Parrot 3.10.0 "Apple Pi" | parrot.org | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 16 November 2011.
00:26 lucian joined 00:57 PacoLinux joined 01:11 woosley joined
dukeleto ~~ 01:32
tadzik hello dukeleto 01:33
01:33 PacoLinux joined
dukeleto tadzik: i am just now getting on the GCI train 01:34
tadzik: you want to be a mentor for Parrot gci?
tadzik dukeleto: of course
dukeleto tadzik: which email do you use for melange? 01:37
tadzik dukeleto: tadzikes at gmail
dukeleto tadzik: melange complained that it was not recognized 01:40
tadzik hmm, maybe I'm not signed up yet
mind if I do it tomorrow? I still have some minix hacking to do 01:41
dukeleto tadzik: do as ye will :) 01:42
msg alester how goes the github issue quest? 01:43
aloha OK. I'll deliver the message.
01:45 nbezzala joined
dukeleto is going to update github.com/parrot/parrot/blob/mast...ssions.pod to prefer Github over Trac 01:50
it is confusing for newcomers
and our new developer guide is still in trac
cotto dukeleto, +1 02:04
dukeleto cotto: i am working on creating gci tasks 02:07
cotto: what is your melange id?
cotto dukeleto, probably something unimaginative like "cotto" 02:08
iirc
dukeleto cotto: i am working on stuff in the github.com/leto/gci 02:09
cotto: if you want to write up and tasks :) 02:11
whiteknight dukeleto: Do you have any more insight about task creation on google-melange? I still don't seem to have any ability to see tasks, much less create or edit them 02:17
msg alvis: If you have any tasks in mind that a GCI student could participate in, let us know and we will set up a task for it 02:21
aloha OK. I'll deliver the message.
dukeleto whiteknight: oy, looking into it now 02:22
whiteknight dukeleto: Because you know I can be counted on to create lots of tickets
dukeleto whiteknight: what is your melange id? wknight8111 doesn't work 02:24
whiteknight that is it
or wknight8111@gmail.com
dukeleto whiteknight: i am keeping track of tasks in github.com/leto/gci
whiteknight ok 02:25
dukeleto whiteknight: you can fork if you like. I copy the task template, tweak it, then paste into the melange webforms
whiteknight: there is a bulk uploader, but it didn't work last year, so I am probably going to stick with this method
whiteknight we can do it that way if we absolutely must. Things will go much faster if I can just submit tasks myself
dukeleto whiteknight: neither wknight8111@gmail.com or wknight8111 works 02:26
whiteknight "You are logged in as: wknight8111@gmail.com (change)
Your username is: wknight8111"
dukeleto whiteknight: you need to create a profile for GCI 2011
whiteknight that's exactly what the melange ui is telling me
dukeleto whiteknight: then i can add you
whiteknight: it is dumb, but true 02:27
whiteknight okay, how do I do that?
dukeleto whiteknight: click on "My Profile" on the left navbar in melange
whiteknight: and fill it out :)
whiteknight no such link
dukeleto blargpony
whiteknight I think my account is crapfaced
dukeleto whiteknight: are you logged in? 02:28
whiteknight yes. I have a "My Dashboard" link which shows the Parrot org application
but no "My Profile" link
dukeleto whiteknight: Username/Email*User wknight8111 must create a profile for this program. 02:30
whiteknight I'm going to send Carol an email
because wtf
02:34 alvis joined
dukeleto alvis: howdy 02:34
alvis Have you had a chance to look at parrot.github.com/ and github.com/parrot/parrot/?
dukeleto: and hello, btw. 02:36
dukeleto parrot.github.com is looking nice these days
alvis: yeah, they look awesome!
alvis: is there some kind of redirect happening?
alvis: could you go to www.google-melange.com/gci/homepage...le/gci2011 and sign up to be a mentor for Parrot Foundation? 02:37
alvis: we can have some GCI students helping you with documentation tasks
i realized i have waited too long to do this. I didn't realized the rules changed until a few days ago. 02:38
alvis dukeleto: Do you mean from 'docs.parrot.org' or via '.htaccess' in the repo?
dukeleto: Ok, I'll signup.
dukeleto alvis: when i go to parrot.github.com, i get redirected to /html/index.hmlt 02:39
html, rather
whiteknight: can you try going to the above link and signing up to be a mentor for Parrot Foundation?
alvis: you may need to just sign up in general. I am not sure exactly how signing up as a mentor works. 02:40
whiteknight going now...
alvis: yes, I have looked at parrot.github.org
dukeleto cotto: what are the most important code tasks to get done in GCI?
whiteknight er, parrot.github.com
dukeleto alvis: i would like some links to Winxed stuff on parrot.github.com
lucian whiteknight: i thought you made it
whiteknight lucian: alvis re-made it
lucian ah, i see. silly me
dukeleto lucian: he started it, but alvis has been hacking on it
whiteknight dukeleto: I'm on that link, but can't see anywhere to go to sign up for anything 02:41
alvis dukeleto: Right, parrot.github.com displays the index.html page, and github.com/parrot/parrot/ contains the doc files.
dukeleto alvis: instead of a .htaccess file, use a symlink from index.html to html/index.html 02:42
lucian btw, it seems to me that parrot.gh.com is already as useful as parrot.org
dukeleto alvis: in my browser, i get a "do you want to allow this redirect?" warning
whiteknight lucian: the bar is set surprisingly low
dukeleto lulz
lucian ponders clojure on parrot
dukeleto well, let's make parrot easier to learn about and surely we will get new contributors
lucian: what is the state of puffin? still waiting on 6model? 02:43
cotto did melange just get worse again or am I missing something?
lucian dukeleto: yeah, i've not touched it
dukeleto cotto: it keeps changing. No comment.
lucian dukeleto: seems pointless, really
cotto I don't see any way to request mentorship with the parrot org
whiteknight cotto: thanks! I'm glad I'm not the only one
lucian dukeleto: also, i didn't have a laptop until a few weeks ago :) 02:44
lucian hugs his 13" thinkpad
alvis dukeleto: Ok, but you'll probably get a redirect message either way. html redirection is, supposedly, depreciated.
dukeleto: Unless, of course, I hink with the relative urls in 'docs/html' or modify 'make_html_docs.pl' 02:45
whiteknight this melange website is maddening 02:46
how the hell is a person supposed to do anything on here?
cotto seriously 02:47
It's like Google's of two minds, one of which loves helping OSS and the other of which wants to drive hackers crazy. 02:48
lucian cotto: it's all an elaborate plan to get hackers to work on melange, thus hooked on app engine 02:49
cotto um....
I'm probably just going to have to get a punching bag and say "no comment" a lot. 02:50
lucian bah, every irc client i've tried is worse than limechat 02:51
dukeleto: anyway, when 6model is in parrot proper and i get some time, i'll have another look at puffin 02:52
dukeleto alvis: if you add a symlink, in git, in parrot.github.com.git, from /index.html to /html/index.html, i think that should simplify things. I will try it right now.
dalek href="https://parrot.github.com:">parrot.github.com: 7738ab2 | dukeleto++ | index.html:
Symlink index.html instead of redirecting
 02:54
href="https://parrot.github.com:">parrot.github.com: 1c09db9 | dukeleto++ | .htaccess:
remove .htaccess
 02:55
dukeleto alvis: is there documentation about how to update the parrot.github.com.git repo after a new parrot release? 02:56
alvis: also, I would like to have each version of parrot's docs available on parrot.github.com
alvis: like parrot.github.com/3.9.0 , parrot.github.com/3.6.0 / etc
alvis: so people can look up the docs for the parrot version they are using 02:57
self.dinner()
alvis dukeleto: (1) Not yet. I wanted to see if everything met with my betters' approval. 02:58
dukeleto: (2) Ok, I'll do that.
dukeleto: May be a day or two (or three 'cause of thanksgiving) 'til I can get it done. 02:59
dukeleto & whiteknight: Thanks for your help/advice/direction. 03:01
03:22 PacoLinux joined 03:30 jsut joined
dukeleto ~~ 04:11
04:23 PacoLinux joined
dukeleto NotFound: ping 04:23
alvis dukeleto: You still 'round? 04:36
well, I'll just ask and you can get back to me: As to winxed docs, like what? 04:40
as far as I'm aware, NotFound has very little way in the way of documentation.
dukeleto alvis: yup 04:41
alvis: a link to whiteknight.github.com/Rosella/winxed/index.html
alvis: and winxed.or
alvis: winxed.org, rather
alvis: actually, we need a page about various languages on parrot 04:42
alvis dukeleto: Oh, ok. That I can do.
dukeleto: Agreed.
dukeleto alvis: but i want to highlight winxed and nqp, because those come in parrot core
alvis dukeleto: Ok, got it.
dukeleto alvis: so winxed and nqp should be mentioned on the homepage of parrot.github.com
alvis: and then we should have a page which lists known-working parrot languages and links about them 04:43
alvis: there are plenty of lists of parrot languages that don't work, we don't need to list them all. Just those that have worked in the last 6 months or some arbitrary time period
alvis dukeleto: Ah, so you want a redesign of the homepage on parrot.github.com? (I didn't understand that.)
dukeleto alvis: no, just add a section to it 04:44
alvis dukeleto: Ah, ok. I understand.
dukeleto alvis: add a section about languges in parrot core (nqp + winxed) and then a link to a page with a list of known-working languages
alvis: i guess this means actually modifying the "make html" process?
alvis: is html/index.html generated? or did you hand write it? 04:45
alvis dukeleto: yeah, probably, to get the functionality your after.
dukeleto: nbd. I was looking at that as a possibility already.
dukeleto: No, it's autogen'd with 'tools/dev/make_html_docs.pl' 04:46
dukeleto alvis: ok, that is what i thought 04:50
alvis: are you comfortable with perl?
alvis dukeleto: very.
dukeleto: been using it, off and on, since ... '92. 04:51
dukeleto alvis: awesome :) then you should have fun fiddling with make_html_docs.pl :) 04:52
alvis: it is a slippery maze of perl modules
alvis dukeleto: yeah, it's fun like that. :) 04:54
dukeleto: Ok, I'm gonna let you go and see if I can figure out this dynpmc thingy to create a cons (may just back it with an unmangedstruct <shrugs>) 04:56
dukeleto well, at least we have 5 GCI tasks in the system 05:00
but I am the only mentor
alvis dukeleto: Oops, I looked, but how did you sign up as a mentor? (Think cotto and whiteknight were wondering as well.) 05:01
dukeleto alvis: oy vey. 05:02
alvis: sorry, i am trying to add tasks as quick as i can, it closes down soon. 05:03
alvis: there should be instructions to sign up on www.google-melange.com/gci/homepage...le/gci2011 05:04
05:16 rfw joined
alvis dukeleto: I agree, there should be, but there's not. Neither is there a register link. 05:27
dukeleto: I'll see if anyone else was successful, tomorrow. 'night #parrot. 05:28
dalek p-parrot-create: afabeec | dukeleto++ | / (3 files):
Change /submit to take a POST and get tests passing
 06:09
06:54 nbezzala joined
NotFound dukeleto: pong 06:59
dukeleto NotFound: i forgot. unpong. 07:07
NotFound k 07:08
dalek p-parrot-create: 4bf3f89 | dukeleto++ | t/002_index_route.t:
add a test for a GET on /submit
 07:18
p-parrot-create: 6f0e0a6 | dukeleto++ | / (2 files):
Get tests passing again and use tempdir() correctly
07:50 PacoLinux joined 07:59 rfw joined 08:46 mj41 joined, nbezzala joined 09:12 lucian joined 09:13 cosimo joined 09:16 lucian joined 09:18 lucian joined 09:20 PacoLinux joined 09:31 bacek joined
bacek ~~ 09:31
dalek rrot: d8b901c | bacek++ | src/ops/ (2 files):
Add deprecated warning to op new_s_i
 10:05
10:10 cosimo joined 11:23 michal joined
michal hello all!! 11:23
ew to parrot, in fact I am new to perl as well. Please advise me, where else - beside parrot.org - I can find some useful informations about the topic. Thaks anyway!! 11:26
I am new to parrot, in fact I am new to perl as well. Please advise me, where else - 11:27
beside parrot.org - I can find some useful informations about the topic. Thaks anyway!!
As You can see I am also new at IRC...:)
tadzik hello michal :) 11:29
parrot.org is probably the best source 11:30
what particulary are you interested in?
michal I am not going to lie to you I need to make a resrearch about parrot and pearl, i already read few e-books and I am reading more, but 11:39
sorry perl
;) 11:40
but abstract knowledge does not lead me anywhere..
mayby you know where i can find practical use of both parrot and perl so i can study it? 11:41
11:54 PacoLinux joined 12:12 Psyche^ joined 12:41 PacoLinux joined 12:47 whiteknight joined
dalek p: 13508b4 | mls++ | src/NQP/SymbolTable.pm:
use direct lexinfo access in install_lexical_symbol()
 12:49
whiteknight msg dukeleto I followed that link from your email and created a GCI profile. Can you please pretty please add me as a mentor so I can start murderlizing some task ideas? We don't have much time, methinks 13:03
aloha OK. I'll deliver the message.
13:08 PacoLinux joined 13:18 aloha joined
dalek sella: ef01076 | Whiteknight++ | s (2 files):
Add in a prototype for a light-weight service locator type to Container
 13:23
sella: 73efbcb | Whiteknight++ | src/container/Container.winxed:
Use a separate __sort_options function for sorting out the options passed to .register() and .resolve(). The two methods have different requirements that can't be satisfied by a single helper method. This fixes one TODO note but creates others
13:28 bacek_at_work joined 13:57 PacoLinux joined 14:27 bluescreen joined 14:37 PacoLinux joined 14:43 PacoLinux_ joined 14:55 eternaleye joined 15:29 woosley joined
dalek sella/container_resolve: 21ef760 | Whiteknight++ | src/path/ (4 files):
A few small cleanups and fixes for Path
 15:41
sella/container_resolve: dde3499 | Whiteknight++ | src/container/ (2 files):
Add in basic recursion-detection for Container
sella/container_resolve: 6e35790 | Whiteknight++ | / (4 files):
Add in a new PerResolution lifetime manager which returns a single instance during a single resolution sequence. Add in a stub test file for it.
sella/gh-pages: 428ed52 | Whiteknight++ | libraries/container.md:
flesh out basic docs on Container classes
sella/gh-pages: 84649bb | Whiteknight++ | libraries/container.md:
Start adding per-class examples to the Container docs
15:41 bluescreen joined
sella: 21ef760 | Whiteknight++ | src/path/ (4 files):
A few small cleanups and fixes for Path
16:30 tty234 joined 16:34 tty234 joined
dukeleto ~~ 16:43
tadzik ~~ 16:44
dukeleto: I'm signed up for GCI, username tadzik
dukeleto tadzik++ 16:45
whiteknight: you are an org admin now
whiteknight w00t
tadzik :)
whiteknight probably too late for any task-making
dukeleto whiteknight: i made 6 tasks last night
whiteknight: many of your ideas that were on the task list
whiteknight okay
dukeleto whiteknight: take a look at them
whiteknight 6 is still small fry. We completed about 150 tasks last year 16:46
total, of course
dukeleto whiteknight: hindsight is 20/20, i guess. I didn't know the rules changed until a few days ago.
whiteknight yeah, it's alright. Not your fault. We're just going to need to have a deluge for the midterm 16:47
Coke Probably warrants a blog post about why the numbers might seem low. 16:48
dukeleto tadzik: think i just added you
Coke: are you volunteering?
dukeleto feels overworked and underpaid (no-paid)
Coke For parrot.org? Sure, I can whip something up, but it won't be until tomorrow.
whiteknight dukeleto: If I'm co-admin now, I can take over most of the work 16:49
Coke That work?
dukeleto Coke: anytime is better than never :)
tadzik dukeleto: accept'd
dukeleto awesome
whiteknight: all tasks are assigned to me right now, because i was the only mentor 16:50
whiteknight: if you can, add yourself to some tasks
whiteknight: there can be multiple mentors per task
whiteknight I will
tadzik wut, I'm an org admin?
dukeleto Coke: also, yes we had lots of tasks last year, which only happened because I and others added tasks for two months straight
tadzik oh, there are >1-day tasks now 16:51
dukeleto tadzik: if so, then run with it. Use the power wisely.
dukeleto is clicky buttons willy-nilly
clicking, even.
16:56 tty234_ joined
whiteknight yay! I added a task 16:57
with no help from this shitty editor 16:58
the melange interface this year seems worse than usual 17:03
tadzik and, since we have a karmabot...
melange--
hmm, we could add some dedeprecator-related tasks 17:05
as in, filling api.yaml with regexes
whiteknight that can be pretty hard, methinks 17:07
17:32 dmalcolm joined 17:41 fperrad joined 17:43 cosimo joined 17:56 AzureStone joined 18:02 lucian joined
cotto ~~ 18:07
whiteknight hello cotto
cotto hi whiteknight 18:11
Coke we are unable to add new tasks until December whatever, yes? 18:15
(just edit existing ones?)
whiteknight I seem to have added a new task this morning, but I don't know what the shutoff date it
I say, if you have any, Just start trying to add them
if they go in, good. If not, we wait till later
tadzik #gci says they'll get published on Dec 16 18:16
whiteknight great 18:18
awesome, even
tadzik not bad
PerlJam so they can get added, but aren't publically viewable until Dec 16?
tadzik it seems so
whiteknight last year students would come to us, we would ask them what kinds of stuff they were interested in working on, and we would create tickets specifically for them on the fly
tadzik yeah
whiteknight and once we saw how many tasks were being developed and at what velocity, we adjusted the difficulty levels (because the students were better than we anticipated) and we increased ticket volume 18:19
18:20 PacoLinux joined 18:30 contingencyplan joined
Coke deadline extension: Nov 28th. 18:44
\\o/
(for 1st round tasks) 18:45
cotto whiteknight: yeah. That's what they're trying to avoid this year. 18:53
whiteknight cotto: I sort of understand the motivation for it, but the end result is going to be far fewer tasks this year for students to complete
cotto yes 18:54
whiteknight if there are anywhere near the number of students, and they move at anywhere near the velocity they did last year, the whole program is going to run out of tasks long before the final call 18:55
cotto I agree that it seems like the net result will be a less productive gci, but I'm not running it. 18:56
whiteknight We'll do the best we can this year, though we only have 6 or 7 tasks queued as of right now to get us through mid-december 18:59
so we're going to have to flood the system with tasks for the second half if we want to get any benefit out of it
Coke whiteknight: initial task deadline extended. 19:00
whiteknight Coke: how long?
Coke Until Nov. 28th.
whiteknight because I will go butt-crazy tonight if I have to
Coke This from rafl, who is handling TPF's queue. 19:01
whiteknight okay, well tonight I'm going to put in as many tasks as I can possibly imagine 19:10
and a few that I can't really imagine, but I vaguely suspect are bad ideas 19:12
19:21 lucian joined 19:27 dukeleto joined 19:32 PacoLinux joined
dukeleto woot, the deadline to add tasks was extended! 19:32
whiteknight doublewoot 19:36
msg NotFound if you can think of any winxed-related tasks that might be good for GCI students to get involved with, please let me know 19:39
aloha OK. I'll deliver the message.
atrodo Would isparrotfastyet improvements make for a good set of gci tasks? 19:40
whiteknight atrodo: yes, if you have any that can be bite-sized and easy to digest for GCI students 19:41
NotFound whiteknight: other than "write X with winxed" for abitrary value of X, I haven't any task in mid.
whiteknight NotFound: okay, I'm putting together a few of those tasks myself.
dukeleto whiteknight: did you see that i have task templates in my gci.git repo? 19:42
whiteknight: if you are going to make lots of similar tasks, make a template for it and then tweak the template. It makes for high-quality tickets that don't take too much time to make
whiteknight dukeleto: yes, I was copying from there
can you just copy+paste the template code into the melange editor box? 19:43
NotFound Reading slashdot I'm tempted to suggest: write a JVM with winxed.
whiteknight I was sort of copying by hand
NotFound The guy said hw wrote the javscript JVM in six months. Surely with winxed it can be done faster ;)
cotto Isn't that still quite incomplete? 19:45
atrodo NotFound> Or we could reach for the moon and write a parrotvm in winxed...
dukeleto whiteknight++ 19:46
NotFound cotto: yes, but I won't blame the student for such incompletitude.
dukeleto whiteknight: yes, click the "html" button, then i use the "It's All Text" FF plugin to use vim and read in the template with :r
NotFound atrodo: that may be doable some day. 19:47
atrodo NotFound> I welcome our winxed written overlords 19:48
NotFound Another interesting idea: a winxed compiler targeting hardware, via a FPGA or somethin' 19:49
Yesteryear I was tempted about creating a website: "April 1st of code" 19:50
dukeleto whiteknight: do you need anything from me gci-wise? Are you on a task-creating spree ? 19:55
whiteknight dukeleto: Just stay out of my way.
I won't go crazy until tonight. Go grab a beer or something and relax
dukeleto whiteknight: i am drinking a double shot at the library, but I get your drift :) 19:56
whiteknight I feel like you're over worked and underpaid :)
dukeleto whiteknight: i need to concentrate on sending GOOG an invoice from PaFo so we actually get the money. It is so close to being done. 19:57
the money from GSoC, that is.
whiteknight dukeleto++
dukeleto Parrot Foundation will be getting $4000 USD for participating in GSoC.
8 slots x $500 = $4000 19:58
GSoC money actually keeps PaFo afloat right now, basically.
dukeleto likes transparency, and figures people should know this.
Coke Given that we have no fundraising, aye. Though I never got to see any of the books, particle kept a tight leash on 'em. 19:59
particle jhorwitz++ took care of the books :P 20:00
dukeleto particle: howdy there partner!
Coke ... My memory is mush. I retract my statement and replace it with "I never saw the books."
whiteknight yeah, gsoc really does cover our costs
particle dukeleto: when did you put me in as temporary admin for gci? 20:01
dukeleto particle: i am going to be submitting an invoice to google for $4000 + reimbursements for travel
particle aloha just told me :)
20:01 preflex_ joined
dukeleto particle: long time ago 20:01
particle dukeleto: i'll deposit ye olde checke
dukeleto particle: yeah, aloha should mention timestamps
particle: we gonna get money electronically. Welcome to the new century!
particle hot damn
dukeleto particle: *but*, cotto++ and I would like to get reimbursed from Parrot Foundation, which requires you to write a check. 20:02
particle: OR, we can pay 25 cents and do it online via something like dwolla.com
particle can do. do you have receipts?
20:02 alester joined
particle we have a reimbursement form somewhere in dropbox iirc 20:02
dukeleto particle: yep
particle just need some paperwtork.
dukeleto particle: ooh, i didn't even know that. 20:03
dukeleto is so allergic to paperwork i break out in hives when i hear the word "form"
particle send it in, pester the heck out of me, and i'll do it
dukeleto particle: consider it done
benabik o/ #parrot 20:08
whiteknight hello benabik 20:11
benabik: can you imagine any PACT- or PCT-related tasks that would be suitable for GCI?
benabik whiteknight: PACT is still in an exploration stage, I think. The biggest thing PCT could use is more/better documentation, but I can't point out where it's needed off the top of my head.
(I just know there are several "dig into the code and find out" things about it.) 20:12
dukeleto GCI tasks need to be well-described. Things that are "figure out how to do this" don't really work well for GCI tasks. 20:16
benabik Yeah, that's why PACT isn't a good one. Some thought might turn up PCT documentation tasks. I can't think of PCT code tasks.
dukeleto looks like the Winxed screencast was the first task requested by a GCI student 20:18
benabik Solid winxed documentation is also a good idea. I loves me documentation. 20:19
dukeleto benabik: yes, please
benabik I don't like writing it, but I love having it. :-D
dukeleto particle: do we have a template for parrot foundation invoices? 20:20
benabik (Although I've been told I'm pretty good at writing documentation. Blah.)
particle dukeleto: i can't find it :( 20:22
20:23 PacoLinux joined 20:31 perlite joined
dalek nxed/inline: 31e59e2 | NotFound++ | winxedst1.winxed:
implement clone in ArrayExpr and SwitchCaseStatement

use it to simplify other clone methods		 20:34
whiteknight There is a GCI category for "Research" and "Documentation". PACT and PCT can be very interesting sources for tasks, I think
we just need to get creative
NotFound: are there any syntax features in winxed that need more testing? I'm going to set up some Winxed documentation tasks for the Rosella website too 20:35
NotFound whiteknight: I can't think of any feature that doesn't need more testing. 20:36
benabik :-D 20:37
NotFound Expresionless switch, for example, has no other test than being used in the compiler. 20:38
20:39 PacoLinux joined 20:41 mj41 joined
dukeleto particle: i have an example. It doesn't need to be fancy. 20:44
whiteknight: research tasks are very time-consuming to describe and judge
whiteknight: they look good at first, but i would rather have code or docs from GCI students
whiteknight: but go for it if you want to create them 20:45
whiteknight NotFound: Those kinds of tasks are very good for GCI. it forces them to read the code, figure out what is going on, then write docs/tests for the feature
dukeleto whiteknight: students didn't seem to want the research tasks last year. I think maybe 1 or 2 got taken
whiteknight I don't think I've ever seen an expressionless switch. I'll have to look into that and maybe create a task for it
dukeleto: The research tasks, what few we had, went quick last year. It was us who couldn't create enough to keep up with demand 20:46
dukeleto whiteknight: sure. But were the research tasks useful?
whiteknight dukeleto: the results weren't always the best, same with the translation tasks
dukeleto whiteknight: yeps, i hear ya
and i think we should get some translations of our New Developer Guide
whiteknight if we set clear goals and well-defined success criteria, it shouldn't be an issue. I don't have too many research tasks in mind this year
dukeleto whiteknight: sounds good to me. Don't let me bog you down with negative energy 20:47
whiteknight bleh, I'm not going to deal with any translations this year. Too much of a hassle. You're welcome to do it
dukeleto whiteknight: go forth and prosper (and make lots of tasks)
whiteknight: sounds fine, translations are on me
whiteknight We're up to 10 tasks now 20:48
cotto how'd we get accepted with less than 10 tasks? 20:49
NotFound whiteknight: is a switch were cases are expressions evaluated to bool.
cotto rhetorical question, I guess
Coke given the age range we're looking at, I'm not surprised at the lack of interest in research.
whiteknight NotFound: yes, I'm looking at it now
NotFound: very interesting. I'm going to make a task for that. Tests and documentation 20:50
20:52 estrabd joined 20:59 Hunger joined
whiteknight 12 tasks 21:03
dukeleto whiteknight++ 21:04
whiteknight I'm listed as mentor for 9 of them 21:05
this is going to be a busy couple of months, methinks :)
dalek nxed/inline: dc89bf0 | NotFound++ | winxedst1.winxed:
implement clone in NewExpr, ClassSpecifierId and InstanceOf

inline one more helper function		 21:07
whiteknight NotFound: I created the task to add tests for expression-less switch (I said to add the tests to t/switch.t, let me know if I need to change it) and to add documentation for it. Don't be surprised if you see a pull request sometime soon 21:08
NotFound whiteknight: no, tests directly in t/ are old ones not cleaned yet, should be in t/basic/ 21:09
whiteknight okay, gotcha. I'll fix it 21:10
dukeleto: Apparently I don't have the ability to approve tasks I create myself. If you would be so kind... 21:11
NotFound: Do you want t/switch.t to be "cleaned"?
I don't know if GCI students will know enough about the project to do that kind of work 21:12
NotFound whiteknight: I don't think that will be interesting for the student or for winxed. The expresionless variant server as example and doc, but the classic switch is well known, and is exercised enough in the compiler. 21:14
One advantage of a self compiler: you get a lot of tests for free ;) 21:15
I'm testing inline that way. 21:16
The trick is easy: stage 0 treats inline the same as function, not inlining at all. 21:17
dukeleto whiteknight: blarg. I made you an admin. What more do they want? 21:18
whiteknight: oh, tickets you create now will get approved Nov 28th
whiteknight: the meaning of "unapproved" changed
whiteknight: i can't approve stuff either
whiteknight: all tasks created now until the 28th will be released on the 28th 21:19
whiteknight okay
some tasks already are approved though, that's what confused me
dalek p/cclass: 0f3407b | diakopter++ | src/QRegex/P6Regex/ (2 files):
changes to support backslash endpoints
whiteknight is signing out, heading home. will think about tasks, add them in tonight when I get home. 21:23
I would like to get up to 25 by tonight, at least. I would like at least 50 by the end of the week 21:24
atrodo whiteknight++
whiteknight don't ++ me yet. Wait until I miraculously don't crash and burn, first
atrodo if not for that, i'm sure you'll deserve it sometime 21:27
whiteknight I'll be back tonight. Later
dukeleto slides about libgit2 that might interest some peeps: speakerdeck.com/u/tanoku/p/a-talk-about-libgit2 22:01
tadzik dukeleto: is (will be?) libgit2 capable of what panda is currently doing with shell "git ..."? 22:06
benabik "libgit2: a magical world of ponies fluffy clouds and cross-platform compatibility" 22:07
22:07 PacoLinux joined
tadzik actually, it only using clone and pull 22:07
NotFound But... no wings? 22:08
tadzik wouldn't that really be http-copying the git object files, and then inspecting them (equivalent of 'checkout master' of some sort) using libgit2>
22:17 PacoLinux joined
dukeleto tadzik: depends on what panda is doing with git 22:20
tadzik: cloning and checking out?
tadzik git clone, git pull 22:23
I suppose libgit2 can give us an equivalent of git checkout 22:24
so we just have to manage with getting files from urls
Coke dukeleto: I am taking myself off the hook for a blog post about the # of tasks we have since we got an extension! 22:25
I will, however, put myself back on the hook for a blog post about the extension!
dukeleto Coke: sounds good 22:30
tadzik: not sure about the status of cloning a remote. If it is not there in libgit2 yet, it will be soon. 22:31
tadzik: it will only implement git plumbing though, not porcelain like git rebase or stuff like that
benabik libgit2 doesn't do networking yet, I think.
Or I could be wrong... 22:33
github.com/libgit2/libgit2/tree/de...transports
dukeleto benabik: the latest release (0.15) has basic remote support 22:42
but yes, there are branches in flight that do more
bacek_at_work ~~
cotto hio bacek_at_work 23:00
bacek_at_work aloha, cotto 23:17
23:39 particle joined
dalek p/cclass: 74ca920 | jnthn++ | src/QRegex/P6Regex/ (2 files):
Simplify and correct the backslashes with .. in charclass patch. Needs more testing.
 23:41
23:45 whiteknight joined
whiteknight good evening, #parrot 23:48
dalek p: 0f3407b | diakopter++ | src/QRegex/P6Regex/ (2 files):
changes to support backslash endpoints
 23:52
p: 74ca920 | jnthn++ | src/QRegex/P6Regex/ (2 files):
Simplify and correct the backslashes with .. in charclass patch. Needs more testing.
p: d4f6e5c | jnthn++ | src/QRegex/P6Regex/ (2 files):
Merge branch 'cclass'
kudo/nom: 55e5294 | jnthn++ | tools/build/NQP_REVISION:
Bump NQP_REVISION to get char class improvements.
 23:54
kudo/nom: 4a57bd7 | jnthn++ | docs/ChangeLog:
Some ChangeLog entries.

Powered by Cro and the Raku® Programming Language.

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

Thank you!