pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, elf: etc.) (or perl6: for all) | irclog: irc.pugscode.org/
Set by Tene on 29 July 2008.
wayland76 Yeah, the question yesterday about whether "handles *" was in relation to delegation 02:24
alldrhwjje_ So, is Perl 6 done yet? :-) 07:25
Just kidding, love rakudo
No one home here? Hellooooooo? 07:26
moritz_ nobodey here, no 07:28
wayland76 not even me :) 07:57
meppl good morning 08:55
wayland76 do you mean that the morning is good whether I want it to be or not, or that it is a morning to be good on, or ... (actually, it's evening here) 09:00
anyway, good morning :)
Btw, does anyone know how we congratulate chromatic on getting all the Rakudo meeting notes up on the rakudo website? :) 09:01
pmurias wayland76: i think it expresses the wish that this morning will be a good one for you (similarly as "good luck") 09:27
wayland76 pmurias_: Yeah, I'm misquoting Gandalf at the beginning of the Hobbit :) 11:12
Googled, and here's what I got: "What do you mean good morning? Do you wish me a good morning, or mean that it is a good morning whether I want it or not; or that you feel good this morning; or that it is a morning to be good on?... Or, now you mean that you want to get rid of me, and that it won't be good till I move off?" 11:13
anyway, bedtime here. 'night all :) 11:14
pmurias @tell ruoso what do you thing of droping the RootNamespace in s1p and using the lexical prelude in it's place? 13:57
lambdabot Consider it noted.
pugs_svn r22638 | pmurias++ | [smop] 14:29
r22638 | pmurias++ | added lexical prelude - using $OUT is lexical HACK
r22638 | pmurias++ | corrected test count in 19_array
r22638 | pmurias++ | 21_code uses lexical prelude
r22639 | pmurias++ | [smop] tests for mildew now live in mildew/t 14:39
pugs_svn r22640 | pmurias++ | [m0ld] m0ld executable now uses the M0ld module 15:01
ruoso Hello! 18:37
lambdabot ruoso: You have 1 new message. '/msg lambdabot @messages' to read it.
ruoso Lexical prelude is the way to go, I'm aware of it... 18:37
pmurias, I was still unsure about if the prelude should be populated by src-s1p already 18:38
ruoso pmurias, but RootNamespace seems odd anyway 18:39
azawawi hi 18:41
@seen moritz_ 18:42
lambdabot moritz_ is in #perl6. I last heard moritz_ speak 11h 13m 48s ago.
pmurias ruoso: hi 18:43
ruoso hi pmurias
pmurias we can always overwrite the s1p stuff in the lexical prelude once better stuff is avalible 18:44
one thing i'm not sure how to solve is that in order for ./m0ld to be able to execute .m0ld code it needs to be linked with smop and smop requires ./m0ld to compile 18:46
ruoso that's bootstrap :) 18:48
pmurias, we need to start delimiting the barrier on the minimal smop feature-set 18:49
and the additional libraries
pmurias it wouldn't help in this case 18:50
ruoso pmurias, ok... maybe I'm not seeing the real problem... please ellaborate
pmurias ./m0ld compiles m0ld to c 18:51
and i want it (have it in my working copy) to also be able to execute .m0ld withought a c compiler 18:52
* without
but for this it needs to have a linked in smop (which needs preprocessing by ./m0ld) 18:53
ruoso we need a 1st-stage m0ld compiler
that doesn't depend on smop
pmurias i could either split it in to executable or add a flag and compile ./m0ld 2 times
s/to executable/2 executables 18:54
ruoso yeah... exactly
ruoso you have a build option FIRST_STAGE that doesn't link to smop 18:54
and after that you can rebuild without that option
pmurias doing it 18:56
ruoso pmurias, btw... you said mildew was blocking on runtime support... 19:01
pmurias having a convenient way to feed .m0ld into smop is required 19:07
ruoso you mean like a bytecode
pmurias well a ./m0ld --exec option will be enough for now
ruoso pmurias, ah... I see... so that mildew can execute directly 19:08
pmurias instead of feeding into the smop build system, yes 19:09
ruoso right... so what you're doing right now solves it
pmurias yes
mildew will block on small stuff such as &say, infix:<~> if we want to run even a small part of the test suit 19:10
ruoso pmurias, I was considering writing that in Perl 6 already 19:11
ruoso pmurias, and that's why I want to have multis soon 19:12
pmurias and it mildew was blocking on the lexical prelude 19:13
ruoso oh yaeh... 19:14
azawawi does STD.pm parse any of t/spec tests? 19:52
azawawi anyone? 19:53
pmurias azawawi: it parses *all* of t/spec tests 19:54
Firelord hey everyone 19:54
masak hello, Firelord.
lambdabot masak: You have 1 new message. '/msg lambdabot @messages' to read it.
masak ooh, messages. :D
Firelord heh
masak TimToady: I dunno, for some reason my fudge line didn't bite. I didn't play around with it long enough to figger it out. 19:55
Firelord hmmm someone already registered Firelord 19:56
azawawi pmurias: im using this command './try5 comp_unit ~/pugs/t/spec/S29-num/abs.t' and it silently dies with an empty try5.err
masak Firelord: what about Overlord?
Firelord anyone got any suggestion for a nick?
Firelord that could work 19:57
masak ponders a set of scripts that'll make it easier to fudge and run Rakudo tests without committing them to the Pugs repo 19:57
masak Overlord: I, for one, welcome you! 19:59
Overlord hmm Overlord is already taken
Cannot keep it
masak figures. 19:59
Guest34606 ya
pugs_svn r22641 | pmurias++ | [mildew] uses the lexical prelude
r22641 | pmurias++ | [m0ld] --exec option if compiled with ./Setup configure --flags=SMOP
Hybrid Wow, that's registered as well 20:01
masak this is a popular IRC server
pmurias Hybrid: you can always try the intitialLetterSurname nick
masak Hybrid: or just the surname, if yours is unique enough. 20:02
ruoso 's is
masak seems mine isn't, but I was lucky, and got here before any japanese with the same name.
ruoso masak, I think that's the definition of "enough" :) 20:03
masak you might be right :)
OTOH, now it isn't unique enough any more. 20:04
[particle]2 masak: you can add them to t/localtest.data and run make localtest
masak [particle]2: ooh.
masak that's nice to know, thanks. 20:04
[particle] i use it frequently
like, as i was developing the 'is export()' tests
pmurias can anyone suggest a build system that isn't a pile of strange syntax and doesn't get in the way when trying to do something moderatly complex? 20:05
ruoso pmurias, how do I run './Setup configure --flags=SMOP'... I couldn't find "Setup" inside m0ld
masak pmurias: ant seems pretty nice.
ruoso noooooooooooooooooooooooooooooooooooooooooo
masak don't know if you count XML as "strange"
ruoso: oh, ok. :)
ruoso please... I've suffered enough on ant's hand... 20:06
pmurias the syntax can be strange the semantics shouldn't be
masak all the semantics I want from a build system is Directed Acyclic Graphs.
pmurias i just want to define custom rules 20:07
ruoso: ghc --make Setup.lhs 20:08
azawawi i think './try5 comp_unit any_file.pl' is currently broken...
ruoso pmurias, hmmm... it says '--flags=SMOP' is not recognized... 20:10
pmurias, about the build system... don't feel bad... very few projects deal with that many specifics 20:11
ruoso masak, you want the build system to understand what is to build a C shared library, what is to build custom things, what it is to build haskell 20:12
actually, not you, we want that :)
masak :)
azawawi what am i doing wrong with try5? 20:15
try5 -e works
but try5 inputfile doesnt
pmurias azawawi: use tryfile 20:16
ruoso: ./Setup configure --flags=SMOP doesn't work?
ruoso pmurias, sorry... dumbness on my part... forgot the "configure" in between 20:17
pmurias, but... Could not find module `M0ld.Eval'
azawawi pmurias: thx
pmurias sorry, dumbness on my part here 20:18
pugs_svn r22642 | pmurias++ | [m0ld] added missing file
pmurias ruoso: scons would have worked, but i just chose cmake-- without looking at the alternatives 20:20
ruoso as long as it isn't ant :) 20:21
btw... pmurias++ # m0ld --exec is soooo cool
ruoso pmurias, I keep receiving unable to fetch <SMOP__S1P__LexicalPrelude> 20:24
pmurias i get even stranger errors 20:28
fixing it 20:30
ruoso: all the ¢ vars need to be harcoded in src/haskel_ffi.c as the dlopen hack didn't work for reasons beyond my understanding (and error message was broken to obscure stuff) 20:46
ruoso pmurias, but that var is there 20:47
and it doesn't find it
ruoso .oO( is that a null-terminated-string problem? )
pugs_svn r22643 | pmurias++ | [smop] hardcoded some more c variables
pmurias ruoso: it's the printf("%s") problem ;) 20:48
ruoso pmurias++ # m0ld --exec is very very much cool 20:50
so we can call mildew --file something.pm | m0ld --exec 20:51
pmurias yes
ruoso amazinh
amazing
pmurias so m0ld --exec works for you now? 20:52
ruoso yes... we still need to add all ¢ variables to haskel_ffi
but it works just fine 20:53
m0ld/dist/build/m0ld/m0ld --exec < m0ld/hello_world.m0ld 20:54
impressive
pmurias, do you think you could add to cmake the rebuild of m0ld after the build of libsmop? 20:55
with the proper flag, I mean
pmurias failed at it, what i'm now trying to do is to make smop a shared library
ruoso ok 20:56
ruoso commuting &
pugs_svn r22644 | pmurias++ | [smop] libsmop is now a shared library (LD_LIBRARY_PATCH required when using m0ld with smop embedding) 21:19

Powered by Cro and the Raku® Programming Language.

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

Thank you!