Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2015-02-23

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

All times shown according to UTC.

Time Nick Message
00:16 mohawk new MR: https://sourceforge.net/p/pdl/code/merge-requests/28/
01:21 mohawk this is a nice fast dev cycle:
01:22 mohawk perl Makefile.PL ; (cd Basic/;make pm_to_blib) && (cd Basic/Core/;make) && (cd Basic/Gen/;make) && prove -b t/pptest.t
01:23 mohawk 14 secs
01:25 mohawk (and i just moved Basic/Test from being installed, into just being extracted and tested in t/pptest.t)
05:24 jberger :o
05:28 sivoais mohawk++ # I'm seeing all these MR in my INBOX
05:47 mohawk ha ha
05:47 * mohawk bows
05:47 mohawk of the 7 bugs chm rated as 6+, i reckon we can close about half now
05:47 mohawk as well as delete a branch
18:06 mohawk sivoais, where are we with bug 347? i thought we fixed it thanks to your travis work?
18:08 mohawk and i thought 346 was fixed by 350?
18:12 sivoais mohawk: let me take a look in a few
18:12 mohawk awesome
18:12 mohawk sivoais++ # the man
18:16 mohawk i'm looking at bug 343, i'll make a *.t test and hopefully a fix
18:17 mohawk i'll probably cherry-pick/merge longlong* over as well
18:17 mohawk byebye branches
18:45 mohawk i'm rebasing longlong* onto master, and it's turning into an epic
18:45 mohawk i think i may need to get "clockwork orange" with chm and craig on "this is how you rebase a branch and keep it up to date"
18:54 sivoais o/~ [Beethoven's 9th] o/~
18:55 mohawk bit of the old ludwig van
18:55 mohawk sivoais, got a view on 346?
18:56 sivoais oh, I had that a bit before... the Travis-CI still passes without Fortran
18:56 mohawk please comment on the bug ticket then :-)
18:57 sivoais done
18:57 sivoais <https://sourceforge.net/p/pdl/bugs/346/>
19:00 mohawk nice!
19:06 * jberger ponders some ultraviolence
20:11 mohawk second commit of longlong, i'm already into some deep PDL shit
20:11 mohawk #     Error:  [PDL->Version: 10 PDL_CORE_VERSION: 11 XS_VERSION: 2.007_11] PDL::Slices needs to be recompiled against the newly installed PDL at /usr/lib/perl5/5.20.1/x86_64-linux/DynaLoader.pm line 216.
20:12 mohawk i may be some time :-)
20:13 mohawk going forwards, i will refer to longlong* as LLDF
20:38 * jberger prepares the eulogy
21:43 opkick [pdl] wchristian pushed 1 new commit to master: http://git.io/APvE
21:43 opkick pdl/master 1e3c95f Craig DeForest: Add some documentation, and make tests and such slightly more forgiving.
22:05 mohawk jberger, you sarcastic bastard
22:05 mohawk jberger++
22:08 Mithaldu out of idle curiosity, any of you math people maybe know a more elegant way to get this graph? https://www.desmos.com/calculator/69llxwdn2f
22:16 sivoais Mithaldu: all I can think of is that you can turn the rational into a Taylor series
22:17 Mithaldu spoiler: i arrived at that formulaa by starting with inverse square and then flailing at it akin to a monkey
22:18 sivoais Mithaldu: something like : <http://www.wolframalpha.com/input/?i=taylor[+1+/+(x-2)^2+]>
22:19 Mithaldu ah
22:19 Mithaldu turns out there's software for that
22:19 Mithaldu $y=-((f-4)*x^2+(2*f-8)*m*x-3*f*m^2)/(3*x^2+6*m*x+3*m^2)$
22:20 Mithaldu including wolfram
22:20 Mithaldu http://www.wolframalpha.com/input/?i=1%2B%28f-1%29%284%2F%283%28%28x%2Fm%29%2B1%29%5E2%29-1%2F3%29
22:21 sivoais and if you want to do it from Perl, I think you can interface with Maxima
22:22 sivoais caveat lector: I haven't tried the CPAN Maxima interfaces
22:23 vicash Mithaldu: plot 1 / log x on Wolfram
22:23 Mithaldu sivoais: i don't need to go that far, it's just a formula for the screenshake in microidium
22:24 Mithaldu vicash: you saying i could approximate it with that?
22:24 vicash Mithaldu:  yes with some variation. you can try "plot |log(x)/x|"
22:25 sivoais microdium ?
22:25 sivoais * microidium
22:25 vicash the first plot that shows up on Wolfram is similar to what you want but the scale is different, so scaling may be needed in some form.
22:25 sivoais ah: <https://github.com/wchristian/Microidium>
22:26 Mithaldu sivoais: yeah that
22:26 vicash the original desmos.com plot is an asymptotic curve.. so anything that has x in the denominator will be needed
22:26 Mithaldu vicash: well, i think inverse square is a specific form of inverse log?
22:27 Mithaldu oh good, i managed to obliterate the youtube link from the readme :(
22:27 vicash actually the Taylor series expansion of log(x)/x is inverse square approximated
22:27 vicash i think :)
22:27 Mithaldu well
22:28 Mithaldu if i need good formulas for effects later on i'll ask you guys :P
22:28 sivoais oh, it's networked! sweet
22:29 Mithaldu crudely
22:29 sivoais have you seen OpenParsec?
22:29 sivoais and XPilot
22:29 Mithaldu sivoais, vicash: https://www.youtube.com/watch?v=KoLMoc5RvQ4
22:30 Mithaldu sivoais: no i hadn't
22:30 Mithaldu and i'm not shooting that high either :)
22:30 Mithaldu openparsec looks kinda dead too
22:31 Mithaldu haha, xpilot looks cute, only i plan to actually have effects
22:31 mohawk Mithaldu, there's a POD error at bottom of your README?
22:31 Mithaldu otoh!
22:32 Mithaldu i can steal ideas from xpilot
22:32 Mithaldu mohawk: yeah, Dist::Zilla being blah
22:32 vicash Mithaldu: you need to write this as an Android app.. use the accelerometer and the user just moves their phone in 3D to do the shooting. it will be pretty awesome then
22:33 Mithaldu vicash: nah, not nearly enough fidelity
22:33 Mithaldu also
22:33 Mithaldu i've done some android development and there are few worse levels of hell
22:34 vicash i agree... i have never tried but there is a CPAN Android module. not sure if it really works on stock Android, or if Perl needs to be recompiled for it to work on Android.
22:34 mohawk i want perl and pdl to work on android
22:35 mohawk once i've finished fixing pdl i may do that next
22:35 mohawk (i know hugmeir's done a lot of work on it, but i mean *work*)
22:35 vicash that would be cool..
22:35 Mithaldu well, i'll happily help anyone port it
22:35 mohawk i want to see Mithaldu's game working on it
22:35 mohawk Mithaldu, port what?
22:36 mohawk does opengl work on android?
22:36 Mithaldu microidium to android :|
22:36 Mithaldu haha, yes android phones ahve opengl
22:36 mohawk Mithaldu, as a background activity, could you get an android emu and put perl on it?
22:36 Mithaldu the issue will be that it needs to talk to the android gui
22:36 vicash well, in theory one could auto-generate the appropriate Android Java code using Perl and make it work that way too. but that would be a subset of perl
22:36 Mithaldu nope
22:36 mohawk i remember seeing some instructions on how
22:36 Mithaldu talk to ...
22:36 Mithaldu castaway
22:37 Mithaldu she's been working on that
22:37 vicash mohawk: android supports OpenGL ES. both in the native SDK and in the java sdk
22:37 mohawk nice
22:38 mohawk i will revisit this question in due course :-)
22:38 Mithaldu at the end of the day you could probably also fork the gui
22:38 mohawk the MI gui?
22:38 Mithaldu i've been trying to separate things nicely, if only because the network stuff necessitates it
22:41 mohawk nice

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