**** BEGIN LOGGING AT Tue Jan 17 02:59:57 2012
Jan 17 03:28:26 <dtzWill> what is it? :)
Jan 17 03:56:45 <PuffTheMagic> dtzWill, i figured out the bug with htop
Jan 17 03:56:53 <PuffTheMagic> and the lines that dont look right
Jan 17 03:56:58 <dtzWill> oh hooray
Jan 17 03:57:03 <PuffTheMagic> when the erase line escape code is sent
Jan 17 03:57:09 <PuffTheMagic> the characters are erased
Jan 17 03:57:28 <PuffTheMagic> but the last graphic state from the cursor is set to the erased cels
Jan 17 03:57:30 <PuffTheMagic> cells
Jan 17 03:57:38 <PuffTheMagic> so if u erase from cursor to end of line
Jan 17 03:57:59 <PuffTheMagic> the graphics state of the cursor cell is set to the all the cells after the cursor
Jan 17 03:58:34 <PuffTheMagic> i've tried a few ways to get this to work but i suck
Jan 17 04:04:16 <dtzWill> oh i see
Jan 17 04:04:20 <dtzWill> and what should it be set to?
Jan 17 04:05:26 <dtzWill> i mean what is the correct behavior
Jan 17 04:05:34 <dtzWill> (esp as compared to the above)
Jan 17 05:44:53 <PuffTheMagic> dtzWill, instead of the erased chars being set to defaultgraphics state they should be the set to the graphics state at the cursor location
Jan 17 05:45:21 <dtzWill> it's been a while, apparently erasing means "set to blank" not "remove"?
Jan 17 05:45:30 <PuffTheMagic> right
Jan 17 05:45:38 <dtzWill> ah
Jan 17 05:45:38 <PuffTheMagic> blank as in replace with a space
Jan 17 05:45:43 <PuffTheMagic> delete is separate
Jan 17 05:45:53 <PuffTheMagic> and this is erase line
Jan 17 05:47:04 <PuffTheMagic> \033[K or  \033[0K or \033[1 or \033[2
Jan 17 05:47:34 <PuffTheMagic> \033[2K
Jan 17 06:28:06 <ka6sox> Brybry, what are you using for esc?
Jan 17 06:36:39 <ka6sox> the basic stuff works...but none of the "special" keys work.
Jan 17 06:38:00 <ka6sox> good start...need to find the keycodes I did earlier
Jan 17 10:06:00 <Brybry> ka6sox, I have the hp bt keyboard so no escape key
Jan 17 10:06:47 <Brybry> I might hook up a usb keyboard again to see (but I think some of the keys for that didn't send any events)
Jan 17 16:28:36 <PuffTheMagic> Brybry, ka6sox, run vttest, which i ship with wterm
Jan 17 16:28:52 <PuffTheMagic> once in it, go to menu 5
Jan 17 16:28:55 <PuffTheMagic> test of keyboard
Jan 17 16:29:07 <PuffTheMagic> then do test 3
Jan 17 16:29:22 <PuffTheMagic> and test 9
Jan 17 16:29:35 <PuffTheMagic> test 3 is basic layout
Jan 17 16:29:41 <PuffTheMagic> 9 is control keys
Jan 17 16:29:54 <PuffTheMagic> let me know what you cant get to in those test
Jan 17 16:30:09 <PuffTheMagic> also bty
Jan 17 16:30:16 <PuffTheMagic> Ctrl+[ == Escape
Jan 17 17:57:53 <PuffTheMagic> dtzWill, ping
Jan 17 18:14:55 <ka6sox> PuffTheMagic, you are aware of the issue when you switch from landscape to portrait and back that anything that was off the screen in portrait is now no longer visible in portrait when you go back again.
Jan 17 18:15:16 <ka6sox> s/last portrait/landscape
Jan 17 18:16:44 <PuffTheMagic> is that an issue or reality of term resizing?
Jan 17 18:16:52 <PuffTheMagic> was this in an app or just at the shell prompt
Jan 17 18:17:07 <ka6sox> shell prompt
Jan 17 18:17:31 <ka6sox> I cat'd a file then switched between landscape to portrait to landscape and its cutoff.
Jan 17 18:17:48 <PuffTheMagic> hmm
Jan 17 18:17:57 <PuffTheMagic> well file it as a bug I will see what I can do about it
Jan 17 18:18:01 <ka6sox> its a wrap issue
Jan 17 18:18:33 <ka6sox> and you are using the redmine @ redmine.webos-internals.org? or github's bugs?
Jan 17 18:18:39 <PuffTheMagic> github
Jan 17 18:18:42 <PuffTheMagic> i dont think its wrap related, wterm passes all the autowrap tests in vttest
Jan 17 18:18:42 <ka6sox> kk
Jan 17 18:18:55 <PuffTheMagic> but I have an idea why this would happy
Jan 17 18:18:57 <PuffTheMagic> happen
Jan 17 18:18:59 <ka6sox> okay sorry
Jan 17 18:19:03 <PuffTheMagic> no, its ok
Jan 17 18:19:04 <ka6sox> this was a vi session
Jan 17 18:19:22 <PuffTheMagic> i've only played with rotating in htop and mc
Jan 17 18:19:25 <ka6sox> just relooked at it.
Jan 17 18:19:31 <PuffTheMagic> and they both resize properly
Jan 17 18:19:34 <ka6sox> I was in a vi session.
Jan 17 18:19:52 <ka6sox> thought I could just move it...
Jan 17 18:19:55 <PuffTheMagic> still using vt102?
Jan 17 18:19:58 <ka6sox> yes
Jan 17 18:20:27 <ka6sox> but bt now instead of VKB.
Jan 17 18:20:44 <PuffTheMagic> try xterm
Jan 17 18:20:50 <ka6sox> kk
Jan 17 18:20:52 <ka6sox> will do so
Jan 17 18:21:23 <ka6sox> I really like the way that VT works on highlighting though.
Jan 17 18:21:32 <PuffTheMagic> hmm, i cant reproduce that with xterm or vt102
Jan 17 18:21:43 <PuffTheMagic> you dont have to keep it at xterm
Jan 17 18:21:45 <ka6sox> in a vi session?
Jan 17 18:21:47 <PuffTheMagic> ya
Jan 17 18:21:52 <ka6sox> wierd.
Jan 17 18:21:53 <ka6sox> okay
Jan 17 18:22:03 <ka6sox> let me go thru the steps again.
Jan 17 18:22:05 <PuffTheMagic> what file you opening
Jan 17 18:22:15 <ka6sox> one on my laptop....
Jan 17 18:22:19 <ka6sox> oh
Jan 17 18:22:19 <PuffTheMagic> ohhh
Jan 17 18:22:23 <PuffTheMagic> so u are ssh'd?
Jan 17 18:22:29 <ka6sox> yup...duh
Jan 17 18:22:41 <PuffTheMagic> ya, there is some weirdness when going though ssh i havent tracked down yet
Jan 17 18:22:42 <ka6sox> he he...
Jan 17 18:22:51 <ka6sox> its probably a termcap issue
Jan 17 18:23:11 <ka6sox> and the terminfo not being passed thru.
Jan 17 18:23:24 <PuffTheMagic> the terminfo doesnt need to get passed though
Jan 17 18:23:34 <PuffTheMagic> the space of possible escape codes is fixed
Jan 17 18:23:39 <ka6sox> kk
Jan 17 18:23:43 <PuffTheMagic> so a term just need to be able to handle them all
Jan 17 18:23:54 <ka6sox> right
Jan 17 18:24:00 <PuffTheMagic> its up to the app to use the right escapecodes for the TERM that is set
Jan 17 18:24:01 <ka6sox> nm...
Jan 17 18:24:04 <PuffTheMagic> which is where terminfo comes in
Jan 17 18:24:24 <ka6sox> I don't remember what I have mine set to on the lappy.
Jan 17 18:24:35 <ka6sox> might be xterm, most likely ANSI
Jan 17 18:24:46 <PuffTheMagic> ANSI?
Jan 17 18:24:48 <PuffTheMagic> never seen that
Jan 17 18:24:59 <ka6sox> let me look @ that.
Jan 17 18:25:14 <PuffTheMagic> ka6sox, while you are here will you do the keyboard tests in vttest with the bt keyboard
Jan 17 18:25:28 <ka6sox> yes, I will
Jan 17 18:25:39 <PuffTheMagic> thanks
Jan 17 19:29:19 <PuffTheMagic> will someone review the saneness of this commit ->
Jan 17 19:29:27 <PuffTheMagic> dwc-, dtzWill stbuehler Brybry ^
Jan 17 19:29:30 <PuffTheMagic> lol
Jan 17 19:29:34 <PuffTheMagic> https://github.com/PuffTheMagic/wTerm/commit/79cb53b1fd9aecb89d3f35de565a60b608a7252e
Jan 17 19:29:40 <PuffTheMagic> guess it would help if i gave the link
Jan 17 19:30:29 <dwc-> nah, we're mindreaders
Jan 17 19:31:27 <dwc-> seems sane.  I don't know much about js/enyo stuff though
Jan 17 19:33:34 <dwc-> oh, this is interesting... touchpad doesn't have very many terms
Jan 17 19:33:56 <dwc-> no xterm actually
Jan 17 19:34:05 <PuffTheMagic> huh?
Jan 17 19:34:12 <dwc-> per /etc/terminfo anyways
Jan 17 19:34:27 <dwc-> oh wait, nm. it's a symlink
Jan 17 19:34:35 * dwc- was grep'ing in /md5sums
Jan 17 19:37:52 <dtzWill> PuffTheMagic: looks fine, assuming it works :)
Jan 17 19:38:28 <dtzWill> is there a difference between those enyo events (which you then forward to the pdk app) and using the sdl active events described here: https://developer.palm.com/content/api/dev-guide/pdk/handling-app-interrupts.html ?
Jan 17 19:43:26 <Brybry> hmm, could that cause a ton of events to pile up and then all try to redraw at once every 10000 ticks?
Jan 17 19:45:37 <dtzWill> Brybry: we should be processing events as much as possible before rendering
Jan 17 19:46:04 <Brybry> yeah, it wouldn't actually *render* anything anyway
Jan 17 19:46:08 <Brybry> so I guess it wouldn't be an issue
Jan 17 19:46:34 <dtzWill> it's possible we're not however, i think redraw() isn't down with the sdl_gl_swapbuffers that's for-sure only done when the event queue is empty
Jan 17 19:58:10 <Brybry> my only worry would be that, say you ran cmatrix, then minimized, and during one of the long SDL_Delays all of the events would pile up, and then you switch back in, and then it maybe has to processes 234324 VIDEOEXPOSE all at once?
Jan 17 19:58:14 <Brybry> I'd have to test it to see
Jan 17 20:02:15 <dtzWill> Brybry: agreed and understood
Jan 17 20:02:26 <dtzWill> that should be solved by moving any/all redrawing to the end of the loop
Jan 17 20:02:37 <dtzWill> (with the sdl_gl_Swapbuffers())
Jan 17 20:02:48 <Brybry> and it wouldn't break out of a long SDL_Delay either if you switch back in right at the start of that
Jan 17 20:02:49 <dtzWill> and use a flag/dirty bit while processing events to indicate we need to redraw
Jan 17 20:02:57 <Brybry> though I dunno how long 10000 ticks really is
Jan 17 20:03:36 <dtzWill> and no, unfortunately sdl_delay doesn't also automagically select() on the event queue or something :)
Jan 17 20:03:46 <dtzWill> idk either, although we could probably still do 1000 and be fine
Jan 17 20:03:54 <dtzWill> errrr assuming ms lolol
Jan 17 20:04:01 <dtzWill> "idk what unit that is, but here's my recommendation"
Jan 17 20:04:11 <Brybry> I think SDL_WaitEvent is probably some of the best sleep behavior, and maybe just have sdlterminal not send expose events if inactive?
Jan 17 20:04:47 <Brybry> unless skipping them does bad things
Jan 17 20:04:48 <Brybry> I don't really know
Jan 17 20:04:49 <dtzWill> lol
Jan 17 20:04:49 <Brybry> hmm
Jan 17 20:05:23 <dtzWill> nah, it might if we actually looked at what was exposed and did a partial redraw or something
Jan 17 20:05:46 <dtzWill> (i have no idea if sdl_Videoexpose contains damage rects)
Jan 17 20:06:05 <PuffTheMagic> dtzWill, only fullscreen PDK apps get the SDL_ACTIVE events
Jan 17 20:06:10 <PuffTheMagic> thus the need for a relay
Jan 17 20:06:40 <dtzWill> PuffTheMagic: great, thanks.  sounds good to me then, modulo things like "is the wait time too long" and ensuring our event loop does what we want
Jan 17 20:06:57 <PuffTheMagic> ka6sox, get a chance to do the keyboard tests in vttest yet?
Jan 17 20:07:40 <stbuehler> PuffTheMagic: i think plugins get a sdk event too when the app is "disabled"
Jan 17 20:07:47 <PuffTheMagic> dtzWill, ya im sure there are things to tune, i just wanted to get something basic in there first so it wasnt wasting cycles when someone wasnt looking at it
Jan 17 20:08:01 <PuffTheMagic> stbuehler, nope
Jan 17 20:08:04 <stbuehler> PuffTheMagic: sometimes it even stops enyo <-> plugin communication completely and throws error on js side
Jan 17 20:08:07 <PuffTheMagic> i confirmed it with ben
Jan 17 20:08:39 <PuffTheMagic> @unwiredben I take it SDL_ACTIVEEVENTevents are not relayed to non-fullscreen PDK plugins?
Jan 17 20:08:48 <PuffTheMagic> @_PuffTheMagic_ no, they only apply to full-screen apps. You'll have to catch activation in JS and pass it one with a method call
Jan 17 20:09:07 <dtzWill> PuffTheMagic: yeah that sounds great re:lower usage when not looked at
Jan 17 20:09:41 <PuffTheMagic> dtzWill, did u understand what I was trying to describe last night about the line erasing?
Jan 17 20:10:11 <stbuehler> well, i'm pretty sure i got errors in my qr decoder if i had a javascript timer call plugin methods while it wasn't fullscreen :)
Jan 17 20:10:24 <stbuehler> but perhaps it was on pre- only
Jan 17 20:10:42 <PuffTheMagic> pdk stuff has evolved a lot since it first came out
Jan 17 20:11:03 <dtzWill> PuffTheMagic: yep i believe so.  haven't looked at any code yet though
Jan 17 20:11:29 <dtzWill> flying out to NY tomorrow for a demo of some stuff i'm working on and...just found out it's my /other/ project that i'm demo'ing not the one i spent the last 6 months on >_>
Jan 17 20:11:41 <dtzWill> so just haven't had a chance to code
Jan 17 20:11:47 <PuffTheMagic> gotta love that
Jan 17 20:11:58 <dtzWill> wonder what it'd take to setup a mini tp-dev setup for wterm hacking on the plane :D
Jan 17 20:12:14 <dtzWill> PuffTheMagic: not really :(
Jan 17 20:12:15 <dtzWill> :P :D
Jan 17 20:17:57 <PuffTheMagic> dtzWill, i was using tide and wterm yesterday to work on wterm
Jan 17 20:18:05 <dtzWill> :D
Jan 17 20:18:18 <PuffTheMagic> can also use vi
Jan 17 20:18:27 <PuffTheMagic> vim from optware is wonky
Jan 17 20:18:30 <PuffTheMagic> idk what is wrong with it
Jan 17 20:18:37 <dtzWill> yeah i'd use vim probably, but i meant build env
Jan 17 20:18:50 <dtzWill> is gcc/etc in optware and such good enough to build the plugin?
Jan 17 20:19:02 <PuffTheMagic> ummm
Jan 17 20:19:04 <dtzWill> PuffTheMagic: oh okay.  fair enough re:broken vim
Jan 17 20:19:32 <PuffTheMagic> dtzWill, if u copy staging armv7 from build.git over
Jan 17 20:19:35 <PuffTheMagic> and then install a toolchain
Jan 17 20:19:40 <PuffTheMagic> u might be good to go
Jan 17 20:19:45 <dtzWill> well
Jan 17 20:19:46 <dtzWill> lol
Jan 17 20:19:57 <dtzWill> if i make a build env with libs and a toolchain and headers
Jan 17 20:19:57 <dtzWill> haha
Jan 17 20:20:18 <dtzWill> yeah, i was hoping for something more automagic, shrug.  worst case i play with it on the plane
Jan 17 20:20:37 <dtzWill> i bet i don't need much other than the headers from the pdk (for SDL, etc)
Jan 17 20:20:47 <dtzWill> since libs are already there...
Jan 17 20:20:50 <PuffTheMagic> echo -e -n "\033[1J\033[10;30H\033[44mBOWL OF DICKS\033[K\033\033[m\n"
Jan 17 20:20:58 <PuffTheMagic> is how u can see this bug i was talking about
Jan 17 20:21:36 <dtzWill> heh urxvt seems to get that wrong
Jan 17 20:21:42 <PuffTheMagic> opps
Jan 17 20:21:48 <PuffTheMagic> there is in extra \033
Jan 17 20:21:50 <dtzWill> assuming you didn't type and konsole gets it right
Jan 17 20:21:50 <PuffTheMagic> not that it matters
Jan 17 20:21:52 <dtzWill> oh then nvm
Jan 17 20:22:20 <dtzWill> with the extra \033 it shows up wrong in urxvt (shows "]m" at the end)
Jan 17 20:22:25 <dtzWill> konsole doesn't seem to care
Jan 17 20:22:48 <PuffTheMagic> echo -e -n "\033[1J\033[10;30H\033[44mBOWL OF DICKS\033[10;29H\033[1K\033[m\n"
Jan 17 20:22:53 <PuffTheMagic> here is the bug going backwards
Jan 17 20:23:16 <PuffTheMagic> replace [1K with [2K to see the bug on the whole line
Jan 17 20:23:37 <PuffTheMagic> basically what ever is erased get its term state set to what ever the termstate is at the cursor
Jan 17 20:27:05 <Brybry> hmm
Jan 17 20:28:18 <Brybry> PuffTheMagic: run top from wterm, set top update to 0.5 (hit s and type 0.5), then minimize wterm to card
Jan 17 20:28:43 <Brybry> and then watch wterm from ssh/novaterm =/
Jan 17 20:28:45 <Brybry> bad things happen
Jan 17 20:29:02 <Brybry> memory leaks to death or something
Jan 17 20:30:33 <Brybry> or maybe the GL stuff just keeps getting added to buffer but it never clears cause swap never gets called?
Jan 17 20:30:39 <Brybry> I don't really know opengl =(
Jan 17 20:32:42 <PuffTheMagic> hmm yess that is not good
Jan 17 20:32:46 <PuffTheMagic> mem keeps going up
Jan 17 20:32:59 <PuffTheMagic> till plugin dies
Jan 17 20:33:35 <Brybry> I think I fixed it
Jan 17 20:33:46 <Brybry> wait, nope
Jan 17 20:33:47 <Brybry> damn
Jan 17 21:40:46 <PuffTheMagic> Brybry, ping
Jan 17 21:40:54 <Brybry> I know what causes it and how to fix it
Jan 17 21:40:59 <Brybry> I'm just thinking the best way
Jan 17 21:41:02 <PuffTheMagic> awesome
Jan 17 21:41:16 <PuffTheMagic> will u do something else for me
Jan 17 21:41:19 <PuffTheMagic> its quick
Jan 17 21:41:31 <Brybry> sure
Jan 17 21:41:44 <PuffTheMagic> run the keyboard test in vttest
Jan 17 21:41:51 <PuffTheMagic> with the bt keyboard
Jan 17 21:42:17 <PuffTheMagic> once in the keyboard menu run test 9 and 3
Jan 17 21:45:33 <PuffTheMagic> i am curious what keys cant be accessed with the bt keyboard
Jan 17 21:48:06 <Brybry> vttest hates me
Jan 17 21:48:17 <PuffTheMagic> whys that?
Jan 17 21:51:09 <Brybry> because I'm on windows and I haven't made a windows-happy makefile for it yet
Jan 17 21:51:26 <PuffTheMagic> lol
Jan 17 21:51:39 <PuffTheMagic> Brybry, just steal it out of an ipk
Jan 17 21:51:43 <PuffTheMagic> and install it in /usr/bin
Jan 17 21:51:55 <PuffTheMagic> or i could send it to u
Jan 17 21:51:59 <PuffTheMagic> either way
Jan 17 21:52:04 <Brybry> I have it stolen from an ipk somewhere
Jan 17 21:52:08 <Brybry> I just have to remember where I put it
Jan 17 21:52:20 <PuffTheMagic> bug the bigger question is why you are on windows
Jan 17 21:52:49 <Brybry> because linux gaming is an oxymoron and dual booting gets tiring
Jan 17 21:52:53 <Brybry> much easier to just run linux VMs
Jan 17 21:53:11 <PuffTheMagic> i dualboot for Civ5
Jan 17 21:53:25 <PuffTheMagic> but i usually play that in 1-2 day binges
Jan 17 21:53:39 <PuffTheMagic> then stop for a few weks
Jan 17 21:57:35 <Brybry> layout test: no escape key, the backspace key is labeled delete and shows up as delete on vttest so no true backspace key I guess
Jan 17 21:57:53 <Brybry> definitely no LF/line feed key
Jan 17 21:58:12 <PuffTheMagic> well that test is for REALLY old keyboards
Jan 17 21:58:14 <Brybry> the HP bt keyboard is pretty sad on keys for such a nice slim keyboard
Jan 17 21:58:15 <PuffTheMagic> so ya ther eis no lf
Jan 17 21:58:25 <PuffTheMagic> and backspace actually sends del in linux
Jan 17 21:58:45 <PuffTheMagic> Brybry, ctrl+[ will send escape
Jan 17 21:58:52 <PuffTheMagic> which u will find when u do the other test
Jan 17 22:01:03 <dwc-> what happens if you stty erase ^H first
Jan 17 22:01:48 <PuffTheMagic> there is a special mode to make esp actually send ^H
Jan 17 22:02:14 <PuffTheMagic> but default linux terms start up as \x7f
Jan 17 22:02:50 <PuffTheMagic> hmm well actually
Jan 17 22:03:03 <PuffTheMagic> stty erase ^H is different
Jan 17 22:03:13 <Brybry> ctrl+@/space doesn't work, ctrl+H gives <127> and kicks me out of the test, ctrl+[ doesn't work and makes the usb mode notification toggle
Jan 17 22:03:17 <PuffTheMagic> from the private mode escape code that sets backspace mode
Jan 17 22:03:23 <dwc-> my ctrl space doesn't work either
Jan 17 22:03:29 <dwc-> vkb
Jan 17 22:03:33 <PuffTheMagic> dwc-, it should
Jan 17 22:03:37 <Brybry> and sometimes the background change/dehighlighting breaks
Jan 17 22:03:39 <PuffTheMagic> on git master
Jan 17 22:03:45 <Brybry> (but it picks up the keys fine)
Jan 17 22:03:47 <dwc-> and can't seem to ctrl-shift anything
Jan 17 22:03:53 <Brybry> I'm not sure if that's a term setting issue or not
Jan 17 22:04:03 <PuffTheMagic> dwc-, wtf is ctrl-shift supposed to do?
Jan 17 22:04:11 <dwc-> I'm running on dfdfcffb98a4c8e7ae7ea03179712fc556ec57ac (remove cruft)
Jan 17 22:04:14 <dwc-> from this morning
Jan 17 22:04:18 <Brybry> oh, and ctrl+A doesn't work
Jan 17 22:04:19 <dwc-> ctrl-shift-@
Jan 17 22:04:37 <PuffTheMagic> Brybry, this is with bt keyboard right
Jan 17 22:04:40 <Brybry> yeah
Jan 17 22:04:47 <Brybry> I think ctrl+H is what breaks my background stuff
Jan 17 22:04:52 <PuffTheMagic> dwc-, what is that supposed to do?
Jan 17 22:04:56 <Brybry> need to verify it first~
Jan 17 22:05:08 <dwc-> hmm, something broke the vkb's keys somewhat
Jan 17 22:05:43 <dwc-> I went ctrl a-z and E and Q didn't work and now ^Y isn't working
Jan 17 22:05:53 <dwc-> and it worked last time when I went ctrl-qwerty
Jan 17 22:06:14 <dwc-> PuffTheMagic: don't you need shift for ctrl-@ ^ ~ _ or ?
Jan 17 22:06:32 <dwc-> or am I supposed to use capslock
Jan 17 22:06:40 <PuffTheMagic> umm
Jan 17 22:07:03 <PuffTheMagic> latest wterm doesnt support the ctrl+ bindings that need shift atm
Jan 17 22:07:16 <PuffTheMagic> but u can pass that whole set of tests with out shift
Jan 17 22:07:18 <dwc-> hmm I see CTRL-E showing up but it's not unhighlighting them
Jan 17 22:07:24 <PuffTheMagic> except for ctrl+space
Jan 17 22:07:27 <PuffTheMagic> which sends null
Jan 17 22:07:30 <Brybry> dwc-: that's the bug I run into
Jan 17 22:07:31 <PuffTheMagic> and is getting ignored for some reason
Jan 17 22:07:34 <Brybry> sometimes unhighlighting breaks
Jan 17 22:07:58 <dwc-> too bad they didnt' have ^L also redraw the screen
Jan 17 22:09:23 <PuffTheMagic> hmmm
Jan 17 22:09:41 <PuffTheMagic> i cant remember what I did with those few shifted ctrl tests
Jan 17 22:10:21 <dwc-> used capslock?
Jan 17 22:10:28 <PuffTheMagic> no
Jan 17 22:10:38 <PuffTheMagic> ctrl+_ works
Jan 17 22:10:39 <dwc-> caps doesn't seem to do "what I expect" for non-alpha keys
Jan 17 22:10:45 <PuffTheMagic> (using shift_
Jan 17 22:10:48 <PuffTheMagic> shift)
Jan 17 22:10:54 <PuffTheMagic> and so does ctrl+^
Jan 17 22:11:06 <PuffTheMagic> thats for US and RS
Jan 17 22:11:11 <PuffTheMagic> the rest dont need shift
Jan 17 22:11:32 <PuffTheMagic> Brybry, so it sounds like u just need arrow key bindings for the bt keyboard?
Jan 17 22:11:48 <dwc-> which key are you hitting first?
Jan 17 22:12:18 <Brybry> well, other people have bt keyboards with escape keys and all
Jan 17 22:12:19 <dwc-> hmm, now my ctrl key is broken
Jan 17 22:12:27 <Brybry> but yeah, for the most part bt keyboards already worked fine
Jan 17 22:12:28 <dwc-> now I'm getting ` -- not a CTRL key
Jan 17 22:12:30 <PuffTheMagic> is it stuck down?
Jan 17 22:12:46 <ka6sox> some keys have caused the unit to reboot.
Jan 17 22:12:46 <Brybry> they just needed arrow key bindings (which I have in my experimental branch)
Jan 17 22:12:51 <dwc-> no, I'm holding CTRL down
Jan 17 22:13:08 <dwc-> and I hit ` and it says "` -- not a CTRL key" but without letting go of CTRL I hit Q and it registers ^Q
Jan 17 22:13:09 <ka6sox> like keys from the apple BT keyboard.
Jan 17 22:13:16 <dwc-> ka6sox: that's awesome
Jan 17 22:13:29 <dwc-> ctrl-openapple reset?
Jan 17 22:13:38 <ka6sox> someone @ palm discovered that.
Jan 17 22:13:50 <ka6sox> can't remember who.
Jan 17 22:13:54 <Brybry> I wonder if the ctrl key ran into the stuck down bug
Jan 17 22:14:00 <Brybry> and so a up never fired
Jan 17 22:14:07 <dwc-> oh look, my TP has been on power long enough that webos is telling me about 3.0.5 now
Jan 17 22:14:37 <ka6sox> and I am ignoring it till AUSMT is sorted out.
Jan 17 22:14:45 <PuffTheMagic> dwc-, ya it seems to be related to missing the touchup events
Jan 17 22:14:46 <Brybry> nope, can't figure out a what where it gets stuck up, hmm
Jan 17 22:15:51 <Brybry> sometimes the keys depress on their own which I don't understand either
Jan 17 22:16:00 <PuffTheMagic> never seen that
Jan 17 22:16:04 <Brybry> like I'll be holding down ctrl or shift or something
Jan 17 22:17:34 <PuffTheMagic> i wonder if there is a way to find out if there are any fingers on the screen
Jan 17 22:17:40 <PuffTheMagic> and poll or something
Jan 17 22:17:49 <dwc-> that'd make a funny app
Jan 17 22:18:07 <dwc-> "take your grubby little paws off me"
Jan 17 22:19:34 <ka6sox> I have noticed for some reason...that CAPS key (aka shift) sometimes takes a fairly HARD PUSH to make it work right.
Jan 17 22:19:47 <ka6sox> and that it gets LOCKED up, like it misses events.
Jan 17 22:20:06 <ka6sox> so I suspect that the BT keyboard driver needs work.
Jan 17 22:20:54 <Brybry> with the vkb ctrl+space@ and ctrl+^~` and ctrl+_? don't work for me
Jan 17 22:20:57 <Brybry> and everything else is good
Jan 17 22:21:14 <dwc-> use capslock and you should be able to pass those
Jan 17 22:21:41 <PuffTheMagic> Brybry, the space/@ issue is because something in the terminal code is not allowing a null to be sent
Jan 17 22:22:04 <Brybry> ah, yes, capslock does fix that
Jan 17 22:23:03 <ka6sox> I don't remember if the code I need to look @ is in the kernel or the way Luna processes the info.
Jan 17 22:23:20 <Brybry> ctrl+shift worked as well for _ :D
Jan 17 22:23:38 <Brybry> and ^
Jan 17 22:23:43 <Brybry> it's just very awkward to do
Jan 17 22:23:56 <dwc-> weird, mine still doesn't like ctrl-shift
Jan 17 22:24:01 <dwc-> maybe it's a 3.0.4 thing?
Jan 17 22:24:23 <Brybry> it depends where I put my fingers
Jan 17 22:25:22 <Brybry> do like lctrl + rshift + _
Jan 17 22:44:34 <Brybry> PuffTheMagic: basically do a check in SDLCore::eventLoop() case SDL_VIDEOEXPOSE: if (active) { redraw(); setdirty(blah); }
Jan 17 22:44:38 <Brybry> that at least fixes the worst case
Jan 17 22:46:52 <Brybry> basically redraw() puts stuff in the buffer and that doesn't get cleared until SDL_GL_SwapBuffers()
Jan 17 23:37:42 <dtzWill> Brybry: oh really? so the memory leak was just a the command bufffer and such getting filled?!
Jan 17 23:37:51 <dtzWill> didn't realize that'd happen, interesting.  good to know and thanks for tracking that down
Jan 18 01:17:28 <PuffTheMagic> Brybry, u found the fix, end a pull request ;)
Jan 18 02:03:13 <PuffTheMagic> or i can do it
**** ENDING LOGGING AT Wed Jan 18 02:59:56 2012