**** BEGIN LOGGING AT Mon Jan 30 02:59:59 2012 Jan 30 03:02:53 kk Jan 30 03:03:52 pull req incoming :) Jan 30 03:04:31 hm Jan 30 03:04:34 needs another moment Jan 30 03:04:35 ? Jan 30 03:04:38 no prob Jan 30 03:05:00 ka6sox, i will push something to alpha feed in ~1 hour after I watch the good wife Jan 30 03:05:18 kk..thansk Jan 30 03:09:29 hm, i didn't "fix" it, but it doesn't crash anymore Jan 30 03:43:56 PuffTheMagic: well, i'm done for tonight - have fun with vt52 :) gn8 Jan 30 04:02:09 night Jan 30 04:06:46 stbuehler, nice, re: vt52 Jan 30 04:30:03 ka6sox, just pushed 0.3.99 to alpha feed Jan 30 04:33:17 ka6sox, let me know what testers think Jan 30 04:33:51 thanks Jan 30 05:47:08 ka6sox, rwhitby wterm is now showing up on http://ipkg.preware.org/alpha/apps/armv7/ Jan 30 05:47:21 ahh Jan 30 05:47:24 cause i mesed up the tag Jan 30 05:48:40 hmm no the tag is right Jan 30 05:49:56 someone broke the makefile ;) Jan 30 07:11:32 ka6sox: its finally in the feeds Jan 30 07:14:56 ty Jan 30 10:11:50 äöü!! ßß? Jan 30 10:12:00 yay :) utf8 working Jan 30 10:14:27 oookay Jan 30 10:15:00 just had to set the irssi option: /SET term_charset utf-8 Jan 30 10:15:16 ah... Jan 30 10:28:05 my normal irssi setup is a little bit too large... (writing this on my tp via ssh+screen+irssi) Jan 30 10:50:59 screen FTW Jan 30 11:37:44 actually i wanted to try tmux for some time :) Jan 30 11:41:00 I want to switch to byobu, its standard now on later distros Jan 30 11:43:15 is it a terminal emulator like xterm and konsole or actually like screen? Jan 30 11:43:57 its the latest incarnation of screen I think Jan 30 11:44:02 it acts like screen Jan 30 11:44:26 why does it depend on screen on tmux? wtf Jan 30 11:44:41 I need to get food...this rsync is going to take 24hrs. Jan 30 11:44:54 I didn't look @ its deps yet. Jan 30 11:44:56 i think this is just a fronted Jan 30 11:45:03 which configures screen or tmux Jan 30 11:45:27 ah Jan 30 11:45:28 okay Jan 30 11:46:32 its going to take 4days to move this builder Jan 30 12:10:41 PuffTheMagic: can you explain why wterm is *not* a phony target in the root Makefile? Jan 30 12:11:23 it does not depend on the source of the binary (src/plugin/wterm), and it creates another file "setup" Jan 30 13:00:42 lol. you should try vttest on a plain linux text console (no X). menu 8 is really funny :) Jan 30 16:55:46 stbuehler, ya the vt52 test is hilarious Jan 30 17:14:40 stbuehler, i thought make targets that made real files were non phony Jan 30 17:14:56 i split plugin into wterm and setup though Jan 30 17:15:04 so they have their own targets now Jan 30 17:20:50 stbuehler, since i added you as a contributor, that give you perms to merge on your own? Jan 30 17:20:56 not that I have a problem with that Jan 30 17:21:09 i just want to know if i am imagining needed to pull Jan 30 17:28:09 NOOOOOOOOOOOOOOOOOOOOO Jan 30 17:28:23 * PuffTheMagic just saw something in the SDL patch from palm Jan 30 17:29:06 oh? Jan 30 17:31:21 i was trying to use the unicode parsing in sdl for our fake keyboard input events a while back and tried reusing some code in sdl which should do it Jan 30 17:31:44 it required using some ioctls on some ttys Jan 30 17:31:48 i tried a bunch of them Jan 30 17:31:59 but in the palm patch they use /dev/console Jan 30 17:32:02 which i never tested Jan 30 17:32:03 :( Jan 30 17:32:10 which might explain why i didnt work Jan 30 17:33:05 but i might not have to reimplement any of that Jan 30 17:33:08 i might be able to use SDL_WebOsEventTranslateKey() Jan 30 17:59:02 PuffTheMagic: yes, contributors have push access Jan 30 18:00:19 PuffTheMagic: well, you are right: phony targets probably shouldn't end up as files Jan 30 18:02:31 cool, re: push acess Jan 30 18:02:44 you can also close bugs via commits Jan 30 18:02:53 with fixes #num Jan 30 18:02:56 or closes #num Jan 30 18:03:00 there are a few others Jan 30 18:03:17 https://github.com/blog/831-issues-2-0-the-next-generation Jan 30 18:03:27 no more redmine? Jan 30 18:03:45 im def not using redmine Jan 30 18:03:52 for wterm Jan 30 18:04:05 i hope we dont use redmine for wirc (if that is gonna move forward) Jan 30 18:04:05 lol Jan 30 18:04:55 lol laforge octopus cat? Jan 30 18:05:05 well, for larger projects redmine is nice Jan 30 18:05:14 but for most things github is good enouch Jan 30 18:05:16 enough Jan 30 18:05:38 unless you don't like having to rely on github or any other 3rd pary for hosting Jan 30 18:06:18 stbuehler, Brybry so we need to set focus to some non plugin event and manually relay keyboard events to the plugin if we want to support phones Jan 30 18:06:42 s/plugin event/plugin object/ Jan 30 18:06:58 phone support++ Jan 30 18:07:20 oilsworkn, it "runs" on my pre3, it just gets no input Jan 30 18:07:33 i havent tried on my veer cause i cant find my cable Jan 30 18:07:40 ohh Jan 30 18:07:40 and on my pixi the plugin doesnt start for some reason Jan 30 18:07:43 need to figure that out Jan 30 18:07:50 bbl Jan 30 18:07:58 oilsworkn, let me be clearer, it gets no input in hybrid mode Jan 30 18:08:06 i can run it standalone and it gets input just fine Jan 30 18:08:08 i think my veer cable is still magnetized to the bottom of my desk for safe keeping lol Jan 30 18:08:26 ben said they didnt test enyo hybrids on non 3.x devices Jan 30 18:08:32 typical Jan 30 18:08:34 ah Jan 30 18:08:48 i wonder what the difference between enyo/mojo is Jan 30 18:08:48 push it out to all devices and ignore a critical component Jan 30 18:08:55 i mean, as far as plugin interface goes Jan 30 18:09:04 its all js, i dont understand how it would be different somehow Jan 30 18:09:21 i just started looking into this yesterday so idk Jan 30 18:09:44 sounds like the "hybrids dont work on pixi" Jan 30 18:09:53 lol Jan 30 18:09:53 until one of us actually got a pixi and tested it Jan 30 18:10:14 well i think i need to go back and look at older mojo wirc code and see how we got the plugin started Jan 30 18:10:17 it hink its a similar issue Jan 30 18:10:20 but idk Jan 30 18:10:29 PuffTheMagic: https://github.com/stbuehler/wTerm/blob/makefile/Makefile - that is what a clean makefile could look like :) Jan 30 18:11:14 stbuehler, i've never claimed to be a makefile expert Jan 30 18:11:23 oilsworkn, you use wterm on TP yet Jan 30 18:11:58 last version i played with was after you first added the keyboard i think? Jan 30 18:12:05 really :( Jan 30 18:12:11 its gotten so much better since then Jan 30 18:13:12 ill have to give the new stuff a go Jan 30 18:13:28 oilsworkn, we are doing to wterm what we did to wirc Jan 30 18:13:39 more features and compliance than most desktop terms Jan 30 18:13:45 lol Jan 30 21:04:25 woot woot Jan 30 21:04:29 Brybry, I've sent this to our planning team for inclusion in the roadmap. Jan 30 21:04:50 ^^ that was in response to my request to allow apps to set how they want keybindings to be handled Jan 30 21:58:06 stbuehler, ping Jan 30 22:04:47 oilsworkn, do you know how to parse U+ unicode notation to \u or something else? Jan 30 22:05:19 what, like regex? Jan 30 22:07:37 like.. i have a string "U+0076" that I want to be a character '\u0076' Jan 30 22:09:20 string.replace('U+', '\u') ? xD Jan 30 22:10:59 nope Jan 30 22:11:17 > 'U+0076'.replace('U+', '\u') Jan 30 22:11:18 'u0076' Jan 30 22:11:18 > '\u0076' Jan 30 22:11:18 'v' Jan 30 22:12:25 \\u? Jan 30 22:14:10 nope Jan 30 22:14:23 > 'U+0076'.replace('U+', '\\u') Jan 30 22:14:23 '\\u0076' Jan 30 22:15:33 lol Jan 30 22:17:46 js? Jan 30 22:18:36 try U\+ Jan 30 22:19:51 dwc-, as the patching pattern or replacement Jan 30 22:19:56 pattern Jan 30 22:20:02 the first one is a regex Jan 30 22:20:27 ya non of that matters Jan 30 22:20:51 where are you even getting U+0076? Jan 30 22:21:05 \u0076 Jan 30 22:21:08 javascript:alert('U+0076'.replace('U\+', '\\u')) Jan 30 22:21:18 * oilsworkn thinks you're looking at the wrong part of the key event Jan 30 22:21:43 oh, you want it to be a character, not a string? Jan 30 22:22:06 he wants 'U+0076' turned into 'v' Jan 30 22:23:09 yup Jan 30 22:25:16 String.fromCharCode() ? Jan 30 22:28:38 or eval :P Jan 30 22:28:42 PuffTheMagic: pong Jan 30 22:29:33 http://code.cside.com/3rdpage/us/javaUnicode/converter.html Jan 30 22:29:53 (view source?) Jan 30 22:30:23 stbuehler, i need to go from U+ unicode notation to something our inputhandler in C++ can handle Jan 30 22:30:39 tying to populate the unicode field for pushKeyEvent Jan 30 22:36:36 where does the U+ stuff come from? anyway, you could parse it in C++ or js; just parse the hex value Jan 30 22:36:55 its coming from document.onkeydown Jan 30 22:39:03 which attribute of the event object? Jan 30 22:39:19 keyIdentifier Jan 30 22:40:39 String.fromCharCode(parseInt(e.keyIdentifer.substr(2), 16)) Jan 30 22:40:45 might need a new plugin call for this Jan 30 22:43:54 intput on the pre3 works now :D Jan 30 22:44:07 let see if it broke bt keyboard Jan 30 22:52:16 Brybry, where are you! Jan 30 23:58:03 stbuehler, what phone webos devices do you have? Jan 31 00:19:29 PuffTheMagic: pre- with 2.whatever, and a tp Jan 31 00:21:08 you have a franken pre-? Jan 31 00:21:09 lol Jan 31 00:36:52 PuffTheMagic, are you trying to take javascript keyboard events and convert them to something the plugin can understand? Jan 31 00:37:06 and I was sleeping! Jan 31 00:37:22 Brybry, pre3 is working Jan 31 00:37:29 bt keyboard on TP is working plugin mode Jan 31 00:37:34 might have broke it for standalone Jan 31 00:37:36 :/ Jan 31 00:39:03 PuffTheMagic: franken? o2 Jan 31 00:40:02 just created a profile for me veer so i could get enyo Jan 31 00:40:11 it's easier to just redirect the events and let SDL work it out than to convert them yourself (or at least that was how I did it when I had to catch the non-focused bt key input) Jan 31 00:40:53 Brybry, things are really broke for enyo hybrids on non webos 3.x Jan 31 00:41:03 so i had to find a different method Jan 31 00:41:12 displatching the event was causing an error Jan 31 00:41:36 Brybry, feel free to play with it more Jan 31 00:41:53 but it took me all day to figure out how to get intput working on the pre3 Jan 31 00:51:07 stbuehler, is your franken pre- 2.2+ Jan 31 00:51:08 ? Jan 31 00:56:50 PuffTheMagic: is Pre 3 wTerm in the alpha or beta feeds? Jan 31 01:00:40 its not in any feeds yet, im working on making a little aux vkb for function and arrow keys Jan 31 01:00:49 then i will push an alpha build Jan 31 01:00:50 ok Jan 31 01:06:32 PuffTheMagic: 2.1 Jan 31 01:06:57 and i'm working on some generic stuff to have async events in sdl Jan 31 01:06:59 so i take it the plugins so start for you Jan 31 01:07:27 didn't try again today Jan 31 01:13:35 hmm, I get double input bt keyboard now (in plugin mode) Jan 31 01:15:14 did you make clean, or just pull and make Jan 31 01:15:57 I use a custom build file and never have old object files but I guess it's possible the install didn't overwrite the old binary for some reason Jan 31 01:15:59 I'll look into it Jan 31 01:16:14 ah Jan 31 01:16:26 it's only double when the terminal/plugin has focus Jan 31 01:16:30 (aka tap it first) Jan 31 01:16:31 which makes sense Jan 31 01:16:45 eww Jan 31 01:16:53 yes I see that now Jan 31 01:17:24 at least that means it still works in standalone! Jan 31 01:17:50 why would that have anything to do with standalone? Jan 31 01:18:58 cause that means it's catching SDL input + javascript fake input? Jan 31 01:20:48 verified stuff is still fine in standalone Jan 31 01:21:08 it did test it and it seemed like it was acting weird for me Jan 31 01:21:27 well, ctrl+c/some other things won't work but that's not our fault Jan 31 01:21:41 even in standalone? Jan 31 01:21:53 only in standalone Jan 31 01:22:09 ctrl+c dont work in plugin mode either i though Jan 31 01:22:17 ctrl+a is eaten everywhere, ctrl+c/v are eaten standalone but are not eaten in plugin Jan 31 01:22:20 it's all very beautiful Jan 31 01:23:32 ctr+c/v doest work in vttest in plugin mode Jan 31 01:24:01 it will if you tap the screen Jan 31 01:24:10 this.pushKeyEvent(1, inEvent.keyCode, String.fromCharCode(parseInt(inEvent.keyIdentifier.substr(2), 16))) <-- eats modifiers Jan 31 01:24:18 or at least it did when I did it a few weeks ago :D Jan 31 01:24:52 why would it only eat ctrl+C/V modifiers? Jan 31 01:25:02 oh, it shouldn't Jan 31 01:25:02 hmm Jan 31 01:26:05 but you're right Jan 31 01:26:08 that's interesting Jan 31 01:26:10 PuffTheMagic: i think keyIdentifier can be different strings (like "Right" and "Left" instead of "U+*"), right ? Jan 31 01:26:20 stbuehler, yes it can Jan 31 01:27:17 i don't see it handling such cases :) Jan 31 01:27:29 which is why it's odd that it works in other cases Jan 31 01:27:39 maybe the 'ctrl' down sets focus on the plugin area Jan 31 01:27:49 well, if you still get the events twice... Jan 31 01:28:07 it only gets the events twice if the terminal plugin has focus Jan 31 01:28:11 kk Jan 31 01:35:58 I wish I had a phone to test with =/ Jan 31 01:45:20 trying to remember how I did this before Jan 31 02:41:44 I wish there were a way to tell if the input was from the phone keyboard or the bt keyboard and then just handle them differently Jan 31 02:43:56 there is Jan 31 02:44:25 err well Jan 31 02:44:27 nm Jan 31 02:45:03 Brybry, i think you can just disable even propagation to stop the double Jan 31 02:45:10 yeah Jan 31 02:45:10 you can Jan 31 02:46:40 enabling allowKeyboardFocus again also fixes it (or setting the tabIndex) but the bigger problem is handling modifiers Jan 31 02:46:55 allowKeyboardFocus fixes it? Jan 31 02:47:12 yeah, that's why I put it in there in the first place Jan 31 02:47:17 iirc Jan 31 02:48:34 when I was trying to get bt focus stuff to work (before I realized I could just pass along the events caught when the plugin isn't focused) I did what we're doing right now Jan 31 02:48:51 with a regex to check if the unicode were U+blah or not Jan 31 02:49:52 but I gave up when I realized I'd have to handle each modifier individually **** ENDING LOGGING AT Tue Jan 31 02:59:58 2012