Parrot 1.0 Released | parrot.org/
Set by moderator on 17 March 2009.
00:01 davidfetter joined
allison gravity: I've added you to the Debian Parrot group, you can subscribe to the list at lists.alioth.debian.org/mailman/lis...rrot-devel 00:03
shorten allison's url is at xrl.us/bekn2i
00:05 shoe joined 00:09 AndyA joined 00:11 davidfetter joined
moritz moritz.faui2k3.org/files/parrot-com...er-day.png average number of commits per day, averaged over release cycles 00:12
updated with better title 00:13
00:13 dmkim joined
moritz feel free to use it under the terms of the WTFPL ;-) 00:13
off to bed now
cotto moritz++
wtfpl++
karma wtfpl 00:14
purl wtfpl has karma of 3
cotto uh-oh 00:15
test failure
purl SOUND THE ALARM! THE HULL OF THE CODEBASE IS ABOUT TO GIVE IN!!
cotto it's minor: t/pmc/packfile.t expects the major version number to be 0
moritz bah. 00:16
bad test
japhb rurban: t/library/opengl.t WFM on i486-linux-gnu -- except that the test plan is off by 1. :-)
cotto nm. it appears to be a trunk issue rather than something that'd mess up the release
rg cotto: are you sure you're all current? 00:17
allison cotto: I updated that test 00:18
cotto I'm checking what's goofy
looks like some stale files
false alarm
00:21 particle joined
particle umm.... i've got a problem. 00:22
so i installed parrot 1.0.0
i'm trying to build lolcode
Can't open perl script "C:/Parrot/lib/parrot/1.0.0-devel/tools/dev/gen_makefile.pl":
No such file or directory
seems the tools aren't installed 00:23
allison particle: did you run "make install-dev"?
particle no, make install
doesn't matter, anyway
if i just do 'nmake', and try to build lolcode 00:24
allison the tools are only installed with "install-dev"
particle ok
if i *don't* install, and try to build lolcode from a worknig copy, it fails
parrot_config libdir
allison particle: yes, it's an example language
particle comes back as the install path 00:25
allison that runs on an installed parrot
particle so there's no dev version of parrot_config?
Tene particle: yes, the language stuff requires an installed parrot and an installed parrot-dev to build
particle ok, fine. i've got that working now.
so, i can pack up, and do my presentation 00:26
excellent, thanks!
cotto what's he presenting? 00:41
nm. probably lolcode like at yapc 00:42
allison cotto: the Seattle Perl Monger's group meets on the 3rd tuesday of the month (same day as parrot releases, by coincidence) IIRC, particle said he's presenting there tonight 00:44
cotto I really wish I didn't have an ongoing previous commitment. 00:45
although it does mean I don't have to care about rush-hour traffic going into Seattle 00:46
00:55 s1n joined
Whiteknight allison: is there going to be some kind of parrot meeting at YAPC? 01:02
jsut there's the workshop, but that's probably not exactly what you mean 01:03
Whiteknight yeah, I'm more interested in something like PDS
allison yes, it's in the plans 01:04
Whiteknight because I have vacation time I'm saving up to do something Parrot-related this year, and I'm trying to get a feel for when that will be
allison there's space to do it the weekend before, but would likely be distracting with people who would be in the meeting having to head over to a different room to give talks 01:05
we often do a hackathon in the days after YAPC
and, may make that the PDS location for the next round
cotto is t/src/extend.t explodey.t for anyone else?
(I did run make realclean and make sure the code is updated) 01:06
allison cotto: all tests pass for me
Whiteknight yeah, test #6 fails for me
#16 that is
allison cotto: platform, etc?
cotto 12-14 fail for me
ubuntu x86
Whiteknight although it's actually marked TODO, so that's no issue 01:07
allison intrepid, jaunty?
that is, 08.10?
cotto hardy
Whiteknight I'm on Intrepid
cotto A message in the error output says 'This Parrot cannot read bytecode files with version 4.0." 01:08
sorry to report and run, but that's what I've got to do
bbl
allison cotto: okay, explore more later
dalek tracwiki: v79 | allison++ | Languages 01:27
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bekobw
dalek tracwiki: v80 | allison++ | Languages 01:31
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bekob8
dalek tracwiki: v81 | allison++ | Languages
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bekoca
01:32 rurban_ joined 01:37 particle joined
rurban particle: Can't open perl script "C:/Parrot/lib/parrot/1.0.0-devel" you had the file DEVELOPING around 01:41
It should be "C:/Parrot/lib/parrot/1.0.0/tools/dev/gen_makefile.pl" 01:42
01:45 jrockway joined 01:47 eternaleye joined
allison rurban: that just means he was compiling from a checkout instead of the tarball, parrot_config is smart enough to hold the right path either way 01:47
rurban Hmm. I cannot think of any other error then. 01:50
Anyway, it's 3am. I'm gonna sleep. Thanks for 1.0! 01:51
01:59 davidfetter joined 02:03 kid51 joined 02:24 dduncan left
dalek rrot: r37538 | jkeenan++ | branches/update_pod:
Branch was merged into trunk at r37193 and so is no longer needed at HEAD.
 02:25
rrot: r37539 | jkeenan++ | branches/remove_pmc_option:
Create branch to work on trac.parrot.org/parrot/ticket/310.
 02:33
rrot: r37540 | jkeenan++ | branches/remove_pmc_option (5 files):
Remove --pmc option to Configure.pl.
 02:45
02:59 davidfetter joined
Tene Spinning on a loop from 1 to 100000000, parrot beats perl 5. 03:09
btw. 03:10
pmichaud: ping 03:15
03:36 janus joined 03:37 jsut joined 03:40 jsut joined 03:42 tetragon joined 03:46 jsut joined 03:48 jsut joined 04:40 Andy joined
cotto allison, the test failure was because of an old installed parrot 04:55
it'd be nice if we had a make uninstall to take care of that
04:57 PacoLinux joined 04:59 NotFound joined 05:20 tuxdna joined
allison cotto: I wondered if that might be it 05:38
Tene allison: do you know who moderates perl6-users?
allison cotto: the solution is to use a versioned libparrot 05:39
Tene: Ask and Robert
purl Ask and Robert are idiots?
allison purl: forget Ask and Robert
purl allison: I forgot ask and robert
allison Tene: what do you need? 05:40
purl hmmm... do you need is rw?
Tene allison: I posted, and my message hasn't gone through and hasn't sent back a rejection, so I presume it's stuck in moderation.
diakopter purl, forget
purl hmmm... forget is not a good thing to have optional
diakopter purl, forget all
purl diakopter: I forgot all
allison Tene: yes, probably 05:41
05:49 rblackwe joined
Tene allison: install-dev doesn't install dynpmc.pl ? 05:50
allison Tene: dynpmc.pl doesn't work on an installed parrot, it has hard-coded paths
Tene Ah.
allison that's a large part of why it's deprecated 05:51
diakopter has 1.0 been tested on msvc
Tene I'm looking at things that use BUILD_DIR in rakudo's makefile.
dalek rrot: r37541 | allison++ | trunk/ports/debian/changelog:
[debian] Update timestamp for cutting real packages.
allison Tene: it's easy to replace with a Makefile template and a call to gen_makefile.pl 05:52
(which is installed)
there should be a Rakudo patch in the queue to do it
diakopter ?
allison Tene: but, it means switching over to building from an installed parrot 05:53
diakopter: yes, it has
(I was about to thank you when I realized you were asking, not reporting that you'd tested it)
diakopter ah ok 05:55
who builds the packages for sourceforge?
dukeleto how do I run a rakudo test through the parrot debugger? I need to convert the test to a .pbc ? 05:57
diakopter oh, sad that ftp.parrot.org does not mirror ftp://ftp.parrot.org
allison diakopter: do you mean the parrot-win32 packages? 06:02
dalek rrot: r37542 | allison++ | failed to fetch changeset:
[debian] Updating docs list for release.
 06:04
Tene *sigh* 06:20
Looks like pmcs are broken with HLL namespaces
still.
dalek kudo: 1ca2109 | (Stephen Weeks)++ | build/Makefile.in:
Fix a bug with rakudo trying to rebuild pbc_to_exe
shorten dalek's url is at xrl.us/beko2y
Tene I don't remember if there was a ticket for it or not. 06:21
cotto out with the old 06:31
dalek rrot: r37543 | cotto++ | trunk (10 files):
[hash] make various hash functions return Hash*, remove now-unused functions
 06:32
cotto and now to kill the Slice PMC 06:34
dalek rrot: r37544 | cotto++ | trunk/DEPRECATED.pod:
[DEPRECATED] remove deprecation notice for src/hash.c update
 06:36
rrot: r37545 | allison++ | trunk/ports/debian (2 files):
[debian] Also wants the Standards-Version increased, to 3.8.1.
 06:40
Tene allison: should the hll pmclass attribute eventually set that class in the approrpiate hll namespace? 06:42
allison Tene: yes 06:43
Tene allison: it looks like it currently doesn't, can you confirm? 06:44
cotto allison, does PMC removal require a PBC_COMPAT bump?
allison Tene: C-level PMCs shouldn't pollute the global namespace any more that PIR-level ones should
Tene: yes, I can confirm that currently C-level PMCs don't respect namespaces at all
Tene I'm running into trouble with c-level pmcs when trying to get rakudo into its own .HLL
allison cotto: yes, a minor version bump 06:45
cotto In general, what changes require that? 06:46
and are they documented somewhere?
allison Tene: yes, you will, though documenting the problem you're having is useful information for the HLL milestone coming up before 1.4
cotto: PBC_COMPAT itself has a list, but it's not written entirely clearly 06:47
cotto: the general rule of thumb is that if it affects what's stored in the bytecode, it needs a version bump 06:48
06:50 particle joined
Tene APL's configure.pl is *completely* useless and broken 07:01
07:01 tuxdna joined
dalek rrot: r37546 | cotto++ | failed to fetch changeset:
[PMC] remove the Slice PMC and bump PBC_COMPAT
 07:01
Tene chdir '../..'; `$^X -Ilib tools/dev/reconfigure.pl --step-gen::languages --languages=APL`
cotto That certainly looks APL-worthy. 07:05
Tene Who has commit rights to paraplegic?
07:06 UltraDM joined
allison Tene: Coke does, I don't think anyone else does yet 07:07
Tene: a lot of the old languages had that (I think that's what the language shell generator used to create 07:08
Tene nods.
I just copied from rakudo and modified.
seems to work.
cotto allison, the Default PMC's get_attr_str has zero test coverage and uses the (also untested and deprecated) Bound_NCI PMC. Is it ok to remove that VTABLE function? 07:20
There's also a similar branch of an if statement in object.pmc's find_method. 07:21
allison cotto: can deprecate it, but can't remove it until after 1.4 07:22
cotto the VTABLE function?
allison cotto: what's it using Bound_NCI for?
cotto: yes, the VTABLE function, the PMC was deprecated
cotto: that is, can it be replaced with something else (the modern equivalent)? 07:23
cotto I don't know if it can be replaced with something modern. 07:24
or if it needs to be 07:26
07:27 uniejo joined
cotto I'll leave it alone for now. 07:31
07:42 bacek joined
bacek g'day 07:42
parrot.org is down...
07:54 pnu joined 07:57 Ademan joined
pnu allison et al: Congratulations on 1.0 release! 07:58
moritz bacek: denial of service? ;-) 07:59
allison pnu: thanks! 08:03
moritz allison: do we have any figures already? number of downloads? or hits on the web page? 08:04
allison moritz: we aren't tracking either, though apparently the server is under quite a load at the moment 08:05
08:06 mj41 joined
mj41 Hi. Down, but not on Slashdot yet. 08:11
08:13 zamolxes joined
zamolxes congrats! :) something wrong with the site though 08:13
08:13 riffraff joined
moritz allison: don't forget the announce on perlmonks 08:16
afk
cotto what's bogging down the server? 08:19
allison cotto: traffic, as far as I can tell 08:21
moritz that's the only good way of brining it down ;-) 08:22
allison cotto: I've pinged the admins
I can't even ssh in
zamolxes varnish, set up varnish
www.catalystframework.org/calendar/2008/14 quick tutorial :) 08:23
08:34 masak joined 08:43 saurabh joined 08:53 tuxdna joined, korshak joined 09:14 teejay joined
teejay wonders if the parrot.org server is overloaded with traffic or just pining for the fjords 09:15
09:15 acme joined
moritz teejay: it's on reddit fronpage (and reddit programming near the top), so likely "just" traffic 09:15
teejay ..for the last 7 hours judging from reddit (yes, I know) comments
it can't be *that* much traffic 09:16
surely it's mostly static content
acme presumably the vm died and we have to wait for americans to wake up? 09:17
teejay and even a small virtual server can saturate a lot of bandwidth
yes, that sounds more like it
GeJ svn seems dead too. I can't even get passed the SSL negotiation.
cotto It certainly didn't take long for the server to melt. 09:19
szabgab on the other hand ftp://ftp.parrot.org/pub/parrot/releases/stable/1.0.0/ answers 09:20
GeJ I can already hear the slashcrowd : "Parrot 1.0 released, already dead" bummer.
teejay chromatic announced on reddit 7 hours ago
seems to be hanging from about 1 or 2 hours ago 09:21
GeJ, jokes like that already in reddit comments
a bit embarrising
shame there are no european admins for the server it's running on 09:22
that's odd 09:24
it's not even mentioned on /. 09:25
but then slashdot is usually up to a week behind even theregister these days
09:26 korshak left 09:34 rurban_ joined 09:41 bacek joined 09:52 Debolaz joined 10:07 pancake joined
pancake is parrot.org down? 10:07
teejay yes, has been for a couple of hours 10:08
somebody prod a yank ;)
10:08 teejay left
cotto only 4-5 hours until business hours 10:08
moderator Parrot 1.0 Released | parrot.org/ (down atm) 10:09
acme left 10:12 kappa joined 10:21 tuxdna joined
xinming_ Bird met hunters... For www.parrot.org's death. :-) 10:31
10:44 kappa_ joined 10:56 dmkim left 10:57 neuralis joined
neuralis hi. the website appears to be melting; would you guys be helped by some immediately available server capacity? 10:58
Nom heh... i noticed that :) 11:02
11:18 ujwalic joined 11:23 gaz joined 11:36 vti joined 11:50 simontwo joined
simontwo would it make sense to have an application server powered by parrot? 11:52
it seems that the popular application servers rely on JVM or .NET and not particular languages. 11:53
and hi. :) 11:58
wow. ubuntu's parrot package: Version: 0.4.13-1ubuntu3 12:00
12:16 Gerd joined 12:21 Gerd left 12:24 explorer joined
moritz that's antcient 12:33
12:34 rg joined
dalek kudo: a197d2a | (Moritz Lenz)++ | t/spectest.data:
more passing test files for t/spectest.data
 12:40
shorten dalek's url is at xrl.us/bekpgo
Patterner "parrot-0.6.1.ebuild" 12:43
moderator Parrot 1.0 Released | parrot.org/ (down atm) | Congratulations and Thanks to the guilty parties! 12:44
gravity allison: Great, thank you! I'm on the list and just saw your preliminary package message. I'll have a close look at the packaging today. 12:45
moritz gravity++ 12:47
Coke tene: if you want commit bits to APL's new home, pm me. 13:02
13:03 rdice joined 13:12 riffraff joined
Coke Unfortunately that parrot.org seems /.'d 13:25
rdice: hey, richard.
purl well, richard is going to Foxtons
neuralis Coke: do you know who administers parrot.org? 13:26
rdice Coke: g'day mate. 13:29
13:32 uniejo joined
Coke neuralis: yes, the OSU folks. a ticket has already been opened. 13:35
tene?
purl i think tene is Stephen Weeks
13:45 gryphon joined 13:49 davidfetter joined
xinming_ Is parrot.org still frozen? 13:55
Coke seems to be. 14:05
davidfetter tries to pour in some antifreeze
14:05 saurabh_ joined
Coke I can't even ssh in to our virtual box. 14:06
just got a reply, OSU is investigating.
ah, I can ssh in. just very slow l y 14:07
mikehh I managed to do an svn up on parrot - after trying all morning - it's 14:11 here 14:11
Coke my ssh is still trying to process my password. 14:21
nopaste "rurban" at 93.82.85.15 pasted "for NotFound, more string encodings" (174 lines) at nopaste.snit.ch/15907
Coke osu reports, not surprisingly, "a lot of web traffic".
nopaste "rurban" at 93.82.85.15 pasted "for NotFound, more string encodings (oops)" (174 lines) at nopaste.snit.ch/15908 14:22
gravity moritz: :-) 14:23
gravity waves to neuralis
neuralis gravity: fancy running into you here 14:31
14:32 jsut joined, teejay joined
teejay wonders when parrot.org will return 14:32
been down for most of the day now 14:33
moderator Parrot 1.0 Released | parrot.org down; unknown ETA for restoration 14:38
eternaleye joined 14:40 particle1 joined 14:41 estrabd joined
jsut it just loaded for me :x 14:42
teejay crikey finally loaded for me too 14:44
moderator Parrot 1.0 Released | parrot.org | Thanks to everyone who helped get 1.0 out the door! 14:46
teejay hmm 14:46
docs.parrot.org/parrot/latest/html/info.html # all links 494
404 even
Coke haru tatsu?
14:46 mvuets joined
Coke haru tatsu is Spring Begins 14:47
(or so a google search leads me to believe)
teejay wonders what makes this 1.0 14:48
i.e. what milestones met vs 0.x
particle1 teejay: i can access that link
Coke particle1: no, the links ON that page.
teejay: looks like the 1.0 docs were not installed as part of the release. I'm doing that now.
particle1 ah, yes. the links don't have the proper root 14:49
Coke AIGH
the ftp link on the release notice is a 404.
teejay oops
Coke particle1: do you have ftp access?
PacoLinux Parrot 1.0 announce in barrapunto (spanish /.) : barrapunto.com/article.pl?sid=09/03/18/0844241
particle1 supposedly
mvuets Hi there! Sorry for the stupid question: is this the same channel as #parrot@irc.parrot.org?
particle1 mvuets: yes 14:50
Coke mvuets: yes.
I'll just grab it from the tag. bother.
mvuets Oh, ok. 'cause i just connected to @irc.perl.org (:
Another dumb one: does exist a mirror of parrot.org? I cannot reach it right now (i assume because of release) 14:52
Coke can someone open a trac ticket for the missing docs and the broken ftp link?
mvuets: try again.
teejay Coke, where is the trac ?
Coke trac.parrot.org
particle1 coke: what's broken?
mvuets Coke, it is down for several hours already.
teejay mvuets, yes
particle1 mvuets: parret.org is back up 14:53
er, parrot.org
teejay been down since 10am uk time
Coke particle1: you should reserve parret.org. =-)
particle1 i may :)
teejay trac. and www. both v v v slow to connect
docs seems quicker
Coke trac is trac, www is drupal, docs is straight apache. 14:54
particle1 just to make sure there hasn't been a dns change... i get parrot.org as 140.211.167.206
teejay yup www.parrot.org still too busy to take connections
jsut perhaps load problems?
Coke certainly load problems.
particle1 teejay: nslookup parrot.org
purl particle1: parrot.org is 140.211.167.206
davidfetter is anybody here on f10?
mj41 Is main page cached?
teejay I assume no caching and drupal not very good at handling load 14:55
mj41 Slow database?
jsut throw squid in front of it
Coke dammit. my svn checkout of the 1.0 tag aborted, too.
mvuets Still cannot reach ):
davidfetter time to git! :)
particle1 wonders if coke has connection problems
teejay mj41, lathos' stock maypole coped with slashdot running on a pish virtual server with mysql/modperl/maypole
Coke particle1: everyone does. or do you mean something specific?
teejay drupal must suck goats to not cope 14:56
Coke oh, regarding the ftp? that might be a bad proxy on my end, sure.
particle1 i can reach the downloads page, save the tarball, and check out parrot
Coke my export (not checkout) died with an svn error. since it was an export, I couldn't resume where I left off.
ok, so nevermind a ticket on the ftp missing. =-)
One of these days I'll remember I can't hit that frmo work. 14:57
still working on the docs.
mvuets Are windows binaries available for 1.0? I have a crash on 0.9.1, wanna check it with the recent version.
diakopter I was gonna suggest making ftp.parrot.org/pub be an http mirror of ftp://ftp.parrot.org/pub
teejay I can't add trac ticket, trac not responding
particle1 mvuets: likely today. we're waiting news from our brazillian parrot windows packager
Coke mvuets: yes. See the mailing list.
moment.
mvuets particle1, i see, tnx.
Coke lists.parrot.org/pipermail/parrot-d...01786.html 14:58
shorten Coke's url is at xrl.us/bekpvo
mvuets Coke, i'm new to all of this, sorry (:
Coke np
welcome aboard.
mvuets (:
Coke I'm not upset, just terse. =-)
teejay can't contribute anything useful, can't resist dissing drupal, probably best to head off and do something more useful ;) 14:59
good luck getting the server back to life
particle1 coke: btw you'll get used to the bike seat, just give it two or three more rides
14:59 teejay left
Coke particle1: ... I've had this seat for a year. 15:00
bah. if I do a checkout of the tag from 1.0.0, the generated html says 1.0.0-devel
trying the tarball through another proxy. 15:01
can someone with edit privs note on the download page that the macports version is out of date? 15:02
particle1 coke: but you have to ride the bike to get used to it!
coke: you may have to make release... does the tag include DEVELOPING? 15:03
maybe our release procedure needs a small fix....
Coke what's the real address for the ftp tarball?
(something with no redirects)
particle1 ftp.osuosl.org/...
Coke wget reports:
No such file `1.0.0'.
ah, needs the filename. wow. can't remember the last time I had to fire up an actual ftp client. 15:04
(why doesn't the link include the filename?)
maybe it's not my proxy, maybe it's my browser. 15:05
particle1 ftp://ftp.osuosl.org/.2/parrot/releases/stable/1.0.0/parrot-1.0.0.tar.gz 15:06
shorten particle1's url is at xrl.us/bekpwn
15:08 frodwith joined
Coke docs.parrot.org/parrot/1.0.0/html/ 15:12
(I'll add in the links shortly)
mvuets Guys, setup-parrot-1.0.0.exe + setup-parrot-1.0.0-rakudo-20090318.exe does not work as well as my old 0.9.1. Under W2K3. But it works under XP. How i can get sources to build debug version myself? 15:17
particle1 www.parrot.org/download 15:18
coke: can you kick particle?
mvuets particle1, sorry, i still cannot hit it.
Coke 11:18 -!- particle #parrot They aren't on that channel 15:19
mvuets Please give me direct repo link.
particle1 ok, thanks :( 15:20
Coke devel, stable, and latest on docs.parrot.org now all point to 1.0.0 15:21
mvuets: 11:06 <@particle1> ftp://ftp.osuosl.org/.2/parrot/releases/stable/1.0.0/parrot-1.0.0.tar.gz 15:22
shorten Coke's url is at xrl.us/bekpwn
mvuets Coke, gotcha.
Coke there's also svn.parrot.org/parrot/tags/RELEASE_1_0_0 (but you'll need to do a make release out of that to get the tarball available via ftp)
15:32 bkuhn joined
NotFound There is some problem with parrot.org servers? 15:34
neuralis NotFound: yeah, it's being investigated. 15:36
NotFound Now it works, but slow.
This morning (CET) I was unable to svn or see the web site. 15:37
particle1 coke++ 15:47
notfound: effectively, we got slashdotted
don't know if we actually did (i *never* go there), but traffic was too high and something popped 15:48
we've had the admins enable caching, but the vm is running out of memory, causing timeouts 15:49
moritz particle1: actually we got redditted 15:51
particle1 aha. never go there, either :)
NotFound Not so bad, it shows that we are catching people's attention :)
neuralis particle1: i popped in here earlier offering extra server resources that can be brought up immediately. offer still stands.
15:52 mvuets left
particle1 neuralis: thanks for the offer! i'm not quite sure how that would work, but i can put you in touch with our server hosts if you'd like to discuss with them 15:53
diakopter may I suggest making ftp.parrot.org/pub an http mirror of ftp://ftp.parrot.org/pub (for those of us without ftp access)?
neuralis particle1: in the immediate term, i can provide e.g. www2.parrot.org, which would allow the osu guys to have apache issue temporary redirects to www2
particle1 neuralis: you can get drupal set up/configured quickly? 15:54
moritz or just proxy the site? 15:55
neuralis particle1: not from scratch, but certainly if you have copies of your database and config
particle1 sure.
moritz have the admins create a real.parrot.org, mirror it on www2 and redirect www there
or something
neuralis moritz: that's what i was thinking, yeah.
particle1 neuralis: who's making this offer? business name? 15:57
neuralis particle1: personal offer, radian.org
Coke I suggested to another proffer of services to email parrot-directors@parrot.org 16:07
as then allison and the board will have a chance to see it/discuss it.
16:08 ruoso joined
Coke some relief from the OSU admins. 16:10
many levels of caching and tweaking, and upgrades to drupal & trac coming "RSN".
osuosl++
of course, i still can't load the page. =-) 16:12
Coke grimaces at the inevitable "dead parrot" comments on the reddit link. 16:13
www.reddit.com/r/programming/commen..._released/
shorten Coke's url is at xrl.us/bekp8k
Infinoid dalek is polling trac twice every 3 minutes, and drupal once every 3 minutes. Should I turn that off for now?
Coke Way old if you ask me. You can already do this on the JVM and the CLR...with ruby and python.
(from the comments page)
allison Infinoid: yes please :) 16:15
Coke "Yup, Duke Nukem forever should be out any minute now."
16:17 dalek joined
Infinoid Ok, please remind me to reenable those when the storm has passed 16:17
16:17 korshak joined, korshak left
Infinoid Ohnoes. "You don't have permission to access /parrot/latest/ on this server." (from docs.parrot.org/parrot/latest/) 16:18
Coke moment.
try now
oh. 16:19
bad Infinoid. wrong url.
docs.parrot.org/parrot/latest/html/
Infinoid ah, ok
Coke go to docs.parrot.org/ and it will redir you to the right url 16:20
Infinoid Yeah, that works. I just had the bad link in my browser history for whatever reason, and it looked right so I selected it
rg coke: how do we know what version that html is from (i.e. has it been updated for 1.0 yet)?
Infinoid It's in the page title, rg 16:21
and yes, it says 1.0.0
particle1 maybe latest/ should redirect to latest/html/
it's in infinoid's browser history, so he's bound to get it wrong again :P
Coke rg: Read the bottom.
particle1: That's something perhaps our apache admins could fix for us. 16:22
(and they can fix the top level html-based redirect at the same time.)
Infinoid It was probably only in my browser history from the time when I was begging and pleading Coke to add the latest symlink... I don't think anything actually links to it.
rg ah. i just got an incomplete page 16:23
but may i say that dark blue on black is *really* hard to notice ;)
Coke rg: oh, i agree. 16:31
please submit a patch against...
docs/resources/*.css ?
allison Coke: parrot.css 16:32
Infinoid "Subversion? FTP!? U2 lyrics!?!? Uh oh." -- reddit comment 16:38
jonathan Clearly git, gopher and metallica woulda made Parrot 1.0 so much better. ;-) 16:41
moritz Infinoid: at least it has low reputation
16:42 kj joined
mikehh rakudo (a197d2a) builds on parrot r37546 - make test/make spectest PASS - Kubuntu Intrepid i386 16:47
Coke any suggestions on when to remove old docs from docs.parrot.org ? 16:48
I'm fine with 'never'. =-)
(but 0.9.1 is now orphaned)
moritz never 16:53
I see no reason to break links
rg what links?
purl links is the linker rel
moritz external links to the docs
rg purl forget links
purl rg: I forgot links
particle1 i'm fine with 0.9.1 being orphaned 16:54
diakopter latest developer download /release/developer points to 0.9.1 16:55
allison diakopter: what url? 16:56
16:56 Psyche^ joined
allison oh, yes, 0.9.1 is the latest developer release 16:56
diakopter: that will change to 1.1 next month
rg but developers should also download 1.0
17:27 rdice joined
Coke allison: I made 1.0.0 the latest 'devel' on docs. 17:28
I think that makes sense.
devel is basically 'latest'
Should be the same for releases, too, I think. 17:29
(otherwise, those trying to get 'bleadparrot' will instead get a potentially 2 month old releaes.
I think stable is for 1.0, 1.4, latest is whatever the latest release is, devel should probably be the snapshot of svn. 17:37
17:43 skv_ joined, shoe left 17:51 DietCoke joined 17:52 parrotCoke left
Coke dukeleto: are you the perl.org gsoc guy? any news? 17:57
17:59 korshak joined
particle1 the announcements to orgs will be made today 18:04
cotto Do we care about the branch_cs instruction? It's buggy and doesn't appear to be used at all during make test. 18:08
tt filed 18:17
18:23 ujwalic joined
Coke particle1: the announcements are already made. 18:23
they're just not public yet.
particle1 sorry, i meant the public announcements. 18:24
Coke which is why i was trying to poke leto.
I just had this SAME conversation with someone in #perl on freenode.
particle1 i'm the backup org admin for tpf
Coke but I bet you a dollar you didn't get the email. =-) 18:26
18:38 Andy joined
particle1 tpf has been accepted into gsoc 2009! 18:38
Coke yay 18:47
cxreg tye is worried that the M&S gc system means that swap is rendered useless due to page thrash inspecting swapped out but still referenced variables. have you guys solved that issue? 18:49
18:50 chromatic joined
cxreg I thought maybe a compact look-aside reference table could help 18:50
or are we just up the wrong tree entirely? :) 18:51
Coke chromatic: you just missed this send:
14:49 < cxreg> tye is worried that the M&S gc system means that swap is rendered useless due to page thrash inspecting swapped out but still referenced variables. have you guys solved that issue?
chromatic Nope, but I have some ideas. 18:52
particle1 trac.parrot.org/parrot/wiki/GSOC2009Tasklist needs a lot of love. it's going to get hit starting today 18:53
chromatic Note that M&S with liveness stored in each GCable element also ruins COW in applications such as mod_parrot.
particle1 www.perlfoundation.org/perl5/index....9_projects too, of course
shorten particle1's url is at xrl.us/bekq26
particle1 coke: did you school with cxreg? 18:54
Coke particle1: ... what? 18:55
purl quietly listens while the crickets chirp
Coke It's possible. a lot of people went to my college.
cxreg chromatic: makes sense. a separate table with that data seems like the logical solution 18:56
Coke wonders how many codingstd tests cotto broke. =-)
it's not reasonable to expect a python script to have well formed POD, izzit?
chromatic cxreg, it also makes copying/compacting much easier. 18:57
Coke (python) nevermind, it was an easy fix. 18:59
particle1 ah, didn't realize cxreg was here in #parrot 19:08
we do expect our c and .ops and .pmc to have well-formatted pod
Coke particle1: no. python. 19:09
someone just add a python script to the repo.
particle1 yeah, cotto++ added a fuzz tester 19:11
Andy tech.slashdot.org/article.pl?sid=09...18/1826201
Oh, the cranks.
"You quoted U2, you suck."
Coke note to self, don't let allison pick the release quote... 19:12
chromatic "Quoting Monty Python -- a group of comedians known for being absurd and subversive -- makes me funny because no one has done it before in the previous ten seconds!"
gravity mumbles something about Duke Nukem jokes being the worst 19:14
Andy Yeah, 'cause they're not funny
Infinoid They were funny in 2001. I think.
chromatic They stopped being funny in 1997. 19:15
Coke thinks someone broke the build on win32. 19:16
particle1: src/pmc/integer.pmc line 405 19:17
19:17 bsdz joined
Coke I'm trying to run a blame right now but svn is slow 19:17
37546 cotto PMC *key;
hopefully that's post-release.
chromatic This one's my favorite: 19:18
www.reddit.com/r/programming/commen...ed/c08b696
shorten chromatic's url is at xrl.us/bekq7n
bsdz Hi, anyone here who might be able to help with an NCI question?
chromatic Go ahead, bsdz. 19:19
particle1 coke: it's post-realease.
Coke whee.
particle1 i'll do a realclean and rebuild 19:20
cxreg "from the take-a-beak dept. " <- beak?
oh duh. :-)
bsdz thanks chromatic. i need to send my dlfunc a pointer to array of ints. I've been using "3" but after some debugging i think i need to use "p" type.
Coke particle1: danke.
chromatic Yeah, a ManagedStruct sent as the 'p' type. 19:21
particle1 socghop.appspot.com/program/accepte...e/gsoc2009 19:22
shorten particle1's url is at xrl.us/bekq77
bsdz yeah thought so. thanks chromatic
Coke particle1: you might want to update www.perlfoundation.org/perl5/index....9_projects to point to the corresponding parrot page. 19:29
shorten Coke's url is at xrl.us/bekq26
Coke 'make fulltest' should probably run configure tests. 19:30
particle1 the idiom is perl Configure.pl --test && make fulltest 19:31
chromatic Frankly, if you can run 'make fulltest', I don't care much about failures in the Configure tests. 19:32
Coke chromatic: funny you mention that, as i can't run make fulltest.
chromatic No tests pass for you in fulltest?
Coke (but that's another issue) 19:33
no. some tests fail.
but I have a work in progress going on atm, so never mind that aside.
chromatic If some tests in fulltest pass, then Configure worked. 19:34
Coke chromatic: I, as a developer, am trying to avoid breaking those tests. 19:35
I certainly agree the end user shouldn't have to run them unless they're unable to build parrot. 19:36
jsut parrot.org is borked again, i take it this is an ongoing problem?
Coke jsut: only due to load.
jsut i figure
Coke admins are working on it.
jsut that's what you get for releasing 1.0 :D
i'll blame it on drupal in the meantime 19:37
particle1 .\\src\\pmc\\iterator.pmc(405) : error C2275: 'PMC' : illegal use of this type as an expression
cotto--
cotto fixing... 19:38
purl fixing is good, definitely.
cotto let's see if the commit works 19:39
Coke cotto: I've been committing things on and off for hours.
allison Andy: Well, you know we asked Brian Eno for a theme song, but he was like in the Bahamas or something, so we had to settle for a U2 quote. ;) 19:40
Coke allison: the trap opcode is being used in src/debug.c
(in a function that is only used in gdbinit)
Andy I'd think Jimmy Buffett would be more appropriate.
parrots and all
Coke I would leave the project in protest. =-)
cotto great. now we're on Slashdot too.
Andy The poor website. :-( 19:41
particle1 we need a nice 404 page 19:42
Coke particle1: you'd still not be able to load it today. =-) 19:43
particle1 www.huffingtonpost.com/theblog/arch...parrot.png
shorten particle1's url is at xrl.us/bekrbk
Coke I wonder why wer're tagged as 'dead' on /. 19:44
is that specifically for the website being /.'d ?
particle1 tinyurl.com/hesresting 19:45
there :)
jsut hah
Coke svn commit slowwwwwwwwwwwww
cotto svn up slooooowwwwwweeeeerrrrr 19:46
It's a good day for a nap.
particle1 when this is all over, i'd like to see some stats on our traffic 19:47
Coke I am removing most of the experimental ops. commit stuck in a time warp. 19:48
chromatic While we have a wishlist, I'd like an implementation of ISRP. I have plenty of DNF jokesters to use as test subjects.
Coke ISRP? 19:49
chromatic Sorry, IRSP 19:50
cotto coke++ 19:51
Coke IRSP?
chromatic Internet Remote Strangulation Protocol 19:52
Coke ACHOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
cotto: ?
cotto hi 19:53
karma was for removing the experimental ops 19:54
purl was for removing the experimental ops has neutral karma
cotto on the off chance that the commit completes
which it appears to have
done
many of those were causing fuzzer failures 19:55
Coke remembers to add it to the NEWS 19:57
TimToady phone 20:00
Coke brt 20:01
20:09 estrabd joined
cotto It's nice that svn.parrot.org still works at all, even though it's really slow. 20:23
Coke corelist++
20:35 Ramereth joined
Ramereth allison: hey (this is lance from the OSL) 20:35
allison hi lance
Ramereth i'm about to make the switch on the trac sites but they'll have a short downtime while I do this 20:36
i'll take just those sites offline while I do it
allison Ramereth: okay, thanks for letting us know
Ramereth but figured being more interactive was better :)
allison Ramereth: it's great to have site admins who understand geeks :) 20:37
Ramereth i'll actually have to do take them down twice (once for wsgi, once for postgres switch)
allison Ramereth: if you'd like, you can just shut down external access to trac.parrot.org for a while you make the transition 20:39
Ramereth ok, i'll just do that
what i've done before is disabled http but not https so that I can still easily access it 20:40
actually, do you mind if I just do that now? it'll probably be down for about 10-20 min while i do this
having it on /'d makes me a bit nervous :)
allison Ramereth: all of our public links go to https so will likely need to shut that off externally too
Ramereth ah ok
allison Ramerth: yes, please do
Ramereth i can just put in an access/deny list then 20:41
so they'll get a 403 while i do this
Coke Ramereth is lance from the OSL
Ramereth will that be fine?
allison in fact, turning off trac.parrot.org will likely help our www.parrot.org traffic
Ramerth: or just redirect trac.parrot.org to www.parrot.org?
Ramereth well, i still want to access it to test it
i'll just use a temp vhost while I do it 20:42
give me a sec, i'll fix it up so it redirects
cotto should we avoid svn commits while this is happening? 20:43
Ramereth nah, thats fine
particle1 someone just suggested i remind you of email2trac... but i thought tomorrow would be a better time :) 20:44
Ramereth email2trac is an utter pain to setup. someone should rewrite that :)
ok, the redirect is up 20:45
particle1 cotto++ for fixing cotto--'s mess
20:45 Theory joined
allison Ramerth: okay, thanks! 20:45
particle1 cotto: that error looked so familiar, but i just couldn't place it. duh! 20:46
20:47 vti left
particle1 chromatic: that link i promised: www.perlfoundation.org/perl5/index....9_projects 20:48
shorten particle1's url is at xrl.us/bekq26
chromatic Thanks. 20:49
Coke removed dan.ops 20:50
adios, dan.
cotto coke++ #cruft removal
Coke I just unstalled all the stalled tickets in RT: everything in RT is stalled, and having that status just hides those tickets. 20:56
anyone have icc on linux? 21:00
(if so, grab RT #58958)
allison I just changed the "Languages" tab on www.parrot.org to go to a Drupal page 21:01
shutting down Trac seems to have completely cleared up the www.parrot.org speed issues
chromatic Also the Trac speed issues. 21:02
Coke allison: of course, all those links go to trac... which is now www. =-)
allison Coke: most of them don't 21:03
szbalint finally a good use for my /. mod points
allison Coke: I can make those non-links
Coke probably not worth the effort.
Infinoid logs into /. for the first time since 2004 21:04
particle1 there's a login feature? 21:05
particle1 has been to /. twice. once today.
Coke it would make sense to regroup parrothist.pod by semi-major release after 0.4.7; it will make sense to group them ending with stable after 1.0
anyone mind if I change the whitespace in there? 21:06
Infinoid particle1: You're not missing much.
allison Coke: go for it
Coke: we'll be changing release managers every month so there's not much point in keeping the grouping by "sequential releases made by the same person"
Coke that's my thinking. done. 21:07
haru tatsu? 21:08
purl hmmm... haru tatsu is Spring Begins
Coke is that right, allison?
allison Coke: yup
Coke google++ 21:09
karma google?
purl google has karma of 507
Coke karma yahoo?
purl yahoo has karma of -6
allison lol :)
21:19 chromatic joined
Ramereth allison: ok, i think i'm ready to make this go 21:21
cross your fingers.. 21:22
allison Ramereth: excellent... 21:23
Ramereth yay, happy trac sites
allison: can you try logging in to make sure everything looks ok? 21:24
allison Ramereth: looks good, source browser is working, wiki is working 21:25
Ramereth and the VM is mostly bored now :)
i love software fixes!
allison Ramereth: hurray!
Ramereth load average: 0.46, 0.99, 0.88 21:26
much happier
allison Ramereth: do you have stats on the traffic we've been getting? would be interesting to see
Ramereth nope, but we could get awstats going, or google analystics 21:27
analytics
purl analytics are just like electrolytics, but with more analytes.
allison Ramereth: that'd be great 21:28
Ramereth which would you like?
we've been doing analytics to all of our drupal sites lately 21:29
allison Ramereth: wow, trac really feels faster now than it did before under a normal load, postgres and mod_wsgi are good
Ramereth yup, it rocks
allison Ramereth: then let's go with analytics
Ramereth my students rock for figuring that out :)
allison Ramereth: three cheers for motivated and curious students :) 21:31
21:31 chromatic joined
Ramereth the only problem was I had to wait for him to get done with a final to tell me how to set it up today 21:31
allison Ramereth: heh :)
Ramereth allison: have you ever setup analytics before? it works best if you setup the account on your end, we can help get the module activated, etc 21:32
allison Ramereth: I haven't, but if you point me at the info I can set it up
Ramereth www.google.com/analytics/
its pretty simple, you just create an account, then add sites to it
it'll generate a "key" that you put into the site 21:33
particle1 Ramereth++ # much faster now
allison Ramereth: should I create a generic Google account for it (and not use my personal google account?)
Ramereth allison: you can give access to other folks with google accounts 21:34
that's how we have ours setup
particle1 yeah, i've done that before
allison Ramereth: okay, will do that
Ramereth: done 21:37
particle1 dang. that's so much faster. wow.
allison Ramereth: then, where in Drupal should I add the javascript?
GeJ Good morning everyone
Ramereth allison: i'll show you here in a sec
allison (turns out, I already had access to the O'Reilly Radar analytics account, just never accessed it) 21:38
Ramereth allison: www.parrot.org/admin/settings/googleanalytics 21:40
(under site configuration)
particle1 funny that i can't access that page 21:41
Ramereth you're probably not an admin :)
allison Ramereth: I don't have access either
Ramereth looks 21:42
allison Ramereth: and particle is an admin... I'll check our permissions...
basic talkin bout us students? :)
particle1 i see 'site information' and 'site maintenance' but not 'site configuration'
students++
Ramereth basic: i didn't even see you there
ha!
particle1 ah, site configuration is above site maintenance
allison Ramereth: yes, I needed to grant permissions for google analytics to the admin group
basic hehe, i've been hiding in here
particle1 found google analytics 21:43
moritz so how many visitors did we have since the release?
particle1 it wasn't there before, so something changed.
moritz: we don't have that history :(
we will, from now on, though
cotto Infinoid, when do we get dalek back?
moritz :(
particle1 i think it's safe to bring dalek back now. 21:44
Ramereth we do have logs, so we can parse those if you want
moritz it's just idle interest
particle1 Ramereth: we're excited, is all. it's nice to see the site went down for the right reasons :) 21:45
moritz but if it's not too much work it'd be nice
basic Ramereth: it may be worth using the http/https redirects to save some bandwidth too
allison Ramereth: okay, I've entered the account number, so should be good to go
Ramereth allison: sweet 21:46
now we just need to get you on drupal 6 :)
but that can wait ;)
allison Ramereth: yes, looking forward to that
Ramereth: hopefully this traffic will die down in a few days, thanks for helping us weather the storm! 21:47
21:49 dalek joined
Ramereth you're pushing about 5-6Mbps right now 21:50
jumped once I fixed trac
basic traffic is good :)
Infinoid cotto: should be back now, was just waiting for the goahead
cotto great 21:53
Infinoid fperrad++ coke++ coke++ coke++ coke++ cotto++ coke++ coke++ cotto++ cotto++ coke++ coke++ coke++ cotto++ cotto++ cotto++ # catching up on dalek karma since r37546 21:54
particle1 hee
infinoid++ # karma spam
cotto unfortunately purl isn't smart enough to evaluate more than one karma change per line
but the thought counts
Infinoid I tried :) 21:55
dalek kudo: f8b6aee | (Moritz Lenz)++ | src/setting/Match.pm:
implement Match.ast
 22:00
22:03 isop joined
isop pmichaud: ping 22:06
dalek rrot: r37563 | cotto++ | trunk (2 files):
[string] make Parrot_encoding_name and a similar function check for negative values
 22:08
22:21 Woody4286 joined 22:26 korshak left 22:31 kid51 joined
Tene isop: pmichaud is currently vacationing, and isn't around much 22:31
dalek rrot: r37564 | cotto++ | trunk/tools/dev/parrot-fuzzer:
[tools] update fuzzer opcode blacklist, minor refactor
cotto Is there a way to get all the opcodes and/or PMCs Parrot knows about from Parrot? 22:41
isop Tene: thanks 22:42
Tene isop: what did you need him for?
isop Tene: to add me to pynie project on google code
:)
cotto you can message him via purl 22:43
moritz or by plain old email ;-)
isop ok
22:52 ruoso joined 23:09 Lennie joined
Lennie first of all congratulations for the first 'real' release 23:09
I do have one question though, is there something like a compiler cache for Parrot/Rakudo/Perl6 ? 23:10
I'm not much of a Python user, but they have it 23:11
chromatic Do you mean PBC files?
jonathan Nothing that I'd really call a compiler cache that I know of.
Lennie I guess so.
jonathan For Rakudo we are able to compile down to bytecode and just run the bytecode to save having to do the compilation every run...
...we use this for Test.pm (the Perl 6 test harness) for example, so we can run the tests faster. 23:12
But maybe that isn't what you're asking.
Lennie jonathan: that's exactly what I was looking for
jonathan: I just don't know the commands yet. :-)
jonathan Lennie: OK. We don't have anything yet that is smart enough to compile a module the first time, save the PBC and only re-compile if the source changes yet, though it'd be a cool thing to do in the future. :-) 23:13
Lennie: Commands for doing this with Rakudo? I can tell you that.
Lennie for example, how do I save/create a PBC file ? 23:14
chromatic ./parrot -o some.pbc some.pir
moritz jonathan: small correction: Test.pm is compiled to PIR, not to pbc 23:15
Lennie (I feel stupid I need to ask, but I didn't see it anyway)
jonathan moritz: Yeah, I was just realizing that when I went to look for the thing...
Lennie: You can do perl6 --target=pir foo.p6 > foo.pir 23:16
And then what chromatic said to get a PBC.
I had thought --target=pbc worked in Rakudo.
But it seems not. :-(
Lennie jonathan/chromatic: ok, thanks, I'll probably get the hang of it that way. ;-) 23:17
23:18 franhp joined
franhp hi 23:18
purl niihau, franhp.
jonathan Lennie: If you compile a MyModule.pm down to MyModule.pir or MyModule.pbc then Rakudo knows to look for those first before the MyModule.p6
dalek rrot: r37565 | jkeenan++ | trunk/t/op/string_cs.t:
Correct error in number of tests planned.
jonathan But it isn't smart enough to choose the newest.
franhp I'm a little confused about what parrot could do for me. I have some .c simple files, would parrot help me skip the gcc compiling step? 23:19
moritz jonathan: but it does seem to work with .pbc... just running the test with Test.pbc and Test.pir deleted 23:20
jonathan moritz: Yes, I just thought we were able to do --target=pbc 23:21
franhp: No, Parrot doesn't have a C compiler targeting it (yet) and it's not really going to be an efficient way to run C anyway, I expect.
franhp: Parrot is more suited for being a target for dynamic languages, e.g. Perl 6. 23:22
franhp jonathan: ok, then I'll run tests with other languages :D thank you!
see you soon1 23:24
23:24 franhp left 23:33 davidfetter joined 23:34 bsdz joined 23:37 rdice joined 23:38 k`sOSe joined 23:45 hudnix joined 23:50 tetragon joined
Lennie jonathan: the reason I asked was, maybe I just have the wrong expectations, because I see a big difference in (startup) performance 23:54
jonathan: between perl 5 and a perl 6 program
jonathan: I have 2 simple test scripts: 23:55
moritz Lennie: parsing is rather slow atm
davidfetter Util, ping
Lennie perl5: strict and warnings turned on: print "Hello\\n";
perl6: use v6; say 'Hello'
time ./test.pl -> real 0m0.011s 23:56
time ./parrot/parrot -r test.pbc -> real 0m1.157s
moritz: I think I'm not parsing at that stage
moritz Lennie: right, not at that stage 23:57

Powered by Cro and the Raku® Programming Language.

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

Thank you!