Camelia, the Perl 6 bug

IRC log for #mojo, 2013-03-17

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:23 MeierM joined #mojo
00:36 riche sri: follow up on the AS perlapp/perlsvc issue
00:37 riche basically I tried every perl compile-stage trickery in the book, no dice.  I posted (instead of talking on phone) about this issue:  http://community.activestate.com/node/9602
00:43 sri riche: i suppose you have a point, guess i'll make it croak
00:49 mattastrophe joined #mojo
00:51 riche yah thanks a lot
01:01 sri you called it sloppy :)
01:02 riche I was being sarcastic ... obviously the file isnt that important or things would be breaking all over, and I don't know if you did that by design anyways
01:02 riche you or whoever
01:03 riche all I know is that this is a painful issue for me, and I am completely backed into a corner.
01:04 riche maintaining a copy myself is worst case scenario, its like openbsd all over again (well, maybe thats a bit dramatic)
01:08 riche there ... fixed ... been a long few days on this problem (and others) so I can see how it didn't come across sarcastic
01:17 Fremean joined #mojo
01:24 good_news_everyone joined #mojo
01:24 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/mlRMEA
01:24 good_news_everyone mojo/master 07a19fd Sebastian Riedel: modernized POD renderer a bit
01:24 good_news_everyone left #mojo
01:25 sri riche: dunno, your case kinda demonstrates that it might be better for detecting a broken mojolicious installation
01:29 sri ok, as far as i can see we now have 3 cases of unchecked open, all are arguable
01:29 sri https://github.com/kraih/mojo/b​lob/master/lib/Mojo/Util.pm#L27
01:29 sri https://github.com/kraih/mojo/blob/ma​ster/lib/Mojo/Server/Hypnotoad.pm#L59
01:30 sri https://github.com/kraih/mojo/blob​/master/lib/Mojo/Exception.pm#L104
01:30 sri marcus, tempire, jberger, crab: maybe worth taking a look
01:31 * jberger looks
01:31 sri Exception.pm being the most arguable... since it has mostly grown that way through trial and error, changes could be dangerous
01:35 jberger I would have said that Mojo::Util's open for entities is a no-brainer, except that riche's case demonstrates why NOT die-ing helped
01:36 sri "helped" is arguable ;p
01:36 jberger well, allowed him to continue
01:36 sri Mojo::DOM is broken there at least
01:36 sri and everything else that needs to handle entities
01:36 jberger absent that argument, I would say that that open should be checked
01:37 riche i agree open should be checked... even tho at this point its self-flagellation for me
01:37 jberger can you really not allow an $ENV{MOJO_ENTITIES_FILE} to allow that to be findable?
01:37 riche PLEASE
01:37 riche that would be awesome
01:37 jberger I suppose that is a security risk?
01:37 sri pointless
01:38 riche lol
01:38 sri look at the big picture
01:38 sri all the other bundled files
01:38 sri they are *all* affected
01:39 riche arent the other bundled files not relevant for production use anyway?
01:39 sri including jquery and prettify.js, all our artwork, built in templates and certificates
01:39 jberger to argue as devils advocate (I have no horse in this race): many of the other **needed** files can be changed in the app startup methods
01:39 riche but isnt that really for development purpose?
01:40 sri ok, i really don't like where this discussion is heading, so i'm out
01:40 sri consider the big picture for every change you're proposing
01:42 jberger I wouldn't change this without your blessing, and I said I was arguing as devils advocate. Like you say every feature needs a core-dev advocate and it wouldn't be me in this case
01:43 jberger personally it seems like activestate's problem to me :-/
01:43 jberger that doesn't help riche tho
01:43 riche jberger: agreed on the activestate thing
01:44 riche but that may never be resolved... seems like priorities are "in the cloud" now
01:47 sri jberger: there is nothing to advocate, i don't think anyone has even done the research to determine which parts actually would have to be changed
01:47 jberger sri: looking at hypnotoad, what would/should happen if STDIN/STDOUT/STDERR can't be re-opened?
01:48 sri jberger: when does that happen?
01:48 jberger the link you sent: https://github.com/kraih/mojo/blob/ma​ster/lib/Mojo/Server/Hypnotoad.pm#L59
01:48 jberger to "close" them
01:48 sri no, i mean when can't you reopen those?
01:49 jberger the question was, should we check success right?
01:49 jberger you would check in case they fail to open
01:50 sri should we really?
01:50 jberger perhaps it couldn't ever happen
01:50 sri is it fatal if those can't be reopened?
01:50 jberger but I'm asking, what would we want it to do if they failed to reopen?
01:50 jberger IIRC it would just silently continue as written
01:50 * sri is asking the same :)
01:51 jberger in that case they are closed because you don't want your daemon to be spitting info
01:51 jberger so its not really a problem if it does, just annoying
01:51 jberger worst case if your terminal can't handle it, just start it on another terminal right?
01:51 jberger seems like you wouldn't want those to be fatal
01:52 jberger Exception ... I have no clear sense on that one
01:52 sri http://grep.cpan.me/?q=%3E%26STDOUT
01:52 sri looking through cpan, non fatal seems to win
01:54 sri that case is prolly as harmless as it gets
01:54 jberger I agree, I think thats the rare case that failed opens are just a nuisance not a problem
01:54 jberger for exception, you are already in a die-like state right?
01:55 jberger it searching for context around a more real error
01:55 sri yes, just collecting source code for context
01:55 jberger seems like die-ing there would only serve to be more confusing
01:55 jberger so my vote is to die on entities and leave the others alone
01:56 jberger (sorry riche)
01:58 sri you know, there is another option
01:59 sri http://pastie.org/6577517
01:59 sri no change in behavior, but one line less code
02:00 sri side effect would be that the warning for riche goes away
02:01 sri downside, it might hide a broken installation
02:01 jberger is that better? fewer complaints, but hidden brokenness
02:01 sri just throwing it out there, not advocating
02:02 riche yah obscurity makes me nervous
02:03 jberger sri: here an odd thought, what about making much of entities.pl available as a method
02:04 sri why?
02:04 jberger and if entities.txt can't be found it builds a hash from the data online?
02:04 sri hell no
02:04 jberger then entities.pl would just be a wrapper around that
02:04 jberger haha ok
02:04 riche yah my customers would super freak on that
02:04 jberger just spitballing
02:04 sri the script breaks every now and then when the w3c template changes
02:05 jberger true, thats bad
02:05 jberger ok, I super take it back
02:05 riche sri: were you asking me why?
02:06 sri no
02:06 riche okee
02:06 jberger no that was to me
02:09 sri http://pastie.org/6577548 # if we want it to croak, we might as well use slurp()
02:11 jberger split on newline? I think I'd rather stick with <>
02:11 duncanthrax2 joined #mojo
02:11 sri why?
02:12 sri \n seems better to me for a file we control
02:12 jberger idiomatic? I've had odd problems with newlines cross-platform too
02:13 jberger since entities.txt is generated elsewhere and bundled
02:13 sri i don't see how input record separator is more portable
02:13 jberger it would probably be alright
02:13 jberger if you wanted to change it, why not Dumper in entities.pl and then do-load
02:14 jberger I guess there is still munging to do
02:15 mattastrophe joined #mojo
02:27 xaka joined #mojo
02:32 good_news_everyone joined #mojo
02:32 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/4xGwdA
02:32 good_news_everyone mojo/master 2e0dd08 Sebastian Riedel: modernized entity handling in Mojo::Util a bit
02:32 good_news_everyone left #mojo
02:32 sri jberger: then lets just do it the correct way ;p
02:34 sri sorry riche, but it got two votes
02:34 riche shrugs ... I didn't have my hopes up
02:35 riche I will just have to keep my own copy until someone does something... I am just gonna stick it in __DATA__ and be done with it
02:37 Fremean it's official... I hate debugging mobile devices
02:37 jberger sri++
02:41 kitt_vl joined #mojo
03:17 good_news_everyone joined #mojo
03:17 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/fbwVqw
03:17 good_news_everyone mojo/master 2eaf19c Sebastian Riedel: slightly better exception handling for Mojo::Exception
03:17 good_news_everyone left #mojo
03:17 sri and that should take care of Exception.pm :)
03:23 Fremean I just updated mojolicious now I have this spamming to stdout - Use of uninitialized value in numeric eq (==) at /usr/lib64/perl5/site_perl/5.12.4/IO/Socket/IP.pm line 651
03:24 Fremean Mojolicious 3.90
04:08 tagg joined #mojo
04:44 preflex_ joined #mojo
05:03 Caelum stupid question, but I'm setting up a dev env for another guy on the same server, I've taken ports 3000 and 3001, what should I give him?
05:04 Caelum I guess I'll give him 3002 and 3
05:08 jamesw or 4000
05:11 Caelum hmm yeah
05:16 crab or 8787
05:18 riche all our dev servers have automated port registries for this ... doing these as one offs is gonna drive you nuts, esp w /etc/services conflicts
05:20 crab how does this automated port registry work?
05:22 yakudza joined #mojo
05:22 riche devs can spin up whatever they want, keeps track of their src trees, sets a env var, and our server startup script sucks that up
05:23 riche culls ports every 24h
05:23 riche if they arent active
05:23 riche its just a sqlite db in /var tracks it
05:24 riche brain dead easy with no admin interaction, they are nonpriv ports after all
05:26 Caelum riche: that is awesome
05:37 crab yes, it sounds nice.
05:59 Vandal joined #mojo
06:06 ryozi joined #mojo
06:09 Caelum TtRenderer makes a /tmp/home/httpd dir for you, so if you have two users you're screwed
06:09 Caelum I should fix that
06:15 Caelum actually it doesn't, and I don't know what's doing that
07:21 marcus Neat
07:22 marcus "The connection to wss://wirc.pl/socket was interrupted while the page was loading" - Wonder why this happens.
07:22 marcus Maybe a nginx ws glitch
07:23 * marcus tries firebug to get wiser.
07:25 * marcus realizes firebug doesn't actually let you debug websockets, reconsiders his choice.
07:28 marcus Wow, WebKit nightly debugger actually looks really good
07:29 marcus it doesn't refresh the websocket frames automatically tho.
07:45 marcus Opera 'Dragonfly' doesn't even seem to show the websocket in the Network log.
07:48 d4rkie joined #mojo
07:54 ka2u joined #mojo
07:55 dod joined #mojo
08:09 dod joined #mojo
08:18 Averna joined #mojo
08:26 * Caelum looks up what the heck a websocket is
08:26 Caelum looks really cool
08:33 Caelum I remember xantus did all these awesome comet demos like 4-5 years ago
08:33 Caelum remember xantus?
08:53 basiliscos joined #mojo
09:02 ka2u joined #mojo
09:11 hrupp joined #mojo
09:15 marcus caelum: I remember him.
09:15 marcus caelum: I stayed at his place a couple of days when I was at WWDC.
09:16 sh4 joined #mojo
09:19 ka2u joined #mojo
09:20 yakudza joined #mojo
09:25 Fremean They call me the nose - neighbor just had his computer emit smoke so he called me over... one sniff - plastic, not electronic... second sniff - top half of case... third sniff - the AC97 cable was shorting to 12v
09:33 crab can you debug software with your nose too?
09:33 crab "this smells like an off-by-one"
09:33 crab damn, that would be cool.
09:34 Fremean I wish lol
09:38 ka2u joined #mojo
09:38 dotan joined #mojo
09:39 newbie joined #mojo
10:08 n0n joined #mojo
10:10 Vandal joined #mojo
10:18 coff joined #mojo
10:38 Britzel joined #mojo
10:38 n0n morning. got an array (1,2,3) and like to insert "connection"-elements ('a') like so: (1,'a',2,'a',3). much like join('a',@array), but returning an array. ideas?
10:43 vervain joined #mojo
11:06 bowtie joined #mojo
11:16 Averna joined #mojo
11:21 inokenty1 split "\0", join "\0a\0", 1, 2, 3
11:21 inokenty1 =)
11:27 inokenty my @a = map { $_, 'a' } 1, 2, 3; pop @a;
12:05 malikai am i right in assuming this example for ws&dbi is blocking?: http://blogs.perl.org/users/joel_berger/2012​/10/a-websocket-mojoliciousdbi-example.html
12:06 malikai or if i need nonblock i just use anyevent?
12:08 malikai i'd like to say $sth->execute(@query, sub { stuff..; }); like in node.js
12:09 mire joined #mojo
12:30 rem_lex joined #mojo
12:36 basic6 joined #mojo
13:01 kitt_vl joined #mojo
13:22 n0n joined #mojo
13:43 jberger malikai, the problem is that dbi isn't non-blocking
13:43 jberger you would use Mojo::IOLoop::Delay I suppose
13:53 newbie joined #mojo
13:56 gryphon joined #mojo
14:36 n0n joined #mojo
14:38 n0n hi. i'd like to generate excel files on-the-fly. and instead of writing them to a string, I wonder if I could write directly to a filehandle. is that possible?
14:38 n0n (passing it to the client...)
14:40 jamesw Excel::Writer::XLSX?
14:41 n0n yeah
14:42 jamesw it accepts a filehandle
14:42 n0n yes, but I ask for the other part of the story.. mojo
14:43 n0n get the mojo response fh and let the excel writer write to it.
14:43 n0n is there such a "response fh" and if yes, howto grab it?
14:51 rem_lex joined #mojo
14:52 malikai jberger, i gotcha.. that's why i was asking about using anyevent::dbi.. at least then it's only blocking in another process..
14:53 jberger I'm actually testing an updated version of the script in my demo you referenced and it seems to work
14:56 jberger https://gist.github.com/jberger/5181913
15:05 sri jberger: Mojo::IOLoop::Delay doesn't turn something blocking into something non-blocking
15:06 sri AnyEvent::DBI could work if you have EV installed
15:07 sri some DBI drivers like Pg also support async, which means you could use a recurring timer to poll them manually
15:09 jberger sri: no I suppose not :-/
15:09 sri the point of delays and "$delay->wait unless $delay->ioloop->is_running;" is to make something non-blocking work outside of our built-in web servers, like a PSGI server, CGI or stand alone
15:09 * jberger needs more coffee
15:10 jberger actually, I need to learn more about non-blocking, its one of those things that I think I understand one day, then it seems to slip out of my head the next day
15:15 jberger at least removing it simplifies the example
15:15 jberger :-)
15:16 jberger sri: your comment about is_running, does that mean that I don't need the is_running tests if I am in a webserver environment
15:16 jberger ?
15:16 jberger s/tests/check/
15:16 basiliscos Hello. I'm moving from java/j2ee world to Mojolicious, so, please, not  blame me on my question :). Is it possible to create in Mojolicious single background thread and performs some actions in background  ? The main (Mojolicious) application will communicate with the thread and provide answers to users... I know, it is possible to done via external command and communicate with it via IPC. Is there "Mojo" way to do that?
15:18 sri jberger: correct
15:18 Molaf__ joined #mojo
15:18 sri i've just added the is_running bit because it mostly "just works" and folks usually don't understand non-blocking ;)
15:19 * sri can actually count the people in the perl community that truly understand non-blocking on one hand
15:20 sri not unlike unicode, where you keep thinking "now i understand it!", only to learn that you really don't the next day
15:22 good_news_everyone joined #mojo
15:22 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/wg74nQ
15:22 good_news_everyone mojo/master 99dff6e Sebastian Riedel: fixed support for RFC 2817 in Mojo::Message::Request (closes #468)
15:22 good_news_everyone left #mojo
15:22 sri or like HTTP :S
15:23 jberger sri: I think I'm starting to get there on unicode, after banging my head against it for galileo, and after the long chats with you about the websocket opcode stuff
15:23 jberger thats not to say I will always do it correctly, but I think I understand it
15:23 sri well, i'm certain i don't understand unicode in perl completely yet :)
15:24 jberger right, its the "how to do it correctly" that I'm still learning, but I think I understand the problems :-)
15:24 jberger on unicode
15:24 * sri wonders how the hell he got the idea that Host was a hop-by-hop header
15:25 jberger we all have bad days
15:25 * jberger hands sri an irish coffee
15:26 sri could have been nginx, which used to follow HTTP 1.0 semantics and rip out Host headers by default
15:26 sri *slurp*
15:29 jberger actually, sri, can I have you check another non-blocking idea I had?
15:29 sri sure
15:29 jberger remember creaktive's non-blocking recipe for downloading?
15:29 jberger he posted it on SO once, so I countered with this:
15:29 sri vaguely
15:29 jberger http://stackoverflow.com/questions/1​5152633/perl-mojo-and-json-for-simul​taneous-requests/15166898#15166898
15:29 jberger is it any better?
15:30 jberger really only the start_urls function is important
15:30 jberger the rest is parsing and appending to the queue
15:31 sri using delay as a condvar is fine
15:31 sri that's how you make chunk of code work inside mojolicious and stand-alone
15:32 jberger cool
15:32 jberger :D
15:32 sri so it's a very good design
15:33 jberger see, sometimes I get it
15:34 jberger this morning, not so much
15:36 newbie joined #mojo
15:43 vervain_ joined #mojo
15:48 mattastrophe joined #mojo
15:50 vervain joined #mojo
16:03 vervain joined #mojo
16:09 vervain_ joined #mojo
16:10 basic6 joined #mojo
16:26 vervain joined #mojo
16:26 suy joined #mojo
16:32 vervain_ joined #mojo
16:41 d4rkie joined #mojo
16:47 mire joined #mojo
16:50 good_news_everyone joined #mojo
16:50 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/BXbfPw
16:50 good_news_everyone mojo/master 4eb1364 Sebastian Riedel: improved bad charset handling in Mojo::DOM::HTML
16:50 good_news_everyone left #mojo
17:00 good_news_everyone joined #mojo
17:00 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/nYo1ZQ
17:00 good_news_everyone mojo/master 3befa92 Sebastian Riedel: fixed small charset handling bug in Mojo::DOM::HTML
17:00 good_news_everyone left #mojo
17:06 vervain joined #mojo
17:10 dod joined #mojo
17:11 malikai anyevent::dbi does not appear to support something like ->fetchrow_hashref()
17:11 malikai i would very much miss fetchrow_hashref()
17:16 malikai sri, my backend is pg.. i'm looking into the async stuff there now
17:18 vervain joined #mojo
17:24 vervain_ joined #mojo
17:29 vervain joined #mojo
17:29 good_news_everyone joined #mojo
17:29 good_news_everyone [mojo] kraih pushed 2 new commits to master: http://git.io/-KpGcw
17:29 good_news_everyone mojo/master a70e919 Sebastian Riedel: fixed proxy detection bug in get command
17:29 good_news_everyone mojo/master 6f002f6 Sebastian Riedel: fixed HTTPS proxy support in Mojo::UserAgent (closes #468)
17:29 good_news_everyone left #mojo
17:41 Adurah joined #mojo
17:53 newbie joined #mojo
17:53 basic6 joined #mojo
17:55 malikai doesn't look like i can do callbacks at all with pg async..
18:05 vervain_ joined #mojo
18:10 basic6 joined #mojo
18:11 asarch joined #mojo
18:11 jberger malikai, metacpan searching found AnyEvent::Pg which might be helpful
18:11 jberger also its see also section has more non-blocking options to look at
18:11 jberger https://metacpan.org/module/AnyEvent::Pg#SEE-ALSO
18:15 malikai oh, the one by salva.. yea that guy's got his postgres-fu
18:15 malikai he did the good one on node.js
18:17 malikai thanks for that jberger :)
18:17 jberger :-)
18:22 sri https://metacpan.org/module/​DBD::Pg#Asynchronous-Queries
18:22 sri that's why i said you have to use a recurring timer and poll, you make your own callback system
18:23 sri it works, but you better be an expert
18:23 sri crab: are you still working on a non-blocking Pg client?
18:29 crab yes, but i've been busy with my daughter's school admissions, so i haven't been able to work on it much recently.
18:32 crab all this exam+school stuff is crazily stressful
18:44 good_news_everyone joined #mojo
18:44 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/y9xVyw
18:44 good_news_everyone mojo/master 0a591b3 Sebastian Riedel: better proxy tests
18:44 good_news_everyone left #mojo
18:44 sri wow, blocking HTTPS requests with proxy were completely broken
18:50 sri just because a few releases ago i moved the ioloop->stop code for blocking requests a little :o
19:08 sri http://www.python.org/dev/peps/pep-3156/ # cool stuff
19:10 rem_lex|pivo joined #mojo
19:25 lammel2 joined #mojo
19:27 Fremean joined #mojo
19:30 moltar joined #mojo
19:33 bpmedley left #mojo
19:48 jberger I like the idea of having explicit "Non-goals"
19:52 xaka joined #mojo
20:05 BeDa joined #mojo
20:09 basic6 joined #mojo
20:25 Fatalnix joined #mojo
20:35 lukep joined #mojo
20:36 riche Sessions Plugin: looking for alternative method of "last known page" --
20:37 riche right now we are using referral url to "get back to" the page the user was on prior to being logged out
20:37 riche but this isn't exactly working all of the time
20:37 riche so is using cookie for storing this ok?  opinions please
20:43 rem_lex joined #mojo
21:10 * sri is more frustrated by reading the PEP than he thought he would be :/
21:16 gryphon joined #mojo
21:20 riche gee i wonder, asynchronous and single threaded
21:29 marty joined #mojo
21:30 riche hypnotoad isnt supposed to work on windows, huh
21:38 sri Bender: trust marty
21:38 Bender OK, sri
21:39 sri Bender: trust crab
21:39 Bender OK, sri
21:39 sri Bender: trust tempire
21:39 Bender OK, sri
21:39 sri Bender: trust marcus
21:39 Bender OK, sri
21:39 sri Bender: trust jberger
21:39 Bender OK, sri
21:40 sri Bender: trust DaTa
21:40 Bender OK, sri
21:40 sri Bender: trust diegok
21:40 Bender OK, sri
21:41 sri Bender: spread the love
21:41 sri oh
21:41 jberger Bender: flummox sri
21:53 riche was wondering if hypnotoad was meant to work on windows... there is a bat file in site/bin for it but it doesnt work
21:53 riche or is this the wrong place to ask about it
21:57 riche left #mojo
21:57 riche joined #mojo
22:03 sri riche: isn't there a message when you run it?
22:04 riche yeps, Time::HiRes::ualarm(): unimplemented in this platform
22:05 riche Hypnotoad.pm ln 16
22:05 riche er
22:05 riche 15
22:06 riche I knew i tried this months and months ago and forgot why I didnt start using it, needed a refresher
22:07 sri ah, you're using an ancient version of hypnotoad
22:07 riche hmm
22:07 riche okay .... lemme see
22:08 riche if thats case cool .. will update
22:12 ka2u joined #mojo
22:12 riche hm running 3.80 ... its that ancient?
22:13 gryphon joined #mojo
22:13 riche unfortunately cant upgrade at this moment ... will take your word for it
22:15 riche I thought maybe I could upgrade it stdalone ... but cant it seems
22:16 sri hypnotoad does not work on windows, but there should be a friendly message telling you so
22:17 riche okay ... is it because you don't have a steward for it on window?
22:18 sri because it can't work on windows
22:18 lammel2 And windows will go away anyway soon ;-)
22:18 basiliscos joined #mojo
22:19 riche lammel12: been hearing that since 98, and been hearing that about mainframe since 91
22:20 riche sighs
22:21 sri there have been attempts at writing a windows version of hypnotoad based on threads, but they mostly sucked
22:21 lammel2 hehe, yes, "soon" is just not adhering to the speed of the internet, just like "Perl is dead" ;-)
22:24 riche sri: i can see how that would happen if using win32 subsystem ... but its been over a decade since I've done anything in windows posix subsystem
22:25 riche anyways maybe i will try via cygwin ... oh well thx
22:39 sri hmm, i guess some bugfixes are important enough for a 3.91 release
22:41 mattastrophe joined #mojo
22:57 Meiermann joined #mojo
22:58 good_news_everyone joined #mojo
22:58 good_news_everyone [mojo] kraih tagged v3.91 at ec37885: http://git.io/8CBDkA
22:58 good_news_everyone left #mojo
23:00 jpn joined #mojo
23:02 ka2u joined #mojo
23:04 bpmedley joined #mojo
23:05 mattastrophe joined #mojo
23:07 Adura joined #mojo
23:13 Averna joined #mojo
23:14 jberger only 8 more numbers left until ...
23:15 jberger sri: do you have a new codename yet?
23:15 sri not yet
23:16 sri but we had multiple good ones last time... could just look into the archive ;p
23:21 sri MILKY WAY, SUNRISE, GLOWING STAR, SHOOTING STAR, BEER MUG, TOP HAT, DRAGON, CROCODILE, OCTOPUS, PENGUIN
23:21 jberger ooooh beer mug!
23:21 sri ALIEN MONSTER
23:22 jberger http://en.wikipedia.org/wiki​/List_of_Unicode_characters <---- takes a really long time to load
23:22 jberger ALIEN MONSTER!!!
23:22 sri MONEY BAG
23:24 sri SMILING FACE WITH HORNS
23:25 sri ROCKET
23:26 jberger money bag might be good marketing :-)
23:27 rem_lex|pivo joined #mojo
23:29 jberger actually I think I'm going to steal ALIEN MONSTER as the symbol for Alien::Base :-P
23:30 sri :o
23:30 jberger HOURGLASS, kinda brings back old (probably bad) memories
23:31 jberger wait ... I got it ...
23:31 jberger BENZENE RING
23:31 * jberger hears crickets
23:33 sri *tumbleweed*
23:35 jberger why oh why are there unicode characters for parenthesized characters?
23:36 jberger I give that idea a â’¡
23:37 suy it could be worse. There could be code points for characters in italics... oh wait.
23:37 ka2u joined #mojo
23:38 jberger SHAMROCK
23:38 jberger RADIOACTIVE SIGN and BIOHAZARD SIGN
23:38 jberger SKULL AND CROSSBONES
23:39 * jberger should keep it to himself
23:39 * jberger is kinda having too much fun with this
23:39 freman SIGHUP
23:40 sri THIS WAS SUCH A PLEASANT ST. PATRICK'S DAY UNTIL THE IRISH SHOWED UP!
23:41 jberger tell them all the beer is gone, they'll move on
23:41 jberger oh I have a new favorite, HOT SPRINGS
23:42 jberger seriously who thought of these?
23:44 jberger apparently my font doesn't have full asian character coverage
23:49 jberger when I get around to making my command-line dice and card game app, I will be set
23:49 jberger SUNRISE OVER MOUNTAINS
23:50 jberger VOLCANO <--- new favorite
23:50 jberger good art too
23:51 ka2u joined #mojo
23:53 anaio joined #mojo
23:55 jberger I think from now on all my classes which already have use utf8 (Mojo::Base-y) are going to return a unicode character for true rather than 1;

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary