pugscode.org «Overview Journal Logs» | r1773/winxp: 587/4256 smoke/win2k (occasional): xrl.us/fqum (91/4373, r2131) Mac OS X (62/4459, r2116) | pugs.kwiki.org
Set by Corion on 19 April 2005.
machack666 hey all 02:08
mugwump hoesit! 02:09
machack666 good, good... 02:10
what's the status on 6.2.1? 02:11
is a junction a formal datatype, and if so, what it its class? 02:15
mugwump but a junction is a superposition of objects!
:)
machack666 we were discussing references of junctions earlier 02:16
mugwump The P6FC still needs building, seen docs/src/ in pugs?
see also src/Pugs/Context.hs (search for "Junction") 02:18
machack666 thx
mugwump The P6FC may be built on that, however note that it is an inheritance tree where it should possibly be a role graph 02:19
machack666 I assume it's docs/src/Junc.pod I'm to be reading? (What, pray tell is P6FC?)
mugwump Well, the "FC" means foundation classes
machack666 gotcha
I'm still cutting my teeth on haskell, but I'll see what I can glean
mugwump just empty your mind and look at the code ... ;) 02:20
you have to think like a maths/compsci geek, like reading -> as "implies" or "derives"
machack666 do you know where autrijus is on creating the internals overview document he was talking about? 02:21
machack666 was a maths geek
mugwump I haven't asked tbh
machack666 some of it is just getting the overall picture of how everything works internally
I don't have any familiarity with p5 internals, other than a basic knowledge of SV, AV, etc. 02:22
not that it is anywhere near the same thing.
mugwump Hopefully autrijus' summary of the internals will very very enlightening. I've got some overall stuff, and a bit of specific knowledge but it hasn't really "meshed" yet :) 02:23
machack666 my guess is that if we are really just spitting out an AST and executing it, then things like junctions are expanded as part of the tree generation process. I'm not sure what primitives are supported at the AST level, but it seems like they would all have to be operators/types supported in parrot. 02:26
machack666 looks at what he just wrote, and wonders if it means anything.
src/Pugs/Context.hs is indeed enlightening 02:28
mugwump that, Prim.hs and Eval.hs are an important trio 02:31
machack666 so what is Prim?
mugwump defines language primitives (functions, etc).
machack666 looks like the referencing of junctions issue has been fixed 02:32
I will still see if a test exists
mugwump issues seem to do that in this tree, it's scary
machack666 is isa() the preferred method of testing reference types?
mugwump see t/builtins/hashes/isa.t, I guess so 02:34
machack666 isa doesn't work with junctions. it'll have to be ref($j) eq 'Junction' or $j.ref 02:52
it tries to autothread the function call in the isa_ok case
pesky little varmits
ah...
when I thought it was fixed earlier, I was mistaken
I was running r1926, which had been installed on the system
there ref $j was 'Junction'
the current build still has problems
checked in 3 tests for junctions' reference types
anyone here who understands Prim.hs? 02:56
I'm wondering what the numeric part of the /op\d/ operator is. It looks like the number of terms it swallows up 02:59
I'm going to go off of that assumption
since everyone else looks to be asleep
jabbot pugs - 2136 - tests for references to junctions 03:01
pugs - 2137 - Typo fix. 03:11
QMario Hello everyone! :) 03:34
Is everyone sleeping? 03:36
\who 03:41
print "/who"
crysflame hi 04:30
yeah, autrijus should be awake soonish
obra 05:13
wolverian interesting, obra 05:24
jabbot pugs - 2138 - clean up a bit 05:31
gaal morning 05:43
Corion r2138 - datenzoo.de/pugs/win2k.html - early morning smoke ;) 05:45
Hi Ho ! Hi Ho ! It's off to $work I go. 05:52
06:10 {s}AINT is now known as chady
Alias_ autrijus: ping? 06:13
autrijus: In case you see this later, you REALLY need to address some of those problems in Module::Install. I just found out it's incompatible with 5.005. eep 06:25
mugwump imagines autrijus saying "yeah yeah, patches welcome" 06:30
Alias_ Pity the guts are pretty opaque
:)
nothingmuch morning 06:47
masak morning nothingmuch 06:51
nothingmuch hola masak
machack666 hey nm 06:52
nothingmuch hola also machack666
#perl6.members».greet
machack666 what is the syntax for nesting arrays? shouldn't this work: my @a=(1,2,[3,4])? 06:54
ie, @a has 3 elems?
pugs insists on flattening this for me 06:55
my @b=(3,4); 06:56
my @a=(1,2,\@b);
this gives the expected nesting of the array in the third slot 06:57
but the direct nesting syntax does not seem to be supported.
castaway_ thought module::install was ingys baby?
morning folks btw ,)
06:57 castaway_ is now known as castaway
machack666 refers to S09 06:58
ok, it looks like in order to get a nested array, I'm having to do this: 07:05
(1,2,\[3,4])
which, unless the nesting rules of perl have changed (and I don't remember that being the case) this would be a bug 07:06
castaway Ugh 07:08
looks like a bug to me
machack666 it's flattening (1,2,[3,4]) by itself. 07:09
castaway are there no tests for it? 07:10
machack666 i'm not seeing any in t/builtins
getting ready to add some
castaway :)
machack666 is (1,2,(3,4)) supposed to flatter in P6 07:15
castaway dont look at me :) whats it do in perl? 07:16
I'd guess yes
machack666 it flattens
but we treat arrays and array references a lot more interchangably in p6 07:17
castaway (3,4) is still just a list, no?
machack666 I suppose so
8 tests should do it for now 07:25
castaway :) 07:27
nothingmuch_ gugod: ping 07:29
07:29 nothingmuch_ is now known as nothingmuch
machack666 (4 of which fail !!) 07:29
castaway oops :)
jabbot pugs - 2139 - Tests for nesting of arrays. 07:31
pugs - 2140 - Renaming arrays/nesting.t to arrays/nest
pugs - 2141 - * nested.t passes; Syn "," now always im 08:41
pugs - 2142 - * numerous cosmetic source cleanups.
castaway thinks jabbot should report blame too :) 08:43
jabbot pugs - 2143 - * flatten RValues at the right hand of a 09:11
nothingmuch (gugod|autrijus|ingy)->pingf 09:27
castaway :) 09:29
jabbot pugs - 2144 - * remove the unexpected successes from f 09:31
Jonathan_ nothingmuch: Just trying to get smoke testing running. When I run yaml_harness.pl, it is trying to run the test scripts with /usr/bin/pugs, which doesn't work out too well on Win32. Ideas for where I should look to fix this? 09:37
nothingmuch set the env variable HARNESS_PERL=$PWD/pugs
or something like that
don't forget PERL6LIB 09:38
gaaah! 09:40
autrijus Jonathan_: use util/run_smoke.pl 09:41
nothingmuch: sorry, I'm out to dinner now, will be back in 20 mins
nothingmuch autrijus: 30 seconds:
where is s5.html?
autrijus no idea... look at original s5 page?
I need to run... & 09:42
nothingmuch ciao
Jonathan_ Hmmm...smoke-run.pl sets HARNESS_PERL etc, but when I run the tests they still look for /usr/bin/pugs. 09:47
nothingmuch www.ngw.nl/int/dld/s/spork.htm
Jonathan_ Seems that it is looking at the #! line at the start of the scripts.
nothingmuch odd
Jonathan_ Very. 09:48
Wonder if I've got older version of some test harness or something.
nothingmuch installs all of spork and friends from svn.kwiki.org 10:01
Jonathan_ Aha, Test::Harness upgrade did it. 10:10
nothingmuch Jonathan_: yay 10:16
Jonathan_ I'll see what HTML is churns out, then I'll set something up to FTP it up to my site. 10:18
Then I will have to do some real work. Ugh. 10:19
nothingmuch rsync with an ssh keys is probably easier
html is spat out to STDOUT by testgraph.pl
and due to a clunky old interface it also creates a 'util/testgraph.css'
Jonathan_ I'm using the run-smoke.pl that Autrijus suggested, so I think that will collect it nicely into a folder I gave it. 10:20
nothingmuch lets hope 10:23
gaaah
Spork::S5 is still not workin
g
autrijus nothingmuch: you can drop ::S5 and use normal spork instead 10:42
nothingmuch but it's not as pretty
i know
autrijus well there's other spork themes that can be pretty 10:43
I have no idea what s5.html are you talking about though.
nothingmuch templates/s5/s5.html
s/templates/template/;
i ktraced a bit
and it's looking in the wrong places
tried to fix that, and i have no CSS 10:44
gave up
btw, how does kwid fit into this?
afaict Spok.slides is ~~ kwid 10:45
autrijus nothingmuch: Spork::Formatter::Kwid let you write kwid instead of spork
in Spork.slides
nothingmuch ah
autrijus main difference being ``
nothingmuch so they are different
autrijus instead of [=...]
nothingmuch okies
autrijus naps some more. brb... 10:46
nothingmuch ciao
gah! header/footer are missing from slide1a.html 10:47
castaway curses oracle. 10:50
jabbot pugs - 2145 - * `=cut` without a newline at EOF is now 11:01
pugs - 2146 - * Setting `$*CWD` to invalid directory i
pugs - 2147 - * `'key' => val` now works as named para
Alias_ pokes autrijus 11:13
11:13 chady is now known as chady_
castaway nudges theorbtwo 11:20
castaway wonders if its weekend yet 11:22
theorbtwo hugs castaway. 11:23
castaway hugs back. 11:24
Can you close the Opera on insel, please?
Daniel_Nee Hi, Autrijus: Could you check the email I just send to you, Thanks. 11:38
Alias_ So... anyone bored? 11:48
castaway wish I were.. 11:50
nothingmuch i hate computers 11:51
i hate software
i hate system administration
theorbtwo hates not having any money. 11:54
nothingmuch is it so hard to include ssmtp in the worlds most popular distro, for crying out loud?! 11:59
theorbtwo wonders what distro nm is talking about.
nothingmuch fedora
theorbtwo debian++ 12:00
nothingmuch indeed
especially for work, where you need something that works
but at my work place everyone knows redhat if at all
and fedora is the next closest thing
Jonathan_ www.jwcs.net/~jonathan/perl6/smokes/win32/
Should now generate daily, after the binary build. 12:01
theorbtwo nothingmuch, do you know if stephan is interested in more people?
nothingmuch you mean stevan? actually the reason he initially wanted to remain anonymous was that he isn't really 12:02
and autrijus didn't know he didn't want it to be too public
castaway more people for?
theorbtwo Oh. 12:07
Nevermind then.
nothingmuch are you on all those social networks?
i have some freelance acquaintances which took LinkedIn very seriously
theorbtwo I'm on LinkedIn via Gaal, I think, but I haven't taken it too seriously. 12:08
I hate working, and I don't think working remotely will work very well for me.
nothingmuch have you worked since coming to .de? 12:09
theorbtwo But I'm in rather a situation where I need the money.
I did some remote work for my parents for a while, but it didn't work very well.
I'm not very good at working in such a situation, and they weren't very good at working with me. 12:10
I doubt I can find work here, though, as I don't speak enough German to ask for directions to the office, much less get through an interview and any sort of meeting. 12:11
castaway You could do what i did.. (look up US companies) 12:12
theorbtwo Also, my visa explicitly prohibits me from working, meaning that they'd have to want me enough to go to the trouble of getting me a work permit.
castaway, you did that, but did you use German at that first job?
castaway No 12:13
(well, eventually, but not to start with)
kungfuftr notes that sophos are currently looking perl programmers in the UK
castaway yeah, BTDT
kungfuftr as well as vancouver iirc
theorbtwo Moving to the UK would be great, but I don't think I can make a livable wage, and they've already interviewed and not hired Jess. 12:14
(No wonder you're still looking to hire if you get wonderful candidates and don't hire them...) 12:15
kungfuftr theorbtwo: Jess?
castaway raises her hand.
Alias_ Jess?
What's with all these UKians in Germany without jobs?
castaway Mostly due to a stupid misunderstanding (or 2).. IMO 12:16
kungfuftr castaway: which post?
castaway "all" ?
theorbtwo She has a job. I do not, and am American, not Birtish.
Alias_ I'm looking at a UKian in Germany called Jody to fly over here to .au
castaway "Senior Perl Developer" or some such, several months back
maybe I should have tried to rectify it.. 12:17
castaway shrugs
kungfuftr castaway: any idea which department?
theorbtwo AV.
kungfuftr Alias_: ah, knewt
Alias_ yes
castaway Umm, not off hand.. 12:18
Alias_ I have to ask, but did want to do it to knewts face
theorbtwo AV research, I think.
Alias_ Is "Jody" a guy or a girl :)
kungfuftr guy
Alias_ I'm thinking guy?
right
Anyone of you worked with or know him?
kungfuftr nice bloke, met him a few times at london.pm meets
Alias_ I need to convince my partner we should stump up the money, but he's saying that he has only one job on his CV... 12:19
theorbtwo castaway, you should do something pugsy and put yourself in AUTHORS, then everybody will know your name from playing hangman.
Alias_ His CPAN modules look OK, but not large of spectacular
castaway ah, Systems Development (with a Jeremy Smith)
kungfuftr Alias_: all depends what you're after at what sort of price 12:20
castaway in the Virus lab
kungfuftr castaway: ah, they are picky, tbh
theorbtwo kungfuftr: Is another department also looking for perl people?
castaway any there that are less picky? (and arent likely to figure out its the same person? :)
kungfuftr theorbtwo: yeah, my department 12:21
castaway which is? :)
kungfuftr Web
=0)
castaway Hmm.
Alias_ kungfuftr: I need to dump someone onto a TT/large-OO/mod_perl project and 3+ months to help get it out the door 12:22
kungfuftr Alias_: local or remote working?
Alias_ here in Syd
local
I don't need brilliance, just speed and competance
kungfuftr Alias_: independent contractor? 12:23
Alias_ employee...
well... it's a bit flexible
kungfuftr has a ponder
castaway bit far away, that
Alias_ But in order to stay in .au and work for more than 3+ months, the job needs to be salary in order to make the working visa stick
castaway at least you've thought of that 12:24
theorbtwo kungfuftr: She has lots of IT experience, speaks German like a native, is a saint and a pmdev on perlmonks, is CPANed.
castaway kungfuftr: also in Abingdon ?
kungfuftr castaway: yeah, where i am currently sitting
Alias_: there's a few NZ folks might be up for it 12:25
theorbtwo www.sophos.com/companyinfo/careers/...prog.html, I assume?
kungfuftr yup 12:26
castaway: if you're intrested, lemme know. 12:27
kungfuftr & #lunch
Alias_ kungfuftr: names?
Juerd Anyone here read Perl Medic?
kungfuftr Alias_: ask mugwump, he's got a btter idea of locals 12:35
Juerd: yup
Alias_ ok
kungfuftr mugwump: OI, wake up biotch 12:36
Alias_ I wish we had purl
Juerd kungfuftr: What's your opinion? 12:37
To be honest, I had high expectations and was rather disappointed. 12:38
kungfuftr Juerd: it's not bad, but it's a little too low level, I was mainly concerned about a very huge legacy system and disecting it peice by peice, which it didn't really cover
castaway: ullo? 12:39
Juerd It doesn't live up to its promises either
Like writing better tests, writing better documentation
It handles the basics of writing tests and documentation, but doesn't explain how best to
kungfuftr Juerd: i found the business reasons for doing things a lot more intresting that the technical stuff in the book 12:40
Juerd Me too
But that's just one chapter
kungfuftr yar
sophos++ # company safari account
=0)
Juerd And could have been summarized in two or three pages
I for a change bought the dead tree version 12:41
Which I now regret
This isn't a book I will be reading again and again
Or be using as a reference
Juerd has a safari account, even 12:42
Limbic_Region has spent next to no money on books or DVDs in the last year but typically always gets dead tree versions 12:46
Limbic_Region is a bit of a bibliophile
theorbtwo - WRT Jess getting in the Authors file so everyone will know her, diotalevi was recently denied a job because he didn't have enough real world Perl experience - go figure 12:47
theorbtwo Ugh. 12:48
Juerd Whuh?
Alias_ & # home 12:49
Juerd Limbic_Region: Do you know how this level of experience was measured?
Limbic_Region Juerd - the real problem was one of awareness - or lack of it
company x wasn't even aware of CPAN
Juerd Impressive.
castaway Limbic_Region: I wonder how they figured thaht? 12:50
Limbic_Region apparently there is a parallel universe that is completely unaware of ours castaway
castaway Most of mine has been personal and not work related
mebbe he didnt put enough on his CV
Limbic_Region castaway - 99% of mine has been personal and not work related, but I doubt I would have a problem convincing someone of my real world experience 12:51
Juerd I've been denied Jr programmer jobs several times because I'd be overqualified
Limbic_Region castaway - I doubt that
Juerd And I got one job (which I had to deny because of my health) mostly based on my public Perl stuff 12:52
castaway Juerd: yeah, i always wondr whether those are worth applying for..
Juerd castaway: Which those?
castaway Junior jobs 12:53
Juerd Apparently not
They don't want people who are good at what they're doing, because they won't "fit in the team" 12:54
castaway odd
Juerd Heard that three times now.
They were afraid I'd eventually take a lead role.
castaway which makes them sound like they want a bunch of people who have no clue?
Juerd Indeed
I asked the third that question almost literally
Because it was the third in a row, and it really got me angry then 12:55
castaway I can see their concern, but if you're applying, surely its because thats what you want to do, and not because you'Re hoping to advance quickly afterwards
castaway has no aspirations to become a leader/manager or some suxh
what did they say?
Limbic_Region Juerd - I had that problem when I was looking for a part-time job to supplement my income 12:56
Juerd They said that basically, they were indeed looking for people who would just code as was structured by others. In other words: translate diagrams into Perl. The difference between sr and jr programmers was, according to them, that sr got something to say in the design process and jrs were just typists who know Perl.
Limbic_Region I kept saying I don't care how much I make since it isn't my primary vocation
Juerd The sort of thing we usually call code monkeys :)
castaway some people actually like that kinda job..
Juerd Had I known that before, I would never have applied for those jobs, because that kind of work is not what I was looking for.
castaway (otoho depends on the level of the "design process" ,) 12:58
Juerd castaway: And re aspirations: I had none either, but it's apparently in any intelligent programmer's nature to want to influence design decisions. And I think they're right.
Limbic_Region is glad he doesn't program for a living and has sympathy/empathy for you guys
castaway nods at Juerd 13:00
I was assuming they meant "we'll tell you what it does, you figure out how and do it" ..
Juerd I really wanted to have that job in Amsterdam, even though it was a two hour drive away from home. 13:01
Limbic_Region oh, you have got to be kidding me
Yahoo let's me receive an email with a particular attachment, but when I try to forward it to my gmail account it bounces for an illegal attachment
what kinda crap is that 13:02
castaway broken crap? 13:04
Limbic_Region no - it was bounced by their spam filters apparently - but only on outbound
Juerd re 13:09
jabbot pugs - 2148 - * Index construct in slices now defaults 13:21
pugs - 2149 - * mark two other tests as working.
pugs - 2150 - * `**` no longer truncates the exponent 14:01
pugs - 2151 - * use a slurpy read for t/force_todo 14:11
pugs - 2152 - * unmark the one unexpected success
pugs - 2153 - * add the `make smoke` target. 14:31
pugs - 2154 - * use a working form to test map on func 14:51
pugs - 2155 - * slurpy params now applies *-flattening 15:11
pugs - 2156 - * code blocks as subroutine arguments no 15:41
Corion Weird. Something really messes up my source tree to the point that ghc doesn't want to link Pugs anymore. Wiping src/ and checking it out again fixes that. 15:50
(even though "svn status" tells of no differences)
Heh. "pugs: src/Pugs/Prim.hs:981:13-36: Non-exhaustive patterns in lambda" for t\builtins\not.t 15:52
theorbtwo Somebody should make that general. 15:54
Sadly, this is easier said then done.
Corion reads the backlog to figure out theorbtwos context 15:58
theorbtwo The patterns around line 981 of src/Pugs/Prim.hs
Corion theorbtwo: Ah. I didn't look at the source code, just at the output of the smoke 15:59
(black smoke, currently :) )
jabbot pugs - 2157 - * Implement throws_ok 16:01
pugs - 2158 - * eval_haskell errors are now catchable.
theorbtwo D'oh!
nothingmuch eh?
theorbtwo I just made almost the same edit as autrijus just did.
I just didn't check in quite soon enough. 16:02
crysflame heh
nothingmuch stevan: ping
Corion #perl6 groupthink is happening 16:03
theorbtwo pings Autrijus. 16:04
nothingmuch anybody know where Spork-S5Theme is? 16:05
i found everything else, but not that
autrijus pongs. 16:10
theorbtwo Should eval_haskell be catchable or self-catching?
autrijus nothingmuch: it's in /gugod/Spork-S5/lib/Spork/S5Theme.pm 16:11
jabbot pugs - 2159 - Return useful error message.
autrijus theorbtwo: that is an excellent, excellent point :)
Corion I'd assume catchable but not self-catching
nothingmuch ah, on svn.kwiki.org?
autrijus theorbtwo: it should self-catching just like normal eval""
nothingmuch: something like that
theorbtwo OK.
autrijus I think. which means my first stab is wrong
Corion Hmmm. Yeah, making it just like perl_eval is a better idea.
autrijus so please fix it :)
nothingmuch mutters something about releasing modules so that they are accessible before telling people to use them
autrijus I'm going through the remaining failures now 16:12
theorbtwo pings "gcomnz", "mattc", and "tobez".
autrijus ping(all<gcomnz mattc tobez>) 16:16
nothingmuch autrijus: do you remember which plugin class (spork? kwiki? spoon? spiffy? what?) implements the extraction of __some/path/file.txt__?
theorbtwo Allo, nr. 16:17
autrijus nothingmuch: Spoon I think. not too sure
nothingmuch autrijus: thanks 16:18
autrijus np
nothingmuch ah, Installer has a method... a clever one 16:19
ingy++; # readable cod 16:22
theorbtwo How can I put more then one statement in a case foo of ... case without introducing a new level of monadicness? 16:25
nothingmuch gah, lack of docu
autrijus: mind if i bug you a bit about this, to keep going fast?
ninereasons hi theorbtwo
autrijus nothingmuch: I don't know anything about it really :)
theorbtwo: just use another do
nothingmuch oh well
ninereasons you're feeling better I hope, autrijus ? 16:26
autrijus theorbtwo: and omit the "return" from lhs of case
ninereasons: yeah, quite a bit better, I think
nothingmuch for some reason plugin-classes is ignored when extracting files, but not when trying to generate
nothingmuch feels so stupid when trying to figure out large apps 16:30
always the "why didn't I think of that" feeling
autrijus heh. it can't be harder than hacking in .assuming :) 16:31
nothingmuch autrijus: dense code is also frustrating 16:32
oh, I see... sub register isn't being called at all
nothingmuch hopes this fixes it 16:33
i need a new computer. It can't handle more than 30 windows well with expose 16:34
blammo 16:35
theorbtwo Yey, t/unspecced/eval_haskell 3 now passes. Now I have to see why 4 fails. 16:38
jabbot pugs - 2160 - * precedence.t all passes; comma is no l 16:41
pugs - 2161 - * hyperops now extends to the longer, no
Corion r2156 - datenzoo.de/pugs/win2k.html - some more breakage than earlier
ingy hi nothingmuch 16:42
theorbtwo Woo, all four pass, time to ci.
autrijus theorbtwo++ 16:43
ingy nothingmuch: are you fighting with Spoon? 16:45
theorbtwo Hm, somebody should fix pugscode.org's DNS so that it works (and not just www.pugscode.org). 16:51
autrijus hm? it worksforme 16:52
Limbic_Region pugscode.org and www.pugscode.org both work fine for me
theorbtwo Hm, something being screwy with my DNS wouldn't be very surprising.
jabbot pugs - 2162 - More eval_haskell fixes: Now passes all 17:01
pugs - 2163 - * `$x = 1|2; $x = 3` no longer treats `1
autrijus woot, I implemented junctive types. 17:21
(neccessary now we need to say Any|Junction instead of Any)
Corion r2163 - datenzoo.de/pugs/win2k.html 17:27
pugscode.org «Overview Journal Logs» | r1773/winxp: 587/4256 smoke/win2k (occasional): xrl.us/fqum (114/4369, r2163) Mac OS X (62/4459, r2116) | pugs.kwiki.org 17:27
Corion & 17:28
nothingmuch ingy: i managed, actually 17:36
pretty fast
ingy ok
Sporking?
nothingmuch very readable code
ingy++ 17:37
yup
i was thinking of maybe completely redoing Spork::Config so it makes more sense
~/.sporkrc/config.yaml is used to generate ./config.yaml
./config/config.yaml doesn't exist, since it's not relevant per sé
and ./config.yaml is guessed, not hard coded (user, copyright year, etc)
makes sense?
ingy yeah
nothingmuch so i'll naturally start using freepan for that 17:38
(and while I'm at it everything else too)
how do I do that?
ingy one sec
jabbot pugs - 2164 - * junctive types landed. 17:41
pugs - 2165 - * the `^` junction opertors now autothre 17:51
machack666 t/junctions/associative.t 10..14 are failing.
Are those bunko tests? (I was the one to write them, wouldn't mind a second look)
autrijus hm? 17:54
those are bunko tests. :)
one(one(@a), one(@b)) is not one(@a, @b).
Limbic_Region machack666 - out of curiosity, did you dream those tests up yourself or did you get them from an A/S/E or perhaps something on p6.L or maybe somewhere else? 17:56
autrijus we are talking to /dev/null :) 17:57
Limbic_Region the point being if we find old/incorrect information out there we should make a best effort to rectify it
duh
thanks autrijus
autrijus np :)
Limbic_Region perlbot has a tell feature though
perlbot tell machack666 about Did you dream those tests up yourself? autrijus says they are wrong but if you got them from somewhere (A/E/S, p6.l, etc) we should try to correct them if possible 17:58
perlbot Did you dream those tests up yourself? autrijus says they are wrong but if you got them from somewhere (A/E/S, p6.l, etc) we should try to correct them if possible isn't something I know about, Limbic_Region
Limbic_Region perlbot machack666 > Did you dream those tests up yourself? autrijus says they are wrong but if you got them from somewhere (A/E/S, p6.l, etc) we should try to correct them if possible 17:59
ok - that appears to have worked 18:00
simcop2387 Limbic_Region: what are you trying to get perlbot to do? talk to someone for you? you'll have to teach it that first 18:01
e.g. perlbot learn blahblahblah as Did you....
perlbot blahblahblah > machack666
perlbot blahblahblah isn't something I know about, simcop2387
ninereasons how do you leave a "pounce" or memo for someone to see when they rejoin, simcop2387 ?
simcop2387 don't think perlbot can do that
perlbot help 18:02
perlbot Syntax: (fact) :: tell (who) about (what) :: (who) > (what) :: learn (what) as (info) :: relearn (fact) as (info) :: facts (search term) :: shorten (url) :: shorten it :: search (module) :: docs (module) :: perldoc -f (function) :: math :: fortune :: flip :: host (type) (record) :: rot13 :: roll (die) :: tempconv (temp) :: scramble (text) :: 8ball :: slap :: diss :: what time is it :: highest karma :: lowest karma
autrijus works on lvalue pairs... 18:09
jabbot pugs - 2166 - * remove fp 18:11
pugs - 2167 - * add back examples/output/fp/fp
autrijus done!
rindolf Hi all! 18:13
autrijus: here?
autrijus yes.
rindolf autrijus: you got a bad review of Module::Install: cpanratings.perl.org/d/Module-Install which raises some questions. 18:14
autrijus rindolf: yeah, Alias has pinged me about it. 18:15
rindolf autrijus: OK.
autrijus: do you need help with Module::Install?
obra Alias' issue with M::I is that it's broken on 5.5? 18:16
autrijus yes.
it's a compound issue
because PPI wants a newer File::Spec than 5.5's
and he did not include EU::AI as suggested
obra Oh. 18:17
autrijus but rather used the auto-download script that was marked as obsolete
clkao i thought he now does.
autrijus well, if he now does, great.
clkao at least for class::autouse. i asked him to do so
autrijus not for PPI. 18:18
so that makes the M::I process broken.
obra Does he have a reason?
Limbic_Region simcop2387 I think I just picked the wrong syntax, the second attempt appears to have worked
autrijus I don't quite know how to do. reject 5.5 for EU::AI autodownloader block
I think
not sure. handling 5005 breakage is not my forte.
rindolf: if you are willing to help communicating with Alias_ and/or duplicating the error and/or find out where to fix, I'd be grateful. 18:19
rindolf autrijus: who's Alias_? Where can I find him? 18:20
autrijus Alias_ is here in #perl6. Alias_ is Adam Kennedy :)
rindolf autrijus: OK.
autrijus rindolf++ # thanks 18:21
jabbot pugs - 2168 - * lvalue pairs!
rindolf Is Module::Install supposed to work in Perl 5.005?
autrijus++ # You're welcome
theorbtwo The AUTHORS is sometimes useful.
autrijus rindolf: yeah, it is.
but the EU::AI part (Module::Install::AutoInstall) may not be handling things correctly.
<- not having a 5.005 to test 18:22
oh btw, the bug reporter is running it on 5.00504.
stevan: I'm commenting out this block from hash_ref.t
# my $hash = ("1st", 1);
# is $hash{"1st"}, 1, 'comma seperated key/value hash creation works';
# is $hash<1st>, 1, 'unquoted <key> fetching works';
castaway 2
autrijus because afaik, it morphs into an Array reference, not a List reference
so the duality no longer holds 18:23
feel free to ask for clarification on p6l though.
castaway oops
Limbic_Region [14:09] * autrijus works on lvalue pairs..., [14:11] <autrijus> done!, [14:21] <jabbot> pugs - 2168 - * lvalue pairs!
Limbic_Region loves the smell of progress in the morning
castaway grins
autrijus :) 18:24
castaway wonders what sort of sin shes committing, downloading opera with mozilla 18:27
Limbic_Region castaway - bootstrap problem, I am sure they will understand 18:34
could be worse - you could be doing it with IE
Odin- Oh, you know. God forgives everything ... except this. >;)
rindolf Hmmm... Alias_ has been idle for 5 hours now.
castaway heh 18:35
actually, its getting me a corrupt tarfile :(
nothingmuch hola gaal 18:36
gaal heya
did you get my mail?
if not that's too bad, cause i won't be near that account for ten days :) 18:37
nothingmuch nope, didn't check for a while
nothingmuch feels shrouded by mystery
nothingmuch goes to see what's up
castaway tries lynx instead
gaal on the other hand goes to see what's in the fridge 18:38
ack! what are you doing in my fridge? 18:39
castaway waves from the fridge 18:40
nothingmuch mutt is 50% done opening the mailbox! 18:41
castaway woo!
castaway makes the eggs explode..
gaal castaway, no, it was nothingmuch in my fridge. i'm hungry :(
castaway oh :)
castaway passes over the pizza
gaal yum yum yum! thanks! 18:42
castaway ponders what pugsish things to do
gaal [off topic] is anyone experiencing weird issues with the location: bar in the latest firefox on windows? 18:43
i switch tabs/follow a link, and the old url still shows up there.
nothingmuch 80%!
castaway no, but my tabs switch focus when moving the mouse over them ;)
gaal that's actually configurable, i just learned today! 18:44
tools > options > tabbed browsing > tab focus > last checkbox in that chunk. 18:45
castaway Tools->Options? (doesnt exist)
this is a plain moz tho, not FF
gaal ah :)
castaway tries edit prefs 18:46
gaal then the pref is prolly different anyway then
castaway yup, no "tab focus" setting(s)
nothingmuch Claim Your FREE Laser Level Pro! 18:53
what's that?
castaway a laser level?
gaal You know you want it!
nothingmuch gaal++ # thanks!
nothingmuch owes a favour
gaal yeah, get out of my fridge :) 18:54
mj41 for @a.values -> $val is rw { $val = 10; } is "Can't modify constant item" today? :-)
nothingmuch want me to order some takeout for you?
gaal nah, i found my freezer still had some pitas. 18:55
which reminds me: must buy pitas before it's too late!
Limbic_Region wonders WTF clones of his wife are doing in gaal's freezer
nothingmuch we have lots of sources of hametz here
Limbic_Region also wonders where this market is that is selling clones of his wife
pita = pain in the ass right? 18:56
nothingmuch Limbic_Region: they're all around israel
castaway also food, Limbic
Limbic_Region castaway - I know, I was being silly
nothingmuch Limbic_Region: why is your wife a pain in the ass?
gaal nm, clones of L~R's wife, my kitchen is full.
shapr Completely off-topic question, any contract programmer websites you guys recommend? 18:57
gaal how wonderful if perl6 would also make the web go away!
jabbot pugs - 2169 - * fix bad sort test 19:01
Limbic_Region nothingmuch - it is an affectionate pet name
nothingmuch pet .. .. affectionate? 19:02
what is 'i' and 't'? 19:03
mj41 hmm ... pugs -e "my @a = ( [1,2], [3,4] ); @a[1][1] = 5; say ~@a;" Can't modify too. Problem is elsewhere. 19:04
autrijus mj41: yeah. it's known bug: array_ref.t line 106 19:05
mj41 but IMHO, for @a.values -> $val is rw { $val = 10; } was ok yesterday. 19:06
autrijus it was?
Limbic_Region nothingmuch it = pita, like sweetie, honey, darling, etc 19:07
nothingmuch Limbic_Region: =P
there's a nice joke 19:08
an american couple and an israeli couple compared
gaal hey
nothingmuch context is some food place
gaal oops wrong window) 19:09
autrijus likes "make smoke"
nothingmuch pass me the honey, honey?
(american husband)
autrijus ok... finally gone thru all the failures.
remaking another round
nothingmuch israeli husband: give me that steak you cow
something like that
the setting was rather dichotomic in the version that i heard, but I can't recall 19:10
any variations
and I'm too lazy to make one up now
jabbot pugs - 2170 - * commenting out the not-really-going-to 19:11
pugs - 2171 - * semicolon AOAs is TODO.
pugs - 2172 - * fix the one last splice bug.
pugs - 2173 - * slice's context depends on the indexEx
pugs - 2174 - * we need that trailing "" for qq<$var>
pugs - 2175 - * add nested assign to force_todo; clean
pugs - 2176 - * update "sort" and "=" to chase prototy
mj41 autrijus: I thing so. For hashs at least. I used "for %conf<index>.values -> $val is rw {"
gaal you know you've been puzzled by haskell too long when you think there's a bug in make smoke when you read "util/run-smoke.pl . smoke.html" 19:12
autrijus gaal: lol
mj41: hm. and it's not working now? 19:13
autrijus waits for "make smoke" to finish
mj41 r2163 pugs -e "my %h = hash ( a => 1, b => 2 ); for %h.values -> $val is rw { $val = 0; }" 19:14
Can't modify constant item .... Val (VInt 0)
autrijus ok. will check
gaal autrijus, releasing tonight? 19:15
autrijus gaal: aiming to
may or may not actually happen
anyone wants to help updating ChangeLog?
Juerd Eeee 19:16
How long have I not compiled pugs?!
I have GHC 6.2
autrijus you need GHC 6.4 nowadays.
Juerd I knew that
But didn't realise it was so long since I last compiled pugs
(I have to admit at this point that I don't usually test the many code examples I use in IRC and mail :P) 19:17
autrijus that's fine :)
gaal autrijus, i'll help with the changelog 19:18
autrijus gaal++ # thanks!
gaal need tea for that though. brb
Juerd Bahh, debian only has 6.2 19:19
autrijus you can get from experimental 19:20
theorbtwo Or get ghc-cvs
Juerd Thanks
ghc-cvs is 6.3 here. But this mirror doesn't have any new packages either, and I'd expect much more in 2 months not upgrading 19:21
So I guess this mirror is outdated
theorbtwo It's 6.5 for me.
Juerd It's T-42s for me :) 19:22
Ah, that's better. 19:25
gaal minor(?) make issue: 'make smoke' doesn't seem to put things in blib before running the smoke tests. so if the harness fails (eg don't haev T:T:M installed), the next make relinks pugs. 19:27
(which on a low-mem machine is quite slow) 19:28
autrijus ah. fixes welcome
Juerd real 2m17.949s
user 1m55.090s
sys 0m3.711s
That used to be less too ;)
autrijus I'm not too worried :) 19:29
Juerd Does Larry's using an operator in example code, outside discussion of that operator, mean the operator has been approved of? 19:31
autrijus nothing is approved of unless it's written into Synopsis :)
Juerd (It's XX, of which I can't find any ruling)
I'd want to write tests for it, but it's a waste of time if the operator doesn't officially exist yet 19:33
mj41 Why output of perl -e "use File::Spec::Functions; print catfile( '', '.\\fn');"
\fn
i prefer fn 19:34
autrijus this is win32?
Juerd This is Perl 5?
mj41 yes
Juerd See channel name
mj41 and pugs -e "use File::Spec; say catfile( '', '.\\fn');" 19:35
\\\fn
Juerd See channel name
crysflame he just said "pugs -e", Juerd.
Juerd Oh, now it gets more interesting :)
autrijus Juerd: see command name :)
Juerd crysflame: Yeah, I was mentally lagged
crysflame enforcer mentality deth 19:36
jabbot pugs - 2177 - * fix "sort"'s prototype. 19:41
Qiang refreshs screen 19:57
jabbot pugs - 2178 - * fix zip test syntax 20:01
pugs - 2179 - * repair %h<x> to mean %h{'x'} instead o
pugs - 2180 - * repair a couple tests
Juerd Is there a web based interface to pugs' versioning system?
With nice coloured diffs and such 20:02
obra rt.openfoundry.org
go to pugs
click on version control
it's running svnweb
Juerd Perlfect
Thanks
theorbtwo pings stevan. 20:04
autrijus gaal: how's changelog? 20:05
gaal about 60% done
autrijus I have only ~4 test files to fix
all in t/data_types/
gaal is it :Perl5 or :P5?
autrijus otherwise I think we're good to go
both works; I like :P5 more. :perl5 also works.
larry warns that he's moving to :P5/:Perl5
but synopsis still says :perl5, so until it changes, we support all three 20:06
gaal does :g work? there's mention of :P5<g>, but not of :g
PerlJam :g has a new name
(assuming you meant g as in global)
gaal PerlJam: for p5 regexps
20:06 [1]metaperl_ is now known as metaperl_
autrijus :g works too. 20:06
gaal thanks
autrijus as does :i 20:07
gaal oh no :) i was changelogging from an older release. timewarping.... 20:10
Juerd operation failed: path does not exist at /usr/local/lib/perl5/site_perl/5.8.5/SVN/Web/Diff.pm line 29.
rt.openfoundry.org/Foundry/Project/...p;rev2=499
jabbot pugs - 2181 - * even more repairing. 20:11
autrijus Juerd: hmm, a bug. file a ticket in the openfoundry queue?
Juerd: hey, maybe you can answer this 20:12
foo(1,2,3);
foo(1,(2,3));
foo((1,2,3));
they are potentially different things, right?
not always the same
gaal i can't find the r of the last release?
autrijus gaal: r1921
r12834 (orig r1921): autrijus | 2005-04-13 03:43:50 +0800 20:13
* This be Pugs, version 6.2.0.
gaal um
then soemone already did half the changelog? and i wans'
t confused after all?
i am now :)
autrijus stevan already did it up to yesterday
or rather, some 40 hours ago
Juerd autrijus: Depends on foo's signature, I think
autrijus Juerd: right. 20:14
Juerd OTOH, no, I think they're all the same.
autrijus Juerd: right.
wait, you can't be right on both times.
gaal ah.
ingy wiki.freepan.org/index.cgi?Changing...anPassword
Juerd Or, wait, no, "," is a constructor now, isn't it? :)
autrijus Juerd: you are confusing me :-/
I'm already confused
Juerd That'd mean they have to be different
But they can't be, because non-subcall parens are only for forcing precedence
HELP
There are two similarly spelled operators, one being , and the other being ,. 20:15
autrijus *exactly*
Juerd One separates sub call arguments, the other list elements
obra ingy++
autrijus right.
Juerd Except of course when lexical (lexical as in how they appear in code) lists do flatten, regardless of context
In which case everything matches up again
In a most useful way :)
autrijus so... 20:16
is((1,2), (3,4), "hey");
is this happy or not?
assuming &is has the signature of <Str,Str,Str -> Str>
Juerd No
Oh
What's ->?
autrijus <Str,Str,Str -> Bool>, rather
the return type
Juerd And doesn't -> in <> clash?
autrijus not something canonical ;)
ninereasons src/Main.o(.text+0x26ab): undefined *** Error code 1 20:17
Stop in /usr/home/markmc/cvs/pugs.
autrijus i.e. it's my notation, ignore it
Juerd Oh!
ninereasons At revision 2181
theorbtwo It's very haskellish.
Juerd It has too many arguments, I think, autrijus
Although
Hm, no 20:18
It's okay!
Because the *,* is in scalar context, not the parens - they just group
Yep, that must be it
autrijus ok. I'm asking p6l :)
Juerd can be extremely confusing when he's thinking out loud
autrijus: I think it's okay and equivalent to is([1,2], [3,4], "hey")
mj41 builtins/arrays/nested.t -> data_types/multi_dimensional_array.t or nested_arrays.t ? 20:19
Juerd Now I'm going home 20:20
(Almost 22:22 and still @office)
autrijus mj41: both works for me; please refactor away 20:21
mj41 Juerd ... 22:22 same in Czechia :-)
theorbtwo 22:22 in Germany and the "flirt" commercials are starting. 20:22
crysflame flirt commercials? 20:23
theorbtwo "Call our flirt-line at 1-900-NNN-NNNN." 20:24
crysflame ah, right 20:27
mj41 some from nested are not multi-dimensionals, IMHO 20:28
perl -e "my @c = (1,2,(3,4)); print $c[3];" # 4 20:29
jabbot pugs - 2182 - move builtins/arrays/nested.t to data_ty 20:31
gaal autrijus, getting close
autrijus cool.
mj41: feel free to refactor mercilessly :) 20:32
ah, you put it in nested_arrays.t. that's good
ninereasons I still can't build r2181 , autrijus 20:33
is that news? 20:34
autrijus ninereasons: tried clean?
ninereasons realclean :-)
autrijus try rechecking out? :)
(that is news.)
ninereasons I'll try rechecking out.
gaal i'm confused about r numbers: when i svn log, the latest is 2264, but jabbot gives something else
autrijus gaal: are you checking out from svn.perl.org? 20:35
mj41 "-> $val is rw" and File::Spec are broken ... so time to live tutorial_gen a go for shower
autrijus mj41: what test tests "-> $val is rw"?
if there's none, please write one?
gaal autrijus, yes, should i svn switch back? 20:36
autrijus gaal: sure!
gaal svn++; # for having switch 20:37
ninereasons is_rw.t, and t/statements/for.t test -> $x is rw 20:38
gaal then again now it doesn't work ;)
ninereasons t/statements/for.t only, actually
mj41 statemests/for.t 23
ninereasons checking out again worked, autrijus. r2184 builds successfully 20:40
autrijus hmm. mysteries
jabbot pugs - 2183 - * remove the unneccessary END block 20:41
pugs - 2184 - * that too.
gaal r2185 - changelog
autrijus gaal++ 20:42
gaal better give it an editorial eye though, autrijus
autrijus k 20:43
gaal sorry it took so long... anyway let's see what's up with my bg smoke test.
theorbtwo gaal++
gaal blushes 20:44
the smoke process really needs some backporting love so i can run it on msys 20:46
mj41 gaal++
Corion We have "make smoke" now? 20:47
(and are we preflight? I'm backlogging and slowly catching up)
pasteling "nothingmuch" at 212.143.91.217 pasted "Spork::Config::Default" (96 lines, 1.7K) at sial.org/pbot/9534
"nothingmuch" at 212.143.91.217 pasted "Spork::Config::Default" (96 lines, 1.7K) at sial.org/pbot/9535
theorbtwo "backslash protections"?
It's unicode URL decoding, not utf8.
gaal Corion, yes 20:49
actually an alternative to porting the new TAP stuff to p5.6.x would be to port it to p6 :) 20:51
and to bundle it with pugs
though until we have YAML...
Corion Bah. We should embed SQLite. Then we can do away with YAML :) 20:52
autrijus mumbless something about patches welcoeme 20:53
Corion Or we should get Data::Dumper ported, or abuse/misuse that Haskell database, what's its name? HaskellDB ? :)
gaal porting yaml.pm would no doubt happen before dbi.pm :)
autrijus I think SQLite is the most attractive :)
PerlJam er, why do we want to get rid of yaml?
gaal SQLite is beautiful (lovely license too)
autrijus right. I like that license. 20:54
PerlJam: it's not either/or... :)
Corion finds cvs.sf.net/viewcvs.py/haskell-libs/...se/Sqlite/
autrijus "getting away without" != "get rid of"
Corion PerlJam: I'm not really enchanted by the stability of YAML for p5
PerlJam autrijus: sorry, I've been suffering the tyranny of OR lately.
autrijus PerlJam: I think you meant XOR
gaal inclusive OR++
PerlJam well, XOR in our usage, OR in "common" usage. 20:55
autrijus whew. I finally worked out the "," semantic.
please fire away your smokes :)
not exactly preflight, but close
autrijus looks at ChangeLog
nothingmuch ingy: wake up
Corion I'm currently running a smoke - still some failures
(win32 of course)
autrijus Corion: abort it?
PerlJam what comma semantic? you mean @a = (1,) ?
autrijus r2186 changed ~everything
gaal ah, we were in prepreflight then, Corion :) 20:56
Corion autrijus: I'm already at "p", so I'll let it fly and check out again immediately
nothingmuch i see 99.26% good
autrijus PerlJam: $x = [$y[1,2,3], $y[1], $y[3,4]]
Corion Pugs. Now with five nines goodness :)
nothingmuch array.t and array_ref.t
nothingmuch.woobling.org/pugs_test_status/
Corion ... which is better than nine fives!
autrijus nothingmuch: right, I just fixed the two
nothingmuch so it should be updated RSN
gaal wonders abotu that 'make smoke' linkarage thing 20:57
nothingmuch nothingmuch.woobling.org/pugs_test_...l#line_326 <-- also fails
Corion BTW, that test failure of pugsrun/10-* comes from not running "make", as that test needs File::Spec copied to blib6/lib.
"test failure" == "complete skip"
nothingmuch compile is sloooow 20:59
it's not declaring skip all though, is it? 21:00
gaal zillions of "pugs.exe: cannot cast from VList [] to Pugs.AST.VCode" in t/builtins/strings (msys)
but these are just warnings
jabbot pugs - 2185 - ChangeLog updates 21:01
pugs - 2186 - * finally, sensible "," flattening seman
gaal anyone else see those?
not just in strings, i see now - once per almost all files. 21:02
*almost each file
Corion r2185 - datenzoo.de/pugs/win2k.html 21:03
r2185 - 34 failed
castaway Please find a copy of my CV attached.
oops ;)
Corion castaway++ # releasing CV with Pugs
gaal lol 21:04
castaway grins
Corion r2186 smoke started ;)
... what better way to promote yourself? At least if you want to work in a shop that uses Pugs ;)
gaal Corion++ # da smoke machine
castaway hmm, tisnt a bad idea actuall;)
+y
theorbtwo "The smoking man."
gaal so *thats* what Burning Man is all about. 21:05
Corion *cough*
stevan nothingmuch: pong
theorbtwo sniffles.
stevan theorbtwo: poing
nothingmuch chimes
theorbtwo tries to remember why he pinged stevan.
autrijus considering that the person behind all technology used in Burning Man is Brian Behlendorf...
Corion Hmmm. Is there an easy shell command/loop to check if a file has been changed? I want to trigger the upload once the html file has been generated ...
stevan autrijus: did I read Junctive types are implemted??? :) 21:06
nothingmuch Corion 'wait'?
autrijus stevan: you read correctly. currently only in subroutine signature
sub foo (Any|Junction $x) { ... }
Juerd Yay!
stevan autrijus++ # that is soooooooo Cool
Corion nothingmuch: The command that I would wait for runs on a different machine
autrijus sub foo (Array&Hash $x) { ... }
Juerd autrijus: I eventually guessed correctly! ;)
nothingmuch ah 21:07
stevan has been longing for sub sigs in perl for several years
nothingmuch rsync is prolly the best way
code wize
Corion Hmmm. Maybe I should simply run rsync in a 1 minute loop, true.
nothingmuch otherwise you have versions of find that know to say newer than
Corion you can check the next smoke result at the usual URL then, because I'll be afk for the next half hour :)
nothingmuch but the only STD way I know of compares file a to file b
gaal wasted 20 minutes debugging *today* becaseu of a forgotten "... = @_;" 21:08
nothingmuch gaal: use Spiffy;
gaal no!
use perl6.
Juerd The yada yada yada operator is lvalue now?
autrijus gah.
Juerd What's the world coming to? :)
Corion Source filters create more problems than they solve :)
nothingmuch i think
source filters give me the heeby jeebies 21:09
castaway gaal: BTDT
nothingmuch but I must admit i'm tempted after about an hour of practice
true that
autrijus Juerd: you are correct after all.
nothingmuch Juerd: AOP ;-)
Corion: true that
autrijus Juerd: so if &is is of signature (Array, Array, Str)
(all three nonslurpy)
gaal castaway, I've BTDT too, which only makes it more aggravating :)
autrijus Juerd: is the inner (1,2) still running in scalar context?
Corion The r2186 smoke test is at /oo/ already. So you'll be able to find it at datenzoo.de/pugs/win2k.html in a few minutes I think.
&
autrijus Juerd: or are the suddenly in Array context? 21:10
castaway heh
Juerd autrijus: Then it's still scalar context (specifically, array context, which surprisingly nicely is not a form of list context) :)
autrijus: Array context is a scalar context
autrijus Juerd: that is hard to believe.
gaal corion, did you not get all those "cannor cast" warnings? only on msys apparently.
autrijus Juerd: I can see "array ref" context, maybe
Juerd I think we should rename scalar context to singular context
autrijus Juerd: but Array is a List
as Hash is a List
Juerd autrijus: Array is a class, classes when used for context expect objects, objects are references, Array implies ref
gaal oh, he left 21:11
theorbtwo wonders why skip is pale green.
Juerd autrijus: That may be, but types have nothing to do with context.
autrijus Juerd: wow. that line is the most clear thing I have read.
Juerd: it also invalidates my hope to release 6.2.1 tonight :-(
stevan gaal: I hope you dont mind I just added a code example to the Junctive types line in the changelog
gaal not at all!
stevan is so excited to get started messing around with them :)
autrijus Juerd: since that means my Eval.hs comma implementation is All Wrong.
stevan I have to go eat though,. I will check in later :) 21:12
Juerd autrijus: I think it'd be good to rename scalar and list context to singular and plural context, to avoid confusion with types.
autrijus Juerd: so, it has nothing to do with types, right?
stevan autrijus++ # for all the cool stuff done today
theorbtwo wonders what line.
autrijus it's either singular, or plural
gaal isn't very proud of his changelog skillz... but somebody's gotta do it and you werent' areound, stevan :)
autrijus literal lists always flatten in plural context
and never flatten in singular context
singular context may be further limited in expecting specific object types 21:13
Juerd autrijus: The main contexts have nothing to do with types, except that scalar(singular) context can expect a certain type of singular value, for example, an Array (reference)
stevan gaal++ # for being brave enough to volunteer :)
autrijus while plural context does not care about type
is that the case?
Juerd autrijus: Exactly
autrijus: Plural context can care about the types of its values
autrijus Juerd: I wish these lines are in the synopsis!
theorbtwo OK, my right hand hurts, and my vision is starting to go out of focus.
It's time for bed.
autrijus Juerd: can you be kind and summarize these line into a reply on p6l to my thread? 21:14
and get another "That's all correct" from Larry
castaway oooh bedtime
mj41 r2185 for.t still not ok 23 :-(
autrijus so, sorry, folks, no 6.2.1 tonight because the new context handling is all wrong :-/
Juerd autrijus: Which lines precisely?
castaway runs off.
autrijus Juerd: everything after
05:08 < autrijus> Juerd: you are correct after all.
Juerd autrijus: That's a lot. 21:15
autrijus Juerd: well, what you said, anyway
I'd do that, but it's 5:14am here and my fingers are on strike :)
Juerd I'll see if I can make a nice bulleted list
autrijus thank you
Juerd++
is Void a special subtype of Singular, then? 21:17
or is Void a context of its own like in p5? 21:18
gaal Nilular? :) 21:19
autrijus Nullar
I'll need to sleep on this new revelation.
journal up... please keep the usual flow of new tests and examples etc :) 21:20
Shillo I think I fixed myself the proper svn access... But I need URL?
gaal (i've seen "nullary" mentioned in some haskell texts, no doubt in different, er, contexts)
autrijus there's no release freeze
Shillo: svn.openfoundry.org/pugs
gaal: it's used in arity context, so "unary" instead of "singular"
jabbot pugs - 2187 - adding some example code for Junctive Ty 21:21
gaal ah. the *opposite* of context.
(in a way.)
autrijus :)
autrijus waves... I'll be back :)
gaal night!
autrijus see you tomorrow, folks
night!
&
mj41 night 21:22
Juerd autrijus: I'd think void context would not be Void context, as you cannot have a reference to nothing, or express nothingness in a variable (we have undef), or meaningfully force void context in an expression 21:25
Shillo Hmm, does pugs with with ghc-6.2.2 ? 21:28
Er, -work- with.
Juerd autrijus: summary sent to p6l
ninereasons it might Shillo , but only >= 6.4 is supported 21:30
jabbot pugs - 2188 - unbreak motd 21:31
Shillo ninereasons: Oif. 21:32
ninereasons: To make things less funny, apt is bugging me about downgrading gcc.
ninereasons I think that ghc 6.4 might be in experimental ? I've overheard this, anyway. 21:35
Juerd I used ghc-cvs from current sid(unstable) and that works well
Shillo ninereasons: I'm on Ubuntu. That leaves me with ghc-cvs from Universe... which also bugs me about gcc. 21:36
ninereasons bummer, Shillo :-(
Corion One last smoke before bedtime 21:37
Somehow, autrijus makes me feel bad, as he goes to bed at the same time I do, but he's one third-day ahead of me ...
Shillo ninereasons: Working on fixing it. :)
ninereasons if you give up, building from sources, which is what I did, using ghc 6.2.2, only took a few hours (gasp) 21:38
Shillo ninereasons: I can do that after I do my hardware upgrade, something I'd ideally want to delay a few months more. 21:50
Juerd put the irc log of the context discussion in a use.perl comment for later reference
Shillo My trusty old Athlon 2000+ got a bit long in the tooth. Can't stand the heating like it used to.
Juerd use.perl.org/comments.pl?sid=26150&...&pid=0
Corion 'night 21:54
Shillo Hmm... finnagled the makemakefile to grok the old ghc. Build fails on unknown pragma. Hmm, now we're in Haskell land... 21:58
nothingmuch autrijus: ping 22:00
ah, too late
anyone else ever used spork?
i fixed it's config system
mugwump I've used spork 22:02
nothingmuch mugwump: want to try my changes, see if they make sense?
mugwump sure, why not 22:03
nothingmuch unidiff?
mugwump shurgs ambivalence
nothingmuch or should i just tar the whole thing up?
mugwump whatever's easiest, mail to [email@hidden.address]
nothingmuch or darcs repo?
mugwump I can't grok darcs :) 22:04
nothingmuch (or get ingy to get my freepan htpasswd stuff fixed ;-)
darcs is easy, but i'm not in an evangelist mode
unidiff against svn.kwiki.org head ca. noon GMT today
the only thing i don't yet know how to do is mask a file from Spoon::Installer 22:05
so that config/config.yaml isn't created
that kwiki stuff is not really necessary
mugwump what's up with FreePAN? I've got admin on that, can fix it
but given the head's in ingy's svn I think I might leave it for now
nothingmuch whose given head? 22:06
anywho, the change basically is: Spork::Config::Default - some default values
all of config.yaml is commented out, because defaults make sense now
./config.yaml overrides ~/.sporkrc/config.yaml which overrides Spork::Config::Default
mugwump Nice one 22:07
nothingmuch hmm... is output of 'svn di' good enough?
mugwump for sure 22:08
nothingmuch nothingmuch.woobling.org/spork_defs.diff 22:09
(i don't feel like waiting for mutt to open maildir)
(i really have to get that under control some time soon)
hmm.. pardon the tabs 22:11
i'll fix that
mugwump Heh, _guessing_defaults looks like fun 22:12
Is Devel::Symdumps core?
nothingmuch yup 22:13
since 5.6 i think
nothingmuch checks Module::Corelist
mugwump not in 5.6.1
nothingmuch hmm, maybe it isn't 22:14
anyway, ingy pondered adding this ability to spiffy anyway
mugwump ok
nothingmuch sorry
next i'll butcher Spork::S5 so that it creates "true"er S5 22:17
with the inc class and all
Juerd ** Proxy scalars: `%ENV<PATH> ~= '/tmp'` 22:20
Should probably be :/tmp
* Experimental support for link external Haskell libraries
Is *anything* in Pugs *not experimental*?
nothingmuch maybe %ENV<PATH>.ref eq 'Array'?
or is that sort of pushing it? 22:21
I'd like Env::Path like support in the core, it makes things easier
Juerd nothingmuch: pushing it. Heh. 22:22
pugscode.org <Overview Journal Logs> | r1773/winxp: 587/4256 smoke/win2k (occasional): xrl.us/fqum (114/4369, r2163) Mac OS X (62/4459, r2116) | pugs.kwiki.org 22:23
Juerd My terminal apparently does NOT like utf-8. 22:23
nothingmuch if we're doing things the right way and not the c library way, and have magic like $*CWD and floating point alarm then why not this too?
Juerd And <<>> is unnecessary there anyway.
nothingmuch: Because %ENV is a very special proxy that you should be able to depend on. 22:24
With NO surprises.
nothingmuch good night Juerd, everyone 22:28
mugwump Hmm, Devel::Symdump isn't even in my 5.8.6 dist 22:30
ninereasons frozen light? sounds cool: snipurl.com/e56d
nothingmuch no, it isn't.. sorry
i was being silly
Limbic_Region hmmm - how would I go about finding out who wrote examples/games/animals.p6 ? 22:31
ninereasons svn blame
svn blame examples/games/animals.p6 22:32
Limbic_Region ninereasons - I don't actually have svn or pugs
Limbic_Region hangs his head in shame 22:33
ninereasons it was iblech, Limbic_Region :-)
Get Pugs!
ninereasons pays for services by proferring uninvited advertising 22:34
Limbic_Region well, it was near the top of my todo list ninereasons
nothingmuch Limbic_Region--; # it's not that hard, just do it man, and you get karma + 2
before i really go too 22:35
nothingmuch brushing teeth
Limbic_Region you know, make sure your wife can stay in the country, keep your job, buy a house, get a car, read HOP, write an SQL parser in P::RD, learn Haskell, get involved in Pugs
nothingmuch brush teeth, ++ Limbic_Region...
Limbic_Region nothingmuch - I remove karma anytime I see it applied to me, but thanks for trying to motivate me
nothingmuch ah
Limbic_Region++; # i'll -- for you, and in the time I save you go get pugs 22:36
Limbic_Region lol - thanks
ninereasons nothingmuch++
Limbic_Region in any event, a good portion of those things will be done by the end of the month
nothingmuch well, good luck
Limbic_Region I will likely skip the P::RD project and combine learn Haskell and get involved in Pugs 22:37
nothingmuch DBI has a pure perl sql fudger
Limbic_Region nothingmuch - the P::RD project was just to become a SME - I don't really need it for anything 22:38
nothingmuch ah
22:43 ihb is now known as Homer, Homer is now known as ihb
Shillo Later, folks! 22:44
ninereasons success, Shillo ?
Shillo ninereasons: Missing Data.Map. Will work on it tomorrow. Prolly have to compile from source.
ninereasons better luck later, then.
Shillo Hope so. :)
stevan is anyone actually using the TEST_LOG_FILE feature in Test.pm? Or is everyone using the smoker now? 23:08
I am considering removing it from Test.pm 23:10
moving it to it's own module
mugwump nothingmuch: I have implemented a quick hack so that change doesn't require Devel::Symdump 23:22
I'll send them both on to ingy as patches 23:24
stevan mugwump: I think he is sleeping
ingy nothingmuch: ping 23:59
nothingmuch: please retry your freepan repos

Powered by Cro and the Raku® Programming Language.

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

Thank you!