00:00:00 --- log: started retro/12.11.13
02:12:22 --- quit: ChanServ (*.net *.split)
02:12:22 --- quit: SimonRC (*.net *.split)
02:12:22 --- quit: crc (*.net *.split)
02:16:03 --- join: SimonRC (~sc@fof.durge.org) joined #retro
02:16:03 --- join: crc (~crc@li125-93.members.linode.com) joined #retro
02:16:03 --- join: ChanServ (ChanServ@services.) joined #retro
02:16:03 --- mode: asimov.freenode.net set +o ChanServ
02:20:57 --- quit: ChanServ (*.net *.split)
02:20:57 --- quit: SimonRC (*.net *.split)
02:20:57 --- quit: crc (*.net *.split)
02:20:57 --- quit: docl (*.net *.split)
02:20:58 --- quit: yiyus_ (*.net *.split)
02:21:00 --- quit: ivan`` (*.net *.split)
02:21:01 --- quit: harrison (*.net *.split)
02:21:03 --- quit: saper (*.net *.split)
02:22:17 --- join: ivan`` (~ivan@unaffiliated/ivan/x-000001) joined #retro
02:22:17 --- join: yiyus_ (1242712427@je.je.je) joined #retro
02:22:17 --- join: saper (saper@wikipedia/saper) joined #retro
02:22:17 --- join: harrison (~quassel@li89-226.members.linode.com) joined #retro
02:22:17 --- join: docl (~docl@unaffiliated/docl) joined #retro
02:22:17 --- join: SimonRC (~sc@fof.durge.org) joined #retro
02:22:17 --- join: crc (~crc@li125-93.members.linode.com) joined #retro
02:22:17 --- join: ChanServ (ChanServ@services.) joined #retro
02:22:17 --- mode: asimov.freenode.net set +o ChanServ
02:27:01 --- quit: tangentstorm (*.net *.split)
02:27:01 --- quit: ncv (*.net *.split)
02:27:01 --- quit: oPless (*.net *.split)
02:27:02 --- quit: docl (*.net *.split)
02:27:02 --- quit: yiyus_ (*.net *.split)
02:27:04 --- quit: ivan`` (*.net *.split)
02:27:06 --- quit: harrison (*.net *.split)
02:27:07 --- quit: saper (*.net *.split)
02:27:08 --- quit: ChanServ (*.net *.split)
02:27:08 --- quit: SimonRC (*.net *.split)
02:27:08 --- quit: crc (*.net *.split)
02:27:35 --- join: oPless (~oPless@lart.doosh.net) joined #retro
02:27:35 --- join: ncv (~quassel@unaffiliated/neceve) joined #retro
02:27:35 --- join: tangentstorm (~michal@108-218-151-22.lightspeed.rcsntx.sbcglobal.net) joined #retro
02:27:35 --- join: ivan`` (~ivan@unaffiliated/ivan/x-000001) joined #retro
02:27:35 --- join: yiyus_ (1242712427@je.je.je) joined #retro
02:27:35 --- join: saper (saper@wikipedia/saper) joined #retro
02:27:35 --- join: harrison (~quassel@li89-226.members.linode.com) joined #retro
02:27:35 --- join: docl (~docl@unaffiliated/docl) joined #retro
02:27:35 --- join: SimonRC (~sc@fof.durge.org) joined #retro
02:27:35 --- join: crc (~crc@li125-93.members.linode.com) joined #retro
02:27:35 --- join: ChanServ (ChanServ@services.) joined #retro
02:27:35 --- mode: asimov.freenode.net set +o ChanServ
04:31:25 --- quit: ncv (Ping timeout: 240 seconds)
06:55:14 <crc> lp:~crc-x/retro-language/keymaps has the image w/partially implemented keymaps as part of kernel.rx
07:16:31 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
10:43:15 --- join: Mat2 (~claude@91-65-144-133-dynip.superkabel.de) joined #retro
10:44:55 <Mat2> god dag
10:49:55 <Mat2> tangentstorm: I have added keyboard scanning to romVDP and fixed some display bugs
10:51:28 <Mat2> you should now be able to try it out (even remoted under Windows)
11:28:44 <Mat2> fixed control character handling
11:37:11 --- nick: Mat2 -> Mat2-work
11:49:57 --- join: foucist (~foucist@ps14150.dreamhost.com) joined #retro
11:49:57 --- mode: ChanServ set +v foucist
11:51:31 <foucist> crc: how's the speed of retro on JVM, taking advantage of hotspot compiler, invokedynamic, etc compared to the regular implementation?
11:52:11 <Mat2-work> hi foucist
11:53:31 <foucist> hi Mat2-work 
11:54:06 <Mat2-work> how you doin ?
11:54:37 <foucist> decent!
11:54:57 <foucist> working on a ruby on rails website.. and looking for work
11:59:25 <Mat2-work> I'm working at current on an SDL based interface for retro
11:59:38 <Mat2-work> in my free time
12:01:32 --- nick: Mat2-work -> Mat2
12:02:02 <foucist> ah, i'm not much of a forther at all
12:02:06 <foucist> these past few years
12:02:27 <foucist> but i still like to see what crc is up to :P
12:02:45 <Mat2> I think you program mainly in ruby, right ?
12:02:50 <foucist> yeah
12:03:22 <Mat2> tangentstorm will bring a pascal compiler to ngaro
12:03:51 <Mat2> and I'm thinking about lua
12:04:38 <foucist> speaking of lua, have you seen luakit?
12:04:46 <foucist> webkit browser with lua
12:05:03 <foucist> how about a retrokit heh
12:05:08 <Mat2> no, but this was only a matter of time
12:05:53 <Mat2> this is on my list
12:07:49 <Mat2> I want to explore the advantages and drawbacks of retro for such applications
12:08:13 <Mat2> but before that, I need to finish my own vm design
12:13:50 <tangentstorm> Mat2 : lua, eh? :)
12:14:05 <Mat2> hi tangendstorm
12:14:15 <Mat2> I have a request
12:14:17 <foucist> imo, i want to combine the best of forth/retro & ruby
12:14:31 <foucist> bring in the ruby-style readability & OO 
12:14:32 <tangentstorm> Back in 2003, I was porting python to the parrot vm... About the same time there was a guy porting lua to it.
12:14:53 <tangentstorm> Mat2: okay... I'm listening :)
12:15:02 <tangentstorm> ( hi foucist )
12:15:15 <foucist> i'm half convinced that a ruby-style syntax could be easily built ontop of forth
12:15:25 <Mat2> it can for sure
12:15:42 <foucist> altho might have to drop the postfix somewhat hmm
12:15:57 <foucist> not sure
12:16:39 <tangentstorm> foucist: i've been transcribing a paper about a pascal parser/compiler built in forth
12:17:05 <Mat2> tangentstorm: I need you for some testing, can you compile the new sources and tell me if keyboard handling is working ?
12:17:27 <tangentstorm> Mat2: you didn't even have to ask! I'm doing that now!
12:18:17 <Mat2> ok
12:19:43 <tangentstorm> Mat2: I've moved xpc, stacks.pas, and a bunch of other general-purpose pascal stuff to a new repository
12:19:56 <tangentstorm> https://github.com/tangentstorm/xpl
12:20:07 <tangentstorm> there's a kind of neat test framework under the test directory
12:21:00 <tangentstorm> ( so you can add it as a git submodule if you want, since it's required for compiling ngaro )
12:23:12 <Mat2> some more structure is always nice
12:24:21 <tangentstorm> I also renamed the ng.*.pas files in the b4 directory the other day, and the {$includes} in this code are out of sync
12:24:25 <tangentstorm> i'll send you a patch
12:24:31 <Mat2> thanks
12:25:48 <tangentstorm> ok! got it to compile... firing it up in 3... 2... 1...
12:26:20 <tangentstorm> HRMF.
12:26:38 <Mat2> ok, don't seem to work
12:26:47 <tangentstorm> a) the typing works!!! :)
12:27:01 <Mat2> that is good
12:27:23 <tangentstorm> b) i use a dvorak keyboard layout and the letters i type aren't the ones that show up on the screen :/
12:27:50 <tangentstorm> but... that's probably something the new keymapping stuff can fix
12:29:13 <tangentstorm> oddly, when i press some letters, i consistently get two keystrokes, but not others
12:29:27 * tangentstorm turns off dvorak setting
12:30:24 <Mat2> what letter is mapped to the z character ?
12:31:17 <tangentstorm> dvorak z is qwerty /
12:31:26 <tangentstorm> qwerty z is dvorak ;
12:31:41 <tangentstorm> it's not specific letters, it's my typing speed
12:31:41 <Mat2> ahhhh !
12:31:58 <tangentstorm> no.. hrm. well, maybe it's two issues
12:32:04 <tangentstorm> i'm trying to type "clear"
12:32:34 <Mat2> this should result in a core dump
12:32:35 <tangentstorm> typing slow, i get "cleear"  ( i turned of dvorak.. this is straight qwerty now )
12:32:57 <tangentstorm> typing fast, i get   cllr   pretty consistently
12:33:06 <tangentstorm> or cler
12:33:13 <tangentstorm> like it's dropping letters if i type too fast
12:33:26 <Mat2> ok, the repeat rate is to high
12:33:28 <tangentstorm> it is an old machine. windows xp
12:33:56 <Mat2> probably 256 ns should work better
12:34:23 <tangentstorm> other than that it does seem to be working. gonna try color now
12:34:39 <tangentstorm> Mat2: wait. where would i change that setting in the code?
12:36:01 <tangentstorm> found it. keyDelay in ng.pas
12:36:05 <Mat2> yes
12:36:23 <tangentstorm> just change the 100 to 256?
12:36:35 <Mat2> try 125 first
12:36:50 <Mat2> that are clocks not time jitters
12:36:55 <tangentstorm> WAY too slow
12:37:05 <tangentstorm> but it doesn't repeat :)
12:37:24 <Mat2> ok, you must find the best fit
12:38:19 <Mat2> sadly the SDL library do not support realtime services
12:38:37 <Mat2> or clock based timers
12:40:53 <Mat2> 130 works fine for my system
12:42:32 <tangentstorm> hrm. so nudging it up to 120-130 solves the repeat issue, but it still drops letters. 
12:42:58 <tangentstorm> i type pretty fast, and i can definitely notice the lag in when the letters appear...
12:44:00 <tangentstorm> if i slow down, everything works fine, but it's consistently dropping letters when i type fast
12:44:03 <tangentstorm> also no backspace
12:44:16 <Mat2> try to input words, if the display update fast or slow ?
12:44:40 <Mat2> ^is
12:44:47 <tangentstorm> hrm. actually the delete key seems to move the cursor backwards, but it doesn't delete the letters
12:45:12 <tangentstorm> you mean "words" to show the word list? it's very very fast
12:45:42 <tangentstorm> it's slow taking the letters in while i type, but once i hit space, the word list shows up very quickly.
12:46:15 <tangentstorm> ( i was using that as a benchmark on the javascript version ... i love the speed on the display :) )
12:46:32 <tangentstorm> ( i mean on the pascal display. the javascript one is still to slow! )
12:46:43 <Mat2> ah, I seem to forgot uploading the last sources, stand by
12:48:03 <Mat2> hmm, there are up to date, backspace should work
12:48:18 <Mat2> (it works here)
12:50:16 <tangentstorm> oh, okay.. backspace works here too, but it doesn't clear the letters
12:50:38 <tangentstorm> it clears them from the buffer as far as retro is concerned but not from the screen 
12:52:22 <Mat2> can you please upload ng.ports.pas and compile again ? I think you have an old file version
12:52:47 <Mat2> because that was a bug I recently fixed
12:53:27 <Mat2> or assembla is out of sync, let's see
12:57:57 <Mat2> nice, the SDL library update the keyboard buffer asynchron, so at fast typing rates characters get lost *arggg*
12:59:14 <Mat2> ok, so the only valid method is the brute force one, refresh the queue after every keystroke
13:00:38 --- quit: karswell ()
13:03:04 <tangentstorm> https://www.assembla.com/code/avm/git/reviews#open
13:06:10 <tangentstorm> "git pull mat2 master" says i'm up to date ( mat2 = git://git.assembla.com/avm.git )
13:07:09 <tangentstorm> hmm. somehow romFont is missing
13:07:33 <Mat2> then the returned backspace code is wrong
13:08:18 <tangentstorm> could be
13:08:49 <tangentstorm> i don't know what happened. romFont.pas just disappeared on me... sometimes i don't understand what git is doing at all
13:09:22 <tangentstorm> it's gone in your assembla too https://www.assembla.com/code/avm/git/nodes/master/metro/backport
13:09:47 <Mat2> yeah, that was a sync problem
13:11:33 <tangentstorm> can i make a suggestion on the coding style?
13:11:37 <Mat2> it's now insode the repro
13:11:42 <Mat2> yes of course
13:12:49 <tangentstorm> i was going to suggest changing romvdp to just vdp... and if you have different versions, put them in different directories, but call them both vdp.pas
13:13:09 <tangentstorm> then you can "use vdp" and you don't have to put a "vdp" prefix on all your functions
13:13:16 <tangentstorm> because you can say "vdp.xxxx" if it's not clear
13:15:42 <tangentstorm> also.. for functions... instead of saying "FUNCTION vdpPollKeyboard; vdPollKeyboard := xxxx"  ...
13:16:04 <tangentstorm> you can now say "result := xxx;"
13:16:30 <tangentstorm> that works with pretty much all compilers that i'm aware of now.
13:17:07 <Mat2> ah ok
13:17:32 <tangentstorm> ( the extended iso standard is:   function whatever : sometype = varname;    I think... but i've never actually seen anyone use it )
13:19:40 <Mat2> probably that is Ada style
13:21:43 <tangentstorm> could be :) there are two ISO standards for pascal... a lot of people kind of ignore the second one, but that's what gnu pascal targets
13:22:18 <tangentstorm> ( but i think gnu pascal is kind of abandoned now anyway... fpc is where it's at :) )
13:23:15 <Mat2> for vector processing Ada is a good choice
13:23:43 <Mat2> yes, fpc is the open source borland compiler
13:25:30 <Mat2> ok, merged your changes, thanks
13:26:27 <tangentstorm> np
13:31:53 <Mat2> ok, characters dont get lost now (have tried to write as fast as possible)
13:32:03 <tangentstorm> :)
13:33:41 <tangentstorm> did you push to assembla?
13:34:00 <Mat2> this moment, uno momento
13:35:10 <Mat2> I must check up your merge first
13:37:37 <tangentstorm> uh-oh. did i break something?
13:40:31 <Mat2> no, but I have now two repros instead of one (do not ask me why, I do not know ?!??)
13:44:31 <tangentstorm> ??? weird
13:48:12 <Mat2> CONFLICT (content): Merge conflict in metro/backport/ng.pas
13:48:12 <Mat2> Automatic merge failed; fix conflicts and then commit the result.
13:51:19 <Mat2> ok upload complete
13:51:45 <tangentstorm> ugh. sorry.
13:51:50 <Mat2> np
13:52:44 <Mat2> that's the git mess of life (otherwise from that strange specialities GIT works fine)
13:52:45 <tangentstorm> not compiling for me
13:53:10 <tangentstorm> ng.pas 127,1  illegal expression... plus a few others
13:53:45 <tangentstorm> oh i see
13:54:07 <Mat2> I do not have changed ng.pas
13:54:09 <tangentstorm> sorry, i didn't mean to commit that change
13:54:15 <tangentstorm> it's the merge conflict
13:54:21 <tangentstorm> from the key delay
13:54:27 <tangentstorm> it's just one line. i'll fix it
13:54:39 <tangentstorm> well. what did you have it set to? 130?
13:56:28 <Mat2> no 80
13:59:47 <tangentstorm> oh. ok. i had set it to 130 because of the repeats. :)
14:00:09 <Mat2> i think 130 is a good value
14:01:21 <Mat2> no for me it must be a value unter 100
14:01:58 <tangentstorm> i'm going to try something... back in a few minutes
14:02:00 <Mat2> 90 is fine
14:02:04 <Mat2> or 85
14:37:29 <Mat2> tangentstorm: the sources are updated, there exist a new variable - keySbDelay for subdelay timing
14:37:57 <Mat2> next problem are the different keymappings
14:38:14 <Mat2> I need support for Dvorak and other layouts
14:42:12 <Mat2> crc: you should test the pascal vm because of SDL terminal support
14:44:20 <tangentstorm> Mat2: uploading a patch. it's not perfect, but it gets rid of the delay, and it seems to work faster
14:44:48 <tangentstorm> i'm just using the PollEvent() from sdl instead of PumpEvents()
14:48:49 <tangentstorm> Mat2: merge conflict... but you can look at what i did here and see if it works for you: https://www.assembla.com/code/michalwallace-avm-fork/git/nodes/master/metro/backport/romVDP.pas#ln275
14:51:34 <tangentstorm> I wonder if we can grab the keymappings out of... FreeBSD maybe?
14:53:02 <tangentstorm> or maybe sdl can just get the os-translated ones somehow?
14:56:02 <Mat2> does your version work better ?
14:56:14 <tangentstorm> Mat2: for me on this machine
14:56:28 <tangentstorm> it types much faster, with no repeats or dropped letters
14:56:57 <Mat2> then lets use this version
14:57:04 <tangentstorm> http://wiki.libsdl.org/moin.cgi/CategoryKeyboard  <- this seems to say that sdl supports keyboard layouts already. maybe only in sdl 2.0 ??
14:57:37 <tangentstorm> :) i'll send a pull request
14:58:27 <tangentstorm> sent
15:01:57 <Mat2> translation only work for alphanumeric and letter characters
15:02:00 <Mat2> sadly
15:03:26 <Mat2> I don't kbnow if keyboard handling in SDL 2 get an usable level
15:06:17 <tangentstorm> i don't think the pascal libraries will work with sdl2 anyway
15:07:48 <tangentstorm> i really don't like the pascal wrapper. it works but everything still feels like C... i might try to clean it up when i have time
15:07:59 <tangentstorm> make it more like pygame ( sdl wrapper for python )
15:08:30 <Mat2> it do not use pascal's capabiities well, yes
15:09:58 --- join: kumul (~kumul@173.215.130.73) joined #retro
15:11:21 <Mat2> no keyboard repeat but it works, thanks
15:11:44 <Mat2> I will left the old routine for key repeat feature
15:12:12 <Mat2> vdpPollKeyboardRepeat
15:15:45 <Mat2> I had to choose Allegro instead of SDL, its keyboard support is much better
15:18:47 --- join: Z_Mass (~quassel@bas1-cornwall24-1242467096.dsl.bell.ca) joined #retro
15:23:38 <Mat2> damn GIT
15:32:26 <Mat2> ok, have merged your changes
15:32:58 <Mat2> you should pull the new version
15:33:22 <Mat2> or we will get in merge trouble again
15:38:21 <Mat2> fine, the SDL routines do not even handle the control and alt keys
15:39:29 <tangentstorm> ??
15:39:47 <tangentstorm> there should be a keydown event when you press it
15:39:50 <tangentstorm> and keyup
15:40:14 <tangentstorm> was afk taking out the trash
15:41:35 <Mat2> no recognition at all with these keys
15:42:40 <Mat2> with both routines
15:46:19 <Mat2> your routine gives me '0' if I press left shift twice and '/' on right shift
15:46:26 <Mat2> oh, I see the problem
15:50:47 <tangentstorm> http://www710.univ-lyon1.fr/~jciehl/Public/SDL_PG/sdlenablekeyrepeat.html
15:51:08 <tangentstorm> ^ SDL_EnableKeyRepeat might be what we need for part of this
15:52:41 <Mat2> my problem is '0' is printed if I press the left shift key, can you try this out ?
15:53:15 <tangentstorm> .. it could be my chr( ) call
15:53:39 <tangentstorm> ah yep
15:54:32 <tangentstorm> let me go to my other computer. i don't have a decent text editor on this one.. back in a few.
15:56:18 <Mat2> ok
15:59:44 <tangentstorm> started using that old laptop because it would run turbo pascal... now i just tend to use it when i'm in the living room :)
16:00:50 <Mat2> so your old laptop is of use, that's good :)
16:11:48 <tangentstorm> it's not even my laptop! I borrowed it from my mom. So I try not to install to much on it...
16:11:59 <tangentstorm> anyway. got this checked out and trying to compile again
16:14:35 <tangentstorm> argh. my makefile had the wrong git url
16:14:50 <Mat2> oh no
16:22:29 --- quit: Z_Mass (Remote host closed the connection)
16:30:07 <Mat2> I think going gradually to tired for troobleshooting
16:30:44 <Mat2> let us finish research tomorrow
16:31:48 <Mat2> I go to bed
16:31:51 <Mat2> ciao
16:31:56 --- quit: Mat2 (Quit: Verlassend)
16:52:50 <crc> multiple keymaps are working now, so once it gets cleaned up a bit it should be ready to merge
17:01:44 <tangentstorm> crc: nice! i got os-mapped inputs working in sdl now
17:02:09 <tangentstorm> which looks like it can provide full unicode input
22:24:22 --- quit: kumul (Quit: WeeChat 0.3.9)
22:37:35 --- join: Z_Mass (~quassel@bas1-cornwall24-1242467096.dsl.bell.ca) joined #retro
23:59:59 --- log: ended retro/12.11.13