**** BEGIN LOGGING AT Fri Feb 03 02:59:57 2012 Feb 03 03:03:59 A MinEnyoVersion/MaxEnyoVersion patch to all of the Preware app, the Preware service, and the build infrastructure and feed definitions would be acceptable. Feb 03 03:18:11 rwhitby, i will look into this Feb 03 03:22:02 lesse... where is my windows vm. want to see if my little novacom/ipkg script is working Feb 03 03:31:29 if this works on windows then its a completely cross-platform script to install an ipkg, including grabbing and downloading novacom. gotta love node. Feb 03 03:37:17 halfhalo: how easy is the install and packaging for node stuff on windows? Feb 03 03:37:38 pretty easy nowadays Feb 03 03:38:16 the hardest part of this is proving to be lining up a cross platform tempfile directory Feb 03 03:38:41 (not counting making a gui, but thats to be done later) Feb 03 04:03:14 windows testing time Feb 03 04:16:45 hmm. I need to come up with a simple way to find out if novacom is already present on the host system. Grabbing it from palm and triggering installation seems to work fine on osx and win Feb 03 04:42:05 huh... windows novacom just... installs, unlike osx which prompts you first. Feb 03 04:49:35 halfhalo: yo Feb 03 04:49:46 mmmmyyyyeeeesssss? Feb 03 04:49:51 there is a way to noprompt intall it Feb 03 04:50:06 i did in novatool Feb 03 04:51:02 eh, I sorta want a prompt, but its a personal preference Feb 03 05:24:15 saweet. almost done. just have to trigger a luna-send command Feb 03 05:55:07 I think this is ready for some people to test to make sure there are no procedural issues with the way it works Feb 03 06:27:50 bleh. I am having issues with a simple thing "running luna-send installNoVerify via novacom". besides that, it works. Doesn't work cleanly, but the core concepts are there Feb 03 06:31:09 https://www.box.com/s/f1slfd8qhxxq5kl4sqtr is a copy of my current code. node 0.6.8 is needed, then just do npm install and node index.js. osx and win are both supported, since I have those OS's on my air Feb 03 09:00:27 operation wrap enyo.dispatcher.dispatch and enyo.dispatcher.dispatchCapture Feb 03 09:00:38 I'm going to find where our little lost lamb touchend goes Feb 03 14:45:34 daaaaaaamn Feb 03 14:45:39 PuffTheMagic, I forgot someone said this earlier Feb 03 14:45:56 but what happens is the touchend events get bundled together Feb 03 14:47:22 accessed by handleTouchend(inEvent) { inEvent.changedTouches } Feb 03 14:47:46 not quite sure how to grab the target yet, doesn't seem to conform entirely to http://www.w3.org/TR/touch-events/#widl-Touch-target Feb 03 14:54:45 also, I dunno if we want this to be true but it's possible to mouse down on the space between keys and not press a key down Feb 03 15:10:15 mmm, I guess it does conform, I just suck at typing Feb 03 16:08:06 stuck key bug fixed \o/ Feb 03 16:08:08 just gotta clean it up Feb 03 16:08:23 nice! Feb 03 16:16:26 uguu Feb 03 16:16:32 catches on the enter key image Feb 03 16:51:01 did we add a new class or something Feb 03 16:51:04 in HEAD Feb 03 16:51:05 hmm Feb 03 16:56:42 i renamed some stuff in the c++ part Feb 03 16:56:49 what is the problem? Feb 03 17:00:00 no problem Feb 03 17:00:26 windows/custom build/makefile so I get a nice surprise every time a .cpp gets added :D Feb 03 17:02:50 does capslock toggling work for you on the vkb? Feb 03 17:03:06 graphically Feb 03 17:14:53 don't think so Feb 03 17:18:32 Brybry, cool Feb 03 17:18:41 so i take it you removed the per-key handler Feb 03 17:18:47 and went with 1 handler for all keys Feb 03 17:18:51 no Feb 03 17:19:07 but hmm Feb 03 17:19:17 capslock thing that mousedown and touchend both get called Feb 03 17:19:26 (in some cases) Feb 03 17:19:34 that can be fixed Feb 03 17:19:49 as we want mousedown to toggle for TP and touchend to toggle for phones Feb 03 17:20:05 no Feb 03 17:20:07 the other way around Feb 03 17:20:12 yeah Feb 03 17:20:14 other way around Feb 03 17:20:17 long night Feb 03 17:20:25 btw Feb 03 17:20:39 can we perhaps put the mod state handling into js? Feb 03 17:20:48 so bt and js don't share those? Feb 03 17:21:17 moving the bt stuff to js is problematic Feb 03 17:21:33 and probably adds latency to bt key input Feb 03 17:21:46 no Feb 03 17:21:59 only the vkb modifier stuff in js Feb 03 17:22:27 oh, I guess we could but that seemed like bad practice to me. I dunno Feb 03 17:22:39 I feel like if a device has multiple keyboards I should be able to hit 'shift' on one and a key on another and get a shift+key Feb 03 17:22:46 not that it really matters Feb 03 17:23:31 not that shift works like that atm, only ctrl would I guess Feb 03 17:29:49 i think only caps would be exempt from that Feb 03 17:30:02 err nm Feb 03 17:30:42 well, a bt caps doesn't reach js next key right now anyway Feb 03 17:31:14 SDL doesn't handle caps as a modifier Feb 03 17:31:38 or at least palms code doesn't allow for it Feb 03 17:34:10 so back to the stuck keybug Feb 03 17:34:38 how were they getting bundled Feb 03 17:34:44 I'm testing my fix, trying to see if I send multiple touchdowns Feb 03 17:34:49 err touchends Feb 03 17:35:17 http://www.w3.org/TR/touch-events/#widl-TouchEvent-changedTouches Feb 03 17:36:02 the annoying thing is the targets in that list are DOM elements Feb 03 17:36:32 that shouldnt be a big deal Feb 03 17:36:40 and aren't necessarily the vbkey dom element either, can be any child of the vbkey (so the control or like for the enter key the image) Feb 03 17:36:42 i def think we might want t consolidate the handlers though Feb 03 17:37:21 Brybry, would it help if we prevent focus on the images? Feb 03 17:37:26 the enter key image made it interesting for me, I had to go up the chain of parentElements until I found one with a id that matched /^wTermApp_vkb_vkbKey\d*$/ Feb 03 17:37:27 no idea Feb 03 17:37:44 If someone can figure out a better way I'm totally down Feb 03 17:37:49 luckily this doesn't fire often/every key Feb 03 17:37:55 only on the bundles Feb 03 17:38:14 a regex test every keyup would be...bad Feb 03 17:40:02 which of those attributes are you using Feb 03 17:40:26 changedTouches? Feb 03 17:41:01 touchcancel??? Feb 03 17:41:06 where did that event comefrom Feb 03 17:41:22 I dunno, I found it when I was looking in webkit Feb 03 17:41:33 which lead me to looking up the touch event spec Feb 03 17:41:54 its a webkit even Feb 03 17:41:55 ? Feb 03 17:41:57 haven't tried looking at touchcancel yet Feb 03 17:41:59 does the TP get it? Feb 03 17:42:08 it's implemented by webkit or webcore or something Feb 03 17:42:15 I'm not sure Feb 03 17:42:26 (if it's actually catchable) Feb 03 17:42:50 I never saw one fire when I had all the event enyo dispatching stuff hooked/logging Feb 03 17:43:29 but then again enyo probably doesn't default register it anywhere Feb 03 17:43:37 it doesnt Feb 03 17:43:42 it doesnt for touches either Feb 03 17:43:59 this.node.ontouchend = enyo.bind(this,'handleTouchend') Feb 03 17:44:04 i had to add that in vkbKey Feb 03 17:44:10 i will add one for touchcancel Feb 03 17:44:14 see if it ever registers Feb 03 17:45:28 i will know the ansert as soon as it binds Feb 03 17:45:36 if it dont throw an error Feb 03 17:45:50 or this.node.addEventListener('touchcancel',function(){console.log("test")}); Feb 03 17:45:51 nice no error Feb 03 17:46:02 oh, needs a ,false on the end of that Feb 03 18:02:28 https://github.com/PuffTheMagic/wTerm/commit/6e663d7692ad502c70fae74a2bae13e26c3e4e05 Feb 03 18:02:32 if you can make it better, feel free Feb 03 18:03:14 or strip the comments as well Feb 03 18:24:30 Brybry, i still get stuck keys Feb 03 18:25:13 unless my prettying it up did something bad you shouldn't :( Feb 03 18:25:17 although maybe it is because of how i merged it with soe cleanuops Feb 03 18:26:43 bleh. someone want to help me find a way to easily run a long/complicated command via novacom Feb 03 18:27:39 also going to need some help with making it work on linux, or at least ubuntu Feb 03 18:28:29 Brybry, will you check to make sure i didnt break something Feb 03 18:29:35 yeah, one sec Feb 03 18:29:41 I was adding a paste menu option again Feb 03 18:29:50 but properly this time Feb 03 18:30:34 i cant seem to trigger is now Feb 03 18:30:42 but i did see it once after adding your code Feb 03 18:30:44 so idk Feb 03 18:30:48 maybe its good Feb 03 18:32:32 someone with linux want to run http://pastebin.com/KG08Lrrk in node 0.6 for me? Feb 03 18:34:30 halfhalo, http://pastebin.com/KqbRJ4mX Feb 03 18:34:50 yeah, seems good to me but I'll keep an eye out Feb 03 18:35:02 halfhalo, ares: '1.7.5-DEV', ??? Feb 03 18:35:05 wtf is ares Feb 03 18:35:22 ares is that enyo editor isn't it? Feb 03 18:35:33 not in this case Feb 03 18:36:06 trickytricky Feb 03 18:36:34 stbuehler, ping Feb 03 18:37:20 oh hey Feb 03 18:37:23 i got a stuck key Feb 03 18:37:30 hmmmm Feb 03 18:37:46 gotta try pretty damn hard though Feb 03 18:37:49 before it was easy Feb 03 18:37:58 that ares is a v8/google thing Feb 03 18:38:04 bleh, I'll have to put in logging again to figure out why Feb 03 18:38:26 stbuehler, i have a request for the GL shader master Feb 03 18:39:43 last step is to get that damn luna-send command to work Feb 03 18:40:41 then just cleaning it up and making it much nicer to use with a cli Feb 03 18:41:02 halfhalo, "novacom -- run file://usr/bin/luna-send ..............." Feb 03 18:41:45 doesn't like novacom run file://usr/bin/luna-send -n 6 luna://com.palm.appinstaller/installNoVerify '{\"subscribe\":true, \"target\": \"/tmp/"+file+"\", \"uncompressedSize\": 0}' 2>&1 Feb 03 18:41:59 halfhalo, add -- after novacom Feb 03 18:42:13 could be the cancels actually fire and then toggle a key up again or something Feb 03 18:42:14 we'll see Feb 03 18:42:27 though, no Feb 03 18:42:28 that's not possible Feb 03 18:59:23 PuffTheMagic didn't work. that makes it not give me any info at all Feb 03 19:01:53 not sure you need the \ for " if it's inside the ' Feb 03 19:03:02 also \"/tmp/"+file+"\", might not resolve 'file' to anything that looks like something came from a .js file? Feb 03 19:03:37 thats dealt with in the node code Feb 03 19:04:34 I need a small child to come mash their hands on my touchpad until a key gets stuck Feb 03 19:06:10 bleh. stupid quotes Feb 03 20:56:17 PuffTheMagic: I think touchend events can happen on the background? row? divs and get bundled/gobble up vkb touchend events Feb 03 20:56:49 the easiest place to test it is by pressing a key and then between the F-keys at the same time and trying to get the timing right Feb 03 20:59:22 whatever the wTermApp_vkb_control[0-9] things are Feb 03 20:59:52 prob the layouts for each row Feb 03 21:00:32 ohhhh. mac/win/linux gui node frigger. I like Feb 03 21:00:52 I bet we can unregister those for touch events Feb 03 21:02:38 touch events arent on by default Feb 03 21:02:42 that i know of at least Feb 03 21:04:23 I dunno, I know I get those in the bundles sometimes so it would only make sense that the bundles could go to them Feb 03 21:04:39 i guess Feb 03 21:04:43 but not get handled and so make it seem like the events are disappearing Feb 03 21:08:16 its a good idea Feb 03 21:08:29 im trying to clean up some of the input on phones so we can get this out of alpha Feb 03 21:11:01 I probably need to fix bt focus again before then too Feb 03 21:11:12 probably a typo Feb 03 21:11:24 type? Feb 03 21:11:38 Brybry, its not just bt focus that needs "fixing" Feb 03 21:11:54 I think focus gets set on this.$.terminal instead of this.$.terminal.$.plugin or something Feb 03 21:11:55 the same issue happens on the phones Feb 03 21:12:02 when going between physical and on screen Feb 03 21:12:21 this.$.terminal.$.plugin does not exist any more Feb 03 21:12:26 I dunno then Feb 03 21:12:28 maybe that handler doesn't get called Feb 03 21:12:34 if you look at plugin.js it IS a hybrid class Feb 03 21:13:05 Brybry, there is some dom/async weirdness getting called Feb 03 21:13:19 because im pretty sure the key handler gets called but focus transfer doesnt happen Feb 03 21:13:45 maybe the up event is restealing foucs Feb 03 21:13:46 idk Feb 03 21:13:47 it worked fine with app events so it's just a matter of figuring out what to catch that also works on phones Feb 03 21:17:04 is allowKeyboardFocus no longer set? Feb 03 21:17:24 look in wterm_app.js Feb 03 21:17:36 ah, I see Feb 03 21:18:42 yeah, that's weird. That's the one thing I knew prevented focus before (indirectly, it was because the tabIndex wasn't set) Feb 03 21:23:54 Brybry, saturday you should see it Feb 03 21:24:10 awesome Feb 03 21:24:13 that'll be useful :) Feb 03 21:37:17 that's so weird, the wtermapp keydownHandler fires...but only for wTermApp_terminal/when that has focus Feb 03 21:37:37 right... Feb 03 21:38:30 but it should be calling from wTermApp not wTermApp_terminal (as that's where it's defined) Feb 03 21:38:49 wait say that again? Feb 03 21:39:01 I do see that now Feb 03 21:39:11 i dont remember it behaving like that Feb 03 21:46:45 Brybry, so tabIndex has to do with focus? Feb 03 21:46:53 what if the keys can never have focus Feb 03 21:47:07 I forget the value then but it's -1 or 0 or undefined or something Feb 03 21:47:14 and I'm pretty sure that's not according to standard Feb 03 21:47:23 and it may only work that way for the plugin object, I didn't test it out Feb 03 21:47:45 TabIndex="-1" looks standard Feb 03 21:47:56 I just know when I gave it an appropriate tabIndex I could focus it (and that's how the enyo documentation says allowKeyboardFocus works) Feb 03 21:48:09 ya the plugin Feb 03 21:48:14 but im saying set TabIndex="-1" to all the vkbkeys Feb 03 21:48:15 I think that's still supposed to allow force focus via javascript per standard Feb 03 21:49:31 worth trying i guess Feb 03 21:51:54 think that did the trick actually Feb 03 21:52:59 question: do phones not have the ApplicationEvents kind or does it just not work Feb 03 21:54:34 its an enyo kind, so they have it if they have enyo Feb 03 21:54:39 it just doesnt seem to work Feb 03 21:54:53 i dont tink palm really tested much of enyo on phones Feb 03 21:54:59 besides that which was needed for maps Feb 03 21:59:20 it's just that it doesn't really do anything novel, unless they don't have dom events at all/document.keydown/etc Feb 03 21:59:25 when I find unlikely Feb 03 22:15:59 bleh. still having install issues Feb 03 22:18:16 Brybry, i think i have an idea whats going on with the keyDown handler in wterm_app Feb 03 22:19:07 i think the focus even needs to be async, but then the even gets missed, so i think we have to cancel auto bubble and then in a callback dispatch the deferred event Feb 03 22:19:48 not that it really matters if your vkbkey tab index bit works Feb 03 22:20:30 cause then we may not even need to catch if the plugin loses focus/do keydown handlers there Feb 03 22:21:09 the index stuff seems to be working on the TP but i have issues on my pre3 Feb 03 22:22:56 ugg. about || close to cheating and jus shoving a bash file onto the device and running that instead of luna-send from the host Feb 03 22:23:52 fail Feb 03 22:30:07 ooooh Feb 03 22:30:15 Brybry, on Keypress works on the Pre3 Feb 03 22:31:53 that's amusing becuase it shouldn't in that case Feb 03 22:32:03 but it sounds like a win in my book \o/ Feb 03 22:32:14 why shouldnt it? Feb 03 22:33:19 actually, I guess it should but just for ascii characters Feb 03 22:33:31 why just for ascii? Feb 03 22:33:50 keypress is supposed to signify like user text input or something Feb 03 22:33:53 http://www.w3.org/TR/DOM-Level-3-Events/#event-type-keypress Feb 03 22:33:56 not that specs mean anything Feb 03 22:35:03 now i wonder if the pre3 will send on keydown for non ascii char Feb 03 22:36:25 looks like on TP it doesn't send keypress for some things, like arrow keys Feb 03 22:36:42 well, looking at dispatch events at least Feb 03 22:36:53 which are what feed application events...I think Feb 03 22:39:13 hmm Feb 03 22:39:20 its like missing the dispatched event Feb 03 22:41:09 now it seems like keydown is working Feb 03 22:41:09 wtf Feb 03 22:42:38 https://gist.github.com/1733436 <-- shove that somewhere if you really like event log spam Feb 03 22:47:57 Brybry, i pushed something Feb 03 22:48:04 seems to work fine now on pre3 and TP Feb 03 22:48:07 but not on my veer Feb 03 22:50:02 eh, screw making it install right now. gonna work on device management for now Feb 03 22:51:40 seems good Feb 03 22:52:49 on the plus side, it can download and try to install any ipk from a url. Feb 03 22:55:25 found a vkb bug: if one tries to edit the input boxes in the misc prefs: our vkb resizes to make room for the palm vkb but doesn't resize again when the palm vkb disappears nor when prefs close Feb 03 22:55:49 ya i need to listen for the resize events Feb 03 22:55:57 dashboard on phones causes issues like that too Feb 03 22:58:06 the last remnant of the stuck key bug can lead to an actual repeating key now :( Feb 03 23:03:36 rwhitby, any hints on how to make novacom run "file://usr/bin/luna-send -n 6 luna://com.palm.appinstaller/installNoVerify '{'subscribe':true, 'target': '/tmp/ipkg.ipkg', 'uncompressedSize': 0}'" run properly? I appear to fail at it. stupid quotes. Feb 03 23:11:48 halfhalo: err, why not just use palm-install? Feb 03 23:13:18 because this script is meant to be used on systems which lack the dev tools. it will download novacom and attempt to install it though, which is all that is needed to talk to the device Feb 03 23:15:30 halfhalo: novacom is made avail by itself? Feb 03 23:15:36 yup Feb 03 23:15:48 or only for linux Feb 03 23:15:48 on the opensource page Feb 03 23:15:53 mac, win, linux Feb 03 23:17:27 hmm that must be a newish thing Feb 03 23:17:57 been there for a number of months Feb 03 23:18:16 since it's a required tool to be able to replace the GPL kernel on the device Feb 03 23:18:48 i miss anything Feb 03 23:18:53 think bip crashed or something Feb 03 23:20:17 rwhitby: yeah but kernel's not GPLv3, they didn't technically have to give us that. Nice that they did though Feb 03 23:20:25 yep Feb 03 23:20:40 what did who give and what was it? Feb 03 23:20:46 we asked for separate downloads like that back in November 2010 Feb 03 23:21:17 PuffTheMagic: I guess novacom is a seperate download on opensource page so our httpunzip thing isn't so important anymore :) Feb 03 23:21:27 PuffTheMagic: novacom driver binaries on opensource.palm.com Feb 03 23:21:35 oh right Feb 03 23:21:50 will be nice when source is available to it too Feb 03 23:23:16 ohh, I think i can use enyo as the gui. saweet Feb 03 23:23:51 will enyo-1 widgets work on enyo-2 base? Feb 03 23:24:29 i doubt it Feb 03 23:25:04 halfhalo: but uh.. linux browsers don't support hybrid apps right? Feb 03 23:25:31 how can you make an enyo app talk to the device? Feb 03 23:25:58 through node i would assume Feb 03 23:26:39 I can wrap the node scripts into a package with node itself and the chrome webapp framework thingie and use node to serve up enyo as the gui Feb 03 23:27:16 could also just npm install the script as well I guess, for a cli version Feb 03 23:27:34 halfhalo: yeah I think python is probably more universally supportable but that is cool Feb 03 23:27:55 except I already have the lowend working on mac, win, and theoretically linux Feb 03 23:28:23 since node is now a fully supported windows thing due to ms's support Feb 03 23:28:45 halfhalo: I'd love to see a writeup on this Feb 03 23:29:06 and/or working example Feb 03 23:29:17 Brybry: ithink we need to add audible BEL support :D Feb 03 23:29:33 soi can hear mentions in irssi Feb 03 23:29:53 I have a semi working one on box of the actual script stuff (ie minus any gui or customizability), but it shows the cross platform stuff Feb 03 23:30:20 except linux. because I didn't get around to setting the command for that. Feb 03 23:30:43 PuffTheMagic: pong Feb 03 23:30:50 but https://www.box.com/s/f1slfd8qhxxq5kl4sqtr runs on node 0.6.8 on mac and win Feb 03 23:30:58 halfhalo: that's really pretty cool considering how easy it makes cross platform thick client stuff for people who already know js etc Feb 03 23:31:28 stbuehler: we need a function that highlights a range of text Feb 03 23:31:34 idk how to write the shader for that Feb 03 23:31:48 the only code that is platform specific is the path to novacom and the url to download novacom. everything else is pretty much the same Feb 03 23:32:27 PuffTheMagic: well, the simple solution would be to invert fg/bg colors on the cells Feb 03 23:32:59 stbuehler: that would conflict with cells that are inverted because of the inverse flag Feb 03 23:33:17 PuffTheMagic: just as an extra step in the redraw loop Feb 03 23:33:30 a separate flag if you want Feb 03 23:34:10 destinal I'm cleaning up and unifying the api to all the internal functions right now, then I'm going to work on packaging it in npm I think Feb 03 23:34:22 i am thinking of something like setHighlight(start_row, start_col, end_row, end_col, color) Feb 03 23:35:27 PuffTheMagic: k, for rects i guess it should be simple Feb 03 23:42:09 my veer crashes now on rotation Feb 03 23:42:10 :( Feb 03 23:52:49 would one really need a new shader for highlighting? I would think it could all be handled from the terminalstate side via fg/bg colors (I assume you're thinking of cursor highlighting?) Feb 03 23:55:35 as i already said, no you don't :) Feb 03 23:56:09 i dont know GL period so.... Feb 03 23:56:12 although we might need some coords -> row/col transformation in sdlfont Feb 03 23:56:13 its all over my head Feb 03 23:57:32 stbuehler: color is gonna come from the same object that contains the font colors and what not Feb 03 23:57:44 so color really doesnt need to be a function param Feb 03 23:57:53 im gonna add a new pref in the UI for color Feb 04 00:02:15 stbuehler: another question Feb 04 00:02:15 case 0x07: // ^G BEL break; // terminator for OSC Feb 04 00:02:35 seems like this is only used for OSC terminator right now Feb 04 00:02:55 how would I know if this was called not part of an escape sequence Feb 04 00:03:21 I would swear that all terminals just invert the BG/FG on highlight (even if they already have inverted state it shouldn't matter, right? because they'll just get inverted again after the highlight is gone) Feb 04 00:03:23 this switch statement is there to handle C0 chars *within* escape sequences Feb 04 00:03:31 you wouldn't need to touch GL/any graphics/anything outside of terminalstate Feb 04 00:03:52 if BEL is found outside a escape sequence it will just be returned as normal char in line 512 Feb 04 00:04:20 yes, you don't need shaders for highlight (i hope i got the message across this time) Feb 04 00:04:47 just check whether the current cell is in the highlight region in wterm::redraw, and swap fg/bg Feb 04 00:05:30 the other thing is, that a mouse event will give you x/y screen coords, but you need row/column to know which text to highlight Feb 04 00:06:00 and i'd do the calculations for this in sdlfont Feb 04 00:07:12 I thought terminal state maybe already tracked that but I don't remember. maybe it only tracks cell coords and not screen coords Feb 04 00:07:17 either way it's not a super complex thing Feb 04 00:07:48 stbuehler: so i guess I couldnt be doing something like this either: Feb 04 00:07:49 case 0x08: // ^H BS \b backspace m_token = CS_ASCII_BS; Feb 04 00:07:55 screen coords are done in sdlfont (for example it centers the output) Feb 04 00:08:43 PuffTheMagic: hm, i don't get the question? we already are doing this, and it looks fine to me Feb 04 00:08:55 but if you want it for BEL Feb 04 00:09:08 ok well for \b, its not used in any escape sequances Feb 04 00:09:18 but BEL is Feb 04 00:09:34 and i want to call a JS method when it happens so I can make a sound Feb 04 00:09:39 ok Feb 04 00:09:41 but im not sure where the best place is to catch it Feb 04 00:10:28 just do if (ST_OSC != m_state && ST_OSC_ESC != m_state) { m_token = CS_ASCII_BEL; return true; } break; Feb 04 00:10:38 in the switch ase 0x07 Feb 04 00:10:58 k Feb 04 00:40:01 ooh Feb 04 00:40:05 found the default keydown sounds Feb 04 00:42:04 thinking if I need to sanitize paste input in any way Feb 04 00:42:08 I think no Feb 04 00:44:50 vkb has sound now :D Feb 04 00:45:14 i hhope there is an option to disable it :) Feb 04 00:52:39 there willbe Feb 04 00:52:51 but actually, this is not the default sound Feb 04 00:52:57 :( Feb 04 01:24:16 I wonder what the pref is for the palm keyboard clicks Feb 04 01:33:44 x_palm_virtualkeyboard_prefs "TapSounds" Feb 04 01:37:28 luna-send -i luna://com.palm.systemservice/getPreferences '{"keys":["x_palm_virtualkeyboard_prefs"],"subscribe":true}' Feb 04 01:50:06 Brybry: did that result in anything? Feb 04 01:51:23 did what result in anything Feb 04 01:51:33 that luna-send? Feb 04 01:51:38 your luna-send exploration of tap sounds Feb 04 01:52:11 it gets the pref that the system uses for if sound should play on vkb and subscribes to changed to it Feb 04 01:52:14 but I haven't found the sound file yet Feb 04 02:02:15 !!! installation is alive!!! Feb 04 02:05:35 http://pastebin.com/Sg3hdKL6 client facing api Feb 04 02:12:55 halfhalo: nice! Feb 04 02:17:01 luna-send -i palm://com.palm.audio/systemsounds/playFeedback '{"name":"default_425hz"}' scared me :( Feb 04 02:17:58 PuffTheMagic: luna-send -i palm://com.palm.audio/systemsounds/playFeedback '{"name":"down2"}' Feb 04 02:18:45 which I assume is /usr/share/systemsounds/down2-ondemand.pcm Feb 04 02:19:27 wait, nevermind Feb 04 02:19:30 that's the sound *you* used Feb 04 02:19:30 down2 is what wTerm uses right now Feb 04 02:19:34 and its nasty Feb 04 02:19:35 lol Feb 04 02:19:47 I didn't notice I was typing on the wrong keyboard to test :D Feb 04 02:20:07 :) Feb 04 02:23:32 hmm. novacom is randomly spitting out crap at me apparently Feb 04 02:23:36 luna-send -i palm://com.palm.audio/systemsounds/playFeedback '{"name":"key"}' Feb 04 02:23:38 for real this time Feb 04 02:23:48 /usr/share/systemsounds/key.pcm Feb 04 02:25:52 sweet. device.list works Feb 04 02:26:25 key works? Feb 04 02:26:27 hmm Feb 04 02:27:02 sure does Feb 04 02:27:03 nice Feb 04 02:28:11 probably should move the sound for keys into the plugin so that we can get sounds for repeat Feb 04 02:29:44 kk ty Feb 04 02:30:11 ka6sox: hows that help scene coming ;) Feb 04 02:33:53 Brybry: seriously nice find on the vkb sound Feb 04 02:34:12 i looked for sounds but didnt think to search for pcm files Feb 04 02:55:24 I spammed vkb keys and saw lunasysmgr using most of the cpu there so I then looked in ls -l /proc/pid/fd trying to see if I could find the filehandle, nothing. so I figured it did something else, did strings on the lunasysmgr binary for audio/sound/mp3/wav/etc Feb 04 02:56:18 found that lunaservice call, and that lead me to the directory by trying a find for the filename of one of the files Feb 04 02:56:58 smart Feb 04 02:58:55 hmm. randomly having the list of apps not being sent back properly **** ENDING LOGGING AT Sat Feb 04 02:59:58 2012