Camelia, the Perl 6 bug

IRC log for #mojo, 2012-12-16

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

All times shown according to UTC.

Time Nick Message
00:00 good_news_everyone joined #mojo
00:00 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/tf371g
00:00 good_news_everyone mojo/master 361ad25 Sebastian Riedel: fixed README formatting
00:00 good_news_everyone left #mojo
00:02 sri markdown looks quite a bit better on github
00:04 chansen sri: at which resolution do you need CLOCK_MONOTONIC support?
00:05 sri chansen: same as EV ideally
00:05 sri i don't need it desparately enough for workarounds though, if Time::HiRes ever gets support i'm happy :)
00:07 chansen I'm looking into it, question is which platforms supports it
00:08 sri the python folks had a really good PEP about that
00:08 chansen I think we could work around it for OS that doesn't support it by caching the previous timeval/epoch and increase it by one if it equals the cached one
00:09 sri hmm, that would be pretty cool
00:09 sri http://www.python.org/dev/peps/pep-0​418/#operating-system-time-functions
00:09 sri that might be the PEP i mean
00:10 chansen and highly unlikely on nix OS which provide usec
00:12 good_news_everyone joined #mojo
00:12 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/4n6V7A
00:12 good_news_everyone mojo/master 9810edf Sebastian Riedel: more links in README
00:12 good_news_everyone left #mojo
00:14 wircus_ joined #mojo
00:15 Molaf__ joined #mojo
00:16 wircus_ /who
00:16 marcus woop
00:17 wircus joined #mojo
00:17 chansen sri: is that PEP backed by code?
00:18 Averna joined #mojo
00:18 wircus chansen: this client is rønning your unicode module nøw.
00:19 chansen wircus: näjs öf ju to rön my cöde!
00:20 sri chansen: no idea
00:22 chansen sri: this would be nice, I could implement it for common OS (Linux/darwin/Windows/FreeBSD) but beyond that I have no clue how portable the code would be
00:22 chansen perhaps you should bring this up on p5p?
00:22 chansen I'll back you up with code if necessary
00:23 luoboiqingcai joined #mojo
00:25 marcus fuudge
00:25 chansen marcus: tried with mixed Latin1 and UTF-8?
00:26 sri i suppose it would have to go through Zefram, since Time::HiRes is dual life
00:26 marcus chansen: ' nåjs øf ju tø wrÓte ∫†' <- this message didn't get through somehow.
00:28 sri markdown looks funny in textmate2
00:28 marcus sri: huge headings?
00:28 sri yea
00:28 sri HUUUUUUUUUUUUGE
00:30 marcus wirc is using 15mb. leaves quite a bit of ram for redis.
00:30 marcus I bet I'm leaking like a russian nuclear sub tho
00:35 chansen marcus: probably on your end, try this: $ perl -MUnicode::UTF8=decode_utf8 -wle 'printf "<%s>\n", join " ", map { sprintf "U+%.4X", ord $_ } split //, decode_utf8("\xE2\x98\xBA \xE5 \xC4 \xF0\x9F\x90\xAB", sub { shift })'
00:35 chansen should output:
00:35 chansen Can't decode ill-formed UTF-8 octet sequence <E5> in position 4 at -e line 1.
00:35 chansen Can't decode ill-formed UTF-8 octet sequence <C4> in position 6 at -e line 1.
00:35 chansen <U+263A U+0020 U+00E5 U+0020 U+00C4 U+0020 U+1F42B>
00:36 chansen no need to shift, can use sub { $_[0] } instead
00:37 chansen use 'no warnings "utf8"' to avoid warnings messages
00:39 chansen sri: either that or we create a new module
00:40 sri dngor already wrote a patch for os x support, the problem is to get Zefram to apply it
00:42 sri Time::Hires currently only has linux support i believe, dies on everything else
00:43 sri (for CLOCK_MONOTONIC)
00:49 chansen sri: yeah, and the Makefile.PL riddled with branches without comments, it's hard to se where it gains knowledge about support for clock_getres()
00:50 chansen and there is also the hints/*.pl
00:53 chansen Makefile.PL looks at $Config{d_clock_getres} or tries to compile a test file using two different c files, one using syscall() and one using gettime()
00:55 chansen We could start a Time::HiRes::Monotonic on CPAN and then try to get it merged with TimeHires
00:58 * chansen anyway, I need some sleep, sri, give it some thought. I'll be back later today
01:01 sri nn
01:14 Mike-PerlRecruiter_ joined #mojo
02:11 wircus joined #mojo
02:34 batman joined #mojo
02:40 sinkovsky joined #mojo
03:20 _xaka_ joined #mojo
03:46 noganex_ joined #mojo
04:00 wircus joined #mojo
05:14 Foxcool_ joined #mojo
05:18 zivester joined #mojo
05:27 mattastrophe joined #mojo
05:46 jberger joined #mojo
05:49 wircus joined #mojo
06:57 egopro joined #mojo
07:15 Miked joined #mojo
07:20 Vandal joined #mojo
07:40 dod joined #mojo
07:41 Drossel joined #mojo
07:56 wircus good morning, fellow hoomans.
08:03 dod joined #mojo
08:23 niczero joined #mojo
09:08 nic the time discussion is a bit confusing
09:08 nic Do you need just monotonic (not decreasing) or strictly increasing?
09:09 nic the 'add one' makes sense if you need strictly increasing
09:28 dotan joined #mojo
09:41 ladnaV joined #mojo
09:43 mire joined #mojo
09:45 kitt_vl joined #mojo
09:46 Vandal joined #mojo
09:59 egopro joined #mojo
10:06 zivester joined #mojo
10:06 ladnaV joined #mojo
10:18 Vandal joined #mojo
10:21 ladnaV joined #mojo
10:30 egopro joined #mojo
10:32 Vandal joined #mojo
10:35 ladnaV joined #mojo
10:55 egopro joined #mojo
11:24 zivester joined #mojo
11:50 Averna joined #mojo
12:37 Molaf_ joined #mojo
12:46 zivester joined #mojo
12:53 dabudabu joined #mojo
12:56 sinkovsky joined #mojo
13:14 Mike-PerlRecruiter_ joined #mojo
13:17 Foxcool_ joined #mojo
13:17 zivester joined #mojo
13:31 gryphon joined #mojo
13:38 Vandal joined #mojo
13:58 rem_lex joined #mojo
14:07 zivester joined #mojo
14:07 ObseLeTe joined #mojo
15:16 Britzel_ joined #mojo
15:30 sri in case anyone wants to help with the http/2.0 preparations https://github.com/kraih/mojo/i​ssues/423#issuecomment-11413551
15:36 lammel2 joined #mojo
15:55 ladnaV joined #mojo
16:02 Miked joined #mojo
16:25 sri would be neat if we could also abstract out connection management somehow, have CONNECT proxy support in its own class and maybe add SOCKS5 in the future :o
16:27 cstamas sri: yeah there are times (e.g when behind firewall) when socks/proxy support would be nice
16:28 sri we have proxy support
16:30 Adura SOCKS5, robot cat.
16:31 sri we'll prolly never have SOCKS in core, but i'm ok with adding a layer that will make proxy management pluggable
16:31 sri if anyone wants to work on it
16:34 mattastrophe joined #mojo
16:35 cstamas sri: one can easily make a socks proxy with 'ssh -D ...'
16:35 cstamas sri: i would not say it is important
16:36 cstamas sri: just as you mentioned it I gave some feedback ;-)
17:14 sh4 joined #mojo
17:27 kongelaks joined #mojo
18:44 sinkovsky joined #mojo
19:17 mire joined #mojo
19:28 keks joined #mojo
19:36 keks left #mojo
19:36 keks joined #mojo
19:37 keks left #mojo
20:04 sinkovsky joined #mojo
20:23 sinkovsky joined #mojo
20:44 jhthorsen_____ joined #mojo
20:59 b2b joined #mojo
21:03 wircus marcus: melding!
21:03 wircus wircus: enda en melding
21:10 jhthorsen joined #mojo
21:23 wircus joined #mojo
21:24 xaka joined #mojo
21:24 chansen wircus: did you find the character encoding issue?
21:27 chansen sri: I just had a look at the monotonic patch in RT, I don't think it will be applied because it's using static globals, which is a bad idea when perl is compiled with threads
21:46 sinkovsky joined #mojo
22:28 marty joined #mojo
23:28 jzawodn joined #mojo

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