**** BEGIN LOGGING AT Sat Jan 28 03:00:00 2012
Jan 28 03:11:55 <ka6sox> PuffTheMagic, 0.4.0? thats been there for a while
Jan 28 03:12:46 <PuffTheMagic> ka6sox, its been updated
Jan 28 03:13:04 <PuffTheMagic> do you see the V4
Jan 28 03:13:17 <PuffTheMagic> that is under the filename to the left of the date
Jan 28 03:14:02 <ka6sox> no
Jan 28 03:14:06 <ka6sox> I do no
Jan 28 03:14:07 <ka6sox> t
Jan 28 03:14:20 <PuffTheMagic> maybe hit refresh
Jan 28 03:14:23 <stbuehler> PuffTheMagic: ready for a review and pull? :)
Jan 28 03:15:00 <PuffTheMagic> stbuehler, what you've done is beyond me, if merges fine I can run it
Jan 28 03:15:13 <PuffTheMagic> dtz is prob the better reviewer for the GL stuff
Jan 28 03:15:24 <ka6sox> PuffTheM`, 0.4.0
Jan 28 03:15:34 <stbuehler> well, you could test it before you merge it upstream :)
Jan 28 03:15:34 <ka6sox> thats what I've been seeing for the last 3hrs
Jan 28 03:16:49 <PuffTheMagic> ka6sox, are you in the wterm folder or just refreshing the link
Jan 28 03:17:04 <PuffTheMagic> ka6sox, http://ompldr.org/vY2h0eA/wterm-box.png
Jan 28 03:23:14 <ka6sox> I go out and in
Jan 28 03:23:35 <ka6sox> I am using the Box app on the TP
Jan 28 03:23:38 <ka6sox> okay nm
Jan 28 03:25:28 <stbuehler> gn8 :) and have fun with shaders :D
Jan 28 03:29:17 <PuffTheMagic> ka6sox, ahh, never really used the app
Jan 28 04:44:34 <PuffTheMagic> stbuehler, i think wterm is no longer based on SDLTerminal any more nearly every part of it has been rewritten now
Jan 28 04:44:46 <PuffTheMagic> time to update/trim copyright headers
Jan 28 05:12:19 <ka6sox> PuffTheM`, lemme know when the dangling participle is gone :D
Jan 28 05:12:40 <PuffTheMagic> dangling participle?
Jan 28 05:12:45 <PuffTheMagic> i cant kill that user
Jan 28 05:12:49 <PuffTheMagic> its not registered to me
Jan 28 05:13:06 <PuffTheMagic> and its somehow logged in at my office
Jan 28 05:33:43 <ka6sox> no I mean your main app icon
Jan 28 05:36:17 <PuffTheMagic> ka6sox, i can get rid of the launchpoints
Jan 28 05:36:22 <PuffTheMagic> and then that issue will go away
Jan 28 05:36:41 <ka6sox> what does that do for my root session?
Jan 28 05:37:24 <PuffTheMagic> you can still get a root session other ways
Jan 28 05:37:41 <PuffTheMagic> or you can reboot luna
Jan 28 05:38:21 <ka6sox> ooooookay....
Jan 28 05:38:35 <PuffTheMagic> its a bug with webos/luna
Jan 28 05:38:39 <PuffTheMagic> idk what you want me to say
Jan 28 05:38:45 <ka6sox> yup yup
Jan 28 05:38:47 <ka6sox> it happens
Jan 28 05:39:03 <PuffTheMagic> i can either remove the launchpoints or deal with the stray icon
Jan 28 05:39:22 <ka6sox> gotta love those kinds of bugs
Jan 28 05:39:30 <PuffTheMagic> i tried creating a special launch arg that deletes the launchpoints
Jan 28 05:39:34 <PuffTheMagic> and it worked with luna-send
Jan 28 05:39:38 <PuffTheMagic> via novaterm
Jan 28 05:39:51 <PuffTheMagic> but when i added it to the prerm script, uninstall would stall
Jan 28 12:52:03 <stbuehler> PuffTheMagic: i think there is a large performance hit for using pdk plugin
Jan 28 12:53:20 <stbuehler> i just converted it to "pdk" and started cmatrix - pretty low cpu load (cmatrix 15%, wterm 12%, nothing else)
Jan 28 12:55:15 <stbuehler> exhibition mode works too :)
Jan 28 13:26:19 <cryptk> stbuehler, I must be reading that wrong
Jan 28 13:26:32 <cryptk> you said there is a large performance hit for using a PDK plugin
Jan 28 13:26:43 <cryptk> then when you said you set it up as a PDK app the CPU usage was low
Jan 28 13:26:49 <cryptk> that's the opposite of a hit to me
Jan 28 13:50:48 <stbuehler> penalty, sry
Jan 28 13:59:52 <cryptk> either way, what you are saying is that the performance go's down for using PDK
Jan 28 14:00:06 <cryptk> but then you said that you set it up as PDK and it had low CPU load
Jan 28 14:00:18 <cryptk> those two statements are opposite of each other
Jan 28 14:00:31 <cryptk> "it performes worse" "look at the low CPU load"
Jan 28 14:11:06 <stbuehler> no
Jan 28 14:11:25 <stbuehler> i said it goes down for using PDK *plugin*
Jan 28 14:12:18 <stbuehler> so the overhead for plugin mode is pretty high, and it looks to me like it is copying the data over the main memory (instead of blitting with gpu)
Jan 28 14:12:58 <stbuehler> i probably should do some real performance measurements
Jan 28 16:45:14 <PuffTheMagic> stbuehler, even if there is a performance difference there is nothing we can really do about it
Jan 28 16:46:54 <PuffTheMagic> stbuehler, i dont understand what your comment was about for dtz bgcolor fix
Jan 28 16:54:06 <PuffTheMagic> stbuehler, please note what your cpu is at when doing your benchmarks
Jan 28 16:54:11 <PuffTheMagic> and if you are overclocked
Jan 28 16:55:44 <PuffTheMagic> stbuehler, because for me, cmatrix running as plugin... the cpu stays at 192000
Jan 28 16:55:52 <PuffTheMagic> and sometimes goes up to 384000
Jan 28 16:56:07 <PuffTheMagic> which isnt bad at all
Jan 28 21:15:12 <stbuehler> PuffTheMagic: how can i see the cpufreq in novaterm?
Jan 28 21:16:22 <PuffTheMagic> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq
Jan 28 21:16:26 <PuffTheMagic> cat /sys/devices/system/cpu/cpu1/cpufreq/scaling_cur_freq
Jan 28 21:17:37 <PuffTheMagic> stbuehler, menu 8 (the vt102 tests) in vttest are broken on HEAD
Jan 28 21:19:46 <stbuehler> i'll have a look
Jan 28 21:20:06 <stbuehler> and you are right, cpu freq goes up with "native" mode, and stays low with plugin mode
Jan 28 21:20:39 <stbuehler> oh, and i have 3 "wterm setup" items now
Jan 28 21:21:19 <PuffTheMagic> ya pulled the code to add launchpoints
Jan 28 21:21:37 <PuffTheMagic> hp/palm doesnt test shit ;)
Jan 28 21:24:12 <PuffTheMagic> stbuehler, the gles2 changes make things wicked smooth
Jan 28 21:24:34 <PuffTheMagic> too bad vttest now fail ;)
Jan 28 21:26:25 <stbuehler> can you tell me what failed, because i didn't see anything wrong?
Jan 28 21:27:10 <PuffTheMagic> weird its working again
Jan 28 21:29:29 <stbuehler> btw: would be nice if you used an editor that visualizes tabs and spaces (or that cleans them up for you) :)
Jan 28 21:30:04 <PuffTheMagic> ya something about aptana is driving me nuts with inserting tabs and shit, it never did that in 2.x but in 3.x it seems to be :/
Jan 28 21:30:06 <PuffTheMagic> i need to fix that
Jan 28 21:30:46 <stbuehler> and you said you need the OSC parsing
Jan 28 21:31:07 <PuffTheMagic> ya I have 2 uses for that text, it would be nice to get it somehow
Jan 28 21:43:26 <PuffTheMagic> stbuehler, btw, thanks for the wiki updates, i linked the new pages to the home page
Jan 28 21:44:16 <stbuehler> well, i need the links myself :)
Jan 28 21:44:22 <PuffTheMagic> no it was a good idea
Jan 28 21:44:29 <stbuehler> btw, can't we hide the setup thing when everything is good?
Jan 28 21:44:39 <stbuehler> i hate clicking it away for each test :P
Jan 28 21:45:06 <PuffTheMagic> well i plan on using it also show changelogs on updates
Jan 28 21:45:39 <PuffTheMagic> but i will try and think of a developer friendly solution
Jan 28 21:45:39 <stbuehler> well, you could check whether the version was actually increased :)
Jan 28 21:45:52 <PuffTheMagic> no i cant
Jan 28 21:46:04 <PuffTheMagic> well depends on if the cookie exists
Jan 28 21:47:09 <PuffTheMagic> i have an idea
Jan 28 21:47:19 <PuffTheMagic> the autobuilder doesnt use palm-package
Jan 28 21:47:33 <PuffTheMagic> so I have have a different post inst for it
Jan 28 21:47:45 <PuffTheMagic> and in the one used by palm-package it wont do the firstUse reset
Jan 28 21:55:00 <PuffTheMagic> stbuehler, last commit fixes it
Jan 28 21:55:54 <PuffTheMagic> stbuehler, so i was thinking about what you were saying about build.git and stub libs + headers
Jan 28 21:56:25 <PuffTheMagic> what if we took a webos doctor and extracted its rootfs
Jan 28 21:56:49 <PuffTheMagic> how hard do you think it would be to walk all the files/dirs in /lib and generate stubs and working headers for everything
Jan 28 21:57:15 <stbuehler> well, you need headers from sdk obviously
Jan 28 21:57:32 <stbuehler> creating stub libs should be easy
Jan 28 21:57:36 <PuffTheMagic> why would u need headers from sdk
Jan 28 21:57:43 <PuffTheMagic> i thought u could generate them
Jan 28 21:57:46 <stbuehler> no
Jan 28 21:58:00 <stbuehler> c libs don't contain any type information
Jan 28 21:58:23 <PuffTheMagic> so you couldnt get SDL to build
Jan 28 21:58:30 <PuffTheMagic> or pulse
Jan 28 21:58:42 <stbuehler> i can show you my sdl and glesv2 stuff
Jan 28 21:58:56 <PuffTheMagic> you didnt use my glesv2 stuff either?
Jan 28 21:59:52 <PuffTheMagic> stbuehler, im gonna try and get build.git moved to github so that you can just pull request that
Jan 28 21:59:53 <stbuehler> i think yours would work
Jan 28 21:59:59 <PuffTheMagic> not having to build all of sdl would be nice
Jan 28 22:00:04 <stbuehler> looked good, i just wanted to test my own method :)
Jan 28 22:00:33 <stbuehler> i think we should only provide the libs from the sdk that way
Jan 28 22:01:05 <stbuehler> other libs from webos shouldn't be used, as hp may break compat with them any time
Jan 28 22:06:16 <stbuehler> PuffTheMagic: do you think we should merge VTTerminalState into TerminalState ?
Jan 28 22:10:53 <PuffTheMagic> probably a good idea
Jan 28 22:11:04 <PuffTheMagic> seems sorta redundant
Jan 28 22:12:27 <PuffTheMagic> stbuehler, what did u do to colors.png
Jan 28 22:15:52 <stbuehler> ups
Jan 28 22:16:09 <stbuehler> damn, didn't notice
Jan 28 22:16:15 <PuffTheMagic> i fixed it
Jan 28 22:16:35 <stbuehler> i run sed on all files to fix whitespace
Jan 28 22:17:00 <stbuehler> sed doesn't recognize the binaries
Jan 28 22:17:18 <stbuehler> just don't break the whitespace again :P (sry, my fault ofc)
Jan 28 22:23:55 <PuffTheMagic> stbuehler, so im thinking we should implement the xterm escape code to load fonts, and use that to load the matrix font into a slot
Jan 28 22:24:19 <PuffTheMagic> we can then change cmatrix to load the font via escape code instead of setfont
Jan 28 22:24:45 <stbuehler> is there a special page that documents this stuff?
Jan 28 22:25:19 <PuffTheMagic> its an OSC command
Jan 28 22:25:23 <PuffTheMagic> http://rtfm.etla.org/xterm/ctlseq.html
Jan 28 22:26:09 <stbuehler> well
Jan 28 22:26:10 <PuffTheMagic> vttest also has some xterm font loading tests
Jan 28 22:26:13 <stbuehler> this selects fonts
Jan 28 22:26:23 <stbuehler> but where is the actual data, what is the format, ...
Jan 28 22:26:39 <PuffTheMagic> well consoles typically use pcf conts
Jan 28 22:26:55 <PuffTheMagic> but im sure we could support ttf instead
Jan 28 22:27:07 <PuffTheMagic> consoles as in non-xterm uses pcf
Jan 28 22:27:46 <stbuehler> so you think this command accepts a filename?
Jan 28 22:28:00 <stbuehler> like ESC]50;fontfile.pcfBEL
Jan 28 22:28:02 <PuffTheMagic> maybe not :(
Jan 28 22:28:17 <PuffTheMagic> but maybe
Jan 28 22:30:04 <PuffTheMagic> https://github.com/trapd00r/pickyfont/issues/3
Jan 28 22:30:21 <PuffTheMagic> https://github.com/trapd00r/pickyfont
Jan 28 22:30:36 <PuffTheMagic> pickyfont lets you change your X-terminal font on the fly, just like
Jan 28 22:30:36 <PuffTheMagic>     setfont can do in a TTY.
Jan 28 22:30:36 <PuffTheMagic>     Normal, bold and italic fontfaces are supported, as well as XFT fonts.
Jan 28 22:31:04 <stbuehler> to me it looks like you specify a font name
Jan 28 22:31:17 <stbuehler> and then use X stuff to get the font
Jan 28 22:31:31 <PuffTheMagic> we cant mimic something like that?
Jan 28 22:31:42 <PuffTheMagic> or is that too much
Jan 28 22:31:44 <stbuehler> well
Jan 28 22:32:00 <stbuehler> you can ship multiple fonts, and map font names to file names
Jan 28 22:32:39 <PuffTheMagic> yes we should ship fonts... but we could also check normal font locations/paths on the system
Jan 28 22:32:47 <stbuehler> hm right
Jan 28 22:46:38 <stbuehler> how am i supposed to scroll up/down?
Jan 28 22:46:47 <PuffTheMagic> with vkb?
Jan 28 22:47:01 <PuffTheMagic> shift+pgup/pgdn
Jan 28 22:47:02 <stbuehler> yes
Jan 28 22:47:14 <PuffTheMagic> but since we dont have direct access to shift u need to add fn in there
Jan 28 22:47:46 <stbuehler> doesn't do anything for me
Jan 28 22:48:01 <stbuehler> apart from keys getting stuck
Jan 28 22:48:39 <PuffTheMagic> prob need to change it to ctrl+pgup/pgdn
Jan 28 22:48:45 <PuffTheMagic> err
Jan 28 22:48:50 <PuffTheMagic> up/dn i mean
Jan 28 22:49:15 <stbuehler> so you didn't test it with vkb, right? :)
Jan 28 22:49:21 <PuffTheMagic> i did
Jan 28 22:49:25 <PuffTheMagic> it works
Jan 28 22:49:30 <stbuehler> hm
Jan 28 22:49:33 <PuffTheMagic> i conformed to standard binding
Jan 28 22:49:35 <stbuehler> i even changed back to default layout
Jan 28 22:49:38 <PuffTheMagic> can it be improved... yes
Jan 28 22:50:09 <stbuehler> i hold fn and shift and press up - nothing happens
Jan 28 22:51:07 <PuffTheMagic> well it def is working here, i will have to look into this after i get back from dinner
Jan 28 22:51:12 <stbuehler> kk
Jan 28 22:51:31 <stbuehler> ah now it works
Jan 28 22:51:57 <stbuehler> doesn't work after i started mc and quit again
Jan 28 22:52:15 <stbuehler> and i think i know why
Jan 28 22:52:25 <PuffTheMagic> bbl
Jan 29 00:21:32 <stbuehler> hm, we have some conflicting esc sequences
Jan 29 01:36:01 <PuffTheMagic> like?
Jan 29 01:40:18 <stbuehler> ESCM
Jan 29 01:40:37 <stbuehler> ESCD, ESCH
Jan 29 01:56:11 <PuffTheMagic> i forget what those do
Jan 29 01:56:41 <stbuehler> well, different things :)
Jan 29 01:56:47 <stbuehler> depending on some mode i guess
Jan 29 01:56:51 <stbuehler> vt52, ...
Jan 29 01:56:56 <PuffTheMagic> i was just gonna say
Jan 29 01:57:03 <stbuehler> i still get endless "wTerm setup" entries
Jan 29 01:57:05 <PuffTheMagic> sounds like a vt52 mode error
Jan 29 01:57:17 <PuffTheMagic> have you pulled
Jan 29 01:57:21 <PuffTheMagic> try deleting the app
Jan 29 01:57:23 <stbuehler> and tp crashes from time to time
Jan 29 01:57:24 <PuffTheMagic> kill luna
Jan 29 01:57:30 <PuffTheMagic> then reinstall
Jan 29 01:57:39 <PuffTheMagic> stbuehler, i dont get crashes though
Jan 29 01:57:43 <stbuehler> well, it just restarted
Jan 29 01:58:43 <stbuehler> and jail config is broken again (it tries to jail wterm)
Jan 29 01:59:06 <PuffTheMagic> i dont get how you cause that
Jan 29 01:59:13 <PuffTheMagic> are you not using palm-install?
Jan 29 01:59:26 <stbuehler> using palm-install and palm-launch
Jan 29 01:59:47 <stbuehler> now i have to delete wterm again, restart, and hope it works
Jan 29 02:10:13 <Brybry> I remember when one time my jailer somehow deleted itself (when I was doing bad things)
Jan 29 02:27:27 <PuffTheMagic> what is wrong with this
Jan 29 02:27:27 <PuffTheMagic> void TerminalState::handle_osc(int value, const char *txt)
Jan 29 02:27:28 <PuffTheMagic> {
Jan 29 02:27:28 <PuffTheMagic> 	char *val = 0;
Jan 29 02:27:28 <PuffTheMagic> 	asprintf(&val, "%d", value);
Jan 29 02:27:28 <PuffTheMagic> 	const char *params[2];
Jan 29 02:27:29 <PuffTheMagic> 	params[0] = val;
Jan 29 02:27:31 <PuffTheMagic> 	params[1] = txt;
Jan 29 02:27:33 <PuffTheMagic> 	PDL_CallJS("OSCevent", params, 2);
Jan 29 02:27:35 <PuffTheMagic> 	if (val) free(val);
Jan 29 02:27:37 <PuffTheMagic> }
Jan 29 02:29:27 <stbuehler> wrong thread i guess
Jan 29 02:29:35 <stbuehler> not sure it is thread-safe
Jan 29 02:30:01 <PuffTheMagic> i dont remember having thread issue in wirc
Jan 29 02:30:30 <PuffTheMagic> got it
Jan 29 02:30:34 <PuffTheMagic> it was a js issue not C
Jan 29 02:46:31 <stbuehler> i think we'll have to fix many functions handling m_data with the scrollback buffer
Jan 29 02:46:58 <PuffTheMagic> ya that is possible, i knew scrollback was going to be invasive
Jan 29 02:52:27 <PuffTheMagic> at some point we need to stabilize 0.4.0 and release it
Jan 29 02:54:02 <stbuehler> yep
**** ENDING LOGGING AT Sun Jan 29 02:59:58 2012