**** BEGIN LOGGING AT Wed May 30 02:59:58 2012
May 30 03:09:50 <mheld> or, does anybody here use activeandroid?
May 30 03:15:27 <dlaroche> Does anyone have a code example of how to get the bearing from one location to another?
May 30 03:16:36 <b1n0ry> "Whenever you fire an AsyncTask from an Activity, which is mostly the point, there’s a chance that by the time onPostExecute() is called the original context for the task no longer exists." <-- likely?
May 30 03:17:05 <b1n0ry> i guess an orientation change during doInBackground() or something?
May 30 03:17:08 <JakeWharton> very
May 30 03:17:08 <pfn> it's possible
May 30 03:17:12 <JakeWharton> precisely
May 30 03:17:45 <pfn> well
May 30 03:17:49 <b1n0ry> pfft... i'm actually learning.
May 30 03:17:50 <pfn> the original context will /always/ exist
May 30 03:17:58 <pfn> it might not do anything if you call methods on it
May 30 03:18:05 <pfn> the asynctask has a strong reference to the context
May 30 03:18:11 <dlaroche> JakeWharton, by putting the AsyncTask in the onResume you could overcome most problems correct?
May 30 03:18:16 <pfn> it may have had its destroy called
May 30 03:18:23 <pfn> dlaroche, irrelevant
May 30 03:18:44 <b1n0ry> whose to say that even after resume, they won't just change the orientation again...
May 30 03:18:50 <JakeWharton> there are various methods of dealing with the context disappearing during an asynctask's execution
May 30 03:18:50 <b1n0ry> who's
May 30 03:19:09 <dlaroche> You would think an AsyncTask would know when the calling context has ended.
May 30 03:19:11 <b1n0ry> i found a blog post that looks promising. looks like the guy abstracted the task to run in its own fragment
May 30 03:19:20 <mheld> can't you just get the application context from the context?
May 30 03:19:28 <JakeWharton> you can hold the parent context in a weak reference
May 30 03:19:29 <pfn> depends on what you're doing
May 30 03:19:42 <pfn> for the most part, having a stale context is really no big deal
May 30 03:19:42 <JakeWharton> re-register with the updated context in onCreate
May 30 03:19:56 <dlaroche> I mean you can't "end" or "stop" an AsynTask can you?
May 30 03:20:04 <b1n0ry> why can't you?
May 30 03:20:04 <pfn> you can request it be cancelled
May 30 03:20:05 <JakeWharton> you can ask it to cancel
May 30 03:20:14 <dlaroche> i see
May 30 03:20:20 <b1n0ry> if your doInBackground is checking whether or not it's cancelled and the ui thread sends a cancel
May 30 03:20:56 <b1n0ry> http://blogactivity.wordpress.com/2011/09/01/proper-use-of-asynctask/ <-- for my lack of knowledge, this guy's solution seems reasonable
May 30 03:21:13 <dlaroche> good information for me. Never experienced any problems with using an AsyncTask so just curious questions from me
May 30 03:22:07 <dlaroche> Does Google employee's ever visit this room?
May 30 03:24:00 <b1n0ry> if you don't write doInBackground() to check cancel status in its loop or during its process, and you send a cancel, the cancel doesn't do anything right?
May 30 03:25:05 <pfn> yes, it's cooperative
May 30 03:25:06 <dlaroche> I would assume not, because a regular thread doesn't either unless you have the loop check the state of a variable before it loops
May 30 03:25:07 <pfn> dlaroche, never
May 30 03:25:12 <g00s> the asynctask won't have a hard reference to the original task if you do things correctly :)
May 30 03:25:26 <g00s> <.<
May 30 03:27:11 <dlaroche> pfn: should have expected that answer...they made a comment in one of their videos about how they were going to start updating their blog more religiously, however, that came and went
May 30 03:30:31 <pfn> dlaroche, fyi, romainguy is one of the main android developers at google
May 30 03:30:55 <dlaroche> oh...lol...sorry didn't realize your were being sarcastic on your answer
May 30 03:31:40 <dlaroche> None the less, they did start neglecting their developer blog though
May 30 03:35:52 <b1n0ry> sarcasm is often difficult to pick up on in text
May 30 03:37:18 <b1n0ry> alright, i was out biking a bit ago so i'm heading to the shower and then to do some building... things are really starting to come together with my app. thank you guys for all the help thus far.
May 30 03:52:36 <CQN> is kinetic scrolling not standard in the android api?
May 30 03:53:10 <CQN> i implemented a TextView with the movement method being ScrollingMovementMethod, but it's basically unusable
May 30 04:38:50 <cr5315> I'm trying to load text into my app from a file stored on the internet, but no matter how I try to implement it, the app always crashes. Here's what I'm trying to do: http://goo.gl/UPvOC
May 30 04:39:07 <bankai_> cr5315: stack trace ...
May 30 04:39:12 <cr5315> One moment
May 30 04:40:35 <bankai_> and do you have internet perms in your manifest?
May 30 04:40:45 <cr5315> Yes
May 30 04:42:11 <dragorn> are you trying to do the network io in your main thread?
May 30 04:42:27 <dragorn> (stack dump will make this obvious if so)
May 30 04:43:41 <cr5315> Logcat isn't logging a stack trace
May 30 04:52:09 <pragma-> i'd like to propose a proposition
May 30 04:52:59 <Knossos> pragma-: I'm sorry, I can't marry you.
May 30 04:54:10 <b1n0ry> i can
May 30 04:54:12 <Knossos> Also, I wish you could get in contact with people who write bad ratings / experience errors. Most of the time the most helpful thing they write is "There was an error!". You don't say!?
May 30 04:54:42 <b1n0ry> as long as you're not in ohio. i'm already married in ohio.
May 30 05:02:24 <bif_1964> i have an app on play. how do you get automatic updates or a notification about there is an update. I have the old version installed on my device and if i go to google play and find my app, i can click update, but i get no notification about it
May 30 05:04:43 <dragorn> there's a pref to control if you get any notifications; other than that, you get them whenever it feels like giving you them
May 30 05:06:41 <g00s> someone asked me to forward them a receipt for their purchase; anybody know what that might be exactly?
May 30 05:10:56 <bif_1964> @dragorn. So i have not missed a setting, when i uploaded the app again to google play? it will happen automatically if people allow it?
May 30 05:11:00 <BloodSkin> its sent to their email address when they make the purchase
May 30 05:11:06 <BloodSkin> if its bought through the market
May 30 05:11:33 <g00s> BloodSkin: i thought so, thats why i am confused
May 30 05:15:01 <g00s> hmm, well i see an order that is 'chargeable' (Order received, authorization successful) but it hasn't transitioned to sharged/shipped … anyone know why that would be ?
May 30 05:15:11 <g00s> *charged
May 30 05:16:08 <cr5315> Is it possible for a crash to not print stack traces?
May 30 05:18:15 <jasta> cr5315: no.  in fact, no way in which the process dies unexpectedly can it avoid printing.
May 30 05:18:27 <cr5315> Well, it's not printing
May 30 05:18:36 <jasta> if it fails in a structured fashion in the VM, it will print the stack.  if it explodes elsewhere, it will print signfiicantly more insane messages
May 30 05:18:37 <doogan> perhaps you're filtering the output
May 30 05:18:39 <jasta> but print none-the-less
May 30 05:19:07 <jasta> cr5315: it is possible that whatever dumbshit manufacturer built your phone they've disabled or obscured the output somehow.
May 30 05:19:18 <cr5315> Curse you HTC
May 30 05:19:25 <doogan> nah
May 30 05:19:37 <cr5315> Oh there it is
May 30 05:20:30 <Daniel-B> hey
May 30 05:20:40 <cr5315> bankai_, so.. here's that stack trace.. https://dl.dropbox.com/u/3270519/log.txt
May 30 05:20:54 <Daniel-B> I'm trying to create application that sets a new apn
May 30 05:21:09 <jasta> cr5315: well, there you go.  tells you plainly what you've done.
May 30 05:21:31 <Daniel-B> I'v got some code but application forces close
May 30 05:21:47 <bankai_> cr5315: fire it off in another thread
May 30 05:21:52 <doogan> the ol NOMTE
May 30 05:22:12 <cr5315> Like an AsyncTask?
May 30 05:22:19 <bankai_> si si
May 30 05:22:22 <cr5315> Alright
May 30 05:22:23 <g00s> jasta: what did you ever decide about the actionbar saga a while back? did you abandon pre-ics altogether?
May 30 05:22:23 <cr5315> Thanks
May 30 05:22:43 <jasta> g00s: i did, i decided to use the old non action bar UI pattern for old devices
May 30 05:22:47 <jasta> keep it hidden away in shitty menus
May 30 05:23:02 <jasta> 6 months from now, no one will hardly care that i've done this :)
May 30 05:23:29 <g00s> GB is still 70% of my user base :)
May 30 05:23:29 <bankai_> Daniel-B: stack trace
May 30 05:23:57 <jasta> g00s: yeah, but the way i figure it, they're used to shitty UX :)
May 30 05:24:21 <jasta> i would've liked to do differently, but actionbarsherlock was just doing it wrong.
May 30 05:24:30 <Daniel-B>  bankai_ I'm sorry what?
May 30 05:24:45 <g00s> jasta: interesting
May 30 05:24:50 <bankai_> Daniel-B: show us a stack trace ....
May 30 05:25:05 <Daniel-B> http://blogs.msdn.com/b/zhengpei/archive/2009/10/13/managing-apn-data-in-google-android.aspx
May 30 05:25:26 <g00s> jasta: were you using abs 4 which dispatches to the native impl on HC and above, or were you using the older 3.x one ?
May 30 05:25:32 <bankai_> that's not a stack trace
May 30 05:25:54 <Daniel-B> I don't know what do you mean
May 30 05:25:58 <Daniel-B> I'm new at this
May 30 05:26:25 <bankai_> Daniel-B: can't help you with a description like "it force closes" ... need code/errors/stack traces etc.
May 30 05:26:34 <Daniel-B> ahh
May 30 05:26:37 <Daniel-B> got you
May 30 05:27:45 <Daniel-B> well it should be in logcat (eclipse) right?
May 30 05:30:32 <g00s> this tim cook guy seems like a likable fella actually
May 30 05:30:49 <g00s> unlike his predecessor 9.9
May 30 05:30:58 <komugi> tim cook is ok
May 30 05:31:06 <komugi> it's the ethics of the company that i have issues with
May 30 05:31:12 <komugi> that was established by jobs
May 30 05:31:28 <g00s> i heard some things are changing there, though
May 30 05:31:38 <g00s> like philanthropy
May 30 05:31:54 <g00s> some things won't change, like secrecy :)
May 30 05:32:09 <komugi> secrecy is fine tbh, to protect company interest
May 30 05:32:12 <g00s> hopefully some things like openness of the platform will change for the better
May 30 05:32:13 <komugi> patent trolling
May 30 05:32:15 <komugi> not so much
May 30 05:33:12 <komugi> g00s the only thing really missing for me on iOS is lack of shared app space
May 30 05:33:17 <komugi> so that all apps can access/modify files
May 30 05:33:28 <komugi> if they open that up, I would be using iOS
May 30 05:33:31 <g00s> komugi: yeah, look no further than "Rockstar consortium" for that
May 30 05:33:57 <komugi> lol
May 30 05:35:14 <g00s> komugi: it will be very interesting. cook thinks the tablet form factor is unique enough it needs its own OS, meanwhile MS is going another way …
May 30 05:35:15 <sunshadow> Hi all, need some help in building libevent on NDK.
May 30 05:35:33 <komugi> I think MS is on to something :)
May 30 05:35:46 <komugi> but I have a bad feeling that the metro app store will fall flat
May 30 05:36:02 <komugi> and desktop interface will alienate their customers
May 30 05:36:03 <komugi> haha
May 30 05:36:19 <g00s> i dunno, but i am excited about a tablet that runs regular apps. if it can run LR4 , for example, i will be happy
May 30 05:36:22 <komugi> well, their enterprise customers
May 30 05:36:39 <komugi> i see Win8 going the way of WinVista
May 30 05:36:46 <komugi> people dragging their feet to upgrade
May 30 05:36:48 <komugi> until MS does a 180
May 30 05:37:00 <komugi> yeah, so am I
May 30 05:37:04 <g00s> seems like every other release is hit or miss
May 30 05:37:12 <komugi> but, as I said, people are resistant to change
May 30 05:38:07 <komugi> well, their cycle seems to be "new interface release" -> people whine whine -> "new interface release rev 2 minor upgrades" -> people gets used to it and say "oh, that's not bad"
May 30 05:38:14 <komugi> ;P
May 30 05:38:27 <giraffeslacks> i'm new to Java and android and i find myself making crazy long names for IDs and strings to keep track of what's what. should i force myself to stop this or is this normal? here's an example in strings.xml: form_add_list_name_title.
May 30 05:39:38 <xa0c> yeah. name them str_1, str_2, str_3, etc. much better.
May 30 05:40:07 <vavirta> :o?
May 30 05:40:12 <g00s> hehe
May 30 05:40:25 <Knossos> Nothing wrong with long names, so long as it doesn't go obsene
May 30 05:40:27 <vavirta> I hope that xa0c is trolling ;)
May 30 05:40:27 <giraffeslacks> xa0c: won't that be really hard to keep track of as my app(s) grow?
May 30 05:40:38 <g00s> giraffeslacks: you can look at the iosched app for best practices
May 30 05:40:40 <giraffeslacks> okaaay i see the funny.
May 30 05:40:41 <Knossos> giraffeslacks: do NOT take that advice :D
May 30 05:40:51 <giraffeslacks> i get jokes.
May 30 05:40:52 <xa0c> %)
May 30 05:41:01 <pragma-> sigh
May 30 05:41:14 <pragma-> "On Android 3.0 and higher, the options menu is considered to always be open when menu items are presented in the action bar. When an event occurs and you want to perform a menu update, you must call invalidateOptionsMenu() to request that the system call onPrepareOptionsMenu()."
May 30 05:41:22 <pragma-> meanwhile:
May 30 05:41:31 <pragma-> "public void invalidateOptionsMenu ()
May 30 05:41:31 <pragma-> Since: API Level 11
May 30 05:41:31 <pragma-> Declare that the options menu has changed, so should be recreated. The onCreateOptionsMenu(Menu) method will be called the next time it needs to be displayed.
May 30 05:41:58 <pragma-> d.android.com reference is full of this inconsistencies
May 30 05:42:08 <pragma-> these
May 30 05:43:12 <g00s> pragma-: "its open source, submit a patch"
May 30 05:43:17 * g00s runs
May 30 05:45:26 <g00s> shit, my app is complicated enough people are asking for tutorials
May 30 05:45:58 <tmho> whats the app?
May 30 05:46:16 <g00s> top s3kr3t :D
May 30 05:46:22 <pragma-> g00s: that's a bad sign
May 30 05:46:29 <pragma-> y u no use good ui design?
May 30 05:46:33 <g00s> yeah, its a complicated app
May 30 05:46:49 <g00s> i thought i used good ui design; i actually read quite a bit about the topic
May 30 05:46:57 <g00s> i made the most common tasks easy
May 30 05:47:08 <g00s> but its still a complicated app. took me about a year to write
May 30 05:47:50 <xa0c> next year for writing tutorials than... :)
May 30 05:47:57 <g00s> haha
May 30 05:48:35 <tmho> good ui = limit app functions to 1
May 30 05:49:03 <tmho> people can't handle more than 1
May 30 05:49:20 <g00s> most people can't even handle 1
May 30 05:49:45 <tmho> sadly true
May 30 05:51:18 <sunshadow> Hi guys, need some help with building libevent on NDK. Help please..
May 30 05:57:33 <JakeWharton> does Java the the "/" symbol somewhere as a constant
May 30 05:57:37 <JakeWharton> for URI/URLs
May 30 06:01:38 <g00s> hmm, all the pictures i took today sucked :(
May 30 06:01:56 <Android03> hi guys, i am creating an app where i need to show some markers on the map and need to change the marker ( blue to orange ) while tapping on it.i have used selector file for this which works fine but When i tap on the blue marker it gets changed to orange one,But when i tap on the map (not on the marker or pin) the orange gets changed to blue again.It should remain orange when i tap on the map .I am not getting how can i resolv
May 30 06:06:06 <Android03> please help me it's bit urget
May 30 06:06:13 <Android03> *urgent
May 30 06:06:48 <jasta> it can't be urgent, you're on irc.
May 30 06:07:42 <Android03> jasta
May 30 06:08:26 <Android03> jasta: i thought people are not giving attention on my question
May 30 06:09:03 <giraffeslacks> Android03: they may not be for multiple reasons. malice, though, is usually not one of them.
May 30 06:09:06 <jasta> Android03: my recommendation is to just change out the marker of the items themselves.
May 30 06:09:08 <Knossos> The way it works, is if people know an answer, they will message. :P
May 30 06:09:23 <jasta> setMarker or whatever the method is
May 30 06:09:27 <jasta> controlling these states yourself
May 30 06:10:17 <Android03> i have tried setMarker also but no luck
May 30 06:10:57 <Android03> giraffeslacks: can u tell me what are the reasons?
May 30 06:11:18 <giraffeslacks> umm... really?
May 30 06:11:55 <giraffeslacks> lack of knowledge on the subject, not looking at the chat channel when your question creeps by are two that come to mind.
May 30 06:12:37 <jasta> Android03: you had no "luck"?  what does that mean?
May 30 06:12:47 <jasta> did you buy a setMarker lottery ticket and your number wasn't called?
May 30 06:13:49 <Android03> jasta: i mean i tried with setMarker also but it didn't work
May 30 06:14:03 <jasta> perhaps you didn't understand the sarcasm
May 30 06:14:15 <Android03> ya i understood
May 30 06:14:22 <jasta> setMarker does function.  it changes markers.
May 30 06:14:44 <jasta> you'll need to explain to me what you did.
May 30 06:15:53 <Android03> jasta : this is the selector file http://pastebin.com/6w9hHpkE
May 30 06:16:15 <jasta> do you have a learning disability?
May 30 06:16:29 <jasta> if you do, i can try to adjust my style to work with that.
May 30 06:18:32 <Android03> jasta : and this is how i m changing the marker http://pastebin.com/KDTaBsu5 on onTap
May 30 06:18:56 <jasta> ok, i'm definitely done with you now.
May 30 06:19:12 <jasta> good luck with your RNG approach to software.
May 30 06:19:14 <Android03> jasta : why u guys are so rude
May 30 06:25:21 <g00s> guess 4.1 is jelly bean
May 30 06:28:46 <g00s> wow, factory reset can break google wallet on that device - for good
May 30 06:29:11 <cr5315> I don't see why Google would release Jelly Bean this soon
May 30 06:29:30 <cr5315> s/release/announce
May 30 06:30:26 <cr5315> ICS is still on < 10% of all devices, and a lot that have GB won't be getting ICS.
May 30 06:31:33 <Knossos> I'll only be able to get the next Android when my contract runs out :(
May 30 06:31:48 <Knossos> Well, unless I decide to root my device, anyway
May 30 06:31:54 <jasta> hehe, subsidies?  people still do that?
May 30 06:32:17 <Ge0rG> g00s: wallet is strange anyway :)
May 30 06:32:32 <g00s> its fucking lame that a factory reset can brick wallet on that device forever
May 30 06:32:34 <Knossos> Is there any news about what will be "special" about jelly Bean?
May 30 06:32:41 <Ge0rG> g00s: it seems they still haven't figured out which data to store on the SE and which in the app prefs
May 30 06:33:26 <g00s> Knossos: there will be new icon and design guidelines i bet :D
May 30 06:33:30 <Knossos> ¬¬
May 30 06:33:35 <JakeWharton> cr5315: it's been 6 months since ICS
May 30 06:33:42 <JakeWharton> hence, Jelly Bean
May 30 06:33:55 <JakeWharton> devices on GB that aren't getting ICS is an OEM problem, not Google
May 30 06:33:57 <Jak_o_Shadows> I personally coulnd't see much difference between the new and old icon design guidelines.
May 30 06:34:03 <jasta> the android team just got hungry, that's all.
May 30 06:34:06 <JakeWharton> ICS is on 100% of Google experience devices
May 30 06:34:08 <wabz> I don't know about that JakeWharton
May 30 06:34:30 <jasta> JakeWharton: not the N1
May 30 06:34:32 <g00s> hm, but google experience isn't what it used to be, at least on VZ
May 30 06:34:40 <wabz> I think it's partly google's problem if they keep pumping out new api's/features taht developers won't/can't use
May 30 06:34:42 <JakeWharton> jasta: you can't buy an N1
May 30 06:34:52 <jasta> JakeWharton: it remains a google experience device.
May 30 06:34:56 <JakeWharton> g00s: that's because the device is from Verizon, not Google
May 30 06:35:00 <jasta> dead dog's still a dog...
May 30 06:35:08 <JakeWharton> no
May 30 06:35:19 <wabz> it'd be interesting to see newer stats for platform distribution than 1st of may
May 30 06:35:27 <wabz> I think a few devices have received ICS this month
May 30 06:35:32 <Ge0rG> wasn't the G1 a google experience device as well?
May 30 06:35:44 <jasta> well......sort of :)
May 30 06:35:55 <jasta> it did have T-Mobile myFaves on it, so no :)
May 30 06:36:38 <JakeWharton> so you're saying Microsoft shouldn't release Windows 8 until Gateway is able to support it on the PC I bought in 2001?
May 30 06:36:47 <g00s> JakeWharton: who cares who it is from, its still a nexus device
May 30 06:37:01 <JakeWharton> that's its name
May 30 06:37:03 <jasta> JakeWharton: im not talking about that, you just said it's on all google experience devices. a curiously wrong statement.
May 30 06:37:16 <JakeWharton> so you're arguing semantics because i'm correct?
May 30 06:37:18 <JakeWharton> gotcha
May 30 06:37:49 <Ge0rG> wabz: my app with ~5000 installs has 15% android 4.x users
May 30 06:38:22 <g00s> nice Ge0rG , i wish i had 5000 installs
May 30 06:38:38 <Knossos> my 20k app is 31.33% ICS
May 30 06:38:59 <Knossos> Barely anything left less than 3.2 nowadays
May 30 06:39:11 <JakeWharton> minSdkVersion = 14
May 30 06:39:15 <JakeWharton> 100% ICS
May 30 06:39:19 <JakeWharton> problem solved
May 30 06:39:48 <Jak_o_Shadows> Apart from the discount devices.
May 30 06:39:49 <tmho> lol
May 30 06:40:03 <Jak_o_Shadows> But then, those who buy a $69 phone arn't likely to spend much money on apps anyway.
May 30 06:40:05 <tmho> then who would use ABS
May 30 06:40:27 <JakeWharton> tmho: no one, and I would actually sleep at night
May 30 06:40:30 <JakeWharton> :D
May 30 06:40:35 <Ge0rG> also, install base growth in the last month: 6.5%, ICS growth in the last month: 34%
May 30 06:40:38 <tmho> lol
May 30 06:40:56 <Ge0rG> 4.0.3+ went from 501 to 695
May 30 06:40:56 <tmho> ABS make you lose sleep?
May 30 06:41:14 <Ge0rG> Knossos: is it a tablet app?
May 30 06:41:15 <Knossos> JakeWharton: Btw, do you actually make anything from ABS? Or are you just Good guy Greg?
May 30 06:41:25 <JakeWharton> The company that I work for uses it
May 30 06:41:32 <JakeWharton> but I have no personal apps that do
May 30 06:41:49 <Knossos> Ge0rG: Both phone and tablet
May 30 06:42:05 <jasta> speaking of common code
May 30 06:42:15 <jasta> i've recently started working on something like iOS' CATiledLayer
May 30 06:42:23 <Ge0rG> Knossos: because most of my apps still peak on 2.3
May 30 06:42:24 <jasta> would anyone find this useful?
May 30 06:42:30 <JakeWharton> tmho: I probably lost 2-3 hours of sleep a night from Novermber 17th to March 7th, the development window of ABS 4.x.
May 30 06:42:35 <jasta> thought about open sourcing it.  nothing preventing me from that.
May 30 06:42:42 <Jak_o_Shadows> jasta: How's the webcam streaming thing going?
May 30 06:42:43 <jasta> just not sure how obscure such a thing would be
May 30 06:42:48 <Knossos> 2.3 for me is 0.40%
May 30 06:42:50 <jasta> Jak_o_Shadows: oh man, if only i had time for that :(
May 30 06:43:01 <Knossos> 2.3.3+ is 50.71%
May 30 06:43:04 <jasta> Jak_o_Shadows: when i'm between jobs next, i've already decided i'm taking an entire month off
May 30 06:43:49 <Ge0rG> Knossos: yeah, 50.19% for 2.3.3 here
May 30 06:44:02 <Jak_o_Shadows> d'aw. I'd help, if I knew what I was doing.
May 30 06:44:35 <jasta> Jak_o_Shadows: it's really just about the availability of my time these days.  i'm stupidly busy at work these days.
May 30 06:45:00 <jasta> frankly though i don't work that many hours, i just burn out fast :)
May 30 06:50:59 <Jak_o_Shadows> I should really at least read through that code.
May 30 06:52:00 <g00s> pretty cool http://phandroid.com/2012/05/29/panasonic-lumix-remote-companion-app-for-dmc-fx90-launches/
May 30 07:06:39 <g00s> jasta: sounds like you are doing iOS stuff now too
May 30 07:12:31 <jasta> g00s: not me personally, but i am lead over our entire mobile team including the ios guys
May 30 07:13:03 <jasta> iOS, btw, has such a surprising number of little UI niceties that android really has no excuse not to provide this far into development
May 30 07:13:17 <jasta> little things like a convenient and simple arbitrary scrolling and pinch zoom controller
May 30 07:13:28 <Knossos> I always heard that iOS UI work was more of a pain. Is that a wrong statement?
May 30 07:13:52 <jasta> but on the other hand, some of this kind of shoehorns iOS devs into the hammer and nail sort of thinking.  it's like when they have to sack the common stuff and write their own controls they just fall apart.
May 30 07:14:10 <jasta> it's as if that is impossible for most iOS devs :)
May 30 07:15:22 <hrnt> it is a pain to go against the iOS default look, much more difficult than doing it on android
May 30 07:16:24 <jasta> well, it is certainly easier to be cohesive
May 30 07:16:30 <hrnt> if you want to have a navigation controller for example that is almost the same as the default one, you would probably need to write one from scratch
May 30 07:17:01 <hrnt> if you do that on iOS 4, then you also need to rewrite all the UI controllers because the default navigation controller uses APIs that you can't use
May 30 07:17:09 <g00s> i think the phone should have system wide themes you can choose from
May 30 07:17:14 <g00s> (as a user)
May 30 07:17:25 <g00s> some may prefer light, others dark
May 30 07:17:33 <hrnt> and all the default UI stuff is more complex than it looks like
May 30 07:17:38 <jasta> g00s: hehe, i wrote that support for CyanogenMod :)
May 30 07:17:45 <g00s> :)
May 30 07:18:08 <hrnt> the table view background is not white, the "white" text is actually slightly blue and so on
May 30 07:18:32 <jasta> hrnt: i will say this, iOS development is suspiciously archaic.
May 30 07:18:49 <jasta> i suppose using Java on mobile devices was a really revolutionary idea afterall :)
May 30 07:19:08 <hrnt> well, i actually like the iOS frameworks a LOT more than the android frameworks
May 30 07:19:16 <alankila> java is clearly a 90s or even 2000s technology (if you are kind)
May 30 07:19:21 <alankila> objC is a 80s tech :-p
May 30 07:19:25 <jasta> i was referring actually to the language, runtime, and tools.
May 30 07:19:30 <hrnt> i even like obj-c more than android's version of java
May 30 07:19:36 <Ge0rG> I wish the java language designers would have been a little more... revolutionary
May 30 07:20:02 <jasta> some of the stuff obj-c developers deal with blows my mind :)
May 30 07:20:08 <hrnt> i can use about 100MB RAM in a iOS app for 3GS. in android, maybe 20MB.
May 30 07:20:18 <jasta> oh, without this linker flag, our app crashes on iOS version <what the hell ever>.  like, are you serious?
May 30 07:20:23 <alankila> hrnt: you can request large heap since 3.0
May 30 07:20:26 <hrnt> i know
May 30 07:20:37 <hrnt> but i can't really target only ICS+
May 30 07:20:47 <alankila> not yet, I admit.
May 30 07:21:02 <alankila> in a year, perhaps you can.
May 30 07:21:19 <hrnt> optimist ;)
May 30 07:21:26 <jasta> hrnt: course, as someone that has worked with an iOS app using 100MB of RAM, you get a whole host of other insane problems on iOS as a result.
May 30 07:21:46 <hrnt> yeah, well, the same thing happens if you use 20MB of RAM consistently in an android app ;)
May 30 07:21:47 <jasta> basically, your cavaliar attitude to memory pressure harms users...
May 30 07:21:49 <hrnt> memory fragmentation for starters
May 30 07:22:08 <jasta> hrnt: the model on android is designed to protect the system from your stupidity, though ;)
May 30 07:22:29 <Ge0rG> on iOS, the app store review team is protecting the system from your stupidity
May 30 07:22:38 <jasta> Ge0rG: hehe, no it's not :)
May 30 07:22:58 <Ge0rG> jasta: "your" as in "the developers"
May 30 07:23:06 <jasta> yes.
May 30 07:23:16 <Ge0rG> ...actually also from novel business models, like flattr :>
May 30 07:23:24 <alankila> but it's lovely that we can talk about a java app and say with seriousness that it runs in 20 MB of memory. Not a whole lot of java applications do, outside android
May 30 07:23:32 <hrnt> alankila: indeed
May 30 07:23:34 <jasta> i'm telling you, i inherited a batch of really shitty iOS apps, they really let you get away with murder on the iOS store.
May 30 07:23:48 <hrnt> that is actually why i like obj-c over java
May 30 07:23:54 <hrnt> obj-c is a much better fit for constrained memory
May 30 07:23:55 <alankila> This must be so embarrassing for all the folks who hate java and call it bloated. Android just isn't like that at all, imho.
May 30 07:24:25 <jasta> hrnt: honestly, the things taking up all the memory on both platforms are exactly the same.
May 30 07:24:33 <jasta> bitmaps, and bloated ass UI hierarchies.
May 30 07:24:33 <hrnt> not true
May 30 07:24:41 <hrnt> that depends on the app
May 30 07:24:54 <alankila> when I wrote a little newspaper viewer for ios and android last year I did not have to even think about memory usage on ios because I got 256 MB to use on that platform
May 30 07:25:08 <alankila> I didn't waste the memory, I freed what wasn't in view, etc. but it never even came up
May 30 07:25:17 <alankila> on android, I upload textures to GPU to get them out of my heap as soon as possible
May 30 07:25:18 <jasta> alankila: see my earlier comment about android protecting users from your stupidity ;)
May 30 07:25:45 <alankila> jasta: nah, honestly, the images I had to deal with were around 8 MB a pop, no way to really get around that
May 30 07:25:51 <jasta> the memory pressure situation on both devices is the same.  android just arbitrarily prevents you from causing it.
May 30 07:26:13 <jasta> alankila: what are you talking about?  there are lots of ways around that.  recommended by both platforms, in fact.
May 30 07:26:29 <tmho> errr 8MB image on the phone
May 30 07:26:30 <alankila> jasta: yes... subsampling, partial decoding, etc
May 30 07:26:31 <tmho> ?
May 30 07:26:41 <alankila> maybe use of ericsson texture compression
May 30 07:26:58 <jasta> just parsing and loading an 8MB bitmap is a terrible idea.  why spend the cycles even if you do have the RAM?
May 30 07:27:09 <jasta> that sort of thinking is exactly what i meant about "your stupidity" :)
May 30 07:27:16 <alankila> Anyway I looked into all that shit and in the end I decided to write a singlethreaded 8 MB-allocating decode and upload to GPU code. The GPU is not at all so constrained as the applications are.
May 30 07:28:10 <jasta> alankila: you do know that the apps themselves are arbitrarily contrained, right?  "working around it" using the GPU texture memory is like working around an ANR by yielding every 4s :)
May 30 07:28:19 <jasta> sure, it won't ANR, but....
May 30 07:28:24 <hrnt> speaking of memory usage: let's say i store a lot of ints. on iOS I would use an NSArray of NSNumbers (ints), on android i would use an ArrayList<Integer>
May 30 07:28:53 <hrnt> the android version stresses the memory a lot more
May 30 07:29:13 <jasta> hrnt: uhh, you know the storage costs of the two aren't like an order of magnitude away from each other, right?
May 30 07:29:42 <hrnt> jasta: uhm, what do you mean?
May 30 07:29:46 <jasta> i think maybe you're just inventing this idea because one has <> in the syntax or something.
May 30 07:29:52 <alankila> jasta: there are few things more pleasing than the ability to allocate memory. Honestly, 2 years ago I held a phone in my pocket with 512 MB of memory, of which 340 MB were usable to applications, usually around 250 MB easily freeable by android if required. Today I have a phone with 1 GB in pocket.
May 30 07:30:02 <jasta> hrnt: what do you mean what do i mean?  storage.  order of magnitude.  not.
May 30 07:30:19 <alankila> we should not be talking about ways to optimize 8 MB allocations imho. Not when we talk about gigabyte of memory on the device and one foreground application at a time.
May 30 07:30:26 <hrnt> jasta: the arraylist version causes more memory fragmentation and needs a lot more memory, right?
May 30 07:30:58 <jasta> alankila: that's fair, but of course the limit is relaxed appropriately by screen density.  i'd like to see it relaxed a bit further, but i can tell you that i don't _require_ it.
May 30 07:31:13 <jasta> hrnt: ...no.
May 30 07:31:32 <alankila> jasta: neither do I. Because there's some unspecified but in general terms fairly large heap available on the GPU side for textures.
May 30 07:31:38 <alankila> :-p
May 30 07:32:00 <hrnt> jasta: the ArrayList<Integer> version would contain an array of pointers which point to Integer objects around the memory, right?
May 30 07:33:03 <alankila> hrnt: the integer objects may be shared. iirc, Integer.valueOf(-128) to Integer.valueOf(127) always returns the same object on java, probably to speed up some autoboxing
May 30 07:33:10 <hrnt> alankila: i know that
May 30 07:33:43 <alankila> but yeah, arraylist<integer> is a sad story on java.
May 30 07:33:52 <g00s> java 10 is unifying the type system and getting rid of primitives all together
May 30 07:34:09 <alankila> interesting ... and scary :-p
May 30 07:34:14 <jasta> hrnt: so, forgive me, but my naive interpretation of NSNumber strongly suggests that this matches the situation with ArrayList<Integer>.
May 30 07:34:16 <hrnt> alankila: on the other hand, iOS NSNumbers are implemented as tagged pointers
May 30 07:34:19 <alankila> nah, just kidding. In honesty, that seems like good idea.
May 30 07:34:19 <g00s> of course, you'll see that in 2017
May 30 07:34:21 <hrnt> http://objectivistc.tumblr.com/post/7872364181/tagged-pointers-and-fast-pathed-cfnumber-integers-in
May 30 07:34:25 <jasta> in that it is an array of pointers, pointing to individual NSNumber instances
May 30 07:34:38 <hrnt> jasta: yes, that would be the naive implementation :)
May 30 07:34:48 <alankila> g00s: so are they going to reify the generics at the same time then?
May 30 07:34:53 <jasta> hrnt: well, keep in mind that if storage is your goal, int[] is what you'd use.
May 30 07:35:05 <g00s> alankila: ha, that cat is out of the bag, i have no idea
May 30 07:35:18 <hrnt> jasta: i intentionally used ArrayList/NSArray, because i'd like to have a mutable container
May 30 07:35:25 <hrnt> (well, NSMutableArray, but you get the point)
May 30 07:35:34 <hrnt> jasta: that is not all - on iOS i could always use std::vector<int>
May 30 07:35:34 <g00s> alankila: oh yeah, java 10 is to have true generics w/o erasure
May 30 07:35:37 <alankila> g00s: ok. was just hoping that there'd be less point to care about int[] vs. arraylist<int> too at the same time
May 30 07:35:49 <alankila> g00s: ah, right, good
May 30 07:36:10 <jasta> hrnt: fair point, however optimizing to this level on java is likely to still yield less code than its obj c equivalent ;)
May 30 07:36:10 <hrnt> jasta: so if i _really_ need to be careful with memory, i can drop down to C or C++.
May 30 07:36:14 <alankila> why only java 10 though?
May 30 07:36:49 <jasta> hrnt: but isn't it amazing how neither of us have found a reason to optimize to this degree coding for mobile phones? :)
May 30 07:36:56 <hrnt> well, actually... :)
May 30 07:37:04 <jasta> i wouldn't believe you if you told me otherwise.
May 30 07:37:58 <jasta> also, it remains true what i said at the start of this discussion: bloated UI hierarchies and bitmaps are the ocean in which your drop of integer array overhead is deposited :)
May 30 07:38:01 <g00s> alankila: java 10 also gets first class functions, i guess
May 30 07:38:18 <alankila> one of the things that might be nice about a new lack of primitives is that you no longer need to allocate 90 % of the bytecode to deal with operators for int, short, char, float, double, ...
May 30 07:38:22 <g00s> its taking them … too damn long
May 30 07:38:36 <hrnt> jasta: i actually have an app in which a substantial amount of memory is taken by integer arrays
May 30 07:38:50 <alankila> if you can rely on runtime generating the optimal sequence anyway
May 30 07:38:50 <hrnt> (and integer->integer hash tables and stuff like that)
May 30 07:39:48 <alankila> still, a bit confused about how the plan for java 10 would look like. Maybe it's nothing particularly adventurous, they just autopromote primitives to boxed objects so you can do 5.toString().
May 30 07:39:49 <hrnt> the memory usage is not actually the problem - fragmentation is
May 30 07:39:56 <jasta> hrnt: substantial being sufficiently large that java autoboxing and fragmentation would cause very real concern for optimization?
May 30 07:40:04 <alankila> hrnt: dalvik doesn't have a compacting collector?
May 30 07:40:08 <hrnt> alankila: no :)
May 30 07:40:12 <alankila> oh. pity
May 30 07:40:14 <hrnt> indeed
May 30 07:40:15 <jasta> i believe such problem domains exist, all of which contained in the frameworks themselves :)
May 30 07:40:37 <alankila> it seems that compacting GC is very difficult as a rule, few runtimes are capable of pulling that off
May 30 07:40:41 <jasta> regardless though, the problem space for java is trivially mitigated.
May 30 07:40:46 <hrnt> i actually had a situation where i used 15MB of heap (of 32MB available, nexus S) and i couldn't allocate a byte buffer on 200kb
May 30 07:41:18 <jasta> int[], SparseIntArray, moving on :)
May 30 07:41:51 <jasta> the design of these sorts of tools aren't even sufficient complex to annoy me :)
May 30 07:41:59 <jasta> sufficiently*
May 30 07:42:16 <hrnt> jasta: see, i've written "smart" arrays that automatically grow their size in C for a long time
May 30 07:42:20 <hrnt> i shouldn't need to do that in java :(
May 30 07:43:15 <jasta> if you pull yourself back a bit from the weeds, you couldn't find any difference at all :)
May 30 07:43:49 <hrnt> alankila: i think i read somewhere that the problem is that in android there is native code that holds references to java objects
May 30 07:43:51 <alankila> there's nothing nicer than not having to worry about memory... so more heap and compacting GC added to dalvik wishlist...
May 30 07:43:57 <hrnt> alankila: which locks the java objects in place
May 30 07:43:59 <g00s> alankila: even mono has had a generational gc ...
May 30 07:44:11 <g00s> dalvik should have it by now
May 30 07:44:26 <alankila> hrnt: eh, maybe. There's certainly some amount of native code but I don't think every little thing has a native counterpart.
May 30 07:44:29 <readme> g00s: i am here!
May 30 07:44:38 <g00s> readme: :O indeed !
May 30 07:45:09 <alankila> many of the platform objects had though... intents did, probably context too
May 30 07:45:13 <hrnt> anyways, the main reason i prefer iOS SDK over android SDK is not even memory management
May 30 07:45:18 <hrnt> it is the activity lifecycle :)
May 30 07:45:46 <jasta> alankila: huh?  intent doesn't have a JNI layer behind it
May 30 07:45:55 <jasta> very little of the framework does, actually.
May 30 07:45:57 <jasta> very, very little.
May 30 07:45:59 <alankila> hm... must have been intentfilter or some other part of the intent delivery mechanics that did then
May 30 07:46:21 <jasta> trust me, i've studied android at this layer a lot :)
May 30 07:46:24 <alankila> sorry, I forgot
May 30 07:47:41 <alankila> I'd support removal of native implementations from android though. Especially if that holds android back from gaining a compacting collector
May 30 07:48:18 <alankila> on the other hand this problem doesn't prevent regular java from having compacting gc, so I'm not sure what ugliness the C++ code potentially does then. Maybe it takes addresses of java objects.
May 30 07:48:31 <alankila> if so, broken.
May 30 07:48:48 <jasta> alankila: most of the reason android has such a huge JNI layer is because its interfacing with third party crap.
May 30 07:48:53 <jasta> that was written in C/C++
May 30 07:49:12 <thepoosh> hello
May 30 07:49:17 <g00s> oh yeah, JNI gets an overhaul in java 9 or 10 also :)
May 30 07:49:25 <jasta> that's stuff like the media layer ( all that packetvideo disaster in the beginning), skia, the binder, ...
May 30 07:49:28 <g00s> 2017 will surely be a year to rejoice !
May 30 07:49:34 <alankila> maybe, but that doesn't have to pin objects in place. Normally the java code is orchestrating the C++ code, so there's the relevant java context passed down
May 30 07:49:57 <jasta> there is some stuff like the aapt/resource system that has a lot of native layer stuff, inexplicably not written in java, but it's rare.
May 30 07:50:18 <jasta> alankila: well, only at a very, very high level.
May 30 07:50:31 <jasta> most of what goes to JNI just has a thin Java layer shell around a whole lot of crud
May 30 07:50:31 <leo2007> Is google to ditch java on android?
May 30 07:50:33 <Drakonite> getting rid of native code support would kill android
May 30 07:50:47 <Drakonite> leo2007: I wish
May 30 07:50:49 <alankila> Drakonite: not arguing that.
May 30 07:51:24 <thepoosh> what Views can register to the GestureDetector.OnDoubleTapListener ?
May 30 07:51:25 <alankila> but removing the portion of native code in interests of improved memory behavior for the java environment itself seems like good idea to me...
May 30 07:51:38 <leo2007> me too. What is the implication of google vs oracle lawsuit?
May 30 07:51:44 <jasta> thepoosh: views dont register to gesture detectors, its the other way around.  views dispatch to them.
May 30 07:52:09 <Drakonite> alankila: what?
May 30 07:52:12 <thepoosh> hmmm, lets say I want to manipulate an image on doubletap
May 30 07:52:16 <alankila> Drakonite: what what.
May 30 07:52:34 <thepoosh> I create a new GestureDetector.OnDoubleTapListener ?
May 30 07:52:38 <jasta> man, it's been a long time since i've wasted my evening arguing on the internet ;)
May 30 07:52:50 * Drakonite is confused as to the topic
May 30 07:53:35 <thepoosh> jasta: ?
May 30 07:53:43 <thepoosh> hmmm, lets say I want to manipulate an image on doubletap
May 30 07:53:44 <alankila> the topic is some pointless internet argument that accomplishes nothing because it's not up to us
May 30 07:53:48 <thepoosh> I create a new GestureDetector.OnDoubleTapListener ?
May 30 07:54:30 <alankila> If we wanted to control future of android, we'd probably work at google or do something so fugging awesome at android level that google would be compelled to copy the feature, perhaps even the implementation itself
May 30 07:56:20 <pragma-> or just file a patent
May 30 07:56:49 <alankila> well, that did not seem to work out so well for oracle at least :-p
May 30 07:56:52 <I440r> ok why does adb push somedir /sdcard/ actually push somedir/* to sdcard/
May 30 07:56:54 <g00s> alankila: or you could just fork android like amazon / chinese carriers
May 30 07:56:58 <g00s> :)
May 30 07:57:51 <alankila> I440r: don't know, hate that as well
May 30 07:58:20 <I440r> scuse me?
May 30 07:58:29 <I440r> so why are you being a dick?
May 30 07:58:57 * alankila raises an eyebrow
May 30 07:59:23 <jasta> yeah, what?
May 30 07:59:27 <jasta> even i'm confused
May 30 07:59:46 <pragma-> he's angry because he uses gentoo
May 30 07:59:52 <g00s> haha
May 30 08:00:03 <g00s> i used to use gentoo, and i'm not angry anymore
May 30 08:00:06 <g00s> :D
May 30 08:00:22 <g00s> i went on to ubunto and got even angrier
May 30 08:00:38 <jasta> alright, im off to bed guys
May 30 08:00:39 <jasta> night
May 30 08:00:44 <g00s> night
May 30 08:00:59 <Ge0rG> what I hate even more about adb push is, if you "adb push somehugefile /sdcard", it pushes the whole damn f*cker over usb, and tells me afterwards that /sdcard is not a valid destination
May 30 08:02:03 <readme> emerge --from-your-parents-basement
May 30 08:03:08 <alankila> Ge0rG: yep, that has happened to me too
May 30 08:03:14 <alankila> adb has these strange rough edges
May 30 08:04:31 <pragma-> android has rough edges, period.
May 30 08:05:20 <pragma-> emulator, adb, framework, etc... all cobbled together by google summer of code interns
May 30 08:05:49 <readme> really?
May 30 08:06:10 <readme> you must be exaggerating, but in the case of the emulator I wouldnt be shocked.
May 30 08:06:13 <pragma-> everything google makes is permanently in beta
May 30 08:06:58 <pragma-> they've just stopped publicly adding "beta" to everything because the board and stockholders get more money if things aren't "beta"
May 30 08:07:09 <Knossos> Hyperbole: Exaggerated statements or claims not meant to be taken literally.
May 30 08:08:33 <pragma-> I'm just ragging on google because I'm annoyed with the sort of crap they are putting out lately.
May 30 08:17:49 <g00s> i get the impression that google is not as mature as MS is, in the sense they have not fully institutionalized certain things across all areas of the company - uh, like testing
May 30 08:19:45 <nism> is there some class similar to android.util.LruCache for 2.3.3?
May 30 08:20:25 <g00s> nism: yeah its in the compat library
May 30 08:21:58 <g00s> whats with all of these 'unknown binding' listings on amazon, which are stupid crazy expensive ?
May 30 08:22:14 <g00s> this book just came  out, there is only one fucking binding
May 30 08:22:17 <drlaban> g00s: Well, users do a lot of testing for them in a sense, I guess. The "It's free"-stamp on things tend to have an effect on stuff like testing.
May 30 08:24:17 <nism> g00s: ty
May 30 08:29:08 <tapas> ñis there a way to have eclipse start the applicaiton project a librarry is referenced in when pressing the run button while working on the library priject?
May 30 08:29:27 <tapas> it's kinda annoying to always have to select the app project while working on changes in the library
May 30 08:29:52 <readme> yes
May 30 08:30:24 <readme> Window->Preferences->Run/Debug->Launching->Luanch the selected resource or active editor, if not launchable: launch the associated project
May 30 08:30:42 <tapas> readme: thanks you're making my day again :D
May 30 08:31:39 <tapas> readme: that's what it's set to....
May 30 08:36:18 <tapas> but if i click run while editing the library a dialog pops up to ask my how to run the project..
May 30 08:40:17 <ech0s7> hi
May 30 08:40:19 <vavirta> Try to select your launch-configuration from the list which popups when clickin that tiny arrow next to run-button
May 30 08:41:51 <ech0s7> hi
May 30 08:42:43 <ech0s7> hi all
May 30 08:43:37 <ech0s7> My Motorola Atrix (Android 2.3.4) allows you to shoot in single shot or panorama. I can launch the android.media.action.IMAGE_CAPTURE intent to capture a photo, but where can I find the extras to pass to intent? To invoke it directly in panorama mode?
May 30 08:46:34 <g00s> ech0s7: well, if it was the stock camera app you could look at the code, but it seems like its the moto photo app … so, i guess you may have to ask on their forums
May 30 08:48:21 <vavirta> ech0s7: intent.putextra()
May 30 08:48:42 <g00s> ech0s7: did you look at the stock camera app? looks like it has panorama already
May 30 08:48:44 <readme> tapas: run configurations->switch manual to automatic
May 30 08:48:50 <ech0s7> vavirta, really ?
May 30 08:49:10 <ech0s7> g00s, ok, i'm looking for stock camera app now
May 30 08:56:23 <ech0s7> g00s, are stock camera sources in official android sources ?
May 30 08:56:31 <g00s> yeah
May 30 08:57:05 <ech0s7> i'm getting android-2.3.4_r1, hoping that panorama mode was introduced into this version
May 30 08:57:36 <JakeWharton> it wasn't
May 30 08:58:03 <g00s> oh, i think i am looking at the ics sources
May 30 08:58:11 <JakeWharton> it was introduced in 4.0.1_r1
May 30 08:58:26 <ixc> motorola has non stock app for panorama
May 30 08:59:04 * Ge0rG hopes HDR will find its way into stock some time...
May 30 08:59:11 <g00s> hehe
May 30 08:59:23 <g00s> i used to be excited about HDR, now not so much
May 30 08:59:43 <Ge0rG> g00s: it is really great if the app is doing all the hard work for you
May 30 09:00:03 <Ge0rG> if MTP was working on my linux box, I'd upload a sample photo :>
May 30 09:00:24 <g00s> my little p&s has hdr, but it sucks because you can't set the white balance
May 30 09:01:01 <Snuffel> i like it for evening pictures of the sky
May 30 09:01:13 <g00s> grad nd filter ftw
May 30 09:01:15 <Snuffel> my HTC One X has it and makes pretty pictures with HDR
May 30 09:01:47 * Ge0rG is using a third-party app
May 30 09:01:54 <Ge0rG> but the UX sucks :(
May 30 09:02:46 <ech0s7> what HDR does it means ?
May 30 09:02:58 <Knossos> High Dynamic Range
May 30 09:03:02 <Knossos> I believ?
May 30 09:03:07 <ech0s7> ok
May 30 09:03:49 <Ge0rG> yeah. great way to make photos that look like painted ;)
May 30 09:03:50 <Snuffel> https://www.dropbox.com/s/g5gabcpfwujgm9n/2012-04-29%2020.31.43.jpg
May 30 09:03:51 <Tsukaza> Can someone tell me how can I make an interface with a line graph in android? I Mean, temperature. We have the last 5 temperature measures of a person in a small graph, and the average(a grey line for example)
May 30 09:03:55 <Tsukaza> Something like this: http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR
May 30 09:03:57 <Snuffel> this is HDR with my HTC
May 30 09:04:14 <Snuffel> i like it
May 30 09:04:22 <g00s> Snuffel: yeah, that came out ok
May 30 09:04:32 <Ge0rG> Snuffel: how large is the delay between individual shots on the X?
May 30 09:04:34 <Knossos> Thats nice
May 30 09:04:39 <readme> Tsukaza: there are libraries, but you can also use a webview
May 30 09:04:51 <readme> I've done it with XSLT + XML -> html -> webview
May 30 09:04:59 <g00s> yeah, doing hdr right means tripod really
May 30 09:05:05 <Snuffel> Ge0rG: too big for action shots
May 30 09:05:05 <readme> obviously css required
May 30 09:05:29 <Tsukaza> readme: Can you name some libraries for research? Hum... I need the graph to be connected to my database, it get the last 5 values and starts drawing it
May 30 09:05:32 <Snuffel> hm that same out wrong :)
May 30 09:05:34 <Snuffel> came
May 30 09:05:39 <Tsukaza> readme: Ill search more into webvies then
May 30 09:05:41 <Knossos> Couldn't you make a View yourself, for a graph?
May 30 09:05:47 <g00s> Snuffel: can you post the medium exposure ?
May 30 09:05:51 <Knossos> Draw directly on the Canvas
May 30 09:05:57 <Tsukaza> Hum...
May 30 09:06:00 <readme> sure
May 30 09:06:02 <Ge0rG> g00s: AFAICT the HDR apps are shifting the individual shots to match and compensate for movement
May 30 09:06:08 <Tsukaza> Is it "hard" to draw in Canvas?
May 30 09:06:13 <readme> no
May 30 09:06:23 <Tsukaza> I want the lightest version of it :P
May 30 09:06:25 <Knossos> I think it is as simple as draw(x, y)
May 30 09:06:27 <Tsukaza> less battery consuming
May 30 09:06:28 <Snuffel> I don't need a tripid or anything for stills
May 30 09:06:36 <Snuffel> tripod
May 30 09:06:52 <readme> you should be using a library, though
May 30 09:06:56 <readme> there's no need to reinvent the wheel
May 30 09:06:57 <Tsukaza> ok
May 30 09:07:01 <Knossos> Haven't really bothered, myself. But I am quite sure that is the simplest route.
May 30 09:07:10 <Knossos> Yeah, there is probably a library for it
May 30 09:07:16 <readme> I *have* done it
May 30 09:07:21 <readme> and I"ve done it with XSLT and XML, too
May 30 09:07:24 <g00s> Snuffel: do you live out there ?
May 30 09:07:24 <readme> I recommend the latter
May 30 09:07:28 <Tsukaza> Which libraries did you used?
May 30 09:07:49 <readme> I didn't use libs. I used the libraries that come with android/java to process XML with XSLT
May 30 09:07:53 <readme> I found some XSLT on the net
May 30 09:07:54 <Tsukaza> oh ok
May 30 09:07:56 <readme> that makes charts
May 30 09:08:03 <readme> theres tons of it
May 30 09:08:06 <Tsukaza> ok ill look into it
May 30 09:08:22 <Snuffel> g00s: i can ride my bike to it
May 30 09:08:25 <readme> http://fgeorges.blogspot.com/2008/04/simple-svg-chart-generation-with-xslt.html
May 30 09:08:29 <readme> here's an svg example
May 30 09:08:31 <g00s> Snuffel: oh nice
May 30 09:09:14 <readme> http://franklinefrancis.github.com/SvgCharts4Xsl/
May 30 09:09:38 <Knossos> readme = hero :)
May 30 09:09:45 <readme> all you've got to do is make some XML to represent your data, process it against that xslt
May 30 09:10:09 <Knossos> I'm saving those URLs for later.
May 30 09:10:12 <readme> the huge benefit of this, is when you want your charts to work the same on multiple platforms.
May 30 09:10:16 <readme> or on a server
May 30 09:10:17 <readme> etc.
May 30 09:10:25 <Tsukaza> ok
May 30 09:10:27 <Tsukaza> readme, ty
May 30 09:11:02 <readme> svg seems not to work on android  webkit
May 30 09:11:09 <readme> but there is android-svg-library
May 30 09:11:13 <readme> avoid svg
May 30 09:11:48 <readme> you could jkust use jquery plugins
May 30 09:11:51 <readme> just*
May 30 09:12:10 <readme> http://www.1stwebdesigner.com/css/top-jquery-chart-libraries-interactive-charts/ --- much better selection than native chart libs
May 30 09:13:39 <Tsukaza> ok, im looking into it
May 30 09:14:57 <TobiasFar> hi..i'm getting this error trying to build goldfish for x86 emulator "arch/x86/boot/header.S:227: Error: can't resolve `input_data_end' {*UND* section} - `input_data' {*UND* section}"
May 30 09:15:01 <TobiasFar> any ideas?
May 30 09:25:35 <g00s> readme: Tsukaza i saw this the other day http://humblesoftware.com/flotr2/
May 30 09:25:56 <Tsukaza> Oh, nice :) ty
May 30 09:42:51 <readme> g00s: those are pretty sick
May 30 09:42:58 <readme> I'm working on a library project now to make using these easier from java
May 30 09:43:41 <g00s> :)
May 30 09:52:20 <readme> basically got it working. refactoring it into a fragment now.
May 30 09:52:23 <readme> oh wait, eclipse froze.
May 30 09:52:26 * readme rage
May 30 10:04:49 <readme> grr assets don't work in library projects
May 30 10:04:52 <readme> i forgot
May 30 10:05:04 <readme> now I have to write complicated code
May 30 10:05:12 <alex_PP> really?
May 30 10:05:14 <alex_PP> wow
May 30 10:05:15 <readme> really
May 30 10:05:30 <readme> http://stackoverflow.com/questions/6346889/how-to-reference-an-asset-in-a-library-project
May 30 10:18:00 <ddreamer0> i have defined an AsyncTask within my activity, how can i call it from another class ( my custom gallery class which is in another file)
May 30 10:18:23 <hrnt> usually that is not something you want to do
May 30 10:18:49 <hrnt> instead you would put the asynctask into a service
May 30 10:18:55 <ddreamer0> hrnt ok, i also tried defining my asyncTask in a separate file
May 30 10:19:11 <ddreamer0> hrnt a service ? dont know that
May 30 10:20:56 <ddreamer0> hrnt when i use my asynctask in another file, the setProgressBarIndeterminateVisibility(true); is broken
May 30 10:21:07 <ddreamer0> i tried with TockitActivity.setProgressBarIndeterminateVisibility(true); but no luck
May 30 10:21:17 <hrnt> yes
May 30 10:21:33 <hrnt> put a method in your asynctask which can be used to associate the asynctask with the activity
May 30 10:22:09 <hrnt> setActivity(Activity a) { this.act = a; } and then use act.setProgressBarWhatever...
May 30 10:23:59 <readme> anyone know what kinds of data are allowed to be retuned by a method in a javascript interface? from java -> javascript
May 30 10:24:04 <readme> returned*
May 30 10:24:19 <readme> so far float[] and float[][] seem to not work
May 30 10:24:21 <readme> but int does.
May 30 10:25:11 <ddreamer0> hrnt aha beautiful :)
May 30 10:25:45 <ddreamer0> hrnt i was trying to access my globarlvars with MyActivity.globalVar and no luck
May 30 10:26:00 <hrnt> global vars?
May 30 10:26:12 <g00s> uh, google earth is such a buggy pile (on mac)
May 30 10:26:28 <g00s> every time i use it, it crashes
May 30 10:39:35 <mdim> how to provide custom Android library (android.jar) when building a project?
May 30 10:41:21 <ddreamer0> hrnt i have a global list, which i use with my adapter to fill a list
May 30 10:41:37 <ddreamer0> and i wanted to fill it asynchronously from a webservice
May 30 10:56:31 <mdim> anyone? ^^
May 30 11:05:41 <thepoosh> is there a way to detect if the phone is sitting upside down?
May 30 11:05:42 <thepoosh> http://androidheadlines.com/wp-content/uploads/2011/02/Samsung-Galaxy-Tab-White_Rogers-back.jpg
May 30 11:05:46 <thepoosh> like this
May 30 11:06:22 <Zider> that's not upside down, is it?
May 30 11:07:01 <thepoosh> well, sort of
May 30 11:07:05 <thepoosh> backwards
May 30 11:07:19 <thepoosh> I want to know when the phone is positioned on it's back
May 30 11:07:27 <thepoosh> (on it's belly)
May 30 11:08:36 <thepoosh> I don;t know even how to describe this
May 30 11:08:38 <thepoosh> :S
May 30 11:08:50 <Knossos> I would describe it as "when the display is facing down"
May 30 11:09:19 <Knossos> Cant help you with the problem though. I'd guess something to do with the gyroscope?
May 30 11:13:41 <Jagdos> Isn't it possible with the proximeter sensor?
May 30 11:14:25 <Knossos> Ahh, yes
May 30 11:14:51 <ddreamer0> hrnt ok i have my activity setter
May 30 11:15:15 <ddreamer0> how do i call it? i tried getCurrentActivity
May 30 11:15:20 <Jagdos> I think it's used when you are talking, so the screen is disabled when you are pressing it against the ear, so I guess most phones can use it..
May 30 11:15:24 <ddreamer0> but that gives a Context reference
May 30 11:15:54 <thepoosh> Knossos: I found the answer
May 30 11:15:54 <thepoosh> http://android-er.blogspot.com/2011/02/detect-phone-flipping.html
May 30 11:16:20 <Knossos> There you go, accelerometer :P
May 30 11:24:25 <mdim> I managed to build custom Android Library (android.jar), but I can't compile an Android project against it because resources are missing; it only contains source code. How do I bundle it with resources too?
May 30 11:25:15 <SimonVT> You don't
May 30 11:25:19 <SimonVT> It's not supported yet
May 30 11:28:04 <mdim> SimonVT: really? How come Google built the original android.jar then?
May 30 11:38:01 <ddreamer0> guys how do i pass my current activity to another object ?
May 30 11:38:27 <ddreamer0> i want to use activity.setProgressBarIndeterminateVisibility(true); in another class
May 30 11:38:37 <ddreamer0> but i don't know how to pass it from my main class
May 30 11:39:01 <appel1> ddreamer0: pass it like any other reference to a object
May 30 11:39:11 <mikedg> ddreamer0: it's basic oo
May 30 11:40:18 <ddreamer0> how do i rever to my activity ? something isn't working here...
May 30 11:40:38 <ddreamer0> my package is called com.dev.tockit
May 30 11:40:49 <ddreamer0> and my Activity: TockitActivity
May 30 11:41:15 <mikedg> have you ever programmed in an object oriented language before
May 30 11:41:18 <ddreamer0> i tried passing it TockitActivity and com.dev.tockit.TockitActivity
May 30 11:41:30 <ddreamer0> i also passed this obviously
May 30 11:42:20 <appel1> ddreamer0: in what context did you try using this ?
May 30 11:42:49 <ddreamer0> problem is, i'm calling this from a GalleryClass
May 30 11:42:58 <appel1> ddreamer0: perhaps you should get a beginners book on how to program in java. once you know the basics things will be much easier
May 30 11:43:12 <mikedg> well then your galleryclass needs to know about your activity too
May 30 11:43:42 <ddreamer0> my gallery has a context reference to the main activity
May 30 11:45:22 <ddreamer0> and i was trying to cast (Activity) context but obvoiusly it's not the way to do it
May 30 11:45:59 <readme> it is amazing how much a simple walk will enhance your thinking
May 30 11:46:41 <appel1> ddreamer0: why not change so that instead of passing the activity as a Context to the gallery, you pass it as Activity if that is what you need
May 30 11:46:45 <readme> I was beating my head against the wall thinking..what type will I return from my javascript interface, into my javascript code from java... it's limited to the basic primitives, objects, but no access to fields, and string. What's the answer. A string of JSON -- herp derp.
May 30 11:49:00 <ddreamer0> appel1 yeah my class receives an activity, but i don't know how to reference my mainactivity to pass it, because i'm trying to pass it from a GalleryClass' onScroll method
May 30 11:51:15 <appel1> ddreamer0: you want to get a reference to an activity that isn't the currently shown activity?
May 30 11:51:47 <ddreamer0> yup..
May 30 11:51:49 <GeeksOnHugs> mornin...coffees on
May 30 11:52:11 <ddreamer0> i have a reference to the context of that activity
May 30 11:52:12 <appel1> ddreamer0: sounds like a really bad design that probably wont even work
May 30 11:52:20 <ddreamer0> can i get a reference to the activity from that ?
May 30 11:52:35 <appel1> ddreamer0: an activity is a context
May 30 11:52:47 <ddreamer0> activity == context ?
May 30 11:52:57 <GeeksOnHugs> are you guys in europe?  its a lotta peeps for 10 to 8 in amrika
May 30 11:53:04 <appel1> ddreamer0: Activity inherits from Context
May 30 11:53:20 <ddreamer0> y tried casting a context to activity but it's nullpointing me
May 30 11:53:59 <appel1> ddreamer0: in any case you probably don't want to touch an activity that isn't the current foreground activity
May 30 11:54:32 <appel1> GeeksOnHugs: yup
May 30 11:54:52 <GeeksOnHugs> ah  kk heh
May 30 11:55:31 <appel1> ddreamer0: if the Context is null then it will still be null after you cast it to Activity.
May 30 11:55:49 <appel1> ddreamer0: what are you trying to do?
May 30 12:07:27 <ocx> hi guys
May 30 12:07:38 <ocx> i have some question regarding inflation layouts
May 30 12:07:59 <ocx> View myWelcome = View.inflate(this, R.layout.welcome, null); will myWelcome point then to the whole ViewGroup or the first view returned from the xml layout?
May 30 12:08:05 <ocx> i am not able to test that
May 30 12:08:37 <hrnt> ocx: what is the root element in your layout?
May 30 12:09:04 <ocx> <RelativeLayout
May 30 12:09:08 <hrnt> ok
May 30 12:09:11 <hrnt> so it returns a RelativeLayout
May 30 12:09:43 <hrnt> which is, of course, a viewgroup
May 30 12:10:03 <ocx> which inherits View
May 30 12:10:16 <hrnt> yes
May 30 12:10:21 <ocx> great thanks
May 30 12:11:54 <mikedg_> ugh
May 30 12:14:17 <mikedg_> my server laptop is refusing my connections
May 30 12:14:19 <mikedg_> wtf :/
May 30 12:14:36 <mikedg_> lamesauce
May 30 12:28:54 <Alcardil> what is the easiest way to create and manage a database for an application? basically i'm trying to create something that keeps track of a student roster and their grades.
May 30 12:29:06 <Alcardil> basically i'm trying to learn the ropes
May 30 12:29:28 <Zharf> basically you say basically too much ;)
May 30 12:29:39 <Alcardil> so if anyone has any good sites to point me to I'd appreciate it
May 30 12:30:26 <Zharf> http://developer.android.com/guide/topics/data/data-storage.html#db
May 30 12:32:35 <pragma-> Alcardil: this is a good site: http://www.explosm.net/comics/
May 30 12:33:38 <Zharf> pragma-, <3
May 30 12:33:55 <Alcardil> thanks Zharf, that was pretty helpful
May 30 12:37:05 <Alcardil> btw, thanks for making me feel like an idiot for referring me to the dev page...lol
May 30 12:37:39 <Alcardil> it has been a long day
May 30 12:37:40 <wongk> it's what we do best ;)
May 30 12:37:55 <pragma-> http://www.rawstory.com/rs/2012/05/29/cyber-attack-concerns-raised-over-boeing-787-chips-back-door/
May 30 12:38:17 <pragma-> attack boeing 787 and military computer chips over the internet!!11
May 30 12:40:35 <hkhalid> pragma-, I think you watch too much tv heh
May 30 12:43:44 <ech0s7> where can i find a readme about how are organized android sources ?
May 30 12:44:34 <wongk> s.android.com
May 30 12:44:55 <readme> well, most of the framework stuff is in 'frameworks/base'
May 30 12:45:03 <readme> the native stuff is in external/
May 30 12:45:19 <readme> stock apps are in 'packages/apps'
May 30 12:45:44 <readme> hkhalid: i've got that book on hold at barnes and noble
May 30 12:46:06 <readme> downloaded the audiobook too but the nararrator's voice is weird and hard to follow
May 30 12:46:31 <wongk> what book are we reading
May 30 12:46:54 <readme> Illuminatus Trilogy
May 30 12:46:58 <wongk> ah
May 30 12:48:03 <pragma-> is that the book about the knights trading their metal armor for business suits and ties to hide and protect the holy grail in the vault on wall street?
May 30 12:49:26 <fmauro> hi all! It appears that some of my users with older android devices have issues with the image-fetching in my app. It force closes (they say) but I haven't seen any reports on my dev console. Does anyone of you have a slow or older phone and the time to see if you can locate the problem?
May 30 12:49:56 <appel1> fmauro: out of memory perhaps?
May 30 12:50:13 <wongk> not able to repro in the emulator, i suppose?
May 30 12:50:47 <fmauro> appel1: yes my first guess too, but proper exception handling is in place and I've provoked an exception by using a huge image and it handles it well.
May 30 12:51:01 <PowerInside> In S Pen SDK samples, when I open a project inside samples (to eclipse), I get errors at all @Override points in the source code.. I can't run the program because of these errors. Why isn't it working out of the box?
May 30 12:51:29 <wongk> fmauro: are you fetching the image over http?
May 30 12:51:31 <fmauro> wongk: no, it works fine with all the android versions I've tried.
May 30 12:51:34 <appel1> fmauro: if it is out of memory it could cause an exception anywhere in your code. when it needs to create a new string or whatever.
May 30 12:51:48 <fmauro> wongk: yes
May 30 12:52:00 <wongk> fmauro: which classes are you using for that?
May 30 12:52:57 <woozly> Guys, I have no animation between Activities
May 30 12:52:57 <fmauro> wongk: URL, URLConnection, then over a bufferedinputstream I fetch the bytes
May 30 12:52:59 <woozly> what I do wrong?
May 30 12:53:07 <wongk> fmauro: http://android-developers.blogspot.com/2011/09/androids-http-clients.html
May 30 12:53:50 <pragma-> woozly: please hold while i polish my crystal to provide me with the best clarity
May 30 12:54:27 <pragma-> woozly: ah yes, i'm seeing... something about a boolean.
May 30 12:55:08 <fmauro> wongk: I guess I got the Http question wrong then, I'm using the URLConnection, do you suppose that is the problem?
May 30 12:55:41 <wongk> fmauro: it seems like a likely candidate, according to this blog post, yes?
May 30 12:57:04 <woozly> Main.java class: http://pastebin.com/cPECMTLa; subActivity.java class: http://pastebin.com/pmGShEGG; main.xml: http://pastebin.com/0yJZMxEp; subactiv.xml: http://pastebin.com/ExdJTUw1; manifest: http://pastebin.com/9fL9Fv39
May 30 12:57:26 <wongk> heh
May 30 12:57:50 <wongk> woozly: create a test case - 1 file, that reproduces the problem
May 30 12:58:03 <wongk> unless someone here is nice enough tp review your 5 files
May 30 12:58:30 <woozly> wongk: what you mean 'create a test case - 1 file' ?
May 30 12:58:35 <woozly> put all into one file?
May 30 12:59:21 <wongk> i mean the least amount of code possible that reproduces the issue
May 30 12:59:27 <woozly> by default, android has animation when switching between activities?
May 30 12:59:34 <PowerInside> Anyone here played around with the S Pen SDK?
May 30 12:59:48 <drlaban> woozly: It depends on if animations are enabled on the device as well
May 30 12:59:56 <woozly> o_O
May 30 13:00:14 <fmauro> wongk: I don't think it's the thing I'm looking for sadly, I am merely fetching files, for which URLConnection is meant. Let me share some code: http://pastebin.com/XriJrn8T
May 30 13:02:12 <wongk> fmauro: what makes you think the bug mentioned in the blog is not the problem?
May 30 13:02:56 <fmauro> Because I'm using neither the regular not the apache HttpURLConnection. but the run-of-the-mill java.net.URLConnection
May 30 13:03:30 <wongk> what subclass do you think is used for the http protocol?
May 30 13:04:47 <fmauro> wongk: URLConnection, for both apache and regular, which brings me to the point where it would affect me either way if using the upper classes, or am I wrong?
May 30 13:05:04 <wongk> that answer is incorrect
May 30 13:05:13 <wongk> URLConnection is abstract, for one
May 30 13:05:20 <fmauro> wongk: please explain
May 30 13:05:31 <wongk> http://developer.android.com/reference/java/net/URLConnection.html
May 30 13:06:29 <fmauro> wongk: ah so you mean that I actually AM already using the Http sub-class?
May 30 13:06:35 <wongk> yes, you are
May 30 13:06:55 <fmauro> wongk: good, got it, and I should use the apache one instead as written in the post?
May 30 13:07:28 <wongk> i would start with the workaround mentioned for HttpURLConnection and send a test build to the users experiencing the issue
May 30 13:08:13 <fmauro> wongk: got it, thanks for the help, I'll try this.
May 30 13:13:09 <lapdis> Is there a way to change the z-order of an animation so it plays out behind a view?
May 30 13:13:44 <ixc> u animate a view
May 30 13:13:55 <ixc> put the view that u animate behind ur top view :)
May 30 13:14:11 <lapdis> yes, but it seems to affect that view to get a higher z-order then it had previously
May 30 13:15:24 <hkhalid> wongk, hello
May 30 13:15:34 <wongk> hola
May 30 13:15:44 <ixc> u mean u add a new view programatically and its over the others?
May 30 13:16:05 <ixc> im not sure but i think u can call on old view: bringToFront()
May 30 13:16:13 <lapdis> no, I have two views side by side and then have the desired z-order. But when I start to animate the view that is behind, the animation plays out in front of the other view
May 30 13:16:35 <lapdis> bringToFront() does not seem to place it ahead of the animation :/
May 30 13:18:48 <gaz`> is there a way to make a chunk of xml, then in a loop load up the view from the xml and add it to a layout say 5 times?
May 30 13:19:24 <ixc> gaz`: inflate?
May 30 13:19:58 <gaz`> only used it in listviews, if that works itll be cool
May 30 13:20:26 <gaz`> i think ur right, thanks
May 30 13:24:41 <osxorgate> anyone know an app that will let me shoot a pic, add some text and share them both at the same time? I'm trying to get it to work but only gmail seems to handle the intent properlyt
May 30 13:26:30 <wongk> osxorgate: something like that sounds like it would be worth 1 BILLION DOLLARS
May 30 13:26:57 <lapdis> lol :)
May 30 13:27:19 <osxorgate> wut
May 30 13:28:13 <lapdis> I might be wrong, but I think he is referensing a speech Picasso held to the spanish parlament in 1966.
May 30 13:31:19 <p_l> actually, some movie... not sure if it wasn't Austin Powers. And later on, a certain photo-sharing startup whose revenue plan was to get bought for 1 billion :>
May 30 13:32:30 <osxorgate> right.. i get it now. what i mean is that apps like say tweetdeck don't seem to properly handle my intent where i ask them to share pic+text for me
May 30 13:33:04 <Ge0rG> osxorgate: maybe you need something like.... https://play.google.com/store/apps/details?id=com.android.lolcat
May 30 13:34:02 <osxorgate> maybe i should go back to coding at night
May 30 13:34:11 <mdim> how do I download Android resources's "sources"?
May 30 13:35:59 <rawroland> Hello
May 30 13:36:14 <rawroland> Is there a way of nesting lists?
May 30 13:37:28 <rawroland> I know listviews cannot be nested, I however have to implement something similar to a nested list.
May 30 13:38:18 <osxorgate> rawroland maybe expandablelistview?
May 30 13:38:27 <osxorgate> that goes 1 sublevel
May 30 13:39:17 <rawroland> osxorgate: But the user has to click on an item to expand,
May 30 13:39:37 <osxorgate> not necessarily
May 30 13:39:46 <osxorgate> you can auto expand
May 30 13:39:51 <rawroland> And I also have two children
May 30 13:40:19 <wongk> 2 children, both one level deep?
May 30 13:41:02 <rawroland> wongk: yes
May 30 13:41:07 <Ge0rG> auto-expanding database-based expandablelists is a little nightmare.
May 30 13:41:26 <wongk> ExpandableListView sounds right up your alley
May 30 13:41:31 <Ge0rG> one day, I'm going to have two children as well...
May 30 13:42:05 <rawroland> Ge0rG: maybe I will have 3 ...
May 30 13:42:23 <rawroland> wongk: I was thinking of trying table layout
May 30 13:47:54 <lapdis> can I reference a local variable such as this, when the anonymous class shadows it? http://pastebin.com/DMzKyE0X
May 30 13:53:27 <wongk> lapdis: uhm, isn't animation == animation in that case?
May 30 13:54:04 <lapdis> yes it is, it is more of a curiosity really :P
May 30 13:54:10 <wongk> oh
May 30 13:54:21 <wongk> i honestly don't know, but it seems like a very bad idea
May 30 13:54:38 <lapdis> yeah :P
May 30 13:57:27 <theNatd> hi folks any good tutorial how to implement in app purchase?
May 30 13:58:52 <Zharf> http://developer.android.com/guide/market/billing/index.html
May 30 13:58:53 <Zharf> ?
May 30 14:02:19 <canadiancow|work> adb takes a shit when i plug in three devices
May 30 14:02:57 <wongk> on windows, it used to throw a fit with a device and an emulator
May 30 14:10:58 <woozly> If I create new AlertDialog.Builder(); what I need to put into constructor?
May 30 14:11:03 <woozly> Context, but what is it?
May 30 14:11:37 <woozly> MyActivity->onCreate->button1.setOnClickListener()->onClick()-> new AlertDialog.Builder(???);
May 30 14:12:13 <evancharlton> woozly: look at the docs?
May 30 14:12:25 <woozly> evancharlton: can't understand.. :/
May 30 14:13:29 <woozly> getApplicationContext() ??
May 30 14:15:51 <canadiancow|work> actually a dialog will likely need an activity's context
May 30 14:15:59 <Jagdos> Try with AlertDialog.Builder(this) maybe
May 30 14:16:01 <canadiancow|work> an application will give you strnage window exceptions
May 30 14:16:04 <readme> it does
May 30 14:16:17 <canadiancow|work> Jagdos, how can you suggest that when you dont even know what "this" is
May 30 14:16:17 <readme> if you try to make a dialog from a service, you need to fake it by using an activity with a dialog theme
May 30 14:16:26 <readme> or by using a shim activity
May 30 14:16:54 <wongk> we know what "this" is, and we know that is not the right answer
May 30 14:16:58 <woozly> (this) not working, because this - View.onClickListener
May 30 14:17:01 <readme> canadiancow|work: the same way a parrot can talk
May 30 14:18:10 <wongk> either way he needs to understand what a Context is, and not just find something that can be passed successfully
May 30 14:18:26 <evancharlton> ==wongk
May 30 14:19:07 <evancharlton> I'm not a big fan of the "let's just put random shit into a .java file until it compiles" programming style
May 30 14:19:16 <Snuffel> hehe
May 30 14:19:23 <readme> BDD - blogspam driven development
May 30 14:19:28 <wongk> hah
May 30 14:19:40 <evancharlton> readme++
May 30 14:21:55 <kakazza> I love the short description of BDD on wikipedia.
May 30 14:22:10 <brian_de> is there a reason why a ldpi drawable doesn't get displayed on a ldpi device? The drawable is present in the ldpi folder and also in other resolutions in the *dpi folders respectively.
May 30 14:22:14 <kakazza> Even if there's a good intent behind it, it just sounds so buzzwordish.
May 30 14:22:34 <readme> kakazza: that's the real bdd
May 30 14:22:36 <evancharlton> brian_de: it sounds like you have other issues
May 30 14:22:39 <kakazza> "BDD is a second-generation, outside–in, pull-based, multiple-stakeholder, multiple-scale, high-automation, agile methodology. It describes a cycle of interactions with well-defined outputs, resulting in the delivery of working, tested software that matters."
May 30 14:22:49 <evancharlton> brian_de: I doubt very much that ldpi has anything to do with it
May 30 14:22:54 <readme> kakazza: a buzzword for common sense
May 30 14:22:57 <kakazza> :D
May 30 14:23:04 <kakazza> Synergize faster!
May 30 14:23:05 <readme> any sane developer working with business focused clients would operate that way
May 30 14:23:35 <brian_de> evancharlton: can you give me one more hint what the issue might be?
May 30 14:23:46 <evancharlton> brian_de: i have no idea; you haven't given us any information
May 30 14:24:39 <canadiancow|work> +1
May 30 14:26:33 <brian_de> evancharlton: sorry, my fault ;) ... well: the drawable is working on all other resolutions prefectly. only on ldpi it is invisible -- really invisible: drawable is bound to an ImageButton whicht receives click events (so it is not related to layout issues)
May 30 14:27:51 <evancharlton> brian_de: take a look at it with hierarchyviewer
May 30 14:28:29 <brian_de> ... we found out that if we remove the very drawable from hdpi it works on ldpi .. o.O
May 30 14:28:29 <readme> evancharlton: is it OK to publically log this channel? If not, is it OK to publish statistics about the channel logs?
May 30 14:29:51 <evancharlton> readme: I would prefer you don't, but I obviously can't stop it.
May 30 14:30:04 <evancharlton> there used to be a bot in here that would; I don't know whatever happened to it
May 30 14:30:15 <readme> evancharlton: what about stats that don't give away identity and such
May 30 14:30:21 <readme> like frequency of actual android discussion etc
May 30 14:30:29 <readme> number of times activity has been discussed in a day
May 30 14:30:31 <readme> stuff like that
May 30 14:30:40 <evancharlton> readme: sounds alright to me
May 30 14:30:47 <readme> ok cool
May 30 14:30:53 <evancharlton> I mean, this is a public channel
May 30 14:31:06 <readme> freenode policy prohibits unauthorized public logging
May 30 14:31:07 <evancharlton> if someone doesn't want to be logged, then they can /wc and get on with their day
May 30 14:31:12 <evancharlton> ah
May 30 14:31:34 <readme> but like you say, there's little you can do abou tit
May 30 14:31:46 <readme> because who knows which guy is logging !? :)
May 30 14:31:52 <evancharlton> I mean, I could play whack-a-mole with banning you, but yeah, not a whole lot
May 30 14:31:58 <readme> it's a good rule when some troll starts pasting your logs, though.
May 30 14:32:03 <evancharlton> yeah
May 30 14:32:20 <readme> i wouldn't publically log without permission, because it's not right
May 30 14:32:23 <readme> I do log privately.
May 30 14:32:49 <Snuffel> readme: do you log our queries? oh my.....
May 30 14:33:10 <evancharlton> readme: yeah, I keep logs too
May 30 14:33:21 <readme> I only started logging recently when I realized it's better to keep logs so I can scrape them for the valuable info later
May 30 14:33:23 <evancharlton> mainly for scrollback and reference
May 30 14:33:28 <evancharlton> yep
May 30 14:34:00 <evancharlton> I should set up a rolling log, though. I don't much care what was talked about in here in e.g. October of 2008
May 30 14:34:48 <readme> yeah I need to write a log rotater
May 30 14:35:59 <evancharlton> readme: pipe it through logrotate
May 30 14:37:52 <readme> yeah
May 30 14:37:55 <readme> I should do that
May 30 14:38:07 <readme> I'll wait until my space starts to run out though
May 30 14:38:22 <readme> i'm on a prgmr vps with like a gig of storage, so that shouldn't be too long
May 30 14:41:38 <evancharlton> uh-oh, I only have 51G left on this VM :(
May 30 14:46:07 <readme> this one is dedicated to irc
May 30 14:46:14 <readme> my linode has a reasonable amount of storage
May 30 14:46:49 <readme> recently upgraded to dropbox pro plan
May 30 14:47:47 <Snuffel> i got 25GB just for registering my HTC One X
May 30 14:48:37 <mikedg_> so i can ping my laptop that i have vnc on, rdp to a vm, and znc on
May 30 14:48:43 <mikedg_> but i cant connect to any of those services
May 30 14:48:44 <mikedg_> wtf
May 30 14:49:06 <mikedg_> it's been working fine for weeks and just lost connections to it this morning
May 30 14:49:27 <evancharlton> mikedg_: router blocking ports?
May 30 14:49:44 <mikedg_> i have another computer on the same network i can RDP to
May 30 14:49:52 <mikedg_> and that's how im pinging the other computer
May 30 14:50:02 <mikedg_> but i cant connect to any of the services even from that computer
May 30 14:50:22 <readme> ssh tunnel to the rescue
May 30 14:51:51 <mikedg_> it's weird cause hey look mikedg is my fucking other computer
May 30 14:52:10 <readme> oh, same network
May 30 15:10:12 <woozly> my AlertDialog displays not on vertical center of screen
May 30 15:10:30 <woozly> it margins + 10-20dps frop top
May 30 15:10:41 <woozly> app not fullscreened
May 30 15:10:49 <woozly> how to fit it at center of screen?
May 30 15:11:08 <readme> you shouldn't
May 30 15:11:14 <readme> that's where the user always see's his alert dialogs
May 30 15:11:19 <readme> you want yours to be like all the other ones
May 30 15:11:23 <readme> principle of least surprise
May 30 15:11:29 <readme> seses*
May 30 15:11:31 <readme> sees*
May 30 15:11:34 <readme> herp derp
May 30 15:13:26 <readme> If you're dead-set on it though. Use the dialog theme on an activity and hack it up a bit .
May 30 15:13:37 <evancharlton> readme: that won't even work
May 30 15:13:46 <readme> why not?
May 30 15:13:52 <evancharlton> you cannot control the position of the window
May 30 15:14:04 <readme> use a transparent window and control the position of a view inside of it
May 30 15:14:20 <john_doe_jr> to add libs to your project...you create a folder named lib and then 'add external jars...' right?
May 30 15:14:51 <wongk> "libs"
May 30 15:15:07 <evancharlton> readme: yes, but that's not a dialog theme :)
May 30 15:15:07 <wongk> that sounds like a lot of work to make an app less user friendly
May 30 15:15:17 <readme> hmm
May 30 15:15:26 <evancharlton> wongk: you'd be amazed at how much work android developers will put in to make their app more stupid
May 30 15:15:45 <evancharlton> see also: every oem "enhancement" ever
May 30 15:15:55 <wongk> i probably wouldn't, but the point stands =]
May 30 15:16:04 <canadiancow|work> evancharlton, "every" is an exaggeration
May 30 15:16:07 <kakazza> On that note, will the Design Guidelines ever be released as a PDF?
May 30 15:16:14 <readme> evancharlton: actually I have a custom dialog right here that is full-screen
May 30 15:16:23 <readme> so that one could be used with a transparent background, for sure
May 30 15:16:41 <oGMo> evancharlton: exemplified by the extra work done to remove the "update all" button in Play when only one app needs updating, thus requiring numerous extra clicks
May 30 15:16:56 <canadiancow|work> oh oGMo thats a good point
May 30 15:17:01 <wongk> it is
May 30 15:17:07 <canadiancow|work> ive noticed that too
May 30 15:17:11 <canadiancow|work> we should submit that as a bug report
May 30 15:17:23 <wongk> vote it up
May 30 15:18:04 <kakazza> Link?
May 30 15:18:32 <canadiancow|work> link?
May 30 15:18:45 <readme> evancharlton: but yeah, I just looked and I'm not using the dialog theme :)
May 30 15:18:51 <oGMo> i haven't bothered to check .. it's clearly intentional since to do it requires an intentional conditional for specifically that case
May 30 15:18:54 <readme> i'm using a custom theme
May 30 15:25:08 <woozly> guys, what to use, for storage app's settings?
May 30 15:25:20 <woozly> SharedPreferences works only for session?
May 30 15:26:56 <evancharlton> ... what?
May 30 15:27:00 <evancharlton> where did you get that impression?
May 30 15:27:30 <evancharlton> right in the docs: This data will persist across user sessions (even if your application is killed).
May 30 15:27:44 <evancharlton> woozly: please consult the documentation from time to time
May 30 15:27:45 <woozly> "This data will persist across user sessions"
May 30 15:27:48 <readme> maybe it works on a per-session basis if you don't call commit
May 30 15:28:00 <readme> "persist ACROSS"
May 30 15:28:06 <readme> I am guessing english is not your first language
May 30 15:28:08 <woozly> sorry, bad english
May 30 15:28:17 <readme> so I don't blame you for being confused
May 30 15:28:17 <woozly> thank you guys.
May 30 15:28:22 <readme> look up prepositions
May 30 15:28:34 <woozly> readme: :) thanks
May 30 15:28:43 <woozly> well, I need to learn english..
May 30 15:28:45 <TobiasFar> is it possible to enable hardware accelleration HAX for self build intel x86 builds of the kernel?
May 30 15:29:26 <danieldeusing> anybody knows how to parse json reponses ?
May 30 15:29:46 <TobiasFar> danieldeusing, there are libraries for that...i.e. gson
May 30 15:35:04 <evancharlton> danieldeusing: Android ships with the json.org library
May 30 15:36:04 <Knossos> Yes, I use that library. JSONObject, JSONArray, etc.
May 30 15:36:13 <Knossos> I've never needed anything else.
May 30 15:36:39 <wongk> the built in JSON is fine if you don't need stream support
May 30 15:42:14 <lapdis> after an animation finishes I change some stuff in my layout. But there is still one frame between the animations finish and the new layout, where it snaps back to the old layout. Is there a way to prevent this? Maybe stop the layout from refreshing temporarily?
May 30 15:43:01 <sleepster> what are some common ways to develop a unique ID for each android phone?
May 30 15:43:07 <sleepster> I need to distingush my clients by an ID
May 30 15:43:13 <sleepster> that needs to be gauranteed unique
May 30 15:44:12 <jrajav> sleepster: You could use the serial number of the device http://stackoverflow.com/questions/2322234/how-to-find-serial-number-of-android-device
May 30 15:44:24 <sleepster> thanks
May 30 15:44:38 <canadiancow|work> no
May 30 15:44:38 <canadiancow|work> sotp
May 30 15:44:41 <canadiancow|work> dont do that
May 30 15:44:46 <wongk> sleepster: Uuid
May 30 15:44:55 <canadiancow|work> http://android-developers.blogspot.ca/2011/03/identifying-app-installations.html
May 30 15:45:29 <jrajav> My bad :)
May 30 15:47:30 <evancharlton> phew, canadiancow|work to the rescue
May 30 15:49:08 <hkhalid> I have a broadacast receiver class for listning to SMS' , and when it gets a SMS it send outs a SMS, but the problem is I dont know how to get the value of an edittext in another Activity
May 30 15:49:25 <hkhalid> as I would want to use that text in the SMS I send out
May 30 15:50:12 <osxorgate> I'm wondering what would happen if i call BitmapFactory.decodeFile() and during that, the sd card becomes unavailable. No mention of exceptions
May 30 15:50:17 <wongk> this all sounds very "big brother"
May 30 15:50:39 <evancharlton> hkhalid: you can't
May 30 15:51:02 <evancharlton> osxorgate: you'll probably get a null bitmap
May 30 15:51:15 <hkhalid> evancharlton, what do you suggest I do
May 30 15:51:17 <osxorgate> that would suit me juist fine
May 30 15:51:41 <john_doe_jr> how do you create a .jar file?
May 30 15:51:43 <evancharlton> hkhalid: I don't know. I don't have the slightest clue what you're actually trying to do.
May 30 15:52:19 <evancharlton> john_doe_jr: http://lmgtfy.com/?q=eclipse+create+jar
May 30 15:52:48 <hkhalid> evancharlton, im listening for incoming SMS', when it gets one, the app im making, auto responds by sending out a text "im busy". The message is hardcoded, I want to use a message that it can pull from an Activity I have that has a small form where you can set your own message.
May 30 15:53:11 <hkhalid> im wondering how I can get that value
May 30 15:53:14 <evancharlton> write their message to sharedpreferences, then read from sharedprefs to get the message to send
May 30 15:53:39 <hkhalid> thanks evancharlton !
May 30 15:54:04 <hkhalid> wongk, your last message was directed at who
May 30 15:54:16 <wongk> hkhalid: you
May 30 15:55:02 <hkhalid> wongk,  strange...I dont get your comment, but I explained it to my friend, my question that is, and he started talking about sec issues....
May 30 15:55:15 <hkhalid> he thought the activity is seperate from the app
May 30 15:55:19 <wongk> we were making the same point then :P
May 30 15:55:38 <wongk> the point being, you're "reading" my text messages
May 30 15:55:49 <evancharlton> wongk: if you installed his app, you gave him permission to
May 30 15:55:54 <wongk> there's a level of trust required there
May 30 15:55:59 <wongk> evancharlton: sure
May 30 15:56:03 <hkhalid> its a sms auto responder
May 30 15:56:04 <wongk> the point being: i wouldn't
May 30 15:57:11 <wongk> hkhalid: just so you're aware that some users would be uncomfortable installing your app, even though the app does nothing "malicious"
May 30 15:58:14 <hkhalid> wongk, im still confused about the security issue, the app just has permissions for checking for incoming calls and sms
May 30 16:00:34 <wongk> hkhalid: i have faced the same issue, i have an app that allows users to log in with their Google checkout merchant ID and key, which makes some users very uncomfortable
May 30 16:01:27 <hkhalid> wongk, still confused...are you saying the app for example, can read incoming calls and sms, and then phone home somewhere with that data or what
May 30 16:01:34 <wongk> yes
May 30 16:01:48 <wongk> if you r app can read their texts, it can do anything it wants with them
May 30 16:01:58 <sonicrules1234> hkhalid: That's what the users are worried about
May 30 16:02:15 <hkhalid> sonicrules1234, but doesnt that require other permissions....like INTERNET at least
May 30 16:02:30 <hkhalid> Im confused on how it seems I dont have enough permissions to make it malicious
May 30 16:02:36 <hkhalid> such as INTERNET
May 30 16:02:39 <dragorn> also if it can send SMS it can send them to premium services and run up the bill
May 30 16:02:43 <sonicrules1234> You should probably mention that in your description
May 30 16:02:50 <dragorn> side note, you don't need internet to leak data on < 3.x
May 30 16:02:57 <hkhalid> thanks guys
May 30 16:02:58 <dragorn> tho most users won't know that
May 30 16:03:16 <evancharlton> anyone installing an sms auto-responder should expect that it can read/send sms...
May 30 16:03:23 <wongk> im not sure what is required to send an emailvia smtp
May 30 16:03:24 <dragorn> hkhalid: you fall into that category where you're not doing anything wrong, but because the permissions aren't that fine grained, you get caught in the net of people being nervous
May 30 16:03:34 <hkhalid> ahhh gotcha
May 30 16:04:00 <hkhalid> cool beans, as this is just a test project for a job interview, make an app where you can set a message, and it will reply to phone calls and sms messages
May 30 16:04:14 <wongk> then nothing to worry about :P
May 30 16:04:27 <hkhalid> they said it usually takes 35 hours for such a project, not sure if its the usual time, or just for the devs from that country
May 30 16:04:34 <kakazza> dragorn: Why only <3.0?
May 30 16:04:46 <hkhalid> trying to move overseas to do Android work
May 30 16:04:49 <dragorn> kakazza: because they fixed it
May 30 16:04:49 <wongk> 35 hours for a job interview... wow
May 30 16:05:46 * wongk lunch
May 30 16:05:46 <Daniel-B> hello @all
May 30 16:06:25 <evancharlton> hkhalid: wait, are you getting help on how to pass a job interview...?
May 30 16:06:32 <evancharlton> something tells me you're missing the point.
May 30 16:07:00 <jrajav> Yeah, everyone knows you don't get access to help after starting a job
May 30 16:07:08 <hkhalid> evancharlton, yea im not going to apply anymore due to this reason and im going to give up on it
May 30 16:08:11 <kakazza> dragorn: Wasn't that the encrypted/unencrpyted thingy?
May 30 16:08:20 <Daniel-B> Well, I got this code http://blogs.msdn.com/b/zhengpei/archive/2009/10/13/managing-apn-data-in-google-android.aspx
May 30 16:08:33 <Daniel-B> when trying to run it Im getting force closes
May 30 16:12:43 <Daniel-B> bankai_?
May 30 16:18:20 <evancharlton> Daniel-B: check logcat.
May 30 16:18:50 <Daniel-B> I will compile it again and msg you
May 30 16:18:55 <Daniel-B> cuse im new at this k?
May 30 16:19:00 <Daniel-B> ty
May 30 16:20:31 <Xonk> When I call setImageDrawable on an ImageButton should that take effect immediately or is there something else I need to call to force a redraw?
May 30 16:20:34 <evancharlton> Daniel-B: don't /msg me. Paste it to pastebin.com and send us the link.
May 30 16:20:41 <evancharlton> and by "us" I mean "the channel"
May 30 16:20:47 <Daniel-B> ok
May 30 16:20:49 <Daniel-B> fine
May 30 16:20:59 <Daniel-B> Im not gonna send prv msg to all the chan lol
May 30 16:21:00 <evancharlton> Xonk: should take effect right away, but you could try calling .invalidate() on the view (button)
May 30 16:23:23 <Xonk> Yeah, I tried invalidate() and refreshDrawableState() so far and it's not changing.  It's within a ListView.  Perhaps the ListView is redrawing itself and reventing the button back immediately.
May 30 16:24:04 <evancharlton> Xonk: hm, that doesn't sound likely
May 30 16:24:19 <evancharlton> Xonk: can you paste us the relevant code? (use pastebin.com or similar, please)
May 30 16:24:26 <Xonk> yeah, one min
May 30 16:28:26 <Xonk> Here's the code for my imagebutton problem: http://pastebin.com/5iHKfvaG
May 30 16:28:56 <Xonk> I've walked through it in Debug and the voteDown() method is executing.  The image just doesn't change on the screen.
May 30 16:35:39 <osxorgate> xonk why refreshdrawablestate?
May 30 16:35:54 <osxorgate> also im sure you can pass a resourceID instead of getResources().get etc
May 30 16:37:26 <Daniel-B> logcat: http://pastebin.com/3DQGYnVE
May 30 16:37:41 <Xonk> osxorgate - I was just trying to force the refresh.  It doesn't work without it either.
May 30 16:37:55 <Xonk> I also tried passing the resource id.  The method wants a Drawable instead of an int.
May 30 16:38:58 <evancharlton> yeah, ImageButton has a weak API, IMO
May 30 16:40:07 <canadiancow|work> how so?
May 30 16:40:16 <wongk> 05-28 00:24:30.713: E/AndroidRuntime(1370): Caused by: java.lang.NullPointerException
May 30 16:40:17 <wongk> 05-28 00:24:30.713: E/AndroidRuntime(1370):     at com.daniel.hotmobile.Main.EnumerateAPNs(Main.java:71)
May 30 16:40:20 <wongk> Daniel-B: ^^
May 30 16:40:23 <evancharlton> canadiancow|work: well it doesn't take a resource, right?
May 30 16:40:27 <canadiancow|work> ?
May 30 16:40:42 <canadiancow|work> it does
May 30 16:40:44 <evancharlton> oh wait, what
May 30 16:40:45 <evancharlton> yeah
May 30 16:40:47 <evancharlton> sorry
May 30 16:40:53 <canadiancow|work> o.O
May 30 16:40:59 <evancharlton> Daniel-B: use setImageResource or whatever
May 30 16:41:08 <Daniel-B> can u fix it for me
May 30 16:41:11 <evancharlton> canadiancow|work: I dunno, I always just style an ImageView instead
May 30 16:41:14 <wongk> no
May 30 16:41:14 <Daniel-B> ?
May 30 16:41:15 <evancharlton> Daniel-B: what, no
May 30 16:41:36 <wongk> and on that note, no more help for you
May 30 16:41:50 <Daniel-B> why?
May 30 16:43:01 <alankila> Daniel-B: we aren't your parents, to put it bluntly. Take care of your own problems.
May 30 16:43:15 <Daniel-B> WTF
May 30 16:43:29 <alankila> besides, there's also always the collective benefit that if you fail then that means less competition for the rest of us.
May 30 16:43:37 <Daniel-B> you should help ppl
May 30 16:43:44 <evancharlton> Daniel-B: I did.
May 30 16:43:48 <Daniel-B> k
May 30 16:43:49 <evancharlton> I'm not going to fix your code
May 30 16:43:52 <evancharlton> I'll give you advice
May 30 16:43:53 <canadiancow|work> do you need us to wipe your ass too?
May 30 16:44:02 <canadiancow|work> like... evancharlton gave you the solution
May 30 16:44:04 <canadiancow|work> but that wasnt enough
May 30 16:44:05 <Daniel-B> omg
May 30 16:44:07 <Daniel-B> such ppl
May 30 16:44:09 <canadiancow|work> you now want him to actually go into your code
May 30 16:44:10 <canadiancow|work> and fix it
May 30 16:44:14 <Daniel-B> never seen
May 30 16:44:14 <jrajav> Daniel-B: There is a null pointer problem at line 71 in Main.java
May 30 16:44:16 <canadiancow|work> i'll do it for you
May 30 16:44:18 <canadiancow|work> i charge $100/hour
May 30 16:44:23 <canadiancow|work> and i charge in increments of 30 minutes
May 30 16:44:24 <Daniel-B> sure dude
May 30 16:45:31 <jrajav> Just look at that line of your code and try to figure it out
May 30 16:45:44 <jrajav> If you can't figure it out, I can suggest some excellent beginner's Java books
May 30 16:46:01 <Xonk> I tried changing it to setImageResource which makes for nicer code, but still getting the same result unfortunately.
May 30 16:46:55 <p_l> canadiancow|work: yay, more people using that method :D
May 30 16:47:06 <canadiancow|work> ?
May 30 16:48:38 <GeeksOnHugs> lol...i just went out to search for an honest appraisel of the differences between android and iOS on the internet....i thought i could find one...but no...some would start off sounding reasonable before going off track but i failed to find one in the end
May 30 16:49:23 <canadiancow|work> fwiw, i think anyone who says devving for ios is easier is on crack
May 30 16:49:33 <canadiancow|work> two people working on a project? deal with that merge hell in the project files
May 30 16:49:37 <canadiancow|work> pbxproject or whatever its called
May 30 16:49:41 <p_l> canadiancow|work: at one point I was responding to people who wanted way too involved solutions with "I can do that for you if you pay"
May 30 16:49:49 <canadiancow|work> oh
May 30 16:50:04 <p_l> also, isn't it possible to *not* use Xcode project files?
May 30 16:50:37 <p_l> personally I'm more annoyed by the UI of Xcode and how you define interfaces
May 30 16:50:45 <p_l> the runtime environment is easier
May 30 16:57:19 <dlaroche> p_l: i still don't like objective-c either....too cryptic
May 30 16:57:52 <joakime> dlaroche: not too bad, once you learn it.  good video lessons on the language and tools available too.
May 30 16:58:24 <dlaroche> I hear the SDK is more straight forward to use then Android's. Any truth behind that one?
May 30 16:59:01 <joakime> no. saying "Straight Forward to Use" is a personal preference.
May 30 16:59:19 <dlaroche> lol...that's kinda what I figured
May 30 17:00:23 <p_l> dlaroche: C++ is cryptic. Objective-C just has annoying level of support outside Apple
May 30 17:00:29 <wongk> never merged xcode project files, but have you ever tried to merge VS project files?
May 30 17:00:44 <wongk> it arbitrarily re-orders the files in the XML
May 30 17:00:46 <p_l> dlaroche: iOS runtime is simpler. Much simpler
May 30 17:01:28 <dlaroche> p_l: yeah I played with it, but never got into learning objective-c. Seemed like it was going to take more time to learn then what I had to spare for it.
May 30 17:02:29 <mitchnull> nah, obj-c is pretty simple if you know C already
May 30 17:02:30 <p_l> as I once figured with some people, iOS developement is plain old writing NeXTSTEP applications, so basically good old nonthreaded GUI app. Android is like developing an EJB or DCOM application where you don't control the main loop at all :)
May 30 17:02:56 <dlaroche> does anyone have any alternative forums worth checking out. Seems like anddev.org is under attack most of the time.
May 30 17:03:24 <wongk> for what purpose?
May 30 17:03:31 <dlaroche> programming questions
May 30 17:03:39 <dlaroche> for Android dev.
May 30 17:03:39 <mitchnull> dlaroche: stackoverflow.com
May 30 17:03:40 <jrajav> stackoverflow
May 30 17:03:51 <wongk> +1
May 30 17:03:54 <Ge0rG> +2
May 30 17:04:00 <dlaroche> cool
May 30 17:04:18 <wongk> i actually just use google and end up at SO, all my questions have been answered :P
May 30 17:04:54 <dlaroche> Yeah. Stackoverflow just doesn't meet my requirements as a forum. But great answers on it though
May 30 17:08:02 <theNatd> ppl in docs http://developer.android.com/guide/market/billing/billing_integrate.html#billing-download some there i can download sample app
May 30 17:08:30 <theNatd> but i can't find link(( some one can help me
May 30 17:10:06 <jrajav> theNatd: There is no link, you download it with the SDK manager
May 30 17:10:43 <jrajav> theNatd: It's explained, with a screenshot, right there in the article
May 30 17:12:49 <theNatd> tnx, you can tell me in there folder it?
May 30 17:13:22 <jrajav> theNatd: Wherever you installed the Android SDK
May 30 17:14:12 <luxurymode> anyway to attach an html file from raw as the body of a share intent for sending an email?
May 30 17:15:18 <theNatd> find thnx alot ypu
May 30 17:15:22 <theNatd> you
May 30 17:26:08 <bara> is there an alternative to WebView with better performace for dynamic content?
May 30 17:36:16 <Daniel-B> I imported this pack "com.android.providers.telephony"
May 30 17:36:22 <Daniel-B> how can i use his functions?
May 30 17:38:19 <canadiancow|work> seriously?
May 30 17:38:29 <t0mless> yo dawg
May 30 17:38:33 <t0mless> I heard you like functions
May 30 17:38:45 <canadiancow|work> http://lmgtfy.com/?q=java+for+noobs
May 30 17:39:06 <Daniel-B> fuck you snob
May 30 17:39:09 <Daniel-B> no ban me
May 30 17:39:12 <Daniel-B> now*
May 30 17:39:14 <Daniel-B> fags
May 30 17:39:21 <canadiancow|work> hey evancharlton, he asked...
May 30 17:39:54 <jrajav> canadiancow|work: Holy crap that site is amazing
May 30 17:40:07 <aboudreault> well.... you have to admit your question is special.
May 30 17:43:13 <Daniel-B> ban...
May 30 17:43:31 <t0mless> lulz
May 30 17:43:44 <canadiancow|work> here we go
May 30 17:43:53 <canadiancow|work> lol
May 30 17:43:58 <canadiancow|work> i didnt realize you had +o'd me
May 30 17:45:14 <wongk> lolz howtoprogramz teh droids?
May 30 17:46:09 <canadiancow|work> not even
May 30 17:46:11 <canadiancow|work> HOW DO I USE A CLASS
May 30 17:46:30 <t0mless> na dawg
May 30 17:46:34 <t0mless> it was a PACK
May 30 17:46:40 <wongk> you're assuming he knew what a class was
May 30 17:46:46 <t0mless> how to use it's functions
May 30 17:46:48 <wongk> i think you're being too generous
May 30 17:47:10 <canadiancow|work> hey evancharlton, i think he's back
May 30 17:47:17 <canadiancow|work> Harley-D's name is too similar
May 30 17:47:28 <wongk> i thought the same thing ;)
May 30 17:47:45 <wongk> his host is completely different :P
May 30 17:47:57 <jayd16> Hey guys, I have a list view set to multiselect.  How do I make it so I can change drawables when an item is selected.  it looks like the default text color changes in a list item text view when the list item is selected, so I assume it changes the state of child views.  Is there some statelist drawable magic I can do here?
May 30 17:48:27 <canadiancow|work> wongk, check the host on me and canadiancow
May 30 17:48:29 <canadiancow|work> they're different
May 30 17:48:31 <canadiancow|work> but they're obth me :P
May 30 17:48:42 <wongk> i'm just sayin
May 30 17:49:08 <luxurymode> i have a broadcastreceiver set up to listen for android.media.AudioManager.ACTION_AUDIO_BECOMING_NOISY and have declared as much in my manifest, but im never receiving any broadcasts. anything else i need to do?
May 30 17:49:14 <evancharlton> canadiancow|work: I don't care if he's back, unless he starts trolling again
May 30 17:49:26 <mikedg> luxurymode: it probably doesnt work in manifest
May 30 17:49:27 <evancharlton> luxurymode: get noisy audio?
May 30 17:49:36 <mikedg> canadiancow|work: check the host on wongk and me, they are different but they're both me :P
May 30 17:49:48 <wongk> snap!
May 30 17:50:01 <luxurymode> mikedg, why not?
May 30 17:50:08 <luxurymode> evancharlton, yeah when you unplug headphones
May 30 17:50:10 <mikedg> cause that would be retarded
May 30 17:50:36 <mikedg> and i mean retarded in the most offensive way
May 30 17:50:53 <luxurymode> mikedg, actually it might because it should be android.media.ACTION_AUDIO_BECOMING_NOISY
May 30 17:51:08 <mikedg> it might be
May 30 17:51:39 <mikedg> i doubt it
May 30 17:51:44 <mikedg> and ill bet you 5 slaps
May 30 17:52:15 <luxurymode> lol
May 30 17:52:19 <luxurymode> why would it be retarded
May 30 17:52:36 <luxurymode> thats how the developer docs said to do it
May 30 17:52:49 <luxurymode> they had a sample under resources or somewhere about playing music with a servie, etc
May 30 17:53:02 <luxurymode> and they said the class works because its declared in the manifest with those filets
May 30 17:53:03 <luxurymode> filters
May 30 17:53:26 <mikedg> it would be stupid because why would you need to do that since you should be running already
May 30 17:53:40 <mikedg> and if you arent why bother triggering
May 30 17:54:48 <luxurymode> huh? if you're playing music and the headphones get disconnected you want to stop the music or whatever… i think im missing what you're getting at.
May 30 17:56:29 <luxurymode> mikedg, i should have to register it somewhere?
May 30 17:56:34 <mikedg> your app is already running so why dont you just register it programattically
May 30 17:56:52 <mikedg> i cant think of a use case where it would be beneficial to broadcast that to an app that is not running
May 30 17:56:55 <wongk> he's saying, the way you're doing it, your app is going to be started every time that event is sent
May 30 17:57:07 <wongk> s/event/intent
May 30 17:57:14 <mikedg> so it seems stupid from a design point to allow it to be registered in the manifest
May 30 17:58:10 <luxurymode> right i hear what you're saying. good point
May 30 17:58:37 <luxurymode> makes sense when you are defining a service that responds to whatever filters you've defined within your app, but not in this case
May 30 17:59:15 <luxurymode> mikedg, so i should just do  registerReceiver(new MusicIntentReceiver(), new IntentFilter("android.media.AudioManager.ACTION_AUDIO_BECOMING_NOISY")); ?
May 30 18:19:45 <Escherial> hey, does anyone here happen to know if the framework for handling external accounts on android is somehow related to oauth?
May 30 18:20:13 <Escherial> e.g. if i were to add oauth authentication to my web service, would that make it easier to use the built-in account system in android to connect to it?
May 30 18:26:21 <f00bar80> Anybody knows why youtube embeded videos cannot be played using the webview ? ? based on this code , >> http://stackoverflow.com/questions/5817805/how-can-we-play-youtube-embeded-code-in-android-application-using-webview
May 30 18:26:41 <[fields]> I was writing apps to my phone yesterday, but today when starting a new project I go Run > As Android Application and my device is always offline.  I found adb in terminal and ran ./adb kill-server  and then ./adb usb  still yields error: device offline    How do I remedy this?  My phone is connected by USB, but the running device is seen as Offline
May 30 18:27:19 <Knossos> Try plugging into a different USB port [fields]
May 30 18:27:28 <[fields]> Knossos:  good suggestion
May 30 18:27:53 <Knossos> I had an issue myself where I plugged into my front ports. Worked just fine for months. Then all of a sudden, it stopped working with those ports on the front and only works with the rear ports.
May 30 18:28:16 <Knossos> f00bar80: What exactly is the error you get? If any?
May 30 18:29:03 <f00bar80> Knossos, no errors , when i click on play the video start loading and then black screen
May 30 18:29:05 <wongk> Escherial: no, you have to implement your own authenticator, which use whatever protocol you want
May 30 18:29:19 <[fields]> Knossos:  unplug > plug into new port > ./adb kill-server > ./adb usb > * daemon not running.  starting it now on port 5037 * > * daemon started successfully * > error: device offline
May 30 18:29:33 <Knossos> I don't know then
May 30 18:29:42 <Knossos> Worth a try though :)
May 30 18:29:52 <[fields]> this is so weird.  I was making apps yesterday, was so much fun.
May 30 18:30:03 <Knossos> You could try reinstalling your usb drivers
May 30 18:30:09 <Knossos> (for your device)
May 30 18:30:31 <Escherial> wongk: would having the web service support oauth make any part of that easier, out of curiosity?
May 30 18:30:33 <wongk> f00bar80: i've seen black videos in the android borwser as well
May 30 18:30:45 <wongk> Escherial: as compared to what?
May 30 18:30:57 <wongk> i don't know oauth
May 30 18:31:31 <mwillbanks> I'm attempting to implement Actionbar tabs (using the FragmentTabs.java) sample; however, the tab content (Fragment) itself may contain a titles and details fragment as well; does this mean I should implement a Fragment within a Fragment?  Generally speaking I've used the TabHost for this previously
May 30 18:31:43 <wongk> f00bar80: i assumed it was a flash and/or codec issue
May 30 18:32:41 <Knossos> My understanding is that you have a fragmentactivity parent, and fragment children that become the tabs
May 30 18:32:59 <Knossos> As its simplest form
May 30 18:33:13 <wongk> fragments in fragments is not supported
May 30 18:33:37 <Knossos> That said, I don't use TabHost. I use ViewPager and ViewPagerIndicator from Jake
May 30 18:33:59 <mwillbanks> hrm; the tabs themselves cannot be fragment activities can they?  from the examples I've seen it looks like they can only be fragments
May 30 18:34:08 <Knossos> They would be fragments
May 30 18:34:15 <f00bar80> wongk, what do you mean
May 30 18:34:35 <wongk> f00bar80: which part of what i said was confusing?
May 30 18:35:27 <Knossos> There are an awfully large number of bugs in the Flash-Android video playback system
May 30 18:35:32 <MDijkstra> fragment activities? what's that?
May 30 18:36:16 <MDijkstra> you really shouldn't write new code using TabHost imo
May 30 18:36:31 <MDijkstra> oh, nevermind
May 30 18:36:44 <f00bar80> wongk, i don't what's the problem
May 30 18:37:01 <Knossos> Some are the fault of Android devs, some are the fault of Flash devs. (my currently big project is heavily invested in this area)
May 30 18:37:17 <wongk> i don't either, i was just pointing out that the android browser has the same problem
May 30 18:37:36 <Knossos> It is entirely plausible it is simply a bug as opposed to a codec issue
May 30 18:37:59 <MDijkstra> Knossos: flash player has been a disaster from the beginning hasn't it?
May 30 18:38:03 <Knossos> :D
May 30 18:38:22 <Knossos> I wish I didn't have to use it. But one of its newer technologies is the backbone of our system
May 30 18:38:40 <wongk>  aflash backbone
May 30 18:38:43 <wongk> how unfrotunate
May 30 18:39:33 <mwillbanks> MDijkstra: i'm attempting to avoid it actually; all i am attempting to do is utilize Fragments for tabs but yet have the fragment basically do what the sample FragmentLayout sample does
May 30 18:40:22 <Knossos> https://github.com/JakeWharton/Android-ViewPagerIndicator
May 30 18:40:25 <Knossos> I recommend that
May 30 18:40:34 <JakeWharton> me too
May 30 18:40:38 <Knossos> https://raw.github.com/JakeWharton/Android-ViewPagerIndicator/master/sample/screens.png
May 30 18:40:46 <JakeWharton> http://viewpagerindicator.com
May 30 18:40:55 <Knossos> There you go
May 30 18:41:00 <Knossos> Right from the horses mouth
May 30 18:41:07 <[fields]> Knossos:  found my solution.  Surprisingly unchecking the USB Debug mode checkbox in the phone and checking it again was the remedy.
May 30 18:41:14 <Knossos> Ha!
May 30 18:41:21 <[fields]> :)
May 30 18:41:43 <wongk> i heard that JakeWharton guy was biased!
May 30 18:41:52 <mwillbanks> JakeWharton: man; always making me happy...
May 30 18:42:01 <Knossos> Seriously though JakeWharton. Do you have a donate a beer button somewhere or something?
May 30 18:42:02 <mikedg> JakeWharton: can you make a android library to make Diablo 3 backwards compatable back to  Android 1.6
May 30 18:42:28 <mikedg> Knossos: yeah paypal him $5 or something to jakew@mikedg.com
May 30 18:42:32 <f00bar80> wongk, is there no way to use opera as the browser ?
May 30 18:42:33 <Knossos> :D
May 30 18:42:52 <wongk> f00bar80: i'm not sure how that question is relevant to anything
May 30 18:42:59 <JakeWharton> Knossos: there's http://abs.io/donate but I'd rather you write an awesome app that I will use as a donation
May 30 18:43:39 <Knossos> JakeWharton: http://goo.gl/Tdep6
May 30 18:44:06 <Knossos> Uses both ABS and VPI
May 30 18:44:11 <evancharlton> JakeWharton is a saint
May 30 18:44:42 <mikedg> JakeWharton: wtf is flatter
May 30 18:46:11 <doogan> how about SeriesGuide
May 30 18:46:14 <doogan> uses both too
May 30 18:47:21 <mikedg> what about Bird Bar
May 30 18:47:26 <mikedg> I don't remember if I use your crap
May 30 18:48:15 <JakeWharton> mikedg: you do
May 30 18:48:27 <JakeWharton> I unrar'd your APK
May 30 18:50:41 <mikedg> im sueing you
May 30 18:50:51 <mikedg> no reverberating engineering
May 30 18:51:12 <mikedg> i will settle for 5/8's of your soul
May 30 18:56:57 <canadiancow|work> Toast created but not shown: did you forget to call show() ?
May 30 18:56:59 <canadiancow|work> best. lint check. ever.
May 30 18:57:28 <orospakr> Why does onDestroyView() and the corresponding loop back to onCreateView() exist in the fragment lifecycle?  Activity instances are reused for what is otherwise an entirely new "instance" of a Fragment?
May 30 18:57:29 <wongk> hah
May 30 18:57:56 <orospakr> s/Activity instances/Fragment instances/
May 30 18:58:54 <mdim> is it ok to delete from the Android Library android.jar android/os/Binder.class and replace it with the android/os/Binder.java, where Binder.java is pre-compiled version of Binder.class?
May 30 18:59:11 <mikedg> ?
May 30 18:59:20 <mikedg> why do you want to do that
May 30 18:59:34 <mdim> mikedg: I'm trying to build custom version of Android Library
May 30 18:59:49 <mikedg> why dont you just try it?
May 30 19:00:02 <mdim> so, if I take default android.jar and replace it with my own implementation of Binder.java, should it work?
May 30 19:00:10 <mdim> mikedg: who says I haven't tried it?
May 30 19:00:29 <mikedg> if you tried it, then why are you asking if it works?
May 30 19:00:30 <mikedg> you should know
May 30 19:00:45 <mdim> mikedg: I know it doesn't work, so I'm wondering what am I doing wrong
May 30 19:01:02 <mikedg> what doesnt work
May 30 19:01:23 <mdim> mikedg: I try to compile an Android project against this custom android.jar, and it fails
May 30 19:02:36 <mdim> mikedg: it says it can't find Binder symbol in: public static abstract class Stub extends android.os.Binder implements net.dimjasevic.jpf.hellobinderipc.service.IHelloService
May 30 19:02:45 <evancharlton> canadiancow|work: Toast should have a warning on it, always
May 30 19:02:59 <evancharlton> "Toast created; are you sure you didn't mean to use a sensible UI?"
May 30 19:03:17 <mikedg> Toast should throw ToastNotShownException
May 30 19:03:31 <mikedg> and then you are supposed to catch it and then do ex.getToast().show();
May 30 19:03:33 <evancharlton> ToastSucksException
May 30 19:05:42 <john_doe_jr> What is the standard format that is used for an android to communicate w/ a server....SOAP?
May 30 19:06:00 <wongk> such a standard doesn't exist
May 30 19:06:13 <john_doe_jr> wongk-home: REST?
May 30 19:06:15 <wongk> if it did, SOAP would be the last format used
May 30 19:06:34 <john_doe_jr> wongk-home: well what would you recommend then?
May 30 19:06:41 <wongk> what meets your use case
May 30 19:07:18 <john_doe_jr> wongk-home: emergency system that alerts everyone in a surrounding area
May 30 19:08:12 <wongk> that doesn't really tell me anything, you need to consider the technical requirements
May 30 19:08:18 <mikedg> why would you communicate with a server?
May 30 19:08:25 <mikedg> just make a really loud sound from everyones phones
May 30 19:09:20 <john_doe_jr> wongk-home: well, I don't understand.....what are some technical requirements that you are looking for?
May 30 19:10:17 <mikedg> john_doe_jr: i recommend HTML over FTP
May 30 19:10:24 <luxurymode> im using a textview as the view returned by getView in my spinner adapter and im setting the gravity to LEFT, but i'd also like some padding on the left of the textview, but it seems to be ignored when i use Gravity.LEFT. any way to get some padding there?
May 30 19:10:30 <mikedg> FHTMLP
May 30 19:10:35 <wongk> mikedg: lol
May 30 19:10:39 <john_doe_jr> mikedg: looking for some security man
May 30 19:11:05 <mikedg> sftp then, SFHTMLP
May 30 19:11:11 <wongk> john_doe_jr: do you not know the technical requirements of your application?
May 30 19:11:57 <luxurymode> evancharlton, lol. yea toasts are super ugly and weird
May 30 19:12:58 <john_doe_jr> wongk-home: I guess I don't...sorry...very important to get the messages to our users from an android device or iphone device
May 30 19:13:30 <wongk> john_doe_jr: how can you be expected to design a system if you aren't given requirements? :P
May 30 19:14:11 <wongk> should the communication be: as small and efficient as possible, or: open, extensible, and able to be consumed by a number of devices and technologies
May 30 19:14:37 <john_doe_jr> wongk-home: large number of devices and technologies
May 30 19:15:02 <wongk> REST w/ json or xml over HTTP spunds great, then
May 30 19:15:10 <john_doe_jr> starting w/ android and iphone...so I was thinking XML (SOAP)
May 30 19:15:26 <wongk> protip: SOAP is never the answer
May 30 19:15:29 <t0mless> ewwww SOAP
May 30 19:15:38 <wongk> *especially* in the mobile world
May 30 19:15:41 <t0mless> Despite it's name... SOAP is dirrrrrttttyyyy
May 30 19:16:41 <mikedg> protobuffs
May 30 19:16:43 <mikedg> json
May 30 19:16:45 <mikedg> buzzword
May 30 19:16:53 <JakeWharton> fragmentation
May 30 19:17:00 <mikedg> +1
May 30 19:17:10 <canadiancow|work> nexus accessories always take too long to hit the market
May 30 19:17:17 <JakeWharton> protobuffs are badass, though
May 30 19:17:20 <canadiancow|work> they finally got the real car docks out for galaxy nexus
May 30 19:17:27 <canadiancow|work> i think the nexus one car dock took just as long
May 30 19:17:30 <JakeWharton> gzip'd json is a great alternative as well
May 30 19:18:26 <wongk> take it to the metal: straight binary streams over TCP
May 30 19:18:54 <g00s> use quantum entanglement !
May 30 19:19:06 <wongk> sounds kinky
May 30 19:20:06 <f00bar80> how to make the dfault browser for webview to be always Opera ?
May 30 19:20:08 <luxurymode> any way to get some padding on text in a textview that isnt inside some other viewgroup?
May 30 19:20:25 <luxurymode> JakeWharton, how can i set custom views for the overflow dropdown?
May 30 19:20:31 <p_l> g00s: ASN.1 :D
May 30 19:20:42 <p_l> f00bar80: not possible, afaik
May 30 19:21:16 <wongk> i fear my metion of the android browser has confused him
May 30 19:27:40 <g00s> when buying a google nexus device, how does one determine whether the carrier or google will be updating it ?
May 30 19:27:50 <g00s> i'm reading VZ is about to update their GN
May 30 19:28:01 <wongk> verizon is the oddball
May 30 19:28:08 <canadiancow|work> not really
May 30 19:28:10 <g00s> if i purchased a devive, i want it updated from google.
May 30 19:28:19 <canadiancow|work> you can flash it to yakju/takju
May 30 19:28:22 <canadiancow|work> and then you'll get updates from google
May 30 19:28:23 <wongk> i just read an article about that yesterday
May 30 19:28:29 <canadiancow|work> but the ones in canada are all like yakjuux
May 30 19:28:32 <canadiancow|work> and get updates from samsung
May 30 19:28:44 <wongk> apparently that's not the case in the US
May 30 19:28:52 <canadiancow|work> ?
May 30 19:28:52 <wongk> they'e calling the verizon nexus a failure
May 30 19:28:56 <canadiancow|work> oh
May 30 19:29:00 <canadiancow|work> verizon is different
May 30 19:29:06 <canadiancow|work> you can just buy it from play.google.com (if you are fortunate enough to live in a certain country)
May 30 19:29:22 <wongk> canadiancow|work: sure, but that can't be used with verizon :P
May 30 19:29:30 <canadiancow|work> so?
May 30 19:29:39 <g00s> ok, so it makes sense any device purchased on play would be updated by google
May 30 19:29:40 <wongk> selection of carrier ? selection of phone, imo
May 30 19:29:47 <canadiancow|work> g00s, just flash a factory image
May 30 19:29:50 <wongk> er, s/?/>
May 30 19:30:01 <canadiancow|work> g00s, flash https://developers.google.com/android/nexus/images and you win
May 30 19:30:15 <g00s> canadiancow|work: does flashing require a factory reset ?
May 30 19:30:23 <canadiancow|work> yes
May 30 19:30:34 <canadiancow|work> well...it wipes data during the process of unlocking the bootloader
May 30 19:30:34 <g00s> i was just reading yesterday that factory reset permanently disabled google wallet on that device foreever
May 30 19:30:38 <canadiancow|work> no
May 30 19:30:43 <canadiancow|work> thats IF you have wallet installed already
May 30 19:30:44 <canadiancow|work> and setup
May 30 19:30:47 <g00s> ok
May 30 19:30:47 <canadiancow|work> and you dont clear its data first
May 30 19:30:55 <canadiancow|work> if you buy a device, and the first thing you do is flash it, you're fine
May 30 19:31:01 <g00s> ah , ok
May 30 19:31:30 <g00s> since most of my users are using GB, i was thinking about getting a Nexus One too
May 30 19:31:42 <g00s> right now i am on Froyo ;)
May 30 19:32:52 <wongk> kickin it old school
May 30 19:33:12 <g00s> wongk: for testing
May 30 19:33:50 <wongk> they have emulators for that :P
May 30 19:34:19 <g00s> not if you need bluetooth :P
May 30 19:34:35 <wongk> =/
May 30 19:34:43 <g00s> although, google said that was coming soon … i hope very soon
May 30 19:35:00 <g00s> well, they said it was coming. they didn't say soon
May 30 19:35:56 <luxurymode> anyone know how to style the overflow menu in the actionbar?
May 30 19:36:46 <bitplane> hey guys, I can't get ADB to recognise a HTC One X under Ubuntu 12.04. I've put the vendorId from lsusb into a udev .rules file and restarted everything
May 30 19:36:54 <bitplane> am I missing anything?
May 30 19:37:16 <wongk> turned on USB debugging on the phone?
May 30 19:37:32 <bitplane> yeah I get the typical "??????? no permissions" thing
May 30 19:37:48 <bitplane> so it's connected at least
May 30 19:42:22 <g00s> what do you guys use for app analytics; to see how people are using your app ?
May 30 19:42:31 <JakeWharton> luxurymode: can't
May 30 19:43:33 <luxurymode> JakeWharton, oh ok. how about applying a style? android:actionDropDownStyle? i'd like a lighter look to it..currently using Theme.Sherlock.Light.DarkActionBar as my parent theme
May 30 19:43:49 <JakeWharton> yeah
May 30 19:43:55 <JakeWharton> that'll do it
May 30 19:44:34 <luxurymode> cool. what themes am i looking for? abs...what?
May 30 19:46:58 <luxurymode> @style/Sherlock.__Widget.Holo.DropDownItem?
May 30 19:47:02 <luxurymode> no that cant be right
May 30 19:47:49 <luxurymode> @style/Widget.Sherlock.Light.ListView.DropDown?
May 30 19:48:24 <g00s> hmm, no googlers
May 30 19:48:33 <JakeWharton> i see two
May 30 19:48:50 <JakeWharton> luxurymode: look in abs__themes.xml and abs__styles.xml and see whatever its doing
May 30 19:48:53 <JakeWharton> i don't remember specifics
May 30 19:49:44 <mikedg> are there any yahooers in here?
May 30 19:49:47 <luxurymode> gotcha, thanks
May 30 19:50:02 <g00s> mikedg: you are an aol'er right ?
May 30 19:50:20 <jrajav> Juno ftw
May 30 19:50:28 <mikedg> gnn.com
May 30 19:51:08 <g00s> actually, was looking for romain
May 30 19:51:21 <mdim> how do I compile android/os/Binder.java only?
May 30 19:52:21 <mikedg> you compile the entire platform
May 30 19:52:38 <mikedg> or get classes for the other files
May 30 19:53:06 <mdim> mikedg: but is it possible to compile one platform source file at a time?
May 30 19:54:54 <mikedg> sure, you need all of its dependencies compiled
May 30 19:56:12 <canadiancow|work> hi ctate
May 30 19:56:20 <mdim> mikedg: so, if I want to put my own implementation of Binder class into android.jar, how do I do that? First extract the whole android.jar, and then somehow compile my own Binder.java?
May 30 19:56:44 <g00s> has anyone seen an order stuck in the 'chargeable' state … for a while ? what does that mean. it looks like teh card was authorized but not actually charged.
May 30 19:56:52 <mdim> I had an idea to replace Binder.class in android.jar with my own implementation
May 30 19:57:43 <Snuffel> isn't it signed?
May 30 19:57:49 <Snuffel> oh wait it's open source
May 30 19:57:54 <ctate> yeah, you can't do that without building the whole phone image
May 30 19:58:02 <ctate> also, what changes do you want to make?
May 30 19:58:16 <g00s> oh hai ctate
May 30 19:58:34 <ctate> Binder / BinderProxy / etc are rather fundamental, and somewhat subtle to get right
May 30 19:59:12 <mdim> ctate: actually, I think it will be android.app.Service I'll have to stub, but that's not important right now
May 30 19:59:13 <wongk> psh. i hack them in my sleep. ;)
May 30 19:59:15 <mikedg> yeah, like if you were making sensible changes ot them, i would hope you could figure out how to compile them
May 30 20:00:14 <luxurymode> how can i send html as the text of action_send? doesn't seem to ever render it as html, like in gmail for example
May 30 20:00:27 <ctate> mdim: the question still stands: what changes are you intending, and to what purpose?  :)
May 30 20:01:12 <mdim> ctate: I'm working on a Google Summer of Code project. The goal is to have a tool for model checking Android apps
May 30 20:01:26 <ctate> interesting
May 30 20:01:33 <mdim> ctate: in order to do that, I'd have to stub some methods
May 30 20:01:40 <mdim> in particular, my job is to model Binder IPC
May 30 20:01:53 <ctate> you'll still need to build your own platform, but you have my permission to proceed.  :)
May 30 20:02:30 <mdim> ctate: as far as I figured it out, I'd need a custom android.jar
May 30 20:02:37 <mikedg> ahh interesting
May 30 20:03:50 <f00bar80> Anybody knows why youtube embeded videos cannot be played using the webview ? based on this code , >> http://stackoverflow.com/questions/5817805/how-can-we-play-youtube-embeded-code-in-android-application-using-webview , somebody mentioned it's a android default browser's flash and/or codec issue, i've tried to use opera as the browser, always it's the same issue.
May 30 20:03:59 <mdim> so, if I need to model Binder IPC, I'd need to stub some methods, probably in android.app.Service, or maybe in android.os.Binder
May 30 20:04:18 <mdim> and I thought I could simply remove android/os/Binder.class from the android.jar, and replace it with my own version
May 30 20:05:36 <mdim> another option is to take all the original Android Library sources and resources, replace some of them with my own implementation, and bundle them into a new android.jar
May 30 20:06:04 <mdim> which one seems easier to you?
May 30 20:09:06 <pragma-> http://ohmyfunny.browntechnolimit.netdna-cdn.com/pictures/7a674153c63cff1ad7f0e261c369ab2c.jpg
May 30 20:09:31 <woozly> guys, how to return some variable from Thread.run() { return myFunc(); } ?
May 30 20:10:14 <wongk> pragma-: tell me that really exists
May 30 20:10:46 <JakeWharton> it exists
May 30 20:10:49 <JakeWharton> we have one floating around
May 30 20:10:58 <wongk> that is awesome
May 30 20:11:00 <wongk> i want one
May 30 20:11:10 <canadiancow|work> lol
May 30 20:11:37 <wongk> can you imagine walking around talking on that
May 30 20:12:10 <pragma-> can you imagine carrying that thing everywhere?
May 30 20:12:31 <wongk> filed under: defeats the purpose
May 30 20:16:42 <g00s> it needs the handsfree extender
May 30 20:17:00 <g00s> so it resist above your shoulder easier
May 30 20:18:03 <wongk> attach like a heaphone band to keep it on your head, and where it like a bluetooth headset
May 30 20:20:33 <pragma-> shoulder extender, pft, real pros just cock their heads against their shoulder while lifting their shoulder up
May 30 20:20:36 <ninjai> achartengine font size is so small on my galaxy nexus I can hardly read it! Any way to change this? I tried looking at the API documentation but have had no luck in changing it so far...
May 30 20:21:33 <wongk> ninjai: i found some great docs here: http://code.google.com/p/achartengine/source/browse/#svn%2Ftrunk%2Fachartengine
May 30 20:22:39 <woozly> guys, how to show Toast message from Thread.run() ?
May 30 20:26:55 <ninjai> wongk, thanks for the link, but I still can't seem to find any way to get achartengine to change the size of the text around the edges of the chart.  The numbers are so teeny!
May 30 20:27:47 <wongk> sometimes you just have to be willing to grock the source
May 30 20:28:26 <ninjai> =\
May 30 20:30:37 <mwillbanks> on the ViewPagerIndicator is there any examples showing how to leverage activities with it from using the ActionbarSherlock?  basically looking to take the adapter and have it bring in an activity? TabHost has a setContent method in which you can pass an intent
May 30 20:30:55 <MarkG123> Hi, bit of an exclipse newbie.  I imported a Android library from my own SVN and it's not showing any Android related stuff from the ADT.  It doesn't list any "builders".  How can I fix it?
May 30 20:32:50 <ironhalik> How can I debug an error with my sqlite? The dubugger tells me that there was runtimeexception and noting more - logcat shows launch timeout
May 30 20:33:19 <ironhalik> and all Im trying to do is to get cursor for my database
May 30 20:34:21 <pragma-> achartengine needs demo apk to download
May 30 20:35:13 <evancharlton> make one and contribute it?
May 30 20:35:13 <t0mless> pragma- I don't think your Network Log works on droid2
May 30 20:35:28 <evancharlton> pragma-: the dev is pretty responsive and a nice guy
May 30 20:36:43 <JakeWharton> mwillbanks: ViewPagerIndicator is just a dumb indicator for the ViewPager, a support library class
May 30 20:36:49 <JakeWharton> no, you cannot use activities with ViewPager
May 30 20:37:12 <pragma-> I need to push my update that fixes the scroll that happens when you lift your fingers after pinchzooming in the graphing library I'm using
May 30 20:37:35 * pragma- wants to get a few more features added and bugfixes to that update, though
May 30 20:37:40 <JakeWharton> ActivityGroup, the long-deprecated class which allows switching between child activities, should be avoided at any and all costs
May 30 20:37:58 <JakeWharton> treat it as radioactive waste that will kick you in the nuts and steal your girlfriend
May 30 20:39:14 <hardyNH> If I have a bunch of C code how is best way to make use of it in an android app? do I need or port it all or can I just compile the c into lib that can be used or something?
May 30 20:39:22 <DammitJim> is it normal to call finish() from a login activity?
May 30 20:39:54 <Gentle> hardyNH: I've never used it but take a look at android NDK (native development kit, the stuff needed to compile C for android)
May 30 20:40:12 <wongk> DammitJim: yes
May 30 20:40:46 <hardyNH> Gentle: thanks thats what I was looking for
May 30 20:41:32 <pragma-> JakeWharton: excellent, i need someone to steal my girlfriend
May 30 20:42:12 <JakeWharton> TabActivity is here for you!
May 30 20:43:22 <DammitJim> how does one normally go about returning to the original activity where the user got redirected to log in?
May 30 20:44:00 <evancharlton> finish()?
May 30 20:44:36 <DammitJim> for example... I am on a search page, but the page requires you to be logged in... so when pressing the search button, it takes you to a login screen (after a dialog warning)
May 30 20:44:51 <DammitJim> how does one normally return to that page?
May 30 20:45:04 <DammitJim> I guess finish would just destroy the current activity and pop it from the stack?
May 30 20:45:04 <canadiancow|work> [16:44:00] <evancharlton> finish()?
May 30 20:45:18 <wongk> startActivityForResult?
May 30 20:46:26 <pragma-> snap, that reminds me -- what's the best way to update an actionbar button when a service starts/stops -- this is proving to be a nightmare of race conditions
May 30 20:47:19 <pragma-> I want a "start service/stop service" single toggle-ish button in the actionbar
May 30 20:47:36 <evancharlton> pragma-: yeah that's perpetually tough
May 30 20:47:44 <evancharlton> pragma-: the best way is to just set a static variable
May 30 20:48:00 <evancharlton> MyService.isRunning = {true,false}
May 30 20:48:20 <pragma-> but what if android kills the service?
May 30 20:48:34 <evancharlton> you'll still go through onDestroy in that case, no?
May 30 20:48:35 <wongk> a listener sound right to me
May 30 20:48:52 <evancharlton> pragma-: you might want to try a broadcastreceiver, too
May 30 20:48:56 <pragma-> evancharlton: my understanding is that onDestroy doesn't get killed -- android just kill -9's with extreme prejudice, right?
May 30 20:49:01 <pragma-> doesn't get called
May 30 20:49:10 <wongk> that's when it kills your entire app
May 30 20:49:14 <evancharlton> pragma-: kill -9 would kill your process, so your activity wouldn't be a live anyway
May 30 20:49:15 <pragma-> oh?
May 30 20:49:17 <evancharlton> alive*
May 30 20:49:23 <evancharlton> you can't kill -9 just a component
May 30 20:50:31 <wongk> i'd call a listener in onStartCommand/onDestroy
May 30 20:50:45 <g00s> if i am writing a file in a service, what is the best way to ensure its closed, since the service onDestroy() is not guaranteed to be called ?
May 30 20:51:47 <lcastro> Hi! I have a small problem... when slidingdrawer is closed i need to hide de handle...
May 30 20:51:59 <canadiancow|work> i would broadcast a message in onstart/destroy
May 30 20:52:02 <lcastro> i use setVisibility but dont work
May 30 20:52:06 <lcastro> any idea?
May 30 20:52:06 <canadiancow|work> and periodically
May 30 20:52:09 <canadiancow|work> (when running)
May 30 20:52:52 <wongk> canadiancow|work: seems like a little much for this purpose, doesn't it?
May 30 20:53:11 <evancharlton> wongk: and how do you bind to this listener?
May 30 20:53:21 <canadiancow|work> depends on a bunch of stuff
May 30 20:53:38 <wongk> evancharlton: the lisener unbinds itself
May 30 20:54:09 <evancharlton> wongk: binding to it would automatically start the service though, no?
May 30 20:54:17 <evancharlton> making a "start" button next to useless
May 30 20:54:29 <wongk> you'd have to have a static "registry" of listeners
May 30 20:54:32 <evancharlton> you'd have to bind, see if it's running (it is), then shut it down (if it's supposed to be stopped)
May 30 20:55:01 <wongk> i wouldn't use binding
May 30 20:56:23 <evancharlton> gotcha
May 30 20:56:48 <canadiancow|work> broadcast an intent!
May 30 20:56:51 <canadiancow|work> O HAI IM ALIVE
May 30 20:56:55 <canadiancow|work> O HAI IM DEAD
May 30 20:57:06 <MDijkstra> O BAI?
May 30 21:01:13 <g00s> unfortunately, the async nature of binding to services often makes the whole scheme an unnecessary pain in the ass
May 30 21:02:12 <g00s> if you don't have to read complex state information from the service, i find the AsyncQueryService model easy (pushing stuff to the service via a queue)
May 30 21:02:28 <g00s>  -- a regular IntentService kind of thing
May 30 21:03:27 <ninjai> with achartengine, what kind of dataset and series do I need to use if I want doubles on my Y axis and Strings on my X axis (dates)?
May 30 21:03:48 <evancharlton> ninjai: you'll almost certainly have to look at the achartengine docs for that
May 30 21:05:30 <alankila> ninjai: might be good idea to just make custom objects. override their toString and implement Comparable<T>
May 30 21:05:50 <alankila> I would guess that it will render any random object using toString & it might sort the set based on comparable
May 30 21:06:01 <alankila> but I have never used it, so this is just all a guess from my part
May 30 21:06:25 <ctate> g00s: and of course if the service is in your own app, then why are you bothering with bindings anyway except for lifecycle/OOM behavior? just make local method calls to it.
May 30 21:06:41 <ninjai> evancharlton, alright. I've been looking at it for a while now to no avail.
May 30 21:07:28 <canadiancow|work> how do you make a method call without binding?
May 30 21:07:38 <readme> to a service?
May 30 21:08:37 <readme> obviously you could do the singleton pattern atrocity
May 30 21:08:48 <readme> and return the instance from a static accessor.
May 30 21:09:15 <canadiancow|work> ok... let me rephrase. how do you make a method call without binding or making everything a global reference?
May 30 21:10:23 <readme> and without sending an intent to the service that tells it to call a method?
May 30 21:10:56 <canadiancow|work> [17:06:25] <@ctate> g00s: and of course if the service is in your own app, then why are you bothering with bindings anyway except for lifecycle/OOM behavior? just make local method calls to it.
May 30 21:10:59 <ctate> "making everything a global reference"
May 30 21:10:59 <wongk> it seems logcal to me to have a static method on the service class than can return a running instance
May 30 21:11:06 <ctate> look, it's a pointer.
May 30 21:11:07 <luxurymode> no no no
May 30 21:11:07 <ctate> to an object.
May 30 21:11:12 <ctate> and you can make method calls using it.
May 30 21:11:28 <ctate> MyService myService = new MyService();
May 30 21:11:39 <ctate> myService.lookMaNoBinder();
May 30 21:11:42 <alankila> just undo the android lifecycle nonsense, rely on process not getting killed, run some random service that prevents android from killing your task. :-p
May 30 21:11:49 <canadiancow|work> public static Myself GLOBAL_STATE; onCreate() { GLOBAL_STATE = this; }
May 30 21:11:50 <luxurymode> wongk, you should do something like Intent intent = new Intent(MyService.ACTION_DO_SOME_METHOD); startService(intent)
May 30 21:11:52 <ctate> if you want the lifecycle
May 30 21:11:55 <alankila> structure your UI on fragments with retainInstance=true
May 30 21:12:03 <alankila> it's almost as good as iOS after that
May 30 21:12:05 <ctate> then insert a bindService in between instantiating it and starting to call methods
May 30 21:12:06 <wongk> luxurymode: unnecessary
May 30 21:12:33 <luxurymode> what does unnecessary mean
May 30 21:12:41 <ericrichardson> Not necessary
May 30 21:12:42 <ctate> it means "not necessary"
May 30 21:13:00 <readme> ctate: interesting. so you can have a service lifecycle without calling startService?
May 30 21:13:05 <alankila> ctate: the problem with that bindservice is that it forces you to write your method calls in asynchronous style though
May 30 21:13:10 <luxurymode> if you want the service to return 1, then sure, you could have a method in there public static int gimmeANumber() {return 1;}
May 30 21:13:18 <ejb> Greetings, I'm getting ready to ad some advertisements to my app and I'm a bit confused. It seems as though most ad providers require that your app already be published in the market. Is this true?
May 30 21:13:37 <alankila> ctate: because you bindservice then it returns, you return to caller, then sometime later the bind happens, you do the work, if there is error you must call an error callback... so tedious.
May 30 21:13:49 <ejb> I'm looking at AdMob, for example.
May 30 21:13:52 <readme> ejb: maybe. but you might as well publish and then figure out ads after.
May 30 21:14:01 <luxurymode> also, there's no reason you cant bind to a service and also have it be started
May 30 21:14:01 <alankila> I explained this before, lamenting about the lack of a synchronous bindservice--one that binds and actually returns immediately with service in tow
May 30 21:14:03 <luxurymode> you can do both
May 30 21:14:22 <ctate> there cannot be a bind error when you're hosting the service yourself
May 30 21:14:22 <luxurymode> bind to it if you want to call methods directly or use intents
May 30 21:14:30 <alankila> ctate: the bind error is not the issue
May 30 21:14:34 <ctate> (yourself == in your own app process)
May 30 21:14:36 <alankila> it is the METHOD you call on the service after bind
May 30 21:14:37 <readme> alankila: the way I deal with that is by creating a callback that gets run after onServiceConnected is called
May 30 21:14:38 <alankila> the bind WILL succeed
May 30 21:14:45 <alankila> but you still don't know if the method you wanted to call succeeds or not
May 30 21:15:04 <ctate> uh, wtf?  of course you do.  it's just a function all; it can return a result however you want.
May 30 21:15:09 <ctate> *function call
May 30 21:15:11 <alankila> readme: indeed, but it's such a motherfucking pile of code which would all go away if we had synchronous bindservice
May 30 21:15:22 <alankila> ctate: you don't get it. If you don't have a service instance, you can't call a method on service.
May 30 21:15:22 <ctate> in the example i gave you aren't calling through a binder interface; you're just calling methods directly
May 30 21:15:30 <MDijkstra> how would that work? without blocking the main thread I mean
May 30 21:15:33 <alankila> you return to the caller, you have launched a bindservice, when the service connects then you call the method
May 30 21:15:55 <ctate> alankila: look at the LocalService example in API Demos
May 30 21:15:58 <alankila> NOW if error happens in that method, you're screwed, you returned long time ago to your original caller -- you need to design an asynchronous API that can report errors from service or success or any other result
May 30 21:16:02 <alankila> THAT is what sucks about it
May 30 21:16:12 <alankila> I can't believe I'm explaining this to you second or third time and we are still talking about this.
May 30 21:16:14 <ctate> i think you're mixing up a bunch of different things
May 30 21:16:22 <readme> i had 2 sales after i turned on "tap for tap" but it was hours later and they don't give  me stats on how long they were actually sending traffic for :|
May 30 21:16:25 <wongk> ctate: if you're not going through the lifecycle, whats the points of extending Service?
May 30 21:16:27 <evancharlton> I haven't been following, but I trust ctate
May 30 21:16:39 <MDijkstra> I wrote a small proxy for it that calls a callback when it's done binding
May 30 21:16:41 <ctate> first off, "Service" lifecycle does not require using a binder call interface in the first place
May 30 21:16:48 <readme> anyone have free sources of inbound app-traffic?
May 30 21:16:57 <ctate> Binder (AIDL) call interfaces are necessary only if you're calling across process boundaries
May 30 21:17:00 <alankila> ctate: I am using a local binder obviously
May 30 21:17:03 <alankila> jesus christ
May 30 21:17:15 <alankila> I get my Service instance nice and tidy into my activity or fragment
May 30 21:17:20 <ctate> if the Service object lives in the same process, you don't actually need to use a binder interface to talk to it
May 30 21:17:23 <evancharlton> chill out a bit, alankila
May 30 21:17:24 <alankila> the trouble is, I don't get that instance WHEN I want it. I get it sometime later.
May 30 21:17:27 <ctate> just treat it like any other object
May 30 21:17:36 <ctate> ahh, yes you do.
May 30 21:17:37 <alankila> I do, but there is no way to get the service instance
May 30 21:17:54 <ctate> you are saying "service instance" but i think what you mean is that you don't get the *Binder* to it
May 30 21:17:57 <alankila> bindService returns but the serviceconnection is still dead
May 30 21:18:00 <ctate> and i am saying you don't need that
May 30 21:18:20 <alankila> the serviceconnection fires sometime later as android executes all those android object state transitions
May 30 21:18:25 <alankila> there must be some kind of queue of operations to do there
May 30 21:18:30 <ctate> yeah but those are just bookkeeping about the lifecycle
May 30 21:18:44 <ctate> they aren't relevant as far as using the object as a local object in your own app
May 30 21:18:50 <alankila> you just don't get it
May 30 21:18:51 <alankila> I give up
May 30 21:18:54 <ctate> "object" meaning the instance of your Service subclass
May 30 21:19:11 <ctate> where are you anticipating an error condition going undiscoverable here?
May 30 21:19:22 <alankila> FooService has method bar()
May 30 21:19:26 <alankila> it might throw or not throw
May 30 21:19:28 <ctate> okay
May 30 21:19:30 <ctate> sure
May 30 21:19:40 <alankila> I would like to do: fooService.bar(); but fooService is null until bind has happened
May 30 21:19:51 <ctate> not if you construct it before you bind
May 30 21:19:56 <alankila> android's bindService does not make it easy for me to get fooService instance created and returned to me
May 30 21:19:56 <ctate> fooService = new FooService();
May 30 21:20:08 <alankila> ctate: WTF? how is android going to manage this for me?
May 30 21:20:13 <b1n0ry> http://pastebin.com/Xv3B1ss9 <-- any idea why my progressbar is visible all the time and never disappears?
May 30 21:20:17 <ctate> "android" doesn't manage anythign but the lifecycle
May 30 21:20:47 <ctate> after you instantiate the object, you then call bindService() and it will do all the lifecycle bookkeeping, and eventually ask *that same instance* for its binder
May 30 21:20:50 <alankila> ctate: well if I do 'new FooService()', how do I tell android that I have constructed this instance
May 30 21:20:55 <ctate> but you never actually care about the binder being returned
May 30 21:21:05 <b1n0ry> i'm running setProgressBarIndeterminateVisibility(false); and the progressbar (actionbar spinner) is still visible.
May 30 21:21:30 <alankila> ctate: and/or: how do I prevent android from returning to me a second FooService in case the platform already has one of those existing?
May 30 21:21:40 <b1n0ry> for the record, ii also tried setProgressBarIndeterminateVisibility(Boolean.FALSE);
May 30 21:22:07 <ctate> just have its onBind() return the local service pointer as the right kind of binder
May 30 21:22:30 <alankila> ctate: let's take a few steps back.
May 30 21:22:35 <DammitJim> is a date in milliseconds always in UTC?
May 30 21:22:40 <DammitJim> no offset calculations needed, right?
May 30 21:22:49 <ctate> there's this Service lifecycle-management thing
May 30 21:22:58 <alankila> my activity does: 'FooService fooService = new FooService()'. If I now do startService(this, FooService.class), how does android know that I already have a fooService instance and returns that?
May 30 21:23:33 <ctate> you actually need to stash 'fooService' into a static or whatever, and then FooService.onBind() returns *that*
May 30 21:23:43 <ctate> [or the associated binder existence or whatever]
May 30 21:23:50 <alankila> ctate: okay... does the onBind happen before the onStart?
May 30 21:24:01 <alankila> so that there's no onStart called on instance that actually gets thrown away?
May 30 21:24:15 <ctate> you need both 'start' and 'bind' semantics?
May 30 21:24:23 <alankila> ctate: not sure, just trying to map your solution
May 30 21:24:25 <ctate> remember that they're orthogonal
May 30 21:24:34 <ctate> i'm specifically talking about the bind scenario
May 30 21:24:35 <DammitJim> Calendar.getInstance().getTimeInMillis() gives you a Long in UTC, right?
May 30 21:24:40 <alankila> additionally how does 'new FooService()' get onStart/onStop called?
May 30 21:24:45 <ctate> where it's an object that exists to have methods called
May 30 21:24:50 <g00s> so it kinda sounds like you have your service be a singleton FooService.getInstance(), and then have the binder return that. but then we explicitly get involved in the semantics of android's notion of service as a singleton
May 30 21:24:57 <ctate> start/stop is a different lifecycle and a very different usage pattern
May 30 21:25:13 <alankila> okay. I've heard enough. Services are still useless to me, this is the short of it. :-/
May 30 21:25:26 <alankila> I rather just use normal java objects and a fake service that just prevents android from killing my app.
May 30 21:25:33 <ctate> that amounts to the same thing
May 30 21:25:42 <ctate> so, you're already doing what i suggested wihtout knowing it :)
May 30 21:26:04 <ctate> [and it doesn't prevent it; it's just a formal notification that you're trying to get work done while in the background so please leave you alone if possible]
May 30 21:26:06 <alankila> but services *would* be useful to me if I had synchronous bind
May 30 21:26:17 <g00s> the contacts app already does that, btw … it has something called DummyService or the like
May 30 21:26:22 <alankila> a bindService call that on return actually hands me a Service instance that android has configured, executed onStart on, etc.
May 30 21:26:32 <alankila> EmptyService, g00s
May 30 21:26:32 <ctate> but you don't *care* about synchronous bind if you're just binding to an extant instance of your Service subclass
May 30 21:26:38 <g00s> alankila: right
May 30 21:26:52 <ctate> it's not hard to write a wrapper around bind/callback that makes it synchronous
May 30 21:27:08 <MDijkstra> won't you block the main thread?
May 30 21:27:10 <ctate> you just can't call that from the main thread, in the general case, because it may take significant time to bring up the objects etc
May 30 21:27:11 <MDijkstra> isn't that a no-no
May 30 21:27:18 <alankila> ctate: I would just like to user Service as an abstraction to cover background processing when no other UI is visible. It's just that this implies some amount of the application's work must be done in that service... and thus I try to call methods on service, which may throw
May 30 21:27:41 <ctate> right.  so what's the problem?
May 30 21:27:51 <alankila> onCreate can't call any methods on a service directly, for starters
May 30 21:27:53 <g00s> ctate: i tried binding from a non-ui thread and using a countdownlatch to block that thread until the service was connected, but the arrangement would not work
May 30 21:27:59 <alankila> because the service is not yet bound, it will be bound sometime later
May 30 21:28:01 <ctate> are you talking about how not all kinds of exceptions get propagated through binder calls?
May 30 21:28:02 <MDijkstra> again, what if you just create a class implementing the service callback interface
May 30 21:28:18 <MDijkstra> and then proxy to a serviceBoundSuccessfully() type of interface
May 30 21:28:38 <ctate> g00s: huh, weird.  i've done stuff like that successfully, so i know it's possible....
May 30 21:28:39 <alankila> but android's problem is that if you for instance go back to a running application you may have whole lot of activity and fragment state transitions queued, you go from oncreate to onstart to onresume before the service binds
May 30 21:28:41 <MDijkstra> then you simply bind from onCreate() and do the stuff you would normally do from there in serviceBoundSuccessfully()
May 30 21:29:08 <g00s> ctate: can you publish a blog entry on d.android.com maybe about all these nifty service tricks?
May 30 21:29:11 <MDijkstra> (and show a loading dialog between those points)
May 30 21:29:14 <g00s> idioms, patterns, whatever
May 30 21:29:20 <g00s> we all know the LocalBinding stuff
May 30 21:29:46 <alankila> so I run into the problem that effectively my UI may already be visible by the point I finally can reach my service and know what I should be displaying in my UI. It makes the code *really* unpleasant to read and brittle
May 30 21:29:47 <ctate> alankila: so then i'm back to saying you should just instantiate your background-processor-object as a static, mostly ignore the results of binding and just treat that as large-scale lifecycle / don't-kill-me handling, and proceed with local method calls instead of using the binder you get from onServiceConnected()
May 30 21:30:29 <ctate> i do keep saying that binding to a service that exists only within your local app process is not actually very important except as far as lifecycle management
May 30 21:31:16 <ctate> you know you can use the object as soon as you've constructed it; the rest is just asynchronous notifications to the OS that it's to be considered a "service" for lifecycle purposes etc.
May 30 21:31:40 <alankila> anyway all this is relatively pointless. I really hate the way android manages object lifecycles, this is truly the long and short of it.
May 30 21:31:54 <MDijkstra> ctate: that isn't strictly true is it?
May 30 21:31:55 <ctate> it makes a lot more sense in the cross-app cases.
May 30 21:32:01 <alankila> I hate that startService doesn't start a service, it just sets a flag somewhere that a service should be started and sometime much later it actually happens
May 30 21:32:01 <ctate> which is what it's mostly for.
May 30 21:32:10 <MDijkstra> ctate: ie, it won't work as a context until then?
May 30 21:32:35 <MDijkstra> at least, I've had that issue with a RecentSuggestionsProvider
May 30 21:32:50 <MDijkstra> I'd like to read the AUTHORITY from a file, but can't in the constructor
May 30 21:33:03 <ctate> so pass it the app context when you construct it?
May 30 21:33:18 <ctate> but yeah that's a reasonable point
May 30 21:33:26 <MDijkstra> ctate: I don't construct it
May 30 21:33:30 <MDijkstra> setupSuggestions does
May 30 21:33:34 <MDijkstra> through an Intent, I guess
May 30 21:33:44 <ctate> oh then you're talking about the cross-process case :)
May 30 21:34:00 <ctate> and then yes you need to let the full lifecycle etc run
May 30 21:35:01 <ctate> i guess i'm really talking about what alankila is already doing: just using their own non-Service object whose role is to run out-of-band stuff; and having a stub Service that doesn't actually do anything but sit there and look pretty to the OS's OOM killer
May 30 21:35:08 <g00s> ctate: but isn't it a bit dangerous to not wait for the binding? if you do FooService foo = new FooService() in onCreate, for example, and then bind in onStart, the service isn't technically alive until some time way after onResume. what if the ui gets paused, you poked your service so it has state, but since it isn't technically started the process gets killed for some reason ?
May 30 21:35:28 <alankila> ctate: I called mine ProtectionMoneyService. The idea is that it pays off the android mafia goons...
May 30 21:35:40 <ctate> g00s: yeah, back it up... i'm talkign about somethign that isn't a Service for OS-management purposes
May 30 21:35:48 <MDijkstra> that and waiting for the binding isn't that different from doing it immediately
May 30 21:35:49 <ctate> alankila: snrk!
May 30 21:36:05 <alankila> not very professional but I was very tired and sick of this part of android after struggling for a long time and half a dozen false starts trying to merely comprehend how do I get services done "properly"
May 30 21:36:30 <alankila> and it turned out that it was not possible, which is very unpleasant state of cognitive dissonance to me
May 30 21:36:34 <ctate> or, if you want to do it that way, the yes it's fairly easy to write a little bit of code that lets you write something like FooService myService = bindSynchronous(FooService.class); or whatever.
May 30 21:37:10 <g00s> ctate: can you put this in a blog, and really go into depth about corner cases & safety :)
May 30 21:37:19 <alankila> ctate: so you say. I would be very curious of your implementation, especially if it actually looks and feels like actual android service, with onstart, stop, that whole lifecycle thing done with it
May 30 21:37:26 <g00s> i have seen many intelligent people struggle with this
May 30 21:37:43 <ctate> again, look at the LocalService example in API Demos
May 30 21:37:44 <alankila> because so far I have not worked out how to do that thing which you claim to be easy to do.
May 30 21:37:45 <MDijkstra> the service model is a bit opaque yeah
May 30 21:37:53 <MDijkstra> took me a while to figure out LocalService as well
May 30 21:38:03 * alankila is using localservice. Doesn't understand the point.
May 30 21:38:14 <g00s> every time i think i might have an epiphany and then get reminded about LocalService example, i doubt we are on the same page
May 30 21:38:45 <ctate> ha ha
May 30 21:38:48 <alankila> localservice has fuck-all to do with the problem which is that writing that bindSynchronous() has stymied any person who has attempted it
May 30 21:38:51 <ctate> okay yeah LocalService is a little odd
May 30 21:39:15 <ctate> alankila: mostly it's discouraged to do any of this stuff synchronously
May 30 21:39:33 <ctate> you're intended to proceed with things in response to the onServiceConnected() callback
May 30 21:39:44 <alankila> yes... but I need some amount of ordering to help me along
May 30 21:40:14 <ctate> the standard pattern is to post work to your looper
May 30 21:40:18 <alankila> for instance, if sometimes service binds before views are created I know that I can use service when I set up views... if sometimes it happens in a different order, then I need a different code to deal with that
May 30 21:40:38 <alankila> I need to know what the order is. Synchronous operation is wonderful in establishing that order.
May 30 21:41:19 <ctate> for example, what you'd do there is to bind in your activity's onCreate() [and maybe put up a placeholder UI], then when the onServiceConnected() came through you'd proceed with setting up your UI
May 30 21:41:20 <MDijkstra> hm, can't you simply show a loading view until binding is complete? then create your views when binding is complete?
May 30 21:41:30 <alankila> I would gladly throw virtually any part of asynchronous nature and callback to the wolves if I can salvage a synchronous model from it. For 99.9 % of cases it will be good enough and much simpler.
May 30 21:41:47 <ctate> you can set up the UI at any time; it' sjust that onCreate() is often the most convenient time to do so
May 30 21:41:53 <ctate> often, not always :)
May 30 21:42:10 <alankila> ctate: tried that, actually
May 30 21:42:19 <MDijkstra> well, in my iOS code I also have a complex state machine because you can often not setup stuff from creation methods
May 30 21:42:25 <MDijkstra> hard to avoid usually
May 30 21:42:38 <MDijkstra> ie, stuff just happens asynchronously
May 30 21:42:41 <alankila> but I ran into problem that I also had fragments and their views were getting created despite the main view was not yet set up or something
May 30 21:42:51 <alankila> It was a murky circumstance and I did not fully understand why
May 30 21:43:13 <alankila> maybe I should revisit it
May 30 21:43:43 <pfn> a setRetainInstance fragment might encounter a situation like you describe
May 30 21:43:46 <alankila> but anyway fragments were being set up when the application resumed from killed state despite the main UI was still blank, no fragments there, just an empty layout
May 30 21:43:54 <ctate> alankila: hmm yeah i can see how having manifest-declared fragments for it to proceed with creating might be a big wrinkle
May 30 21:43:57 <alankila> and because fragments needed the service they all died of course
May 30 21:44:27 <ctate> they do potentially add a fair bit of complexity
May 30 21:45:19 <MDijkstra> heh, binding to a service from a fragment is also fun
May 30 21:45:44 <MDijkstra> (ie, the fragments used to be activities that bound to the service)
May 30 21:45:56 <alankila> but complicated problems like these can be solved by just never letting android's task killer loose at your application. I'm sad to say but this solution is surprisingly robust. To use it, you would want some kind of limitations like maybe remove your protectionmoneyservice 5 minutes of idle in your application
May 30 21:46:15 <alankila> so that your application will eventually die
May 30 21:47:13 <alankila> it's a programming model never intended by android designers, but the primary idea of the design is that you take control of the lifecycles of the objects when they get into the way of what you are doing. Anybody can write working application after that.
May 30 21:47:30 <alankila> but with android managing lifecycle, the complexity tends to shoot through the roof whenever any form of interaction between lifecycle-managed objects is required
May 30 21:47:32 <g00s> MDijkstra: yeah, i had that situation too. and even more weird, those fragments (binding to the service) are not in ViewPager or whatever
May 30 21:47:35 <ctate> you can't actually keep the OS from killing your app
May 30 21:47:54 <alankila> ctate: that's alright. The protectionmoneyservice solution is still practically perfect :-p
May 30 21:47:57 <ctate> even if you're showing a notification and proclaiming yourself a *foreground* service, it just means you'll be killed relatively late
May 30 21:47:59 <ctate> heh
May 30 21:48:22 <ctate> next question, why do you particularly care?  is the app that incapable of bringing itself back up coherently after being killed?
May 30 21:48:32 <alankila> It's OK for the app to die really. I don't mind. It's possible to even throw all that view/activity/fragment state saving code out too if you look at it this way, your application is semiperpetually there
May 30 21:48:35 <birbeck> ctate: i want my app to always run, even if the user removes the battery!
May 30 21:49:01 <g00s> ctate: i think all these lifecycle things, which to me seem part of android's philosophy of memory management, are becoming moot since devices have enough memory, a lot of this footwork seems irrelevant - or will be irrelevant soon
May 30 21:49:06 <alankila> ctate: just thinking that I don't like android's way to manage lifecycles and offer a suggestion that there is an escape from it
May 30 21:49:19 <ctate> devices absolutely do not have "enough memory" to avoid killing apps.
May 30 21:49:31 <ctate> there are games out there with 100M memory footprint
May 30 21:49:53 <birbeck> games *sigh*
May 30 21:50:02 <ctate> birbeck: also, browsers.
May 30 21:50:04 <ctate> which includes book readers.
May 30 21:50:17 <g00s> yeah - i guess a more traditional approach to virtual memory would be nice
May 30 21:50:23 <birbeck> ctate: i think i have 17 tabs open in chrome on android :)
May 30 21:50:29 <ctate> g00s: you mean including a hard disk?  :)
May 30 21:50:47 <p_l> g00s: android keeps rather traditional approach with a bit more sensible OOM Killer
May 30 21:50:47 <ctate> birbeck: ha ha yeah chrome on android is waaaaaay more memory-hungry than the non-chrome browser
May 30 21:50:56 <alankila> I think that android could actually fairly easily swap applications to disk instead of killing them. Reloading them from swap would be of comparable time penalty most likely to just restarting them
May 30 21:51:06 <ctate> alankila: if we had disk we'd use it
May 30 21:51:12 <ctate> swapping to flash devices is a nonstarter
May 30 21:51:25 <ctate> it's much too slow and it would burn the flash out too fast
May 30 21:51:33 <ctate> [it's been tried]
May 30 21:51:40 <ctate> MUCH too slow
May 30 21:51:41 <alankila> hardly. I've used flash swap on memory-starved laptop for years and the stick is still working.
May 30 21:51:49 <ctate> i mean, there are flash parts out there that have like 1.5 second write times.
May 30 21:52:01 <alankila> and the performance is like 10 MB/s. If you think swapping is too slow, then how come you don't think loading applications from disk is too slow?
May 30 21:52:07 <ctate> alankila: laptop & desktop grade flash ram is a whole different ball game
May 30 21:52:12 <ctate> it's much too expensive for phones
May 30 21:52:35 <alankila> ctate: eh? didn't android just go recently with this sort of generic flash chips with hardware block remapping layer and all that stuff?
May 30 21:52:38 <ctate> also tends to have big ram caches fronting it etc
May 30 21:52:45 <alankila> my sgs at least seems to run ext4 on mmcblk0psomething
May 30 21:52:53 <ctate> alankila: emmc-type flash includes a controller that does the block remapping etc, yes
May 30 21:52:59 <ctate> but hte churn rate is still really high
May 30 21:53:05 <ctate> and emmc is still *slow*
May 30 21:53:08 <ctate> really, really slow.
May 30 21:53:09 <alankila> I think today modern android phones just have exactly same chips as those usb sticks
May 30 21:53:23 <jreck> putting your swap file on a USB stick also kills it in a hurry
May 30 21:53:36 <ctate> plus you're at the mercy of hte emmc controller; that's where those specific devices hit their 1.5-second write delays sometimes.  SIGH.
May 30 21:53:39 <alankila> jreck: so you say but I did it for years and I *still* use that same stick
May 30 21:54:16 <alankila> I had 500 MB in swap and the root filesystem on it too. All the talk about how flash can't stand write cycles is probably more than a little obsolete these days.
May 30 21:54:40 <ctate> tell that to our bricked sample phones :)
May 30 21:54:47 <alankila> either because flash got better, or because there's spare capacity to migrate blocks to transparently when the chips do develop unrecoverable errors
May 30 21:54:52 <p_l> alankila: most PCs have much less swapping than phone would get
May 30 21:54:57 <alankila> either way, they seem to last for a long time, much longer than people seem to expect
May 30 21:54:58 <ctate> [seriously, the parts are just not the same in phones; they're much cheaper.]
May 30 21:55:16 <g00s> ok, so flash is limiting right now, but it won't be around forever. i remember reading about HP's memristor memory that would dram/sram and be suitable for ssd - supposed to be here in 2013 i thought. samsung was working on something similar
May 30 21:55:32 <ctate> yeah, the next gen will help al ot
May 30 21:55:35 <p_l> g00s: memristors would make SRAM
May 30 21:55:35 <ctate> also batteries
May 30 21:55:39 <alankila> p_l: I know, but I had 1 GB of memory and I had around 500 MB of swap constantly in use. It was the fastest usb stick I could find, it advertised 20 MB/s read/write speeds and by god it could do them.
May 30 21:55:46 <ctate> [power being the other significant issue that makes the phone parts different]
May 30 21:56:02 <ctate> p_l: bubble ram!
May 30 21:56:23 <p_l> ctate: MRAM! Let the core memory rule again! :P
May 30 21:56:25 <alankila> I was able to extend the life of that laptop for years because of this memory enhancement trick. Everyone was horrified, telling me how it's going to die right away, and it still works perfectly as far as I can tell, although the laptop itself has gone to pieces
May 30 21:56:49 <p_l> hmm... I might put my swap on some usb key as well
May 30 21:57:02 <p_l> might make for smoother sailing when I run java or chrome
May 30 21:57:27 <alankila> I am now using 100 GB SSD in front of 1.5 TB drive, it probably sees good deal of overwriting every day. I haven't measured, but it's probably at least 200 GB written to it every day. I'll see how long it lasts
May 30 21:57:37 <g00s> ctate: so, to sum up this FooService = new FooService() … way; is this industrial strength and safe ?
May 30 21:57:54 <alankila> g00s: no, remove Service name, no point to make it one
May 30 21:58:02 <alankila> :-p
May 30 21:58:25 <alankila> and by "in front" I mean I have a software driven hybrid harddrive solution
May 30 21:58:26 <g00s> MDijkstra: what are you using for a state machine on IOS ?
May 30 21:58:36 <g00s> i used SMC its pretty nice
May 30 21:59:09 <ctate> ==alankila
May 30 21:59:19 <ctate> [re remove Service, no need to make it one]
May 30 21:59:28 <ctate> and then use a dummy do-nothing service just to track/manage lifecycle
May 30 21:59:37 <ctate> i.e. what alankila was already doing :)
May 30 22:00:03 <ctate> i get the impression that people not infrequently use Service when all they really need is something like HandlerThread
May 30 22:00:55 <alankila> So ... mmc versus emmc, the difference is?
May 30 22:01:26 <ctate> the comparison is usually emmc vs nand
May 30 22:01:42 <ctate> nand parts are raw flash; you need to do wear levelling etc yourself
May 30 22:01:58 <ctate> emmc parts have a controller that fronts the raw flash and does its own wear levelling out from under you
May 30 22:02:15 <ctate> making it suitable for e.g. unmodified ext4
May 30 22:02:33 <ctate> whereas on raw nand you need to use a levelling filesystem like yaffs2 or whatever
May 30 22:02:34 <alankila> ... so, if I have a device through a linux's mmcblk* driver stuff, and it uses ext4, does it have a wear-leveling controller or not?
May 30 22:02:46 <ctate> recent Google-involved Android phones have used emmc parts and ext4
May 30 22:02:57 <alankila> mmc is apparently just the interface, emmc is explicitly stating that it has a controller
May 30 22:02:59 <ctate> if it uses ext4 it almost certainly has emmc hw
May 30 22:03:04 <ctate> otherwise you're gonna have a bad time
May 30 22:03:13 <ctate> ext4 is *not* suitable for raw nand
May 30 22:03:30 <ctate> i've heard that iOS devices use raw nand and apple's own proprietary filesystems
May 30 22:03:49 <alankila> I'm going to guess you are right and there is a hardware wear-leveler there
May 30 22:04:09 <p_l> ctate: afaik iOS goes with HFS+ and some translation layer
May 30 22:04:14 <alankila> I don't exactly want to test it, but I imagine I find out in some unpleasant way relatively quickly if not
May 30 22:04:15 <g00s> interesting story about apple, nand, clouds, and flash http://www.appleinsider.com/articles/12/05/30/apple_likely_to_use_nand_storage_in_upcoming_macbook_pros_analyst_says.html
May 30 22:04:28 <p_l> ctate: they do, however, have batshit insane arrangements of flash chips
May 30 22:04:38 <g00s> of course, s/apple/google, s/ios/android it has same theme
May 30 22:05:22 <ctate> p_l: phone innards are pretty wack in general :)
May 30 22:05:42 <p_l> ctate: there's "wacky" and there "apple-level bugfuck crazy for nothing"
May 30 22:06:01 <kakazza> Why, is there Macbook Air storage that bad?
May 30 22:06:06 <alankila> it's interesting that the nand stuff apparently got removed... maybe the emmc-like chips just got so good that there was no downsides to having them, and there aren't that many kickass filesystem for linux to use on raw nand
May 30 22:06:24 <ctate> p_l: it's usually so that they can shave a tenth of a millimeter somewhere
May 30 22:06:26 <alankila> iirc both yaffs and yaffs2 suffer from slow mounts over large volumes, and ubifs from nokia might or might not work, no idea
May 30 22:06:27 <p_l> kakazza: Macbook Air is simple block storage. iPhone has weird flash connection
May 30 22:06:46 <ctate> alankila: there aren't *any* kickass linux filesystems to use on raw nand.  :)
May 30 22:06:56 <ctate> we used yaffs2, and unfortunately it has significant issues
May 30 22:07:02 <p_l> ctate: sometimes it's for "let's make it complex and hard to crack" except the hard to crack part could have been done by enabling trusted boot instead
May 30 22:07:17 <alankila> the slow mount at least should hurt with yaffs2. Not sure if there are other problems.
May 30 22:07:22 <ctate> it's slow
May 30 22:07:24 <kakazza> ARM TrustZone! \o/
May 30 22:07:27 <ctate> it doesn't support concurrent access
May 30 22:07:33 <kakazza> I wish they'd call it FriendZone, tho
May 30 22:07:37 <ctate> it doesn't included in-use blocks in its wear levelling
May 30 22:07:41 <p_l> alankila: yaffs(2) has global lock
May 30 22:07:57 <ech0s7> i would launch the panorama activity stocked in android, i have tried with this: http://ideone.com/QUsHT
May 30 22:08:00 <ech0s7> but i get error
May 30 22:08:02 <ctate> it has linear performance degradation as fs utiliziation rises
May 30 22:08:12 <ech0s7> java.lang.ClassNotFoundException: com.android.camera.panorama.PanoramaActivity
May 30 22:08:12 <alankila> ctate: okay, I get it. So it isn't at the level of quality expected from a phone filesystem.
May 30 22:08:16 <ctate> [until you hit 80-85% utilization, when performance falls off a CLIFF]
May 30 22:08:33 <ctate> alankila: basically, yeah.  it's adequate but not what i'd call *good*
May 30 22:08:42 <ctate> and the edge case behaviors are quite bad
May 30 22:08:53 <alankila> know anything about ubifs?
May 30 22:09:37 <p_l> ctate: the use of yaffs2 always made me scratch my head
May 30 22:09:38 <ctate> i don't other than that it exists; it might be getting good
May 30 22:09:52 <ctate> p_l: reliability, mostly; in 2008 there wasn't an obviously better alternative
May 30 22:10:27 <alankila> I remember that galaxy s 1 had nand, regular internal emmc-like storage and external emmc-like storage
May 30 22:10:40 <p_l> ctate: well, when I was looking at it, I thought it might have been better to go with software remapping and putting "normal" fs on top
May 30 22:10:48 <alankila> cyanogenmod did away with yaffs2 on that as well... I wonder maybe they just threw that internal nand away completely
May 30 22:11:04 <alankila> should look into it, a bit curious what exactly they did
May 30 22:11:36 <ctate> p_l: effort vs risk; writing your own wear levelling / bad-block lockout / etc is not trivial
May 30 22:11:43 <ctate> and yaffs2, y'know, *existed*.  :)
May 30 22:11:44 <alankila> p_l: that was done at least by samsung on galaxy s 1, there was rfs on top of wear-leveling abstraction on top of mtd
May 30 22:11:54 <ctate> nowadays yes ubifs looks like it'd be a reasonable thing to try
May 30 22:12:23 <ctate> the magic trick is remapping in-use blocks so you an level across the whole flash part even when the disk is largely committed
May 30 22:12:34 <ctate> if ubifs does that then it's a big yay
May 30 22:12:37 <alankila> I reinstalled the OS on galaxy s 1 many times. Each time I did that I wondered if it formatted over that wear-leveling portion as well
May 30 22:13:02 <p_l> ctate: I believe there were more solutions than yaffs2, including existing wear-leveling for mtd devices
May 30 22:13:11 <alankila> ctate: that does leave the question: where do you store the wear-leveling information...
May 30 22:14:16 <ctate> that too :)
May 30 22:15:00 <alankila> stochastic wear-leveling starts to sound appealing to me... just throw the blocks in randomly
May 30 22:15:12 <alankila> and never allow files to be deleted
May 30 22:15:26 <alankila> what, this isn't a good filesystem? Sucks to be on MTD.
May 30 22:17:07 <alankila> although honestly, once you start wear-leveling with nearly full filesystem you're going to have to do really ugly shit like move data from one erase block to another
May 30 22:18:06 <alankila> so that stuff which is not changing is placed on the flash block that happened to see a whole lot of erase cycles. And this sort of thing works best if you set aside some % of the flash so that you can have free erase blocks when you need them too, otherwise you may end up doing something like defragmenting and that throws your i/o predictability to hell
May 30 22:19:16 <ctate> right. i.e. purpose-built fs.
May 30 22:19:32 <alankila> maybe if the erase blocks are large, say, 512 kB, you don't really need a whole lot of space to record the number of erase cycles seen on one particular block. For instance, a 32 GB filesystem with 512k blocks only needs 32 * 1024 * 2 = 65536 bytes to record the erase counts
May 30 22:20:06 <alankila> so you can store 8 erase block updates inside one erase block before you have to erase that bookeeping block
May 30 22:20:31 <alankila> and if you aren't very militant about storing that info but rather keep it in memory, you can maybe get away with just saving updates once a day or something
May 30 22:20:36 <p_l> alankila: or you can use a different method to store erase history
May 30 22:20:46 <alankila> p_l: oh, sure
May 30 22:21:05 <p_l> some scheme that let's you reconstruct that data
May 30 22:21:06 <alankila> some kind of ranged updates against a snapshot
May 30 22:21:53 <p_l> alankila: a function f(block,init-state) which can derive the data based on block allocation algorithm
May 30 22:22:56 <p_l> with init-state being kept in rarely-updated config block
May 30 22:23:24 <alankila> there's still that remapping block you also need which tells where each logical copy is, but maybe that doesn't need a whole lot of updates either. Or maybe these data structures could be combined. In any case it's not a problem to spend even 128 MB on this sort of thing, you can really waste a lot of space if you want to and just keep on appending updates
May 30 22:23:40 <alankila> 128 MB against, say, 32 GB is practically nothing
May 30 22:24:16 <g00s> hm, i guess Technicolor is the new patent troll
May 30 22:24:23 <p_l> g00s: hmm?
May 30 22:24:43 <g00s> p_l: http://yro.slashdot.org/story/12/05/30/2053220/technicolor-takes-aim-at-apple-samsung-others-for-patent-infringement
May 30 22:24:59 <alankila> I'm still mildly attracted to the stochastic wear-leveling idea
May 30 22:25:24 <alankila> you store data on random erase blocks and you rearrange the erase blocks randomly to prevent overwearing any particular block
May 30 22:25:28 <alankila> sounds crazy but it might just work
May 30 22:25:46 <p_l> ... can't we just agree to dispose of patent trolls by putting them in place of R-36's warheads so that we can have nice fireshow instead of just dismantling those rockets? ;)
May 30 22:27:46 <p_l> cause we can't just put them at strike site thanks to comprehensive nuclear test ban treaty
May 30 22:28:19 <readme> but, I have patented that nuclear warhead technology and, and
May 30 22:29:31 <p_l> readme: at least in USA it's not patentable, because even figuring it on your own is enough to make you disappear
May 30 22:29:37 <p_l> and patents require publication :D
May 30 22:30:11 <p_l> (the disappear part is thank to one of biggest pieces of legal bullshit I ever learned of, "born secret")
May 30 22:32:56 <ech0s7> i'm trying to launch internal activity com.motorola.Camera.Camera
May 30 22:32:57 <ech0s7> http://ideone.com/86fqJ
May 30 22:33:19 <ech0s7> it's strange because it find the class, but it doesn't load it
May 30 22:33:42 <readme> p_l: interesting. i wasn't aware. although this makes perfect sense.
May 30 22:33:54 <dragorn> ech0s7: why not just send a startactivity
May 30 22:33:54 <ech0s7> the output of my test case is: found!, is null!!
May 30 22:34:14 <p_l> readme: I prefer the polish approach, where IIRC military gets first pick on everything submitted to patent office
May 30 22:34:14 <ech0s7> dragorn, because at line 31 cameraClass is null
May 30 22:35:02 <dragorn> ech0s7: you don't need to load code from it (i'm actually really not sure what you're trying to do there) if you just want to start the camera app
May 30 22:35:40 <ech0s7> yes i know, but i need to launch a Panorama activity developed from motorola team
May 30 22:35:53 <ech0s7> that haven't a public intent
May 30 22:36:54 <g00s> ech0s7: no public intent = no public contract
May 30 22:37:21 <g00s> the best you can hope for, is that they use the same intent / extras as the stock app
May 30 22:37:31 <g00s> this is where things get murky though
May 30 22:37:33 <ech0s7> yes g00s, but this app is dedicated for this device
May 30 22:50:43 <readme> p_l: i prefer the anarchist approach :P
May 30 22:50:59 <readme> no patent office -- no problem, since no one will register their nukes with you anyway
May 30 22:51:22 <readme> it's not like the presence of a patent office can stop me from developing a nuclear weapon. i can choose to not patent it
May 30 22:51:33 <p_l> :)
May 30 22:58:03 <jayd16> hey, is there some kind of trick to get list items to show as checked in list views pre SDK11?  on my galaxy nexus, setting selector drawables with an activated version works exactly like I want it to.  How do I get that behavior on early versions?
May 30 22:58:32 <readme> when I search my app on google, the first hits are warez sites offering pirated versions.
May 30 22:59:44 <wongk-home> i guess that
May 30 22:59:48 <wongk-home> 's a good thing?
May 30 23:01:13 <readme> it would be, if they didn't outrank my actual app.
May 30 23:01:18 <readme> I
May 30 23:01:19 <bubbleguuum> lol @ search result polluted by warez links. so true
May 30 23:01:27 <readme> I don't think I'm going to bother with asking them to take it down
May 30 23:01:35 <readme> that would just put my name and email in their posession.
May 30 23:01:45 <readme> currently, I only get spam in my support email addr.
May 30 23:01:58 <bubbleguuum> wait until you get support requests
May 30 23:02:07 <readme> I've gotten around 6 that i've solved
May 30 23:02:15 <readme> 6 good ones
May 30 23:02:26 <readme> except, there were only 2 issues. 3 requests for each.
May 30 23:02:34 <wongk-home> report it to google, they'll suspend the application
May 30 23:02:40 <readme> huh?
May 30 23:02:45 <readme> I don't want my app suspended :)
May 30 23:02:50 <wongk-home> oh, this isn't through the market?
May 30 23:02:53 <readme> right
May 30 23:02:55 <wongk-home> ok
May 30 23:02:57 <readme> it's just being offered on mediafire
May 30 23:02:58 <readme> etc
May 30 23:03:05 <wongk-home> how do you know how popular it is?
May 30 23:03:22 <readme> all I know is my google play stats :)
May 30 23:03:24 <wongk-home> or you just mean in the search results
May 30 23:03:29 <readme> if I google my app
May 30 23:03:38 <readme> the first 2 results are warez sites offering it for download.
May 30 23:03:47 <wongk-home> i'd be more concerned with where you fall out in a search within google play
May 30 23:03:57 <bubbleguuum> the day a cracked version is released for your app pollutes search results for 1 week
May 30 23:04:01 <wongk-home> most people are going to search from their phone
May 30 23:04:08 <readme> bubbleguuum: yeah, its 5 days old :)
May 30 23:04:28 <bubbleguuum> congrats =)
May 30 23:04:54 <wongk-home> my problem is my app doesn't appear for like 20 pages in what i speculate would be a common search by users
May 30 23:05:04 <readme> mine is on the 5th or so page
May 30 23:05:12 <readme> send me your app, i'll review it and bump it up
May 30 23:05:19 <wongk-home> it's not free
May 30 23:05:19 <readme> i jumpstarted mine by having friends install & review
May 30 23:05:29 <readme> thats fine as long as it's not like 5 bucks
May 30 23:05:44 <readme> 5 bucks and we're talking big dollars man
May 30 23:05:58 <wongk-home> https://play.google.com/store/apps/details?id=com.kwproductions.checkouttrackerv2
May 30 23:06:10 <wongk-home> thanks bunches =]
May 30 23:06:17 <readme> nice!
May 30 23:06:19 <wongk-home> i think i have your app installed, what is it called again
May 30 23:06:19 <readme> I wanted one of these
May 30 23:06:24 <readme> LED caller id
May 30 23:06:33 <wongk-home> no, then im thinking of something else
May 30 23:06:52 <wongk-home> ahh, i think it's pragmas app i have installed
May 30 23:07:17 <wongk-home> who'da thought he'd be good for anything? ;)
May 30 23:07:33 <readme> feel free to return the favor http://tinyurl.com/bqrsxga
May 30 23:07:44 <g00s> wongk-home: haha
May 30 23:07:47 <readme> also my app is useful and you may like it.
May 30 23:07:53 <wongk-home> readme: i was just searching the market =]
May 30 23:07:56 <readme> yours is EXACTLY what I wanted
May 30 23:08:03 <readme> got to watch the sales
May 30 23:08:31 <wongk-home> fwiw, it's the first result when i type in the actual app title :P
May 30 23:08:46 <wongk-home> glad to hear you'll actually find it useful :P
May 30 23:08:55 <wongk-home> the problem is no one can find it
May 30 23:09:59 <wongk-home> looks interesting
May 30 23:10:24 <readme> let me know if it does/doesn't work. for some phones you'll need to have wifi on.
May 30 23:10:53 <wongk-home> one of your commenters is "licking it so far" ;)
May 30 23:11:08 <readme> :P
May 30 23:13:42 <readme> had to enter merchant key & id manually, but it's working nicely
May 30 23:13:44 <readme> thanks
May 30 23:14:07 <wongk-home> rated & reviewed, ill have to wait to see how it works for me =]
May 30 23:14:42 <readme> nice
May 30 23:15:06 <readme> if it doesn't ill fix it. i offer good support and i'm friends with the guy who runs the service it uses
May 30 23:15:25 <wongk-home> i try to do the same, for my 12 customers ;)
May 30 23:15:55 <readme> what phone do you have?
May 30 23:16:06 <wongk-home> i actually ran into several issues with international users
May 30 23:16:09 <wongk-home> thunderbolt
May 30 23:16:17 <readme> and carrier?
May 30 23:16:20 <wongk-home> verizon
May 30 23:16:30 <readme> oh yeah. you'll def need wifi on to make any use of it.
May 30 23:16:39 <wongk-home> thunderbolt is LTE
May 30 23:16:51 <readme> oh sick
May 30 23:16:52 <readme> I didn't know
May 30 23:17:00 <wongk-home> it was the first
May 30 23:17:02 <readme> the future is good :)
May 30 23:17:08 * readme doesn't even have LTE
May 30 23:19:18 <wongk-home> readme: what would you search for when looking for an app like that in the play store?
May 30 23:19:56 <readme> probably google checkout
May 30 23:20:04 <wongk-home> yeah, exactly
May 30 23:20:17 <wongk-home> a ton of crap is picked up with that search
May 30 23:20:19 <readme> what category did you list it under
May 30 23:20:30 <wongk-home> productivity i think
May 30 23:21:38 <readme> tracker is the word that's killing it, I think
May 30 23:21:42 <readme> tons of stuff is called tracker
May 30 23:21:58 <wongk-home> at first i called it "Google Checkout Tracker" and it got suspended
May 30 23:22:02 <wongk-home> made no sense
May 30 23:22:16 <readme> #15 for google checkout tracker
May 30 23:22:19 <wongk-home> there are several apps in the market using the google checkout logo and the google name
May 30 23:22:28 <readme> dumb luck I guess
May 30 23:22:44 <g00s> apple has an app that does this, not sure why google doesn't
May 30 23:22:48 <wongk-home> at least i didn't use the logo and made it clear it wasn't from google
May 30 23:22:55 <readme> you're high up for 'checkout tracker'
May 30 23:22:58 <readme> you're on the first page
May 30 23:23:08 <wongk-home> yeah but no one searches for that :P
May 30 23:23:12 <readme> #7
May 30 23:23:15 <readme> what do people search
May 30 23:23:20 <readme> "google checkout"?
May 30 23:23:31 <wongk-home> g00s: becuase they don't have a standard API for this that works for all checkout users
May 30 23:23:42 <wongk-home> readme: i can't say for sure, but i'd guess so
May 30 23:24:01 <wongk-home> g00s: my app has to interact with 2 different APIs
May 30 23:24:10 <wongk-home> with 2 different login methods
May 30 23:24:37 <wongk-home> and they operate completely differently
May 30 23:26:32 <wongk-home> i'
May 30 23:26:48 <wongk-home> i'd really like to add tracking of installs, ratings, and comments
May 30 23:27:17 <wongk-home> but that is proprietary to the play app, uses protobuffs, and has to be completely reverse engineered
May 30 23:27:40 <readme> would be nice if google let you see your own app on your own device in the market
May 30 23:27:52 <readme> I don't see why they have to hide it. it's not like I can do anything that sinister.
May 30 23:28:24 <wongk-home> what do you mean?
May 30 23:29:06 <readme> if I search my app on my own device, it doesn't show up in play
May 30 23:29:31 <wongk-home> weird, mine do
May 30 23:29:37 <wongk-home> always have
May 30 23:29:38 <readme> actually I think it's the CM9 issue i'm having
May 30 23:29:44 <readme> for some reason, even though my app works on CM9
May 30 23:29:50 <wongk-home> i just can't buy them
May 30 23:29:53 <readme> it shows as incompatible on some phones with CM9 installed
May 30 23:30:01 <readme> so it's probably not showing up for that reason
May 30 23:30:02 <wongk-home> yeah, that's probably it
May 30 23:30:11 <readme> i had one user want it on his cm9 phone
May 30 23:30:14 <readme> so I sent him an apk directly
May 30 23:30:20 <readme> he bought it twice somehow
May 30 23:30:27 <readme> i am not sure how, but he bought two copies anyway.
May 30 23:31:05 <g00s> hmm, this looks kinda cool http://www.ibm.com/developerworks/java/library/j-javaactors/index.html?ca=drs-
May 30 23:31:08 <wongk-home> not sure how he pulled that off
May 30 23:32:51 <wongk-home> i got started working on supporting refunds, but i've been so busy interviewing for the past month
May 30 23:33:06 <wongk-home> and trying to do some probono work for a locak charity
May 30 23:34:11 <readme> ah yeah, that would be a cool feature
May 30 23:34:18 <readme> fortunately I haven't had any one request a refund yet
May 30 23:34:23 <readme> one that wasn't automated, anyway
May 30 23:34:35 <readme> i should have some more contract working starting tomorrow
May 30 23:34:42 <readme> it was supposed to be today but my client is lazy
May 30 23:34:44 <wongk-home> i haven't either, but i offered them to current customer after it got suspended
May 30 23:35:05 <wongk-home> since you have to relist after that
May 30 23:36:42 <wongk-home> happy selling, time to cook some dinner
May 30 23:39:12 <readme> later
May 31 00:04:30 <sirellyn> How long does it take for google to authenticate you for google play?
May 31 00:04:39 <sirellyn> I signed up and paid here.  https://play.google.com/apps/publish/signup
May 31 00:04:59 <sirellyn> And day 2 I've heard nothing new.  Going to that page just brings me the blank form again.
May 31 00:08:16 <Leeds> I think it was at least a couple of days for me, but it was a long long time ago
May 31 00:08:48 <sirellyn> ok at least I'm not imagining things.
May 31 00:09:03 <sirellyn> I'm just used to.  You sign up on this website and viola!  You can now use their services.
May 31 00:09:23 <sirellyn> or at least a notice somewhere saying.  Please wait a few days while your account is being processed. :-/
May 31 00:09:28 <luxurymode> im calling context.startService(intent) from a broadcastReceiver and i'm never seeing my service's onStartCommand get called. any ideas?
May 31 00:09:54 <readme> service is not declared in manifest
May 31 00:10:01 <readme> or is declared incorrectly
May 31 00:10:51 <luxurymode> its in the manifest, have a bunch of other intentfilters on it and it works just fine
May 31 00:11:17 <readme> show the intent code and the manifest lines I guess
May 31 00:11:35 <readme> you're certain the broadcastreceiver is being called?
May 31 00:15:08 <luxurymode> readme, yeah it us
May 31 00:15:10 <luxurymode> logged that
May 31 00:15:12 <luxurymode> its getting called
May 31 00:15:14 <luxurymode> https://gist.github.com/2839749
May 31 00:15:16 <luxurymode> sorry for the formattng
May 31 00:15:25 <luxurymode> had to strip some stuff out for NDA purposes bla bla
May 31 00:16:06 <luxurymode> crap didnt delete enough had to delete that lol
May 31 00:16:09 <luxurymode> but you see it?
May 31 00:17:03 <ctate> NEVER seeing the service started, even after you wake up the phone or whatever?  nothign in the log about the startService attempt?
May 31 00:17:30 <luxurymode> nope
May 31 00:17:38 <luxurymode> im gonna try using the class with setAction
May 31 00:19:07 <luxurymode> still no luck
May 31 00:19:34 <readme> luxurymode: sorry, was reading useless reddit links
May 31 00:20:03 <readme> The Gist you were looking for has been deleted. Sorry about that!
May 31 00:20:57 <luxurymode> yeah had to remove it
May 31 00:20:58 <luxurymode> sec...
May 31 00:25:10 <luxurymode> https://gist.github.com/2839793
May 31 00:25:15 <luxurymode> buncha stuff stripped out but yea
May 31 00:25:29 <luxurymode> same way ive started the service a million times...
May 31 00:31:38 <luxurymode> any thoughts?
May 31 00:31:48 <readme> i'm thinking maybe try android:exported="true"
May 31 00:31:54 <readme> but I'm looking it up in the docs
May 31 00:32:18 <readme> I figure since you're presumably inside the same app
May 31 00:32:20 <readme> you don't need it
May 31 00:32:34 <readme> but it might be worth a try anyway
May 31 00:32:39 <ctate> with an intent filter the default is android:exported="true"
May 31 00:32:41 <ctate> so hm.
May 31 00:32:52 <ctate> does that exact code work elsewhere in your app?
May 31 00:33:17 <ctate> again, i'm curious whether you see any logs about the service launching [or error messages] when the receiver runs
May 31 00:34:56 <b1n0ry> why don't they take you to the vet, you're obviously not their favorite pet... smelly cat.
May 31 00:35:14 * b1n0ry bows. thank you, i'll be here all night.
May 31 00:35:32 <etcetera_> any good tutorials on android's new gpu accel animation stuff?
May 31 00:35:47 <etcetera_> more real-world implementation focused, not this is how it works.
May 31 00:36:09 <b1n0ry> could always try to find an open source app that uses it...
May 31 00:36:18 <etcetera_> I'd settle for sample code, yes.
May 31 00:36:39 <etcetera_> I mean if the standard bearer is uiview animation...I'd be happy to look at 10 lines of sample code :-P
May 31 00:37:30 <ctate> there's always the ICS source code....
May 31 00:38:10 <etcetera_> ctate: you mean they would release new apis without documenting them?
May 31 00:38:30 <etcetera_> shocked and appalled, I say. SHOCKED!
May 31 00:40:00 <ctate> there's documentation and then there's example code and then there are tutorials
May 31 00:40:23 <lov> Dear ctate: Please send me teh codes.
May 31 00:40:26 <lov> Sincerely, lazy internets.
May 31 00:40:42 <lov> Addendum: Get off my lawn.
May 31 00:41:36 <etcetera_> lov: docs don't exist if they aren't easy to find.
May 31 00:43:05 <etcetera_> in fact, this is a pretty substantial feature, with very little google juice. I would be paying people to talk about it.
May 31 00:43:05 <ctate> well, there are http://android-developers.blogspot.com/2011/02/animation-in-honeycomb.html and http://android-developers.blogspot.com/2011/05/introducing-viewpropertyanimator.html and http://android-developers.blogspot.com/2011/11/android-40-graphics-and-animations.html
May 31 00:43:37 <ctate> which also have links to e.g. youtube video of conference presentations
May 31 00:45:41 <etcetera_> ctate: so the api's haven't changed, they are just get hardware accelerated now?
May 31 00:45:49 <ctate> that is my impression
May 31 00:46:07 <etcetera_> that's just bad marketing.
May 31 00:46:24 <ctate> note that ViewPropertyAnimator was new in 3.0
May 31 00:46:30 <etcetera_> hey we released the api before it's ready, now that it's ready we're talking about the cerebral bits
May 31 00:46:33 <ctate> so on *phones* it's new in ICS
May 31 00:47:07 <JakeWharton> ViewPropertyAnimator is 3.2+
May 31 00:47:20 <ctate> okay 3.2 :)
May 31 00:47:34 <ctate> nowait, it was 3.1
May 31 00:47:40 <canadiancow> 3.2 was like "o hai honeycomb isnt total suck now"
May 31 00:48:05 <JakeWharton> etcetera_: there's ample documentation, samples, and presentations on the subject
May 31 00:48:54 <etcetera_> JakeWharton: I'm just saying, it's not terribly easy to find.
May 31 00:49:02 <JakeWharton> http://d.android.com
May 31 00:49:07 <JakeWharton> found it
May 31 00:49:19 <canadiancow> +1
May 31 00:49:36 <JakeWharton> there are api demos, other samples, blog posts, and the javadocs
May 31 00:49:42 <canadiancow> my maguro docks are in mississauga at fedex :)
May 31 00:49:51 <etcetera_> google juice -> "ios animation" vs "android animation"
May 31 00:51:14 <canadiancow> [20:46:31] <etcetera_> hey we released the api before it's ready, now that it's ready we're talking about the cerebral bits
May 31 00:51:17 <canadiancow> wat are you smoking
May 31 00:51:19 <canadiancow> i want some
May 31 00:51:27 <JakeWharton> *google guice
May 31 00:51:30 <JakeWharton> problem solved
May 31 00:51:31 <canadiancow> they made everything work
May 31 00:51:39 <canadiancow> and then they made it work faster
May 31 00:51:46 <canadiancow> and you're complaining it wasnt that fast from the start?
May 31 00:51:57 <JakeWharton> bind(Intelligence.class).to(Brain.class);
May 31 00:52:01 <canadiancow> if i had that mindset, i would have killed myself after seeing new hardware come out every year
May 31 00:52:14 <canadiancow> those are some strange method names JakeWharton
May 31 00:52:22 <JakeWharton> you've never used Google Guice?
May 31 00:52:31 <canadiancow> no?
May 31 00:52:35 <JakeWharton> watch the I/O talk from Jesse Wilson
May 31 00:52:39 <JakeWharton> you'll never not use it again
May 31 00:53:22 <etcetera_> like most things ios vs android, it's differing philosophies. I'm just saying, it's not terribly easy to find.
May 31 00:53:41 <canadiancow> really?
May 31 00:53:44 <ctate> i happen to agree that the docs have room for improvement
May 31 00:53:58 <ctate> both in content and in ease of discovery
May 31 00:54:06 <canadiancow> http://developer.android.com/guide/topics/graphics/animation.html
May 31 00:54:10 <canadiancow> that is not hard to find
May 31 00:54:29 <ctate> that said, it's definitely worth your while to follow the android-developers blog :)
May 31 00:54:38 <canadiancow> oh i follow that shit
May 31 00:54:43 <canadiancow> i read it every day
May 31 00:54:47 <canadiancow> even if there's nothing new
May 31 00:54:52 <etcetera_> http://developer.android.com/reference/android/animation/package-summary.html
May 31 00:54:56 <etcetera_> http://developer.apple.com/library/ios/#documentation/windowsviews/conceptual/viewpg_iphoneos/animatingviews/animatingviews.html
May 31 00:55:23 <canadiancow> http://www.apple.com http://developer.android.com/guide/topics/graphics/prop-animation.html
May 31 00:55:30 <canadiancow> look i can show urls that are useless and useful the other way around too
May 31 00:55:38 <JakeWharton> http://lmgtfy.com
May 31 00:56:09 <readme> http://www.youtube.com/watch?v=l2-UuIEOcss
May 31 00:57:11 <birbeck> working for a mobile payments company, you would think the ceo and vp of business development wouldnt have to come ask me questions about reloading google wallet
May 31 00:57:15 <tmho> lol everytime i look here there is an ios vs android debate
May 31 00:57:34 <etcetera_> canadiancow: I'm showing the top results for "ios animation" and "android animation" respectively.
May 31 00:57:50 <etcetera_> this is a more fundamental problem.
May 31 00:58:11 <readme> tmho: in a debate there is intelligent argument and rebuttal
May 31 00:58:19 <readme> that's not what this is
May 31 00:58:28 <etcetera_> this is me complaining. and wondering why it's this way.
May 31 00:58:39 <tmho> lol fair anough
May 31 00:58:44 <tmho> enough*
May 31 00:58:48 <JakeWharton> birbeck: what company?
May 31 00:59:00 <b1n0ry> ok, anyone have an idea why my action bar/progress bar (spinner) shows up when the activity is created but i can't seem to make it disappear?
May 31 00:59:00 <pfn> ugh, how do I parse this kinda json with gson
May 31 00:59:02 <birbeck> JakeWharton: boku
May 31 00:59:10 <pfn> http://pastie.org/3998752
May 31 00:59:11 <JakeWharton> pfn: define kinda
May 31 00:59:11 <b1n0ry> like... ever disappear
May 31 00:59:17 <readme> tmho: i guess it's a debate if you compare it to any of the republican primary debates.
May 31 00:59:29 <JakeWharton> pfn: what's wrong with it?
May 31 00:59:29 <canadiancow|work> so etcetera_, you want to animate an android?
May 31 00:59:31 <pfn> the problem is the Events object
May 31 00:59:36 <pfn> the keys are indeterminate
May 31 00:59:39 <canadiancow|work> learn to use proper search queries
May 31 00:59:46 <JakeWharton> pfn: you can have custom evaluators
May 31 00:59:47 <pfn> I wonder if gson will support a Map<String,DomainObject>
May 31 00:59:55 <pfn> yeah, I need to read it and figure it out
May 31 00:59:56 <pfn> meh
May 31 01:00:17 <b1n0ry> http://pastebin.com/hifz95WV <-- here's the code. progress spinner shows up, never goes away.
May 31 01:00:36 <ctate> y'all need to back off and accept that etcetera_ is basically right here
May 31 01:00:39 <canadiancow|work> JakeWharton and birbeck: FIGHT
May 31 01:00:41 <ctate> the android docs are not that great
May 31 01:00:52 <ctate> there's a lot of info out there, which is not the same as the docs being excellent :)
May 31 01:00:53 <canadiancow|work> hey ctate, i have an idea for your 20% time
May 31 01:00:57 <birbeck> why fight?
May 31 01:01:02 <JakeWharton> i work for Square
May 31 01:01:06 <ctate> bwahahahahahaha 20% time ha ha ha ha
May 31 01:01:07 <birbeck> oh, lol
May 31 01:01:09 <ctate> snrk
May 31 01:01:19 <readme>             requestWindowFeature(Window.FEATURE_INDETERMINATE_PROGRESS);
May 31 01:01:22 <birbeck> thats why pay with square uses abs :)
May 31 01:01:27 <readme> and you're surprised, b1n0ry ?
May 31 01:01:41 <b1n0ry> readme, tutorial i followed used that
May 31 01:01:54 <pfn> ctate, the docs are great, you speak sacrilege!
May 31 01:02:01 <b1n0ry> should i just be using FEATURE_PROGRESS?
May 31 01:02:19 <ctate> pfn: bah!
May 31 01:02:45 <birbeck> JakeWharton: do you work downtown sf?
May 31 01:02:46 <b1n0ry> readme, i've not implemented one of these before. this is my first app using action bar... actually, my first app period.
May 31 01:02:48 <ctate> it does not help that they're quite a moving target, what with all the new stuff all the time
May 31 01:02:51 <etcetera_> ctate: I think that's part of it. The other part is, it's just bad developer marketing.
May 31 01:02:52 <JakeWharton> birbeck: they were using it in an unreleased version when I showed up on my first day
May 31 01:02:54 <JakeWharton> birbeck: yes
May 31 01:03:05 <birbeck> JakeWharton: then why havent i bought you a beer yet?
May 31 01:03:06 <canadiancow|work> ctate lies...there haven't been any new apis since like december!
May 31 01:03:15 <pfn> JakeWharton, how are you liking sf anyway?
May 31 01:03:22 <ctate> engineers tend not to realize that writing takes *time*
May 31 01:03:23 <ctate> :)
May 31 01:03:27 <etcetera_> "Hey look, new animation APIs you should use to make amazing apps!" sounds much better than, "Hey, we fixed animation!"
May 31 01:03:28 <readme> b1n0ry: yeah it seems odd
May 31 01:03:30 <canadiancow|work> psh
May 31 01:03:32 <readme> i think it's not supposed to show
May 31 01:03:36 <canadiancow|work> code.document()
May 31 01:03:39 <pfn> ctate, writing doesn't take time  :p
May 31 01:03:41 <etcetera_> technical writing takes * a lot * of time.
May 31 01:03:41 <pfn> you think it
May 31 01:03:43 <pfn> and you're done
May 31 01:03:44 <pfn> ;-)
May 31 01:04:23 <b1n0ry> readme, especially with the very next line: setProgressBarIndeterminateVisibility(false);
May 31 01:04:49 <b1n0ry> (and for the record, JakeWharton, i tried with Boolean.FALSE, too, per the ABS example)
May 31 01:05:19 <canadiancow|work> evancharlton you around?
May 31 01:05:25 <readme> b1n0ry: https://github.com/JakeWharton/ActionBarSherlock/blob/master/samples/demos/src/com/actionbarsherlock/sample/demos/Progress.java
May 31 01:05:29 <readme> that's how you should probably do it
May 31 01:05:32 <readme> since you are using ABS
May 31 01:05:39 <readme> and there is an example of how to do it in the ABS source tree
May 31 01:05:53 <t0mless> b1n0ry is you're using ABS you should be using setSupportProgressBarIndterminateVisibility()
May 31 01:06:14 <readme> https://github.com/JakeWharton/ActionBarSherlock/blob/master/samples/demos/src/com/actionbarsherlock/sample/demos/IndeterminateProgress.java
May 31 01:06:14 <b1n0ry> wow, that's a heck of a function name.
May 31 01:06:17 <readme> that evn
May 31 01:06:24 <b1n0ry> let me give that a shot t0mless
May 31 01:06:38 <b1n0ry> readme, it is indeterminate, though. not sure how long it will take and no real way to measure progress.
May 31 01:06:43 <readme> second one is
May 31 01:06:50 <readme> t0mless pinned it down already
May 31 01:06:52 <b1n0ry> ah, ok, didn't notice that thanks!
May 31 01:06:56 <readme> you needed to call the one with _support_ in it
May 31 01:06:59 <b1n0ry> i'll give it a shot. hadn't seen mention of that function.
May 31 01:09:40 <b1n0ry> i can't seem to find any function called setSupportProgressBarIndeterminateVisibility
May 31 01:10:31 <canadiancow|work> ?
May 31 01:10:44 <canadiancow|work> getSupporActionBar().setProgre...
May 31 01:10:47 <canadiancow|work> probably something like that
May 31 01:10:56 <t0mless> nah, its a method of SherlockActivity
May 31 01:11:21 <canadiancow|work> i know i did it...
May 31 01:11:50 <b1n0ry> i'm in a FragmentActivity, technically in the fragment itself. getActivity().setSupport... doesn't resolve.
May 31 01:11:51 <t0mless> it's prolly a method on the AB too
May 31 01:11:58 <t0mless> b1
May 31 01:12:08 <amoxibos> hello guys
May 31 01:12:11 <amoxibos> i have an activity
May 31 01:12:13 <t0mless> b1n0ry you need to to getSherlockActivity or cast your activity to a SherlockActivity
May 31 01:12:17 <t0mless> or SherlockFragmentActivity
May 31 01:12:31 <t0mless> cast your getActivity() call that is
May 31 01:12:34 <b1n0ry> ah! i didn't realize that was a library
May 31 01:12:40 <b1n0ry> er class, er whatever
May 31 01:12:41 <t0mless> wut?
May 31 01:12:44 <b1n0ry> nm.
May 31 01:12:50 <JakeWharton> there's probably a getSherlockActivity()
May 31 01:12:53 <b1n0ry> i didn't realize there was a getSherlockActivity()...
May 31 01:13:04 <t0mless> It makes it easier than casting
May 31 01:13:21 <amoxibos> i have an activity which has to have a lot of stuff on it, and it cannot fit on a single screen, so i need to make it possible for the user to scroll downards to see all the options on this activity - how do i go about this?
May 31 01:13:39 <t0mless> ScrollView
May 31 01:14:09 <amoxibos> thanks man
May 31 01:15:24 <b1n0ry> that would have helped if i extended SherlockListFragment instead of just ListFragment
May 31 01:15:38 <b1n0ry> resolved it, thanks guys! still have to test it...
May 31 01:17:41 <amoxibos> i want to have a label next to a radio button - do i make two listviews next to each other or?
May 31 01:17:56 <b1n0ry> getSherlockActivity().setSupportProgressBarIndeterminateVisibility(false); <-- bada bing. you guys are geniuses.
May 31 01:19:25 <giantpune> hi, i am looking to buy my first cell phone since 1999.  am i right in understanding that if i get one with the android OS, that i can use the android sdk and make my own apps for the phone?
May 31 01:19:27 <b1n0ry> amoxibos, you make one listview and use a list item layout
May 31 01:19:45 <pfn> giantpune, yes
May 31 01:19:54 <b1n0ry> giantpune, do you know java? or are you willing to learn? if so, then you're in luck!
May 31 01:20:12 <giantpune> i know c and c++ pretty well.
May 31 01:20:20 <amoxibos> aha aha
May 31 01:20:22 <amoxibos> thanks man
May 31 01:20:27 <giantpune> im sure i can pick up java if thats all that is supported
May 31 01:20:55 <t0mless> you can use c and c++ via the ndk
May 31 01:21:00 <t0mless> but not everything is available to the ndk
May 31 01:21:01 <b1n0ry> the jump from c++ to java isn't insane. i'm struggling a bit as i'm learning as i've been living in C#/JavaScript/RAD land for awhile now.
May 31 01:21:11 <ctate> no, it's not that big a leap
May 31 01:21:21 <pfn> the jump from c# to java should feel quite natural
May 31 01:21:25 <pfn> at least the other way around certainly does
May 31 01:21:27 <ctate> mostly you have to get used to actually having some decent container classes by default :)
May 31 01:21:29 <t0mless> indeed
May 31 01:21:35 <b1n0ry> classes, methods, instantiation. OOP to OOP is exactly that. the rest is just syntax.
May 31 01:21:47 <b1n0ry> pfn: mostly, my problem is just learning the nuances of android
May 31 01:21:59 <ctate> oh, and no operator overloading, so you need to remember not to use == to compare strings :)
May 31 01:21:59 <b1n0ry> the java itself isn't hurting too badly ;) just the normal learning curve
May 31 01:22:07 <giantpune> how are you guys debugging your code?  do you use an emulator on your pc or do you connect your phone to your computer and sent a bunch of printf() or what?
May 31 01:22:19 <b1n0ry> stringA.equals(stringB)
May 31 01:22:24 <ctate> i use the actual debugger, eclipse + ddms
May 31 01:22:35 <ctate> which works with either the emulator or with a phone, thanks to the magic of ddms
May 31 01:22:50 <b1n0ry> giantpune, at any given point, i have a phone and a tablet hooked to the PC and en emulator running. all different OS versions.
May 31 01:22:53 <ctate> (the ddms stuff is built into the eclipse andorid integration package)
May 31 01:23:52 <giantpune> and will i be able to build a hello world and run it on a new phone rightout of the box?  or do i have to root/jailbreak/whatever the phone before i can run my own code?
May 31 01:23:59 <lov> yes, you can do it out of the box.
May 31 01:23:59 <b1n0ry> no, out of box
May 31 01:23:59 <canadiancow|work> no
May 31 01:24:01 <canadiancow|work> this isnt iphone
May 31 01:24:10 <lov> giantpune: http://developer.android.com/guide/developing/device.html
May 31 01:24:15 <canadiancow|work> hi lov
May 31 01:24:18 <b1n0ry> there's a setting right in the settings to say "use my phone for development" - check the box and you're good to go.
May 31 01:24:27 <lov> canadiancow|work: good timing, I'm going afk :>
May 31 01:24:36 <canadiancow|work> </3
May 31 01:24:44 <canadiancow|work> "i less than scrotum you too" yea yea
May 31 01:24:45 <b1n0ry> basically, it lets you install "unsigned" apps more than anything and sends debug information (when called properly) back to your development environment
May 31 01:24:50 <canadiancow|work> err slash scrotum
May 31 01:24:52 <canadiancow|work> or whatever you say
May 31 01:25:10 <b1n0ry> slash scrotum... ouch
May 31 01:25:26 <b1n0ry> reminds me of a particuarly nasty chapter of Stephen King's "The Dark Half"
May 31 01:26:19 <giantpune> awesome.  i guess thats all i needed to know for now.  now i gotta decide which phone to buy and actually buy it
May 31 01:26:24 <giantpune> thanks for the info
May 31 01:26:32 <b1n0ry> giantpune, i got my tablet, took off the screen protector, plugged in a usb cable, and installed my app...
May 31 01:26:38 <b1n0ry> simple as that
May 31 01:26:54 <b1n0ry> good luck!
May 31 01:27:09 <b1n0ry> the learning curve is addicting, i haven't seen my wife in at least 3 weeks...
May 31 01:27:15 <giantpune> that will be a pleasant change. im used to wii/ps3 where you have to jump through so many hoops to run your own code
May 31 01:27:25 <ctate> another week and you won't see her...EVER.  :)
May 31 01:27:34 <JakeWharton> giantpune: I strongly, strongly recommend a phone with Android 4.0 Ice Cream Sandwich
May 31 01:27:41 <b1n0ry> oh yeah, i've built flash apps for the wii and never did get very far with actually trying them on the console
May 31 01:27:43 <ctate> <3 Galaxy Nexus :)
May 31 01:27:49 <JakeWharton> You can buy an unlocked Galaxy Nexus from the Play Store
May 31 01:27:58 <Xonk> Does anyone know of someone/somewhere cheap to get about a dozen icons drawn up?
May 31 01:27:59 <b1n0ry> ctate, that's because i'll be in Las Vegas next week
May 31 01:28:25 <giantpune> JakeWharton, im only looking at phones i can use with a prepaid deal
May 31 01:28:42 <b1n0ry> Xonk, put out a bid on deviantart
May 31 01:29:07 <b1n0ry> Xonk, or on freelancer.com (and make sure to ask for a sample portfolio)
May 31 01:29:53 <b1n0ry> i think most of the major carriers have prepaid plans now
May 31 01:29:57 <Xonk> Huh.  I didn't know diviant art had a market system.
May 31 01:30:09 <Xonk> b1n0ry - I'm talking about dirt cheap.  Like $5-10
May 31 01:30:23 <tmho> where can you buy the nexus from the play store?
May 31 01:30:32 <Xonk> For just simple icons like search and thumbs up.
May 31 01:30:36 <b1n0ry> they don't technically, as far as i know. but i know a lot of artists who used to do freelance work. someone would see their work on deviantart and ask what they'd be willing to take to do some work...
May 31 01:30:47 <b1n0ry> Xonk, just grab the ones out of android.jar ;)
May 31 01:31:00 <JakeWharton> tmho: there's a dropdown
May 31 01:31:02 <JakeWharton> select devices
May 31 01:31:07 <Xonk> Can you tell me how b1n0ry?
May 31 01:31:15 <b1n0ry> or go and search for an opensource/apache/MIT license icon library for linux. you'll get most of them in PNG format.
May 31 01:31:22 <JakeWharton> https://play.google.com/store/devices
May 31 01:31:31 <readme> b1n0ry: creative commons, maybe
May 31 01:31:35 <b1n0ry> yeah
May 31 01:31:43 <tmho> "We're sorry, the document you requested is not available in your country." - sigh
May 31 01:32:06 <Xonk> yeah, that's a good point.  As long as i can find some that don't require citation.
May 31 01:32:13 <b1n0ry> Xonk, just go to the android.jar file for the android sdk (on your PC) and open it up. android.jar is just a .zip file with a different extension. 7-zip or Winzip will be able to open it up just fine.
May 31 01:32:36 <JakeWharton> you don't need to unzip android.jar for icon assets
May 31 01:32:43 <JakeWharton> they are provided separately in the SDK download
May 31 01:33:02 <b1n0ry> true, and most of them you don't have to even do that.. just reference android.R.drawable.<whatever>
May 31 01:33:09 <readme> wait so, how do i unzip again winzip?
May 31 01:33:20 <readme> wat
May 31 01:33:21 <b1n0ry> although there you risk some inconsistency by OEM
May 31 01:34:12 <b1n0ry> found out my droidx doesn't have an "unselected" ic_search icon
May 31 01:34:49 <b1n0ry> so when the tab is selected, it's dark on dark... nice
May 31 01:35:01 <b1n0ry> gotta love Verizon's custom theme for 2.3.3
May 31 01:38:02 <amoxibos> can i have scrollview in alertbox>/
May 31 01:39:07 <JakeWharton> no, but you can make a custom dialog with a ScrollView as its content view
May 31 01:39:22 <JakeWharton> i'm sure googling 'android scrollview dialog' will give you plenty of examples
May 31 01:40:37 <b1n0ry> btw, if anyone is in vegas and interested, i'll be performing at Dal Toro and LAVO (the Palazzo), Gilly's Salloon (Treasure Island), and Mandalay Bay... all next week, then I'm flying out.
May 31 01:41:04 <pfn> ah, gson supports deserializing into a Map<String,DomainObject>
May 31 01:41:07 <b1n0ry> always gotta plug when there could be folks around interested
May 31 01:41:07 <pfn> excellent
May 31 01:41:26 <pfn> performing what?
May 31 01:41:30 <b1n0ry> magic
May 31 01:41:31 <pfn> a chipndale show?
May 31 01:41:47 <pfn> I see
May 31 01:41:57 <b1n0ry> <-- magician
May 31 01:42:01 <readme> oh man, chip and dale are awesome
May 31 01:42:07 <amoxibos> can some one point me to some short tutorial on shared preferences with code example
May 31 01:42:08 <amoxibos> ?
May 31 01:42:19 <readme> http://upload.wikimedia.org/wikipedia/en/thumb/3/38/ChipDaleLogo.jpg/250px-ChipDaleLogo.jpg
May 31 01:42:46 <b1n0ry> amoxibos, http://saigeethamn.blogspot.com/2009/10/shared-preferences-android-developer.html
May 31 01:42:56 <readme> amoxibos: we're too busy denigrating the conversation... oh wait, b1n0ry to the rescue
May 31 01:43:11 <b1n0ry> it's what i'm here for
May 31 01:43:16 <b1n0ry> denigrating conversation
May 31 01:43:18 <amoxibos> b1n0ry, thanks man
May 31 01:43:32 <b1n0ry> yw
May 31 01:43:34 <amoxibos> http://www.thefreedictionary.com/denigrate
May 31 01:43:45 <readme> amoxibos: stop denigratin me
May 31 01:43:54 <amoxibos> 1. To attack the character or reputation of; speak ill of; defame. 2. To disparage; belittle: The critics have denigrated our efforts.
May 31 01:43:59 <readme> i'll denigrate your face
May 31 01:44:00 <b1n0ry> it was more fun when i didn't know the definition and just trusted readme's interpretation
May 31 01:44:01 <tmho> amoxibs: http://www.vogella.com/articles/AndroidFileBasedPersistence/article.html
May 31 01:44:31 <readme> i looked it up after I used it
May 31 01:44:35 <readme> I was gonna try to back pedal
May 31 01:44:46 <readme> but amoxibos had to denigrate me
May 31 01:45:04 <b1n0ry> thus, i denigrate thee
May 31 01:45:04 <amoxibos> what i love about this is that not only do we learn andro dev but we learn new words and stuff
May 31 01:45:30 <b1n0ry> i think that's in the bible somewhere
May 31 01:45:53 <readme> andro dev? like an androgynous device?
May 31 01:46:14 <b1n0ry> a device that you can't really tell if it's a computer or a phone
May 31 01:46:19 <readme> or are you saying you ahve androgynous developments
May 31 01:46:38 <b1n0ry> maybe he developed androgynously
May 31 01:46:51 <readme> I did that today
May 31 01:46:59 <readme> in java and javascript simultaneously
May 31 01:47:23 <b1n0ry> i didn't realize you worked in hp service manager!
May 31 01:47:46 <b1n0ry> three languages in one platform: java, javascript, and their proprietary RAD
May 31 01:48:34 <b1n0ry> talk about a headache. i avoid any of the java manipulation in there like the plague. stick with javascript. rad only when needed.
May 31 01:49:25 <readme>  this work I'm doing is going to eliminate the need to write javascript to use a thing in a thing, with a thing.
May 31 01:56:36 <b1n0ry> readme, care to be a little more vague, please?
May 31 01:57:32 <t0mless> Abstract that shit.
May 31 02:02:58 <b1n0ry> alright, i'm checking out Sherlock on Netflix... so far, it's a pretty good show.
May 31 02:03:09 <b1n0ry> didnt' think a remake/modernization of sherlock holmes would be any good
May 31 02:05:11 <andreluiz> is there a way to use an photo contact uri to display the photo on an imageview?
May 31 02:05:20 <readme> I didn't like it. Too vague and thin plot. Lots of imagery.
May 31 02:05:34 <readme> I could only watch for like 10 minutes before I had to shut it off
May 31 02:05:55 <readme> Luther, is pretty good
May 31 02:06:03 <b1n0ry> i'm 15 minutes in like it so far
May 31 02:06:12 <b1n0ry> but i tend to be a shallow, thin and vague person myself.
May 31 02:06:26 <b1n0ry> i spent monday evening watching two headed shark attack
May 31 02:07:04 <b1n0ry> hot girls + unbelievable monsters + lots of blood/screaming = typical sci-fi original movie
May 31 02:08:55 <b1n0ry> if it weren't for sci-fi original movies, i wouldn't have cable.
May 31 02:09:53 <p_l> now make those girls intelligent, and responsible for a lot of enemy blood being spilled, and you got yourself a cult following
May 31 02:10:04 <JakeWharton> b1n0ry: it's good
May 31 02:11:08 <JakeWharton> Sherlock, that is
May 31 02:12:00 <b1n0ry> JakeWharton, i assumed you didn't mean two headed shark attack :)
May 31 02:13:14 <b1n0ry> nobody calls that rubbish "good". only cheesily awesome.
May 31 02:18:52 <appamatto> Hmm, the in-app billing sample seems incredibly complicated
May 31 02:19:53 <appamatto> Is there some sort of library for in-app purchase?
May 31 02:20:12 <JakeWharton> yes
May 31 02:20:17 <JakeWharton> you can download it via the SDK manager
May 31 02:20:33 <appamatto> hmm... I see that the sample code is incredibly long
May 31 02:20:42 <appamatto> I thought there might be a library with a simpler API
May 31 02:20:49 <JakeWharton> write a wrapper specific to your use case
May 31 02:21:50 <appamatto> https://github.com/robotmedia/AndroidBillingLibrary looks promising
May 31 02:25:32 <andreluiz> could anyone help me? i'm trying to fetch a contact photo, i see the name of the column is "photo_id" but when i try to even get the type cursor.getType(2) it crashes
May 31 02:25:36 <andreluiz> and yes, 2 is the index
May 31 02:25:53 <readme> why are you hardcoding the column index
May 31 02:26:22 <readme> also I relaly hope you're not using the magic string "photo_id" anywhere
May 31 02:26:32 <readme> really*
May 31 02:27:04 <andreluiz> ahm, i dont quite get you?
May 31 02:27:11 <andreluiz> i'm trying several ways o fetch that photo hehe
May 31 02:27:23 <andreluiz> i'm sure i'm doing it the possible worst ways
May 31 02:27:45 <readme> http://developer.android.com/reference/android/database/Cursor.html#getColumnIndex%28java.lang.String%29
May 31 02:27:57 <readme> read docs, not blogspam
May 31 02:28:55 <andreluiz> i'm using getColumnNames() inside a for
May 31 02:29:07 <andreluiz> and it shows data1, data2 and photo_id
May 31 02:29:10 <g00s> yuck, come on :)
May 31 02:29:19 <andreluiz> then getColumnIndex
May 31 02:29:32 <andreluiz> =\ i'm a rookie hehe
May 31 02:29:55 <g00s> get your column indexes outside your loop
May 31 02:29:56 <readme> "data1" and "data2" etc are generic names that you shouldn't be caring about
May 31 02:30:15 <readme> use the constants that the Contacts provider gives you
May 31 02:30:41 <g00s> uh oh, the contacts content provider ...
May 31 02:30:43 <readme> next up, try showing us some code and a stack trace.
May 31 02:30:45 * g00s makes his exit
May 31 02:30:47 <readme> g00s: i've mastered it.
May 31 02:30:48 <andreluiz> g00s i'm watching for those indexes only for 'debug' hehe
May 31 02:30:52 <readme> I know exactly how it works
May 31 02:31:07 <readme> at least, from the perspective of an API user.
May 31 02:31:09 <g00s> readme: which one, there are like 3 right ?
May 31 02:31:47 <readme> I prefer the generic one
May 31 02:31:51 <g00s> if you want to see some bad ass code, head over to the contacts content provider :)
May 31 02:31:57 <g00s> bad, as in terrible as hell
May 31 02:31:59 <readme> yes
May 31 02:32:05 <readme> it's pretty monstrous
May 31 02:32:11 <readme> and to use the generic one
May 31 02:32:19 <readme> you have to compare strings just to get your selection
May 31 02:32:43 <readme> because, the entity type is determined by the mimetype col
May 31 02:32:59 <readme> multiple entity types in same "table" basically
May 31 02:33:07 <readme> icky
May 31 02:33:53 <readme> but the generic one is the only way you can get all the data you want in one query.
May 31 02:34:43 <andreluiz> readme
May 31 02:34:43 <andreluiz> https://gist.github.com/2840565
May 31 02:35:16 <andreluiz> there's not an traceback.. but that InputStream is null all the time
May 31 02:37:32 <readme> andreluiz: have you considered that maybe it is null when the contact doesn't have a photo
May 31 02:37:40 <amoxibos> i have radio buttons group - i want to use shared preferences to save which radio button user chose - shared Preferences, Preference category - how do i go about this?
May 31 02:37:49 <andreluiz> readme but the contact has a photo
May 31 02:37:54 <andreluiz> i'm testing on my device
May 31 02:38:33 <readme> http://stackoverflow.com/questions/2383580/android-how-do-i-load-a-contact-photo
May 31 02:39:30 <andreluiz> ahm, i've been to that page
May 31 02:39:34 <andreluiz> but i'll try it again then
May 31 02:39:57 <readme> if you couldn't help yourself after visiting that page, I don't know why you think any of us can :)
May 31 02:40:23 <readme> I'm leaning towards photo is actually null.
May 31 02:40:26 <readme> let me know what you find out.
May 31 02:41:01 <andreluiz> ok
May 31 02:48:44 <andreluiz> readme i was able to use the photo_id
May 31 02:48:52 <andreluiz> thanks for your help and i'm sorry to bother =P
May 31 02:49:19 <readme> :D
May 31 02:50:14 <b1n0ry> "i'm not a psychopath. i'm a high-functioning sociopath. do your research." wow.
May 31 02:50:18 <andreluiz> did you read that on pvt?
May 31 02:53:17 <linuxgecko> what would it take for me to make my phone expose it's usb storage the moment it detects it's connected by usb, instead of waiting for my per-instance authorization?
**** ENDING LOGGING AT Thu May 31 02:59:58 2012