svn switch --relocate svn.openfoundry.org/pugs svn.pugscode.org/pugs/ | run.pugscode.org | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com
Set by avar on 16 November 2006.
nperez something i've never considered 03:24
what happens if the underlying object in a POE::Session suddenly becomes undef? 03:25
using object_states
crap
wrong window
:)
bobsaget DCC SCHAT isoiledmypantaloon 05:10
audreyt dmq: ok, so the y2k entry didn't win me beer 06:22
lambdabot audreyt: You have 8 new messages. '/msg lambdabot @messages' to read them.
audreyt I now claim a winning entry. it's on CPAN now :)
@tell dmq perlcabal.org/~audreyt/tmp/re-engin....01.tar.gz
lambdabot Consider it noted.
audreyt @tell putter do you perhaps have tuits to turn re::override to use re::engine plumbings dmq hacked so we have a generic perlland overrride hook? see my PCRE entry for some readymade code :) 06:24
lambdabot Consider it noted.
gaal audreyt: needs an 'unimport' method too 06:40
i wonder, is undef $^H{regcomp} sufficient for starters?
also the filename says 0.01, but the Changes suggests 0.03 :)
*poof*
Khisanth at least it doesn't have mismatching versions of different modules in different versions of the tarball! 06:41
audreyt ^Wgfixed and uploading 06:55
gaal yeah, hate software #68712: when installing Net::SSLeay with PREFIX, you have to say perl Makefile.PL -- PREFIX=... because it bundles another module tarball inside it and the directive doesn't propagate otherwise 06:59
*repoof*
{ temp unpoof; note("--") } 07:00
wolverian gaal, I wonder how you found that out. seems pretty arbitrary. 07:30
ayrnieu thinks of we.hates-software.com 07:45
gaal wolverian: trial and tribulation... 08:03
avar gaal: yeah, I noticed that too 10:18
that pisses me off every time I install Net::SSLeay
avar@Arsia:/tmp/Net_SSLeay.pm-1.30$ perl Makefile.PL PREFIX=/my/usr/local/is/here 10:20
Checking for OpenSSL-0.9.6j or 0.9.7b or newer...
I could not find your OpenSSL in `PREFIX=/my/usr/local/is/here'
oh that's another issue actually 10:21
it doesn't assume that your PREFIX also has the openssl libs (should it?)
broquaint IIRC, PREFIX is where stuff is going, it shouldn't change how it looks for libs. 10:23
avar yeah 10:24
gaal that's just a default where-to-look-for-them thing. you can enter e.g. /usr at the prompt there and it'll work. the problem i was moosing about was that come install time, N:S is installable but its bundled N:S:Handle is not, because it didn't inherit PREFIX and so wants to be installed in the default system location. 10:39
xinming hello all 11:40
kolibrie fglock: I'm installing v6.pm from CPAN, but most (if not all) tests are failing to compile 15:47
pasteling "kolibrie" at 66.239.158.2 pasted "fglock: test compilation failures for CPAN v6.pm" (21 lines, 952B) at sial.org/pbot/21624
kolibrie fglock: any ideas what I should try? 15:48
pmichaud are there any accessors/methods defined for Range objects? 15:49
(i.e., in the synopses)
fglock kolibrie: looking 15:54
pmichaud: not afaik; there are some old notes in docs/notes 15:58
pmichaud fglock: okay, thanks. That's sorta what I figured.
fglock kolibrie: try updating Module::Compile ?
kolibrie fglock: okay
fglock pmichaud: imo, Range should behave just like List 15:59
pmichaud fglock: sure, but one might want to have special methods to obtain the min/max of the range
fglock and also deal with '~~' 16:00
pmichaud exactly 16:01
from t/operators/range.t: 16:04
is ~(1 ..^ 4.9), "1 2 3" , "upper exclusive limit is truncated";
that looks wrongish to me -- I think it ought to be "1 2 3 4"
kolibrie fglock: removed all copies of Module::Compile, reinstalled Module::Compile 0.20, v6.pm tests still getting same errors 16:11
:( 16:12
fglock pmichaud: some info scattered in the specs: s03 - .min and .max values (which change as they are iterated); s09- You can use ranges as types; :by(2); 2..1:by(-1); alphabetic ranges; Range object smartmatches the endpoints
pmichaud fglock: excellent
fglock kolibrie: another possible problem, you may have a devel version of PCR installed 16:13
wolverian pmichaud, I suppose it's an integer range, so the exclusivity drops the 4
pmichaud wolverian: but later in the test we have is ~(1.9 ^..^ 4.9), "2 3", "both exclusive limits are truncated";
er, 16:14
is ~(1.9 ^..^ 4.9), "2 3", "both exclusive limits are truncated";
wolverian pmichaud, that looks correct to me as well. it's 1 ^..^ 4
pmichaud, (that is, assuming integers.)
kolibrie fglock: I deleted all 'Pugs' directories from my site_perl, thinking that might be the case. Are there other namespaces to delete?
fglock maybe infix<..> has signature (int, int) 16:16
pmichaud wolverian: oh, so the arguments to infix:.. has in... yes, what fglock just said
TimToady I don't think so 16:16
pmichaud I wouldn't think so either :-)
TimToady ~(1.9 ^..^ 4.9) should produce 2.9, 3.9
pmichaud oooooh 16:17
wolverian ah. that is more comfortable indeed.
TimToady with handwave at floating-point issues
pmichaud and ~(1.9 ^..^ 4.5) would produce the same?
TimToady yes
pmichaud okay, that helps a bunch
I'll see about updating the tests 16:18
fglock kolibrie: I'll reinstall
wolverian hrm. rationals?
TimToady rats would presumably help with that.
pmichaud I haven't done a lot of test updates in pugs; any conventions I should be aware of ?
kolibrie TimToady: and what if you want 2.0, 2.1, 2.2,... 4.4?
pmichaud kolibrie: 2.0..4.4:by(0.1) ? 16:19
kolibrie is that :by(0.1)
TimToady why not?
kolibrie okay 16:19
wolverian TimToady, would all floating point values be rats, or just when the signature says so?
TimToady I think maybe .. stores all its args as rats but produces floats/ints
pmichaud looks like I'll need a rat type in parrot :-) 16:20
a parrat :-)
TimToady by and large I don't like rats
pmichaud (I'll obviously cheat and use float for a while)
TimToady I think that's fine. 16:21
pmichaud TimToady++ # thanks for quick answer
wolverian I've had some rats under my shirt. it wasn't unpleasant, surprisingly.
TimToady I had a lizard up my pantleg once...
wolverian (now, pants would have been something else)
TimToady that was not pleasant
wolverian ouch. did you hop around? 16:22
TimToady no, I stood there hollering at my mom with my hands around my thigh to keep it from going higher.
she told me to hobble over to the bathroom and take my pants off, which I did.
fglock kolibrie: same error here 16:23
TimToady the poor lizard thought it had found a nice dark cave to hide in...
clkao audreyt: doh, you should have bought this: news.bbc.co.uk/go/rss/-/1/hi/entert...209658.stm
lambdabot Title: BBC NEWS | Entertainment | Auction frenzy over Hepburn dress
kolibrie fglock: glad it's not just me 16:25
wolverian heh. I didn't realise at all you could use your hands like that. my survival points are negative, I think.
fglock kolibrie: downgrading PCR to 0.17 works 16:31
kolibrie fglock: okay, will try that 16:32
kolibrie does pugs use the installed PCR, or the dev version in the pugs repository? 16:33
fglock kolibrie: not sure 16:35
svnbot6 r14866 | pmichaud++ | Updates to range.t tests.
r14866 | pmichaud++ | # 16:16 <TimToady> ~(1.9 ^..^ 4.9) should produce 2.9, 3.9
r14866 | pmichaud++ | # 16:17 <pmichaud> and ~(1.9 ^..^ 4.5) would produce the same?
r14866 | pmichaud++ | # 16:17 <TimToady> yes
pmichaud hmmmm 16:37
my $x = 3.1^..6; say $x.min produces ... ?
kolibrie hmm, doesn't look like pugs wants to compile for me either: ghc-6.6: unknown package: mtl 16:38
TimToady pmichaud: still produces 3.1 16:42
the min and max are just stored. the ^ says how it's applied. 16:43
(likewise the :by is just stored, if any)
hmm, I wonder if :by should have any influence on a ~~ comparison... 16:44
$a ~~ 0..10 is defined as 0 <= $a <= 10, but 16:45
$a ~~ 0..10:by(1) might mean $a == any(0..10)
well, === really, since range might be nonnumeric 16:46
or it might just ignore the :by 16:47
TimToady same distinction arises for strings, actually, but I don't know how to write the increment... 16:49
"bbb" ~~ 'a'..'z' # true
"bbb" ~~ 'a'..'z':by($string_increment) # false
I suppose it's better just to ignore the :by and force people to write any() explicitly if they mean that. 16:50
or carp about a "useless use of :by", or some such.
pmichaud TimToady: agreed (with all points) 16:53
so, in 3.1^..6 the range object has to "know" that the lower bound is excluded
that makes sense, since we need it to smartmatch that way 16:55
okay, works for me
eviltwin_b kolibrie: make sure you have the ghc-extralibs package installed 16:56
pmichaud afk # lunch
kolibrie eviltwin_b: I don't see a package called ghc-extralibs in debian 17:01
eviltwin_b hm, I wouldn't know what debian calls it 17:03
but ghc is typically distributed in stripped-down form, stuff like mtl lives in an extra package
kolibrie eviltwin_b: is there a specific filename I can look for? 17:04
eviltwin_b $prefix/lib/ghc-6.6/imports/Control/Monad/Cont.hi should do 17:05
eviltwin_b (on debian I presume $prefix == /usr) 17:06
kolibrie maybe usr/lib/mtl-1.0/ghc-6.6/Control/Monad/Cont.hi ? 17:07
that's listed under libghc6-mtl-dev
eviltwin_b that might well make sense if they unbundled it. I don't have a debian (or indeed a linux) box to check; I'm on a Mac 17:09
but that certainly sounds like the right file
kolibrie I'll try it
Limbic_Region hangs his head in frustration 17:10
Limbic_Region no longer has perl, pugs, mingw, cygwin, firefox, gvim, haskell, parrot, etc 17:12
kolibrie fglock; with PCR 0.17, v6 fails 3 subtests (two in 07-simple-multisubs.t and one in t/builtins/arity.t). Anything I have to worry about? 17:14
TimToady Limbic_Region: are you programming a Java toaster or something? 17:18
Limbic_Region If it wasn't meant to be funny then no 17:21
grrr - this client is eating half of what I type
TimToady - if that's funny then I guess I should lookup what a Java toaster is
the 3rd line should have preceded the 1st 17:22
in any case - just a difference of opinion between local ISSO and parent organization. Until the differences are worked out I have removed all software not installed from image 17:23
wolverian Limbic_Region, ...and soon they'll let you go because you're not productive anymore? 17:25

Powered by Cro and the Raku® Programming Language.

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

Thank you!