for it to fail 13:09
Jonathan Argh.
nothingmuch 6.03 is from before i was born, i think
Jonathan ActiveState's ppm happily doesn't have anything later than 6.03. :S
nothingmuch yup
perlwise i was born around 2002
shortly before 6.05 13:10
;-)
kungfuftr Jonathan: should be able to install directly from cpan instead of ppm
nothingmuch booyakasha 13:11
Jonathan Righty, that upgrade rids me of the warning. 13:18
pasteling "Jonathan" at 82.133.67.25 pasted "Now I get this..." (21 lines, 717B) at sial.org/pbot/8717 13:19
hcchien should I add the test for multidimensional array? or it exists already? # didn't see it. 13:27
kungfuftr hcchien: t/data_types/multi_dimensional_array.t ? 13:31
Thomas_Yeh hi Autrijus 13:32
mugwump he's sleeping 13:34
Thomas_Yeh mugwump: can you tell him to reply mail to danielnee of s-team, thanks 13:36
Limbic_Region Does anyone know if p6 is supposed to allow this? www.perlmonks.org/index.pl?node_id=443062 13:38
pdcawley used to do the weekly p6 summaries so I would have guessed he would know what he is talking about
though in a follow up he didn't sound so sure
hcchien kungfuftr: oh, thanks. but it doesn't include the data type declare, like my num @nums is shape(2,4)? 13:41
kungfuftr hcchien: add to that test then i'd reckons 13:47
hcchien kungfuftr:ok, I will do. 13:48
13:54 cnhackTNT|away is now known as cnhackTNT
jabbot pugs - 1311 - Require a version of Test::More that kno 13:57
chip kungfuftr: seeking me? 13:58
Limbic_Region was looking for a snack, but he really isn't fond of chips 14:03
shapr Too bad, I have several chips on my shoulder.
kungfuftr chip: yar, 2 secs 14:04
Limbic_Region I'm glad to see you are using a use.perl journal chip - will be nice to see how all this Pugs development spills over into Parrot - thanks for putting on the hat
chip Limbic_Region: thanks ... I'm pondering my own sanity, but I couldn't resist
Limbic_Region getting Leo involved in IRC will be a huge boost I think 14:05
chip My first issue is figuring out a basic architectural question of big register set vs. continuations. It's ... fun
Limbic_Region the man is phenominal, but he works vaccuously too much IMO
shapr cheers for continuations
kungfuftr chip: my housemate has been looking for stuff to hack on in parrot, are there certain areas he should look towards? 14:07
jabbot pugs - 1312 - More MetaClass musings...
Limbic_Region I know Sam Ruby was getting just a tad bit frustrated and without our fearless leader (Dan was MIA at the time) no resolution was made
Limbic_Region volunteered to forward links to autrijus for any Pugs related questions that arose at the Monastery/use.perl - I can do the same for Parrot if you want chip 14:09
chip kungfuftr: I'm not the guy to ask that question. Leo is (among his other other roles) the traffic cop, and he's been occasionally posting to p6i with tasks in search of volunteers 14:13
Limbic_Region: Yes, please, that'd be fine; I like much of the Monastery, but there are only so many minutes in a day 14:15
Limbic_Region ok - you prefer email or /msg here in IRC? 14:19
chip - to give you an idea of the volume, there have been 4 specific related posts at the Monastery in the last month and a half 14:22
interestingly enough, one of them is titled "Parrot on a chip" 14:23
chip bwa 14:26
bwaha
msg on irc is fine ... my screen session sees all 14:27
thanks
Limbic_Region shame on you chip - your away message is triple x-rated 14:29
kungfuftr chip: thanks 14:30
Corion pugscode.org <<Overview Journal>> | pugs.kwiki.org | logged: xrl.us/e98m | FBSD_4.11:r1302(183/3383) win2k:r1312(185/3383) Linux:r1302(193/3383) MacOSX:r1197(105/3116) 15:05
elmex hi 15:13
Limbic_Region lo
Corion la 15:21
ter
15:50 PerlJam is now known as PerlPilot
nothingmuch if anyone wants to see reference impls of ref counting and copy sweeping garbage collection: darcs get nothingmuch.woobling.org/gc 15:54
i'm working on zoned GC
and then I will script usage of a memory structure
and then i will benchmark
16:07 PerlPilot is now known as PerlJam
theorbtwo Does anybody know if the hackathon around YAPC::NA is going to be before or after? 16:18
Apparently not, sigh. 16:21
gaal_ to2, i recall autrijus said before, but search these logs to make sure. 16:46
Corion Did anyone actually hack on %*CONFIG ? I only see mention of it by integral, gaal and me, and nobody mentioned it's been done... 17:10
(I'd "simply" expose the FMHash (or whatever it is) in Config.hs to the Perl side of life)
... and all of that just to get -V:foo_bar to work :) 17:11
jabbot pugs - 1313 - TODO test for %*CONFIG hash (tentative) 17:37
integral Corion: I've got half the code done 17:44
Also autrijus suggested that it should be %?CONFIG
the issue I'm working on is that the FiniteMap in Config.hs uses Haskell rather than perl types
Corion integral: Oooo - even cooler! I'm currently wondering about why %*ENV works and %*CONFIG doesn't compile (Haskell side)
integral: Aaah - maybe that's my problem too :-) Well - I'll change my tests to use %?CONFIG then :-) 17:45
integral there's no actual patches on svn yet
Corion integral: ... except my patches - I accidentially also committed my Run.hs which has a stub hash for %*CONFIG :-) 17:46
integral ah
Corion integral: But filling that hash with life is what doesn't work for me currently :)
integral I'll send my patch to the Config.hs stuff 17:47
Corion I'm undoing my Run.hs patch (resp. comment it out)
jabbot pugs - 1316 - TODO test for pugs -V:foo 17:57
pugs - 1315 - Add pugs_versnum, pugs_version, and pugs
pugs - 1314 - Rename %*CONFIG to %?CONFIG
pugs - 1317 - Added tests for unpacking array paramete 18:17
18:24 Corion_ is now known as Corion
integral *sigh* ghc eats so much ram :-/ 18:25
Corion Heh - I think I tricked Haskell into exporting %?CONFIG :-) 18:43
Heh - I think I tricked Haskell into exporting %?CONFIG :-) 18:48
Ooops - sorry :)
hi nothingmuch :) 18:49
nothingmuch evening 18:50
hola Corion
Corion integral: ping ?
shapr sings cheerfully 18:57
I'm so happy there are inspirational projects in the world like pugs!
jabbot pugs - 1318 - First draft of a 'working' %?CONFIG
shapr see! jabbot thinks so too!
integral Corion: pong 18:59
Corion integral: I just fudged in my version of %?CONFIG (but I think I broke something) - are you still working on %?CONFIG ?
integral yep, I'm still fiddling 19:00
Corion integral: I hope your solution will be better than my fudging :)
integral it's probably not :-)
integral looks
Corion integral: I use let confFM = listToFM $ [ (k, VStr v) | (k, v) <- fmToList config ]
integral ah, that's what I'm using but with a map rather than a comp. 19:01
Corion ... but I didn't test before committing, so I broke the build :(
Gaah - and another stupid thing committed before testing it :(
... I should set me up a local repository and only commit when I've tested it all locally :( 19:02
integral ./pugs -e'say "This is the %?CONFIG<pugs_version>"' # now works 19:04
jabbot pugs - 1320 - Fixed all my stupid not-test-before-comm 19:07
pugs - 1319 - I should test before commit :(
Corion integral: Heh - I'm just hacking in -V:pugs_version :) 19:08
integral: What implementation did you use / is it similar to what I did ? 19:09
integral It's exactly what you did
Corion Yay. pugs -V:foo now works, at least for known foo :-)) 19:10
integral I just used map with (\(k,v)->(k,vStr v))
Corion Ooops - except that I get path_sep: path_sep - which means I left out a FM lookup somewhere :-(
gaal hey integral, Corion 19:12
Corion integral: I went that way because the line above looks exactly the same :)
gaal!
integral hmm, should revision be "r123" or just "123"?
nothingmuch hola gaal
gaal sup? hey, cool, %CONFIG :)
hey there
nothingmuch integral: IMHO just 123
integral hacks
nothingmuch that's easier to pass around to svn
Corion gaal: Close... My -V:foo stuff is broken, but otherwise it seems to work ;)
gaal svn ups 19:13
Corion d'oh - reversing the parameters to lookupWithDefaultFM is stupid :) 19:14
gaal wow, we have lots of tests
Corion, you know about the accessor, yes?
Config.getConfig
Corion gaal: That one doesn't help me.
gaal because it defaults to ""
oh
too bad :)
nothingmuch is there a safe/efficient way to climb up two branches of a tree 19:15
and find their parent?
Corion gaal: It only looks up one value. And I want to use the same variable that is used for %?CONFIG :)
gaal wow, we have hella lots of tests.
nothingmuch lowest common parent
gaal does each node know its depth from the root, nm?
nothingmuch no 19:16
gaal then i don't know :)
nothingmuch otherwise i'd just level, and then climb a step at a time
gaal true.
i only have answers to the easy questions. that's why i want to make everythign easy.
nothingmuch i'm guessing I could climb all the way up, stack parents
integral I remember seeing something like that in the type code, iirc it just turned each node into it's path (top -> grandparent -> parent -> node), and pulled out the common prefix
nothingmuch and then pop, till there's a split 19:17
Corion Ah - a Perl6 question - should @array = ([1], [2,3]) work like in Perl5 or should it flatten ?
nothingmuch the reason I want to avoid that is that i"m reeeaaaalllly afraid of latency
gaal oh, no doubt there's a way to get the haskell or ocaml compilers to produce an error message that expresses a solution to this problem :)
Corion (because Pugs flattens, and I'm not sure if that's a bug :) ) 19:18
nothingmuch i think that's a bug
gaal i don't know, Corion, but I'd guess like p5.
nothingmuch but it could be that what you really mean is '@array = (1; 2, 3)
gaal what do the tests say?
Corion nothingmuch: Would that create an array of two arrays ?
nothingmuch yup
in theory 19:19
i'm not sure it's implemented
Corion nothingmuch: Then I think I want that :)
nothingmuch i think i want @array = ([1], [2, 3])
because otherwise there's no real way to say @array = ($aref, $aref2); 19:20
kungfuftr autrijus: yullo?
nothingmuch considers balanced zoning, a la btrees
makes things simpler
but also more complicated 19:21
Corion What should become the equivalent to perl -V:unknown_option ? In Perl5 it is 'UNKNOWN'.
nothingmuch i think unknown_option=undef; 19:22
that means false in p5 %CONFIG
gaal why not raise an exception on that kind of thing?
nothingmuch but i don't think that's a good thing
perl -Mfatal -V:unknown_option 19:23
Corion nothingmuch: It must return a string to the command line (I'm talking about -V: , not about the internal value, which will be undef IMO) 19:24
gaal if i'm doing low level io, and have a binary buffer, how do i substr off it w/o unicode semantics? i know there was some debate but has there been a bottom line? 19:25
in that case, UNKNOWN makes better sense than an empty string
nothingmuch sh compatible?
i think you want 'my str $buffer'
gaal though stderr or some nonzero exit code might be better 19:26
thanks, nm
jabbot pugs - 1322 - -V:foo works for known values of foo 19:27
pugs - 1321 - Change %?CONFIG{pugs_revision} to be jus
gaal can i cast a probably unicode string onto that buffer?
my str $nonuni = "this contains unicode chars"?
Corion Currently the output will be "\tfoo: \n" - ugly but "good enough" so far 19:29
gaal that's like giving an empty string, Corion, if for other values you return "\tMoose: elk\n"
Corion gaal: Yep. Not ideal, but good enough for a first try I think 19:30
gaal sure, but i'd still prefer UNKNOWN :) 19:31
Corion gaal: No problem - I'll put that in and it'll go in with the next commit :) 19:33
That way, it'll stick out in generated shell scripts or makefiles, true!
gaal and there'll be a way to differentiate it between values that really are "". 19:34
Corion Hmmm. As a side effect, "pugs -V foo" works too. Is that good or bad ? 19:36
jabbot pugs - 1323 - Unknown -V: options output the empty str 19:37
gaal my Code $callback = smth. how do i invoke $callback? $callback.(args)? $callback.call(args)? 19:45
kungfuftr gaal: $callback() ?
gaal i don't know, couldn' 19:46
t find it in the Syns.
kungfuftr callback is just an anonymous closure, aye?
gaal yes
kungfuftr yeah... it's $callback()
gaal $x.() makes sense to me, because in p5 i was a fan of $x->() 19:47
thanks, kungfuftr!
kungfuftr gaal: hhhmmm.. it's not specific, but see about currying in S06 19:50
gaal: ah ha! it's in S04 within "Relationship of blocks and declarations 19:52
gaal thanks again :) 19:53
kungfuftr np 19:54
kungfuftr has a crazy idea for pugs/perl6 in general 20:01
gaal we're listening :) 20:03
kungfuftr soz, coffee called
translating "Learning Perl" directly into the perl 6 equivalent (ie: Learning Perl 6) 20:04
gaal we certainly want docs and books, but you know, learning perl isn't in the public domain :) 20:05
kungfuftr this is true, but oreilly might be up for it, especially since it's a format they already know and somewhat guarantees they'd be the first publisher to have a book 20:06
gaal what about the author?
kungfuftr i believe that the publisher gets nist if the rights 20:07
gaal i mean, we can certainly start writing docs and tutorials
manpages
and you're free to write a book if you like
kungfuftr heh
Corion pugscode.org <<Overview Journal>> | pugs.kwiki.org | logged: xrl.us/e98m | FBSD_4.11:r1302(183/3383) win2k:r1324(187/3474) Linux:r1302(193/3383) MacOSX:r1197(105/3116) 20:24
Corion Hmmm. OSX is falling behind :) 20:24
Should I promote the tests in unspecced/config.t from todo_is() to is() ?
gaal if they pass :) 20:25
Corion gaal: They pass, but it's uncharted country :) 20:26
gaal that's why it's in unspecced... 20:27
jabbot pugs - 1324 - -V: 'works now'
Corion gaal: OK - I'll promote them then :)
Corion is amazed how far he gets with cargo-culting stuff 20:29
... and I think I just had an idea how I can make the config parser work better and still keep the pattern matching of Haskell... 20:31
gaal yay!
Corion gaal: I think I'll first expand all clustering options (-dwl0e) and then regroup the stuff into nested arrays again, so I have [ [ "-e", fragment ], [ "-e", fragment2 ], ... ] which I can then sort with ease. 20:33
This will also make the -V: parsing less convoluted.
And all of this, just to get to -p and -n ...
gaal Corion pwns 20:34
Corion gaal: No - I'm just efficient at spotting stuff I copy-and-paste and where I then change a symbol to the symbol I need :)
theorbtwo BTW, Corion: -I path/name/ does wierd things, last I looked. 20:35
Corion theorbtwo: It's -I/path/name (no space)
theorbtwo: Err -Ipath/name (still no space)
theorbtwo: But a test case would be welcome ...
theorbtwo: Aaah - I think I should handle -I (nothing) as an error, then 20:36
Oh. Perl5 allows -I path/to/dir . So I should better allow that too. Feh. I hoped I could delay the move to my new schema. :( 20:37
First, let's add a (failing) test case :)
gaal tdd++
say, does @_ include the invocant in method calls?
(i suppose so...) 20:38
jabbot pugs - 1326 - -I path/to/dir does not work 20:57
pugs - 1325 - Promoted t/unspecced/config.t from todo_
kungfuftr gaal: =0)
Corion Gah. These "Couldn't match `Char' against `String'" will be the death of me ! 21:02
gaal [[[[death]]]]
Corion gaal: Indeed :)
But Haskell poetry has nothing on Perl poetry :)
gaal yeah! 21:03
CATCH being inside the try takes getting used to. 21:05
Corion Hah! Adding a judicious :[] makes Haskell understand what I mean :)
gaal: CATCH is inside try{} ? Ugh! 21:06
gaal it's consistent. just not with other languages. :)
Corion gaal: Its pattern matching takes some getting used to, yes - the last parameter always is a list. If you don't want that, specify a list after it :) 21:07
gaal :) 21:09
in p5 /x-modified regexps, what's the idiomatic way to say "one space"? [ ] ? '\ ' ? 21:11
e.g. /this is an example of an unhappy regex/x 21:13
Corion gaal: \s if you don't care, [ ] otherwise, I think.
gaal i'll definitely use m:w in this particular p6 rx, then. 21:14
:)
does \0 mean NUL in a regexp like it did in p5? 21:37
nothingm1ch nothingmuch.woobling.org/pugs_test_status is now updated automatically 21:52
gaal nothinglikegoodolnicksmuch++ 21:53
nothingm1ch is on the other comp =).
does vim know :TOhtml from the command line? 21:54
preferably to STDOUT?
gaal yes, let me remember how 21:55
pass it the -c arg
don't remember if they chain. if so, vim -c :tohtml -c :q file 21:56
if not, you'll need a function, i guess
when i declare a method, is the first positional param always the invocant? or can i omit it if i don't nead it? 21:57
nothingm1ch i think if it's omitted it goes somewhere special
i'm not sure where. 21:58
gaal well, there's the debate over .meth and $_.meth and o.meth etc,
PerlJam gaal: If you don't put a colon, then all of the required parameters are invocants.
gaal er
no grok
nothingm1ch to get what you asked for: sub foo (: $x, $y, $z) 21:59
invocant chucked
without the colon, $x, $y, $z are invocants
right, PerlJam?
gaal hmm
that's not mentioned in t/Synopsis 22:00
i think
if i call my code like so somewhere in my class: $self.othermeth(1, 2) 22:01
and my othermeth is declared 'method othermeth($x, $y) { ... }
won't that silenty work?
PerlJam nothingm1ch: um ... no, and yes.
method foo ($a,$b,$c) { ... } All of $a, $b, and $c are invocants. 22:02
method foo ($a: $b,$c) { ... } Only $a is an invocant
or you can think of it as in the first case all of parameters participate in MMD determination, whereas in the second case, only the $a does. 22:03
gaal PerlJam: and if i'm not doing mmd, and dont' need a 'this' in the method, what syntax would you recommend?
PerlJam gaal: method foo ($a,$b,$c) { ... } 22:04
gaal good :) 22:05
nothingm1ch gaal: ideas for yaml harness: 22:06
it appends a whole YAML doc per run 22:07
to tests.yml
with only the revision
then you can run several tests at a time on SMP and you don't care
xargs -P
then we have pugs_info
which generates the big platform/r/-V yml 22:08
and smoke_report, which just posts the data to a webserver
the webserver reads the yml, and merges it
consolidation of results:
disagreement causes yellow 22:09
and onmouse over you see all reports
as platform: ok
.
if there is agreement, then there is agreement and we don't care
testgraph.pl will take in a slightly more complicated yml, of merged/sorted results 22:10
and someone will take care of generating that yml for it,
and that's it
good enough?
gaal er, what manages contention against tests.yml? 22:12
nothingm1ch some merge and sort script
either taking data in from sqlite, or from the ymls themselves 22:13
gaal no, on the client side
nothingm1ch ah
uhm
gaal you can't just xargs -P and have everyone write to the same place
nothingm1ch yes you can, you write for a short time
gaal bleh :)
nothingm1ch YAML probably has a lock mode
so it doesn't matter if locks block for a short while 22:14
gaal well, okay.
(sorry i'm slow, i'm also just ciing something)
nothingm1ch hmm, it doesn't
well, maybe ingy is nice enough to add that ;-)
gaal i was just about to suggest that 22:15
nothingm1ch either way, you can just flock
gaal sure
ok
nothingm1ch do what parts do you want to cover?
what parts do you want to leave to me?
and what should we spill on stevan?
gaal i can do everything on the client side no problem
nothingm1ch including http post? 22:16
gaal it's a fair organization
nothingm1ch we just have to agree on a URL
gaal sure, why not
nothingm1ch ok
pugs.woobling.org/smoke
it just uploads tests.yml, with normal cgi file upload wizardry
gaal yeah, that's easy enough. 22:17
Limbic_Region did autrijus forget to journal today?
gaal i think the man deserves some sleep :)
you know, once a month
they say it's healthy
how do i un-svn-add a file? 22:18
nothingm1ch svn revert?
gaal (i didn't ci it yet)
nothingm1ch exactly
svn revert tosses on ci'd changes
i'm guessing
gaal yes, that workd. thanks :)
worked 22:19
i need sleep :)
Limbic_Region wasn't complaining - just inquiring
actually it was a concern for his lack of sleep that prompted the question -
gaal oh, i was kidding, mean :)
man
i can't type
Limbic_Region as in - is he still up and just hasn't yet 22:20
gaal esp when i'm asleep
Limbic_Region no worries
gaal ++ports; please review if you can. i'm off to bed 22:22
nothingm1ch ciao
kungfuftr gaal: merlyn++ 22:23
gaal kff: ? merlyn++ indeed but why do you mention it?
kungfuftr gaal: llama book stuff 22:24
gaal oh yes 22:25
i didn't learn from it, but i taught with it :)
kungfuftr gaal: depending on confirmation with publisher... a companion guide fer perl6 is well possible... =0) 22:26
gaal i'm sure, but it sounds to me like it's merlyn's job to write merlyn's books (actually, tom phoenix coauthored ed. 3 iirc) 22:27
anyway, i 22:28
'm outtahere. night!
kungfuftr yar, but that won't happen for a long while
gaal &
theorb G'night, gaal.
nothingm1ch ciao gaal 22:30
nothingm1ch is not sure whether to leave the test reporting on his osx or linux box for now 22:31
22:32 theorb is now known as theorbtwo
theorbtwo Why not both? 22:33
(BTW, discovered the obvious the other day: diffing two test run yaml files produces rather readable results.) 22:34
nothingm1ch theorbtwo: because it isn't consolidated yet 22:35
human readable data is very good
lots of tools made to eat it over the years
diff, less, wc, grep, awk, sed
theorbtwo Tools for eating human-readable data tend to be named with an initial capital... 22:36
nothingm1ch ?
theorbtwo James, Bob, Frank, Jess, Yuval, Autrijus...
nothingm1ch ah
heh 22:37
jabbot pugs - 1328 - Comment out STDERR prints
pugs - 1327 - Perl6 port of Cache::Memcached
nothingm1ch theorbtwo: think you can handle pod->html escaping madness, and folding of <code> in the synopses? 22:38
crysflame memcache perl6?!
crysflame cheers 22:39
nothingm1ch L<foo/"blah C<for> ding"> does not resolve
folding of <code> should also be done for reexes
but it's a bit tricky
theorbtwo Nice use of the scalar ... operator. 22:48
nothingm1ch loves scalar .. and ... 22:51
i also couldn't manage to wrangle Text::VimColors 22:53
i'll try to run :TOhtml myself
since it's smarter
optionally, ofcourse
anyway, good night
jabbot pugs - 1329 - oops, accidentally removed line 22:57
crysflame can jabbot tell us who committed?
Darren_Duncan question: anyone know what the semantic difference is between 'my $foo ::= 1;' and 'my $foo is constant = 1;' -- I'm using the former when declaring constants and wondering if I should use the latter instead
theorbtwo Suggestions for improvements to jabbot should go to gugod. 22:58
crysflame ok 22:59
Darren_Duncan the reason I used '::=' for constants was because '::=' did compile-time binding and so thought that if the bound value was a literal, then the container on the left was effectively a constant declaration 23:05
If they do mean the same thing, which one is easier to understand, the shorter or longer one 23:06
all ?
theorbtwo I'd say the longer one is easier to understand.
Darren_Duncan thanks for one answer 23:07
also, many of my constants are hashes of hashes of hashes etc ...
would declaring the root 'is constant' mean that all sub-elements can't be modified too? this is the desired effect
theorbtwo I'm not sure.
Darren_Duncan in any event, I'll convert to 'is constant' for readability 23:08
(at the time I used '::=', I forgot that 'is constant' existed)
kungfuftr you can also `constant $PI = 3.142;` 23:13
iirc 23:14
Darren_Duncan thanks kungfuftr, thats even better
right now I'm looking for the synopsis that talks about this stuff ... started in S02
kungfuftr you should read Perl 6 and Parrot essentials at some point... slightly neater format than the synopses
Darren_Duncan I thought those were out of date 23:15
kungfuftr yeah, bulk of it is correct though
Darren_Duncan in any event, while I want one for each year, I only have the first edition, currently
kungfuftr but if you think about it... `my Int $foo = blah;` so constant $foo seems reasonable to have a variable of 'constant' datatype 23:16
Darren_Duncan now looking in A12
kungfuftr anyways
Darren_Duncan I mean S12
theorbtwo tunes out for a bit. 23:21
jabbot pugs - 1330 - Reorder some text 23:27
Darren_Duncan okay, change committed ... all '::=' changed to 'is constant =' in my modules ... seems every other module already does 'is constant' 23:33
jabbot pugs - 1331 - constants declarations in LKT and SRT al 23:37

Powered by Cro and the Raku® Programming Language.

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

Thank you!