Parrot 2.8.0 released | parrot.org Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | smoke GC-related branches and attack GC tickets
Set by moderator on 1 October 2010.
00:19 theory left 01:30 hercynium joined 01:42 theory joined 01:49 PerlPilot joined 01:54 PerlJam left 02:35 janus left 02:36 janus joined 04:15 hercynium left 04:30 AzureStone left 04:40 AzureStone joined
dukeleto AzureStone: howdy 04:46
AzureStone Hi 04:47
I am Japanese. 04:48
dukeleto AzureStone: all are welcome here :) How did you hear about Parrot ? 04:55
dukeleto has 10 minutes of batter left
battery, even.
AzureStone Perl6 04:56
dukeleto AzureStone: that is how I got interested in Parrot as well 04:57
AzureStone: there is a large overlap between #parrot and #perl6
AzureStone Perl6 doing some research and learned parrot. 04:58
dukeleto found a power outlet. Yay. 05:01
AzureStone: what kind of stuff are you doing in Perl 6? 05:04
05:26 plobsing_ left
dukeleto I found a simple-to-fix coredump in ByteBuffer. 05:28
dalek rrot: r49487 | dukeleto++ | trunk (2 files):
[pmc] Fix coredump in ByteBuffer when initialized with a negative size and add a test
 05:49
05:55 theory left
dukeleto coverage? 06:09
The Iterator PMC has very little test coverage. 06:27
sorear The Iterator PMC has very little documentation, too 06:29
or I should say, documented semantics; it's one of the inconsistent bits of the aggreggate API I was complaining about long ago 06:30
dalek rrot: r49488 | dukeleto++ | trunk/t/pmc/boolean.t:
[pmc] Add tests for converting Boolean PMC to a numeric, which should bring the coverage up to 100%
 06:35
dukeleto sorear: it seems like a fairly important PMC 06:40
Also, even the Integer PMC has < 60% test coverage. We need to fix that. 06:41
Socket PMC's should have a get_bool vtable, right? They don't currently. 07:27
cotto What would it represent? 07:29
open/closed? 07:30
dukeleto actually, it exists in the Socket PMC but not the Sockaddr PMC, but cloning a Socket seems to give a Sockaddr, which is odd 07:32
cotto: it is documented to return if the socket is open
cotto Ah. Following the docs is a good idea. 07:33
dukeleto Yep, cloning a Socket returns a Sockaddr. Odd.
cotto quite odd 07:34
dukeleto cotto: i think it was returning the wrong value from it's vtable, seeing if I can make my test pass now 07:36
cotto return SELF for clone doesn't make sense
dukeleto cotto: yep, it should return data_struct, i think 07:37
dalek rrot: r49489 | dukeleto++ | trunk/t/pmc/socket.t:
[t] Add a test for cloning a Socket PMC
cotto I'd think "copy".
data_struct not being a PMC
dukeleto cotto: hmmm. indeed. 07:38
cotto It makes me sad that that's been around since r37705 07:41
dukeleto++ for increasing test coverage.
dukeleto no one really uses that PMC 07:42
cotto We should pick a couple PMCs and bump up their coverage for the next weekly goal.
dukeleto i want to apply the ipv6 patch we have waiting in Trac, so I am adding some basic tests to Socket
cotto +1
dukeleto cotto: returning "copy" still is a Sockaddr 07:43
cotto Oh my. That's a sizable patch.
dukeleto cotto: SUPER of a Socket must be Sockaddr 07:44
cotto We should start a branch for it so it doesn't get too stale.
dukeleto cotto: yes, large patch with no tests. But they are cool features. If they work.
cotto: that is an idea
cotto Or not. I don't think the files it touches get modified very often.
dukeleto cotto: no, they don't. Not currently. 07:45
cotto: where do i look up what SUPER exactly means? 07:47
cotto It calls the same VTABLE function of the PMC's parent.
iirc
verified 07:48
dalek rrot: r49490 | dukeleto++ | trunk (2 files):
[t] More tests for Socket PMCs, including one TODO test for cloning a Socket
 08:07
rrot: r49491 | dukeleto++ | trunk/t/pmc/socket.t:
[t] Add tests for creating TCP and UDP sockets
 08:23
cotto www.slideshare.net/cog/chmod-x-chmod 08:50
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#517) fulltest) at r49491 - Kubuntu 10.10 RC amd64 (g++-4.5 with --optimize) 08:51
08:53 M_o_C joined 09:25 tadzik joined
mikehh rakudo (668719b) builds on parrot r49491 - make test PASS, make spectest_smolder(#518) FAIL - Kubuntu 10.10 RC amd64 (g++-4.5 with --optimize) 10:16
27,049 ok, 3 failed, 611 todo, 1,888 skipped and 0 unexpectedly succeeded
t/spec/S09-typed-arrays/arrays.rakudo seems to be having problems 10:17
moritz that file works on my machine 10:18
got a test output at hand?
mikehh moritz: I have run it a couple of times and it seems to hang in different places, in the test I did I was about to kill it when it completed 10:20
moritz mikehh: we should probably disable the test, types arrays are horribly broken at them moment anyway 10:22
mikehh moritz: it has hung on test 6 for me and now after not ok 35
moritz: it has now completed - 14m35.571s 10:23
moritz wow
mikehh so test 36 seems to take ages 10:24
but it gave me ok 36
moritz: re-running it gets to test 36 fairly quickly, but is now waiting on that 10:26
moritz: 10:30
dies_ok { @x[0, 2] = 2, 3},
'can not assign values of wrong type to a slice';
moritz runs into the recursion limit 10:31
mikehh moritz: up tp 402.4 MiB and 8 minutes so far 10:33
moritz mikehh: I've disabled that test for now 10:36
mikehh moritz: it completed in 14m13.912s this time - I re-run the build and tests again later 10:48
moritz nearly all of that time was in that test 30
10:50 tadzik left 11:14 allison joined
allison celebrates release day on other project is.gd/fUEo0 11:15
11:26 allison left 11:30 ellefu left
moritz I can see why the want to release their 10.10 on 2010-10-10 :-) 11:34
s/the/they/ 11:35
or maybe "you" :-)
11:53 tadzik joined 12:21 M_o_C left 12:34 ruoso joined 13:18 cotto left 13:20 Psyche^ joined 13:21 Patterner left, Psyche^ is now known as Patterner 13:49 he_ left 14:18 theory joined 14:21 hudnix left 14:28 cotto joined, he joined 14:29 theory left 14:41 kid51 joined
dalek rrot: r49492 | jkeenan++ | failed to fetch changeset:
Merge tt1810_missing_step_tests branch into trunk. Only report missing step tests when --test or --test=build is used.
 14:47
kid51 Getting failures in t/tools/mk_language_shell.t. See smolder.parrot.org/app/projects/rep...etails/522 14:49
14:50 hercynium joined
nopaste "kid51" at 192.168.1.3 pasted "2 failures in mk_language_shell.t" (26 lines) at nopaste.snit.ch/24177 14:51
mikehh damnit Jim - I thought I tested that on i386 14:56
14:58 preflex left
mikehh I did and it passed for me - smolder.parrot.org/app/projects/rep...etails/512 14:59
15:02 preflex joined 15:04 jan left
kid51 mikehh: Hey, I'm just the messenger here! 15:04
mikehh: This is the first time I've looked at that test file. 15:07
Why is it testing $? rather than $build_status and $test_status (variables defined but not used)?
mikehh kid51: yeah - it still passes for me but looking at smolderr reports it is failing on other platforms - I think it might be a realclean problem 15:08
kid51 OTOH, I get the same failures with $build_status and $test_status 15:09
mikehh: I did make realclean;perl Configure.pl; make; prove t/tools/mk_language_shell.t. Got same FAILs. 15:11
mikehh kid51: it is new - first introduced yesterday - somewhere around r49484
kid51: quite a bit of discussion on irc yesterday 15:12
kid51 One problem: The two failing tests rely on Perl system calls. There are a myriad (is a myriad?) of reasons why the system call might return a non-zero value.
kid51 is on vacation and was mostly up in the air yesterday 15:13
kid51 has not backscrolled for several days 15:14
15:16 lucian joined
mikehh kid51: well enjoy the vacation, got to go out for a bit, will check later 15:16
dalek rrot: r49493 | jkeenan++ | branches/tt1810_missing_step_tests:
Branch has been merged into trunk and is no longer needed at HEAD.
 15:18
15:35 plobsing joined
dukeleto 'ello 15:54
It seems that calling clone on a Socket actually calls the clone VTABLE on Socket's parent, Sockaddr, which is why cloning a Socket returns a Sockaddr 16:01
Also, I think SockAddr is a horrible name. SocketAddress anyone? 16:02
16:10 kid51 left 16:11 jan joined
dukeleto jan: howdy 16:15
16:15 tadzik left
nopaste "moritz" at 192.168.1.3 pasted "what's wrong here?" (10 lines) at nopaste.snit.ch/24190 16:21
moritz Method 'reverse' not found for invocant of class 'String' # in the nopaste 16:22
but PDD28 mentions that there should be one in the String PMC
and that it's even implemented for ASCII strings 16:23
dukeleto looks 16:25
moritz: off the top of your head, do you know if Rakudo uses Parrot Sockets or Sockaddr's at all?
moritz dukeleto: I'm pretty sure it does 16:26
# Create the socket handle 16:27
sock = root_new ['parrot';'Socket']
from src/core/IO/Socket/INET.pm
dukeleto moritz: i can reproduce your reverse issue, not sure what is going on 16:29
moritz: i will look in the string tests to see what I can see 16:30
moritz only mention of reverse there is in reverse_index 16:31
dukeleto moritz: string.pmc does not seem to implement a reverse method 16:33
moritz opened tt#1821 16:35
dukeleto moritz++ 16:38
moritz: just a note, we have a bug in our Sockets such that cloning them doesn't work properly. Keep that in mind if you see any funny business with sockets. 16:39
16:47 hercynium left 16:52 particle left 16:58 cognominal joined
dalek rrot: r49494 | dukeleto++ | trunk/t/pmc/socket.t:
[t] Add some tests for closing Sockets
 17:05
dukeleto cotto: can you add something about your improvements to the profcore to NEWS? 17:22
17:29 mikehh left
dalek rrot: r49495 | dukeleto++ | trunk/NEWS:
[doc] Give NEWS some love
 17:36
rrot: r49496 | dukeleto++ | trunk/t/pmc/socket.t:
[t] Add tests for creating raw TCP and UDP sockets
 17:51
rrot: r49497 | dukeleto++ | trunk/t/pmc/socket.t:
[t] Add tests for creating IPv6 sockets
17:57 tadzik joined
dalek rrot: r49498 | dukeleto++ | trunk/t/pmc/sockaddr.t:
[t] Add a TODO test for get_bool on a SockAddr PMC
 18:06
18:17 particle joined
dalek rrot: r49499 | dukeleto++ | trunk/docs/pdds/pdd22_io.pod:
[pdd] Fix a fib in PDD22 about get_bool on a Socket
 18:22
18:44 PerlJam joined 18:48 PerlPilot left
dukeleto goes on a socket-related test binge 19:29
cotto dukeleto, most definitely 19:34
19:39 davidfetter left
cotto done 19:41
dalek rrot: r49500 | cotto++ | trunk/NEWS:
[NEWS] add PCore improvements to the news
 19:54
20:25 davidfetter joined 20:26 M_o_C joined 20:27 perlite left, perlite joined 20:32 eternaleye left 21:11 darbelo left 21:16 M_o_C left 21:30 bacek joined 21:53 bacek left
dalek rrot: r49501 | dukeleto++ | trunk/t/tools/mk_language_shell.t:
[t] Improve diagnostic messages in tests for mk_language_shell.t
 21:57
23:45 szabgab left, szabgab joined 23:51 silug left

Powered by Cro and the Raku® Programming Language.

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

Thank you!