**** BEGIN LOGGING AT Sun Jan 22 02:59:57 2012
Jan 22 03:23:18 <dtzWill> PuffTheMagic: ping me if/when you have cmatrix committed please. no hurry but don't wanna miss it.
Jan 22 03:24:09 <PuffTheMagic> oh shit forgot about that
Jan 22 03:25:04 <PuffTheMagic> dtzWill, http://ompldr.org/vY2RybA
Jan 22 03:25:08 <PuffTheMagic> there is the arm binary
Jan 22 03:25:13 <dtzWill> ty
Jan 22 03:25:56 <PuffTheMagic> i am in the middle of shit
Jan 22 03:27:32 <dtzWill> sure thing, and ty. gl with your stuff :)
Jan 22 03:27:41 <PuffTheMagic> i got the parsing working
Jan 22 03:27:51 <PuffTheMagic> adding the enyo stuff to set it now
Jan 22 03:28:07 <PuffTheMagic> i figure this will be useful for when I connect Tide to wterm
Jan 22 03:28:11 <PuffTheMagic> and dont need to run bash
Jan 22 03:28:17 <PuffTheMagic> just need to execute a make file
Jan 22 03:28:37 <PuffTheMagic> or... when i want to run cmatrix in exhibition mode ;)
Jan 22 03:42:11 <PuffTheMagic> dtzWill: did u push your changes for building on device
Jan 22 03:42:32 <dtzWill> oh it's in the 'native' branch and not much effort for generalization was there
Jan 22 03:42:50 <dtzWill> https://github.com/dtzWill/wTerm/tree/native fwiw
Jan 22 03:43:00 <PuffTheMagic> would u pull request it
Jan 22 03:44:16 <dtzWill> done
Jan 22 03:46:15 <PuffTheMagic> bummer, i have no ext3fs
Jan 22 03:46:39 <dtzWill> oh well i just wanted it to be on ext3fs for the purpose of things like to have symlinks and such
Jan 22 03:46:43 <dtzWill> but don't know that the includes do
Jan 22 03:46:47 <dtzWill> and that's all that's needed
Jan 22 03:48:20 <PuffTheMagic> since gcc comes from opt i think  better loc mght be /opt/local/...
Jan 22 03:48:37 <dwc-> what do you have symlinks to, just out of curiousity?
Jan 22 03:48:53 <PuffTheMagic> ^^ it was an if
Jan 22 03:49:09 <dwc-> ok, what do you have in /media/ext3fs/dev
Jan 22 03:49:14 <dtzWill> dwc-: includes doesn't need anything afaik, so nothing. i meant that in general i tend to put my build stuff onto ext3 (vs fat) so it can have symlinks, and the big user there is for liibraries
Jan 22 03:49:33 <dtzWill> but since all the libraries we need are already in /usr/lib or whatnot, really no need for it.
Jan 22 03:49:39 <dwc-> since ROOT=.../dev I assume that's still being used for the location of lib/ and such that it's looking for things
Jan 22 03:49:55 <dtzWill> oh, i'm not looking there.  it's probably added to the library search path, yes.
Jan 22 03:50:28 <dtzWill> honestly it's a silly little patch that puff seemed interested in having on his radar.  what makes sense for a general-purpose on-device dev environment might be different :)
Jan 22 03:50:42 <dwc-> ahh. I don't have my phone handy, so I couldn't check mine... thought maybe palm didn't ship the library headers and you had to add them
Jan 22 03:50:53 <dtzWill> dwc-: oh. yes, that's exactly it, sorry.
Jan 22 03:51:28 <PuffTheMagic> i wanted to get setup to dev on my tp
Jan 22 03:52:23 <PuffTheMagic> but i dont have a ded8cated ext3fs
Jan 22 03:52:37 <dtzWill> then put the include stuff on /media/internal then
Jan 22 03:52:42 <PuffTheMagic> i do have an opt though
Jan 22 03:53:39 <dwc-> so palm didn't include them? maybe we should package up the PDK headers etc
Jan 22 03:53:47 <PuffTheMagic> i know i can do manynthings, i guess im just trying to get input for an official loc that is good for all
Jan 22 03:53:53 <dwc-> the ones that we can redist anyways
Jan 22 03:54:43 <dwc-> actually, I guess if we can't we might just have to tell people where to put it and they'll have to do it themself
Jan 22 03:54:55 <PuffTheMagic> actually
Jan 22 03:55:15 <PuffTheMagic> we could make an ipk of build.git's staging dir
Jan 22 03:55:47 <PuffTheMagic> which could be nstalled on cryptofs
Jan 22 03:55:49 <dtzWill> could we? doesn't that involve artifcats pulled from doctors, etc?
Jan 22 03:55:59 <PuffTheMagic> nope
Jan 22 03:56:02 <dtzWill> we might need to go metadoctor style and automate it
Jan 22 03:56:05 <dtzWill> oh? mmkay then.
Jan 22 03:56:09 <PuffTheMagic> its all fro  source
Jan 22 03:56:20 <dwc-> it is?
Jan 22 03:56:27 <PuffTheMagic> ya
Jan 22 03:56:50 <dwc-> I could have sworn I ended up with a mac doctor
Jan 22 03:56:56 <PuffTheMagic> build.git yes, cross-compile.. idk
Jan 22 03:57:08 <dwc-> oh
Jan 22 03:57:22 <dwc-> yea, my bad
Jan 22 03:58:24 <dwc-> btw, puff, if I have a directory (~/src/webos) full of git repos, is there an easy way to update them all... svn, I'd do svn up * and it'd update all of them, even if they had different repos
Jan 22 03:58:42 <dwc-> best I've been able to find is for foo in * ; do (cd $foo && git pull) ; done
Jan 22 03:58:51 <dtzWill> dwc-: afaik that's the best too :/
Jan 22 03:59:08 <PuffTheMagic> ugly
Jan 22 03:59:33 <PuffTheMagic> what happens whwn u have conflicts
Jan 22 04:00:04 <dwc-> it would fail to update that one and go on, ideally
Jan 22 04:00:52 <dwc-> I don't need the interactive conflict resolution to show up immediately, tbh, I can go back and do those by hand or whatever.
Jan 22 04:00:56 <dwc-> that's what svn does anyways
Jan 22 04:02:22 <PuffTheMagic> i cant believe i eeded to use strtok in c++
Jan 22 04:02:41 <PuffTheMagic> what a useless lang
Jan 22 04:04:31 <dwc-> looks like boost is one solution
Jan 22 04:05:03 <dwc-> I see someone else using an interator<string>
Jan 22 04:05:14 <PuffTheMagic> im not adding boost to build.git
Jan 22 04:06:43 <dwc-> http://stackoverflow.com/questions/236129/how-to-split-a-string-in-c
Jan 22 04:06:49 <PuffTheMagic> its already retarded enough i need to recreate what gentoo/portage already does for each new staging dep
Jan 22 04:07:09 <dwc-> why don't you just replace build.git then with ports/portage
Jan 22 04:07:12 <PuffTheMagic> dwc-: i know how to do it in c
Jan 22 04:07:25 <dwc-> read the link, it's for c++
Jan 22 04:07:40 <dwc-> blame stackoverflow, it probably makes + a space
Jan 22 04:08:33 * dwc- doens't c++ either
Jan 22 04:08:36 <PuffTheMagic> dwc-: if we didnt uave to support windows and osxi migt have
Jan 22 04:09:21 <PuffTheMagic> i got emerge working on ubuntu
Jan 22 04:09:33 <PuffTheMagic> but i wasnt clean
Jan 22 04:09:44 <PuffTheMagic> and disnt follow through
Jan 22 04:10:02 <PuffTheMagic> doubt rod would have accpted it even if i did
Jan 22 04:11:50 <PuffTheMagic> dtzWill: we need to merge the char buffer and graphics state shit for so many reasons
Jan 22 04:12:21 <PuffTheMagic> scrolling woild be easy
Jan 22 04:16:41 <dtzWill> yes we do. giving it go atm, but no promises, not sure i have the time tonight to see this to completion :)
Jan 22 04:16:57 <PuffTheMagic> no prob
Jan 22 04:17:17 <dtzWill> this code makes many implicit assumptions on everything haha
Jan 22 04:17:32 <dtzWill> so we'll see how well i can replace something like this :)
Jan 22 04:17:34 <PuffTheMagic> i am just using irssi seeing how ad scrollig is
Jan 22 04:18:04 <dtzWill> oh scrolling is broken
Jan 22 04:18:10 <dtzWill> mostly because we don't even try to handle garphics states atm
Jan 22 04:18:42 <dtzWill> roflcopter our insertChar is ridiuclously complicated
Jan 22 04:18:43 <dtzWill> hahaha
Jan 22 04:19:13 * dtzWill happily rips out his contribution ot that mess
Jan 22 04:19:45 <PuffTheMagic> ya so much is so complicated
Jan 22 05:08:50 <dtzWill> PuffTheMagic: saving me the bother of tracing it, do you know offhand what LOGLEVEL to set while building under the new regime so that i get things like LOG_DEBUG back?
Jan 22 05:11:08 <PuffTheMagic> 7
Jan 22 06:22:40 <dtzWill> PuffTheMagic: offhand, what vttests presently pass? should top-level menu 1 pass? 2?
Jan 22 06:22:49 <dtzWill> iirc all of one of those used to pass
Jan 22 06:58:43 <dtzWill> latest in git seems to have problem setting screen resolution on startup
Jan 22 06:58:54 <dtzWill> text goes behind vkb--toggling vkb on/off fixes it
Jan 22 06:59:10 <dtzWill> since you've been making commits about initilaization order and such, just thought i'd make sure that was known
Jan 22 08:23:44 <dtzWill> PuffTheMagic: https://github.com/dtzWill/wTerm/tree/unified_cellstate is first attempt at what we discussed
Jan 22 08:23:59 <dtzWill> seems to work pretty well, but some tests don't work any longer. mc/vim/etc seem to work good
Jan 22 08:24:06 <dtzWill> anyway i probably have some off-by-one in there somewhere, idk
Jan 22 08:25:13 <dtzWill> if you get some time and wanna give that a look/feedback/something that'd be great
Jan 22 08:25:17 <dtzWill> needa leave it alone for now :)
Jan 22 08:26:18 <dtzWill> curiously i'm still seeing a bug when scrolling, bleeding graphics state in a strange way, which since i'm not explicitly associating each cell with its state means something somewhere isn't tracking state properly (as opposed to not updating the hard-to-reason-about graphics state datastructure properly)
Jan 22 08:26:48 <dtzWill> but haven't looked into it much.  fwiw test case is 'git log' using a shell other than the default busybox 'sh' (zsh works for me, but suit yourself)
Jan 22 08:27:57 <dtzWill> hopefully it'll be overall a step forward, it did let me entirely rip out huge swaths of code xD :)
Jan 22 13:10:00 <Brybry> <dtzWill> latest in git seems to have problem setting screen resolution on startup
Jan 22 13:10:05 <Brybry> what test am I looking at for this?
Jan 22 13:16:58 * cryptk just sent a pull request to add the Dvorak Simplified layout to wTerm
Jan 22 13:17:22 <cryptk> https://github.com/cryptk/wTerm/commit/65df14f020ba96563ad0f99205e8341171139e6e
Jan 22 13:20:32 <stbuehler> nice
Jan 22 13:20:46 <stbuehler> although the question is now whether to put all layouts in the same file
Jan 22 13:21:21 <cryptk> that is how PuffTheMagic told me to do it
Jan 22 13:21:40 <stbuehler> my basic idea was that they could be put into json files, but i doubt that would work now that there are the SDLK.* constants involved.
Jan 22 13:22:59 <stbuehler> i'd still split them into different files in the long run, whether loaded with enyo.xhr or depends.js
Jan 22 13:23:24 <cryptk> yeah, eventually it would be nice to have something like src/models/vkblayouts/*.js
Jan 22 13:23:31 <cryptk> each one gets it's own file
Jan 22 13:24:29 <stbuehler> the idea with async loading is that it wouldn't need all layounts in memory
Jan 22 13:24:46 <stbuehler> perhaps we can call enyo.depends, that might solve the constants problem
Jan 22 13:29:55 <cryptk> I don't know much about that stuff, I jsut happen to use a dvorak keyboard and PuffTheMagic told me how to add it
Jan 22 13:29:56 <cryptk> lol
Jan 22 13:33:14 <stbuehler> :)
Jan 22 15:39:16 <PuffTheMagic> stbuehler, this being the first custom vkb i really didnt know where to put it ;)
Jan 22 15:40:01 <PuffTheMagic> Brybry, i moved the initial set resolution code in main
Jan 22 15:40:04 <PuffTheMagic> prob have to fix that
Jan 22 15:48:58 <stbuehler> PuffTheMagic: hm, i get very strange errors with loading scripts... it somehow wipes the <head> node
Jan 22 15:49:28 <stbuehler> and i think you should remove all SDKL.* symbols for "normal" keys
Jan 22 15:52:50 <PuffTheMagic> what are normal keys?
Jan 22 15:54:33 <stbuehler> everything that isn't handled in SDLTerminal::handleKeyboardEvent
Jan 22 15:55:24 <stbuehler> you either need "sym" or "unicode" - never both
Jan 22 15:55:49 <stbuehler> ideally we would move everything to javascript and only forward strings
Jan 22 15:56:10 <PuffTheMagic> well the reason for using all the SDL key stuff was that I wanted to try and use SDL unicode translation stuff
Jan 22 15:56:20 <PuffTheMagic> so that we didnt need to hardcode the unicode stuff
Jan 22 15:57:19 <PuffTheMagic> but that didnt work out
Jan 22 15:57:27 <PuffTheMagic> so if u have ideas on how to simplify things go ahead
Jan 22 15:57:53 <stbuehler> kk
Jan 22 15:58:25 <PuffTheMagic> but for someone working on keyboard stuff its easier to use NAMES rather than numbers
Jan 22 15:58:47 <PuffTheMagic> so something like the SDLK file I think is good
Jan 22 15:59:44 <stbuehler> yes sure
Jan 22 15:59:55 <stbuehler> but why need a symbol for "1" or "q" ?
Jan 22 15:59:58 <PuffTheMagic> and the sdlk file doesnt prevent us from splitting layouts into multiple files
Jan 22 16:00:07 <stbuehler> F1, ESC, SHIFT, ... sure, you want a symbol for that
Jan 22 16:00:21 <PuffTheMagic> consistency?
Jan 22 16:00:26 <PuffTheMagic> idk
Jan 22 16:00:34 <PuffTheMagic> what happens for non ascii chars?
Jan 22 16:00:41 <stbuehler> no; if you don't use it, remove it. otherwise it contains errors, and that might bite you later :)
Jan 22 16:00:51 <stbuehler> like the '|' symbols is BACKSPACE
Jan 22 16:01:15 <stbuehler> i added utf-8 handling, so i can send chars like 'ä'
Jan 22 16:01:31 <stbuehler> i just can't display them yet :)
Jan 22 16:01:44 <PuffTheMagic> ya well sending was never an issue
Jan 22 16:01:45 <PuffTheMagic> ;)
Jan 22 16:01:51 <PuffTheMagic> but go ahead, do what u want
Jan 22 16:01:57 <PuffTheMagic> i will see it when its ready i guess
Jan 22 16:02:15 <stbuehler> and i'm not sure how to handle the layout itself; sometimes i want more keys or other key sizes
Jan 22 16:02:29 <PuffTheMagic> more keys is easy
Jan 22 16:02:40 <stbuehler> would be easier if box-stretch would be working
Jan 22 16:03:21 <PuffTheMagic> if u can get both orientations working with box stretch go for it
Jan 22 16:03:49 <stbuehler> how could i do more keys otherwise?
Jan 22 16:04:04 <stbuehler> for example i need one more key in the row with backspace
Jan 22 16:04:04 <PuffTheMagic> new css classes for some keys
Jan 22 16:04:23 <stbuehler> and setting the class depending on the layout, hm
Jan 22 16:05:27 <PuffTheMagic> instead of using small: true to set that its a small layout each layout could define the master classes
Jan 22 16:05:35 <cryptk> PuffTheMagic, bear in mind, I didn't test that layout at all, since I'm not sure how to build wTerm
Jan 22 16:05:44 <PuffTheMagic> masterClasses: ['large','small']
Jan 22 16:05:49 <cryptk> if you want to build me a copy with the dvorak in it, I can test it out for ya
Jan 22 16:05:56 <cryptk> make sure all the keys actually work properly
Jan 22 16:06:09 <PuffTheMagic> cryptk, do u have build.git?
Jan 22 16:06:10 <cryptk> and I decided to go with dvorak simplified rather than straight dvorak
Jan 22 16:06:20 <cryptk> I don't now, but I can clone it
Jan 22 16:06:43 <PuffTheMagic> cryptk, if u clone it into the standard location u just need to do make toolchain
Jan 22 16:06:46 <PuffTheMagic> and then u can build wterm
Jan 22 16:06:53 <PuffTheMagic> cryptk, http://ompldr.org/vY2U0bQ
Jan 22 16:06:53 <cryptk> kk, lemme get that done
Jan 22 16:07:04 <PuffTheMagic> there is an ipk to play with while it compiles
Jan 22 16:07:19 <PuffTheMagic> cryptk, after u complete make toolchain just do export DEVICE=1
Jan 22 16:07:27 <PuffTheMagic> before building wterm
Jan 22 16:10:13 <stbuehler> PuffTheMagic: did ctrl-c work? it doesn't right now^^
Jan 22 16:10:36 <PuffTheMagic> ya it works fine for me
Jan 22 16:10:41 <stbuehler> hm
Jan 22 16:10:50 <stbuehler> the sleep 100 should be done by now anyway
Jan 22 16:11:11 <PuffTheMagic> which sleep 100
Jan 22 16:11:21 <stbuehler> that i tried ctrl-c with
Jan 22 16:11:32 <stbuehler> ah, now it works
Jan 22 16:17:34 <cryptk> PuffTheMagic, Shift+V and Shift+Z gives me < on the dvorak keyboard
Jan 22 16:18:55 <cryptk> everything else seems to work properly
Jan 22 16:19:14 <cryptk> 		{symbols: [['V',SDLK._v],['<span class="fnBind">Del</span>',SDLK._DELETE]], printable: true},
Jan 22 16:19:14 <cryptk> 		{symbols: [['Z',SDLK._z],['<span class="fnBind">Ins</span>',SDLK._INSERT]], printable: true},
Jan 22 16:19:23 <cryptk> those are the lines in my dvorak config for those keys
Jan 22 16:19:27 <cryptk> not sure what is wrong with them
Jan 22 16:19:31 <PuffTheMagic> i do sorta
Jan 22 16:19:42 <PuffTheMagic> they are printables with more than 1 option
Jan 22 16:19:54 <PuffTheMagic> all the other letters that are printable only have 1 option
Jan 22 16:19:58 <cryptk> yeah
Jan 22 16:20:01 <PuffTheMagic> thats hwo i did lowercase
Jan 22 16:20:19 <cryptk> and the Fn key plus those two just print the lowercase letter rather than triggering Del or Ins
Jan 22 16:20:34 <PuffTheMagic> ya all this new layout stuff is still pretty rough ;)
Jan 22 16:20:42 <cryptk> btw, is it just me, or is your keyboard more responsive than the built in keyboard, lol
Jan 22 16:20:55 <PuffTheMagic> lol, in portrait its slower
Jan 22 16:21:00 <PuffTheMagic> but i think thats GL being slow
Jan 22 16:21:05 <PuffTheMagic> having to do rotations of the screen
Jan 22 16:21:06 <PuffTheMagic> idk
Jan 22 16:21:24 <cryptk> could be
Jan 22 16:21:26 <PuffTheMagic> cryptk, im gonna wait to see what stbuehler does with the keyboard code before I fix this
Jan 22 16:21:30 <cryptk> ok
Jan 22 16:21:40 <cryptk> that is the only bug I found with dvorak, everything else works great
Jan 22 16:21:48 <stbuehler> i'd like to remove the <span> things
Jan 22 16:21:50 <cryptk> I feel more at home typing on it now, lol
Jan 22 16:22:43 <PuffTheMagic> stbuehler, the current setup was my attempt a more flexible setup that supported up to 4 chars per key
Jan 22 16:22:54 <PuffTheMagic> i know it needs improvement
Jan 22 16:22:55 <stbuehler> yes, i get that :)
Jan 22 16:23:08 <PuffTheMagic> i need to distribute my efforts in lots of places
Jan 22 16:23:16 <stbuehler> hm
Jan 22 16:23:24 <stbuehler> i guess for now i'll leave the span in
Jan 22 16:23:48 <PuffTheMagic> we need to be able to specify a class of each of the char locations on a key
Jan 22 16:23:49 <PuffTheMagic> ideally
Jan 22 16:24:36 <PuffTheMagic> if u are getting rid of the unicode and sym for just  1, use the other cell for a class name?
Jan 22 16:26:56 <cryptk> ok, but the problems with dvorak aren't anythign with my config, they are something else, which means that it isn't something I can fix, lol
Jan 22 16:27:37 <PuffTheMagic> i know what the issue is with that dont worry
Jan 22 16:27:39 <stbuehler> cryptk: are you happy with how it looks?
Jan 22 16:27:51 <cryptk> with how the layout is arranged? yeah
Jan 22 16:28:00 <PuffTheMagic> cryptk, what does a non simplified dvorak look like?
Jan 22 16:28:01 <cryptk> it matches my keyboard perfectly, which is a dvorak simplified layout
Jan 22 16:28:01 <stbuehler> k
Jan 22 16:28:37 <stbuehler> PuffTheMagic: what is the state of Fn+X support?
Jan 22 16:28:41 <stbuehler> i can't find the code for it :)
Jan 22 16:28:46 <cryptk> PuffTheMagic, the main differences is that a dvorak simplified maintains the QWERTY key pairings, (so ` and ~ are on the same keys like on a QWERTY) and on a simplified the numbers are in numeric order
Jan 22 16:28:58 <PuffTheMagic> stbuehler, minimal
Jan 22 16:29:08 <PuffTheMagic> its in sdlterminal
Jan 22 16:29:12 <cryptk> on a straight Dvorak, that isn't the case, things are more re-arranged and the numbers are arranged so that the more commonly used numbers are easier to type
Jan 22 16:29:18 <cryptk> lemme get you a pic of a true Dvorak
Jan 22 16:30:00 <PuffTheMagic> stbuehler, again, if u see ways to improve any of the code relating to input just do it
Jan 22 16:32:06 <stbuehler> PuffTheMagic: yes, but first i need to know what it is doing now :)
Jan 22 16:32:24 <PuffTheMagic> all the modifiers are handled in the sdlterminal.cpp
Jan 22 16:32:35 <PuffTheMagic> except for shift on ascii chars
Jan 22 16:32:36 <stbuehler> i see the SDL_GetModState(), but where is it used for translating "." to "Del" for example?
Jan 22 16:32:43 <stbuehler> hm
Jan 22 16:33:14 <stbuehler> i see
Jan 22 16:34:14 <stbuehler> i can see how this would be useful for bt keyboards
Jan 22 16:34:21 <Brybry> some of the Fn stuff works, right?
Jan 22 16:34:30 <Brybry> like home/end
Jan 22 16:34:45 <stbuehler> but i'd like to customize the fn handling in layouts
Jan 22 16:35:18 <cryptk> PuffTheMagic, I can't find a pic of it, but the one that I set up is the "commonly used" one
Jan 22 16:35:38 <cryptk> it became more popular because it can be done by just re-arranging qwerty keys for the most part
Jan 22 16:35:41 <PuffTheMagic> stbuehler, the modstate in vkb code should be accurate if u want to do stuff in js
Jan 22 16:36:05 <PuffTheMagic> but i think it should either all be js or all in C
Jan 22 16:36:20 <stbuehler> i'm for all in js :)
Jan 22 16:36:23 <PuffTheMagic> stbuehler, i was doing stuff in C so that I could support BT keyboards with the same code
Jan 22 16:36:31 <cryptk> PuffTheMagic, if it is done in C, then the layouts could still be done with config files
Jan 22 16:36:37 <cryptk> or perhaps, here's a novel idea
Jan 22 16:36:42 <cryptk> use the same keymaps that X uses
Jan 22 16:36:47 <cryptk> and have a folder in /media/internal
Jan 22 16:36:53 <stbuehler> cryptk: well, but you would need two files, one for js and one for c per layout
Jan 22 16:36:54 <cryptk> so that users can drop whatever keymaps they want into there
Jan 22 16:37:34 <PuffTheMagic> im saying out of this, as long as its less sucky than it is now im fine
Jan 22 16:37:43 <PuffTheMagic> and the Bt keyboard people need to be happy too
Jan 22 16:37:49 <cryptk> yeah
Jan 22 16:37:58 <cryptk> I would love to see it be able to use standard X keymaps
Jan 22 16:38:43 <cryptk> and there are libraries out there already to parse those and work with them
Jan 22 16:39:05 <cryptk> granted, I'm not a C coder, so feel free to ignore me if that idea isn't practical
Jan 22 16:40:22 <stbuehler> well, i think you can't mix the "customize" concept with standard x keymaps
Jan 22 16:53:14 <PuffTheMagic> hmmm
Jan 22 16:53:20 <PuffTheMagic> busybox need chmod u+s /bin/busybox for su to work
Jan 22 16:54:28 <PuffTheMagic> and there is no pam
Jan 22 16:54:33 <PuffTheMagic> so no wheel group
Jan 22 16:54:35 <PuffTheMagic> awesome
Jan 22 16:57:20 <cryptk> stbuehler, the x keymap idea would have been replacing the customized ideas
Jan 22 16:57:32 <cryptk> although it wouldn't preclude someone from making a custom x keymap
Jan 22 16:57:35 <cryptk> which isn't difficult
Jan 22 16:57:57 <stbuehler> the point is that you have to fit the keys somehow on the display
Jan 22 16:58:12 <stbuehler> and assign them symbol value, that the x keymap can use
Jan 22 16:58:20 <cryptk> yeah, I see what you are saying
Jan 22 16:58:37 <cryptk> you would have to basicaly impliment almost all of X's keyboard input stuff to make that work
Jan 22 16:58:45 <cryptk> and then have something render that in the app...
Jan 22 16:58:52 <cryptk> while not impossible, it would be a ton of work
Jan 22 16:59:06 <cryptk> and likely impractical
Jan 22 17:05:24 <stbuehler> where is "@" located on your keyboards?
Jan 22 17:12:09 <PuffTheMagic> ok yay!
Jan 22 17:12:14 <PuffTheMagic> i solved the su issue
Jan 22 17:12:40 <PuffTheMagic> it involved deleting the symlink of /bin/su (to /bin/busybox)
Jan 22 17:12:49 <PuffTheMagic> then copying busybox to su
Jan 22 17:12:54 <PuffTheMagic> and then setting the perms on it properly
Jan 22 17:14:14 <PuffTheMagic> would anyone object to me adding this to the postinst of wterm
Jan 22 17:14:24 <PuffTheMagic> and creating a non-root wterm user
Jan 22 17:14:48 <PuffTheMagic> then we can tell people if they want to su then need to add wterm user to the root group in /etc/group
Jan 22 17:14:59 <PuffTheMagic> rwhitby, ^^
Jan 22 17:29:05 <PuffTheMagic> cryptk, ping
Jan 22 17:36:39 <stbuehler> ok, layouts looking good :)
Jan 22 17:36:47 <PuffTheMagic> D:
Jan 22 17:36:48 <stbuehler> also supporting pageup/down,insert,delete
Jan 22 17:36:49 <PuffTheMagic> :D
Jan 22 17:40:24 <stbuehler> only my umlaut keys are a bit useless, they "should" work, but only got crap over netcat
Jan 22 17:44:12 <cryptk> PuffTheMagic, I wouldn't want wTerm to change things in the core OS of my device personally
Jan 22 17:45:01 <PuffTheMagic> well if I cant do that there is no way to allow safe su usage
Jan 22 17:45:01 <cryptk> my new project has officially been started
Jan 22 17:45:11 <cryptk> an online mysql slow query log analyzer
Jan 22 17:45:21 <cryptk> the file upload portion is almost done...
Jan 22 17:46:07 <cryptk> it enforces a max file size, and makes sure that the file that is being uploaded is of type text/plain
Jan 22 17:46:16 <cryptk> and it stores them in an area outside of the doc root
Jan 22 17:46:40 <PuffTheMagic> cryptk, i suppose I could create a "new" su that is on path
Jan 22 17:47:22 <PuffTheMagic> that way it would be overridden
Jan 22 17:50:08 <PuffTheMagic> cryptk, ever get wterm building?
Jan 22 17:50:28 <cryptk> I didn't try, you sent me the ipk before make toolchain finished, lol
Jan 22 17:50:45 <cryptk> I don't know how to write in C, or enyo, or mojo
Jan 22 17:51:03 <cryptk> so there isn't really a point to me being able to build it myself, since my code contributions are likely limited to that Dvorak layout
Jan 22 17:51:17 <cryptk> trust me, that's the only thing you want me to touch, I would break it, lol
Jan 22 17:51:30 <cryptk> there is a reason that I am "that one guy who works on the servers"
Jan 22 17:51:31 <cryptk> lol
Jan 22 17:51:39 <cryptk> now if these apps were written in PHP, I would be all over it
Jan 22 17:54:26 <stbuehler> wtf, what did you psuh ? takes me ages to update my github fork...
Jan 22 17:55:04 <PuffTheMagic> it always takes ages
Jan 22 17:55:46 <stbuehler> not on my usual projects
Jan 22 17:56:27 <stbuehler> perhaps you should keep your history more linear instead of al those merges :D
Jan 22 18:02:00 <stbuehler> have fun with testing :D
Jan 22 18:02:13 <PuffTheMagic> hmmmmm
Jan 22 18:02:13 <PuffTheMagic> cp /bin/busybox ${APPDIR}/bin/su
Jan 22 18:02:13 <PuffTheMagic> chmod 750 ${APPDIR}/bin/su
Jan 22 18:02:13 <PuffTheMagic> chmod ug+s ${APPDIR}/bin/su
Jan 22 18:02:16 <PuffTheMagic> the cp works
Jan 22 18:02:25 <PuffTheMagic> but the file is not getting chmod'd
Jan 22 18:02:27 <PuffTheMagic> :(
Jan 22 18:03:55 <cryptk> is the postinst getting run as root?
Jan 22 18:04:06 <PuffTheMagic> i thought that was the hole point of postinst
Jan 22 18:04:10 <cryptk> true
Jan 22 18:07:53 <PuffTheMagic> anyone know what the filemode for 750 then ug+s ?
Jan 22 18:08:53 <stbuehler> 01750 ?
Jan 22 18:09:21 <PuffTheMagic> isnt it max 4 numbers?
Jan 22 18:09:45 <stbuehler> wait
Jan 22 18:09:50 <stbuehler> 04750
Jan 22 18:09:58 <PuffTheMagic> 6750
Jan 22 18:09:59 <stbuehler> well, the first 0 is for "octal" mode
Jan 22 18:10:08 <stbuehler> 6 is for suid + sgid
Jan 22 18:10:20 <PuffTheMagic> i thought ug+s was suid and sgid
Jan 22 18:10:21 <stbuehler> it will probably work without the leading 0 too
Jan 22 18:10:25 <stbuehler> ah yes
Jan 22 18:10:29 <stbuehler> sry, didn't read :)
Jan 22 18:11:12 <stbuehler> chmod will probably always use octal mode, but in c (or any other "normal" language) you need the 0 prefix
Jan 22 18:13:34 <PuffTheMagic> hmm
Jan 22 18:13:41 <PuffTheMagic> package.properties is not working
Jan 22 18:17:23 <PuffTheMagic> maybe i need a new sdk
Jan 22 18:38:36 <dwc-> PuffTheMagic: would also need to set a root password then too?
Jan 22 18:38:57 <PuffTheMagic> dont think so
Jan 22 18:39:24 <PuffTheMagic> not to su
Jan 22 18:39:33 <PuffTheMagic> dont need to need to set a passwd for the non-root user
Jan 22 18:41:22 <dwc-> I would say that would add a security hole then
Jan 22 18:41:30 <dwc-> (making su work)
Jan 22 18:41:34 <PuffTheMagic> well
Jan 22 18:41:50 <PuffTheMagic> it only works if the user manually adds the wterm user to the root group in /etc/passwd
Jan 22 18:42:10 <dwc-> does it?
Jan 22 18:42:11 <PuffTheMagic> but i suppose we can also add a root passwd
Jan 22 18:42:24 <dwc-> most linuxes I know of, by default doesn't require a wheel group to su
Jan 22 18:42:34 <PuffTheMagic> this has no wheel group
Jan 22 18:42:39 <PuffTheMagic> i said root group
Jan 22 18:43:15 <dwc-> oh, I see, you don't have o+x
Jan 22 18:44:36 <PuffTheMagic> hmm now chmod ug+s /media/cryptofs/apps/usr/palm/applications/us.ryanhope.wterm/bin/su
Jan 22 18:44:41 <PuffTheMagic> chmod: /media/cryptofs/apps/usr/palm/applications/us.ryanhope.wterm/bin/su: Operation not permitted
Jan 22 18:44:42 <PuffTheMagic> wtf
Jan 22 18:46:52 <PuffTheMagic> ok i think cryptofs is breaking perms on files
Jan 22 18:48:56 <dwc-> just the setuid part?
Jan 22 18:48:57 <PuffTheMagic> yep
Jan 22 18:49:12 <dwc-> cryptofs on /media/cryptofs type fuse.cryptofs (rw,nosuid,nodev,user_id=0,group_id=0,allow_other)
Jan 22 18:49:12 <PuffTheMagic> well
Jan 22 18:49:16 <PuffTheMagic> its worse than that
Jan 22 18:49:49 <PuffTheMagic> less of the day, install wTerm to /
Jan 22 18:49:57 <PuffTheMagic> s/less/lesson/
Jan 22 18:50:00 <cryptk> PuffTheMagic, you know what you need?
Jan 22 18:50:09 <cryptk> something made by a company named "Stone Brewery"
Jan 22 18:50:27 <cryptk> drinking an IPA from them as we speak
Jan 22 18:50:37 * PuffTheMagic goes down stairs and fetches a Double Bastard
Jan 22 18:51:01 <dwc-> cryptk: not at scale10x?
Jan 22 18:51:12 <cryptk> dwc-, unfortuantely not
Jan 22 18:51:16 <cryptk> couldn't make it this year
Jan 22 18:51:29 <cryptk> Too muck going on in my department at Rackspace currently
Jan 22 18:51:43 <dwc-> I hear ya
Jan 22 18:52:35 <PuffTheMagic> so looks like i need to install wterm to /, any objections?
Jan 22 18:55:20 <cryptk> PuffTheMagic, to / or to somewhere outside of cryptofs?
Jan 22 18:55:41 <PuffTheMagic> where else will it work?
Jan 22 18:56:02 <cryptk> well, what I meant was, where exactly are you planning to install it?
Jan 22 18:56:19 <cryptk> and does wTerm need to be in / or does the copy of busybox need to be there?
Jan 22 18:56:19 <PuffTheMagic> i would have to go in /usr/palm/applications wouldnt it?
Jan 22 18:57:09 <cryptk> yeah, somewhere in there would work
Jan 22 18:57:23 <PuffTheMagic> i guess its just the busybox copy that needs to move
Jan 22 18:59:25 <PuffTheMagic> /usr/local/bin is on path already
Jan 22 18:59:28 <PuffTheMagic> guess that would work
Jan 22 19:01:10 <cryptk> that is a less intrusive change
Jan 22 19:01:23 <PuffTheMagic> yes it works
Jan 22 19:01:25 <cryptk> since you aren't changing anything that already exists, just placing a copy somewhere
Jan 22 19:01:39 <PuffTheMagic> now...
Jan 22 19:01:43 <Brybry> can't you just make a symlink?
Jan 22 19:01:57 <PuffTheMagic> symlink of what to where?
Jan 22 19:02:15 <Brybry> actually, I dunno if that would work
Jan 22 19:02:42 <Brybry> a symlink to busybox in the path named su with the correct permissions?
Jan 22 19:02:53 <PuffTheMagic> no that wont work
Jan 22 19:02:58 <PuffTheMagic> su in /bin is alraedy a symblink
Jan 22 19:03:07 <PuffTheMagic> change the perms on that and the main busybox bin changes
Jan 22 19:03:13 <Brybry> oh :(
Jan 22 19:03:31 <PuffTheMagic> so now we just need a way to get users to set a root password
Jan 22 19:03:40 <Brybry> hmm
Jan 22 19:03:59 <Brybry> doesn't that do bad things for development stuff?
Jan 22 19:04:06 <PuffTheMagic> no
Jan 22 19:04:18 <cryptk> things that run as root, run as root
Jan 22 19:04:33 <Brybry> ah, i guess dev mode password != root password
Jan 22 19:04:37 <cryptk> you will only need the password if you are running something as non-root and want it to run something else as root
Jan 22 19:05:00 <cryptk> so it /shouldn't/ affect anything, except things running as non-root that need to run something as root
Jan 22 19:05:57 <PuffTheMagic> is there a way I can display a motd when a user logs in
Jan 22 19:06:00 <PuffTheMagic> but only for that user
Jan 22 19:06:47 <dwc-> put it in their .shellrc
Jan 22 19:07:25 <dwc-> another possibility might be to put "start root shell" in the wterm menu
Jan 22 19:07:25 <PuffTheMagic> so I could set a random root password if one is not already set and then tell the user what it is when they login
Jan 22 19:07:27 <PuffTheMagic> i suppose
Jan 22 19:07:37 <PuffTheMagic> or that
Jan 22 19:08:45 <PuffTheMagic> so if root had a password, then i could add the wterm user to the root group when its created
Jan 22 19:15:41 <cryptk> being able to share texts to the TP is so handy...
Jan 22 19:18:53 <PuffTheMagic> stbuehler, you have a bad row of keys on your new layout
Jan 22 19:19:04 <PuffTheMagic> one of the keys is cut off
Jan 22 19:24:44 <stbuehler> yes, i know. but still reachable, and i need to think of a good solution :)
Jan 22 19:37:27 <cryptk> PuffTheMagic, I would leave creating the password, and adding wterm to the goup up to the user
Jan 22 19:37:49 <cryptk> it opens a big security hole doing that, and as such, it should require a bit of know-how to open it
Jan 22 19:38:31 <cryptk> at that point, if someone does the work to open the hole, and they bork their device because something exploits it, then you have plausible deny-ability
Jan 22 19:38:48 <cryptk> they have gone through extra steps to make your application work in a way other than how you provided it
Jan 22 19:39:20 <cryptk> and they have read the appropriate warnings and documentation on what doing those extra steps means
Jan 22 19:40:12 <PuffTheMagic> ok fine
Jan 22 19:41:59 <cryptk> haha, I just read the ingredients on this stone brewery IPA
Jan 22 19:42:20 <cryptk> "Nothing but Barley, Hops, Water & Yeast."
Jan 22 19:42:25 <cryptk> lol
Jan 22 19:42:52 <cryptk> and I believe it, because by federal law, the ingredients list has to be accurate
Jan 22 19:44:08 <PuffTheMagic> doesnt say the strain of yeast
Jan 22 19:44:17 <PuffTheMagic> so cant do much there
Jan 22 19:44:31 <cryptk> yeah
Jan 22 19:44:40 <cryptk> it doesn't reveal anything
Jan 22 19:45:09 <cryptk> federal law requires that the list be accurate, but it does allow for enough ambiguity to protect proprietary recipes
Jan 22 19:49:00 <PuffTheMagic> cryptk, im adding a "safe root access" procedure to the readme
Jan 22 19:51:33 <dtzWill> *yawn*, morning all :). looks like you've been having some fun.
Jan 22 19:53:19 <dtzWill> PuffTheMagic: have a chance to take a look at that branch i linked ya?  as we discussed i think that's the direction we wanna go in.... but there's something slightly wrong atm.  not entirely surprising since i had to touch most of the escape code helper functions
Jan 22 19:53:33 <PuffTheMagic> not yet, been busy with security stuff
Jan 22 19:53:50 <dtzWill> ah, the wterm user stuff? kk
Jan 22 19:53:55 <PuffTheMagic> ya
Jan 22 19:55:26 <PuffTheMagic> cryptk, https://github.com/PuffTheMagic/wTerm
Jan 22 19:55:32 <PuffTheMagic> how's that
Jan 22 19:55:40 <PuffTheMagic> the "Safe Root Access" section
Jan 22 19:56:17 <cryptk> PuffTheMagic, you shouldn't have them vi the group file directly
Jan 22 19:56:23 <cryptk> is usermod or moduser available?
Jan 22 19:56:39 <PuffTheMagic> nope
Jan 22 19:56:55 <PuffTheMagic> neither is available
Jan 22 19:56:55 <cryptk> ahh
Jan 22 19:57:17 <cryptk> I guess the only option is to have them vi directly then
Jan 22 19:57:28 <cryptk> that section should have a warning with it though
Jan 22 19:57:31 <dwc-> oh come on now, it's not unix if you've got something to stop them from borking an entire file
Jan 22 19:57:56 <cryptk> since you are calling it "safe root access" but it isn't necessarily safe
Jan 22 19:58:05 <cryptk> dwc-, you are right... and this isn't unix
Jan 22 19:58:08 <cryptk> it's linux
Jan 22 19:58:14 <cryptk> OOOH BURN!
Jan 22 19:58:22 <dwc-> just be glad they don't have nano
Jan 22 19:58:34 <PuffTheMagic> cryptk, how is it not safe?
Jan 22 19:58:38 <dwc-> or someone would bork their pw file when it saves it out linewrapped at 80 chars
Jan 22 19:59:13 <PuffTheMagic> safe has nothing to do with preventing login entirely ;)
Jan 22 19:59:36 <PuffTheMagic> s/preventing/breaking the ability to/
Jan 22 20:00:05 <cryptk> because if you follow those instructions, then can't the just type actions exec rm -f /
Jan 22 20:00:06 <cryptk> >
Jan 22 20:00:07 <cryptk> ?
Jan 22 20:00:24 <cryptk> which means if you follow those instructions, things can exploit the application
Jan 22 20:00:32 <PuffTheMagic> you lost me
Jan 22 20:01:21 <cryptk> ok, if you add the wterm user to root group, then can't an application send the command `rm -rf /` to wTerm through the just type interface?
Jan 22 20:01:28 <cryptk> granted, it would have to be a malicious application
Jan 22 20:01:46 <PuffTheMagic> wouldnt they still have to be root to do that?
Jan 22 20:01:47 <cryptk> but if I follow those instructions, what is preventing an app from cross launching wTerm with that command?
Jan 22 20:02:06 <dwc-> isn't it more if you don't follow those instructions?
Jan 22 20:02:18 <cryptk> they would, but what if wterm is added to the root group, and I send this command
Jan 22 20:02:32 <cryptk> su -c "rm -rf /" root
Jan 22 20:02:55 <cryptk> although, it should prompt me for a password there
Jan 22 20:03:01 <cryptk> if I follow those instructions
Jan 22 20:03:10 <dwc-> yes
Jan 22 20:03:33 <cryptk> I would still pair that up with a popup showing what command is being requested
Jan 22 20:03:46 <cryptk> not sure if you have one (now that I think about it, I think you do)
Jan 22 20:03:54 <dwc-> it's dangerous if you /don't/ follow those directions
Jan 22 20:03:59 <PuffTheMagic> i do have a pop up for launch args
Jan 22 20:04:00 <dwc-> and you have just type on
Jan 22 20:04:03 <PuffTheMagic> currently
Jan 22 20:04:21 <PuffTheMagic> i dont like it though ;)
Jan 22 20:04:29 <PuffTheMagic> but i didnt remove it yet
Jan 22 20:05:50 <cryptk> yeah, I would leave the current one, at least until you come up with a replacement that you do like
Jan 22 20:06:01 <cryptk> having something that you don't like is better than having nothing
Jan 22 20:06:26 <cryptk> I would still put a quick warning statement on those instructions about the dangers of running commands as root
Jan 22 20:06:28 <cryptk> something like...
Jan 22 20:06:48 <PuffTheMagic> i dont like it because its a popup/warning period
Jan 22 20:06:52 <PuffTheMagic> there will never be one i like :D
Jan 22 20:07:03 <cryptk> "Any command run as root has the potential to wipe your device, alter your data, or kick puppies"
Jan 22 20:07:25 <dwc-> yes! puppy kicking! *installs*
Jan 22 20:07:29 <cryptk> I would add something like that to the README near the safe root instructions
Jan 22 20:07:36 <cryptk> haha dwc-
Jan 22 20:07:45 <dwc-> I would add that to first app startup
Jan 22 20:07:54 <dwc-> nobody in preware reads readmes
Jan 22 20:07:58 <PuffTheMagic> Enabling this option will allow any trojan or virus to execute destructive commands on your device without your knowledge!
Jan 22 20:08:03 <PuffTheMagic> ^^ is what Rod said to add
Jan 22 20:08:23 <cryptk> Enabling this option will allow any trojan or virus to execute destructive commands on your device or kick puppies without your knowledge!
Jan 22 20:08:26 <cryptk> ^^fixed
Jan 22 20:08:27 <PuffTheMagic> cryptk, added you to the contributors btw
Jan 22 20:08:33 <cryptk> thanks PuffTheM`
Jan 22 20:08:35 <cryptk> argh
Jan 22 20:08:45 <cryptk> PuffTheMagic, what is up with PuffTheM` ???
Jan 22 20:09:03 <cryptk> my bouncer can handle over 10 people on freenode now btw, it has been abel to for a long time, if you want on it, let me know
Jan 22 20:09:33 <cryptk> it go's down for a server reboot maybe once every 2 months, and I only do it if a kernel update is pending, and only if nobody is on IRC that uses it
Jan 22 20:10:05 <PuffTheMagic> cryptk, i decided to stay with bip
Jan 22 20:10:13 <PuffTheMagic> u can kill that user on your server
Jan 22 20:10:21 <cryptk> there was one time where I meant to shut down my laptop, but I forgot that specific terminal was SSH'd into my primary server, and it went down for that, lol
Jan 22 20:10:46 <cryptk> killed
Jan 22 20:10:55 <cryptk> any specific reason for going with bip rather than ZNC?
Jan 22 20:10:59 <PuffTheMagic> dtzWill, ok i can look at your branch now
Jan 22 20:11:24 <cryptk> s/branch/fork/g
Jan 22 20:11:33 <cryptk> wha!!! no infobot in here!?!
Jan 22 20:11:52 <dwc-> infobot won't s/ other people's text
Jan 22 20:11:54 <cryptk> PuffTheM`, user killed
Jan 22 20:11:56 <dwc-> and it doesn't take /g anyways
Jan 22 20:12:04 <PuffTheMagic> lol
Jan 22 20:12:04 <cryptk> dwc-, it will if you put g at the end
Jan 22 20:12:18 <cryptk> but only if it matches what the person immediately before you said
Jan 22 20:12:19 <PuffTheMagic> dtzWill, is your branch safe to merge or will it kill kittens?
Jan 22 20:12:35 <cryptk> s/kill kittens/kick puppies/
Jan 22 20:12:40 <cryptk> fixed ^^^
Jan 22 20:14:46 <PuffTheMagic> terminal/terminalstate.cpp:85: warning: comparison is always true due to limited range of data type
Jan 22 20:14:47 <PuffTheMagic> /tmp/.private/ryan/cck12Do1.s: Assembler messages:
Jan 22 20:14:47 <PuffTheMagic> /tmp/.private/ryan/cck12Do1.s:609: rdhi, rdlo and rm must all be different
Jan 22 20:14:47 <PuffTheMagic> /tmp/.private/ryan/cck12Do1.s:4961: rdhi, rdlo and rm must all be different
Jan 22 20:14:56 <PuffTheMagic> ^^^ dtzWill what did u do?
Jan 22 20:16:33 <cryptk> he did what I would do... broke it, lol
Jan 22 20:16:49 <stbuehler> wtf
Jan 22 20:18:35 <PuffTheMagic> dtzWill, looks like all the things that worked in vttest still do
Jan 22 20:20:52 <cryptk> PuffTheMagic, I deleted my fork, mainly because if I contribute again, my fork will be horrendously out of date, and it would be best for me to re-fork it then
Jan 22 20:21:53 <PuffTheMagic> not sure that is accurate but ok
Jan 22 20:23:13 <cryptk> well, my fork won't stay up-to-date with the changes that you make in your repo
Jan 22 20:23:42 <cryptk> so if I go to make another contribution, my fork won't match the main repo, which could make merging my pull request a PITA
Jan 22 20:24:15 <dwc-> you can update your fork
Jan 22 20:24:38 <cryptk> I can, but that is more work for me than just going to PuffTheMagic's repo and clicking fork
Jan 22 20:24:47 <dwc-> indeed
Jan 22 20:24:53 <cryptk> granted, forking takes 2 clicks for me...
Jan 22 20:25:16 <cryptk> click fork, click where I want to fork it to
Jan 22 20:26:20 <PuffTheMagic> cryptk, i guess deleting your fork, and recloning is way less work than; git reset --hard; git pull
Jan 22 20:26:52 <cryptk> for me it is
Jan 22 20:26:53 <dwc-> wouldn't that just update his local copy to his fork?
Jan 22 20:26:59 <cryptk> yeah, it would
Jan 22 20:27:06 <cryptk> since the forks are a github thing
Jan 22 20:27:15 <cryptk> how would I make my github repo do that?
Jan 22 20:27:29 <cryptk> I don't know, I am sure you can, but this is easier than me learning how to do it, lol
Jan 22 20:27:44 <cryptk> I know git basics, more than most people actually, but I am no git master
Jan 22 20:27:49 <PuffTheMagic> if u say so
Jan 22 20:27:54 <cryptk> lol
Jan 22 20:28:26 <PuffTheMagic> dtzWill: im using irssi on your new code and the srolling is much better
Jan 22 20:29:22 <PuffTheMagic> i cant tell if te minor issuea i do see are wterms fault or the nick coloring scripts fault
Jan 22 20:41:24 <Brybry> I just do like git fetch PTM; git reset --hard PTM/master
Jan 22 20:41:58 <Brybry> and then I guess I would push that if I cared to do it on my remote/github
Jan 22 20:43:05 <cryptk> but would that get a fork on github caught up to the repo it was forked from?
Jan 22 20:43:31 <cryptk> I guess it would, but my way is easier for me
Jan 22 20:43:32 <cryptk> lol
Jan 22 20:43:44 <cryptk> anyway, I gotta run
Jan 22 20:43:45 <cryptk> later all
Jan 22 20:46:56 <Brybry> PuffTheMagic, I can get a new login prompt to show up after typing exit but I get a "-sh can't access tty; job control turned off message" in the process that I'm reading up on to see what I missed
Jan 22 20:47:16 <PuffTheMagic> k
Jan 22 21:44:01 <PuffTheMagic> damn wterm crashes luna now
Jan 22 21:44:08 * PuffTheMagic converted app to headless
Jan 22 22:02:05 <Brybry> oh god what did I do that just made this work
Jan 22 22:02:11 <Brybry> shots in the dark :D
Jan 22 22:02:41 <PuffTheMagic> git diff
Jan 22 22:09:13 <PuffTheMagic> dtzWill, pinger
Jan 22 22:30:57 <PuffTheMagic> Brybry, gonna have something to pull soon, i want to tag 0.3.4
Jan 22 23:21:52 <Brybry> I'm killin zombies
Jan 22 23:24:01 <dtzWill> not to be that guy, but if we're distributing cmatrix we need the source somewhere too, probably.
Jan 22 23:24:28 <PuffTheMagic> dtzWill: in due time
Jan 22 23:24:49 <PuffTheMagic> it willrequire ncurses in build git
Jan 22 23:24:57 <PuffTheMagic> which i not easy
Jan 22 23:25:02 <PuffTheMagic> s/i/si
Jan 22 23:25:43 <PuffTheMagic> dtzWill: scrolling is almost perfect for TERM=rxvt, terrible for xterm
Jan 22 23:25:58 <dtzWill> PuffTheMagic: those assembler warnings are benign afaik
Jan 22 23:26:12 <dtzWill> and i think they go away with a newer toolchain
Jan 22 23:26:19 <PuffTheMagic> but why did they show up after your merge
Jan 22 23:30:06 <dtzWill> well we don't track state separately any longer
Jan 22 23:30:08 <dtzWill> garphics state i mean
Jan 22 23:30:23 <dtzWill> so if it's bad for anything, or wrong, it's b/c we didn't set it right, not because we didn't update it right.
Jan 22 23:30:38 <PuffTheMagic> oh i know
Jan 22 23:30:45 <PuffTheMagic> not blaming u
Jan 22 23:30:49 <dtzWill> which is a comfort of sorts :D
Jan 22 23:31:09 <dtzWill> interesting, didn't realize TERM changed the scrolling
Jan 22 23:31:19 <PuffTheMagic> oh
Jan 22 23:31:34 <PuffTheMagic> xterm has 1000 extra eacape codes
Jan 22 23:31:39 <dtzWill> oh, lol!
Jan 22 23:32:47 <PuffTheMagic> hey, i think if we are gonna have bold+bright fonts we need to set the base font spacing based on thw bold font
Jan 22 23:33:21 <PuffTheMagic> ithink we use the normal font rigt now and it makes bold text a litte crampt
Jan 22 23:33:45 <dtzWill> oh, okay.
Jan 22 23:33:51 <PuffTheMagic> but that might break oter shit
Jan 22 23:34:00 <Brybry> yay no more zombies
Jan 22 23:34:05 <PuffTheMagic> maybe bold is a bad idea
Jan 22 23:34:37 <dtzWill> PuffTheMagic: so scrolling and such work, but i think i slightly misimplemented some things like insertchar or something
Jan 22 23:34:47 <dtzWill> since vttests don't run quite like they used to :(
Jan 22 23:35:11 <dtzWill> otherwise it's a clear win in that the colors are much better in many cases now, and the color-handling code is zomg simpler
Jan 22 23:35:13 <PuffTheMagic> i thought all the basic vttests looked good
Jan 22 23:35:22 <PuffTheMagic> menu 1 ans 2 that is
Jan 22 23:35:58 <PuffTheMagic> sans scrolling behavior, which i know needs wrk
Jan 22 23:36:09 <dtzWill> tp is dead and booting up on touchstone atm, i'll double-check. hopefully you're right
Jan 22 23:36:10 <dtzWill> kk
Jan 22 23:36:39 <dtzWill> well menu 8 doesn't work like it used to i don't think, fwiw
Jan 22 23:37:23 <PuffTheMagic> lolu are right
Jan 22 23:37:29 <PuffTheMagic> i can fix that
Jan 22 23:37:39 <dtzWill> oh okay. i was all @.@ wtf
Jan 22 23:37:45 <PuffTheMagic> its part related to th term dimensions isue
Jan 22 23:38:07 <PuffTheMagic> dtzWill: dead?
Jan 22 23:38:12 <dtzWill> haha
Jan 22 23:38:24 <PuffTheMagic> do u have fast drain issues too/
Jan 22 23:38:30 <dtzWill> i just was confused, unable to track it down--even if it seems like it's an off-by-one
Jan 22 23:38:40 <dtzWill> PuffTheMagic: oh, i just hadn't put it on a charger since thurs
Jan 22 23:38:50 <PuffTheMagic> ahhh
Jan 22 23:39:06 <PuffTheMagic> mmmine dies as quick since updTE TO 305
Jan 22 23:39:14 <dtzWill> :/
Jan 23 00:00:18 <dtzWill> PuffTheMagic: when debugging applications that don't display correctly, any tips for figuring out offending escape codes? have wterm syslog what it sees and then bounce through that?
Jan 23 00:00:41 <dtzWill> also tell me it's not nice to see all that graphics state code ripped out ;) :D
Jan 23 00:01:59 <dtzWill> also, when you remove the vkb, there's some strange rendering strangeness that needs attention.  so some things to sort out before it's ready :)
Jan 23 00:02:53 <Brybry> I think I'll have something for you soon, PuffTheMagic
Jan 23 00:03:18 <Brybry> and then you can change the behavior to however you like
Jan 23 00:03:33 <Brybry> atm an exit will just throw you right back into another login/prompt
Jan 23 00:06:34 <dtzWill> PuffTheMagic: oh it's merged in ^.^ guess better fix those things soon hehe
Jan 23 00:06:45 <dtzWill> it /is/ forward progress in a number of ways, hehe :3
Jan 23 00:07:30 <dtzWill> lmk if you notice any performance changes.  also the portrait shouldn't be slower due to rotations, although iirc the way plugins work involves a level of indirection that at least at one point didn't rotate as well as one'd hope
Jan 23 00:08:08 <dtzWill> Brybry: oh, you're handling shell death? hooray
Jan 23 00:25:52 * Brybry glares at utf8.cpp for freaking him out
Jan 23 00:26:23 <stbuehler> hey!
Jan 23 00:26:30 <stbuehler> don't touch that :P
Jan 23 00:27:20 <Brybry> nah, I just don't use the Makefile so when I saw linker errors I was very confused for about three seconds
Jan 23 00:27:25 <stbuehler> i learned some utf-8 handling some time ago, so that code isn't exactly new
Jan 23 00:27:34 <stbuehler> hehe
Jan 23 00:28:53 <Brybry> something broke bluetooth keyboard support for me =(
Jan 23 00:31:47 <Brybry> hmm, our idling seems a lot less clean now
Jan 23 00:31:50 <Brybry> I wonder if that's me or something else
Jan 23 00:34:57 <Brybry> good news: not me bad news: not me
Jan 23 00:51:46 <PuffTheMagic> Brybry, whats the sleep 1 about?
Jan 23 00:54:46 <dtzWill> aww, y'all updated the enyo so i have to actually install via ipk or something xD
Jan 23 00:54:56 <Brybry> it's not really necessary
Jan 23 00:55:16 <Brybry> but letting the shell login stuff go through first seemed like best behavior
Jan 23 00:55:25 <Brybry> it shouldn't bug out without it there
Jan 23 00:56:18 <Brybry> now I have to look into this idling stuff and bt keyboard input :(
Jan 23 00:57:00 <dtzWill> Brybry: does it not work at all? and does vkb work?
Jan 23 00:57:19 <Brybry> looks like enter and backspace work
Jan 23 00:57:21 <PuffTheMagic> Brybry, stbuehler messed with keyboard shit
Jan 23 00:57:28 <PuffTheMagic> blame it on him ;)
Jan 23 00:57:41 <Brybry> but I don't think any of the ascii keys works
Jan 23 00:57:47 <PuffTheMagic> ya
Jan 23 00:57:49 <Brybry> well, alpha keys I should say
Jan 23 00:57:51 <PuffTheMagic> he changed mad shit
Jan 23 00:57:54 <dtzWill> Brybry: yeah, nothing seems to work on vkb here, even after updating the enyo stuff
Jan 23 00:57:57 <dtzWill> kk so it's not just me
Jan 23 00:58:02 <Brybry> vkb works
Jan 23 00:58:09 <PuffTheMagic> ya vkb works
Jan 23 00:58:17 <Brybry> though I haven't tried a clean install
Jan 23 00:58:21 <dtzWill> oh vkb doesn't work here :( i probably updated wrong, i'll poke at it later. just wanted to test redraw fixup
Jan 23 00:58:33 <Brybry> and our idling is back to terrible
Jan 23 00:58:38 <PuffTheMagic> dtzWill, oh come on
Jan 23 00:58:44 <PuffTheMagic> cant u install the right way?
Jan 23 00:58:44 <Brybry> :D
Jan 23 00:58:56 <dtzWill> PuffTheMagic: novacom bugs out on libusb on my computer
Jan 23 00:59:05 <dtzWill> since i have to force it to use some ubuntu libusb or some crap
Jan 23 00:59:08 <dtzWill> LD_PRELOAD and such
Jan 23 00:59:09 <PuffTheMagic> use palm-install!
Jan 23 00:59:13 <dtzWill> and that apparently doesn't play nice
Jan 23 00:59:16 <dtzWill> palm-install requires running novacomd
Jan 23 00:59:26 <PuffTheMagic> i thought u meants precom
Jan 23 00:59:30 <PuffTheMagic> do u have latest novacom?
Jan 23 00:59:42 <dtzWill> and that makes my system unstable.  so stable way to use novacom is to plug in my stuff to my media center across the room... and i'm much too lazy for that as part of my dev cycle, yes haha :)
Jan 23 00:59:57 <dtzWill> uh i think so. i can double check. i did as of when the TP came out, doubt i've updated since then
Jan 23 01:01:01 <PuffTheMagic> Brybry, there is something wonky about the respawn code
Jan 23 01:01:34 <Brybry> I tested it pretty thoroughly D:
Jan 23 01:01:41 <Brybry> though I had to modify it to handle the new login stuff
Jan 23 01:01:52 <Brybry> but I assume that didn't break anything
Jan 23 01:01:56 <PuffTheMagic> shell history seems to not get saved
Jan 23 01:02:13 <PuffTheMagic> if I type dmesg; exit
Jan 23 01:02:26 <PuffTheMagic> when i log back in the last command is not exit
Jan 23 01:03:57 <Brybry> history seems to work great here with sh/busybox =/
Jan 23 01:04:30 <PuffTheMagic> not with bash :/
Jan 23 01:06:51 <Brybry> you could try uncommenting the terminal_save_settings lines in newLogin()
Jan 23 01:07:09 <Brybry> I haven't really looked up what they do so I didn't include them
Jan 23 01:07:17 <Brybry> because everything looked super
Jan 23 01:07:58 <Brybry> I have run into the dtzWill vkb input broke bug
Jan 23 01:08:12 <PuffTheMagic> ohhh
Jan 23 01:08:15 <Brybry> well, it works but the graphics aren't updating with it
Jan 23 01:08:19 <PuffTheMagic> i know what it is i think
Jan 23 01:08:22 <PuffTheMagic> delete the app
Jan 23 01:08:24 <PuffTheMagic> then reinstall
Jan 23 01:08:33 <PuffTheMagic> or
Jan 23 01:08:34 <PuffTheMagic> before that
Jan 23 01:08:37 <PuffTheMagic> if u have latesty
Jan 23 01:08:40 <PuffTheMagic> try going into prefs
Jan 23 01:08:43 <PuffTheMagic> and hit restore defaults
Jan 23 01:08:51 <PuffTheMagic> but if that dont work delete then reinstall
Jan 23 01:08:54 <Brybry> am I going to regret this
Jan 23 01:09:33 <Brybry> and what is this supposed to fix
Jan 23 01:10:08 <dtzWill> Brybry: it's not just not being updated, see if you hit 'enter' the display updates
Jan 23 01:10:26 <Brybry> what happened for me was what I typed stopped echoing
Jan 23 01:10:27 <dtzWill> for me, anyway, definitely doesn't send most of the characters
Jan 23 01:10:31 <dtzWill> PuffTheMagic: i'll try
Jan 23 01:11:09 <dtzWill> restore defaults is greyed out :(
Jan 23 01:11:15 <PuffTheMagic> pull
Jan 23 01:11:15 <Brybry> but like ctrl+c worked fine and updated the display fine and 'ls' output showed up fine just not the typed ls characters
Jan 23 01:11:30 <PuffTheMagic> there have been 1000 updated today
Jan 23 01:11:38 <PuffTheMagic> so people need to update the right way ;)
Jan 23 01:12:15 <dtzWill> also, Brybry: lmk if the redraw touchup puff just merged in helps with idle times at all
Jan 23 01:12:21 <PuffTheMagic> cmatrix killed wterm
Jan 23 01:12:28 <PuffTheMagic> after merge
Jan 23 01:12:32 <dtzWill> should for /sure/ stop us from getting behind when om nom'ing on updates
Jan 23 01:12:34 <PuffTheMagic> after running for a min or 2
Jan 23 01:12:57 <dtzWill> the change just delays a redraw! lol
Jan 23 01:13:14 <dtzWill> https://github.com/PuffTheMagic/wTerm/commit/0debda4e79fee70786bf1d8417f4aa434dce1404 is what i'm taling about, FWIW
Jan 23 01:13:16 <PuffTheMagic> well it could be related to the term dims issue
Jan 23 01:13:28 <Brybry> just going to do a clean install given how much has changed
Jan 23 01:13:45 <PuffTheMagic> all im saying is that wterm running cmatrix on head crashed the plugin
Jan 23 01:13:51 <stbuehler> i really hope this postinst crap doesn't destroy anything...
Jan 23 01:14:15 <dtzWill> PuffTheMagic: yes, and good to know. thanks for sharing, sorry haha :)
Jan 23 01:14:35 <PuffTheMagic> stbuehler, why would it?
Jan 23 01:14:48 <PuffTheMagic> i dont touch anything related to the system
Jan 23 01:15:05 <Brybry> PuffTheMagic: does / get remounted rw automatically before postinstall is called or something?
Jan 23 01:15:10 <PuffTheMagic> ya
Jan 23 01:17:43 <PuffTheMagic> dtzWill, the dimensions issue needs to get fixed asap
Jan 23 01:17:47 <Brybry> I can confirm history works fine with sh/busybox on a clean install
Jan 23 01:17:53 <Brybry> I suppose I can install bash later and see if I can see what's up
Jan 23 01:18:22 <dtzWill> PuffTheMagic: which is this? the not setting the correct dimensions on startup issue, or the race between whatever changes the dimensions and other events, causing crashness?
Jan 23 01:18:47 <PuffTheMagic> what race?
Jan 23 01:18:53 <PuffTheMagic> there shouldnt be a race
Jan 23 01:19:06 <dtzWill> oh i'm assuming there is one, as opposed to something being just wrong.
Jan 23 01:20:00 <Brybry> hello invisible input bug
Jan 23 01:20:04 <Brybry> maybe login is just broken
Jan 23 01:20:04 <dtzWill> we used to crash while running cmatrix and changing dimensions--i always assumed it was because cmatrix was relenetless updating the screen, and we changed the nature of the buffer underneath it, causing breakage.  Sorry, i don't know that's the case a tall
Jan 23 01:20:08 <Brybry> it's very possible
Jan 23 01:20:17 <PuffTheMagic> i guess its just not getting set right at startup
Jan 23 01:21:45 <Brybry> dtzWill: you probably only need one good clean install with an ipk before you can go back to your updating binary only ways
Jan 23 01:22:00 <dtzWill> Brybry: yep, for sure.
Jan 23 01:22:16 <dtzWill> and it looks like mismatch between input handler and some updated keyboardness
Jan 23 01:22:32 <dtzWill> since changing the 'case 0' in the keysym switch to 'default' made things work again
Jan 23 01:22:51 <dtzWill> (appraently we explicitly set sym to 0 when using unicode input now)
Jan 23 01:22:56 <dtzWill> probably some stale file(s), shrug
Jan 23 01:35:52 <PuffTheMagic> i dont get why the font size is being weird
Jan 23 01:36:43 <stbuehler> hrmpf, my seqparser rewrite isn't working -.-
Jan 23 01:36:59 <PuffTheMagic> stbuehler, u are rewriting that?
Jan 23 01:37:11 <PuffTheMagic> because there are changes in it that I need supported
Jan 23 01:38:04 <stbuehler> ah, now it works. forgot to reset numValues
Jan 23 01:38:26 <PuffTheMagic> the array of values needs to be char*'s not int*
Jan 23 01:38:29 <stbuehler> utf-8 support is near, at least for terminalstate.
Jan 23 01:38:51 <stbuehler> for what do you need chars ?
Jan 23 01:39:21 <PuffTheMagic> echo -ne "\033]0;${USER}@${HOSTNAME}: ${PWD}\007"
Jan 23 01:39:24 <stbuehler> you do not really want to support OSC? ?
Jan 23 01:39:24 <PuffTheMagic> sequences like that
Jan 23 01:39:44 <PuffTheMagic> is that a question or statement?
Jan 23 01:39:47 <stbuehler> do you really need the value?
Jan 23 01:39:53 <stbuehler> my new parser just throws it away
Jan 23 01:40:03 <PuffTheMagic> throws what away?
Jan 23 01:40:09 <stbuehler> skips the input
Jan 23 01:40:19 <PuffTheMagic> ya, then it shows up in the terminal
Jan 23 01:40:22 <PuffTheMagic> which is not right
Jan 23 01:40:23 <stbuehler> no
Jan 23 01:40:28 <stbuehler> i said "skip"
Jan 23 01:40:34 <stbuehler> not "handle as text"
Jan 23 01:40:52 <stbuehler> as it will skip unsupported CSI sequences
Jan 23 01:40:57 <stbuehler> no need to display them imho
Jan 23 01:41:20 <PuffTheMagic> might find a use for them some day
Jan 23 01:41:32 <stbuehler> then we will implement it that day :)
Jan 23 01:41:43 <PuffTheMagic> there is also a vt52 command I cant implement with the current parser
Jan 23 01:41:53 <stbuehler> you could store the string from osc sequences and parse them too if you really want to
Jan 23 01:42:01 <stbuehler> i even thinkn it will be easier in my new parser
Jan 23 01:42:38 <PuffTheMagic> CS_VT52_CURSOR_POSITION
Jan 23 01:43:04 <PuffTheMagic> that one is tricky because the values come after the identifying token instead of before like eveyrthing else
Jan 23 01:43:32 <PuffTheMagic> its like \033Yrc
Jan 23 01:44:09 <stbuehler> how is it terminated?
Jan 23 01:44:21 <stbuehler> do you have an example?
Jan 23 01:44:37 <PuffTheMagic> its fixed length
Jan 23 01:44:41 <PuffTheMagic> just like i showed
Jan 23 01:44:54 <stbuehler> hm
Jan 23 01:44:55 <PuffTheMagic> the r and c are ascii chars
Jan 23 01:45:01 <PuffTheMagic> they get converted to a row and col
Jan 23 01:45:02 <stbuehler> http://rtfm.etla.org/xterm/ctlseq.html says otherwise
Jan 23 01:45:12 <stbuehler> Ps: A single (usually optional) numeric parameter, composed of one of more digits.
Jan 23 01:45:19 <stbuehler> ESC Y Ps Ps    Move the cursor to given row and column.
Jan 23 01:45:42 <stbuehler> but what you say makes more sense somehow
Jan 23 01:46:05 <PuffTheMagic> CS_VT52_CURSOR_POSITION
Jan 23 01:46:46 <PuffTheMagic> Direct Cursor Address
Jan 23 01:46:46 <PuffTheMagic> ESC  Y   line  column
Jan 23 01:46:46 <PuffTheMagic> 033 131  ***   ***
Jan 23 01:46:46 <PuffTheMagic> Moves cursor to specified line and column. Line and column numbers are ASCII character whose codes are their octal value plus octal 037. For example, line 1 column 8 parameters are octal 040 (first line) and octal 050 (eighth column).
Jan 23 01:47:37 <stbuehler> uhm?
Jan 23 01:47:53 <stbuehler> so is it \033Y040050 or \033Y\040\050 ?
Jan 23 01:48:13 <PuffTheMagic> lthe latter
Jan 23 01:48:36 <PuffTheMagic> i believe
Jan 23 01:48:45 <stbuehler> so just value + 31
Jan 23 01:48:50 <PuffTheMagic> but i havent been able to test that because i cant parse it
Jan 23 01:48:59 <stbuehler> so you don't end up with control chars
Jan 23 01:49:00 <PuffTheMagic> but there is a test in vttest for vt52
Jan 23 01:49:03 <stbuehler> k
Jan 23 01:49:05 <PuffTheMagic> its a hilarious test too
Jan 23 01:49:13 <PuffTheMagic> write foobar all over the screen
Jan 23 01:56:16 <stbuehler> if i'd only had something to compare.. i never know what i did break and what didn't work anyway :)
Jan 23 01:56:52 <PuffTheMagic> compare vttest on your dev machine to vttest in wterm
Jan 23 02:02:01 <PuffTheMagic> stbuehler, there is something else u need to consider for your rewrite
Jan 23 02:03:13 <PuffTheMagic> in vttest
Jan 23 02:03:16 <PuffTheMagic> menu 1
Jan 23 02:03:29 <PuffTheMagic> the "Test of cursor-control chars inside ESC sequences"
Jan 23 02:03:35 <PuffTheMagic> we currently cant do that
Jan 23 02:03:59 <PuffTheMagic> would be nice to pass that
Jan 23 02:04:07 <PuffTheMagic> dtzWill, fixed the term not starting on right size issue
Jan 23 02:05:00 <stbuehler> already on it
Jan 23 02:05:11 <PuffTheMagic> awesome
Jan 23 02:13:21 <dtzWill> PuffTheMagic: great. does cmatrix still die?
Jan 23 02:28:15 <stbuehler> i don't get why it blanks my line when i press up (with my new parser9
Jan 23 02:28:22 <PuffTheMagic> doesnt look like it
Jan 23 02:29:29 <PuffTheMagic> cmatrix is now smooth like butter
Jan 23 02:29:33 <PuffTheMagic> dtzWill: ^^
Jan 23 02:30:49 <dtzWill> PuffTheMagic: ^.^ :D
Jan 23 02:34:28 <PuffTheMagic> cant wait formnew parser
Jan 23 02:37:54 <dtzWill> o_O whoa wterm user haha
Jan 23 02:37:59 <dtzWill> also, installing via ipk did the trick
Jan 23 02:38:04 <stbuehler> if i push some commits, can you pick the ones you need? i'm really tired, no time to fix them (3:36am here)
Jan 23 02:39:53 <Brybry> mmm, dtzWill 0 should be SDLK_UKNOWN
Jan 23 02:40:06 <stbuehler> PuffTheMagic: have a look at https://github.com/stbuehler/wTerm/commits/wip (not all commits should be merged, be careful)
Jan 23 02:40:09 <Brybry> we probably just need to pass the sym in default: right? because then it's known
Jan 23 02:40:52 <stbuehler> Brybry: i removed most of he sym passing, as the plugin needs sym only for special keys like F*, ESC, SHIFT, ENTER, ...
Jan 23 02:41:06 <stbuehler> gn8
Jan 23 02:41:11 <Brybry> usb/bt keyboards are broken~
Jan 23 02:41:37 <dtzWill> Brybry, stbuehler: just put the 'default' above the 'case 0'....?
Jan 23 02:41:46 <Brybry> noooo
Jan 23 02:41:54 <Brybry> see case 0 is SDLK_UKNOWN
Jan 23 02:41:59 <Brybry> (Which should probably actually be changed to that)
Jan 23 02:42:08 <Brybry> and so if it's NOT unknown it'll hit default
Jan 23 02:42:19 <Brybry> and then we just pass it from there
Jan 23 02:42:29 <stbuehler> either figure it out yourself or wait until i'm up again :P
Jan 23 02:42:37 <dtzWill> stbuehler: kk night
Jan 23 02:42:50 <dtzWill> Brybry: what i'm saying would work, but is equivalent to reverting the use of the case 0 to just 'default'
Jan 23 02:43:50 <Brybry> I don't think case 0 would ever be hit then
Jan 23 02:43:56 <PuffTheMagic> dtzWill: ya, wterm user
Jan 23 02:43:58 <Brybry> unless you don't break;
Jan 23 02:44:02 <Brybry> which I guess would work
Jan 23 02:44:06 <Brybry> but it's ugly and not really needed
Jan 23 02:44:10 <Brybry> as every keyboard keypress has a sym
Jan 23 02:44:19 <Brybry> kinda the whole point of SDL input~
Jan 23 02:44:51 <dtzWill> Brybry: except stbuehler's migrating the input to use unicode for everything, in which case it generates events with sym==SDLK_UNKNOWN, and unicode set appropriateyl
Jan 23 02:45:01 <dtzWill> anyway the solution is the same, and idc what you fancy better lol
Jan 23 02:45:11 <dtzWill> just combine the 'default' case with the 'case 0' case and all should be well, yes?
Jan 23 02:45:38 <dtzWill> (including doing so by replacing the text "case 0" with "default")
Jan 23 02:45:56 <dtzWill> why are we having this discussion.  anyway, hopefully that fixes bt and such and all is good.
Jan 23 02:46:04 <Brybry> ah, I get your point though I'm unsure if there is proper unicode generated for every keyboard input
Jan 23 02:46:07 <Brybry> I'll have to look at it again
Jan 23 02:47:32 <Brybry> if setting the base case to default works fine then I see no reason not to do that
Jan 23 02:47:33 <dtzWill> well it's how we've been handling it before stbuehler's changes, although i don't know either.
Jan 23 02:47:34 <Brybry> it makes the most sense
Jan 23 02:47:38 <Brybry> and you're quite right
Jan 23 02:54:09 <dtzWill> \o/
Jan 23 02:56:40 <PuffTheMagic> wterm cpu usage has increased
Jan 23 02:56:56 <Brybry> yeah, I'm going to look into it :(
Jan 23 02:57:14 <Brybry> assuming we dont' have new threads then something is firing events/causing us not to sit blocking
Jan 23 02:57:26 <dtzWill> cpu usage increased while active, or doesn't idle as well?
Jan 23 02:57:30 <dtzWill> (or both?)
Jan 23 02:57:37 <Brybry> it was idle for me, I dunno about active
Jan 23 02:57:46 <Brybry> I also still have invisible input stuff
Jan 23 02:58:03 <PuffTheMagic> idk, ignore my comment foe now
Jan 23 02:58:22 <Brybry> if I let the login prompt sit around for like 2 minutes my input eventually doesn't echo
Jan 23 02:58:23 <dtzWill> PuffTheMagic: lol at 2bd3b0568e16253ae61b47f8818b235756fc049d xD
Jan 23 02:58:50 <dtzWill> Brybry: is this a new development, do you know?
Jan 23 02:58:57 <PuffTheMagic> dtzWill: is that the makefile.in commit?
Jan 23 02:59:02 <dtzWill> PuffTheMagic: yeah haha
Jan 23 02:59:08 <PuffTheMagic> :)
Jan 23 02:59:23 <Brybry> dtzWill, I think so but we didn't use login interactively before
Jan 23 02:59:28 <Brybry> and for all I know it's just /bin/login misbehaving
Jan 23 02:59:37 <Brybry> I should try a 'reset' next time it happens
**** ENDING LOGGING AT Mon Jan 23 02:59:57 2012