Parrot 1.8.0 Zygodactyly released | Latest modified TT's: icanhaz.com/parrotbugs | Parrot Languages: icanhaz.com/parrotlang
Set by moderator on 23 November 2009.
Coke svn cp => git ??? 00:21
davidfetter leto? 00:23
purl leto is Leopold Toetsch
davidfetter hrm
anybody seen leto lately?
Coke no, that's leo.
no, leto is also dukeleto
purl okay, Coke.
Coke leo? 00:24
purl leo is possibly leo lapworth eq ranguard or one of parrot's older contributers. or Leopold Toetsch
davidfetter i'm pretty sure leopold toetsch *isn't* leto 00:25
Coke davidfetter: yes, that's what I just said. =-) 00:26
davidfetter is there a way to tell purl about that? 00:27
Coke ... I just did?
dalek kudo: 7ce13d8 | (Solomon Foster)++ | build/Makefile.in:
Change master spectest to use t/spec revision 29188.
Coke leto?
purl leto is Leopold Toetsch or dukeleto
Coke no, leto is <reply>See dukeleto 00:28
purl okay, Coke.
Coke (no, purl syntax fooled me.)
ls
davidfetter leto? 00:29
purl See dukeleto
davidfetter excellent!
purl zwooshes
00:33 mikehh joined
mikehh All tests PASS (pre/post-config, smoke (#30243), fulltest) at r42820 - Ubuntu 9.10 amd64 (g++ with --optimize) 00:37
Tene purl: seen pmichaud? 00:41
purl pmichaud was last seen on purl 8 hours, 14 minutes and 45 seconds ago, saying: <private message>
Coke yah, but what was the private message? 00:48
Tene purl: msg fperrad If you're going to set +x on something, please give it a useful shebang line 00:49
purl Message for fperrad stored.
00:50 vad joined
vad hi, I have some "parrot-to-c" question. 00:51
Tene vad: Go ahead and ask them. I'll see if I can answer them, or maybe other people are around. :)
vad ... Actually I just want to pass some obj** to some function... 00:52
.... and I am a bit lost
Tene you're calling your C function from Parrot?
vad Tene, thanks for your time. yes, I'm calling C from parrot 00:53
... and (still typing:) ...
... and, say, I have 5 objects of type obj* which I collect from elsewhere, and then I want to pass a stack of them into some 00:54
... C function.
Actually I already have connection to many C funciton in library of interest, but this obj** makes just hard...
I've tried to create an UnManagedStruct containing 1 field repeated N times, but I can't fill it with values. 00:56
Tene I'm afraid I can't quite help... :( 00:57
vad Actually more generic question - maybe my Q is too large for IRC, and, given that parrot-dev is for developers, and my question is like user question, is there exists a parrot for users ML?
Tene Yes, there is parrot-users.
vad ok, maybe that's the right place, then...
Tene lists.parrot.org/mailman/listinfo/parrot-users
Coke parrot-users?
purl parrot-users is starting to get lots of talk too, which is cool to see
Coke no, parrot-users is lists.parrot.org/mailman/listinfo/parrot-users 00:58
purl okay, Coke.
Tene if you can put together some example code and post to parrot-users, I'll try to look at it tonight.
vad ok, thanks to all, I'll subscribe to that list.
Coke does nqp have labels? 01:05
(and goto?)
(or, does while() in nqp support break/continue control exceptions? 01:06
Tene Coke: loops in nqp support break/continue, but nqp doesn't have builtins to throw the right exceptions. 01:10
so you have to use inline pir to do it.
nqp does not have labels or goto.
Coke Tene: what are the "right" exceptions?
CONTROL_BREAK;CONTROL_CONTINUE ? 01:11
Tene CONTROL_BREAK/CONTROL_CONTINUE
Coke =-)
Tene Yes.
Coke ok. doesn't seem to work.
if I have no CONTROL in my while loop, and throw a break, I get "No exception handler and no message 01:12
Tene Oh? Really?
Gimme 5m
Coke I just pushed some stuff for partcl-nqp; you can duplicate my problem with: 01:13
dalek rtcl-nqp: 22d94dc | (Will Coleda)++ | src/PmTcl/Commands.pm:
add initial [unset]
rtcl-nqp: 612d63f | (Will Coleda)++ | TODO:
update TODO
rtcl-nqp: cc2b053 | (Will Coleda)++ | src/PmTcl/Commands.pm:
Add [break] and [continue]
nopaste "coke" at 72.228.52.192 pasted "for tene++" (18 lines) at nopaste.snit.ch/18894
01:16 cconstantine joined
Tene Coke: I lied. They handle CONTROL_LOOP_{NEXT,LAST,REDO} 01:18
I don't know tcl semantics enough to know if break/continue are loop_last and loop_next or not. 01:21
nopaste "tene" at 24.10.252.130 pasted "loops example for Coke++" (25 lines) at nopaste.snit.ch/18895 01:22
Coke why did we duplicate those? *sigh* 01:46
ah well. guess I'll ditch my use of CONTROL_BREAK and CONTROL_CONTINUE. 01:47
tene++
dukeleto What does this actually mean?: add_attribute() not implemented in class 'Tapir;Stream' 02:21
i though i could add an attribute to any class?
Tene dukeleto: are you trying to call that on the class object or on an instance of that class? It looks like the latter... 02:22
dukeleto Tene: instance 02:27
purl instance is probably a blessed reference to a thing or called an "object"
Tene dukeleto: add-attribute is a method on the class.
dukeleto Tene: if i do it on the class I get "Modifications to classes are not allowed after instantiation.
Tene dukeleto: that's right. add_attribute doesn't work after you've instantiated the class. 02:28
dukeleto Tene: do i have to do it to the class in an _initialize?
Tene: i want to add attributes as early as possible, not runtime 02:29
Tene dukeleto: you just have to do it before instantiating the class.
dukeleto Tene: what is the syntax? 02:30
purl it has been said that the syntax is not experimental.
Tene dukeleto: Eh?
class.'add_attribute'('foo')
dukeleto i am using the addattribute opcode 02:32
and i keep getting "Modifications to classes are not allowed after instantiation."
02:32 joeri joined
Tene dukeleto: I think I must be missing something you're saying... you can't add attributes after instantiating objects in that class, just as the error says. You have to do it before instantiating any objects in that class. 02:33
dukeleto :load is not working for me 02:38
Tene curious.
purl gives the small curious key to Bilbo. Thorin sits down and starts singing about gold.
dukeleto a .sub marked with :load should be run before other .sub's, right?
Coke only if you're loading it.
Tene you might also want :init 02:39
Coke if you're running the file directly, then you want :init.
Tene Coke++
Coke (or, perhaps :immediate, depending)
dukeleto Coke: i am creating an object in the class the file describes. I wanted :init, thanks! 02:40
Coke tene++
Tene Coke++
Coke karma "will coleda"
dukeleto tene++
purl "will coleda" has neutral karma
Coke karma coleda
purl coleda has neutral karma
Coke hurm.
dalek TT #1339 created by coke++: eliminate redundant exceptions CONTROL_CONTINUE and CONTROL_BREAK
02:41 Essobi joined
Tene Coke: in Perl 6, at least, break is different from last 02:41
Coke ... how?
Tene break is used to exit the innermost containing construct that sets $_ 02:42
like a 'given' block.
and 'continue' is used to fall out of a 'when' block without throwing a 'break' exception. 02:43
Coke "fall out" ?
dukeleto Tene,Coke: Thanks guys, you got me unstuck!
Tene Perl 6 'when' blocks end with an implied 'break'
Coke (makes it sound like break again.)]
Tene: I think your break is not my break.
Tene so: when 1 { do stuff; } default { do more stuff; } 02:44
the 'default' doesn't get executed.
but: when 1 { do stuff; continue; } default { do more stuff; }
the default block does get executed
Coke that continue is also not like my continue.
Tene So the tcl continue/break are the Perl 6 next/last 02:45
Coke so you can keep break and continue, I suppose, but the way you describe them there are useless to me.
just annoying that I "reserved" them and they were co-opted. :P
Tene It's not the first time that things have different names in tcl and perl 6
Coke be nice if these were documented. =-)
Tene Where would you expect to find that?
Coke runtime/parrot/include/except_types.pasm 02:46
Tene Oh! I know! I'll document it in the ticket rejection!
Coke or perhaps in the C enum that generated it.
Tene that should be sufficient, right?
Coke but the .pasm is the file I'm actually using.
necessary but not sufficient.
document it, leave the ticket open as a "needs docs before closing"
thanks for the clarification. 02:47
Tene ... how do I do that?
I see "invalid"...
Coke just don't close the ticket. =-)
perhaps set the component to 'docs' 02:48
Tene what address do I want to cc?
Coke none.
Tene Okay.
I just pasted from irc into a comment.
and.. formatting fail.
Tene trac fail. 02:49
dalek rtcl-nqp: 3dea1b5 | (Will Coleda)++ | src/PmTcl/Commands.pm:
Avoid CONTROL_BREAK/CONTINUE - these have been co-opted by the perl6 folks

   (See parrot's TT #1139) - Advantage? we don't have to explicitly deal with
   these exceptions in the nqp version.		 02:56
Tene Coke: fwiw, the reason I co-opted those and added the CONTROL_LOOP exceptions was exactly because they weren't documented. ;) 02:57
Coke water under the bridge. 02:58
Tene I was just entertained by that in the context of your request for docs.
Coke Hoist by my own petard. 02:59
bah. converting all this stuff over to NQP is so far, not a huge win for me. 03:08
It's like the 5th major rewrite I've done since '02. :|
(stable my ass) 03:09
Coke completes his rant and returns you to our regularly scheduled channel. 03:13
pmichaud hello 03:17
dalek rtcl-nqp: 2802664 | (Will Coleda)++ | build/Makefile.in:
found another passing test.
 03:19
rtcl-nqp: aa315ca | (Will Coleda)++ | TODO:
some updates on the task list.
pmichaud fwiw: irclog.perlgeek.de/parrot/2008-12-16#i_765095 # already anticipated the naming issue almost a year ago :) 03:20
03:22 chromatic joined
Tene hi pmichaud! I've been excited for you to see my macro POC in steme. 03:25
pmichaud ...macro? 03:26
Tene 'sec
pmichaud POC?
purl POC is this chat, which is running under same in vmware or proof of concept. or piece of crap
Tene github.com/tene/steme/blob/master/t/05-macro.t
pmichaud oh, porof 03:27
okay
Tene Proof of concept
That test passes in steme.
pmichaud interesting.... the first part is a grammar rule, second part is an action method?
Tene Yes.
pmichaud clever.
Tene So I just need a more scheme-like language for expressing those, and it'll look a bit more natural. :) 03:28
Which is the easier part.
(thanks to PCT)
pmichaud so, do you have steme working on the new nqp or .. ? 03:30
Tene Yes, I do. 03:31
pmichaud nice.
Tene The first hunk of github.com/tene/steme/commit/62d78b...5bc9187baf shows the implementation of 'macro'
pmichaud 03:08 <Coke> bah. converting all this stuff over to NQP is so far, not a huge win for me. 03:32
...even though more people are able to contribute now?
Coke pmichaud: "so far"
purl somebody said "so far" was not too far in Hungary ;)
pmichaud :)
Coke sure, more people can contribute. but this version is very very very incomplete compared to the other one. 03:33
(sadly, I am no longer one of the people who can contribute well. =-)
pmichaud well, I'm not sure how to jumpstart the degree of completion, alas 03:34
dalek rtcl-nqp: 4c699d5 | (Will Coleda)++ | (2 files):
The only bit of status we had reported so far is now built into 'make test'
 03:36
Coke karma (Will Coleda) 03:37
purl (will coleda) has neutral karma
Coke karma Coleda)
purl coleda) has neutral karma
dalek rtcl-nqp: 9b30569 | (Will Coleda)++ | CREDITS:
Attempt to fix karma matching for dalek bot in #parrot
 03:42
03:42 cconstantine joined
dalek rtcl-nqp: 7c5a26d | (Will Coleda)++ | TODO:
Throw some more on the pile.
 03:47
rtcl-nqp: 6718aef | (Will Coleda)++ | src/PmTcl/Commands.pm:
fix [proc]'s return value.
rtcl-nqp: 7948a6f | (Will Coleda)++ | src/PmTcl/Commands.pm:
coerce to string for string concat.
Coke pmichaud: well, certainly the effort to test nqp-rx wouldn't have gone anywhere without you. 03:48
test *the conversion to, that is.
pmichaud what sort of documentation is needed for "how this version works"? 03:49
I was working on that a bit over the last couple of days... 03:50
Coke how do we go from tcl down to pbc? pointers to existing docs fine. references to any existing docs very helpful. 03:51
pmichaud I'll add --target=pbc to HLL::Compiler tomorrow
then it'll just be pmtcl --target=pbc xyz.tcl 03:52
Coke that doesn't help much, given how much happens at runtime (more)
... and I meant more generally; we have this tcl program: what stages does it go through on the way to being run in the vm?
pmichaud same as most programs (more)
it's parsed using the grammar, as it's being parsed it produces an ast 03:53
the ast is then compiled to PIR
Coke (most programs) a pointer to existing docs on how this works is fine. I'm not sure those exist for someone who isn't already familiar with how it works.
pmichaud the PIR is then compiled to PBC and executed
there aren't any up-to-date existing docs, no. 03:54
the closest would be the squaak tutorial, I guess. 03:55
I can spend some time writing up-to-date description of how it works for "most programs", which would generally apply to partcl-nqp
Coke I think having something more general is better, yah. 03:56
pmichaud anyway, yes, at the moment much of the compilation happens at runtime 03:57
for constant proc strings, we can get that to happen at compile-time though. 03:58
Coke except when we can't. =-)
pmichaud when can't we?
Coke when the body of the proc contains a syntax error.
pmichaud oh, that's no problem.
Coke right now, that's blocking programs from running that should. (overeager.... +1, then. =-)
pmichaud okay, I don't understand then. 03:59
Coke I /think/ there are cases now where we're throwing a syntax error about a block because we're treating it as code too early.
(though this could just be a problem with wrong exception types.) 04:00
I haven't diagnosed it fully yet, or it'd be in the TODO.
let me find an example...
pmichaud afaik, nothing gets treated as code until it's executed.
(currently)
which would be higher priority... documentation, or ARE? 04:01
Coke sorry, s/as code/as a list/
Tene pmichaud: is :s lexically scoped inside of []s in nqp? 04:02
pmichaud Tene: should be.
Coke look at t/cmd_list.t , e.g.
(hurm. that may be just be a parse error.)
nevermind. I'll put concrete issues in TODO when I think I know what they are.
pmichaud rebuilds partcl-nqp to test 04:03
yes, that looks like a parse error to me. 04:04
Coke I imagine docs would be higher priority if the goal is to get more people able to hack on partcl. (I didn't really have that as a goal, though. I figured we had already saturated any potential contributors)
pmichaud Coke: I'm following whatever goals you have for partcl, primarily :)
is [list "} {"] {\\}\\ \\{} {braces with spaces} 04:05
....how the heck is that supposed to parse?! 04:06
ahhhhh
Coke [list "} {"] generates a single element list.
pmichaud "(however, if an open brace or close brace within the word is quoted with a backslash then it is not counted in locating the matching close brace)" <-- I didn't see this part when I read the first time
fixing.
Coke whee.
(goals) too much stuff missing atm for any goals. I don't know how I'm going to get it back to where it was, let alone take it further. 04:07
so I've just been slapping stuff in TODO as I come across it and either cannot fix it or think it might be doable for someone else.
dalek rtcl-nqp: 7168d84 | pmichaud++ | src/PmTcl/ (2 files):
Fix \\{ and \\} inside of braced words.
 04:10
rtcl-nqp: d60c099 | (Will Coleda)++ | src/PmTcl/Commands.pm:
add args-handling to [proc]
rtcl-nqp: 94f2ae9 | pmichaud++ | :
Merge branch 'master' of git@github.com:partcl/partcl-nqp
pmichaud latest push fixes \\{ and \\} in braced words.... t/cmd_list.t parses now. 04:11
04:11 lucian joined
Coke danke. 04:12
04:12 zak_ joined
pmichaud I'll add a few more backslash escapes, too. 04:15
Coke if it's easier to reuse standard p5 regex for now, instead of rolling our own, I would say let's do that, since we cannot test the ARE until we're can run tcltest.tcl again.
pmichaud it's the same amount of work either way. for partcl, rolling our own is simpler. 04:16
Coke ok
pmichaud++ 04:18
# zzz
04:19 theory joined, s1n joined
dalek rtcl-nqp: 6cfa1cf | pmichaud++ | src/PmTcl/ (2 files):
Add a few more backslash sequences: \\a \\b \\t \\r \\v \\f .
 04:27
rtcl-nqp: e784606 | pmichaud++ | src/PmTcl/ (2 files):
Add \\x and \\o backslash substitutions.
 04:34
rtcl-nqp: 6a74cf4 | pmichaud++ | src/PmTcl/ (2 files):
Add \\u backslash substitution.
04:38 theory joined
dalek rtcl-nqp: 343feaf | pmichaud++ | src/PmTcl/Grammar.pm:
Allow leading whitespace in expressions.
 04:40
rtcl-nqp: db66abf | pmichaud++ | TODO:
Update TODO with a fix message.
04:47 davidfetter joined 04:51 patspam joined 05:02 JimmyZ joined 05:23 nopaste joined 06:26 rhr joined
Coke msg pmichaud looks like 'backx' should not be doing a substr - doesn't the $<x> already contain just the digits? 06:36
purl Message for pmichaud stored.
Coke (trying to get t/tcl_backslash.t to work)
whoops. you were mostly right, but it has to deal with the case when the length is < 2. =-) 06:44
pmichaud++ 06:49
dalek rtcl-nqp: 3b259e5 | (Will Coleda)++ | src/PmTcl/Commands.pm:
add args handling to [append]
 06:56
rtcl-nqp: 750eea8 | (Will Coleda)++ | src/PmTcl/Commands.pm:
add args handling to eval
rtcl-nqp: 3205dc7 | (Will Coleda)++ | src/PmTcl/Actions.pm:
When converting \\xNNN, don't try to take more digits than we have.
rtcl-nqp: f4076cb | (Will Coleda)++ | TODO:
Add a new blocker.
 06:57
Coke wonders why the CREDITS is still not translating Will Coleda to coke.
(maybe it is pulling from just parrot after all.) 06:58
cotto apparently generating a 12G pprof and filling up the disk is a condition the profiling runcore doesn't handle gracefully. 07:18
Meh. I should be encapsulating that anyway. 07:21
dukeleto 'ello 07:24
cotto hi 07:33
purl privet, cotto.
cotto Mmmm. purl sounds all foreign and sophisticated.
It's a thin and obvious veneer, but I'll take it.
07:34 plobsing joined
dukeleto cotto: hola 07:35
purl que tal, dukeleto.
dukeleto plobsing: howdy
dukeleto is hacking on a pure-PIR test harness 07:39
cotto dukeleto, will it be as crazy as the perl-based version? 07:40
because that'd be both awesome and terrifying
07:41 JimmyZ joined
dukeleto cotto: github.com/leto/tapir 07:43
cotto: it eats TAP and prints out a summary. i think everything that Parrot::Harness does is outside the scope, but tapir runs tests and eats the output 07:44
cotto: my hope is that it makes running the parrot test suite *a lot* faster and maybe gets rid of some perl 5 deps
cotto: tapir is being written to be testable from the ground up, unlike nqpTAP 07:45
as well as being a no-dependency test harness for things built on Parrot 07:46
07:47 fperrad joined
dukeleto fperrad: hola 07:47
cotto pmichaud, ping
dalek tracwiki: v4 | cotto++ | WhyDoesNQPGenerateInefficientCode 07:48
tracwiki: better formatting for those of us with non-big screens
tracwiki: trac.parrot.org/parrot/wiki/WhyDoe...ction=diff
07:50 fperrad_ joined
fperrad_ dukeleto, good morning 07:50
dukeleto tapir should be the fastest test harness around
fperrad_: good localtime()!
fperrad_: thank you for setup.pir, it rocks! 07:51
cotto msg pmichaud Do you know of a system that gets assignment vs binding right? It seems like this would be a productive task to add to ItsABugHunt since it's causing some pessimization in nqp.
purl Message for pmichaud stored.
fperrad_ dukeleto, what's about your 2 msg 07:52
Tene cotto: it would also be useful to ask how it works in other vms.
that would be a good research task for someone.
dukeleto fperrad: what do you mean? 07:53
cotto Tene, that's a good idea. I imagine that the PyPy guys have dealt with something like this problem. 07:54
fperrad dukeleto [Fri Nov 27 23:41:52 2009] said: parrot setup.pir plumage does not work for me
dukeleto [Fri Nov 27 23:48:48 2009] said: i would like to add custom steps to setup.pir, how hard is that?
Tene cotto: please add those questions to the wiki.
dukeleto fperrad: i would like something like "parrot setup.pir readme" to print out a readme or something 07:58
fperrad: and, parrot setup.pir plumage does not work for me :) 07:59
fperrad dukeleto, there are : 08:03
$ parrot setup.pir usage
and you could overload the message in setup.pir :
$P0['usage'] = 'your helpful message'
cotto Tene, done 08:04
dalek tracwiki: v6 | cotto++ | ItsABughunt
tracwiki: add deficient lvalue model
tracwiki: trac.parrot.org/parrot/wiki/ItsABu...ction=diff
fperrad dukeleto, if you want a real custom step, see for example : 08:10
github.com/vadrer/tcl-bridge/blob/m.../setup.pir
dukeleto fperrad: thanks! 08:13
fperrad dukeleto, what happens with : parrot setup.pir plumage 08:15
dukeleto fperrad: unknown target : plumage 08:19
dalek l: 4df1b40 | fperrad++ | setup.pir:
fix shebang
 08:20
fperrad dukeleto, target 'plumage' was added in r42763 08:22
08:36 nopaste joined
dukeleto fperrad: how do I update setup.pir? which files are involved? 08:37
08:39 lucian joined
fperrad dukeleto, only one file in parrot tree : runtime/parrot/library/distutils.pir 08:40
nopaste "vad" at 195.250.178.68 pasted "I've mailed to parrot-users, but it isn't appearing there... (premoderation?)" (48 lines) at nopaste.snit.ch/18897 09:00
dukeleto fperrad: what do i do to upgrage setup.pir in Kea? what do i need to do with distutils.pir? 09:01
vad hi, can anyone please look at nopaste.snit.ch/18897 ?
dukeleto vad: i saw your email go through to parrot-users 09:02
vad: your question is very specific and I am of no help
fperrad dukeleto, nothing in Kea, in parrot : make world install 09:03
vad dukeleto: thanks, probably I'm not seeing my own e-mail due to mailer settings. okay :) 09:10
Tene fperrad: I'm really pleased with setup.pir, thanks! 09:22
dukeleto fperrad: i see, parrot needs to be upgraded. thanks! 09:23
fperrad: i see it now, thanks! 09:25
fperrad++
09:28 iblechbot joined
fperrad Tene, you succeed to convert steme to NQP-RX, have you seen TT #1319 & TT #1321 ? 09:30
Tene fperrad: No, I haven't.
fperrad: Yeah, I can do those. 09:31
probably not tonight.
If you ping me about it tomorrow, I'll try to make time. 09:32
Thanks for pointing that out to me.
I don't know where squaak is located, I think...
abc is in parrot's examples dir, maybe? 09:33
fperrad Tene, both in examples/languages 09:34
Tene ah
fperrad seen chromatic 09:37
purl chromatic was last seen on #parrot 2 days, 5 hours, 27 minutes and 19 seconds ago, saying: There are a few tickets for related issues. [Nov 27 04:09:08 2009]
09:42 mikehh joined
dukeleto squaak? 09:45
purl squaak is languages/squaak or the language in kj's tutorial at parrotblog.org or svn.perl.org/parrot/trunk/languages/squaak/
dukeleto has anyone written getopt for PIR yet? 09:46
Tene dukeleto: I thought I saw plumage using it... 09:48
dukeleto: runtime/parrot/library/Getopt/Obj.pir 09:49
dukeleto cool!
Tene It's in the right namespace, looks like, so you should even be able ot use it from Parrot.
erm
HLLs.
Oh, not quite... 09:51
that could use some fixing... I should write a doc on that.
>.>
I mean... "someone" should.
You didn't see anything.
Is there a ticket about that? If so, does the ticket say where that documentation should live? 09:52
If someone can point me in the right direction on that, I'll write it.
10:13 nopaste joined 10:15 nopaste joined 10:36 zak_ joined 10:53 Zak joined 11:29 mikehh joined
dalek a: e90f9cf | fperrad++ | (2 files):
update generated Plumage description
 11:35
a: 197f0cd | fperrad++ | :
Merge branch 'master' of github.com:fperrad/lua
a: b6ce733 | fperrad++ | (2 files):
convert test to PIR (from Perl)
11:42 luobo joined 12:34 lucian joined 12:41 Zak joined 12:47 lucian joined 12:55 barney joined
dalek rrot: r42821 | fperrad++ | trunk/runtime/parrot/library/distutils.pir:
[distutils] help is an alias of usage
 12:55
13:06 lucian joined 13:16 joeri joined, bacek_at_work joined 13:24 masak joined 13:39 bacek_at_work joined 13:56 Zak joined 14:08 bacek_at_work joined 14:09 JimmyZ joined 14:10 Whiteknight joined
dalek 9plus: 886664c | bernhard++ | (10 files):
Adapted file structure to the structure proposed
 14:23
9plus: 01a5ab9 | bernhard++ | (2 files):
Configure.pl is no longer used, but a second hq9plus.pir is needed.
9plus: 0b160c2 | bernhard++ | lib/Parrot/Test/Hq9plus.pm:
Removed support for testing in 'languages'.
9plus: 262854f | bernhard++ | lib/Parrot/Test/Hq9plus.pm:
Use the installed 'parrot'.
9plus: 5d85288 | bernhard++ | Makefile.in:
Merge with version from current mk_language_shell.pl.
9plus: 071e84b | bernhard++ | README:
Updated README.
9plus: 479ece9 | bernhard++ | src/hq9plus.pir:
gen_builtins.pir is no longer used.
9plus: d212b6d | bernhard++ | (8 files):
Removed copyright statements.
9plus: 2f438ca | bernhard++ | (9 files):
Removed CVS-Id tags.
14:31 bacek_at_work joined
dalek nxed: r232 | julian.notfound++ | trunk/winxedst1.winxed:
operator post ++ in stage 1
 14:32
9plus: 2983ddd | bernhard++ | hq9plus/.gitignore:
Added hq9plus/.gitignore, so that the dir is created.
 14:34
tracwiki: v114 | barney++ | Languages 14:40
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
14:58 patspam joined 15:07 bogen left
dalek a: 1cbd39d | fperrad++ | t/env.t:
more tests in PIR
 15:31
15:32 patspam joined 15:57 Psyche^ joined 16:14 cognominal joined 16:16 patspam1 joined 16:25 elmex joined 16:27 jan joined 16:28 tetragon joined
dalek a: b427908 | fperrad++ | t/env.t:
convert CRLF to LF
 16:44
16:45 tetragon_ joined
dalek rrot-linear-algebra: 1e236ea | andrew++ | (5 files):
fix the build so it dumps the library into the ./ folder instead of ./dynext/. Fix tests so they can run without having to make install
 16:50
rrot-linear-algebra: f27ef82 | andrew++ | README:
remove notice in README about the broken build/test process. It's fixed
16:54 vad joined
dalek a: 5c10f3c | fperrad++ | t/lua-TestMore:
update submodule lua-TestMore
 17:07
tracwiki: v115 | barney++ | Languages 17:24
tracwiki: Note that unlambda works with Parrot 1.8.0.
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
17:43 jsut joined 17:48 elmex joined
japhb fperrad: I've been in Thanksgiving Land. You rang? 17:49
17:51 iblechbot joined
dalek lambda: 4076342 | bernhard++ | (4 files):
Make unlambda work with an installed parrot.
 17:53
lambda: 89010cc | bernhard++ | (5 files):
Use Configure.pir instead of Configure.pl.
lambda: 7bb32d9 | bernhard++ | Configure.pl:
Configure.pl is no longer needed.
rtcl-nqp: 8add5bf | pmichaud++ | (6 files):
Add initial version of AREs (only matches regex literals for now).
 18:14
rtcl-nqp: ccab549 | pmichaud++ | src/ARE/ (2 files):
Add *, +, and ? quantifiers.
rtcl-nqp: c61dca5 | pmichaud++ | src/ARE/ (2 files):
Add dot metachar to ARE (metachar:sym<.>).
rtcl-nqp: 2bff6f8 | pmichaud++ | src/ (3 files):
Add \\d, \\s, \\w, \\D, \\S, \\W backslash escapes to ARE.
18:15 cognominal joined
dukeleto when does dalek update the list of langauges from the wiki? seems like it takes a while 18:15
pmichaud afk, lunch
18:38 theory joined 19:02 joeri joined 19:05 mtk1 joined
dalek rrot: r42822 | jonathan++ | trunk/runtime/parrot/library/P6object.pir:
[p6object] P6Object needs to also recognize new Perl 6 named for Object (left recognition of existing one in, no deprecation issues).
 19:05
19:06 mtk1 left 19:15 Hunger joined
Coke pmichaud++ 19:47
dalek rrot-linear-algebra: ceba80e | andrew++ | src/pmc/pla_matrix_types.h:
fix indexing for complex.
 20:42
rrot-linear-algebra: 8ae9618 | andrew++ | t/pmc/complexmatrix2d.t:
add some tests for ComplexMatrix2D.tranpose and .mem_transpose so we don't regress on those things
rrot-plumage: cdb7a36 | japhb++ | :
[METADATA] GIL and Steme files, courtesy of fperrad++
 20:44
pmichaud Coke: if you'll tell me which regexp constructs are most useful for partcl at this point, I'll add them. 20:48
20:54 davidfetter joined 21:23 PacoLinux joined
dalek rrot-linear-algebra: 7606f31 | andrew++ | (2 files):
add a conjugate method to ComplexMatrix2D and add tests to verify it
 21:29
trixy: 7542fa4 | andrew++ | t/functions/mtimes.t:
skip the t/function/mtimes.t tests, they all fail for now
trixy: 811296c | andrew++ | t/functions/conj.t:
add test for conj()
trixy: f79ad83 | andrew++ | t/functions/ctranspose.t:
add tests for ctranspose.t
21:56 hercynium joined
Coke pmichaud: I don't have any priority on those. 21:56
so, whichever ones look fun. =-) 21:57
21:58 nopaste joined
Coke I am wondering if this is a good time to copy over the switch handling code. 22:04
pmichaud: ping 22:11
22:18 jsut joined 22:19 cognominal_ joined 22:31 lucian joined 22:36 nopaste joined 23:10 nopaste joined 23:25 lucian joined
pmichaud Coke: pong 23:36
japhb pmichaud, I'd like to re-up my pitch for full hash literal syntax in NQP-rx. Aside from Plumage, not having hash literals is also affecting use of fperrad's setup system. 23:39
pmichaud japhb: noted
japhb pmichaud, thanks!
pmichaud japhb: I'd be willing to entertain patches from others for it, though. I looked at writing it a couple of weeks ago and it wasn't coming out very pretty 23:40
japhb pmichaud, if you can describe what you're looking for (by which I mean, any limits on how it should or shouldn't work), I *might* be able to learn enough to do it. 23:45
pmichaud I'm just looking for a cleanish implementation. Generating the code to individually set the hash elements is a bit of a pain.
japhb Might be a good excuse to really try to learn the NQP-rx internals
I'm not entirely sure there exists a cleaner method for non-constant multi-level hashes. (I mean, I can think of some variant implementations, but they are all of equal or greater complexity than just inlining the hash-building code.) 23:48

Powered by Cro and the Raku® Programming Language.

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

Thank you!