»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
sorear mberends: when you say "sometimes crashes" how badly do you mean? 00:01
mberends sorear: application level errors are not caught, such as division by zero 00:02
sorear I see 00:03
mberends also keystroke events still need to be caught. I've not found docs about how to read the $args parameter in an event handler.
sorear I got it to fail by trying to parse "Infinity2"
00:05 patspam joined
mberends I'm quite pleased that it's getting the right callbacks anyway. I'm kinda hoping @other will play with it and come up with much better little GUI apps. A text editor would be quite feasible. 00:06
sorear mberends: what is $args.GetType?
mberends sorear: I don't know
sorear I guess $args is a reference to some event structure
00:06 patspam left
sorear perhaps play with some code like 00:08
niecza> my $r = CLR::System::Random.new; my $ma = $r.GetType.GetMethods; for ^$ma.Length { say $ma.Get($_) }
there's something slightly ironic here, since the introspection for native Perl 6 objects is so much weaker than that :) 00:10
00:10 bbkr left, bbkr joined
mberends sorear: that looks like "shotgun programming", not far removed from the original Monkey Typing. 00:10
sorear has pondered making CLR arrays somehow look like Perl 6 iterables
00:11 patspam joined, risou is now known as risou_awy
mberends sorear: I'll continue searching tomorrow. All the docs found so far are too simple. It's time for sleep. 00:15
sorear mberends: can you point me to said docs? 00:16
'night
00:17 whiteknight left
mberends sorear: only the same mono pages you also linked, and various DuckDuckGo hits... 00:17
'night
00:26 packetknife joined 00:27 wamba left, clairvy_ left 00:42 packetknife left 01:03 replore joined 01:16 patspam left 01:41 HarryS left 02:18 patspam joined 02:29 JimmyZ joined 02:48 daniel-s joined 02:49 HarryS joined 02:50 molaf joined 03:01 wolfman2000 joined 03:03 packetknife joined 03:09 aloha left 03:22 JimmyZ left 03:24 packetknife left
tadzik 'morning 03:28
03:28 soh_cah_toa_ left
sorear hello tadzik 03:35
03:51 birdwindupbird joined 03:57 aloha joined 03:59 kaare_ joined, aloha left 04:00 aloha joined 04:02 uasi joined 04:15 kaare_ left 04:17 packetknife joined 04:18 skangas joined 04:19 molaf left 04:36 satyavvd joined 04:39 skangas left
moritz good morning 04:40
tadzik good morning moritz 04:42
04:45 bluescreen10 left 04:52 mikemol left 04:59 frhodes joined 05:01 frhodes_ joined 05:04 simcop2387 left 05:05 REPLeffect left 05:06 simcop2387 joined 05:07 uasi left 05:09 plobsing left 05:10 plobsing joined, envi joined 05:13 wamba joined 05:17 REPLeffect joined 05:21 packetknife left 05:30 itz_ left 05:31 itz_ joined 05:32 frhodes left 05:39 orafu left, skangas joined 05:40 ranguard left, orafu joined 05:42 ranguard joined
sorear moritz: good morning! 05:45
05:48 _jaldhar joined, jaldhar left 05:52 uasi joined 05:53 uasi left 05:59 wtw joined 06:05 orafu left, orafu joined 06:06 REPLeffect left, wamba left 06:08 masak joined
masak 哥们们,你的问候。 06:08
06:11 alvis left, SHODAN joined 06:18 frhodes joined, frhodes_ left 06:19 REPLeffect joined 06:26 alvis joined
jlaire masak: fyi, "no" is "ei" in Finnish, but it's inflected so that "I don't" is "en" 06:27
why im2ee or anyone would want to learn this language, I don't know...
06:34 koban` joined, koban` left
masak jlaire: ah. kiitos. :) 06:42
06:44 uasi joined
jlaire heh :) 06:46
06:49 uasi left
pnu whoah.. Huomenta Suomi! :-) 06:57
jlaire haha, huomenta huomenta 07:04
07:09 pernatiy left
masak Eilen oli puolalainen teema. Nykyään ilmeisesti on suomalainen teema. :) 07:14
mberends goedemasak #perl6 07:15
jlaire masak: erinomaista!
montakohan suomalaista tääl on lurkkimassa 07:18
pnu Puolalaisia on melkein kahdeksan kertaa enemmän kuin Suomalaisia.ihmisiä kuin Suomessa. 07:19
masak goedemoritz mberends 07:21
mberends :-) 07:23
pnu actually it would be nice to see a geographic mapping of users' ip locations found in this channel. 07:25
07:25 uasi joined
moritz I don't think the IRC servers tell us your IP 07:31
pnu moritz: Germany Bayern Nuernberg Herr Florian Forster ? 07:34
moritz pnu: the first two are correct :-)
pnu: the latter come from the assumption that IP == identity
07:34 wamba joined 07:35 uasi left
moritz pnu: how did you get that information? 07:35
moritz is pretty much still an IRC noob
pnu moritz: dig + iplocation.net - which was just the first one i googled
moritz pnu: and how did you get the IP? 07:36
pnu oh, and ofcourse with WHOIS.. "/whois moritz"
moritz dumb
I just did a /whois, but got a domain name instead of an IP back
and thought "how do I get the IP?" :-) 07:37
mberends moritz: IRC join messages contain some kind of host name or IP address (you have yours turned off)
moritz mberends: am I not displayed as moritz [moritz@faui2k3.org] or so?
mberends yes
moritz faui2k3.org is the host name :-) 07:38
mberends it is probably derived from reverse DNS, going by the IP addresses shown with many DHCP configured clients
moritz faui2k3.org is one of the names of the host which holds my screen+irssi session 07:39
mberends last examples, * wamba (~wamba@158.194.149.143) has joined #perl6, * uasi (~uasi@131.183.244.43.ap.yournet.ne.jp) has joined #perl6 07:40
pnu moriz: Herr Florian is the registrant of that domain, and it seems to be operated by noris.net AG at Thomas-Mann-Str.
mberends yes, it's where your irssi appears to connect from (modulo client NAT) 07:41
snarkyboojum I use an IRC bouncer, so it looks like I live in Ohio :) 07:42
moritz masak: did you know about www.nature.com/nature/journal/v436/...6776a.html and www.nature.com/nature/journal/v437/...7318d.html ? :-) 07:43
pnu snarkyboojum: but you use an norwegian irc server..? hmm.. :)
snarkyboojum pnu: ke? 07:44
pnu: I like lag it seems :) 07:45
moritz isn't the server just determined by DNS roundrobin? 07:46
pnu snarkyboojum: ah.. nevermind. i was just wondering if you're located in oslo but that's a long shot.. -> what moritz just said.
snarkyboojum pnu: yeah - nowhere near there
pnu anyway, for most cases, the client address is close enought (for interesting mapping). :-) 07:47
could i be more off topic? :)
masak moritz: no! cool!
moritz pnu: in the widest sense it's about the Perl 6 community :-) 07:48
pnu: now write a Perl 6 based IRC bot which looks up all the IPs and draws a map
:-)
mberends webchat users are the largest group whose real locations are unknown
07:51 mux left, mux joined
moritz you can shell out for the geoip lookup: host -t TXT 193.99.144.80.lookup.geo.opera.com 07:51
pnu moritz: hmm.. true. :) it's the perl6 geodiversity we're exploring here. 07:53
masak jnthn is probably the most geodiverse among us. 07:54
moritz nom: say any(1, 2).WHAT 07:57
p6eval nom 90a619: OUTPUT«Junction()␤»
07:58 wamba left, wamba joined 08:19 bbkr left
jnthn morning, #perl6 08:19
...huh, is masak saying I'm fat?
:P
08:20 uasi joined
moritz jnthn: no, just spread out 08:20
jnthn :P
moritz
.oO( unfocused )
mberends good mortiz jnthn
jnthn o/ mberends 08:22
masak jnthn: from now on, I'll take any compliment I get and shoot back "huh, are you saying I'm fat?" :P 08:25
jlaire this kinda reminds me of xkcd.com/559/ 08:26
except the other way :P
jnthn Urgh. I actually haz to do some work... 08:27
jnthn slurps the coffee and digs in to it
08:29 wamba left 08:30 M_o_C joined, Woodi left, Woodi joined
moritz imagines jnthn digging into the coffee 08:31
08:31 wamba joined
masak imagines him digging in to it :) 08:33
maybe the coffee is under a pile of $work
08:34 dakkar joined, daxim joined
moritz ./perl6 t/spec/S14-roles/parameterized-basic.t 08:48
===SORRY!===
Cannot use 'NarrownessTestB.new' as an argument to a parametric role as its value is not known at compile time at line 65, near " { }\nclass"
08:50 M_o_C_ joined
jnthn huh, I thought I fixed that test. 08:52
oh, committed, forgot to push 08:53
dalek ast: dceb8e3 | jonathan++ | S14-roles/parameterized-basic.t:
A couple of parametric role test updates.
08:53 M_o_C__ joined, M_o_C__ left
jnthn Sorry 'bout that; moritz++ for noticing. 08:53
08:53 M_o_C left 08:54 M_o_C_ left 08:56 wamba left 09:03 pnu left 09:04 pnu joined
masak I think perl6.org/compilers/features should have a row for "separate environments" under "Module management". 09:12
jnthn "seperate environments"? 09:13
As in, seperate compilation?
moritz has no idea what that could mean
masak one error that I keep making is that I 'use' modules once somewhere, and then don't use them again downstream from that in later files, even when it'd be necessary from a "real Perl 6" perspective.
both Niecza and Rakudo get this wrong presently.
jnthn nom should get it correcter.
moritz transitive use?
masak jnthn: it's related to separate compilation, but not identical.
moritz accidental transitive use, that is 09:14
jnthn masak: I don't see the distinction.
masak moritz: sort of.
jnthn: maybe there isn't one, then.
dalek ast: fcbbbbe | moritz++ | S16-filehandles/io.t:
clean up behind masak; be more explicit in testing the retrieved line
jnthn New NQP is a bunch more sensitive to that also
masak moritz: thanks :)
moritz masak: you're welcome :-) 09:15
jnthn When I started doing Rakudo atop of new NQP, I had to add a bunch of use statements.
masak not "clean up after", but "clean up *behind*" :P
shows what you think of my output :P
jnthn :D
tadzik freedom!
masak tadzik: William Wallace, is that you? 09:16
moritz masak: no, just a word by word translation of a German idiom
tadzik no, my head is fine 09:17
but my physics exam is over :) 09:18
jnthn tadzik++
tadzik well, not sure if it's passed :)
but it's over :P
09:26 [Coke] left
moritz
.oO( the end is nigh (
09:26
09:27 replore left
moritz sometimes it really itches me to write a Perl 6 FAQ for perl6.org 09:27
09:28 [Coke] joined
frettled at least the answers ;) 09:31
moritz well, the questions are determined by what people ask here on IRC 09:32
or on the mailing lists, blog comments, reddit, HN etc.
09:34 russellw joined
daxim "when is it ready?" bet ya never heard *that* one 09:34
jnthn
.oO( Frequently Questioned Answers )
09:35
moritz jnthn: I fondly remember the C++ FQA :-) 09:36
masak they don't seem to appreciate it over at programmers.stackexchange.com/quest...ed-answers 09:44
moritz well, it's a controversial piece of art 09:45
jlaire agrees with stackexchange answers 09:48
09:56 xinming_ joined 09:59 xinming left
masak somehow the stackexchange points in defense of C++ ("you can use the nice parts and it can be pretty nice") remind me of point in defense of Perl Classic. 10:19
or points in defense of JavaScript. 10:20
jlaire well, it's true :) 10:26
10:29 wamba joined
masak sure, but there's a point at which you might as well use Java, or Perl 6, or Coffeescript. 10:31
where a lot of the deficiencies have been fixed but you still get most of the good parts.
jlaire I strongly prefer C++ over Java and JavaScript over CoffeeScript; the deficiencies don't bother me at all once I know them 10:38
they just make learning a bit slower
masak that's an interesting point.
jlaire but it's good there are all kinds of languages
masak I don't really have any problems with Perl 5, either.
jlaire especially in case of JS, which is a fairly small language
C++ probably has lots of traps I'm still unaware of :)
masak there's probably very few people who couldn't claim that about C++ :) 10:40
10:43 [Coke] left, wamba left 10:44 [Coke] joined 10:46 replore joined 10:50 hanekomu joined 10:52 replore left 10:58 pernatiy joined 11:11 [Coke] left 11:13 [Coke] joined 11:15 skangas left
flussence JS is a pretty unfortunate case; it's a small and fairly decent language tied to one of the most bloated APIs ever invented. 11:17
11:18 mikemol joined
arnsholt It's also blemished by a syntax with some horrid warts, IMO 11:18
flussence yeah, lack of string interpolation is a pain... 11:19
jlaire fortunately jQuery provides a nice wrapper for DOM, and server-side JS doesn't need to care about it 11:20
if the syntax really bothers you there's CoffeeScript, but /me doesn't care 11:21
arnsholt flussence: And the whole newline or semicolon thing. Ew. 11:22
But the language itself is cool. Functional programming with prototype OO 11:23
masak loses himself in the C++ FQAs
jlaire if you always use semicolons, there's no problem... except reading others' code without semicolons maybe
but fortunately most people agree it's bad style to rely on the automatic semicolon insertion 11:24
masak arnsholt: JS is hardly functional programming in any strong sense.
arnsholt Yeah, but I find it quite Lispy (modulo syntax) 11:25
daxim news.cnet.com/8301-30685_3-20103843...r-the-web/ 11:28
11:30 [Coke] left 11:32 [Coke] joined 11:35 Psyche^ joined
masak wake me up when they actually debuts it :) 11:36
debut*
11:38 Patterner left, Psyche^ is now known as Patterner 11:39 bbkr joined 11:51 satyavvd left
masak finds he's not as good at golfing Niecza bugs as he is golfing Rakudo bugs 11:51
moritz why not? what's different? 11:52
11:53 alvis left
snarkyboojum Niecza sounds like a kind of bug, but then so does Rakudo :) 11:54
masak moritz: my familiarity with the system, most like.
snarkyboojum "Those pesky Niecza's last night. I know.. I was bitten by a whopping great big Rakudo!"
jnthn "Then I got chased by a load of Pugs!" 11:55
moritz "and smashed frontally into the smop"
snarkyboojum "only to slip in a core dump" 11:56
11:57 donri joined
snarkyboojum I always take it to the gutter :S 11:58
12:03 clairvy joined
flussence you're not in the gutter until you've got wasted on PHP 12:05
12:06 ab5tract joined, alvis joined 12:07 lestrrat left 12:08 lestrrat joined 12:12 [Coke] left 12:14 [Coke] joined 12:18 [Coke] left 12:19 [Coke] joined 12:28 patspam left 12:32 uasi left 12:34 patspam joined 12:36 uasi joined 12:45 patspam left 12:53 bluescreen10 joined 12:56 Holy_Cow joined
[Coke] thinks it would be nice if he could log into the irc log web app and mark "I've read through here", so he can easily catch up. 13:00
13:00 molaf joined 13:01 Holy_Cow left
moritz [Coke]: I know it's not the same, but if you click on the time column of a line in my logs, it's hilighted 13:01
13:01 russellw left
moritz so if you reload, you see how far you got 13:01
13:01 russellw joined
[Coke] moritz: doesn't help, since my normal client doesn't even tell me what day I'm looking at. ;) 13:02
moritz (the hilight is preserved through reload by adding the ID as hashtag to the URL)
[Coke] so if I'm backscrolling, I have to first find the line (with google search if I'm lucky, which I never am),
13:02 jjore left
[Coke] moritz: ah, so if I went /just/ web app, I could do that. 13:02
and if I kept browsing from the same place.
moritz right.
13:03 jjore joined, sbp left, [sbp] joined
[Coke] Having played with mojo recently, I'm very psyched about perl based web dev for the first time since about 1998. 13:03
13:03 [sbp] is now known as sbp
masak and remember: nobody can take your mojo. 13:07
moritz I have entertained the thought of reimplementing some of the supplemental Mojo:: classes (URL, Cookie, ...) to Perl 6 13:10
s/to/in/
again, way too many good ideas, too little time
arnsholt Indeed. I'd like to make some Mongrel 2 libs for Perl 6
There're even zmq bindings for Parrot 13:11
13:11 patspam joined
mberends masak: did you manage to run the Niecza Gtk GUI examples locally? You error message looked like how an assembly loader would spell 'file not found'. On Debian the required gtk-sharp.dll file is in the libgtk2.0-cil package. 13:20
*Your 13:21
masak mberends: oh, ok! I installed a gtk-sharp package, and when that didn't work either, I gave up. 13:23
moritz .u trademark
phenny moritz: Sorry, no results for 'trademark'.
moritz .u tm 13:24
phenny moritz: Sorry, no results for 'tm'.
flussence .u ™
phenny U+2122 TRADE MARK SIGN (™)
13:26 wolfman2000 left
moritz www.mahdiyusuf.com/post/9947002105/...g-syntaxes the last image is quite telling :-) 13:29
13:31 patspam left 13:34 tokuhirom joined 13:36 PacoLinux_ joined
flussence would be more interesting if it showed data for the shift keys too... 13:36
moritz and backspace
flussence depends where it gets the data from, it'd be kinda hard to do that for pre-existing source code :) 13:37
jlaire and I wonder which characters were typed manually and which were inserted by the editor
PerlJam It's odd (to me) that "e" is so hot in programmatic texts
moritz jlaire: it seems somebody just read in finished source code and mapped it to a keyboard
PerlJam: it appears in a lot of identifiers and keywords 13:38
PerlJam I guess that's just a reflection of english as the "base language" that we program in.
jlaire moritz: ok, so if it recorded actual keypresses while coding the stats could look quite different
I also wonder why 'x' is so little used... it's a great variable name :) 13:39
daxim makes a face like a horse
moritz jlaire: aye, hence my remark about backspace
jlaire oh, right
13:40 JimmyZ joined, packetknife joined
jlaire vim/gedit/emacs would probably make a bigger difference than the programming language 13:41
arnsholt PerlJam: I think e is the most common vowel in all Western European languages at least 13:48
Probably related to the fact that it's kind of the average vowel I think
13:49 icwiener joined, ingy joined
PerlJam arnsholt: aye, "e" is the most common letter in the language even. I just didn't think it so prevalent in source code. 13:49
arnsholt I think most code is alphanumerics (excepting pathological cases like obfus) 13:50
Especially if you start including comments
jlaire en.wikipedia.org/wiki/Letter_freque..._languages
moritz interestingly, the letter frequency doesn't differ much by the subject you write about 13:51
jlaire yep, 'e' is pretty high in many languages
masak which makes "La Disparation" all the more impressive.
arnsholt I have that on my shelf, but haven't actually read it yet 13:52
I tried several years ago, but gave up after I had to look up ten words on the first page alone
I suppose I could try "A Void" though
13:54 ingy left
masak yes :) 13:54
13:54 ingy joined
moritz en.wikipedia.org/wiki/A_Void#Plot_summary I like that the appraisal in The Times also doesn't include any e 13:54
13:57 PacoLinux left
daxim and that paragraph has only 6 es and 1 E 13:57
arnsholt Oulipo is a slightly insane literary movement 13:58
I really like it =)
sorear good * #perl6
masak: erm, what are you seeing exactly? niecza has "separate environments" 13:59
masak: note that if A defines GLOBAL::A::X, and B uses A, and C uses B, then C will be able to see GLOBAL::A::X because globals are always copied into calling units (S02:2692) 14:01
moritz if I write class A::X { }, does that create a GLOBAL::A::X? 14:02
jnthn Yes
(by spec)
If you "use A; use B;" then the compilation of B won't see stuff dragged in by A 14:03
moritz but it will see A itself, right?
jnthn But if A did "use C" and that had some globals, then yes, you'd see them.
moritz: No
moritz hmkay
jnthn moritz: We compile the module with a "clean slate"
Then merge its idea of GLOBAL with the current one. 14:04
And a clean slate means a clean idea of GLOBAL
moritz doesn't make a pun on clean/dirty ideas
masak sorear: oh, ok. 14:07
sorear: actually, I'm having a different problem right now, but I'm unable to golf it. 14:08
sorear: the exception I'm getting is "System.InvalidCastException: Cannot cast from source type to destination type"
moritz now please add source and destination type to that error message :-) 14:09
14:12 patspam joined
tadzik yay, ram arrived 14:13
jnthn That'll help ewe with your Rakudo hacking \o/
tadzik yep :0
:)
moritz \o/
tadzik seems that the exam is failed though :\
14:13 tokuhirom left
moritz /o\ 14:14
jnthn :(
tadzik tomorrow will be the begging day
we'll see
bbkr tadzik: what kind of exam?
tadzik bbkr: physics
14:15 mtk left
masak jnthn++ # horribly sheepish punning 14:16
bbkr tadzik: don't worry about physics unless you plan to build wormhole one day :) it's totally useless in IT. 14:18
tadzik bbkr: naah, I don't really care. I didn't even take my matural exam in physics
and I have still 3 semesters to pass it :) 14:19
moritz should preach against the unholy disregard of the most noble and ancient subject of physics :-) 14:20
14:21 wtw left
bbkr ok, i will rephrase: "it's totally useless in IT unless you are Angry Birds developer" 14:22
masak allow me to disagree. 14:24
physicists get one very useful tool for practically all other fields: that of thinking in models.
14:25 mtk joined
masak they go on to become great bioinformaticians. I bet they go on to become decent programmers as well. 14:25
moritz hopes he is on that way too
jnthn
.oO( A compiler is like a perfect, uniform sphere :-) )
moritz jnthn: that's what the mathematicians say :-)
arnsholt jnthn: Not a spherical cow? 14:27
masak .oO( cowmpiler ) 14:28
daxim frictionless cowtipping 14:29
sorear masak, moritz: mono --debug=casts 14:30
there's also the backtrace, a lot of functions have only one cast that could possibly fail 14:31
14:35 tokuhirom joined
masak sorear: "Unhandled exception: System.InvalidCastException: Unable to cast object of type 'Niecza.SubInfo' to type 'Niecza.STable'" 14:36
at Niecza.RuntimeUnit.FixupSubs
14:41 abercrombie joined 14:42 BooK left
masak in chasing Niecza errors down the golf chute, I get the distinct impression that the same input doesn't always generate the same error. 14:43
is there a compilation cache involved somewhere?
14:44 pmurias joined
pmurias sorear: hi 14:44
sorear: as people have seem to have problems with different mono version would it make sense to write a monobrew/perlbrew script? 14:46
masak: obj/ 14:48
masak pmurias: thank you.
I'd very much like a monobrew/perlbrew script for Niecza.
14:48 ab5tract left 14:53 icwiener left
pmurias masak: for installing a mono version or a niecza version? 14:53
pmurias meant to type nieczabrew
masak I assumed Mono, but installing Niecza would perhaps be even nicer :) 14:54
flussence would like a perl6brew script in general 14:55
14:58 tokuhirom left, birdwindupbird left
masak decommutes 14:59
15:00 masak left, uasi left, alvis left, wamba joined 15:01 simcop2387 left, SHODAN left 15:03 mberends left, thou joined 15:04 aloha left 15:05 simcop2387 joined 15:09 aloha joined 15:16 mkramer joined
dalek kudo/nom: ee12e5a | moritz++ | t/02-embed/01-load.t:
[t] fix deprecated use of run()

t/02-embed/01-load.t still fails for other reason though
15:23
15:25 alvis joined 15:31 benabik left 15:33 im2ee joined
im2ee Hi! :) 15:35
moritz \o
[Coke] thinks a good grasp of physics is helpful in general. But then, I have an engineering degree, and would recommend that to everyone as well, instead of comp sci. (helps with generic problem solving skills) 15:37
[Coke] likes math, but his interest petered out after partial differential equations. 15:38
15:39 jevin left 15:41 jevin joined
moritz would prefer it if physics worked fine without partial differential equations. It doesn't. 15:42
15:42 sayu joined
sorear phenny: tell masak Yes there is a compilation cache. Looks like you have problems of that nature 15:50
phenny sorear: I'll pass that on when masak is around.
[Coke] took 3 semesters of physics without needing diffeq. 15:51
sorear is something of a finitist and dislikes most continuum mathematics
arnsholt moritz: And next you're gonna tell us infinite frictionless planes don't exist? =p
dalek ast: 65541e3 | Coke++ | / (2 files):
nom fudging
15:53
15:54 daxim left
sorear to me, Banach-Tarski is a sign that \mathbb{R} is fundamentally wrong for describing the real world 15:54
dalek kudo/nom: 08ef941 | Coke++ | t/spectest.data:
track failures/run fudges
15:55
15:55 wamba left
sorear pmurias: I don't really follow you 15:55
15:55 wamba joined
sorear I have multiple mono versions installed side-by-side, it works rather well 15:56
15:57 alester joined 15:59 simcop2387 left
[Coke] is there a way to squash something into the first git commit in a repo? 16:00
(rebase -i won't let me include the first commit in the list.)
PerlJam [Coke]: there is, but I don't recall how to do it right off.
[Coke]: pretty sure you can find the answer on stackoverflow though
16:02 simcop2387 joined 16:09 daniel-s left
pmurias sorear: how do you switch between them? 16:18
16:19 packetknife left
sorear pmurias: I change the command I run 16:25
pmurias sorear: you don't set all env variables in www.mono-project.com/Parallel_Mono_Environments? 16:27
i had to do that to install gtk#
16:32 static_perl left 16:34 mkramer left 16:35 risou_awy is now known as risou
pmurias sorear: i'm not familiar with how mono works, but i assume that being able to easily install a tested version of mono would make installing niecza much easier 16:36
16:37 packetknife joined
[Coke] PerlJam++ 16:38
16:38 Trashlord joined, MayDaniel joined 16:47 daniel-s joined 16:51 mj41 joined, Chillance joined
pmurias sorear: easily meaning with a single command ;) 16:55
16:55 pmurias left 16:56 mj41 left 16:59 frhodes left 17:01 bluescreen10 left 17:03 wamba left 17:05 xinming_ left, frhodes joined 17:09 localhost joined 17:14 masak joined
masak sorear: if you're a finitist, what do you think of the surreal number system? 17:15
phenny masak: 15:50Z <sorear> tell masak Yes there is a compilation cache. Looks like you have problems of that nature
17:15 REPLeffect left
masak yes, I think so. 17:15
17:15 molaf left 17:20 wamba joined
TimToady the interesting news this morning is that NetLogic is being acquired by Broadcom 17:22
17:22 thou left 17:24 packetknife left
masak .oO( the merger, briefly known as "Broad Logic", quickly changes its name to "net.com" ) 17:26
17:27 REPLeffect joined 17:28 thou joined
sorear masak: since I don't like limit ordinals, my surreals are the dyadic rationals. 17:28
17:30 pernatiy left
masak sorear: :P 17:30
I can understand the skepticism towards limit ordinals. the ancient Greeks would probably agree wholeheartedly. 17:31
17:36 frhodes left, ab5tract joined 17:37 dakkar left 17:41 mj41 joined
diakopter TimToady: interesting 17:48
TimToady I knew the instant I saw that our stock price went up 50%, and said to Gloria, "Looks like we've been bought." :) 17:50
masak oh, it's your employer? then I see the relevance :) 17:51
diakopter hopefully they'll believe in Perl 6 too
PerlJam TimToady: what exactly do you do for NetLogic? 17:52
er, did
sbp do you did?
TimToady at the moment, still sitting in the company meeting :)
17:52 envi left 17:53 thou left
PerlJam hopefully you aren't decended upon by a hoard of bean-counters intent on "trimming the fat" or some such. 17:53
er, but with my spelling fixed
17:53 JimmyZ left
PerlJam :) 17:53
TimToady well, I'm not allowed to speculate publicly, but I believe I can point out this is not a hostile takeover 17:54
PerlJam blah, I've got a meeting in about 5 minutes 17:55
TimToady though of course the lawyers are already very busy :)
TimToady has a tremendous grasp of the obvious :) 17:56
meeting over, moving to my cubicloid now...
biab & 17:57
17:59 mspaulding joined 18:01 thou joined
diakopter PerlJam: TimToady provides pattern-matching expertise for NetLogic, which is a chimpmaker 18:02
er, chipmaker
masak oh right! 18:04
18:04 _jaldhar left
masak sounds like a perfect match, for some reason. 18:04
diakopter also I'm sure they like to be able to say he's a employoid 18:05
*an
18:05 im2ee left
diakopter so, name-dropping 18:05
18:05 bluescreen10 joined
masak if they don't, we're hiring! :P 18:05
18:05 ab5tract left
diakopter we're hiring too, and nearly infinitely closer :) 18:06
masak dang :P
jnthn Yeah but we have drunk elk.
Which, apparently, is a big news story...to people reading BBC news. :) 18:07
(Somehow, it managed to hit most read story at some point for several days running...)
18:11 alvis left 18:18 im2ee joined
TimToady funny think, almost exactly the moment you said "name-dropping", I knocked someone's nameplate off, and said "That's name-dropping." 18:18
*thing
diakopter that's name-dropping-dropping 18:24
TimToady same to you, but more of it 18:25
do grade school kids still say that?
diakopter I've never heard it 18:26
18:26 alvis joined
TimToady "Anything you can do, I can do meta." --Annie Oakley 18:26
"I can do anything meta than you!" 18:27
diakopter btw, I think I wrongly thought I was a hypochondriac this weekend
TimToady you're a sick person
TimToady wonders whether hypo or hyper chondria is more adaptive... 18:28
diakopter :) 18:30
18:30 sayu left
TimToady the answer is likely to be contextual 18:31
diakopter yeah depends on the (lack of) illness
masak this is like Epimenides exclaiming "I'm afraid I'm far too truthful a person!" 18:32
18:34 hanekomu left 18:35 bluescreen10 left
diakopter masak: I can't ever make absolute claims about anything 18:36
masak diakopter: :P :P
mikemol bounces 18:39
masak there's paradox in the air... 18:40
18:48 birdwindupbird joined 18:52 Trashlord left
masak rakudo: my $epimenides = (True + False) / 2; say $epimenides == (True - $epimenides) 18:54
p6eval rakudo 08ef94: OUTPUT«Bool::True␤»
masak :)
18:57 MayDaniel left 19:00 _jaldhar joined 19:01 PacoLinux_ left 19:04 soh_cah_toa joined
[Coke] perl6: say True + false 19:08
p6eval pugs: OUTPUT«*** No such subroutine: "&false"␤ at /tmp/3G1lnW2tWj line 1, column 5 - line 2, column 1␤»
..niecza v9-23-g1a8efca: OUTPUT«===SORRY!===␤␤Undeclared routine:␤ 'false' used at line 1␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) ␤ at /home/p6eval/niecza/src/STD.pm6 line 1136 (STD P6.comp_unit @ 36) ␤ …
..rakudo 08ef94: OUTPUT«Could not find sub &false␤ in <anon> at /tmp/4SODIDi4ff:1␤ in <anon> at /tmp/4SODIDi4ff:1␤␤»
[Coke] perl6: say True + alse // even
p6eval niecza v9-23-g1a8efca: OUTPUT«===SORRY!===␤␤Bogus term at /tmp/jlgCCXrN2S line 1:␤------> say True + ⏏alse // even␤␤Parse failed␤␤»
..pugs: OUTPUT«*** ␤ Unexpected "\ACK"␤ at /tmp/R4Zs1AOls3 line 1, column 12␤»
..rakudo 08ef94: OUTPUT«===SORRY!===␤Confused at line 1, near "say True +"␤»
[Coke] perl6: say True + False # fix many mistakes. ;)
p6eval pugs, rakudo 08ef94, niecza v9-23-g1a8efca: OUTPUT«1␤»
19:09 hanekomu joined
[Coke] masak: ping 19:10
phenny: ask masak to ping me about the old non-hague grant when he gets in. 19:11
phenny [Coke]: I'll pass that on when masak is around.
19:12 soh_cah_toa_ joined 19:13 soh_cah_toa left 19:22 PacoLinux_ joined 19:25 ilogger2 joined, ChanServ sets mode: +v ilogger2 19:27 stepnem joined 19:28 wallberg joined 19:44 gatinueta joined 20:13 y3llow joined, pothos_ joined 20:14 pothos_ is now known as pothos 20:16 y3llow left, pothos left, y3llow joined 20:17 pothos_ joined, pothos_ is now known as pothos 20:24 HarryS joined 20:28 supernovus joined 20:30 benabik joined 20:34 simcop2387 joined
supernovus I'm not sure if I'm doing something wrong, or if there is a genuine bug in the IO::Socket::INET in the nom branch, but when I try to create an instance, I get a "Could not find sub &INET" error. In the library itself, there are a few references to INET as a raw term. I couldn't find any bugs related to this on RT, and am not sure this is even a "bug". So yeah, just bringing it up, if it is a bug, I'll see if I can remember how to 20:41
submit to RT...
jnthn supernovus: Hm, oddness. May be somehow busted. 20:43
supernovus: I did at least get it to make a HTTP request.
masak rakudo: IO::Socket::INET.new; say "alive"
phenny masak: 19:11Z <[Coke]> ask masak to ping me about the old non-hague grant when he gets in.
p6eval rakudo 08ef94: OUTPUT«Could not find symbol 'IO::Socket::&INET'␤ in <anon> at /tmp/26CV9KuQC4:1␤ in <anon> at /tmp/26CV9KuQC4:1␤␤» 20:44
masak huh.
jnthn Hm
masak maybe p6eval doesn't have that class.
20:44 eternaleye joined
jnthn Oh 20:44
Yeah
It won't
masak for shcur'ty.
jnthn Yeah
The lexically nested IO package in SAFE.setting totally hides the outer one.
Which is what we want but breaks things here... 20:45
20:51 gatinueta left
supernovus The calls that seem to be breaking it are all nqp::getattr and nqp::bindattr calls. An example being -- nqp::bindattr(self, INET, '$!PIO', $PIO); 20:55
jnthn Ah
s/INIT/$?CLASS/ is probably more robust. 20:56
er, INET 20:57
masak rakudo: say (1..10).list.fmt('%d', ' | ')
p6eval rakudo 08ef94: OUTPUT«1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10␤»
masak rakudo: say (1..10).fmt('%d', ' | ')
p6eval rakudo 08ef94: OUTPUT«Too many positional parameters passed; got 3 but expected between 1 and 2␤ in method fmt at src/gen/CORE.setting:1410␤ in <anon> at /tmp/QoMWwNMEAJ:1␤ in <anon> at /tmp/QoMWwNMEAJ:1␤␤»
masak should the latter work? discuss.
it seems to have worked in b.
b: say (1..10).fmt('%d', ' | ') 20:58
p6eval b 1b7dd1: OUTPUT«1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10␤»
masak yup.
jnthn Not sure
masak I vote "yes".
jnthn Wonder where fmt lives in that base
masak fwiw
jnthn *case
b: say Range.^methods(:local)
supernovus changed all occurences of INET wit $?CLASS and recompiling CORE.setting, fingers crossed :-)
p6eval b 1b7dd1: OUTPUT«newboundsfromtoiteratorperlACCEPTSpostcircumfix:<[ ]>pickrollat_posofminexcludes_minmaxexcludes_max␤»
jnthn b: say Range.^mro 20:59
p6eval b 1b7dd1: OUTPUT«Method 'mro' not found for invocant of class ''␤ in main program body at line 22:/tmp/u5jupLXUX5␤»
jnthn b: say Range.^parents
p6eval b 1b7dd1: OUTPUT«Iterable()Cool()Any()Mu()␤»
jnthn b: say Iterable.^methods(:local)
p6eval b 1b7dd1: OUTPUT«itemNumericIntNumStrelemsfmtlist␤»
masak fmt!
jnthn aha, there's a fmt in iterable
masak huh.
jnthn nom: say Iterable.^methods(:local)
p6eval nom 08ef94: OUTPUT«elems infinite item Int Num Numeric Str␤»
masak submits rakudobug
jnthn nom: say Range.^mro
p6eval nom 08ef94: OUTPUT«Range() Iterable() Cool() Any() Mu()␤»
masak at the very least, this is something that needs looking into. 21:00
jnthn Probably easy to patch.
masak very probably LHF, yes.
jnthn Provided we feel that b's anwer is OK.
*answer
masak does :)
otherwise I'll have to patch connect-4 in a way I don't like.
jnthn I agree it should work, and Iterable seems a sane enough place to put it to me.
So no objects on my part if we make that work.
diakopter no objects indeed 21:03
jnthn er, objections :P 21:04
masak: Adding .fmt to Iterable now. 21:05
masak ooh! 21:07
jnthn supernovus: Trying a patch for your issue too 21:08
supernovus: oh, you're also doing so
OK :)
Imagine you have a class with a method m() { ... } 21:10
In a subclass you write a multi method m() { ... }, maybe some more cands
masak hm, there's an interesting new bug in connect-4 in which the first player wins after her first move!
jnthn Anyone disagree with the notion that that it should just generate a proto in the subclass?
masak well, I guess I'll have to track it down... 21:11
jnthn (At the moment it complains that there's an only method in the tree...)
masak jnthn: I agree with the notion.
supernovus Well, running SCGI-test.p6 doesn't crash now, but I can't connect to the port it should be listening on yet. So, the patch seems to have fixed the crashing issue, but I may need to play with it some more yet :-)
jnthn supernovus: OK, I'll push that patch.
dalek kudo/nom: 831b481 | jonathan++ | src/Perl6/Actions.pm:
Fix thinko in fakesignature action; another little step for putting :(...) := ... style binding back.
21:14
kudo/nom: ad9639a | jonathan++ | src/core/IO/Socket/INET.pm:
Socket fixes; supernovus++ for discovering the issue and testing the fix.
kudo/nom: e3a2fd7 | jonathan++ | src/core/Iterable.pm:
Make (1..10).fmt(...) work again; masak++ for noting it.
masak jnthn++ 21:15
supernovus Okay, I can confirm I can connect to the port, but now I get a nice neat new error: get_integer() not implemented in class 'Socket'. So, I will go digging again :-) 21:16
Hmm, not that I can see where there are any calls to get_integer anywhere in the core... must be a nqp/pir issue :( 21:20
jnthn supernovus: Run it with --ll-exception
supernovus: It'll give more info on what's up
cotto_work I need something that depends on Rakudo to start fleshing out how to make Parrot's allhlltest work with Perl 6 modules. 21:21
What's a useful and well-tested Perl 6 module that won't eventually end up as part of Rakudo?
soh_cah_toa_ irc::utils ;)
cotto_work It doesn't matter too much what it does as long as its tests pass and it doesn't too many external dependcies.
*dependencies
soh_cah_toa_ heard that was a pretty awesome module by pretty awesome guy 21:22
tadzik awesome
soh_cah_toa_ totally
plus, all tests pass on my machine at least 21:23
cotto_work how about Web.pm? 21:24
masak: ping
masak cotto_work: pong 21:25
cotto_work masak: do you see Web.pm as something that you'll be maintaining for a while?
masak ...as opposed to...?
supernovus Hmm, it seems the accept() method has an issue. It seems to be trying to return a nqp::p6bool(), but the expected output is an object representing the connection. 21:26
masak cotto_work: I'm not sure I understand the question.
cotto_work masak: I'm looking for a starting point for testing Perl 6 module as part of Parrot's allhlltest.
masak ah.
Web.pm has lots of code, but it's not guaranteed to be up-to-date. 21:27
jnthn supernovus: Maybe the fool who ported it from ng to nom knew nothing about sockets ;)
masak cotto_work: something like Pies might be a better choice.
cotto_work masak: does that mean "likely to break" or just "sub-optimal"
masak cotto_work: a bit of both.
cotto_work I see no "Pies" on modules.perl6.org/ 21:28
where does it live?
supernovus jnthn: I'll play with it. I don't know much about nqp, but I'll dig through the code.
cotto_work: see Panda 21:29
tadzik cotto_work: panda
masak panda :)
cotto_work cotto_work: panda
masak :P
tadzik cotto_work: github.com/tadzik/panda/blob/master/lib/Pies.pm you want something implementing this protocol 21:30
masak I should've said Panda, not Pies.
tadzik possibly using the roles already provided by Panda, as in Panda::Builder and so
cotto_work panda looks like a good first step
tadzik panda is just an installer
Pies is supposed to be a general implementation for module mangling
I once wrote something for smoketesting using the Pies "protocol" 21:31
dalek ast: ff5b76b | jonathan++ | S12-methods/lastcall.t:
Update lastcall.t for current spec.
21:32
im2ee Good night! :) 21:34
masak im2ee: dobranoc 21:35
supernovus Okay, maybe "playing with it" wasn't such a good idea... now instead of a useful error message, I get "Segmentation fault (core dumped)"... I should avoid playing with internals :-P 21:36
jnthn Congrats! 21:40
supernovus I wish I understood the bindattr/getattr stuff better. Comparing the ng class to the nom class it interested, yet ultimately futile, as I don't know what to "box" the getattr results into. Any NQP primers? 21:47
21:48 benabik left
cotto_work its test suite is the best I know of 21:48
tadzik I think it's about wrapping a low-level object in a Perl 6 object
jnthn supernovus: What do you expect to come back from it?
masak b: my @a = [1], [2], [3]; say (map { @a[1 - $_][0] }, 0 .. 3).perl
21:48 kaare_ joined
p6eval b 1b7dd1: OUTPUT«(2, 1, Any, Any)␤» 21:48
masak nom: my @a = [1], [2], [3]; say (map { @a[1 - $_][0] }, 0 .. 3).perl
p6eval nom 08ef94: OUTPUT«(2, 1).list␤»
masak submits rakudobug
tadzik here we go again
supernovus A Socket object. 21:49
It may be an IO::Socket::INET object actually, representing the current connection.
21:49 benabik joined, wallberg left
jnthn supernovus: ah. 21:50
masak tadzik: this change in semantics causes connect-4 to give the first player the win on the first move.
supernovus I think my attempt to "fix" it ended up returning the raw parrot Socket object, which of course rakudo looked at and promptly died ;-)
tadzik masak: serious business 21:51
masak tadzik: seems connect-4's winning condition relied on being able to do negative indexing.
tadzik: I have no trouble believing that this isn't spectested.
and I have no horse in this race. I'll gladly rewrite that part of connect-4 if need be. 21:52
supernovus I'm guessing the accept() method hasn't been tested up until now as most uses of the library are for client connections, whereas I'm trying to port my SCGI daemon library to nom :-)
masak but until there's a verdict, it goes into RT.
niecza: my @a = [1], [2], [3]; say (map { @a[1 - $_][0] }, 0 .. 3).perl
jnthn nom: my @a = [1], [2], [3]; say (map { say @a[1 - $_][0].WHAT; @a[1 - $_][0] }, 0 .. 3).perl
p6eval niecza v9-23-g1a8efca: OUTPUT«(2, 1, Any, Any).list␤»
nom 08ef94: OUTPUT«Int()␤Int()␤Nil␤Nil␤(2, 1).list␤»
masak there'a another data point, I guess.
jnthn nom: say (map { Nil }, 0 .. 3).perl
p6eval nom 08ef94: OUTPUT«().list␤»
jnthn masak: There's the underlying thing. 21:53
tadzik supernovus: make sure to make friends with IO::Select :)
masak jnthn: I suspected that.
jnthn: why Nil and not Any?
jnthn masak: I can believe that's actually correct. But I'm not quite sure if the Nil you got before is wrong.
masak: That is, it's not map's fault..but @a[1 - $_][0] returning Nil may be wrong.
masak aye.
I know.
21:54 daniel-s joined
jnthn Well, so that golfs if. :) 21:54
nom: my $num = '4'; say (+$num).WHAT 21:55
p6eval nom 08ef94: OUTPUT«Int()␤»
jnthn b: my $num = '4'; say (+$num).WHAT
p6eval b 1b7dd1: OUTPUT«Num()␤»
jnthn Pretty sure nom's right here. Comments? 21:56
masak nom++
jnthn ok, then I'll fix the multi test that wrongly relies on it :)
21:58 wolfman2000 joined
supernovus tadzik: IO::Select looks nice, I'll have to dig into it a bit more. Hopefully I can get my SCGI library working under nom, then I can start extending its feature set. Since writing it, the only other time it got updated, was to port it from alpha to ng... it's a pretty simple library at the moment. 21:58
[Coke] (smoketesting, pies) should that have been mentioned in re: my ping to the list last week? 22:00
dalek ast: 48a5530 | jonathan++ | S12-methods/multi.t:
Start updating S12-methods/multi.t for current semantics.
22:01
22:02 wolfman2_ joined
dalek kudo/nom: a5d0ca2 | jonathan++ | src/core/control.pm:
Fix handling of lack of candidate to defer to in nextsame/nextwith.
22:03
kudo/nom: ef8480f | jonathan++ | t/spectest.data:
Run S12-methods/lastcall.t.
kudo/nom: bb2a5ad | jonathan++ | src/Perl6/Metamodel/MultiMethodContainer.pm:
Remove a die in a place where we should just go ahead and generate a proto. Fixes compilation of S12-methods/multi.t, though it doesn't yet pass.
tadzik [Coke]: what ping?
22:04 wolfman2000 left 22:06 alvis joined
supernovus is crossing fingers and trying something crazy... 22:11
sorear good * #perl6 22:13
diakopter / 22:15
masak goodsorear 22:16
22:17 wolfman2000 joined
masak niecza: my @a; say @a[-3][1] 22:18
p6eval niecza v9-23-g1a8efca: OUTPUT«===SORRY!===␤␤Unsupported use of [-3] subscript to access from end of array; in Perl 6 please use [*-3] at /tmp/Yrr1qF_V0Q line 1:␤------> my @a; say @a[-3]⏏[1]␤␤Parse failed␤␤»
masak niecza: my @a; say @a[$_][1] given -3
p6eval niecza v9-23-g1a8efca: OUTPUT«Any()␤» 22:19
masak sorear: Any(), or Nil() ?
22:20 wolfman2_ left
sorear masak: tentatively, I will say Any is correct, but I am open to arguments; that code predates Nil 22:25
masak I'm also leaning towards Any. 22:26
reason I'm asking is nom does Nil, but b does Any.
I just submitted a rakudobug about it. 22:27
sorear I thought b did &die
b: my @a; say @a[$_] given -1;
p6eval b 1b7dd1: ( no output )
sorear b: my @a; say @a[*-1].perl
p6eval b 1b7dd1: OUTPUT«undef␤»
sorear hrm
jnthn b did some cute things...
undef? really? :) 22:28
benabik What undef?
sorear wonders wtf the (no output) was about
shouldn't &say always at least make a newline?
benabik Perhaps evalbot interprets "\n" as nothing
nom: say '' 22:29
p6eval nom e3a2fd: OUTPUT«␤»
benabik Or not.
diakopter nom: print ''
p6eval nom e3a2fd: ( no output )
diakopter heh
sorear once upon a time, p6eval had a bug which would interpret segfaults as a lack of output, but that's been fixed
b: my $a := $a; say $a;
p6eval b 1b7dd1: OUTPUT«Segmentation fault␤»
jnthn nom: my $a := $a; say $a; 22:30
p6eval nom e3a2fd: OUTPUT«Any()␤»
jnthn phew ;)
diakopter hm, I suppose that's a known segfault
oh
jnthn diakopter: b did some...interesting things...with binding.
(where interesting means "omg WRONG") 22:31
cotto_work what's b? 22:32
jnthn cotto_work: beijing release
cotto_work thnaks
jnthn cotto_work: The last release we did from the old master.
22:32 patspam joined 22:33 wolfman2000 left 22:37 patspam left, wolfman2000 joined 22:44 wolfman2000 left, wolfman2000 joined
masak 'night, channel 22:48
TimToady o/
supernovus Well, fixing accept() in IO::Socket::INET is beyond me. I'm not sure how to wrap the parrot Socket in a Perl 6 IO::Socket. I fail. 22:56
jnthn supernovus: hm 22:57
supernovus: Did you try something like
my $new_sock := nqp::create($?CLASS);
nqp::bindattr($new_sock, $?CLASS, '$!PIO', $the_parrot_socket);
?
22:58 thou joined
supernovus Nope, that looks much nicer than the horrid mess I did try... 22:58
jnthn oh, but there's a bunch of other attrs that need initializing too...
but maybe you get away with it... 22:59
supernovus I need to read the NQP docs more. 23:01
jnthn I suspect we need to write the NQP docs more...
23:02 kaare_ left
soh_cah_toa_ loves anything w/ the phrase "more documentation" 23:02
benabik soh_cah_toa_: You should write more documentation. 23:03
soh_cah_toa_ benabik: i am. working on a chapter on testing for perl6/book
23:09 tokuhirom joined
supernovus jnthn: Well, I'm getting closer. I need to initialize $!buffer too. 23:12
jnthn nqp::bindattr with '$!buffer' and setting it to '' should work, it seems. 23:13
(e.g. the empty string)
23:16 patspam joined
supernovus that's what I'm trying actually, didn't know if it would work or not, but figured why not try :-) 23:16
Hmm, apparently that doesn't quite work, "Cannot assign to a non-container". Ah well, I'll keep playing :-) 23:19
jnthn oh 23:21
yeah, that's trickyish :)
You can try this:
nqp::getattr($obj, $?CLASS, '$!buffer') = '';
(where $obj is the object you're populating)
It'll already have a container thanks to having done nqp::create 23:22
So you just grab it and assign to it.
So maybe not so trickyish after all :)
23:28 wolfman2000_ joined 23:30 wolfman2000 left 23:50 whiteknight joined
supernovus Well, it seems really really slow, but connections do seem to be accepted now. They are so slow that mod_scgi times out before the request is received, but that's better than crashing I guess. 23:52
diakopter that's interestingish
jnthn That's...odd.
jnthn wonders what could be causing connection acceptance to be so slow 23:53
supernovus: I need to sleep now. Feel free to send a patch or pull request for the fixes so far. 23:54
supernovus: Would be good to get to the bottom of the slowness. 23:55
Can hopefully find tuits to help with that in the coming days.
diakopter jnthn: are bvalues a slushy portion of the spec, or frozen 23:56
jnthn diakopter: bvalues?
diakopter lol I guess not 23:57
jnthn diakopter: As in, b for bind?
diakopter yeah at some point I read about bvalues
supernovus jnthn: Okay, I'll keep testing more. For some reason, the recv() seems to only get the string when the remote connection is closed, so it's not getting the string from apache until apache has timed out... that is definitely weird.
jnthn diakopter: If so, those were a SMOP idea that I'm not really sure I believe in. Binding in Perl 6 is meant to be performant, and is constrained somewhat to make that possible. So there's no "action at a distance" binding. 23:58
supernovus: That sounds eerily familiar...
diakopter jnthn: news to me; it would be good if your last sentences are "binding" canon
jnthn supernovus: Sadly I can't remember any more, but maybe somebody will read this and do so... :)
supernovus I tried telnet localhost 8118 and typing the manual QUIT command for the SCGI daemon, and the request was never recv()ed... oO
jnthn diakopter: :)
supernovus anyway, I'll send a patch for the IO::Socket::INET changes. 23:59
jnthn Many thanks.
And really, sleep...or I'll be doing way too much coffee tomorrow...again :)
diakopter jnthn: I wish binding/assignment/captures/containers were better explained somewheres
jnthn night o/