Camelia, the Perl 6 bug

IRC log for #parrot, 2008-04-24

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:24 cotto_home rillian, thanks
00:24 cotto_home that looks like the talk I attended
00:25 cotto_home (separate from lugradio live)
00:27 particle skids: you're much more likely to get perl 6 questions answered in #perl6 on freenode
00:28 skids problem is most of my questions are really aimed at getting an idea of what is going on behind the scenes, so it ends up with a pugs-centric answer.
01:03 Infinoid and my problem is, I'm usually on floor 3 or 4 of the parrot building, and I rarely take the elevator up to floor 10 (rakudo)
01:05 particle third floor: cage cleaners, bytecode emitters, test files. going up!
01:06 kid51 joined #parrot
01:09 Infinoid to get to the GC in the basement, you have to take the stairs.
01:09 particle :D
01:21 davidfetter joined #parrot
02:04 Zaba_ joined #parrot
03:17 tetragon joined #parrot
03:21 japhb When does work on the GSOC stuff begin?  June?
03:32 AndyA joined #parrot
03:52 cotto_home http://code.google.com/opens​ource/gsoc/2008/faqs.html#0.1_timeline
03:52 shorten cotto_home's url is at http://xrl.us/bispn
03:53 Zaba joined #parrot
03:55 dalek r27152 | pmichaud++ | trunk:
03:55 dalek : [docs]:
03:55 dalek : * Fix pod error in LANGUAGES_STATUS.pod .
03:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27152
04:08 Infinoid Community Bonding Period.
04:08 * Infinoid bonds with dalek
04:11 * pmichaud bonds with cyanoacrylate
04:12 cotto_home group hug
04:16 cotto_home I'm a little surprised purl didn't have a response for that.
04:18 Infinoid you sound like a volunteer to add one :)
04:20 cotto_home I'm trying to think of something fitting.
04:40 dalek r27153 | pmichaud++ | trunk:
04:40 dalek : [core]:
04:40 dalek : * Remove user stack opcodes (RT#52842)
04:40 dalek :   (save, restore, lookback, entrytype, depth, rotate_up)
04:40 dalek : * Update NEWS
04:40 dalek : * There is still dead code and structures to be removed -- this
04:40 dalek :   just eliminates the opcodes.
04:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27153
04:43 Zaba_ joined #parrot
04:48 Psyche^ joined #parrot
05:14 dalek r27154 | pmichaud++ | trunk:
05:14 dalek : [core]:  dead code removal - PDB_print_user_stack()
05:14 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27154
05:40 dalek r27155 | pmichaud++ | trunk:
05:40 dalek : [core]:
05:40 dalek : * Remove what's left of the user stack.
05:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27155
05:57 uniejo joined #parrot
06:11 peepsalot joined #parrot
06:26 nopaste joined #parrot
06:40 TonyC joined #parrot
06:58 allison joined #parrot
07:37 iblechbot joined #parrot
07:55 Psyche^ joined #parrot
07:59 bradleh joined #parrot
08:29 IllvilJa joined #parrot
09:00 uniejo joined #parrot
09:04 DarkWolf84 joined #parrot
09:11 Zaba joined #parrot
09:16 mire joined #parrot
09:21 dakkar joined #parrot
09:55 wknight8111 joined #parrot
10:22 Zaba_ joined #parrot
10:52 ruoso joined #parrot
10:55 rdice joined #parrot
11:51 wknight8111 joined #parrot
11:55 dalek r27156 | coke++ | trunk:
11:55 dalek : [docs]
11:55 dalek : Remove deleted items from the list of deprecation notices.
11:55 dalek : Note in the NEWS file that some of these were removed ahead of schedule.
11:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27156
12:13 Zaba_ joined #parrot
12:15 Coke group hug is <reply>get off me!
12:16 nopaste "dakkar" at 85.45.142.2 pasted "trying classes in rakudo" (20 lines) at http://nopaste.snit.ch/12800
12:17 nopaste "dakkar" at 85.45.142.2 pasted "output from 12800" (10 lines) at http://nopaste.snit.ch/12801
12:17 dakkar what am I doing wrong?
12:19 nopaste "dakkar" at 85.45.142.2 pasted "PIR for 12800" (164 lines) at http://nopaste.snit.ch/12802
12:20 dakkar looks like the parameter to Bar::set does not get its value, and that assigning the resulting undef is a bad idea...
12:32 GeJ dakkar: shouldn't that be    Line 11  method set(:$v) { $.x=$v }
12:32 GeJ ?
12:32 dakkar uh
12:33 GeJ .o0(Just a guess, I don't have a rakudo near me)
12:33 dakkar no, that marks $v as an optional parameter
12:33 dakkar that's what I infer looking at the generated PIR
12:34 dakkar this is reminding me of the times I tried to understand what gcc was doing by looking at the assembly output :)
12:37 iblechbot joined #parrot
12:40 UltraDM joined #parrot
13:08 Zaba_ joined #parrot
13:15 skids joined #parrot
13:17 MagNET joined #parrot
13:49 gryphon joined #parrot
14:04 rdice joined #parrot
14:17 Coke "remember that time you tried to drill a hole in your own head?"
14:22 jalbo Nast
14:22 jalbo Uh, sorry.
14:30 Tene I remember the time I tried to drill a hole in *your* head...
14:33 Coke ... clearly no ghostbusters afficionados.
14:34 Tene None, or one?
14:36 Coke ... pedant.
14:37 japhb I didn't get the Ghostbuster's reference.  I just immediately thought of a Stargate episode involving trepanning.
14:39 Coke ... I am a rabid stargate fan and am trying to remember which episode this is.
14:39 Coke ... though I did miss season five.
14:40 * japhb goes to look it up
14:40 DarkWolf84 joined #parrot
14:41 japhb Looks like season 3, episode "Demons": http://www.rdanderson.com/st​argate/database/databas3.htm
14:51 Infinoid Stargate SG-1 is one of the things on my watch-queue.
14:55 jalbo http://en.wikipedia.org/wiki/L​ist_of_Stargate_SG-1_episodes
14:57 japhb Ah, I see there is a release date for Stargate: Continuum.  Cool.
15:28 Tene Hm.  Looks like I'm not scheduled to teach the week of yapc::na.
15:28 Tene I might even have the week off.
15:30 Tene Now I have to actually consider going.
15:32 Coke dooooo it
15:33 Tene Is there a list of talks posted anywhere yet?
15:33 pmichaud Tene: I haven't seen one
15:33 Tene Hmm.  I wonder if I could arrange to teach in Chicago the week before, and then just stay there.
15:35 pmichaud parrot hackathon on sat and sunday before the conference :-)
15:37 Tene Although, I could just get my employer to schedule my return flight to go to Chicago instead of SLC.
15:37 Tene That's friday, though.
15:37 * particle arrives friday night
15:38 * pmichaud arrives saturday morning
15:38 Tene I also need to arrange for a place to stay.
15:38 particle everybody stays in the dorms
15:38 particle i don't know rates yet, but you can't book it anyway until next week
15:39 Tene Oh.
15:39 particle ~$40/night iirc
15:39 Tene Hm.
15:39 particle http://yapcchicago.org
15:39 Tene That should be possible.
15:40 particle the hackathon is in the dorms' common area
15:48 particle I found your resume online from awhile back. Please let me know if you’re and interested in a 6month config mgmt project in Brazil.
15:48 particle ...tempting...
15:52 particle why on earth, in a south american city of 18M people, they're looking internationally, i have no idea
15:52 * Coke arrives friday night, and is staying at the dorms, if they ever open up reg on that.
15:53 particle here's hoping
15:55 particle i'd like to get a block of rooms together, and near a common room, for us parrot hackers
15:56 particle pinged josh about it, but no response
15:56 pmichaud that would be great, yes
15:59 Coke I especially like the part where someone else handles that. =-)
16:12 particle my god, my soccer game *starts* at 9:15p tonight
16:14 particle anyone here use the kate editor?
16:14 * particle wonders if there's a directive syntax for kate like vim and emacs
16:15 particle ...similar to # vim: shiftwidth=4 expandtabs:
16:17 pmichaud my soccer scrimmage starts at 8:30p tonight
16:18 ambs joined #parrot
16:18 particle so, we'll start at about your halftime
16:18 particle er, no
16:18 pmichaud nope
16:18 pmichaud I'll be home before you even start :-)
16:18 particle right
16:42 pmichaud afk # lunch
16:48 gryphon joined #parrot
16:51 particle can't decide if it's better as a tpf grant or microgrant
16:51 particle oops
16:58 dakkar joined #parrot
17:08 Tene http://conferences.mongueurs.net/yn2008/talks
17:08 dalek r27157 | fperrad++ | trunk:
17:08 dalek : [crypto]
17:08 dalek : - minor improvement
17:08 dalek : Courtesy of Jim Keenan
17:08 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27157
17:17 * Coke ponders how nice it would be to have an expansion for $Coda$ (along with the long awaited $Copyright$)
17:18 particle i'd settle for $Year$
17:18 Theory joined #parrot
17:19 particle since svn 1.1 i've been waiting
17:22 dalek r27158 | chromatic++ | trunk:
17:22 dalek : [src] Tidied code; no functional changes.
17:22 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27158
17:24 dalek r27159 | chromatic++ | trunk:
17:24 dalek : [GC] Consolidated duplicate flag tests into a single path.
17:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27159
17:37 Coke particle: I think this last release was "close".
17:44 Coke (to having that feature.)
17:45 particle so was 1.3.
17:45 particle svn--
17:45 Zaba hmm..
17:45 Tene Hmm... perhaps it's time for me to play with mod_parrot.
17:45 Zaba what will be the next parrot release? 0.6.2? 0.7?
17:45 particle zaba: depends on which features are implemented
17:45 particle we don't know the release number until near the release date
17:46 Zaba hm, I see.
17:46 particle releases are date-based, release numbers are feature-based
17:46 Tene Zaba: what release would you like it to be?
17:46 * particle would like 1.5
17:47 Zaba Tene, I don't know how exactly they are numbered.. so I can't say..
17:47 Tene purl: roadmap
17:47 purl roadmap is http://www.xray.mpe.mpg.de/mailing-lis​ts/perl5-porters/2004-11/msg00115.html
17:47 Tene purl: you lie!
17:47 purl Tene: huh?
17:47 particle there's a wiki page....
17:48 Zaba purl, that's for perl 5.10!
17:48 purl Zaba: huh?
17:48 Tene purl: parrot roadmap
17:48 purl Tene: excuse me?
17:48 particle http://www.perlfoundation.org/pa​rrot/index.cgi?release_planning
17:48 Zaba Tene, I'll stick with trunk anyway :>
17:50 * Tene pondering mod_lolcode
17:51 Zaba Tene, a good idea
17:51 Zaba (except I don't use apache)
17:51 * particle thinks jhorwitz started mod_lolcode, and dropped it
17:52 particle you should ping him for his work so far
17:52 particle shouldn't be hard to do from scratch though, registry-style
18:07 Debolaz People who use apache for anything except maintaining legacy applications needs to be mutilated and killed. Not neccesarily in that order. Imho. :)
18:08 Zaba I run lighty on my server
18:08 Zaba since I need nothing but static content
18:09 * Debolaz uses lighty for dynamic content as well.
18:10 jalbo I use HTTP::Daemon
18:10 Zaba Debolaz, it has some problems with cgi, for example checking for extension '.cgi' instead of executable bit
18:11 Zaba and well I don't need anything but static stuff anyway
18:11 Debolaz I mostly use FastCGI.
18:27 Ivatar joined #parrot
18:37 donaldh joined #parrot
18:38 gryphon joined #parrot
18:40 PJMODOS joined #parrot
18:40 PJMODOS e
18:40 Tene f
18:41 davidfetter heh
18:48 itz joined #parrot
18:48 itz I've found what appears to be an odd bug in perl6
18:48 pmichaud there are probably many of those :-)
18:50 Tene itz: details?
18:50 purl Where's the beef??
18:50 itz what was the paste url?
18:50 pmichaud nopaste?
18:50 purl i heard nopaste was at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://paste.husk.org/ or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or don't bother me while I'm eating
18:51 Tene purl: parrot roadmap is http://www.perlfoundation.org/pa​rrot/index.cgi?release_planning
18:51 purl OK, Tene.
18:51 shorten Tene's url is at http://xrl.us/7nmt
18:51 itz http://nopaste.snit.ch/12804
18:52 sjansen joined #parrot
18:52 Tene purl: sjansen?
18:52 purl no idea, tene
18:54 * sjansen is a riddle wrapped in a mystery inside an enigma
18:54 pmichaud itz:  does it simply output nothing, or ... ?
18:58 itz maximum recursion depth exceeded
18:58 Tene itz: can you nopaste the PIR dump?
18:59 pmichaud are you actually creating an object of type Foo and invoking .print() somewhere?
18:59 itz yeah should have put that in the original sorry
18:59 itz http://nopaste.snit.ch/paste
18:59 pmichaud it might be a problem of having a method named 'print'
19:00 itz sorry there is a missing line Foo.print();
19:00 itz ah
19:00 pmichaud I think the print @foo.elems() line is acting as a recursive call
19:01 pmichaud (whether it *should* act that way is something I'd need to check the synopses for)
19:01 Tene itz: run it with parrot perl6.pbc --target=pir foo.pl
19:01 Tene and nopaste the output
19:03 dakkar is inheritance working in rakudo?
19:03 dakkar http://nopaste.snit.ch/12800 and http://nopaste.snit.ch/12801 and http://nopaste.snit.ch/12802 for reference
19:03 nopaste "Infinoid" at 96.238.213.50 pasted "error output and pir for itz's test script" (128 lines) at http://nopaste.snit.ch/12806
19:04 pmichaud I think there may be a bug in the way that rakudo is currently handling attribute definitions
19:04 dakkar pmichaud: oh
19:05 pmichaud I'm not sure that "is Foo" is picking up the appropriate class object, especially since the way attributes are currently handled is creating a new class object for each attribute
19:05 itz http://nopaste.snit.ch/12807
19:05 pmichaud (i.e., I'm not sure the protoobject gets updated to a new instance.)  It may be -- I haven't checked completely yet
19:06 pmichaud itz:  rakudo methods might be defaulting to "is export"
19:06 dakkar the "is Foo" maybe work, but what does not work (and I noticed just now) is that $x is made a *private* attribute
19:06 pmichaud has $.x always creates a private attribute
19:06 dakkar at least, that's what I read from the PIR (and I'm not sure I can read it correctly)
19:06 dakkar oh
19:06 dakkar with a public accessor method?
19:06 pmichaud yes
19:06 dakkar ok, so that part works as expected
19:08 dakkar http://nopaste.snit.ch/12802 at the end: that's the code that sets up my classes, right?
19:08 pmichaud yes
19:08 pmichaud and in looking at it here, it appears that it is doing inheritance as I would expect
19:08 pmichaud (not the actual code I would use, but good enough to produce what I think are correct results)
19:09 dakkar hmmm
19:09 Infinoid itz: so, it looks like the problem is that Foo.print() method is calling itself rather than the "print" builtin, right?
19:10 pmichaud dakkar: it's probably worthwhile to send this example to rakudobug
19:11 itz is there a way of calling the "print" builtin more explicit?
19:11 pmichaud itz:  I don't know if it works yet, but one could conceivably do   &*print
19:11 pmichaud (i.e., grab the "global" print)
19:12 dakkar pmichaud: I'll do another couple of checks, then submit it as a bug
19:12 pmichaud dakkar:  that would be great, thanks
19:12 Infinoid I think itz's bug might be a parrot namespace issue, rather than a rakudo one
19:12 Infinoid comparing a standalone script's PIR to his test script's PIR, the op is the same
19:12 pmichaud Infinoid: I know that at one time Parrot would automatically stick methods into the namespace
19:13 pmichaud Infinoid: then there was a time when it wouldn't, and I complained
19:13 Infinoid its just "print"($P15)  in both cases
19:13 pmichaud there does need to be a way to create methods that don't automatically appear in the namespace.  Last I checked that was parrot's default, but perhaps that default changed again.  I don't know how/if it was resolved.
19:14 Infinoid &*print plays some weird games with get_hll_global.  it seems to work, but it does require parens to be added to the test script too
19:15 pmichaud I think parens may be required when the & is used.  I'd have to check STD.pm and the synopses for that
19:16 Infinoid so, &*print(@foo.elems()) is a valid workaround
19:16 Infinoid and ... it prints newlines, despite being print() instead of say()
19:18 Infinoid pmichaud: I'd gladly add a test for the namespace thing.  any pointers on how to go about it?
19:19 pmichaud in pir?  I'm not sure we ever resolved what the correct behavior would be.
19:19 * Infinoid adds pdd21_namespaces.pod to his reading queue
19:20 pmichaud actually, I think the question is how the :method flag is being handled in Parrot
19:20 pmichaud from S02:  Unlike in Perl 5, the notation &foo merely stands for the foo function as a Code object without calling it. You may call any Code object with parens after it (which may, of course, contain arguments):
19:20 Infinoid that's getting into IMCC territory, right?
19:20 pmichaud Infinoid: yes.
19:22 pmichaud Infinoid: anyway, I'd be curious to know how Parrot is working now
19:22 pmichaud so, create a class, define a method in the class, then do a lookup of the method name in the class' namespace
19:23 pmichaud if it returns PMCNULL, then :method isn't automatically placing the method into the namespace (good)
19:23 pmichaud if it returns a Sub object, then :method is placing the method into the namespace
19:23 particle didn't we have some discussion about :method :anon ?
19:23 Infinoid it pretty obviously is being added to the namespace, based on my reading of http://nopaste.snit.ch/12806
19:23 pmichaud particle:  discussion, yes; decision -- I don't remember
19:24 pmichaud I remember advocating that :method :anon would be the standard mechanism for "method not in namespace"
19:24 Infinoid hmm.  but the print method didn't get :anon
19:24 pmichaud however, we could also do that the default is method not in namespace, and :export puts it into the namespace
19:24 pmichaud Infinoid: right -- I don't think rakudo has dealt with the issue of exporting (or not exporting) methods yet
19:25 particle http://rt.perl.org/rt3/Tic​ket/Display.html?id=48631
19:25 Infinoid does there need to be a special syntax to differentiate methods from functions?
19:25 Infinoid (or builtins, in this case)
19:25 pmichaud in PIR?  We have one -- it's  :method
19:25 particle right.
19:26 Infinoid so if I hack it to add :anon, this will suddenly work? :)
19:26 Coke this reminds me: did we ever have a decision on class methods other than "no, we don't have those" ?
19:26 particle pmichaud: as per that ticket, you requested that :method also be entered in the namespace. it is.
19:26 pmichaud Coke:  I think it's "no, we don't have those".
19:27 Coke pmichaud: any idea how we're going to support languages that have that OO model?
19:27 pmichaud Coke:  it will depend on the language (more)
19:27 particle coke: derive your class from Object
19:27 pmichaud In the case of Perl 6, I would end up creating a protoobject for the class, and then adding the class methods to that specific instance (possibly via roles or some other)
19:28 pmichaud or, just use the protoobject as the invocant for a "class method"
19:28 pmichaud at any rate, my understanding of protoobjects and Perl 6 evolved beyond having to worry about class methods, at least for now :-)
19:29 Infinoid $P54 = $P53."print"()  is supposed to call a method, but  "print"($P37)  is supposed to call a standalone.  But  Foo.print :method  ends up being called in both cases here, causing the recursion exception.  That tells me :method didn't work
19:29 Infinoid sounds like a parrotbug
19:29 pmichaud Infinoid: it depend on how :method is defined
19:30 pmichaud Infinoid: and that's what we're saying hasn't been resolved yet
19:30 Infinoid ok, thanks
19:30 pmichaud or perhaps I'm misunderstanding what you're pointing out -- just a sec
19:30 Infinoid I'm mostly PIR-clueless
19:31 pmichaud okay, here's my guess
19:31 pmichaud :method "worked" in the sense that it created a method for 'print' in the class 'Foo'
19:31 pmichaud however, it also added 'print' into Foo's namespace.  (This is the part that is unresolved.)
19:31 pmichaud so, $P53."print"()   does exactly the correct thing by calling the print method
19:32 particle pmichaud: it's not unresolved.
19:32 pmichaud inside of that method we have "print"($P37)
19:32 particle "Kevin Tew started investigating this ticket, and he discovered
19:32 particle that :method subs are already being placed as entries in the
19:32 particle namespace by default, which is the behavior I was looking for.
19:32 particle So, the issue turns out to be a non-issue (for me at least),
19:32 particle unless we decide to make it one." --pmichaud
19:32 pmichaud particle:  ah.  I just hadn't gotten to the end of that post.
19:33 pmichaud particle:  you're correct, we resolved it as the current behavior being okay for now.  But it's no longer okay for now -- we need a way to create methods that aren't in the namespace
19:33 pmichaud and I'm now going to switch my position and ask that :method not place methods in the namespace by default, and that we have an explicit :export flag to get it to do so
19:33 particle try :method :anon. see if that works today
19:34 Zaba_ joined #parrot
19:34 Infinoid pmichaud: sorry, I'm still just trying to catch up with you guys and understand what's going on.  In the PIR for a method, if you call "foo"($P37), is that obviously a non-method invocation, or will it be translated to $P37."foo"()?
19:35 pmichaud Infinoid: method can be invoked in two ways
19:35 pmichaud $P37."foo"()  is always a method
19:35 pmichaud sorry
19:35 pmichaud $P37."foo"()  is always a method call
19:35 pmichaud however
19:35 pmichaud methods can also be treated as subs, where the first argument is the invocant
19:35 pmichaud thus  "foo"($P37) can also be used to invoke a method
19:36 Infinoid that sounds a lot like p5 :)
19:36 Infinoid ok, and "isa"($P37, "foo") will end up calling the isa vtable method on whatever class P37 belongs to?
19:36 pmichaud no
19:37 particle please don't use the words "vtable method". there's no such thing.
19:37 pmichaud not precisely
19:37 particle it's "method" and "vtable function"
19:37 pmichaud for "isa"($P37, "foo"), Parrot does
19:37 pmichaud $P99 = find_name "isa"
19:37 pmichaud $P99($P37, "foo")
19:38 pmichaud the (Sub) object associated with "isa" can be either a normal sub or a method -- the caller doesn't care
19:39 pmichaud however, the "isa" method that gets invoked depends strictly on the symbols in the current and global namespaces -- it doesn't depend on the class of $P37 at all
19:39 pmichaud this is different from  $P37."isa"("foo"), in which case we would grab the "isa" method that is associated with $P37
19:40 pmichaud in other words   "isa"($P37, "foo")   doesn't look at the type of $P37 at all (unless "isa" is defined with mmd/:multi), while $P37."isa"(foo)  never looks in the namespaces
19:40 Infinoid ok, thanks.  if I'm understanding correctly, that sounds like a strong argument for not listing methods in namespaces
19:40 pmichaud at this point I agree.
19:41 pmichaud particle++ # finding the earlier thread on the topic
19:42 pmichaud wow, :anon seems to work :-)
19:43 pmichaud nopaste coming up
19:43 gryphon joined #parrot
19:44 particle ...as i suspected.
19:45 pmichaud I'm thinking that better would be to switch the default
19:45 gryphon joined #parrot
19:45 contingencyplan joined #parrot
19:46 particle i don't understand why you wanted it the other way around in the first place.
19:46 particle but, we have something that works today. should be easy to modify the code generator to support that, no?
19:46 particle ...then argue your case to have it flipped.
19:46 pmichaud I wanted it the other way because otherwise it was difficult to get a method into a namespace  (basically it involved the use of :load :init subs and some very bizarre lookups)
19:47 pmichaud yes, we can fix the code gen immediately
19:47 pmichaud ideally what I'm after now is
19:47 pmichaud .sub 'foo' :method                   #  creates a method, no namespace entry
19:47 pmichaud .sub 'bar' :method :export          # creates method, also entry as 'foo' in namespace
19:48 pmichaud .sub 'baz' :method :export('abc')      # creates 'baz' method, entered as 'abc' in namespace
19:48 particle yep
19:48 pmichaud although all I need for the moment is the :export flag
19:48 particle .sub 'baz' :method :export('abc') :vtable('def')
19:48 cognominal pmichaud, speaking of namespace, I am not familiar to the compilation chain to add it to regexen
19:49 nopaste "pmichaud" at 76.183.97.54 pasted ":anon does what we need for now (particle++)" (37 lines) at http://nopaste.snit.ch/12808
19:51 pmichaud uh oh, no it doesn't
19:51 purl oh yes it does!
19:51 pmichaud :anon also means it's not a method
19:51 nopaste "pmichaud" at 76.183.97.54 pasted ":anon doesn't do what we need for now :-( :-(" (44 lines) at http://nopaste.snit.ch/12809
19:52 pmichaud I'll re-open #48631 with the new recommendation
19:52 pmichaud or I can make a new ticket
19:54 PerlJam good afternoon
19:55 PerlJam What's new in parrot-land today?
19:55 pmichaud exporting methods as subs (or not exporting them)
19:55 Coke pmichaud: new ticket with a see also is probably best, since you're asking for something else. =-)
19:57 pmichaud Coke++ # thanks
20:01 Zaba joined #parrot
20:01 allison joined #parrot
20:04 ambs Anybody knows what means: sengsv?relse ?
20:14 Debolaz ambs: If danish, it means bedroom.
20:14 ambs yes, danish
20:15 ambs strange
20:15 ambs one room type is "standard class"
20:15 ambs the other is "standard 4 sengsvaerelse"
20:19 Debolaz Hmmm.. Yeah, that didn't quite add up.
20:19 ambs hehehe
20:19 * Debolaz is a norwegian, danish not the first language. :)
20:20 ambs hehehe
20:20 ambs thanks
20:20 ambs asking on #perl :)
20:23 Debolaz It seems to simply mean number of beds.
20:26 donaldh single bed
20:26 donaldh dobbelværelse
20:26 donaldh presumably means double
20:26 ambs yes, somebody on #perl said it was a room with four beds
20:26 ambs although that is kind of strange
20:26 ambs :)
20:29 wknight8111 joined #parrot
20:30 * ambs sent an email to the hotel asking for information
20:30 ambs Debolaz, and donaldh, thanks for your help
20:31 nopaste "donaldh" at 213.123.171.12 pasted "definition of seng" (1 line) at http://nopaste.snit.ch/12810
20:32 donaldh oops http://da.wiktionary.org/wiki/seng
20:35 lichtkind joined #parrot
20:35 ambs :)
20:40 Zaba_ joined #parrot
20:47 Theory joined #parrot
21:29 mj41 joined #parrot
21:43 ewilhelm joined #parrot
22:07 cotto_work does anyone know offhand the revision when the perl6 fakecutable was introduced?
22:08 cotto_work I'm becoming increasingly certain that it never worked with c's mmd example
22:09 pmichaud well, the mmd stuff is quite recent
22:09 pmichaud I think it might have worked for a couple of revisions, and then something broke in the perl6 executable
22:10 cotto_work bwt, my definition of "worked" is "didn't cause a backtrace"
22:10 pmichaud right
22:11 pmichaud I think something happened to parrot that causes the fakecutable to not function properly in some cases
22:11 pmichaud I haven't had a chance to track it down yet, though.
22:11 cotto_work I'm trying to bisect it
22:11 pmichaud find the revision where mmd was introduced
22:11 pmichaud it will be on or about the date that c posted the article
22:12 pmichaud I have to run to dinner, bbiaw
22:12 cotto_work I've been trying to find the first revision where the example doesn't cause a bt
22:31 particle joined #parrot
22:40 cotto_work this is beyond my skill to heal
22:40 Limbic_Region joined #parrot
22:40 Infinoid cotto_work: what does your test look like?
22:41 skids joined #parrot
22:42 cotto_work pick a revision, apply http://www.parrotvm.org/svn​/parrot/revision?rev=26173, make perl6, check if it pukes on chromatic's mmd example
22:42 cotto_work repeat
22:43 ruoso joined #parrot
22:44 cotto_work I'm getting the impression that I'm missing something
22:56 Limbic_Region looks like the Win32/Cygwin and Win32/MinGW build is broke again - unless someone invalidated the bytecode - in which case.....make reaclean now
23:03 Infinoid Limbic_Region: pmichaud has been removing ops, which I'm pretty sure will invalidate the bytecode
23:03 particle joined #parrot
23:19 Limbic_Region Infinoid - yeah, it built fine after the realclean
23:22 dalek r27160 | pmichaud++ | trunk:
23:22 dalek : [core]: more dead code elimination
23:22 dalek : * Remove STACK_ENTRY_INT, STACK_ENTRY_FLOAT, STACK_ENTRY_STRING,
23:22 dalek :   and STACK_ENTRY_POINTER, as they are no longer used after
23:22 dalek :   eliminating stack.ops .
23:22 dalek : * Resolves RT#52842.
23:22 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27160
23:32 Limbic_Region any oracle gurus about that would mind answering a question in /msg ?
23:35 * davidfetter not a guru, but i've worked with the beast. what's up?
23:38 Limbic_Region will /msg since it is completely OT
23:39 Limbic_Region ok, /msg(s) sent
23:44 kid51 joined #parrot
23:47 purl joined #parrot
23:48 Zaba joined #parrot

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced