|
Parrot 5.3.0 "W00tstock Parrot" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC Set by moderator on 23 May 2013. |
|||
|
00:06
darbelo joined
00:36
darbelo joined
02:45
cooper joined
03:50
preflex_ joined
04:26
Psyche^ joined
04:52
zby_home joined
05:58
PacoAir joined
06:16
sivoais joined
06:47
davidfetter joined
07:21
contingencyplan joined
07:35
denisboyun joined
|
|||
| dalek | p: c557c27 | moritz++ | Configure.pl: be smarter about guessing Darwin architecture Closes #100 |
07:50 | |
|
08:31
bouncy joined
10:22
schm00ster joined
12:22
denisboyun joined
12:41
PacoAir joined
|
|||
| Coke wonders if that will fix some nagging os x test failures coke has had. | 12:55 | ||
|
13:01
PacoAir joined
|
|||
| dalek | p: 2a557de | (Pawel Murias)++ | docs/serialization_format.markdown: Document invocation spec in the serialization format description. |
13:12 | |
|
13:36
woolfy joined
13:37
Liz joined
13:39
woolfy_ joined,
Liz joined
14:24
darbelo joined
14:44
bluescreen joined
14:55
zby_home joined
15:24
davidfetter joined
|
|||
| Coke | anyone here able to review some GC code? | 15:56 | |
| github.com/perl6/nqp/blob/master/s...object.pmc is showing up in some persistant rakudo failures when running on parrot. (specifically in destroy, but it could be an initial setup issue) | 15:57 | ||
|
16:23
particle joined
16:27
bluescreen_ joined
17:11
davidfetter joined,
jsut_ joined
17:32
davidfetter joined
|
|||
| Util | Coke: I can echo your request in the #ps meeting in 2 hours, or you could mention it yourself if you attend. | 17:38 | |
| In either case, could you provide a summary of steps to reproduce the problem? | |||
| Also, (unrelated) are your "nagging os x test failures" documented anywhere that you could point me to? | |||
|
17:41
denisboyun_ joined
|
|||
| Coke | it was in response to nqp, errors are on rakudo on nqp on parrot, not parrot itself | 18:06 | |
|
18:08
sa1 joined
18:13
particle joined
18:35
ggherdov joined
18:49
isBEKaml joined
18:54
alester joined
|
|||
| isBEKaml | Hey guys, was there a parrot release this month? I don't see that on parrot.org. | 19:04 | |
| Util | isBEKaml: There was a release, and I got the word out on parrot-dev, but not parrot.org. Will fix today. | 19:21 | |
| ftp://ftp.parrot.org/pub/parrot/releases/devel/5.4.0/ | |||
| isBEKaml | Util: okay, great. | 19:24 | |
|
19:25
contingencyplan joined
19:27
bluescreen joined
19:34
brrt joined
|
|||
| Util | #ps time | 19:34 | |
| sorear wonders how pmurias' gsoc thing is related to PaFo | 19:50 | ||
|
19:56
isBEKaml joined
|
|||
| masak | sorear: that has already been answered somewhere. | 19:57 | |
| sorear | that's very helpful | 19:59 | |
| Util | sorear: Two answers; one from me, and one from pmurias. | 20:05 | |
| From me (starts at 20:20; bookmark taking me directly to the line): | |||
| irclog.perlgeek.de/perl6/2013-05-01#i_7008132 | |||
| From pmurias, via an addendum to his proposal, in response to that very question: | |||
| The main benefit to the Parrot Virtual Machine is that a JavaScript backend for QAST makes it a much more attractive target for compiler writers (they would be able to target JavaScript, JVM and Parrot with little extra effort over targeting Parrot and JVM). | |||
| As QAST is abstract enough to support Perl 5 and Perl 6 it should be able to support many other dynamic languages with minor extensions (Perl 6 is in a way a superset of dynamic languages). | |||
| The convenience of using Perl 6 grammars for parsing and being able to target multiple backends at once using QAST, might encourage compiler writers to (indirectly) target Parrot even though they don't have a strong belief in the Parrot VM vision. | |||
| Additionally having the languages target a form abstracted from the minor details of the different VM should free Parrot make the necessary changes it will need to evolve. | |||
| END_QUOTE | |||
|
20:07
quias joined
20:25
zby_home joined
20:55
elisa joined
|
|||
| elisa | hi parrots | 20:55 | |
| crocket | Is parrot suitable for such languages as haskell, standard ML, and scala? | ||
| masak | crocket: how do you mean? | 20:56 | |
| crocket | Can people make a compiler that compiles scala to parrot bytecode easily? | ||
| masak | I see no theoretical limitation. Scala is probably at about the same level of difficulty as Perl 6, which has an implementation on Parrot. | 20:57 | |
| crocket | masak, Some people argue that standard ML is not implementable on JVM. | 21:03 | |
| Do you think they are correct? | |||
| a functor-free subset of standard ML was implemented on JVM a long time ago, but it wasn't a complete subset. | |||
| masak | crocket: I'm afraid I don't know enough about standard ML to give a good answer. | 21:04 | |
| crocket | masak, Can people make a parrot bytecode compiler for any language available? | 21:05 | |
| masak | crocket: but claims of non-implementability tend to be based on a lack of imagination -- or maybe they're really claims of "cannot be efficiently implemented". | ||
| crocket | without sacrificing their functionalities. | ||
| masak | crocket: yes, the idea is to target almost any language. that said, Parrot is especially targeted towards dynamic languages. | 21:06 | |
| crocket | masak, What about static languages? | ||
| masak | I don't think it's impossible -- it's just that Parrot wasn't built primarily for them. | 21:07 | |
| the way I see it, it's easier to build a static-language compiler for a dynamic VM than the other way around. | |||
| darbelo | A static language would probably need to bypass or otherwise avoid the parts of parry that aim to support dynamic languages. | 21:08 | |
| crocket | parry | 21:09 | |
| This is not encourating. | |||
| encouraging. | |||
| What VMs are more versatile? | |||
| darbelo | Alternatively, you could just ignore the extra features and not think about how it would be more efficient if the dynamic stuff wasn't there in the way :) | 21:11 | |
| crocket | JVM nowadays support dynamic langauges, too. | 21:13 | |
| Has parrot gained enough momentum to make an ecosystem? | 21:14 | ||
| Or does it already have an ecosystem? | |||
| darbelo | How large does something need to be to get called an ecosystem? | 21:15 | |
| masak | Parrot used to have more active High-Level Languages than it does today. | ||
| darbelo | The biggest language using parrot is Rakudo Perl 6. | 21:16 | |
| masak | I struggle to think of others. | ||
| crocket | Rakudo Perl 6? | ||
| masak | crocket: rakudo.org/ | ||
| crocket | darbelo, enough ecosystme for commercial web applications. | 21:17 | |
| masak | or maybe better rakudo.org/about/ | ||
| to the extent that there are web applications out there running Parrot, they are running Rakudo. | 21:18 | ||
| crocket | perl??! | ||
| Can parrot run all existing perl programs? | 21:19 | ||
| masak | note that `perl` refers to the Perl 5 implementation, while `Rakudo Perl 6` refers to a Perl 6 implementation. | ||
| sounds like a difference in major versions only, but it's more like a difference between two languages. | |||
| `perl` doesn't run on Parrot. it's more it's own little VM. | 21:20 | ||
| its* | |||
| darbelo | Think of the python 2/3 split, but bigger. | 21:21 | |
| Much bigger. | |||
| crocket | masak, You know what I mean. | 21:22 | |
| masak, I don't know perl well enough. | |||
| masak | crocket: *I* know what *you* mean. I'm pointing out to you that there's a difference. you seemed not to be aware of it. :) | ||
| crocket | masak, Can parrot run all existing perl6 programs? | ||
| masak | crocket: if you meant Perl 5, then the answer is "no". Parrot cannot run all existing Perl programs. | 21:23 | |
| crocket: if you meant Perl 6, then the answer is "mmmnnn". Rakudo Perl 6 is still chasing the spec. it can do a lot, but not everything. | |||
| on the other hand *all* current Perl 6 implementations are chasing the spec. | |||
| darbelo | I'm tempted to say "Only to the extent that perl6 can run all existing perl6 programs." | ||
| masak | that's why we haven't seen a "final, for realz, totally ready for production ultimate release of Perl 6" yet, I guess. | 21:24 | |
| crocket | It seems rakudo is the biggest perl6 implementation. | 21:26 | |
| masak | Niecza isn't half-bad either. see perl6.org/compilers/features | 21:27 | |
| crocket | It seems that perl6 is targeting parrot mainly. | 21:30 | |
| darbelo | Parrot started out as the "VM for the next perl" | 21:31 | |
| A long time ago. | |||
| Also, Rakudo is growing multi-vm support nowadays. | 21:33 | ||
| I think it's on its way to bootstrapping on the JVM. | |||
| masak | aye. | ||
| darbelo | Or maybe it's there already, haven't checked in a while. | 21:34 | |
| masak | nqp is bootstrapped already. Rakudo parses the core setting and can print "Hello World". | 21:36 | |
| crocket | nqp? | 21:57 | |
| darbelo | "Not Quite Perl" | 21:58 | |
| It's a subset of perl used to implement perl. | 21:59 | ||
| An older version of that is part of the parrot toolkit. | 22:00 | ||
| See the bit about "PCT Overview" in docs.parrot.org/parrot/devel/html/d...s.pod.html | 22:03 | ||
| For an idea of what it's used for. | |||
| dalek | kudo/nom: be60ebe | dagurval++ | src/core/IO.pm: fix order of directory/volume used in dir() |
22:06 | |
| kudo/nom: 2f8ebe1 | (Brent Laabs)++ | t/spectest.data: add S02-names/bare-sigil.t to spectest |
|||
| kudo/nom: a418e53 | dagurval++ | src/core/Temporal.pm: Fail on default constructor |
|||
| kudo/nom: 2788f30 | dagurval++ | src/core/Temporal.pm: empty constructor in DateTime class, not Date... |
|||
| kudo/nom: f2cce1c | dagurval++ | src/core/Temporal.pm: Added gist to DateTime |
|||
| kudo/nom: ee38e16 | labster++ | src/core/Temporal.pm: Merge branch 'nom' of github.com:dagurval/rakudo into nom |
|||
| kudo/nom: 571ae98 | (Brent Laabs)++ | src/core/IO/Spec/ (3 files): try to fix $*TMPDIR/zavolaj on feather |
|||
| kudo/nom: 8677d32 | labster++ | / (12 files): Merge branch 'nom' of github.com:labster/rakudo into nom |
|||
| kudo/nom: 118018c | labster++ | / (2 files): Merge branch 'nom' of github.com:rakudo/rakudo into nom |
|||
| kudo/nom: 8a08591 | labster++ | src/core/IO.pm: Add IO::Handle meth: gist, perl, path (now returns IO::Path), IO (a no-op) |
|||
| darbelo | Also: From the README of the 'new' NQP Rakudo uses: | ||
| NQP is the successor implementation of "nqp-rx" [2]. Unlike nqp-rx, which aimed to have almost no runtime component whatsoever, this new version of NQP accepts that a minimal Perl 6 object metamodel, multidispatcher, and regular expression engine are needed on top of the underlying virtual machine | |||
| nqp-rx would be the one that ships with parrot. | 22:08 | ||
|
22:09
davidfetter joined,
crocket left
22:45
PacoAir joined
23:03
kid51 joined
|
|||