6.2.10 released! xrl.us/hxnb | geoffb's column (/. ed): xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: paste.lisp.org/new/perl6 or sial.org/pbot/perl6 | www.geeksunite.net Set by integral on 25 October 2005. |
|||
gaal | putter: except that we do have it working :) (well, not without some hack) | 00:30 | |
but i'm not here. zzZZZ & | |||
01:14
stevan__ is now known as stevan
|
|||
stevan | any Bostonians around? In particular MIT people? | 02:06 | |
I am trying to find out information about the MIT Little Languages workshop this year | 02:07 | ||
putter: if you backlog,.. IIRC that is your neck of the woods,.. maybe you know something :) | 02:09 | ||
ingy | stevan: putter == MIT iirc | ||
stevan | putter: If it is happening, and I go, we should meet up | ||
ingy: yeah I think you are right,.. I know he is in Mass | 02:10 | ||
ingy | he went to MIT | ||
stevan | ah | ||
ingy | at least he had an mit email addy once | ||
Mitchell Charity. Know him from his many contributions to Inline | 02:11 | ||
well never met him physicallt | |||
y | |||
stevan wonders how many people ingy *doesn't* know :) | 02:12 | ||
buu | ME! | 02:56 | |
dduncan | ?eval @foo = [2,3]; @bar [10,20]; @foo >>+<< @bar | 05:05 | |
?eval 1 | |||
whence goes evalbot? | 05:08 | ||
QtPlatypus | I think it was lost in the last netsplit. | 05:13 | |
06:31
peterson_ is now known as dhpeterson
|
|||
dduncan | ?eval 1 | 07:28 | |
evalbot_7755 | 1 | ||
dduncan | ?eval kick it | ||
evalbot_7755 | Error: No compatible subroutine found: "&kick" | ||
dduncan | evalbot came back, but svnbot didn't alas | 08:01 | |
well, nightey night | |||
10:39
ods15_ is now known as ods15
|
|||
r0nny_ | yo | 13:33 | |
anyone alive ? | |||
xinming | yes. | ||
r0nny_ | any new comments on the SixPanUtils ideas ? | 13:35 | |
xinming | r0nny_: Maybe, I am the worng person for you. :-P | 13:37 | |
r0nny_ | hmm | ||
maybe | |||
anyone else any comment ? | 13:41 | ||
xinming | r0nny_: what that is plesae? | 13:42 | |
r0nny_ | sixpan.perl6.nl/sixpan/index.cgi?SixPanUtils | ||
basically cpan utils for perl6 | |||
oh - and the module build utils | 13:43 | ||
xinming | r0nny_: hmm, can we have a option to install module into ~/.perl/perl6/modules such kind of dir? | 13:45 | |
r0nny_ | read - its allready there .. | ||
Juerd: got some time ? | 13:48 | ||
hmm - seems to be away | 13:49 | ||
well - im going now - latter | 13:50 | ||
xinming | r0nny_: hmm, can module be install into a specified dir? | ||
r0nny_ | xinming: should be easy to manage latter in the process - initial versions might be a bit unflexible about such stuff | 13:52 | |
first is to manage globla install, then add user-hope install wich basically should make it flexible enough to install anywhere | |||
global ;P | 13:53 | ||
xinming | how about multi versions of modules? just like can 1.0 co-exist with 2.0 ? | ||
r0nny_ | im not sure about this | ||
casue it might create name-collisions | 13:54 | ||
xinming | hmm, perl 6 can choose modules for you if you specify, so,I wanna the module can be co-exist. So that, some module upgrade won't break the whole program which need the module. | 13:55 | |
rafl_ | Some debian users in here? I've got some parrot packages for you that should be tested. | ||
xinming | perl6/modules/module-name/versions/module_files. | ||
r0nny_ | xinming: well - how do i tell it wich version to use ? | 13:57 | |
xinming | rafl_: I am using debian, I'll be glad to be the "laboratory animal" | ||
rafl_ | xinming: Great. Uploading it.. | ||
xinming | r0nny_: hmm, in perl 6, if you just specify the name of the module, choose the latest, if with version number is specified, then, choose that one... | 13:58 | |
r0nny_ | hmm | ||
then it should be no problem | |||
rafl_ | xinming: perlcabal.org/~rafl/debian/ | 13:59 | |
r0nny_ | how does perl manage if u use 2 different modules, and each of them uses a different incombatible version of the same module | ||
xinming | r0nny_: hmm, why I wish to install to a specified location, is because, we might wish to be the project on a different partition, which is filled of projects, and you need `ln -s` all the modules to a project lib dir, maybe, lib/CPAN sometimes | 14:01 | |
rafl_: hold on plz | |||
eric256_ | juerd: ping | ||
xinming | r0nny_: and the modules has a working version specified for the needed module... | 14:02 | |
r0nny_: and prefer to choose the latest if not found. | |||
hmm, Just guess. :-P | 14:03 | ||
parrot_0.3.0-1_i386.deb | |||
Is this one I should try? | |||
rafl_ | parrot-dev and parrot-doc as well. | 14:04 | |
xinming | s/this one/this the one/ | ||
ok | |||
rafl_ | It's not really 0.3.0, btw. It's a current svn snapshot. | ||
Limbic_Region | for anyone interested - perlmonks.org/index.pl?node_id=503589 is a p6 related post on combinatorials and lazy evaluation and what not | 14:05 | |
first p6 specific post I have seen in a while | |||
*hint hint* | |||
xinming | rafl_: hmm, Do I have to upgrade to sid? :-( | 14:06 | |
rafl_ | xinming: I don't know. They're at least built against sid. | 14:07 | |
xinming | rafl_: I can't install it on testing. | ||
got the dependency problem. | 14:08 | ||
rafl_ | xinming: What does it complain about? | ||
xinming | dpkg: dependency problems prevent configuration of parrot: | ||
parrot depends on libicu34; however: | |||
parrot depends on libstdc++6 (>= 4.0.2-3); however: | |||
Version of libstdc++6 on system is 4.0.2-2. | |||
and, If I try to apt-get install libcstdc++6 or the needed module, I will also get the dependency problem | 14:09 | ||
rafl_ | Yes. Nevermind then.. | ||
xinming | rafl_ parrot-dev and parrot-doc as well. | 14:10 | |
xinming s/this one/this the one/ | |||
oops... | |||
root@HomePc:/home/xinming# apt-get install -f | |||
Reading package lists... Done | |||
Building dependency tree... Done | |||
Correcting dependencies... Done | |||
The following extra packages will be installed: | |||
libicu34 | |||
The following packages will be REMOVED: | |||
parrot | |||
The following NEW packages will be installed: | |||
libicu34 | |||
after I `dpkg -i parrot*`, that's what apt-get install -f will show | 14:11 | ||
anyone here can tell me where to find the url for autrijus' journal? | 14:17 | ||
Jooon | somewhere here use.perl.org/~autrijus/journal | 14:18 | |
xinming | Jooon: thanks. | ||
Limbic_Region | there used to be a link on the pugs homepage | ||
but now it links to planet pugs | |||
xinming | Limbic_Region: No, It is removed... | ||
Limbic_Region | errr | ||
xinming | yes. | ||
Limbic_Region | planet perl6 | 14:19 | |
xinming wonders what autrijus do these days. | |||
Limbic_Region: But, I have problem to access there. | |||
Limbic_Region | err - pugs even # I was right the first time | ||
confusing - the link implies pugs but the page says planet perl six | 14:20 | ||
xinming - you using planetsix.perl-foundation.org/ ? | |||
xinming | planet.pugscode.org/ | ||
Limbic_Region | that just forwards to the URL I provided | 14:21 | |
xinming | I can access planetsix.perl-foundation.org/ | ||
But, can't access planet.pugscode.org | |||
Limbic_Region | so you are getting to where you wanted to go | ||
it is a phantom URL anyway | |||
xinming | The Great FireWall in China blocks too many sites. :-/ | 14:22 | |
Limbic_Region | xinming - I don't understand, you are getting to the right place so why not cut out the middle man? | ||
xinming | Limbic_Region: hmm, If I use planet.pugscode.org, I will get the documents contains no data, But If I use planetsix.perl-foundation,org It will opened correctly. | 14:25 | |
rewriting the world, What does rewriting here mean? rewrite most grammars of perl 6? | |||
rafl_ | autrijus: ping | 14:28 | |
Juerd | r0nny_: ? | 14:31 | |
eric256_: pong | |||
eric256_ | hey. is there a timelimit on cron jobs? make smoke, seems to die after 5 minutes, i wasn't sure if the smoke was broken or the server just shut it down | 14:32 | |
Juerd | There is no time limit | ||
eric256_ | okay. /me scans error log | ||
Juerd | Which means that if you're not careful, the processes will overlap | ||
eric256_ | i only run the smoke every 6 hours | 14:33 | |
Juerd | ok | ||
eric256_ | should be okay.. might goto 12 dunno | ||
hehe it took too long and the pugs rebuild swiped the pugs executable out form under it | 14:35 | ||
any ideas? | |||
Limbic_Region | xinming - as I said before planet.pugscode.org redirects to planetsix so there is NO content at planet.pugscode.org | 14:36 | |
forget that link and just go directly to planetsix | |||
xinming - rewriting the world is meant to mean changing the world | |||
xinming | Limbic_Region: Ok. | 14:37 | |
r0nny_ | re | 14:41 | |
eric256_ | Juerd...for now i'm creating a pugs-smoke checkout that will build and smoke every 6 hours and just keep it seperate | ||
r0nny_ | Juerd: wanted to have a short talk about the development steps for the build sys and package manager | 14:42 | |
since it will be a bigger project and i hace no clue where to start since im used to think c++ | 14:47 | ||
well - i will write the basic steps i think will be good to the wiki | 14:49 | ||
sixpan.perl6.nl/sixpan/index.cgi?Si...opmentPlan <- please edit | 14:56 | ||
rafl_ wonders why pugs needs a parrot src dir to build. | 15:19 | ||
Shouldn't it be enough to have the headers? | |||
eric256_ | svn check | 15:20 | |
is svn bot feeling okay today? | |||
hehe. he isn't even present... thats a crime! ;) | |||
rafl_ | eric256_: I don't think so.. | ||
15:20
rafl_ is now known as rafl
15:21
eric256_ is now known as eric256
|
|||
nothingmuch | we should add a new pugs logo | 15:26 | |
www.axis-of-aevil.net/archives/2005...fcare.html | |||
Limbic_Region chuckles | 15:27 | ||
rafl | How to check if program foo is in path in perl5? Does a nice idiom or module exist? | 15:28 | |
Limbic_Region | define path | 15:31 | |
rafl | $ENV{PATH} | ||
Limbic_Region | check to see if which has been ported in the PPT | ||
it has | 15:32 | ||
search.cpan.org/~cwest/ppt-0.14/ | |||
see which | |||
search.cpan.org/~cwest/ppt-0.14/bin/which | |||
rafl - does that do what you want or do you need something else? | 15:34 | ||
sili | ppt makes me sad. | 15:36 | |
Juerd | r0nny_: Sorry, I'm not interested in the packaging thing enough | 15:37 | |
Limbic_Region | because? | ||
sili | because you reusing the code that was written isn't really possible | ||
since each tool is its own script, basically | |||
Juerd | r0nny_: Please discuss your ideas with the people who do care (rafl, kane) | 15:38 | |
sili | and ``... isn't pretty | ||
r0nny_ | Juerd: ok | ||
rafl | r0nny_: I am. There's already some kind of prototype. | ||
r0nny_ | rafl: where ? | ||
eric256 | sili...port ppt to p6 and base them on a nice module ;) | ||
r0nny_ | want to take a look | ||
sili | eric256: i wish i had the time | ||
rafl | r0nny_: On my disk. Let me finish a quick hack and I'll upload it. | ||
Juerd | rafl: Wow, you're quick | 15:39 | |
Limbic_Region | sili - ok but making it available inside a perl module wasn't a goal of the project | ||
theorbtwo | I could have sworn there was a core module for PATH and friends, but I don't see it... | ||
15:39
brother_ is now known as brother
|
|||
sili | Limbic_Region: maybe not, but it would have been so benefitial | 15:40 | |
rafl | Juerd: Well, kane wrote most of it. I only fixed the parts which required some deeper Debian knowledge. | ||
r0nny_ | it required debian knowdegde ? | ||
well - need to go away for some time | 15:41 | ||
rafl | Yes. It uses the .deb format from Debian and currently also the debian tools for that (dpkg, dpkg-deb) because there no perl equivalents for them yet. | ||
r0nny_ | rafl: omg - hope u get a rid of needing to use .deb soon ;P | 15:42 | |
rafl | r0nny_: Why? | ||
r0nny_: It's really mature, flexible and does what we want. | |||
r0nny_ | well - i will discuss about this issue when im back | 15:43 | |
theorbtwo | Ah, File::Spec->path is what I was thinking. | ||
rafl | r0nny_: Of course, preferably in the Wiki. | ||
theorbtwo | Using .deb sounds completely reasonable to me. | ||
rafl | To kane and me as well. | 15:44 | |
Limbic_Region | theorbtwo - does File::Spec->path tell you if a file exists in the path? | ||
if not, I found a module that does | |||
Env::Path has a whence method which will tell you if a file exists in the $ENV{PATH} | |||
rafl | Limbic_Region: I solved it in another way which doesn't require a module. | 15:45 | |
Limbic_Region: I hope which is something we can assume to be there everywhere.. | |||
But I don't think so.. ;-) | |||
Limbic_Region | which can't be assumed to be everywhere | ||
Win32 for instance | 15:46 | ||
rafl | Limbic_Region: OK then. How does windows solve that? | ||
theorbtwo | Windows? Solve something? You must be joking... | ||
rafl | Limbic_Region: Can you take a look at my recent commit to Makefile.PL and make it portable? | ||
Limbic_Region | rafl - 1 question at a time | ||
rafl - windows solves this by erroring out when it isn't in the path | 15:47 | ||
theorbtwo | L~R, File::Spec->path only tells you what $ENV{PATH} is; it doesn't search for things in it. | ||
Limbic_Region | AFAIK, native windows doesn't provide the ability to tell you where in the path - only if it is in the path | ||
theorbtwo | I seem to recall that there is something in inc:: that's supposed to do that, but I think I remember that only because it was buggy. | ||
Limbic_Region | if you want to know the location portably (Win32 and *nix) then use the "whence" method in Env::Path | ||
second question, WRT Makefile.PL - I am not doing any Pugs hacking anymore but my last comment should help solve the problem | 15:48 | ||
rafl goes home. | 15:49 | ||
xinming | ?eval class A { has @.a; sub BUILD { @a = ( "a", "", '' )} }; ?A.new.a[0]; | 15:52 | |
evalbot_7757 | Error: Undeclared variable: "@a" | ||
xinming | .... | ||
Limbic_Region | rafl - I think your change to Makefile.PL is b0rk | 15:53 | |
oh, you went home | |||
hrmm | |||
anyone who is listening and has an opinion - feel free to speak up | 15:54 | ||
apparently knowing the location of a file in $ENV{PATH} is required for something to work correctly | |||
the way to do this on Win32/*nix is to use Env::Path - which probably won't work on Mac, VMS, etc | 15:55 | ||
not that I think Parrot works on VMS yet | |||
anyway - copy/pasting the code from the module is fairly easy as it is compact which wouldn't require adding the module | |||
but it still won't be 100% portable | 15:56 | ||
the code could be adapted to combine the search routine and File::Spec which would be portable | |||
theorbtwo | Sounds like a plan, Limbic. | ||
Does it properly handle extensions under win32? | |||
xinming | how to test if a value is undefined? | 15:57 | |
Limbic_Region | theorbtwo - if File::Spec is providing the file names and paths then it doesn't matter | ||
xinming | just if $a { true } else { false } ? | ||
Limbic_Region | the question is then who is going to adapt the search code ;-) | ||
xinming - defined | |||
?eval my $foo; say "undefined" if ! defined $foo; | 15:58 | ||
evalbot_7758 | undefined bool::true | ||
Limbic_Region | xinming - undefined didn't change behavior from p5 to p6 | ||
theorbtwo | It looks like no, from a quick read of the source (it is, indeed, quite short). | 15:59 | |
Limbic_Region | theorbtwo - so that means you are volunteering right ;-) | ||
theorbtwo | What I meant is that if you're looking for a command called "foo" under win32, you really want foo.{exe,com,bat,cmd,pif}. | ||
Limbic_Region | theorbtwo - that isn't needed for this specific problem | 16:00 | |
and besides - the whence routine allows for patterns not just exact file names | 16:01 | ||
chomp(my $parrot_config = `which parrot-config.imc 2>/dev/null`); | |||
that's the line that needs to be made portable | |||
theorbtwo | Hm. | ||
OK. | 16:02 | ||
Sounds like Env::Path can do that. | |||
Limbic_Region | and that is certainly doable by combining File::Spec and Env::Path | ||
theorbtwo - it could if you didn't care about OSes other than Win32 and *nix | |||
but I think we do | |||
theorbtwo | Oh, right -- the path might not be in $ENV{PATH}. | 16:04 | |
Limbic_Region notes the time and heads for lunch | |||
theorbtwo | Have a good one. | ||
Limbic_Region | actually - I think Perl makes $ENV{PATH} portable - that's not the problem | ||
the problem is searching in $ENV{PATH} | |||
delimiters and extensions and what not | |||
Luna | hola | ||
Limbic_Region | salutations | 16:05 | |
theorbtwo - if you read the source for Env::Path - it is doing some / and \ conversions prior to searching - that needs to be extended beyond Win32 | |||
and *nix | |||
in any acct - I am off for lunch | 16:06 | ||
rafl | Limbic_Region: b0rked in which way? | 16:10 | |
xinming | ?eval "asdfg/" ~~ rx:perl5{^(\w+)?/(\w+)?}; $1; | 16:12 | |
evalbot_7758 | \MkMatch {matchOk = True, matchFrom = 0, matchTo = 0, matchStr = "", matchSubPos = [], matchSubNamed = {}} | ||
xinming | ?eval "asdfg/" ~~ rx:perl5{^(\w+)?/(\w+)?}; $1 ?? "true".say !! "false".say ; | ||
evalbot_7758 | true bool::true | ||
xinming | Is this the right behaviour? | ||
rafl | I don't think so. | 16:14 | |
xinming | It seems, that it returned a null string with `but true` | 16:15 | |
eval? my $a; $a = "abc" ?? "bcd" !! $a = "false"; | 16:19 | ||
?eval my $a; $a = "abc" ?? "bcd" !! $a = "false"; | |||
evalbot_7758 | Error: Can't modify constant item: VStr "bcd" | ||
xinming | how about this? | 16:20 | |
nothingmuch | do roles have no instance data? | 16:24 | |
xinming: ?? !! is lvalue | |||
xinming | ?eval my $a; my $b; $a = "abc" ?? "bcd" !! $b = "false"; | 16:25 | |
evalbot_7758 | Error: Can't modify constant item: VStr "bcd" | ||
xinming | Do we have to write this using if statement? | ||
nothingmuch: ?? | |||
nothingmuch | ?eval my $x = 0; my $y; my $z; $x ?? $y !! $z = "foo"; $y.perl ~ " " ~ $z.perl | ||
evalbot_7758 | "\\undef \\\"foo\"" | ||
nothingmuch | ?eval my $x = 1; my $y; my $z; $x ?? $y !! $z = "foo"; $y.perl ~ " " ~ $z.perl | 16:26 | |
evalbot_7758 | "\\\"foo\" \\undef" | ||
nothingmuch | what you wantes is probably !! ($a = "foo") | ||
xinming | thanks. :-) | 16:28 | |
nothingmuch | what were you trying to do, btw? | 16:30 | |
?eval my $x = 1; my $y; my $z; $x ?? $y !! $z = "foo"; ($y, $z) | |||
evalbot_7758 | (\"foo", \undef) | ||
nothingmuch | that's simpler ;-) | ||
xinming | nothingmuch: hmm, I want to try if we can assign a value using $x ?? true !! false statement instead of if. ;-) | 16:32 | |
theorbtwo | Except the $y is being used in void context for some reason... | ||
nothingmuch | yes, you can | ||
theorbtwo | ?eval my ($x, $y); $x = $y ?? 'true' !! 'false'; $x | ||
evalbot_7758 | \"false" | ||
theorbtwo | ?eval my ($x, $y)=1; $x = $y ?? 'true' !! 'false'; $x | ||
evalbot_7758 | \"false" | ||
xinming | theorbtwo: hmm, I don't mean this. | ||
nothingmuch | oh, that's what you're checking | ||
theorbtwo | ?eval my ($x, $y)=(0,1); $x = $y ?? 'true' !! 'false'; $x | ||
evalbot_7758 | \"true" | ||
nothingmuch | ?eval my ($x, $y) = 1; ($x, $y) | 16:33 | |
evalbot_7758 | (\1, \undef) | ||
nothingmuch | that's expected | ||
xinming | ?eval my( $x, $y ) = 1..Inf; ($x, $y); | ||
evalbot_7758 | (\1.0, \2.0) | ||
xinming | ?eval my( $x, $y ) = 1..Inf by 3; ($x, $y); | ||
evalbot_7758 | Error: unexpected "b" expecting term postfix, operator, "does", "but", "is", "^..^", "cmp", "<=>", "^..", "..^", "..", ";" or end of input | ||
xinming | hmm... | ||
bbl | 16:34 | ||
can we make a step? | |||
Just like, 1, 5, 9, 13, 17, 21 such kind of list? | 16:35 | ||
?eval my( $x, $y ) = 1..Inf :by(3); ($x, $y); | 16:36 | ||
evalbot_7758 | Error: unexpected ":" expecting term postfix, operator, "does", "but", "is", "^..^", "cmp", "<=>", "^..", "..^", "..", ";" or end of input | ||
theorbtwo wonders why it lists "operator" and several individual operators, and why he doesn't remember those ^..^ smiliey things. | 16:37 | ||
ajs_ | In S03 under new operators: The .. binary range operator has variants with ^ on either end to indicate exclusion of that endpoint from the range. | 16:44 | |
So 1^..^3 would be (2) | |||
xinming | amazing... | 16:46 | |
?eval my( $x, $y ) = 1^..^Inf; ($x, $y); | 16:47 | ||
evalbot_7758 | (\2.0, \3.0) | ||
xinming | ?eval my( $x, $y ) = 1..^Inf; ($x, $y); | ||
evalbot_7758 | (\1.0, \2.0) | ||
xinming | ?eval my( $x, $y ) = 1..^..^..Inf; ($x, $y); | ||
evalbot_7758 | Error: unexpected "^" expecting term | ||
nothingmuch | xinming: err, what are you trying to do? | 16:50 | |
xinming | nothingmuch: I wish I will be able to get a step for 1..Inf | 16:51 | |
Just like It will auto generate 1, 5, 9, 13, 17, 21, ... Inf | |||
nothingmuch | ( grep { ($_ - 11) % 4 == 0 } .. Inf ))[1 .. 10) | 16:56 | |
?eval( grep { ($_ - 11) % 4 == 0 } .. Inf ))[1 .. 10] | |||
evalbot_7758 | Error: unexpected ")" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input | ||
nothingmuch | ?eval (grep { ($_ - 11) % 4 == 0 } .. Inf )[1 .. 10] | ||
evalbot_7758 | Error: cannot cast from VCode (MkCode {isMulti = False, subName = "<anon>", subType = SubBlock, subEnv = Just (MkEnv {envContext = CxtItem (mkType "Any"), envLValue = False, envLexical = MkPad (padToList [("$?1",[(<ref>,<ref>)]),("$_",[(<ref>,<ref>)]),("$code",[(<ref>,<ref>)]),("$lang",[(<ref>,<ref>)]),("&?BLOCK_EXIT",[(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>)]),("&break",[(<ref>,<ref>),(<ref>,<ref>)]),("&c | ||
nothingmuch | eep, i'm drunk | ||
?eval grep { ($_ - 1) % 4 == 0 } ... Inf | |||
evalbot_7758 | Error: unexpected "I" expecting operator, ":", ",", term postfix, postfix conditional, postfix loop, postfix iteration, ";" or end of input | ||
nothingmuch | ?eval grep { ($_ - 1) % 4 == 0 } 1 .. Inf | ||
?eval (grep { ($_ - 1) % 4 == 0 } 1 .. Inf)[0 .. 10] | 16:57 | ||
i think i fried it | |||
xinming | you are drunk... | ||
nothingmuch | is grep lazy? | ||
evalbot_7758 | (no output) | ||
nothingmuch | ?eval (grep { ($_ - 1) % 4 == 0 } 1 .. 1000)[0 .. 10] | ||
(not spec grep, current prelude grep) | |||
evalbot_7758 | (no output) | 16:58 | |
[1, 5, 9, 13, 17, 21, 25, 29, 33, 37, 41] | |||
nothingmuch will do a haskell version with fold | 16:59 | ||
no, that doesn't work out well | 17:01 | ||
Limbic_Region | rafl - b0rk in that it isn't portable | 17:07 | |
rafl | Limbic_Region: Yes, indeed. It only works on UNIX platforms. How to do that better? | 17:08 | |
Limbic_Region | you can immediately improve the portability by using 'whence' from Env::Path | ||
that will cover *nix and Win32 | |||
the problem is supporting other platforms | 17:09 | ||
to do that - you have to combine code from 2 modules - File::Spec and Env::Path | |||
that is what I was discussing right before going to lunch | |||
the code from Env::Path is quite small so modification shouldn't be that difficult | 17:10 | ||
and theorbtwo was just about to volunteer | 17:11 | ||
svnbot6 | r7756 | eric256++ | Add doc links for several tests in /t/operators | 17:12 | |
r7757 | eric256++ | Fixed typo in last commit | |||
r7758 | rafl++ | r18808@ata: rafl | 2005-10-28 17:44:31 +0200 | |||
r7758 | rafl++ | * debian/control: Build-Depend on parrot-dev for parrot embedding. | |||
r7758 | rafl++ | * Makefile.PL: Some fixes for parrot header detection. | |||
rafl | Limbic_Region: Commiting another possibility.. | 17:13 | |
Limbic_Region sees what it takes to hack Env::Path's whence and File::Spec to make it work universally | 17:15 | ||
svnbot6 | r7759 | rafl++ | r18815@ata: rafl | 2005-10-28 19:24:35 +0200 | 17:24 | |
r7759 | rafl++ | * Make pugs detection a bit more portable. | |||
r7759 | rafl++ | * Embed parrot in the debian package as default. | |||
Limbic_Region | rafl - I figured out how to make it portable without using Env::Path at all | 17:32 | |
and without invoking a shell | 17:33 | ||
perlbot nopaste | |||
perlbot | Paste your code here and #<channel> will be able to view it: sial.org/pbot/<channel> | ||
pasteling | "Limbic_Region" at 129.33.119.12 pasted "Fix the `which` portability issue" (15 lines, 304B) at sial.org/pbot/13992 | 17:34 | |
Limbic_Region | Do you want me to update Makefile.PL or are you going to? | ||
rafl | Please do it. | ||
Limbic_Region | ok | ||
I just need to know what to do if it isn't found in the PATH? | 17:36 | ||
nevermind - I see an or die later on that should handle it | 17:39 | ||
checking in now | 17:41 | ||
hmm - this is taking a while | |||
svnbot6 | r7760 | iblech++ | PIL2JS: Prelude::JS::Proxy: Unbreak PIL2JS (forgot a +$var to :$var change). | 17:42 | |
Limbic_Region | grrr - update failed | ||
Limbic_Region tries again | |||
rafl - your attempt was also b0rk | 17:45 | ||
PATH delimiter is not portable either - : on *nix and ; on Win32 | |||
I have checked in something that should be portable | |||
please verify it actually works (I don't have Parrot on this machine) | |||
actually - I think it is still b0rk but not because of the parrot-config.imc thing | 17:47 | ||
svnbot6 | r7761 | Limbic_Region++ | Made `which parrot-config.imc` portable | ||
Limbic_Region | (-d $base and -e "$base/parrot-config.imc") isn't portable either | 17:48 | |
that needs fixing too | |||
that whole section is non-portable | |||
Limbic_Region goes to work | |||
svnbot6 | r7762 | Limbic_Region++ | Corrected many more portability issues with parrot-config.imc | 17:52 | |
Limbic_Region | but still not all of them | ||
eric256 | "10 */6 * * * " in crontab should run every y6 hours at 10 after the hour right? | ||
Limbic_Region | the more I look the more problems there are - is it because no one is running Pugs AND Parrot on anything other than *nix? | 17:53 | |
eric256 - huh? | 17:57 | ||
svnbot6 | r7763 | Limbic_Region++ | Corrected copy/paste error in adding portability | ||
Limbic_Region | there are more unix only style paths in the PUGS_EMBED section but I am not sure why so I am not touching them WRT portability - anyone know what is going on? | 18:00 | |
eric256 wants to smack PerlMouse | 18:10 | ||
webmind | evening | 18:54 | |
r0nny_ | re | 19:04 | |
eric256 | yea automated feather smoke worked ;) | ||
r0nny_ | rafl: ping? | ||
rafl: ping2? | 19:16 | ||
rafl: rofl - i had read how xactly deb files are made - dey definitive dont fit sixpan well - since all it needs is a tarball (preferable bz2 compressed) | 19:31 | ||
Limbic_Region | eric256 - why do you want to smake PerlMouse? | 19:32 | |
eric256 | smack...and i dunno. i think the wording he uses just grates on my nerves. | 19:38 | |
haven't you ever had someone you just didn't like and couldn't point to a reason. ;) | |||
theorbtwo wonders who this perlmouse person is. | 19:40 | ||
Limbic_Region | eric256 - I had my own run in with PerlMouse | ||
eric256 | hehe. just someone on perlmonks. | ||
Limbic_Region | PerlMouse ended up being right but for the wrong reason | 19:41 | |
which kinda made the burn worse | |||
fwiw - lazy evaluation also has the benefit of memory consumption | |||
s/memory/efficient memory/ | |||
eric256 | i don't mind being wrong. i don't mind being proven wrong, or argueing about right and wrong. with me it is more a tone issue, i need to focus on remembering that most people (including myself) arn't good at conveying emotions well in text. | ||
Limbic_Region | well, PerlMouse never made their case - japhy ended up doing it | 19:42 | |
eric256 | a guy on pm awhile back ticked me off, and i decided to private message him about it, turned out english wasn't his first language and he had no idea that his tone held negative connotations. | ||
Limbic_Region | I had shown that while I was right, I wasn't as right as I could have been | 19:43 | |
fwiw - never under-estimate the cost of compiling regular expressions un-necessarily | |||
theorbtwo | Though for longer-running programs that tends to get amoritized out. | 19:44 | |
Oh. Unless I'm forgetting how often regexes get compiled, which is quite possible. | 19:45 | ||
Limbic_Region | theorbtwo - not when they are in a loop that IS the body of a long running program | ||
eric256 is still waiting for theorbtwo to help cross reference tests. ;) i'm doing a couple test files a day but its time consuming ;) | 19:46 | ||
Limbic_Region | I followed the general rule - when there is a performance problem fix the algorithm and don't spend as much time on shaving individual ops | ||
PerlJam | Limbic_Region: you were just guilty of a lapse in critical thinking. You had tunnel vision. No big deal; it happens. | ||
Limbic_Region | in this case, the shaving outweighed changing the algorithm | ||
and doing both outweighed either combined ;-) | 19:47 | ||
theorbtwo | Rules of thumb are useful things, but don't forget about the rest of your fingers! | ||
r0nny_ | rafl: ping agin | 19:52 | |
Limbic_Region | so does anyone have any insight into the non-portability of the EMBED_PUGS stuff in Makefile.PL ? | 19:53 | |
I helped rafl out earlier with a specific issue and found that there was a whole lot of non-portable code | |||
I started fixing it and then took a step back and wondered why it was that way (and stopped) | |||
theorbtwo - fwiw, Env::Path was entirely unnecessary | 19:54 | ||
20:51
buu is now known as o
20:52
o is now known as buu
|
|||
meppl | gute nacht | 22:16 | |
nothingmuch | good night | ||
meppl | ;) | 22:17 | |
Juerd | ja, goede nacht | ||
meppl | ;) | ||
nothingmuch goes to play violent computer games | 22:30 | ||
rafl | r0nny_: pong | 23:07 | |
r0nny_ | re | 23:30 | |
rafl: ping? | |||
rafl | r0nny_: pong, again. | 23:32 | |
r0nny_ | rafl: took a look at the wiki ? | 23:33 | |
os - i see | 23:34 | ||
i know a tar file could contain the same scripts | 23:35 | ||
rafl | The tools to extract/install debs and especially to extract meta data etc already exist. | ||
It doesn't matter if it's a tarball or an ar achive. | 23:36 | ||
r0nny_ | its an ar archive of multiple tarballs | ||
theorbtwo | And a magic file. | ||
However, like rafl said, it doesn't matter. | 23:37 | ||
That's nice to know if you need to deal with things on a low level, or without the proper tools. | |||
rafl | A deb can contain arbitrary data, btw. Not only binary packages. You can also put a src package in with the according scripts to build it when installing. | ||
theorbtwo | Most of the time, you don't. | ||
r0nny_ | well - cpan is about source installs - i dont see the need of all the stuff deb brings in | 23:38 | |
rafl | Most src packages on cpan only contain pure perl stuff. Let's say 80%. Why build packages that must be built to contain the same files afterwards? | ||
theorbtwo | It brings in a well-defined set of metadata, with years of careful evolution. | 23:39 | |
r0nny_ | well if u just install source modules - just make it copy | ||
but i think perl6 will have binary modules | 23:40 | ||
where a build process is usefull | |||
rafl | It is, for 10% of all packages. Let's optimize for the default case and provide src pkg's for the rest. | 23:41 | |
theorbtwo | I think this is a really bad place to try the 90/10 game. | 23:42 | |
r0nny_ | i want to compile all perl modules i install | ||
with specieal optinations just for my system | |||
rafl | r0nny_: Feel free to do. I don't want to. | ||
r0nny_: Yeah.. --omg-optimized ;-) | |||
theorbtwo | r0nny_: I bet you use gentoo, too. | 23:43 | |
rafl | theorbtwo: I didn't say we won't support src packages. I think it should be the default to provide binary packages for pure perl stuff and fall back to src packages when it's needed. | ||
theorbtwo | Compiling perl modules with special optimizitaions just for you system doesn't make any freaking sense. | ||
r0nny_ | rafl: i just want to ensure the tools dont get to debinaish, casue the debian from shouce stuff feels horrible | ||
eh source | 23:44 | ||
theorbtwo | r0nny_: The second half of that sentance made no sense to me. | ||
rafl | r0nny_: Debian is optimized for binary packages. We only adopt the package format. | ||
r0nny_ | even the package fromat is optimized for binary ... | ||
they put stuff like archs in | |||
theorbtwo | There are two completely different formats in deb -- one for binaries and one for source. | ||
rafl | r0nny_: It isn't. | 23:45 | |
theorbtwo | Source ones aren't .deb at all -- they are .tar.gz, patch.gz, and .dsc. | ||
rafl | r0nny_: Architecure is just one kind of meta information in the package. Evaluating and using it is left to the package management. | ||
OK, I'll go to bed now. Let's discuss tomorrow or in the wiki. Bye! | |||
r0nny_ | ok | 23:46 | |
n8 then | 23:47 | ||
need to sleep. too | |||
theorbtwo | "nate then"? | ||
r0nny_ | night ! | ||
theorbtwo | Damn l33t idiots... | 23:48 |