**** BEGIN LOGGING AT Thu Aug 13 02:59:57 2009 Aug 13 03:03:02 got filezilla, where do i go to get root status Aug 13 03:03:34 ie sudo /opt/libexec/sftp-server Aug 13 03:03:38 ie sudo /opt/libexec/sftp-server Aug 13 03:06:30 sorry... what? Aug 13 03:06:45 you mean you've logged in but need to be root to change stuff? Aug 13 03:07:21 I never finished this config but there's a wiki page on it http://www.webos-internals.org/wiki/Setup_SFTP Aug 13 03:07:58 i have it set up Aug 13 03:08:09 ive done quite a bit, but that was on my windows box Aug 13 03:08:19 now i need root sftp acces Aug 13 03:10:09 I never got that far. I suppose sftp-server could be configured to allowed root login? Aug 13 03:10:44 hmm Aug 13 03:10:44 lest see Aug 13 03:15:45 looks like the change I linked to indeed does it.... question is just how to configure filezilla. page uses winscp as example Aug 13 03:16:22 i cant even edit anything in filezilla Aug 13 03:16:36 im not sure how to set gedit as the default editor Aug 13 03:16:39 :/ Aug 13 03:17:01 works for me if I just right click > Open Aug 13 03:17:09 opens gedit Aug 13 03:18:03 in filezilla? Aug 13 03:18:07 i get open/edit Aug 13 03:18:22 yeah Aug 13 03:19:00 hmm guess it depends a lot on the filetype. Aug 13 03:19:43 no esta bien Aug 13 03:33:40 I need help Aug 13 03:33:51 here is my topic http://forums.precentral.net/web-os-development/198144-cant-get-my-pre-launcher-menu-open-after-doing-add-remove-pages-hack.html Aug 13 03:35:43 you sure you restored the original file? Aug 13 03:36:18 if thats the only file you touched, and you restored the original.... then we're at a standstill. something has to give Aug 13 03:36:34 Traveler: are you *sure* you did the 4 steps correctly? Aug 13 03:36:51 yes, it was just deleting those */ and /* parts Aug 13 03:37:10 i was also using the Notepad++ program too Aug 13 03:37:19 just those characters, correct? Aug 13 03:37:24 yep Aug 13 03:37:27 upload your modified file Aug 13 03:37:27 not the whole line Aug 13 03:37:35 no Aug 13 03:37:45 I even added them back in to see if that would work, and nothing Aug 13 03:37:49 can you pastebin the file? Aug 13 03:38:08 what does that mean Aug 13 03:38:20 http://predev.pastebin.com/ Aug 13 03:38:31 paste the contents of the file there Aug 13 03:39:09 the whole file or just the lines taht were affected? Aug 13 03:40:08 wow, I didn't realize that file was so big Aug 13 03:40:19 http://predev.pastebin.com/m30db2ef6 Aug 13 03:40:22 how about 10 or so lines on either side of the changes Aug 13 03:40:23 i think I did it right Aug 13 03:41:59 did the link work? Aug 13 03:42:34 link to the instructions you were following Aug 13 03:42:58 Would anyone know how to send a key event from a widget after gettinga mouse event Aug 13 03:43:42 Traveler: this is the original, correct? Aug 13 03:44:08 The file that I pasted there is the changes I made reverting back as mentioned in here http://www.precentral.net/how-apply-system-hacks-without-linux-access Aug 13 03:47:50 :( Aug 13 03:48:50 the file you pasted is correct, or at least until the part where you stopped pasting Aug 13 03:49:02 after a reboot, it's not working? Aug 13 03:50:39 no, when i click the launcher arrow, the icon just lights up and the top left launcher heading appears, but the screen stil shows the start screen with the icons on the bottom Aug 13 03:51:00 i haven't installed anything today, and only did this one modification Aug 13 03:52:02 Traveler: do you get the little page indicator bars at the bottom of the screen? Aug 13 03:52:11 and there are only 548 lines. the last line 549 keeps saying NULL NULL NULL NULL Aug 13 03:52:14 no, nothing at all. Aug 13 03:52:21 Is it possible you're on a blank page? Aug 13 03:52:40 hmm Aug 13 03:52:52 wait.... Aug 13 03:53:02 there are 548 lines where? Aug 13 03:53:09 in the file you edited? Aug 13 03:53:14 in that one file Aug 13 03:53:16 yes Aug 13 03:53:26 the launcher js one Aug 13 03:53:32 the original is 834 lines Aug 13 03:53:38 o_o Aug 13 03:53:50 launcher-assistant.js? Aug 13 03:53:58 just to be sure, you're talking about /usr/lib/luna/system/luna-applauncher/app/controllers/launcher-assistant.js Aug 13 03:54:04 yeah Aug 13 03:54:10 834 lines Aug 13 03:54:13 what the F- Aug 13 03:54:24 would you like me to email you the original? Aug 13 03:54:41 yes Aug 13 03:54:44 storino03@gmail.com Aug 13 03:55:03 i don't know if the notepad++ program truncated the lines or something funky. Aug 13 03:56:15 yeah I just diffed them and noticed that. you're missing a lot of stuff Aug 13 03:56:28 that's odd Aug 13 03:56:29 sent Aug 13 03:56:40 i was about to restore my phone, glad I was persistent Aug 13 03:56:41 the part you pasted is perfect Aug 13 03:56:53 but it's missing 300ish lines Aug 13 03:57:23 in the future if you forget to make a backup file, you can get them from the webos doctor download. the whole filesystem is there if you dig deep enough Aug 13 03:58:44 ok, thanks. i'm going to try the file now Aug 13 03:59:15 may the schwartz be with you Aug 13 03:59:20 :P Aug 13 04:02:10 biab... out of smokes Aug 13 04:03:22 restarting Pre now. Aug 13 04:06:07 thank the heavens, it works! :) Aug 13 04:06:15 i'm relieved. Thanks a lot guys! :) Aug 13 04:06:21 welcome Aug 13 04:24:10 \msg decimator Aug 13 04:24:13 bleh Aug 13 04:24:14 sorry Aug 13 04:32:31 when i edit the autoreplace file, what do i have to do to get it to refresh the changes Aug 13 04:32:38 or is rebooting the only thing Aug 13 04:33:54 anyone here wanna hop over to #webos and answer a few SDK questions? Aug 13 04:34:48 is it against the rules to ask em here? Aug 13 04:35:58 Well... I dunno about against the rules, I just figured, its not really "internal" :) Aug 13 04:36:15 Its quite simply "WebOS" Aug 13 04:36:30 I asked the question there already but no response Aug 13 04:36:55 funny how many people hang out in these two rooms but never say anything :) Aug 13 04:37:07 oh wait, it says right there in the topic Aug 13 04:37:15 oh hey lookie there. Aug 13 04:37:27 well I guess I was right then :D Aug 13 04:37:34 We're all waiting for genius to strike :) Aug 13 04:38:05 Actually, I got semi-reprimanded yesterday for not knowing enough to do basic tasks. So I'm here, but mostly reading up on what I should already know. Aug 13 04:38:20 hm.. sometimes genius is in the answer to a newbie's question, and not the unsolicited epiphany :) Aug 13 04:38:37 semi-reprimanded?? Aug 13 04:39:00 yeah heisenman, stop sucking so much Aug 13 04:39:03 Truly, I should read the website a little more thoroughly before I shoot my mouth off here about things. Aug 13 04:39:06 asap Aug 13 04:39:14 Techno: Trying very hard, I swear. Aug 13 04:39:23 Just got the phone last Saturday, and only rooted it on Monday. Aug 13 04:39:36 (Which is about when I found the website) Aug 13 04:39:46 it's okay, i'm a mostly n00b too. ask me some things before you ask the voices Aug 13 04:41:15 did you find the terminal program yet? Aug 13 04:41:57 http://www.webos-internals.org/wiki/Application:Terminal <-that thing's sexy as hell Aug 13 04:42:02 Oh, I always read until I get so A.D.D.'d I can barely see the words Aug 13 04:42:02 if you aks me Aug 13 04:42:06 it is indeed. Aug 13 04:42:12 then I start asking questions... its that human element thingy Aug 13 04:42:14 I notice that arrow keys work for it... Aug 13 04:42:29 And then I wonder why this has not been implemented as a general feature. Aug 13 04:42:38 yeah but they're a little tricky at first, you hold down the orange button and hit the number keys like they're arrow keys Aug 13 04:42:42 But maybe it has and I haven't read enough. Aug 13 04:43:44 woah woah woah... has someone got a front-end client working on pre? Aug 13 04:43:48 terminal i mean Aug 13 04:43:51 yes Aug 13 04:43:55 it's on the website Aug 13 04:44:00 creo, that's it right there Aug 13 04:44:16 destinal started it, and he says a couple other guys did a lot of work to wrap it up Aug 13 04:44:59 i wonder why the guys don't credit themselves and each other on each of these little apps Aug 13 04:45:37 wow Aug 13 04:47:08 looks like they also got rsync pretty well handled too Aug 13 04:47:19 which i plan on putting to good use here probably tonight Aug 13 04:53:01 kay, messed up more apps per page, does anybody have the original files so that i can just patch them? i didnt use the patch for add/ remove launcher pages, i did it by hand. Aug 13 04:54:11 or can i do that with quilt Aug 13 04:54:36 i'm sur eyou can do it with quilt, but i'm not sure how Aug 13 04:55:27 :/ Aug 13 04:56:02 maybe remove add/remove launcher pages Aug 13 04:56:07 then re patch Aug 13 05:08:32 yeah, probably just remove that and repatch Aug 13 05:09:02 I tried patching and it didn't work Aug 13 05:09:10 strange, 'cause the email patch worked Aug 13 05:09:57 stopping/ starting luna Aug 13 05:10:00 lets see Aug 13 05:10:08 jcrawford: 0.4.0pre5 was released mind checking the the update manually. Aug 13 05:10:18 nope Aug 13 05:10:22 still borked Aug 13 05:10:38 haha, i got lazy and wrote a script that sends /sbin/initctl stop LunaSysMgr && /sbin/initctl start LunaSysMgr Aug 13 05:10:46 contemplating turning it into an app Aug 13 05:11:00 * Templarian zzZZzzz will check it out in the morning. Aug 13 05:11:11 acydlord, a service Aug 13 05:16:02 acydlord: u trying that from Terminal? Aug 13 05:16:17 currently yes Aug 13 05:17:48 u can create a file in /etc/profile.d/ and have a line somethink like this: alias restart='/sbin/initctl stop LunaSysMgr && /sbin/initctl start LunaSysMgr' Aug 13 05:18:10 or u can just do pkill LunaSysMgr Aug 13 05:18:16 and it will auto restart Aug 13 05:18:40 ... u can also add it as an alias, everytime u log in that alias will work Aug 13 05:19:39 true Aug 13 05:22:17 can someone email me a file? Aug 13 05:22:29 ./usr/lib/luna/system/luna-applauncher/app/controllers/launcher-assistant.js Aug 13 05:22:42 it would be much appreciated Aug 13 05:22:58 you didn't back it up first? Aug 13 05:23:11 haha, i didnt back mine up either Aug 13 05:23:19 <3 my internet Aug 13 05:23:23 thelane3@gmail.com Aug 13 05:23:32 lane, is a pre-modded one fine? Aug 13 05:23:35 +10 internets to whoever does ir Aug 13 05:23:39 mine has more icons + add/del pages in it Aug 13 05:23:43 yeah Aug 13 05:23:45 perfect Aug 13 05:23:54 k, lemme send it Aug 13 05:23:59 :D Aug 13 05:24:40 step 1 should always be backup Aug 13 05:25:05 i did Aug 13 05:25:17 but it already has add/remove launcher pages Aug 13 05:25:26 and quilt seems to hat me atm Aug 13 05:25:30 and i have no launcher Aug 13 05:27:29 ouch Aug 13 05:28:52 mail is sent Aug 13 05:29:01 backup should be of the original Aug 13 05:29:06 not of a modified version Aug 13 05:29:13 lol Aug 13 05:29:25 yeah idk where that went :/ Aug 13 05:29:39 thank you acylord Aug 13 05:30:02 doesnt the webos doctors root image contain an unmodified version of the file? Aug 13 05:31:03 yeah, it does Aug 13 05:37:49 hey I just installed quilt and when I try push or push -a it says "The topmost patch /opt/src/patches/enable-browser-downloads.patch needs to be refreshed first." Aug 13 05:38:58 anybody run into this before? Aug 13 05:41:22 I think you need to git pull Aug 13 05:41:25 but ask rwhitby Aug 13 05:42:00 sorry, it's probably a quilt command you need to run, not git Aug 13 05:42:21 I have done git pull and it says it's up to date Aug 13 05:43:20 try quilt refresh Aug 13 05:44:12 !!!!!!!!1111eleventy!!! Aug 13 05:44:26 it worked, thanks much Aug 13 05:44:59 im ocd about my icon order, this is gonna take a while Aug 13 05:45:43 what does it mean when it says 2 out of 3 hunks failed? Aug 13 05:45:56 means the patch didnt apply correctly Aug 13 05:47:35 why wouldn't the patch apply correctly? Aug 13 05:48:11 line numbers could be wrong in the patch for the file on your local system Aug 13 05:48:56 edektor: the more icons per row patch edits the same file as the add/delete pages patch Aug 13 05:49:11 ahh I get ya Aug 13 05:49:37 rwhitby: do you think on the semantic patch systems would work better though? I've been thinking a little about this Aug 13 05:49:40 since there was a conflict, the author of the more icons per row patch incorporated the add/delete pages into his patch Aug 13 05:50:01 (hunks 2 and 3) Aug 13 05:50:16 how can I remove individual patches? Aug 13 05:50:37 edektor: that means you've modified the files that you patched, but have not asked quilt to save those modifications back into the patch before trying to apply another one Aug 13 05:50:47 edektor: I haven't figured that out... but the work-around is: Aug 13 05:50:52 quilt -p patchname.patch Aug 13 05:51:00 tmzt: dunno what a semantic patch system is. Aug 13 05:51:05 quilt pop -a Aug 13 05:51:16 err quilt pop patchname.patch Aug 13 05:51:21 pop -a removes all Aug 13 05:51:26 quilt delete /opt/src/patches/badpatch Aug 13 05:51:33 quilt push -a Aug 13 05:51:54 rwhitby: I saw an example on a kernel list mail, it essentially is a pattern to apply to the programing language syntax rather than the raw ascii, so no raw numbers Aug 13 05:52:05 no line numbers Aug 13 05:52:07 acydlord: it unapplies them all, but leaves them in your series Aug 13 05:52:16 tmzt: got a reference? Aug 13 05:52:29 they are mostly research projects now but are being used to patch security issues Aug 13 05:52:30 when i'm less lazy i need to add to the messgener part of the wiki, got it so it only sends the divider/timestamp on a protocol switch instead of every 5min Aug 13 05:52:35 let me see if I can find one Aug 13 05:52:36 then you can delete the one you want to remove and reaplly all Aug 13 05:53:38 I'm off to bed Aug 13 05:53:50 have a good one y'all Aug 13 05:54:17 night Aug 13 05:56:34 holly shit Aug 13 05:56:47 i thought i bricked my palm Aug 13 05:57:11 i was using pusb from omap-u-boot-utils Aug 13 05:57:18 to upload x-loader Aug 13 05:57:25 and i got into bootie mode Aug 13 05:57:35 and 4 tty serial devices showed up Aug 13 05:57:40 lol Aug 13 05:57:41 but i couldnt send it u-boot Aug 13 05:57:52 and then i coudnt get it to turn back on Aug 13 05:57:57 but its good now Aug 13 05:58:39 i tried to reboot my centro on accident yesterday Aug 13 05:59:05 sent commands to the wrong com port since i still have it hooked up to port it to alltel lol Aug 13 06:03:52 lol now i cant turn my pre off Aug 13 06:03:55 it just restarts Aug 13 06:03:57 great Aug 13 06:07:30 good jork Aug 13 06:07:32 jorb^ Aug 13 06:10:31 hey uh... Aug 13 06:10:36 Question for anyone out there... Aug 13 06:10:58 Just tried to restart LunaSysMgr and ended up with the "you are completely screwed" 2 pages of errors ending in a segmentation fault. Aug 13 06:11:23 heisenman, wtf are you doing Aug 13 06:11:40 enable-downloads patch fails to install. Aug 13 06:11:50 So I tried to do it manually. Aug 13 06:12:03 (Yes, I did back up the files before I edited them, but I did so... Aug 13 06:12:18 I'm hoping I caught myself before I made a change and didn't back up a file). Aug 13 06:12:29 So then I reverted to old files, but I think I forgot something. Aug 13 06:12:50 Yes, I'm still remembering what it's like to mess around with things I should be VERY CAREFUL with. Aug 13 06:15:02 So um... Aug 13 06:15:20 Anyone know how to reinstall the web browser? Aug 13 06:16:29 try typing reboot see if that fixes all Aug 13 06:17:06 Zuch: thank you, you bring up an option I'd used before. Aug 13 06:17:09 Let's see if this works. Aug 13 06:17:46 MAN that was close. Aug 13 06:17:53 pkill LunaSysMgr is a lifesaver. Aug 13 06:18:33 I've forgotten my NickServ password... First time using IRC was about a month ago... can anyone direct me on what I need to do to obtain my password? Aug 13 06:19:28 #freenode Aug 13 06:20:31 freenodes website should tell you also Aug 13 06:21:06 Great thanks. Aug 13 06:23:23 So I don't know if this is relevant to the discussions on this channel... Aug 13 06:24:02 I've found that most of the patches claiming to be available fail to install on v1.1 of the firmware. Aug 13 06:25:38 I've managed to install one with quilt (the one from yesterday making launcher icons saner), and one from source (making developer mode app visible). Aug 13 06:26:17 Why are patches failing to install so frequently right now? Is it something I should be reading on the website. Aug 13 06:33:47 is there a hyperterminal like program in linux Aug 13 06:33:50 i hate minicom Aug 13 06:37:42 heisenman i've noticed copying patches off the wiki and making a patch file in notepad++ fails for me every time Aug 13 06:38:03 anything that isnt in the repo i've had to do by hand Aug 13 06:39:05 PuffTheMagic, gkterm? Aug 13 06:39:58 err gtk term Aug 13 06:40:48 http://sourceforge.net/projects/gtkterm/ Aug 13 06:42:40 xterm, gtkterm... Aug 13 06:42:51 acydlord:thanks for the info.. Aug 13 06:43:03 heisenman, welcome Aug 13 06:43:05 I'm trying to figure out what repos have what, what they're relevant for... Aug 13 06:43:23 yeah i've been looking through those for a couple days now Aug 13 06:43:30 Possibly even a compatibility list patches->OS versions? Aug 13 06:43:37 could be that as well Aug 13 06:44:02 i've run into it a couple times as the code that needs to be modified is on a different line in 1.1 than 1.04 for some things Aug 13 06:44:11 nod Aug 13 06:44:16 I've only used 1.1 ever... Aug 13 06:44:29 rwhitby: here's two: Aug 13 06:44:57 http://www.diku.dk/hjemmesider/ansatte/jespera/doc.html Aug 13 06:45:11 http://coccinelle.lip6.fr/ Aug 13 06:45:20 the second is the one I saw an example of Aug 13 06:45:54 i'm glad i still have a spare battery from my old centro, the battery life on these stock palm batteries suck Aug 13 06:56:46 yeah..... Aug 13 07:06:55 sweetsauce, someone is starting up a hackerspace here in phx Aug 13 09:55:55 hi folks - looking for a little help trying to ssh to my emulator. Wiki says that I should use the user/pass that I created in the optware installation. I installed optware, but didn't get a user/pass. Is it different on emulator? Aug 13 10:04:24 yes, on the emulator you just ssh in as root Aug 13 10:05:16 dropbear supports logging in without a password right? Aug 13 10:05:18 even as root Aug 13 10:05:33 rwhitby: did you see what I posted earlier? Aug 13 10:05:44 think we could adapt something like that for js? Aug 13 10:08:03 tmzt: the dropbear running on the emulator allows root login, yes. Aug 13 10:08:12 tmzt: yep, looks interesting Aug 13 10:10:39 thanks for the response - I'm still not able to get in though. I got the ip address from an ifconf and am trying ssh root@10.0.2.15 -p 222 -but it just times out Aug 13 10:10:48 any ideas? Aug 13 10:12:25 all I'm really trying to do is get a copy of the internal apps' sourcecode to have a look. If there is a simpler way to do that, please let me know! Aug 13 10:27:02 anyone? Aug 13 10:33:14 good morning guys Aug 13 10:33:39 im trying to set up a cron job on the webos where do i put my script Aug 13 10:33:39 ? Aug 13 10:38:17 anywhere? Aug 13 10:39:46 is there a way to set up cron on the pe Aug 13 10:39:47 ? Aug 13 10:40:16 I think you have to add an event.d for it Aug 13 10:40:20 but doesn't wiki say this? Aug 13 10:42:54 oh ok Aug 13 10:42:58 thanks man Aug 13 10:43:02 sorry about that Aug 13 10:43:05 your right Aug 13 11:33:53 ConfusedVorlon_: the emulator doesn't use port 22 Aug 13 11:33:59 sorry, doesn't use port 222 Aug 13 11:34:14 ConfusedVorlon_: there's instructions on how to ssh into the emulator on the wiki Aug 13 11:34:46 5522 Aug 13 11:35:13 rwhitby: what port? -or can you point me to the wiki instructions. I didn't find them with a search Aug 13 11:35:26 just ssh to localhost at port 5522, use root with no password Aug 13 11:36:12 what blau-mikeDg said. Aug 13 11:37:48 ConfusedVorlon_: you're right - the instructions for ssh into the emulator do not seem to be on http://www.webos-internals.org/wiki/Portal:Accessing_Linux like they used to be. But are you sure they are not on the Palm emulator doco site? Aug 13 11:38:05 We don't duplicate information which is on Palm's site. Aug 13 11:38:40 thanks Aug 13 11:57:59 ok, so I can ssh in - but my gui ssh apps fail. Do I need to install sftp seperately? Aug 13 12:17:47 ConfusedVorlon_: I doubt that Palm included sftp on the emulator. Aug 13 12:17:55 just use scp instead Aug 13 12:32:36 brill - that did it. thanks Aug 13 12:50:28 ping egaudet|Away Aug 13 12:51:23 egaudet|Away Eric, I had a thought as I woke up this morning. Can you use /n o /r in insertCharacterAtCursor() to do an enter? It accepts  ...... Aug 13 13:41:42 yo Aug 13 13:54:56 yo back Aug 13 13:55:25 God willing, and if the FREAKING PHONE lets me to it this afternoon, the framework for Preware installer gets started today. Aug 13 13:55:31 Asuming my customers let me Aug 13 14:01:23 I am in the same boat Rick... this week has been very busy at work, no time for JS stuff Aug 13 14:07:05 can someone with a pre with ssh do me a favor Aug 13 14:07:25 PuffTheMagic: like? Aug 13 14:07:37 shh in, then do ls /boot Aug 13 14:07:47 and tellme the default uImage filename Aug 13 14:08:23 i'm jetting out in a couple minutes, but had a chance to upgrade to term 0.17. opening font option then not selecting (swiping back twice) freezes Pre hard, requires reboot. Aug 13 14:08:25 uImage-2.6.24-palm-joplin-3430 Aug 13 14:08:38 will give a more proper bug report when i'm free tonight. Aug 13 14:09:01 thanks Aug 13 14:09:56 np Aug 13 14:11:18 nice... Aug 13 14:11:22 in bootie Aug 13 14:11:36 setevn bootfile Aug 13 14:11:39 fsboot Aug 13 14:11:50 instead of trashing the symlink Aug 13 14:12:29 new kernel? Aug 13 14:12:54 nothing special Aug 13 14:13:00 i just built a kernel with my own toolchain Aug 13 14:13:02 and booted it Aug 13 14:13:05 to get the feel Aug 13 14:13:08 ah ok Aug 13 14:13:10 .... now i am gonna buil .31 Aug 13 14:13:11 :D Aug 13 14:13:15 cool Aug 13 14:13:18 have fun with that Aug 13 14:13:24 busy on too many things right now Aug 13 14:13:24 i will Aug 13 14:17:43 PuffTheMagic: don't forget BlueZ :) Aug 13 14:18:15 touchestone ftw Aug 13 14:19:13 i wish pa;lm would fix battery life Aug 13 14:19:21 something bad goes on in the background with the data connection Aug 13 14:19:26 the phone is fine with no data Aug 13 14:19:27 and dsp, I think you will have fun with that Aug 13 14:19:44 seems the dsp memory has to be reserved in the kernel Aug 13 14:19:54 but beagle kernels (and git) should be able to do it Aug 13 14:23:38 n e body know how to show how fast the battery is charging ? Aug 13 14:30:59 blau-mikeDg, data takes a lot of power Aug 13 14:33:36 jcraw: but it shouldnt be using data all the time Aug 13 14:33:38 raeb, there is a device for the battery - a onewire interface .. and you can cat some stats (curent consumption, etc) Aug 13 14:43:27 what is the pre's wireless card? Aug 13 14:43:38 is it a Marvell 8xxx Aug 13 14:45:06 and does anyone know what the touchscreen is? Aug 13 14:57:58 TSC2046? Aug 13 15:02:01 hmmm Aug 13 15:02:11 well first .31 kernel attemp is built Aug 13 15:02:19 lets see how this works Aug 13 15:03:11 touchscreen Cyprus CY8MRLN Aug 13 15:03:25 thanks Aug 13 15:04:37 * PuffTheMagic is rebooting to bootie Aug 13 15:06:42 wsup people Aug 13 15:09:23 zodttd: sup Aug 13 15:09:34 PuffTheMagic: But the toutch screen controller is a CP6944BA Cypress Semiconductor Aug 13 15:09:37 heya! Aug 13 15:11:07 I need to figure out how to build installation packages for Palm Pre apps...I've been so distracted by other projects as of late. Aug 13 15:11:38 hmmm Aug 13 15:11:43 webos ipk's right? Aug 13 15:12:57 is there a way to power off the pre with out taking out the battery when not in luna Aug 13 15:13:07 power button and orage+sym+r do not work right now Aug 13 15:13:13 shutdown -r now Aug 13 15:13:22 i am not at a console either Aug 13 15:13:28 i need a hw reboot Aug 13 15:13:32 lol if thats not working Aug 13 15:13:37 hold power for a really long itme Aug 13 15:13:37 lol Aug 13 15:13:45 take out the battery Aug 13 15:13:52 i said with out taking out the batter Aug 13 15:13:54 i hate taking out my battery Aug 13 15:13:58 zodttd: you'll need to create a webos project in eclipse, or something alike Aug 13 15:13:59 i have a case on it Aug 13 15:14:01 with the sdk Aug 13 15:14:11 zodttd: or use palm-package Aug 13 15:14:26 http://www.webos-internals.org/wiki/Tutorials_webOS_Hello_World Aug 13 15:14:40 basic overview of the structure Aug 13 15:15:04 Ok. Thats for something like psx4pre even? Where i compile with a GNU Toolchain? Aug 13 15:15:13 http://www.webos-internals.org/wiki/Portal:Tools Aug 13 15:15:17 Yeah the IPKs Aug 13 15:15:39 zodttd: i think psx4pre is beneath Luna, so it won't be an ipk Aug 13 15:15:45 errr, thats not what i meant Aug 13 15:15:54 Right though Aug 13 15:16:05 I need to disable Luna Aug 13 15:16:08 yup Aug 13 15:16:13 then reenable when closed Aug 13 15:16:20 stop LunaSysMgr Aug 13 15:16:29 right, a shell script can handle that Aug 13 15:16:33 yes but how will he invoke the executable once luna is stopped Aug 13 15:16:44 anybody ever notice in the audio class theres Aug 13 15:16:44 Media.AudioClass.VOICEDIAL Aug 13 15:17:13 initctl stop LunaSysMgr && initctl start LunaSysMgr is what i've used in the past Aug 13 15:17:25 I can whip up a shell script. Basically make a .sh Aug 13 15:17:36 and give the commands in there Aug 13 15:17:44 question will be how to invoke from luna, if even possible Aug 13 15:17:47 have that script stop Luna, run app, then start Luna Aug 13 15:17:57 otherwise it'll need to be executed via nova Aug 13 15:18:02 (guess) Aug 13 15:18:07 Well right now I didnt know you could invoke from Luna our sorta apps Aug 13 15:18:18 Ive been doing it from Nova sadly Aug 13 15:18:19 ya im not sure Aug 13 15:18:20 via SSH Aug 13 15:18:21 through service you can Aug 13 15:18:34 does service stay when luna stops? Aug 13 15:18:41 u can invoke sh scripts though Aug 13 15:18:47 and no Aug 13 15:18:53 then it's no good Aug 13 15:19:05 but the script will be running Aug 13 15:19:08 whether luna is up or not Aug 13 15:19:11 true Aug 13 15:19:45 http://www.webos-internals.org/wiki/Com_Palm_Appinstaller Aug 13 15:19:48 fyi Aug 13 15:20:29 though i'm thinking your ipk will just be a non-palm-specific armel Aug 13 15:25:22 checking, thanks nebula Aug 13 15:46:08 :D so .31 is gonna take some work ;) Aug 13 16:05:17 PuffTheMagic: oh, why? Aug 13 16:05:44 cause the only thing upstream looks like basic omap3 support Aug 13 16:05:50 but every other driver is not Aug 13 16:06:04 all the spi and i2c and gpio and mmc shit Aug 13 16:06:05 what drivers? Aug 13 16:06:09 mmc isn't? Aug 13 16:06:13 nope Aug 13 16:06:18 what mmc? Aug 13 16:06:25 oh, the internal one Aug 13 16:06:34 and some nand shit Aug 13 16:06:40 so it needs embeddedsd pdata or an equivalent Aug 13 16:06:43 no accelerometer Aug 13 16:06:44 but it should work Aug 13 16:06:45 no vibez Aug 13 16:06:47 nothing Aug 13 16:06:54 those should just copy to .31 cleanly Aug 13 16:07:00 yeah i know Aug 13 16:07:05 thats what im doing now Aug 13 16:07:09 copying over the easy shit Aug 13 16:07:15 1 at a time and make sure it builds Aug 13 16:07:17 oh, I interpreted "some work" over Aug 13 16:07:19 then i will look at the core stuff Aug 13 16:07:22 badly Aug 13 16:07:34 "some work" is an understatment Aug 13 16:07:48 but its not gonna be that bad Aug 13 16:08:20 what's on spi? Aug 13 16:08:50 the tsc2101 Aug 13 16:08:51 the tsc2102 Aug 13 16:08:54 the tsc2301 Aug 13 16:09:10 and ssi too Aug 13 16:09:13 what are they used for? Aug 13 16:09:17 ssi? Aug 13 16:09:40 nfc im just starting to dig around on this Aug 13 16:10:15 tmzt: git.webosinternals.org/kernel.git Aug 13 16:10:23 if you want to play around Aug 13 16:10:29 master has the .24 kernel Aug 13 16:10:39 master-nopalm is where i am working on the .31 stuff Aug 13 16:10:46 well, ok Aug 13 16:10:48 but what is ssi? Aug 13 16:10:54 good luck Aug 13 16:10:58 geist: lol Aug 13 16:11:00 it's a big pain to move the code up Aug 13 16:11:08 why? Aug 13 16:11:10 i think we've decided to more or less throw away the old kernel and just start over Aug 13 16:11:33 and just reapply all the stuff we need Aug 13 16:11:42 geist: any reason that can't be done in the open? Aug 13 16:11:46 geist: that i sorta was i was thinking Aug 13 16:11:52 just lack of infrastructure? Aug 13 16:11:55 geist: is there a way to disable the boot/logo.tga in bootie Aug 13 16:11:55 tmzt: because we dont have a public git server Aug 13 16:12:16 mostly just infrastructure, we're working on it Aug 13 16:12:30 we have to have a way to push what we're working on without revealing future stuff too early Aug 13 16:12:44 so we're restructuring our git trees to try to accomodate this Aug 13 16:13:01 geist: the freedesktop people did that for all their nda radeon stuff Aug 13 16:13:01 but we absolutely want to be as transparent as possible with linux Aug 13 16:13:33 sorry to use the phrase "in the open" implying it wasn't, I'm talking specificly about a public tree Aug 13 16:13:40 yeah, that'swhat i mean Aug 13 16:13:49 i got your meaning Aug 13 16:13:57 but yeah, we want to set up a public facing git server too Aug 13 16:14:13 thinking of using the model android does Aug 13 16:14:26 since they're smart, we know a bunch of them, they're just down the street, etc :) Aug 13 16:15:07 well, the had a public tree on git.android.com for the dev board they were working with since late 2007 Aug 13 16:15:26 right after the G1 was released things moved to android.git.kernel.org Aug 13 16:15:46 now qualcomm has something setup at codeaurora.org, but google still uses the kernel one Aug 13 16:51:18 new terminal 0.1.7 no workie Aug 13 16:51:24 just big black screen, no text Aug 13 16:51:49 should I not have removed x-webosinternals-termplugin? org.webosinternals-termplugin replaced it, no? Aug 13 16:55:28 t3rmin: r u taking from the HEAD? Aug 13 16:55:46 nope, just ipkg repo Aug 13 16:55:58 I just removed both packages and reinstalled Aug 13 16:56:01 rebooting now Aug 13 16:56:16 dunno what's there now, when was it updated last? do u know Aug 13 16:56:31 0.1.7 wasn't there yesterday Aug 13 16:56:43 wiki page is updated too Aug 13 16:57:00 i am working from the latests mods in the HEAD and they work fine Aug 13 16:57:42 back up and it's working Aug 13 16:57:48 but there's a scrolling issue Aug 13 16:57:58 have to scroll to see first line Aug 13 16:58:09 but the auto-login to root is nice Aug 13 16:58:51 it's one line short? Aug 13 16:58:55 hmm scrolling issue may be because I had things in the notification area when I opened it Aug 13 16:59:03 cleared those out and re-opened, and it's right Aug 13 16:59:16 ok, that's a known issue which PuffTheMagic wanted to address Aug 13 16:59:20 k Aug 13 16:59:52 dunno if that's what happened first or if there was some issue with me removing the old x-webosinternals.termplugin Aug 13 17:00:07 I *think* I tried scrolling before re-installing but I'm not sure Aug 13 17:00:29 the last instruction on the wiki is: initctl stop LunaSysMgr && initctl start LunaSysMgr Aug 13 17:00:58 as otherwise it crashes Aug 13 17:02:59 anyobdy tried Aug 13 17:03:00 http://gitorious.org/webos-internals/applications/trees/master/accelexample Aug 13 17:03:18 zuchmir2: I did that Aug 13 17:03:29 after the first upgrade Aug 13 17:04:30 atlanta yes Aug 13 17:04:42 i wrote it Aug 13 17:04:47 egaudet_work: cool cool Aug 13 17:04:57 does the accelrator work pretty fast Aug 13 17:05:00 using the service Aug 13 17:05:03 ? Aug 13 17:05:25 i thought that service was for setting the polling freq Aug 13 17:05:29 the service allows you to change it Aug 13 17:05:30 since the eventListener isnt fast rnough Aug 13 17:05:39 oh the polling rihgt Aug 13 17:05:51 right?** Aug 13 17:05:56 so by default the events trigger every 250ms, the accelservice allows changing it to whatever Aug 13 17:05:58 1 - 1000 Aug 13 17:06:12 egaudet_work: you might want to consider a name change palm/mojo already has a accelerometer service i think Aug 13 17:06:17 the accelexample uses the service to show the differences but only changes up to 50hz Aug 13 17:06:44 PuffTheMagic, named accelservice? Aug 13 17:07:02 close enough Aug 13 17:07:46 http://developer.palm.com/index.php?option=com_content&view=article&id=1554 Aug 13 17:07:51 hey egaudet_work whats the best way to install the service from git Aug 13 17:07:52 ? Aug 13 17:08:04 PuffTheMagic: that one is slow! Aug 13 17:08:10 scp the jar file to /usr/lib/luna/java/ Aug 13 17:08:17 atlanta: ? Aug 13 17:08:30 PuffTheMagic: the palm accelator service is slow Aug 13 17:08:45 atlanta: no you are confused Aug 13 17:08:46 or i am Aug 13 17:09:02 you still need to use palm accelerometer service Aug 13 17:09:16 egaudet's just lets you change the polling freq Aug 13 17:09:23 they are mutually exclusive i beleive Aug 13 17:09:39 unless i missed what he actually did Aug 13 17:09:51 no they are not mutually exclusive Aug 13 17:09:55 he actually gets the readings from the dbus correct? Aug 13 17:10:09 you do not call palm's service, i don't even know what the dbus service name is Aug 13 17:10:15 you set up a listener to listen to acceleration events Aug 13 17:10:23 you call the accelservice to change the fequency of those events Aug 13 17:10:36 egaudet_work: i though you just wrote a service to write to the /sys interface to change the frequency Aug 13 17:10:40 write Aug 13 17:10:47 right* Aug 13 17:11:05 oh ok Aug 13 17:11:09 so how are they not mutually exclusive? Aug 13 17:11:24 so you just change the frequency so that palm one reads faster Aug 13 17:11:24 ? Aug 13 17:11:25 Palm does not have a service that does that Aug 13 17:11:30 right Aug 13 17:11:36 i know that Aug 13 17:11:38 ok gotcha Aug 13 17:11:50 but your service relys on palms Aug 13 17:11:57 not really Aug 13 17:12:03 I don't see how they are mutually exclusive? Aug 13 17:12:06 you need them both Aug 13 17:12:16 there is no accelerometer dbus service from palm Aug 13 17:12:23 yea there is Aug 13 17:12:29 ++ Aug 13 17:12:39 egaudet_work: i sent you the link to it Aug 13 17:12:44 the polling is just slow Aug 13 17:12:55 I do not see any serviceRequest() calls Aug 13 17:13:10 the polling happens in the kernel Aug 13 17:13:34 which is transfered via a service Aug 13 17:13:44 i know theres a way to get accelometer readins from palm Aug 13 17:13:46 to whom? we never use this service, we set up an event listener Aug 13 17:14:21 egaudet_work: palms "service" might not use dbus... Aug 13 17:14:27 but its in the services api section Aug 13 17:14:31 If palm has a service that exports the accelerometer data to you i do not know of it Aug 13 17:14:37 and palm provides ways to get accelerometer data Aug 13 17:14:47 they provide an event to give you the data Aug 13 17:14:50 which must mean it uses some type of service in the background Aug 13 17:14:54 egaudet_work: so what Aug 13 17:14:57 egaudet_work: same thing Aug 13 17:15:02 not exactly Aug 13 17:15:06 yes exactly Aug 13 17:15:15 yea it has to have a service to interface with the dbus Aug 13 17:15:21 at some point and time Aug 13 17:15:37 they dont "need" to use dbus to generate events Aug 13 17:15:45 they could be doing this with libsysfs and a native lib Aug 13 17:15:50 built into mojo framework Aug 13 17:15:52 or that Aug 13 17:16:15 the only thing palm did NOT do is expose a method to change the polling frequency Aug 13 17:16:22 egaudet_work: which is what i though you did Aug 13 17:16:25 Ok well has anyone tracked down this service? Aug 13 17:16:33 i did not realize you exported other stuff Aug 13 17:16:37 i did not PreGame Aug 13 17:16:40 err PuffTheMagic Aug 13 17:17:11 idk whats going on now.... Aug 13 17:17:16 * PuffTheMagic goes back to kernel hacking Aug 13 17:17:18 Ok so palm must have some service that supplies accelerometer data, but I have never seen it Aug 13 17:17:29 egaudet_work: is part of the services api Aug 13 17:17:33 but i dont htink its a dbus service Aug 13 17:17:40 they don't document the service, they document the event listener Aug 13 17:17:51 same thing... Aug 13 17:17:55 how??? Aug 13 17:17:58 just a different interface to the data Aug 13 17:18:07 it dont matter if they use events or dbus hooks Aug 13 17:18:12 you get the same data Aug 13 17:18:13 the Event framework will handle this service Aug 13 17:18:41 I guess I am just confused on the Event layer then Aug 13 17:19:13 If i tap the screen, the kernel picks it up and sends it over to the web framework right? Aug 13 17:19:37 as an event Aug 13 17:20:45 the event framework then triggers and the handlers Aug 13 17:20:56 s/and/all Aug 13 17:25:40 Does anyone know what process parses out the SMS messages? Say you get a concatenated SMS message or one that is meant for a particular SMS port, or the magic SMS that wipes the user data. What is the process that reads the message and figures out what to do (and does it). Aug 13 17:26:59 magic SMS Aug 13 17:27:00 ? Aug 13 17:27:04 where did you get that from Aug 13 17:27:04 lol Aug 13 17:27:22 Hehe, there is a particular SMS that you can sign up to receive that wipes your phone (I heard) Aug 13 17:27:32 I'll try to find the reference Aug 13 17:27:36 no its through data no sms Aug 13 17:27:43 not** Aug 13 17:28:19 omg this .24 patch from palm is do dirty Aug 13 17:28:27 its like 1/3 whitespace changes Aug 13 17:29:06 lmfaooo Aug 13 17:29:15 patch for the kernel changed Aug 13 17:29:19 changes**? Aug 13 17:29:23 This page: http://www.palm.com/uk/en/products/phones/pre/ says 10 Remote erase deletes all data from your phone including files stored using USB drive mode. Remote erase command sent via SMS and must be received by activated phone within 24 hours. Wireless coverage area only. Requires data services at additional cost. Aug 13 17:30:06 but regardless, I'm not as interested in that as what process is responsible for parsing out SMS messages (PDU's) Aug 13 17:43:35 cmiller: they lied Aug 13 17:43:43 lol Aug 13 17:44:07 have you seen how the remote wipe works? Aug 13 17:44:53 kinda scary Aug 13 17:47:00 its hard coded and does checks Aug 13 17:47:17 not too scary since i can send a sms to iphone users to wipe there shit clean Aug 13 17:47:54 what code does the checks/wiping? Aug 13 17:52:51 its hard coded Aug 13 17:55:08 Something has to read the data and decide to do the wipe (or not). Do you know what process (or is it the kernel) is responsible? Aug 13 17:55:17 kernel Aug 13 17:55:52 thanks. Aug 13 18:07:33 hi Zuchmir Aug 13 18:23:10 PuffTheMagic: did you figure out whether the binary-only pieces that were listed on the wiki will be an issue for a kernel replacement being compatible with the rest of webos / fully functional / etc? Aug 13 18:23:36 i am no where close to that Aug 13 18:23:47 ah, ok Aug 13 18:23:57 i am working on getting something to Aug 13 18:24:01 "just boot" Aug 13 18:24:03 at this point Aug 13 18:26:02 PuffTheMagic: I haven't started digging into it but I figured the first thing to dry to do is get the boot messages in text form -- do you know if the animated boot logo just part of palm's kernel patch? (I know the static one is bootie) Aug 13 18:26:18 s/dry/try/ Aug 13 18:26:18 destinal meant: PuffTheMagic: I haven't started digging into it but I figured the first thing to try to do is get the boot messages in text form -- do you know if the animated boot logo just part of palm's kernel patch? (I know the static one is bootie) Aug 13 18:26:45 yeah i know how to get kernel on console now on the pre using the .24 kernel Aug 13 18:26:57 hold volup+c at boot Aug 13 18:27:03 nice Aug 13 18:27:06 but that wont be in .31 till i add it Aug 13 18:27:15 destinal: thank geist for that Aug 13 18:27:36 destinal: and if u hold down power and wiggle the lock 3 times is will hard reboot Aug 13 18:27:41 so u dont have to pop the battery Aug 13 18:27:53 geist: thank you, sir. you officially rock. :) Aug 13 18:28:08 ? Aug 13 18:28:09 :) Aug 13 18:28:24 awww yeah Aug 13 18:28:30 made my day Aug 13 18:28:38 destinal: so the kernel.git repo is where my work will be if you want to look/help Aug 13 18:28:46 yay boot output on console :) Aug 13 18:28:55 i have a branch for each "chunk" i port to .31 Aug 13 18:28:59 then i merge that into master Aug 13 18:29:01 ah so the c key thing worked? Aug 13 18:29:05 only did the accelerometer so far Aug 13 18:29:08 geist: yeah it works Aug 13 18:29:09 trouble is as soon as the ui comes up it blows it away Aug 13 18:29:17 actually that reminds me, maybe i'll look into it today Aug 13 18:29:18 right Aug 13 18:29:27 meant to figure out how to get the linux console swapping bits working Aug 13 18:29:30 geist: yeah, we could just disable luna if need be Aug 13 18:29:33 can't be that hard Aug 13 18:29:40 geist: what about a way of getting the boot console over usb or something Aug 13 18:29:48 no, that's super hard Aug 13 18:29:55 even using novaterm? Aug 13 18:29:59 anyone know how to find the last commit that changed line N in file X Aug 13 18:30:05 well, not until everything is up and going Aug 13 18:30:08 epv: git blame Aug 13 18:30:10 novaterm server doesn't come up until linux is booted Aug 13 18:30:16 thanks Aug 13 18:30:18 but by then the kernel concept of console is pointing somewhere else Aug 13 18:30:33 having bootie do it and pass something to the kernel to get initial boot messages would be non-trivial Aug 13 18:30:33 though... i think there may be some ioctl somewhere that lets you change the console Aug 13 18:30:41 should make a novacom kernel module :) Aug 13 18:30:56 but what we do do is have an in memory log of the kernel, seperate from dmesg Aug 13 18:30:57 what's the kernel's initial idea of console?one of the arm serial ports? Aug 13 18:30:59 klog Aug 13 18:31:07 iirc, we have it pointing at tty1 Aug 13 18:31:13 ahh Aug 13 18:31:13 which essentialy goes no where Aug 13 18:31:17 right. Aug 13 18:31:45 somewhere on the main board there are a couple of pads that are tied to uart3, so if we need to debug real hard we can get a tech to bring the serial port out Aug 13 18:31:54 vty switching would be cool. :) Aug 13 18:31:55 and then route the console to ttyS3 Aug 13 18:32:00 er ttyS2 Aug 13 18:32:17 but all in all debugging production devices are a pain Aug 13 18:32:31 a real incentive to just Get It Right the first time Aug 13 18:32:43 oh ok i was wondering why there was a getty on ttyS2 :) Aug 13 18:33:00 yah, the upstart script looks for console=ttyS2 or something and enables it Aug 13 18:33:03 sounds like a fun microsurgery experiment for some day in teh future Aug 13 18:33:45 but yeah, for the most part we debug stuff after the fact. we use klog a lot Aug 13 18:34:06 it essentially keeps an in memory buffer of the kernel's output, and it survives a warm boot Aug 13 18:34:11 so bootie can read it out on the next boot Aug 13 18:34:21 for production we just have the system reboot on oops Aug 13 18:34:32 but in debug mode it halts in bootie and we can extract the klog out of it Aug 13 18:36:05 god damn chromium in linux keeps crashing Aug 13 18:39:03 I somehow hosed my messaging app and it's no longer showing conversations in the main list. Is it possible to download the stock versions of those files using Terminal and then reboot to get it working again? Aug 13 18:39:13 geist, how was the vacation ? Aug 13 18:39:18 hot Aug 13 18:39:30 got on a plane at 75F, stepped off at 105F Aug 13 18:39:31 relaxing? Aug 13 18:39:35 eew Aug 13 18:40:13 yeah, apparently texas gets warm in the summertime Aug 13 18:40:23 usually i dont visit this time of year Aug 13 18:40:25 but i had put it off Aug 13 18:41:35 105 is still pretyt hot .. did it stay that warm the entire time ? Aug 13 18:41:43 yeah :) Aug 13 18:41:47 how cool does it get in the evenings? Aug 13 18:41:58 oh it'll get down to 75 or so Aug 13 18:42:03 i remember going to AZ during a heatwave with 120 during the day, but it went down to 65 at night Aug 13 18:42:03 it's just hot in the afternoon Aug 13 18:42:08 ah Aug 13 18:42:09 ew Aug 13 18:42:15 so a couple cold brews could help that out :) Aug 13 18:43:10 i have my annual review in 20 minutes .. but a) salary freeze across the board for the year and b) i have no idea what to put down for my 'individual development plan' for next year :\ Aug 13 18:43:38 in other words it's a Waste Of Time Aug 13 18:43:57 reviews with no hope of salary ups are completely pointless Aug 13 18:44:11 hopefully your boss will think so too and you can just get out of there after 30 seconds Aug 13 18:45:11 haha Aug 13 18:45:13 a complete waste of time - my boss is a real cool guy too Aug 13 18:45:18 so smart Aug 13 18:45:28 "my development plan is to sit around collecting a salary until something interesting happens" Aug 13 18:46:32 it will be interesting to get his take on what's happening here .. since he's been here for so long Aug 13 18:46:41 but yeah - a sad waste of time :( Aug 13 18:47:08 Zuchmir2: I can confirm the hard freezing issue with font change in term 0.1.7 :( Aug 13 18:47:19 i think it's any menu activity Aug 13 18:47:21 anyone know where the messaging app files are located? Aug 13 18:47:47 will, there is an apps directory under /usr/palm i thought? something like that Aug 13 18:48:06 i haven't played around with my pre in so long Aug 13 18:48:35 btw does anyone know where the term app finds out about orientation change? I'm trying to find out why setSize doesn't get called anymore on rotation Aug 13 18:49:12 in the app itself there should be a callback to the accelerometer orientation change method ? Aug 13 18:49:56 ok it looks like it's in /usr/palm/applications/com.palm.app.messaging/ Aug 13 18:49:58 i thought so but didn't see it Aug 13 18:50:11 at least, the string "accel" doesn't appear anywhere in it : ) Aug 13 18:50:17 ah Aug 13 18:50:18 if I obtain the stock versions of those files and overwrite them, should that fix the app? Aug 13 18:50:40 will, plus a reboot of the phone/reloading of Luna Aug 13 18:51:10 right Aug 13 18:51:28 I don't suppose there's a quick way to redownload that app is there? Aug 13 18:52:09 you can take apart the webos doctor image, then tar up the messaging app on your host PC. . Aug 13 18:52:17 use novacom to copy the file over .. Aug 13 18:52:23 then use novacom to untar it in place Aug 13 18:52:44 create an ipk of the webos image and keep it around for installing ? Aug 13 18:52:51 yeah I was trying to do it while I'm at work. all I can use is terminal on the phone since I can't load the novacom driver on this PC Aug 13 18:53:09 scp works too Aug 13 18:53:33 many ways to skin the cat with such a connected phone Aug 13 18:53:33 I guess I'll just wait until I get home and use the WebOS doctor Aug 13 18:54:03 that works Aug 13 18:57:24 thanks Aug 13 19:03:11 oh gitk is useful Aug 13 19:16:35 geist: is there a way to perma change env vars in booties Aug 13 19:16:46 so that they stay after reboot Aug 13 19:28:34 whats the free for all webos channel? Aug 13 19:33:59 webos is pretty free for all Aug 13 19:34:10 but someone made a chatter or something Imm not sure Aug 13 19:34:29 i mean the channel where they dont care about copyright issues or whatever Aug 13 19:34:55 i know there was this channel, perhaps it was on efnet or something Aug 13 19:57:22 lol Aug 13 19:57:22 include/linux/vibrator.h Aug 13 19:57:26 best file ever Aug 13 19:57:48 haha Aug 13 20:02:43 that's on efnet. Good luck with getting much help there. Aug 13 20:03:18 ok im a newb... can't figure out how to install novaterm on ubuntu, I do not have the /opt/nova dir that the wiki seems to assume I have when following the pre root instructions, and novaterm is no where to be found in the packages or even on the web, unless im looking under the wrong name or something? Aug 13 20:04:42 im starting to think the sdk was suppose to install novaterm? Aug 13 20:04:45 did you install the sdk? Aug 13 20:05:09 wall--- Aug 13 20:05:12 the skd installed a dire /opt/PalmSDK/* however no novaterm inside of it Aug 13 20:05:28 is there nova anything in it? Aug 13 20:05:33 yes Aug 13 20:05:35 wait - linux. Aug 13 20:05:44 you don't use novaTERM on linux that's mac Aug 13 20:05:50 you use ssh with novacom. Aug 13 20:06:01 oh Aug 13 20:06:01 Rick_work: no, linux uses novaterm as well. Aug 13 20:06:05 hmm Aug 13 20:06:07 that'sright. Aug 13 20:06:09 nevermind Aug 13 20:06:13 bad day Aug 13 20:06:17 windows doesnt use novaterm Aug 13 20:06:23 it does NOW... Aug 13 20:06:26 that we have one. Aug 13 20:06:27 :-) Aug 13 20:06:31 but doesnt it call novacomm tty:// or something like htat? Aug 13 20:06:33 right Aug 13 20:06:36 nope Aug 13 20:06:46 Geist wrote a novaterm for windows one night for me Aug 13 20:06:53 in response to a politely wordedrequest Aug 13 20:07:04 sweet guy Aug 13 20:07:13 well, I can't find any nova* stuff on my ubuntu install, or in the sdk Aug 13 20:07:15 yeah, i just saw it on the accessing linux page Aug 13 20:07:15 amazing piece of work. Aug 13 20:07:36 needs to be recomipiled for win32, butthat's all. Aug 13 20:07:40 works fine Aug 13 20:08:06 wall--- that's very odd, because without it you ain't talking toyour pre. Aug 13 20:08:13 ubunto 64 by any chance? Aug 13 20:08:18 i found it Aug 13 20:08:19 http://www.kludgesoft.com/c64/nt10.html Aug 13 20:08:31 nodnod Aug 13 20:08:44 you have 64 bit ubuntu, which you didn't mention. :-) Aug 13 20:08:52 rick; yeah thats why I want to get it, lol Aug 13 20:09:52 that site is for commodore 64 novaterm, lol Aug 13 20:10:40 i have 32bit ubuntu on my laptop maybe ill just use that... Aug 13 20:11:07 i believe installing the liba32 or whatever its called, solves the problem Aug 13 20:23:05 wall---: yeah that's not the same novaterm Aug 13 20:23:06 lol Aug 13 20:23:34 yeah I figured this much.... Aug 13 20:23:36 wall---: you need to download install the novacom deb from palm Aug 13 20:23:44 ok Aug 13 20:26:24 http://www.crombiez.com/category/palm-pre/ Aug 13 20:32:37 i guess it used to be done internal to termplugin maybe Aug 13 20:34:59 I got it working using this http://rs604.rapidshare.com/files/250160873/palm-novacom_0.3-svn162448-hud7_i386.deb.ipa Aug 13 20:36:37 root@castle:/# :) finally.... Aug 13 20:36:46 yay Aug 13 21:00:13 When a webos app calls some sdk function, say getCurrentPosition to get GPS coordinates, this must correspond to a function in some binary or library on the phone. Does anybody know which binary/library does the low level work for functions in the SDK? Aug 13 21:07:07 ok so I'm trying to switch over to bash, so I create a /etc/shells file with vi and when I go to save it I get read-only file system and it doesn't save, even when I use :w! ... and I am using sudo... Aug 13 21:07:38 cmiller: if it's a service (ie palm:// ) you can use dbus-util --list, so if it's com.palm.location that's the service name, you can dbus-util --list |grep -i com.palm.location Aug 13 21:09:52 in this case you'll see it's a java service, ipkg list | grep -i location shows a java-com.palm.location and ipkg files java-com.palm.location will list what it's made up of Aug 13 21:09:56 Hey destinal, thanks for the information. Yea, it seems to be some Java thing. Aug 13 21:10:52 cmiller: look into "system services", they can be native binaries but luna also includes a java mechanism to provide services as well Aug 13 21:11:32 services communicate over the lunaservices interface which goes over the palm bus (dbus-based) Aug 13 21:11:44 Ok, thanks again for the info. Yea, in this case it looks like it gets me to a jar file. Aug 13 21:12:16 I thought they would all be based on native binaries, but I guess not. Aug 13 21:12:20 yeah you could use jd-gui to decompile it and figure out what it does Aug 13 21:12:43 cmiller: some _are_ native binaries, but most are java Aug 13 21:14:20 wall---: have you read our wiki? Aug 13 21:16:03 wall---: just about every article that mentions modifying things on the root file system points out that it's read-only by default and list the command to remount it read-write Aug 13 21:18:32 ahh... yeah I do remember reading that I just am slow, ty Aug 13 21:26:19 egaudet|Away destinal rwhitby Eric's on-screen-keyboard is now on the front page of precentral Aug 13 21:27:00 rwhitby destinal egaudet|Away and they linked TO THE WIKI and not to the thread!!! Aug 13 21:27:05 Yeah!!!! Aug 13 21:27:25 i never got that worked... i should look at it again. Aug 13 21:30:46 if i applied the patch, should the onscreen keyboard automatically come up when entering a text field? Aug 13 21:30:53 no Aug 13 21:30:55 haha, i like the warning at the top of the wiki page Aug 13 21:31:03 how do i open it then? Aug 13 21:31:08 when you are in a text field, tap the gesture area and the kb comes up Aug 13 21:31:47 the PLAN is to have the kb NOT come up in portrait if you tap gesture unless the hardware kb is CLOSED. that doesn't work yet. Aug 13 21:32:06 also, any gesture in the gesture area right now brings up the kb not just taps. Aug 13 21:32:12 That should change to. Aug 13 21:32:35 hm... doesnt seem to work, i remember having trouble with quilt, it wanting to refresh the patch or something, perhaps it didnt apply correctly Aug 13 21:35:06 hte biggest problem of course is that since MOST of this is a framework.js patch, new versions of luna will invalidate the patch. :-( Aug 13 21:35:22 and of course, ANY OTHER framework patch, makes the diff of this one not work. Aug 13 21:35:36 yeah... Aug 13 21:35:36 we need a patching system that is not line number dependent Aug 13 21:35:47 Rick_work: make one. :D Aug 13 21:35:50 ha Aug 13 21:35:56 I've tried over the years. Aug 13 21:35:59 Several times. Aug 13 21:36:14 and? Aug 13 21:36:18 'find this set of lines... it should be somewhere afterthis line number.' Aug 13 21:36:29 "now, make that set of lines look like this: " Aug 13 21:36:44 I'm not a good enough programmer Aug 13 21:37:06 lol Aug 13 21:50:48 morning Aug 13 21:51:20 morning Aug 13 21:52:20 hello Aug 13 21:54:51 hi all... was wondering if someone was willing to point me in the right direction... updated to 1.1 and dont have PmModemFactory anymore Aug 13 21:57:46 so I extracted it from 1.0.3 and know i need to chmod 755... but stuck after that Aug 13 21:58:16 zodttd: I can show you how to do binary packages for the Pre Aug 13 21:59:13 k thanks Aug 13 22:03:00 Rick_work: cool Aug 13 22:05:09 ok, Here's the plea I keep seeing on precentral re patches. (This on the on-screen-keyboard thread) " can someone make this into one simple command? Is that even possible? " Aug 13 22:05:33 So, I've been thinking ... quilt only sort of works. Aug 13 22:05:43 and is sort of huge for the pre. Aug 13 22:05:58 and we're starting to have conflicting patches Aug 13 22:06:04 and that's only going to get worse. Aug 13 22:06:58 so, Shouldn't we think about an EXTERNAL patch utility, that pulls patches off a feed, uses novacom to extract the "interesting" files from the pre, backs them up, and then installes the patch and puts the patched files back? Aug 13 22:07:12 Rick_work: I think an ipkg for each patch with install uninstall Aug 13 22:07:24 ok Aug 13 22:07:40 Rick_work: incidentally I've been saying this since before we started down the quilt direction Aug 13 22:07:52 * rwhitby chuckles that DRG is working with PmModemFactory, but doesn't know chmod Aug 13 22:08:05 there's a bricked Pre waiting to happen ... Aug 13 22:08:16 (well, bricked on the phone side) Aug 13 22:08:22 rwhitby: yeah, poor baseband Aug 13 22:08:27 destinal if we do that, with a post-install script, we really don't need quilt right, just apost install script for the set of patches that simply walks through and applies each patch. Aug 13 22:08:57 * thatdude likes where this is going. Aug 13 22:08:57 destinal: ipkg for each patch with postinst and prerm is what I am working towards Aug 13 22:09:43 I've been working with the WebOS Quick Install author to get proper postinst and prerm support into it, just like we're going to do in ipkgservice. Aug 13 22:10:42 the trick is to have a package that *all* the patch ipks depend upon, and which marks the webOS version. then you just ipkg -recursive remove that package, and it undoes all the patches before the OTA upgrade. Aug 13 22:11:27 (you have to make sure all the patch dependencies are captured in the ipkg Depends: and Conflicts: field - this is where quilt helps by being able to give you that information when creating the patch Aug 13 22:13:16 webos quick install apparently is functional enough at the moment to be able to fully install Terminal without needing manual Linux access (of course, the program itself does the Linux Access bits in the background, via running the postinst or prerm script over novaterm at the right time) Aug 13 22:22:34 hi Aug 13 22:23:58 hi Aug 13 22:24:09 So http://home.leakdroid.com/2009/08/11/unauthorized-use-of-palm-inc-email-released-to-the-public/ leads me to believe that Palm's trademarks are on "Palm Pre" and "Palm webOS" rather than just "Pre" and "webOS". Anyone know for sure? Aug 13 22:24:46 http://www.uspto.gov/main/profiles/acadres.htm Aug 13 22:25:34 http://tess2.uspto.gov/bin/showfield?f=doc&state=4005:5i5jhk.2.4 Aug 13 22:26:09 see also http://tess2.uspto.gov/bin/showfield?f=doc&state=4005:5i5jhk.2.1 Aug 13 22:26:17 it definitely reinforces the need for the care we're taking with not redistributing Palm copyright files ... Aug 13 22:26:19 which is also live and prevents the bare webos word Aug 13 22:27:26 Interestingly, the trademark has a capital W in webOS Aug 13 22:27:39 see also http://tess2.uspto.gov/bin/showfield?f=doc&state=4005:5i5jhk.2.2 Aug 13 22:27:41 which is dead Aug 13 22:28:55 rwhitby the all upper case version is automatically inclusive of any other case, this is defined as a "word mark" and the word "palm webos" is the trademark irrespective of caps Aug 13 22:29:02 see also "palm pre" http://tess2.uspto.gov/bin/showfield?f=doc&state=4005:5i5jhk.3.1 Aug 13 22:29:38 Rick_work: yes, but the WebOS trademark (no Palm prefix) would be case-specific, no? Aug 13 22:30:20 the IMAGE mark of Palm's VERSION of WebOS is an IMAGE seperate from the word trademark Aug 13 22:30:47 oh, hang on, http://tess2.uspto.gov/bin/showfield?f=doc&state=4005:5i5jhk.2.1 is STT WebOS, unrelated to Palm. Aug 13 22:31:09 so there is no live trademark on just "WEBOS" (without the Palm or STT prefixes) ? Aug 13 22:31:23 not as a word, no Aug 13 22:32:00 good, so we have no problems with webos-internals, WebOS-Internals, "WebOS Internals" or "PREWARE" Aug 13 22:32:07 nope Aug 13 22:32:30 preware comes back "none found" Aug 13 22:33:08 and this publicly logged conversation is now in the public record :-) Aug 13 22:33:26 plus, it's really easy to search tess any time you want to Aug 13 22:39:08 I note that "palm eon" returns no hits. Aug 13 22:41:06 * rwhitby bbl Aug 13 22:44:01 * Rick_left bbl going home Aug 13 22:58:39 rwhitby: destinal ping!!! Aug 13 22:58:47 we should not be making ipk's of patches Aug 13 22:59:04 !!! Aug 13 23:01:24 PuffTheMagic: pong Aug 13 23:01:43 ipks of patches are just a bad idea Aug 13 23:01:57 granted they provide nice features Aug 13 23:02:01 PuffTheMagic: the dependencies as well as ability to easily install and uninstall are nice Aug 13 23:02:09 but they are and should remain an advanced hack Aug 13 23:02:46 ipks should be for stand alone apps that dont touch palm code only Aug 13 23:03:01 PuffTheMagic: have you ever seen cydia on the iphone? they have things you can install from their gui that totally change the operation of the system, their world hasn't ended yet Aug 13 23:03:20 the equivilent of patches basically Aug 13 23:03:56 so lets let patchA, patchB and patchC are applied on system1 Aug 13 23:04:00 and i release a patch Aug 13 23:04:11 that touches the same code as patchA and patchC Aug 13 23:04:17 and there are conflicts Aug 13 23:04:21 how are we gonna deal with that Aug 13 23:04:56 I see your point Aug 13 23:05:17 PuffTheMagic, I think you kind of hinted at a solution yesterday Aug 13 23:05:27 egaudet: the git thing? Aug 13 23:05:30 IMO we need a strict policy on patches, period, no way around it Aug 13 23:05:31 that is more for developers Aug 13 23:05:32 yea PuffTheMagic Aug 13 23:05:32 runtime patching Aug 13 23:05:37 and people who apply patches manually Aug 13 23:05:40 is also an idea Aug 13 23:05:50 destinal: what is runtime patching Aug 13 23:05:52 ohhhh Aug 13 23:05:53 why can't we require users to have a unpatched system before they start with their first patch Aug 13 23:06:04 manual patching should be FROWNED upon Aug 13 23:06:11 egaudet: u talking about the overriding existing code thing Aug 13 23:06:12 for non devs Aug 13 23:06:28 what do you mean? Aug 13 23:06:31 patches are really an advanced topic Aug 13 23:06:42 we should not automate it Aug 13 23:06:53 we can though Aug 13 23:06:54 cause its gonna involve manual interaction to resolve conflicts Aug 13 23:07:07 so people who patch palm code should be able to handle it manually, period Aug 13 23:07:17 PuffTheMagic: the only alternative I can see is people maintaining "versions" of patchsets Aug 13 23:07:18 manual patching and patching from multiple places will never work Aug 13 23:07:35 like, CoolDialer 1.9 for Dialer 1.1 Aug 13 23:07:55 destinal: so i have to maintain a a version of my patch for every possible combination of other patches that touch the same code Aug 13 23:07:55 What does package managing do for you on a debian distros Aug 13 23:07:57 then it's exclusive and provided Aug 13 23:07:57 that is crazy Aug 13 23:08:15 like people running alternative "distros" Aug 13 23:08:18 Cydia doesn't have "patches" in it Aug 13 23:08:18 of individual files Aug 13 23:08:24 like, there are some, in third party repositories Aug 13 23:08:29 but I don't support them in the main repositories Aug 13 23:08:34 we change code all the itme Aug 13 23:08:36 We support webos-internals patches, and webos-internals patches only Aug 13 23:08:37 but we do it at runtime Aug 13 23:08:38 egaudet: package management systems try their best to avoid multiple packages touching the same code Aug 13 23:08:39 which is more intelligent Aug 13 23:08:43 that is a BIG no no Aug 13 23:08:53 the very idea of changing the files on disk is downright stupid Aug 13 23:08:56 saurik: so the question I had would be how can we change mojo code at runtuime Aug 13 23:08:57 no distro has a good/any solution for that Aug 13 23:09:00 they all avoid it Aug 13 23:09:00 well lets create patches into packages Aug 13 23:09:01 we should too Aug 13 23:09:02 destinal: welcome to javascript Aug 13 23:09:02 s/tui/ti/ Aug 13 23:09:04 destinal meant: saurik: so the question I had would be how can we change mojo code at runtime Aug 13 23:09:18 research "monkey patching" Aug 13 23:09:24 framework patches depend on each other and should be patched against each other Aug 13 23:09:28 etc... down the line Aug 13 23:10:22 why cant users just apply patches them selves and deal with conficts them selfs and not put all the work on developers for maintaining 1000 different patches Aug 13 23:10:25 the most important work I did on the iphone was to construct a platform where you could modify C code at runtime in a dynamic fashion; for javascript you get that for free Aug 13 23:10:56 PuffTheMagic, because users don't operate that way Aug 13 23:10:59 it's just not the way of the world Aug 13 23:11:08 epv: pong Aug 13 23:11:11 there's no reason you should ever touch the original files: that's just asking for a world of hurt when dealing with upgrades from palm and patches that conflict with each other, etc. Aug 13 23:11:16 egaudet: u are confusing users with hackers Aug 13 23:11:16 saurik: now the trick might be that we may still need to "patch" on the target system to avoid redistributing a copy of a method that's 90% the same as stock Aug 13 23:11:25 users install standalone apps Aug 13 23:11:27 and use them Aug 13 23:11:29 ie use patching techniques to create the target file Aug 13 23:11:32 hackers modify existing code Aug 13 23:11:40 destinal: what's the issue w/fereezing u mentioned? Aug 13 23:11:44 PuffTheMagic, homebrew "hackers" are users in every sense of the word Aug 13 23:11:49 most of them Aug 13 23:12:01 egaudet: no they are a league above "users" Aug 13 23:12:01 most of them follow word-by-word instructions knowing nothing of what they are doing Aug 13 23:12:02 destinal: functions are normally not so large that you actually /need/ to patch the insides of them; putting more thought into the problem almost always allows you to find a more intelligent way around the problem Aug 13 23:12:07 hackers know what they are doing Aug 13 23:12:10 users dont give a shit Aug 13 23:12:16 i dont give a shit about users and patches Aug 13 23:12:20 destinal: although maybe's palms engineers don't organize code as well as apple's (which I'd be willing to believe) Aug 13 23:12:22 no one should Aug 13 23:12:24 PuffTheMagic, that is what I am saying... people installing the patches are users who dont give a shit Aug 13 23:12:30 just check the precentral forums Aug 13 23:12:52 yes they dont give a shit, so i dont give a shit about them Aug 13 23:12:55 they can learn Aug 13 23:12:59 and patch them selves Aug 13 23:13:03 PuffTheMagic, that doesn't work Aug 13 23:13:08 making ipks of patchs is a bad idea Aug 13 23:13:13 saurik: so it looks like I should try to learn more from how the iphone community resolved these issues Aug 13 23:13:13 egaudet: it dont work for them sure Aug 13 23:13:14 anyone want to help a noob like me connect using novaterm? im getting some weird debug error Aug 13 23:13:21 PuffTheMagic, I think that's the wrong attitude to have Aug 13 23:13:21 but i dont care about people who dont care Aug 13 23:13:28 why should we care if they dont? Aug 13 23:13:34 who really has the bad attitude here Aug 13 23:13:39 maybe they do care, they just don't know enough to understand what they are doing or what to care about Aug 13 23:13:46 I don't even know how the framework you describe for runtime patching of C code works, and it may also come in very handy for some of the lower level stuff Aug 13 23:13:51 we are the ones doing the work of creating the patch Aug 13 23:13:59 if they dont care enough to learn how to apply it Aug 13 23:14:02 why exclude countless people from using useful hacks just because we don't want to go through the trouble? Aug 13 23:14:04 then they dont deserve to use it Aug 13 23:14:11 linux wouldn't be on the desktop if that's how everyone treated it Aug 13 23:14:16 egaudet: i am not doing this for "users" Aug 13 23:14:19 Complicated things have to be dumbed down and made seamless Aug 13 23:14:20 i am doing this for ME Aug 13 23:14:23 and as a result Aug 13 23:14:32 i am making it available to others who care Aug 13 23:14:44 i dont give a shit about people who dont want to learn or care to learn Aug 13 23:14:51 Well then you shouldn't have an opinion on patches and ipks Aug 13 23:14:52 PuffTheMagic: yeah but at some level someone is going to want to package up all the deeply technical stuff we use in a consumable format Aug 13 23:14:55 egaudet: complicated things can remain complicated Aug 13 23:15:01 PuffTheMagic: so even if you don't, someone will Aug 13 23:15:05 since anyone who cares/knows will know how to resolve the issues that arise Aug 13 23:15:43 ok, I just felt the need to throw that in there, as someone claimed that Cydia was somehow managing a world of patches without problem, and I wanted to make it clear that the only reason cydia works as well as it does is that I am so anal about "no patches" ;P Aug 13 23:15:45 destinal: they can package it up if they want, we should not go out of our way to make more work for out self for all these people who cant be bothered to learn Aug 13 23:15:47 I'll leave you back to your users vs. hackers argument Aug 13 23:16:08 saurik: :D Aug 13 23:16:11 saurik: glad I pulled you into responding, I learned a few things that make a big difference Aug 13 23:16:11 * saurik goes back to working on html'izing the hell out of Cydia Aug 13 23:16:56 some things are just NOT EASY, and they should stay that way Aug 13 23:17:34 PuffTheMagic: you don't need to be an RF engineer to run a microwave oven. at some point someone took something deeply technical and packaged it up in an easy to use format and the world is better for it Aug 13 23:17:43 if a patch is so great and people want it, palm will implement it Aug 13 23:18:17 destinal: but you dont see any commercial microwave to ham radio conversion kits Aug 13 23:18:20 that is all homebrew Aug 13 23:18:22 and custom Aug 13 23:18:23 and manual Aug 13 23:18:54 yeah but what I'm getting at is that the actual customizations people want to grab and install, people are doing in cydia just fine on the iphone Aug 13 23:19:30 so we can figure out a solution that solves the problem without creating a disaster for someone who's not highly technicakl Aug 13 23:19:33 well saurik just said they dont use patches Aug 13 23:19:35 akl/al/ Aug 13 23:19:44 PuffTheMagic: so use a runtime patching approach like they do Aug 13 23:19:58 well that is what i mentioned last night Aug 13 23:19:58 override functions / methods dynamically Aug 13 23:20:03 which we dont have yet Aug 13 23:20:12 and im not against that Aug 13 23:20:17 im against patches in ipks Aug 13 23:20:34 PuffTheMagic: yeah, I'm convinced that you have a good point Aug 13 23:20:37 im against patch developers maintainsing 10000 patches Aug 13 23:20:53 we're still on a small scale, as we scale up it becomes a disaster Aug 13 23:21:25 destinal: i am coming from a world of gentoo users, who i would say are among some of the better linux users in general, who cant handle patching their kernel with a new driver Aug 13 23:21:39 and we want non-linux pre users to patch all this palm code Aug 13 23:21:43 that is insane Aug 13 23:22:17 So duck punching / monkey patching sounds promising Aug 13 23:22:33 donkey punching Aug 13 23:22:34 :D Aug 13 23:23:00 quick question Aug 13 23:23:05 there are still some potential issues with version changes but nowhere near as many Aug 13 23:23:11 can i revert a file to the patched file via quilt? Aug 13 23:24:12 you can pop all the patches in the series back past the patch u want to remove Aug 13 23:24:16 then comment it out in the series Aug 13 23:24:31 the re push all the other patches Aug 13 23:24:33 with minimal issues Aug 13 23:26:06 ok so what to do if i want to undo the current modifications that are not in a patch yet Aug 13 23:26:16 haha Aug 13 23:26:23 thats why quilt sucks Aug 13 23:26:27 you cant Aug 13 23:26:37 if you make changes and you are using quilt Aug 13 23:26:42 make sure you make a patch out of it Aug 13 23:26:49 git would auto track that shit Aug 13 23:26:54 and you could reset all those changes Aug 13 23:26:58 you are basically fucked Aug 13 23:27:05 hope you made a backup Aug 13 23:27:09 this is dumb Aug 13 23:27:11 i did not Aug 13 23:27:19 how can i do a diff of the current stuff and the quilt patch? Aug 13 23:27:20 yes quilt is "dumb" Aug 13 23:27:36 egaudet: i dont get what you mean Aug 13 23:27:48 i want to KNOW what i changed since the last patch Aug 13 23:27:49 you need to tell quilt to start tracking a file BEFORE you start editing it Aug 13 23:27:55 see above Aug 13 23:28:10 if you edited a file with out telling quilt to track it you can do anything about that Aug 13 23:28:11 i have to tell quilt to start tracking EACH time? Aug 13 23:28:15 yes Aug 13 23:28:21 that is why git is superior Aug 13 23:28:29 WTF Aug 13 23:28:42 quilt it an old ass tool Aug 13 23:28:44 quilt files lists the files though Aug 13 23:28:58 that is not what it means Aug 13 23:29:11 quilt files list files that were touched by files in the series Aug 13 23:29:13 basicallyt Aug 13 23:29:29 in what series Aug 13 23:29:33 :D Aug 13 23:29:42 you should have a series file Aug 13 23:29:49 in you working dir somewhere Aug 13 23:29:55 that is a list of all the patches that are applied Aug 13 23:29:59 in the order they are applied Aug 13 23:30:13 all quilt does is make it easy to apply a shit load of patches Aug 13 23:30:20 its not a great version control tool Aug 13 23:30:24 its not one at all Aug 13 23:30:31 its a patch manager thats it Aug 13 23:31:09 so what do i do now Aug 13 23:31:24 revert, patch, add, refresh Aug 13 23:32:01 no Aug 13 23:32:06 cause you will have a dirty file Aug 13 23:32:12 like i said, you are fucked Aug 13 23:32:22 the only thing you can do Aug 13 23:32:27 its pop ALL your patches Aug 13 23:32:42 then ask someone to send you a clean version of the file you fucked up Aug 13 23:32:45 so u can replace it Aug 13 23:32:52 then you can push all your files again Aug 13 23:33:10 if it's webos stuff you can just extract from the doctor image Aug 13 23:33:21 pop patches, diff with the clean tree Aug 13 23:33:23 or ask someone to send it to you Aug 13 23:33:26 how do i know if im "tracking" files to patch Aug 13 23:33:38 egaudet: if you did quilt edit Aug 13 23:33:42 then its being tracked Aug 13 23:33:43 if not Aug 13 23:33:47 then its not being tracked Aug 13 23:33:55 what if i don't remember if i did that Aug 13 23:34:03 then you a sol Aug 13 23:34:06 use git Aug 13 23:34:09 I think git is definitely the thing to use if we're going to use patches Aug 13 23:34:15 if u are gonna do dev work Aug 13 23:34:22 yeah Aug 13 23:34:32 destinal: git is only good for devs Aug 13 23:34:41 its not gonna make an end user happier Aug 13 23:34:53 nothing can really make patches easier for end users Aug 13 23:35:00 conflicts are conflicts Aug 13 23:35:10 neither git or quilt is gonna handle them well Aug 13 23:35:13 git a little better Aug 13 23:35:17 ubuntu updates never patch the kernel? Aug 13 23:35:18 than just plain patch Aug 13 23:35:28 egaudet: what? Aug 13 23:35:35 egaudet: the kernel is easy, you ship a new binary Aug 13 23:36:13 but the kernel is open source, so you can Aug 13 23:36:14 Linux distros never patch kernels? Aug 13 23:36:30 they do Aug 13 23:36:35 egaudet: in their own source trees Aug 13 23:36:47 egaudet: not lots of projects patching the same files, generally Aug 13 23:36:52 and their patches dont conflict Aug 13 23:36:54 or if they did Aug 13 23:37:05 someone who knows what they are doing fixes the confict manually Aug 13 23:37:12 yes so why can't we do that? Aug 13 23:37:19 egaudet: we cant Aug 13 23:37:25 quilt isnt gonna help u with that though Aug 13 23:37:34 quilt does not resolve conflicts Aug 13 23:37:39 git can to a degree Aug 13 23:37:47 but that is only when pulling other git updates Aug 13 23:37:56 i mean why can't we control patches like linux distros Aug 13 23:38:00 and even then users till need to fix oms conflcits Aug 13 23:38:10 egaudet: distros dont "control" patches Aug 13 23:38:12 egaudet: linux distros add their own patchsets on top if the kernel Aug 13 23:38:19 they can a team of devs who do shit manually Aug 13 23:38:22 and know what they are doing Aug 13 23:38:30 and what destinal said Aug 13 23:38:30 which we are analogous to? Aug 13 23:38:37 ipkg Conflicts: field will handle conflicting patches Aug 13 23:38:39 we can support our patches and our patches only Aug 13 23:38:52 hold on i can epxlain this Aug 13 23:38:55 rwhitby: still, runtime patching is probably much better Aug 13 23:38:57 i did this for the past 8 years Aug 13 23:38:59 and not have them conflict or resolve them when they do Aug 13 23:39:08 egaudet: when "distros" release their own kernel version Aug 13 23:39:18 they take a tarball of a vaniall linux kernel Aug 13 23:39:28 a released kernel from kernel.org Aug 13 23:39:33 and they apply patches to it Aug 13 23:39:41 and most of those patches dont conflcit with each other Aug 13 23:39:43 and if they do Aug 13 23:39:50 they have to manually fix them Aug 13 23:39:53 * rwhitby reads the backlog and sees monkey patching mentioned Aug 13 23:39:57 then when they ae done Aug 13 23:40:02 they create a new patch Aug 13 23:40:04 of all their work Aug 13 23:40:09 PuffTheMagic, that is what I am saying we do Aug 13 23:40:11 which is what they distrbute Aug 13 23:40:19 we "should" do Aug 13 23:40:27 that is impractical Aug 13 23:40:34 I disagree Aug 13 23:40:45 egaudet: think of it like this Aug 13 23:40:59 1 in 10000 kernel patches touch the same files Aug 13 23:41:08 there is "rarely" a conflict Aug 13 23:41:17 on the pre though Aug 13 23:41:31 many people are gonna be modifying the same code Aug 13 23:41:41 and not all those people are in this channel Aug 13 23:42:05 so that would mean i would have to make a new patch for every possible combination of other patches that touch the same code Aug 13 23:42:11 that is way too much work Aug 13 23:42:16 too bad for those people then Aug 13 23:42:19 no Aug 13 23:42:22 its too bad for us Aug 13 23:42:33 cause you are saying we need to account for all that Aug 13 23:42:36 we can be a "webos-patch" distro Aug 13 23:42:43 and people can use us or not Aug 13 23:42:48 but they can't go willy nilly combining anything with us Aug 13 23:42:48 :/ you are completly missing the issue Aug 13 23:43:12 and then we make cumulative patches only Aug 13 23:43:19 no.... Aug 13 23:43:22 bad idea Aug 13 23:43:25 for each "package" like frameworks/ Aug 13 23:43:43 you are missing the boat on this completly Aug 13 23:43:50 and i cant really explain it any better Aug 13 23:44:43 egaudet: if you want to get a real idea of the potential issue, clone the git repo i started for the pre kernel Aug 13 23:44:45 The bottom line is patches are issues because of conflicts, and if we support only our patches we control the conflicts for our patches, and people using our software can use our patches easily Aug 13 23:44:51 and try to rebase it to 2.6.31 Aug 13 23:44:55 and look at all the conflicts Aug 13 23:45:11 egaudet: that is not good enough Aug 13 23:45:19 egaudet: and each patch ipkg depends on a virtual ipkg that tracks the webos version. Aug 13 23:45:40 to unapply them all, you recursively remove that virtual ipkg Aug 13 23:45:43 rwhitby: please dont bring ipks into this Aug 13 23:45:52 and when people are using this stuff en masse, other patchers/hackers will want their stuff to work with ours so they will submit them here Aug 13 23:45:53 that will only confuse the issue even more Aug 13 23:46:07 egaudet: that is not the right assumption Aug 13 23:46:10 and as a developer Aug 13 23:46:22 egaudet: yep, just like debian developers have to comply with the debian packaging standards Aug 13 23:46:23 i dont want to maintain that many patches Aug 13 23:46:36 PuffTheMagic: then don't. let someone else do it. Aug 13 23:46:50 we have to assume that EVERYONE has a modifies userland Aug 13 23:47:02 that is why we need to avoid touching palm code Aug 13 23:47:21 let the use manage conflicts between 2 patches Aug 13 23:47:25 that is not the devs issue Aug 13 23:47:27 PuffTheMagic: if they modified userland without using a patch ipkg, then they will get a conflict and the ipkg installation will fail, leave the palm file untouched. Aug 13 23:47:57 rwhitby: even with patch ipk's there will be conflicts Aug 13 23:48:01 ipks dont solve anything Aug 13 23:48:08 they just make it easier for issue to occur Aug 13 23:48:12 "patch ipkg foo didn't install" "did you manually modify anything?" "yes" "tough, use the doctor and start again" Aug 13 23:48:23 rwhitby: no Aug 13 23:48:28 this is not something that has to do witha doctor Aug 13 23:48:37 lets say user A applied patch A Aug 13 23:48:40 a feature they want Aug 13 23:48:44 yep Aug 13 23:48:50 and then they install patch B using an ipk Aug 13 23:49:00 does the ipkg have a Conflicts: A line or not? Aug 13 23:49:03 who is to say that our patch B trumps what they arready installed Aug 13 23:49:06 PuffTheMagic, look at us as a distro and then it comes clear Aug 13 23:49:10 yes there is a conflict Aug 13 23:49:15 PuffTheMagic: did they install A using an ipkg? Aug 13 23:49:18 egaudet: but we are not Aug 13 23:49:28 that's the only way we can release patches cleanly Aug 13 23:49:31 rwhitby: it dont really matter if they used an ipk Aug 13 23:49:38 is by releasing a "distro-like" patch system Aug 13 23:49:39 rwhitby: cause it alreaduy installed Aug 13 23:49:48 where we control things Aug 13 23:50:05 egaudet: as rwhitby would say, we are not the center of the universe there Aug 13 23:50:10 PuffTheMagic: if they used an ipkg, then there is a bug in A and B ipkgs since they were not marked as conflicting. If they didn't use an ipkg for either of A or B, then they didn't meet the prerequisites and are told to use the doctor and start again. Aug 13 23:50:28 you don't have to be the center of the universe to release a well supported distro Aug 13 23:50:37 bzhou * r10556 optware/trunk/make/mkvtoolnix.mk - mkvtoolnix: 2.9.5 -> 2.9.8 Aug 13 23:50:38 rwhitby: even if they both use ipk, and there is a conflict Aug 13 23:50:45 that is the whole point of what im saying Aug 13 23:50:53 there WILL be conflcts galore Aug 13 23:50:59 PuffTheMagic: yep, they have to choose one or the other Aug 13 23:51:01 and that will require manual interaction Aug 13 23:51:03 no Aug 13 23:51:09 there does NOT have to be conflicts Aug 13 23:51:11 they can merge the two manually Aug 13 23:51:21 egaudet: there will nbe Aug 13 23:51:22 be Aug 13 23:51:27 there only has to be conflicts when patch developers do things with a blind eye on everything and everyone else Aug 13 23:51:31 PuffTheMagic: users can't merge patches Aug 13 23:51:37 if we say "webos-internals patches only" Aug 13 23:51:42 rwhitby: LOL Aug 13 23:51:52 egaudet: that is anti everything we stand for Aug 13 23:51:59 HOW? Aug 13 23:52:03 and im surprised rwhitby is not behind me on this Aug 13 23:52:15 users can install ipkgs, and if they ipkgs are marked correctly with Conflicts: and Depends: then everything works. Aug 13 23:52:21 patches only work with a central control system, we either create one or someone else will Aug 13 23:52:22 period Aug 13 23:52:28 rwhitby: no Aug 13 23:52:39 those users who access Linux manually to apply patches manually have to deal with manually resolving conflicts Aug 13 23:52:45 a patch conflict is not the same as a package conflict Aug 13 23:52:54 a patch conflict means the code touches similar areas Aug 13 23:52:59 PuffTheMagic: patch conflicts are recorded as package conflicts Aug 13 23:53:00 it does not mean the code is incompatable Aug 13 23:53:13 it just means patch isnt smart enough to merge all the chages Aug 13 23:53:27 a "user" can resolve this conflict manually most of the time Aug 13 23:53:40 a package conflict literally means package A breaks package B Aug 13 23:53:40 should I follow the applying patches wiki to a T Aug 13 23:53:43 http://www.webos-internals.org/wiki/Applying_Patches Aug 13 23:53:47 PuffTheMagic: so what's the alternative that you have in mind for all this? and saying "users aren't l33t enough to have patches" is not a good enough answer. Aug 13 23:53:54 i want to apply my patch now that i reverted to clean webos image Aug 13 23:53:57 for frameworks/ Aug 13 23:54:17 rwhitby, said it best Aug 13 23:54:30 rwhitby: no i am saying, we should not package diffs in ipks, and leave patches to plam code up to the user to apply manually them selves Aug 13 23:54:34 say I have a linux distro, and the distro maintainers are controlling their patches and conflicts Aug 13 23:54:38 PuffTheMagic: seriously, what's your alternative that meets the requirements of putting patches into the hands of aunt minnie in a safe way? Aug 13 23:54:47 well if i go by myself and hack the kernel... tough shit, now it's on me, not on the distro Aug 13 23:54:52 i dont give a shit about aunt minnie Aug 13 23:54:59 i've said that 10000 times Aug 13 23:55:03 patches are advanced Aug 13 23:55:09 they are not for the masses Aug 13 23:55:13 PuffTheMagic: sorry, but you're just too elitist. Aug 13 23:55:21 bzhou * r10557 optware/trunk/make/libpng.mk - libpng: 1.2.38 -> 1.2.39 Aug 13 23:55:24 yea PuffTheMagic come on now Aug 13 23:55:25 this isnt elitist Aug 13 23:55:29 PuffTheMagic: they are *already* being applied by the masses. we're making it safer. Aug 13 23:55:39 we are not making it safer Aug 13 23:55:50 we are saying package A is better than package B cause its already installed Aug 13 23:56:04 you can't stop it PuffTheMagic, you can sit back and say screw everyone who doesn't know, or you can help create a system that makes it easier for them Aug 13 23:56:04 and we are implying that package A and B cant be used together when they really can be Aug 13 23:56:13 thus providing an avenue for devs code to reach broader audiences Aug 13 23:56:15 sure we are. we are providing a package mechanism that is safe for aunt minnie, cause it records patch conflicts as package conflicts. Aug 13 23:56:29 rwhitby: but it is giving preference to things Aug 13 23:56:33 and that is not fair Aug 13 23:56:57 PuffTheMagic: anyone can refine the two patches to work together, or create a new patch that as a preference Aug 13 23:57:17 it doesn't have to be fair, if anyone can create a new patch or patches that fix the conflicts Aug 13 23:57:22 rwhitby: why should developers worry about that, they already did enough work creating something that people want Aug 13 23:57:29 market forces then determine which patches are available Aug 13 23:57:30 why do they need to care abou what every other dev does? Aug 13 23:57:56 PuffTheMagic: why are you so selfish? Just don't participate. Let someone else do it. Aug 13 23:58:01 im not being selfish Aug 13 23:58:04 im trying to be fair Aug 13 23:58:16 a patch can fail to apply with out it really conflicting with existing code Aug 13 23:58:19 example Aug 13 23:58:25 PuffTheMagic: so what is your proposal for a fair system of applying patches for aunt minnie? Aug 13 23:58:36 patch one insersts a new function at the end of the file Aug 13 23:58:37 rwhitby: runtime patching, patch in memory Aug 13 23:58:49 patch two insets the a different function at the end of the file Aug 13 23:58:54 has this page been updated to use quilt-lite? http://www.webos-internals.org/wiki/Next_steps Aug 13 23:58:57 the functions dont conflicts Aug 13 23:58:59 destinal: right - if someone works out how to do that without palm copyright problems, I'm all ears. Aug 13 23:59:02 but the patchest will fai Aug 13 23:59:03 fail Aug 13 23:59:12 just ebcause they get inseted in the same spot Aug 13 23:59:21 PuffTheMagic, again you are proving what happens with uncooperative patches Aug 13 23:59:27 egaudet: no Aug 13 23:59:30 PuffTheMagic: right, so that is identified and the two patch ipks are marked as conflicting Aug 13 23:59:32 that is real life Aug 13 23:59:36 when dealing with pathces Aug 13 23:59:45 patch A and patch B cannot be created and released at the same instant in time Aug 13 23:59:50 rwhitby: they are not conflicting Aug 14 00:00:14 now in my idea, patch A and patch B would be resolved by webos-internals BEFORE being released Aug 14 00:00:14 there's another way to deal with this also Aug 14 00:00:14 they just "fail to apply" Aug 14 00:00:14 there is a HUGE difference Aug 14 00:00:39 treat a patch collection as an app Aug 14 00:00:44 egaudet: forget webosinternals, what about all the other non webosinternals devs that make a patch tht we cant control Aug 14 00:00:51 PuffTheMagic, what you seem to be talking about is the inability to control a global package system for patches that all devs outside of webosinternals can work with Aug 14 00:00:53 so someone says, I totally want to do x y and z Aug 14 00:00:55 and you are right, it's not possible Aug 14 00:01:10 are u saying i need to search the internet for every patch that exists and maintain patches for everything? Aug 14 00:01:11 rather than downloading patches for x, y, and z, download NovaDialer which is a patchset Aug 14 00:01:18 * rwhitby has a meeting now - please "show me the code" - write a wiki page or create a full workflow that replaces the quilt workflow that we have now (which can be packaged in ipkg files as I have described) Aug 14 00:01:20 why should I the dev go thought that much work Aug 14 00:01:23 then you have centralized management Aug 14 00:01:27 PuffTheMagic, the same issue with desktop linux. Can't reach Aunt Minnie until you split off into distros and control things Aug 14 00:01:36 and the NovaDialer maintainer doesn't have conflicts Aug 14 00:01:58 If you want to apply patches from webosinternals, supermegapre.com and joe-schmoe-hack-a-man... then YOU have to know what you are doing Aug 14 00:02:11 if you want to install just webosinternals patches.. then WE have resolved all conflicts within our own stuff Aug 14 00:02:22 Not expert enough to participate in this conversation, but would anyone mind helping me troubleshoot a ddns problem I'm having in a private chat? Aug 14 00:02:42 Devs will WANT to push their stuff into a central control system that helps get their stuff to the masses Aug 14 00:02:48 egaudet: +1 Aug 14 00:03:00 egaudet: but why do u think people are gonna pick our management system over another Aug 14 00:03:04 so joe-schmoe-hack-a-man will start submitting to webosinternals Aug 14 00:03:19 because we currently have the foothold Aug 14 00:03:24 no we dont Aug 14 00:03:33 who does? Aug 14 00:03:43 does the more_icons_per_row patch need add-delete-pages-in-the-launcher to be removed? Aug 14 00:03:46 precentral has a shit load of ipks for apps Aug 14 00:03:49 we dont have shit yet Aug 14 00:04:02 precentral doesn't know how to handle patches Aug 14 00:04:04 APPS are separate Aug 14 00:04:07 forget apps Aug 14 00:04:12 anyone? Aug 14 00:04:14 or services or plugins Aug 14 00:04:22 Apps do not, cannot and shall not depend on patches Aug 14 00:04:26 commandercup: I believe they are orthogonal now Aug 14 00:04:28 commandercup: I would remove it b/c the add/delete pages is built into the more icons per row patch Aug 14 00:04:39 and if they do... it's a rare case Aug 14 00:04:41 commandercup: Decimation fixed that dependency Aug 14 00:04:48 so the patch in the repo is updated? Aug 14 00:04:59 * PuffTheMagic give up Aug 14 00:05:02 do what ever you guys wany Aug 14 00:05:02 but that's besides the point, precentral has the largest homebrew apps Aug 14 00:05:04 want Aug 14 00:05:05 commandercup: dunno - ask Decimation or look at the git log Aug 14 00:05:12 ok Aug 14 00:05:13 thanks Aug 14 00:05:19 PuffTheMagic: seriously, put forward an alternate proposal Aug 14 00:05:31 i have, dont put ipks in patches Aug 14 00:05:35 Is there a way to reboot or shut off the pre without taking out the batttery? Aug 14 00:05:36 that is my proposal Aug 14 00:05:46 Ok PuffTheMagic switch topics Aug 14 00:05:47 leave patches to people who know what they are doing Aug 14 00:05:53 hello Aug 14 00:05:58 get me set up on git'ing correctly from my pre Aug 14 00:06:00 i am all for stand alone apps as ipks Aug 14 00:06:04 patches are just a bad idea Aug 14 00:06:04 PuffTheMagic: that's a "don't do it cause you're not l33t enough" proposal. not good enough. Aug 14 00:06:07 so i dont have to revert with webos image again! Aug 14 00:06:34 rwhitby: this isnt about being l33t this is about, patches are gonna cause more problems than the solve in the long run Aug 14 00:06:42 egaudet: the Applying_Patches page shows how to do that Aug 14 00:06:42 i dont want to get involved with them Aug 14 00:06:47 PuffTheMagic: then don't Aug 14 00:06:48 what's up Aug 14 00:06:55 PuffTheMagic: let those who do care do it Aug 14 00:07:07 rwhitby, i want to track my whole rootfs/ Aug 14 00:07:19 egaudet: optware has stgit if you want to go down that route Aug 14 00:07:19 but you solution is l33t to Aug 14 00:07:22 so i can always have a tracking system on what Im editting Aug 14 00:07:27 fuck everyone elses patches Aug 14 00:07:28 egaudet: stgit Aug 14 00:07:36 they are no good enough if they are no on webosinternlas Aug 14 00:07:39 that is what u are saying Aug 14 00:07:44 stgit wont help anything Aug 14 00:07:50 no PuffTheMagic, it's not fuck everyone elses Aug 14 00:07:52 PuffTheMagic: everyone can put their patches in webosinternals repo Aug 14 00:08:01 rwhitby: that is l33t Aug 14 00:08:09 nope, there is no barrier Aug 14 00:08:13 it's sorry we are incapable of supporting every possibility in the universe so we can only officially support what we control Aug 14 00:08:18 everything else is a risk to not be compatible Aug 14 00:08:39 egaudet: stgit is what you want for the scenario on your Pre that you want Aug 14 00:08:55 Anybody mind helping me troubleshoot ddns problem? In private chat is fine. Aug 14 00:08:57 no stgit does not help Aug 14 00:09:03 its a quilt wrapper Aug 14 00:09:04 thats it Aug 14 00:09:11 does not solve any issues Aug 14 00:09:23 PuffTheMagic: which of the two conversations are you responding to? Aug 14 00:09:33 how do you remove a patch? Aug 14 00:09:35 forget it, I'm creating my own git repos on my pc Aug 14 00:09:37 idk i thought there was 1 Aug 14 00:09:48 PuffTheMagic: there's the problem. there are two conversations happening Aug 14 00:10:01 egaudet: stgit works fine on the pre Aug 14 00:10:04 or the emulator Aug 14 00:10:06 is it webOS.tar that is the root image? Aug 14 00:10:22 egaudet: stgit does not help you though Aug 14 00:10:23 egaudet: it's in a tar inside that Aug 14 00:10:28 you are not working with a series Aug 14 00:10:43 I quit quilt Aug 14 00:10:44 its just at much work to type patch < Aug 14 00:10:59 egaudet: stgit or guilt will not help yout Aug 14 00:11:06 I can't stand the stupid pre for developing Aug 14 00:11:11 they are for applying a series of patches Aug 14 00:11:13 I am gitting the webos image locally on my pc Aug 14 00:11:20 * rwhitby bbl Aug 14 00:11:22 i will edit there and then mv the directory to the pre for testing Aug 14 00:11:30 egaudet: thats why i suggesting tracking your root with git Aug 14 00:11:31 commit and patch locally and get it up to webos Aug 14 00:11:43 egaudet: but you should really do dev work in the emulator Aug 14 00:11:46 and not your pre Aug 14 00:11:55 PuffTheMagic, I can't because the files are different Aug 14 00:12:02 the frameworks doesn't have the same structure Aug 14 00:12:08 egaudet: they shouldnt be Aug 14 00:12:14 emu has 1.1.0 Aug 14 00:12:18 my pre has 1.1.0 Aug 14 00:12:27 you are talking about /opt/PalmSDK/Current right? Aug 14 00:12:31 what are you hacking in the frame work anyway? Aug 14 00:12:37 the onscreen keyboard Aug 14 00:12:50 PuffTheMagic: have you ever used stgit? Aug 14 00:12:54 yes i have Aug 14 00:12:59 i have used stgit and guilt Aug 14 00:13:09 they are quilt wrappers Aug 14 00:13:13 they manage series files Aug 14 00:13:20 stgit is not a quilt wrapper Aug 14 00:13:31 it is essentially Aug 14 00:13:37 i have 100 patches and a series file Aug 14 00:13:42 i want to apply them to my git tree Aug 14 00:13:44 that's a very narrow view of stgit Aug 14 00:13:46 and i want nice commit messages Aug 14 00:13:50 stgit does that Aug 14 00:13:51 so goes Aug 14 00:13:52 guilt Aug 14 00:14:04 that is not gonna solve egaudet's problme Aug 14 00:14:09 his problem is the same as mine Aug 14 00:14:15 we forget to use quilt edit Aug 14 00:14:18 anyone willing to tell me how to remove a patch? Aug 14 00:14:24 so we edit files and they dont get tracked Aug 14 00:14:27 commandercup: patch -R Aug 14 00:14:30 ok tys Aug 14 00:14:31 reverts a patch Aug 14 00:14:32 ty* Aug 14 00:14:44 DDNS troubleshooting? Should be an easy fix...anybody? Aug 14 00:14:44 PuffTheMagic: stgit tracks all the files using git internals Aug 14 00:14:59 no git is tracking all the file Aug 14 00:15:04 stgit isnt doing anything Aug 14 00:15:05 its all git Aug 14 00:15:06 how do i create a repo with git Aug 14 00:15:12 egaudet: see pm Aug 14 00:16:01 PuffTheMagic: exactly, it's git, not a quilt wrapper Aug 14 00:17:17 thats not what i meant either Aug 14 00:17:25 egaudet: had an issue cause he was editing files Aug 14 00:17:29 PuffTheMagic: if you use git on the phone, and want to track any file under /, you will have a .git directory which will fill / Aug 14 00:17:34 and they there were not being tracked by quilt Aug 14 00:17:52 PuffTheMagic: right. stgit will track them, cause it's built on git, not quilt. Aug 14 00:18:02 rwhitby: i suggested last night putting the .git dir on media-interlal Aug 14 00:18:08 that way / does not get filled Aug 14 00:18:18 PuffTheMagic: .git dir works on vfat? Aug 14 00:18:24 now the question is Aug 14 00:18:35 i beleive so Aug 14 00:18:42 i dont see why not Aug 14 00:18:54 you dont need xattr or quota for .git Aug 14 00:19:09 what about symlinks? Aug 14 00:19:12 or hard links Aug 14 00:21:02 well then put it in /opt or something Aug 14 00:21:06 or use the emu Aug 14 00:21:07 if .git can work fine on vfat, then patch developers can just use stgit and still export the patches for users to apply using the simpler apply-patches workflow Aug 14 00:21:22 why would patch devs use stgit Aug 14 00:21:32 its not gonna give them anything Aug 14 00:22:06 stgit only add functionality if you have a series files and a bunch of patches Aug 14 00:22:10 if you are just editing files Aug 14 00:22:13 ok, please write up the workflow for patch developers that uses plain git, and provides patches exported in a form that a user can apply them without using git. Aug 14 00:22:13 and craeting diffs Aug 14 00:22:16 you dont need stgit Aug 14 00:22:57 you need to look at the whole picture to determine the workflow, not just the developer side of things. Aug 14 00:23:18 you continually talk in terms of tools, not workflows. Aug 14 00:23:31 how big should wallpapers be? Aug 14 00:23:33 git add .; git commit -a -m "initial seed"; start editing some files.... git add .; git commit -a; git show > /tmp/somechanges.diff Aug 14 00:23:34 or Aug 14 00:23:38 look in git log Aug 14 00:23:42 get the commit id Aug 14 00:23:57 then do git show > /tmp/somefile.patch Aug 14 00:23:59 or Aug 14 00:24:07 if ou want to roll up a bunch of patches Aug 14 00:24:39 git diff .. > /tmp/another.path Aug 14 00:24:40 patch Aug 14 00:24:48 and how do you refresh a patch? Aug 14 00:25:05 that is no something "git" does Aug 14 00:25:11 you can merge patches toether Aug 14 00:25:13 i mean Aug 14 00:25:16 merge "commits" Aug 14 00:25:18 again, talk about the complete workflow lifecycle of patches, not just one part of it. Aug 14 00:25:38 ok i have a clean, pure, beautiful rootfs git commit Aug 14 00:25:43 rwhitby: git is too complex to be forced into a single "patch" paridigm Aug 14 00:25:54 PuffTheMagic: what do you do with git when a webos upgrade occurs? Aug 14 00:26:06 how to do refresh a patch easily after the rebase? Aug 14 00:26:17 these are all things that stgit is designed to handle Aug 14 00:26:31 you cant mix commits and patches as the same thing Aug 14 00:26:36 you can create a patch from a commit Aug 14 00:26:41 or from a bunch of commits Aug 14 00:26:45 but they are not the same Aug 14 00:26:50 you are too wrapped up in quilt Aug 14 00:27:02 ahahaha I get it Aug 14 00:27:04 quilt is ok for the end users Aug 14 00:27:05 wrapped up in a quilt Aug 14 00:27:12 git is for developers Aug 14 00:27:13 no, I'm too wrapped up in looking at the whole picture, not just one part. Aug 14 00:27:17 i will help any dev use git Aug 14 00:27:20 to create patches Aug 14 00:27:30 rwhitby, he is right for developers Aug 14 00:27:42 well I guess I should research stgit first Aug 14 00:27:42 PuffTheMagic: as long as the patch is rooted at /, then that's fine. the interchange format is compatible. Aug 14 00:27:56 rwhitby: i've been using git for ~2-3 years and i dont know everything git can do Aug 14 00:28:21 rwhitby: yeah ... cd /. git init; git add .; git commit -a -m "initial seed" Aug 14 00:28:25 so everything looks like a nail, right? Aug 14 00:28:26 start editing files Aug 14 00:28:27 git add . Aug 14 00:28:29 git commit -a Aug 14 00:28:32 got a commit Aug 14 00:30:12 I'm happy to see a writeup of how git solves all the problems across the whole workflow and whole lifecycle of patch management Aug 14 00:30:52 including how aunt minnie applys the patches before and after a webos upgrade Aug 14 00:31:59 rwhitby: do NOT mix patch managemnt with git Aug 14 00:32:08 git is source/version control Aug 14 00:32:15 and happens to be able to create diffs Aug 14 00:32:26 quilt is patch managment Aug 14 00:32:39 PuffTheMagic: how do you distribute the modifications without violating palm copyright if you don't use patches? Aug 14 00:32:41 i dont have a "better" way for users to manually deal with patches Aug 14 00:32:47 i am only advocating git for devs Aug 14 00:33:03 rwhitby: like is said, devs can output patches Aug 14 00:33:13 but using patches sucks in general Aug 14 00:33:20 but you haven't said how they refresh them easily Aug 14 00:33:22 but considering our situation with palm Aug 14 00:33:35 rwhitby: you cant in git Aug 14 00:33:38 that is a quilt thing Aug 14 00:33:44 git can generate a new patch Aug 14 00:33:50 based on x number of commits Aug 14 00:33:55 exactly. that's why you should use stgit Aug 14 00:33:56 git is NOT a patch management system Aug 14 00:34:05 stgit wont help you either Aug 14 00:34:38 damnit i just ran out of space on /home Aug 14 00:34:43 rwhitby: see PM Aug 14 00:34:55 i have a "semi solution" for you Aug 14 00:35:01 I believe it will, and have prototyped that solution on the emulator when I ported stgit to optware Aug 14 00:35:10 just got to pm Aug 14 00:35:13 anyone use rtorrent on their Pre? Aug 14 00:40:28 z Aug 14 00:40:40 unzzzz Aug 14 00:43:08 what is the difference between the roam-control.patch and phone-enable-roam-only.patch? Aug 14 00:51:19 WillJitsu: you wuld need to ask the folks who wrote them, or compare the code. Aug 14 00:51:37 you can check the commit logs on git to see who committed them Aug 14 00:53:02 can anyone give me some help to get the onscreen keyboard working? Aug 14 00:53:15 actually, let me try a restart first, just in case Aug 14 00:53:33 what does this patch: message-tone-sounds-and-alerts.patch do? Aug 14 00:56:46 yep, restart did it. sweet Aug 14 00:56:59 :D Aug 14 00:57:22 (more) functional keyboard coming soon to a repository near you Aug 14 00:58:02 though, i guess there is no backspace, or space, or enter. but great job Aug 14 00:59:37 backspace, space, enter currently working, will have patch tonight Aug 14 00:59:46 oh, cool. Aug 14 01:02:17 commandercup: http://www.webos-internals.org/wiki/Patch_Messaging_Sounds Aug 14 01:02:27 egaudet: let us know when its up Aug 14 01:02:39 will do Aug 14 01:03:42 oh, have you guys seen the video a guy who shoots an iphone? Aug 14 01:05:23 Wow. I'm new and fair to say lost Aug 14 01:05:44 lots of names, some of which look familiar Aug 14 01:06:31 I feared this would be chaotic and I wouldn't be able to get a word in edge-wise Aug 14 01:06:53 now, I'd guess I'm alone Aug 14 01:07:19 Colonel: no, this channel is pretty good Aug 14 01:07:33 okay Aug 14 01:08:16 I'm here to find out what I need to do to submit a hack for the Calendar app Aug 14 01:08:17 in fact, all the freenode channels im in are nothing like the "stereotypicall" freenode channels, like ##c, or #scheme Aug 14 01:08:34 orange key is what alt? Aug 14 01:08:37 Colonel: check the wiki? Aug 14 01:09:18 Colonel: Applying_Patches page, second half. Aug 14 01:09:29 egaudet: orange key, if anything, more like the shift key but just for numbers, since it allow us to write symbols Aug 14 01:09:39 yeah, but I'm not a Linux guy and any instructions I'd provide would be devoid of that Aug 14 01:09:46 gkatsev, we have a shift key Aug 14 01:09:54 gkatsev, I think I sat next to you in Boston this past weekend Aug 14 01:09:59 I mean what is the orange key on the actual keyboard map to in palm's framework Aug 14 01:10:02 ctl or alt Aug 14 01:10:17 egaudet: it doesnt is my point Aug 14 01:10:39 Colonel: did you make BADDFlashlight? Aug 14 01:11:09 gkatsev, it does tho i think its alt Aug 14 01:11:22 that's me Aug 14 01:11:31 Colonel: then, yes, you did Aug 14 01:11:49 how you doing? have you seen the new site that John has set up? Aug 14 01:11:56 Colonel: yep Aug 14 01:12:05 and the twitter and irc channel Aug 14 01:13:12 I'll find and follow the twitter and look into the irc channel later Aug 14 01:13:26 webosboston Aug 14 01:14:04 rwhitby, I'll read the second half of the Applying Patches page to see what sense I can make out ot it Aug 14 01:14:24 John has a new site up? What? Aug 14 01:14:51 my suspicion is that much of it will be beyond me, so i guess I'm here to ask how I'd get help getting it legit Aug 14 01:14:55 Rick_home: jcrawford, webosboston.org Aug 14 01:15:18 sort of like submitting a drawing to an architect to get it done right Aug 14 01:16:32 I used webOS Quick Install to pull/edit/push some non-idiotic event reminder times into the Calendar app Aug 14 01:16:58 and I don't see that listed on the site so I wanted to inquire Aug 14 01:19:35 that's the nickname I meant to choose Aug 14 01:19:59 I told you I was new Aug 14 01:21:50 rwhitby, 2nd patch process? Aug 14 01:22:33 rwhitby, remove first patch and then re-patch yes? Aug 14 01:25:34 egaudet: yeah, pop -a, then add new patch. then push -a to make sure there are no conflicts Aug 14 01:25:56 bah Aug 14 01:27:36 well, I've got some reading to do, so I'll come back when I have some specific questions Aug 14 01:27:49 thanks for your time Aug 14 01:27:57 talk later Aug 14 01:29:08 rwhitby, if i do a pop -a, do i then have to re-add all the files? Aug 14 01:29:16 or simple pop -a push -a does the job Aug 14 01:29:34 egaudet: pop -a, then push -a and all the patches are back Aug 14 01:29:46 its nifty Aug 14 01:32:28 hello Aug 14 01:39:54 are you guys honestly still going on about the usefulness of git to manage patches? Aug 14 01:43:25 arrow keys in vim produce funky escape code looking artifacts Aug 14 01:43:38 in terminal Aug 14 01:44:05 t3rmin-home: create the .vimrc in your home Aug 14 01:45:17 I have... anything special I should put in it? Aug 14 01:46:32 also I'm ircing w/ BX on phone and noticed another thing Aug 14 01:46:41 t3rmin-home: you could always use h, j, k and l to move around Aug 14 01:47:19 when you type a message thatls longer than available space it scrolls funky Aug 14 01:47:31 t3rmin-home: nope, just create it and it should be fine, vim runs in compatability mode when running without a vimrc and creating it turns it off Aug 14 01:47:33 and you loose status bar Aug 14 01:48:14 gkatsev: I'm actually already on nocompat Aug 14 01:48:17 Has anyone else attempted to change colors? Aug 14 01:48:31 it's locking up the terminal for me Aug 14 01:48:50 JackieRipper: yeah, for me it locked up the whole device Aug 14 01:48:51 t3rmin-home: and you are still getting the letters when using arrow buttons? Aug 14 01:49:14 gkatsev: yup Aug 14 01:49:27 TERM=linux btw Aug 14 01:49:34 t3rmin-home: is this when you're in insert mode? Aug 14 01:49:47 t3rmin-home: try #vim, they can help you more. Aug 14 01:49:52 JackieRipper: nope Aug 14 01:50:04 but for me i got that problem till i created my vimrc Aug 14 01:50:34 gkatsev: interesting. what's your TERM? Aug 14 01:51:09 I've never had that problem except in insert mode Aug 14 01:51:27 t3rmin-home: vt100 Aug 14 01:51:37 you guys have colors/syntax on? Aug 14 01:52:05 I do Aug 14 01:52:22 i didnt do anything like that... Aug 14 01:52:29 so, i guess not Aug 14 01:52:47 I don't have a .vimrc either Aug 14 01:53:09 JackieRipper: and your $TERM? Aug 14 01:53:15 vt100 Aug 14 01:53:18 do you get the letters when using arrows? Aug 14 01:53:43 ok Aug 14 01:53:50 it's the TERM Aug 14 01:54:07 switched to vt100 and it's good Aug 14 01:54:11 ok Aug 14 01:54:14 no color tho Aug 14 01:55:24 more ansi quirks I guess Aug 14 01:55:52 t3rmin-home: have you done anything special to your profile? Aug 14 01:56:03 still, BX on the phone is pretty nifty :) Aug 14 01:56:37 JackieRipper: just removed hardset TERM and let it auto set Aug 14 01:56:59 that way it's happier when I ssh in from an xterm Aug 14 01:58:23 brb gonna try BX in vt100 Aug 14 02:00:06 well unlike vim BX still has color Aug 14 02:00:31 line length wrap issue still evident tho Aug 14 02:01:53 did anyone catch what geist said earlier about the kernel logs for crashes - where they get dumped during a soft reboot ? Aug 14 02:15:01 JackieRipper: yes, we're still working out an improved workflow for patches. Aug 14 02:15:27 any improvements? Aug 14 02:15:48 er ideas? Aug 14 02:15:53 no complete proposals yet Aug 14 02:17:22 hey guys hows everything going Aug 14 02:17:31 I've got ideas about ipkgs. PuffTheMagic has serious concerns about them. PuffTheMagic doesn't want to consider the case of aunt minnie installing patches from a repository using something like filecoaster. Aug 14 02:19:10 I still don't understand the objection to git... yes we're underutilizing the tool, but its primary purpose is revision control of files, and that's all we're using it for Aug 14 02:20:56 JackieRipper: no objection from me. it's just that we can't host git repos with Palm files in them. Aug 14 02:21:10 we can only distribute patches Aug 14 02:21:50 and git merges are way too complex for an end user. I'm scared of git merges myself. Aug 14 02:22:06 right... but patches are files... we're not looking using git to create patches, simply to control the revisions Aug 14 02:22:23 rwhitby, developers create patches Aug 14 02:22:47 we need to coordinate "patch" developers to using a fully patched git tree locally that cannot be pushed up Aug 14 02:23:02 developers create FILES. We have a repository to store those files. Aug 14 02:23:08 that the files are patches makes no difference. Aug 14 02:23:10 They're files Aug 14 02:23:13 we store them. Aug 14 02:23:13 when a patch is released by another developer, you must patch your git tree and check it in before you work on those files Aug 14 02:23:20 egaudet: JackieRipper, egaudet, Rick_home: I think we're all agreeing. Aug 14 02:23:32 If they contain an error, they need to be "patched" -- git is good for this. Aug 14 02:23:33 patches are the interchange format Aug 14 02:23:36 nodnod Aug 14 02:23:47 developers create patches. users don't create patches Aug 14 02:23:51 It's difficult situation for sure since we cannot push our working trees up Aug 14 02:24:33 developers in this case may be people who have never used an SCM before (otherwise we don't get a large enough base of people contributing patches) Aug 14 02:25:00 egaudet: later tonight is when for you? (just wondering when the updated onscreen keyboard patch is available) Aug 14 02:25:11 developers need to be able to manage and refresh (or regenerate) patches across webos upgrades. Aug 14 02:25:30 users need to be able to unapply all patches before an upgrade, and reapply them afterwards. Aug 14 02:25:33 as soon as i get this patch created gkatsev Aug 14 02:25:35 it's ready Aug 14 02:25:40 i just have a screwed up tree and quilt right now Aug 14 02:26:02 egaudet: heh, cool. Aug 14 02:26:52 egaudet: any tool (or set of tools) which meets all those requirements is fine by me. the simplest one I found so far is quilt, but I'm sure that people smarter than me will find a simpler way, as long as they concentrate on the whole workflow and lifecycle and use cases, and not just one part of it. Aug 14 02:27:24 quilt doesn't work for developing though, since it doesn't remember anything until you do quilt add Aug 14 02:27:45 I think we -need- git for this purpose Aug 14 02:27:58 let's assume down the road we have 85 patches in the stream Aug 14 02:28:12 egaudet: I'm happy to be convinced of that. As long as the developer can develop and manage a patch on the Pre. Aug 14 02:28:22 all the developers need a way to know what they are working on is going against the same tree Aug 14 02:28:38 (since the webos doctor lags the Pre updates, and developers need to get updated patches out quickly after a pre update) Aug 14 02:28:42 A developer does not -have- to develop on the pre, but of course they can Aug 14 02:29:00 what do you mean webos doctor lags the updates? Aug 14 02:29:07 our workflow *must* support developers refreshing (or recreating) patches on the Pre after a webos upgrade. Aug 14 02:29:10 did we have a win32 tool to extract .ipk files Aug 14 02:29:25 wouldnt 7-zip be able to handle it? Aug 14 02:29:28 egaudet: the webos doctor for webos version N lags the Pre update to webos version N by days Aug 14 02:29:41 the emulator may too. Aug 14 02:29:47 (in the last case it didn't) Aug 14 02:29:56 7zip extracts ipk's just fine Aug 14 02:30:00 im confused Aug 14 02:30:30 but someone tell me quickly how to get a git diff of local changes Aug 14 02:30:32 egaudet: yes, using the emulator is the best way to develop patches, but our workflow must support refreshing patches on the Pre. Aug 14 02:30:56 the Pre part is purely testing/user area Aug 14 02:31:26 You can use the pre, or your pc or wherever you want as a developer to maintain the git tree of the rootfs Aug 14 02:31:34 rwhitby: I truely am coming to the conclusion that the correct workflow for preparing DISTRIBUTIONS of patches is a desktop too. Aug 14 02:31:37 tool Aug 14 02:31:51 bzhou * r10558 optware/trunk/make/vsftpd.mk - vsftpd: 2.1.2 -> 2.2.0 Aug 14 02:31:53 egaudet: I'm saying that our workflow must support using the Pre to maintain the patches. Aug 14 02:31:54 safe your files to your desktop Aug 14 02:32:00 crap, the webos doctor rootfs image is not version 1.1.0 is it Aug 14 02:32:03 (in addition to the normal case of using emulator or desktop) Aug 14 02:32:10 edit your files. Aug 14 02:32:17 copy the edited files to the desktop Aug 14 02:32:27 run a diffing tool that produces a diff AND the script to install it Aug 14 02:32:47 There is a webosDr that is 1.1 Aug 14 02:32:51 Rick_home: that's already 5 too many steps Aug 14 02:33:00 ok Aug 14 02:33:33 how do i know if i have webosDR 1.1? Aug 14 02:34:00 what file do you need, I'll email it Aug 14 02:34:09 egaudet: check the md5sum on our webos doctor versions wiki page Aug 14 02:34:14 the whole rootfs tree lol Aug 14 02:34:23 i need to make sure that is the base of my git tree Aug 14 02:35:02 where on wiki Aug 14 02:35:38 ohh winrar coudln't, Aug 14 02:35:45 * Robi_ tries 7-zip Aug 14 02:36:38 http://www.webos-internals.org/wiki/Webos_Doctor_Versions, first hit for searching on "webos doctor versions ;-) Aug 14 02:38:06 k good Aug 14 02:40:38 KEYBOARD patch pushed Aug 14 02:41:08 differences? Aug 14 02:41:11 egaudet: Aug 14 02:43:00 just added space, backspace and enter Aug 14 02:43:52 I wasted just about the entire day yesterday fighting with enter because I kept trying to trigger the event to the scene rather than the text target :( Aug 14 02:45:51 gkatsev Aug 14 02:45:59 If anyone tests the patch let me know Aug 14 02:46:23 egaudet: rebooting right now Aug 14 02:46:30 :D Aug 14 02:46:48 egaudet: I'll test it on the emulator now - how do you get the keyboard to show on the emulator? Aug 14 02:46:55 Who wants to make a "restart Luna" service & app? :) Aug 14 02:47:04 can you simulate a gesture tap? Aug 14 02:47:44 im still seeing the old keyboard, let me try again Aug 14 02:48:13 remove first patch first? Aug 14 02:48:42 egaudet: nope, I don't think you can do gestures on the emulator Aug 14 02:49:09 well I dont even think you can use the patch for the emulator Aug 14 02:49:10 perhaps we need a real keyboard way of opening the virtual keyboard ;-) Aug 14 02:49:27 egaudet: why not (apart from not being able to do a gesture) ? Aug 14 02:49:36 rwhitby, i dont think the framework structure is the same Aug 14 02:49:48 really? that would be news to me. Aug 14 02:49:48 the emulator uses /opt/PalmSDK/share right? Aug 14 02:49:53 no Aug 14 02:50:04 the emulator has a rootfs disk image Aug 14 02:50:05 ok I haven't even used the emulator so tell me Aug 14 02:50:20 because i really should be dev'ing on the emulator if that's the case Aug 14 02:50:27 err testing not dev'ing Aug 14 02:50:29 the emulator is basically the same rootfs as on the Pre, except binaries are compiled for i686 instead of arm7. Aug 14 02:51:15 egaudet: I can tell you whether the patch applies or not in 30 seconds Aug 14 02:51:21 ok Aug 14 02:52:10 egaudet: patch format is slightly different - our process requires patches rooted at / - yours is rooted at a and b Aug 14 02:52:20 damnit Aug 14 02:52:38 something weird happened to my phone yesterday Aug 14 02:52:59 my phone screen was off but the little button on the gesture area was lit up Aug 14 02:53:08 i had a notification too when i turned it on Aug 14 02:53:17 * rwhitby installs emacs on the emulator to do a search and replace on the patch ... Aug 14 02:53:18 but then later i got another notification and it was off Aug 14 02:53:28 maybe just a coincidence Aug 14 02:55:01 diff --git usr/palm/frameworks/mojo/builtins/palmInitFramework191_15.js usr/palm/frameworks/mojo/builtins/palmInitFramework191_15.js Aug 14 02:55:01 index 4307893..97bd500 100644 Aug 14 02:55:01 --- usr/palm/frameworks/mojo/builtins/palmInitFramework191_15.js Aug 14 02:55:01 +++ usr/palm/frameworks/mojo/builtins/palmInitFramework191_15.js Aug 14 02:55:07 rwhitby, is that how it should look Aug 14 02:55:43 --- .orig/usr/palm/applications/com.palm.app.camera/app/controllers/capture-assistant.js Aug 14 02:55:44 +++ /usr/palm/applications/com.palm.app.camera/app/controllers/capture-assistant.js Aug 14 02:56:02 your format may well work just as well Aug 14 02:56:14 I think it does Aug 14 02:56:20 try it Aug 14 02:56:25 should do, as long as patch is run from / Aug 14 02:57:03 yea it will work Aug 14 02:59:04 hmm - emulator has palmInitFramework191_14.js, patch is looking for /palmInitFramework191_15.js **** ENDING LOGGING AT Fri Aug 14 02:59:57 2009