**** BEGIN LOGGING AT Thu Feb 12 02:59:58 2009
Feb 12 04:00:17 <zhobbs_> anyone else think Toast.LENGTH_LONG is way too short?
Feb 12 04:00:31 <zhobbs_> 3.5 seconds
Feb 12 04:01:57 <zhobbs_> and there's no way to customize it....
Feb 12 04:02:42 <zhobbs_> I really wanted to write a class called BurntToast too...
Feb 12 04:03:26 <DJTachyon> lol
Feb 12 04:04:37 <DJTachyon> yeah i like defining ways to exit an application
Feb 12 04:17:49 <DJTachyon> how can you define that the entire application activity you are running is a service?
Feb 12 04:18:21 <DJTachyon> this.startService(this.getIntent()); in the onCreate?
Feb 12 04:21:17 <zhobbs_> you extend Service
Feb 12 04:21:26 <zhobbs_> not Activity
Feb 12 04:23:40 <DJTachyon> does that mean if I have a UI, I can't extend Service for the main activity?
Feb 12 04:24:05 <DJTachyon> do i have to create a sub-service for holding the parent activity as a service?
Feb 12 04:24:37 <DJTachyon> i just don't want the use to be able to hit home and close the app and prevent the music from stopping
Feb 12 04:26:35 <zhobbs_> Activities and Services are completely seperate
Feb 12 04:26:47 <zhobbs_> they can communicate with each other via AIDL
Feb 12 04:26:53 <DJTachyon> well what about the built in music app
Feb 12 04:26:57 <DJTachyon> that plays music in a service
Feb 12 04:26:59 <zhobbs_> yeah
Feb 12 04:27:07 <DJTachyon> and then you can go back and see the progress and everything
Feb 12 04:27:13 <zhobbs_> and the activity tells the service what to play via aidl
Feb 12 04:27:26 <zhobbs_> and the activity gets all the playback status info via aidl
Feb 12 04:27:50 <DJTachyon> wow thats gonna take some work
Feb 12 04:28:05 <zhobbs_> :)
Feb 12 04:28:45 <zhobbs_> it's kinda old, but may help:
Feb 12 04:28:47 <zhobbs_> http://www.helloandroid.com/node/140
Feb 12 04:28:53 <DJTachyon> have to look for existing services matching my spawned service
Feb 12 04:29:01 <DJTachyon> and whatnot
Feb 12 04:29:15 <DJTachyon> heh M3
Feb 12 04:29:20 <zhobbs_> you won't have 2 instances of a Service running
Feb 12 04:29:38 <DJTachyon> right
Feb 12 04:29:39 <zhobbs_> well, I don't know how to create 2 instances anyway...
Feb 12 04:29:53 <DJTachyon> what if two apps spawn the same service name?
Feb 12 04:30:04 <zhobbs_> well, you assume different package names
Feb 12 04:30:09 <zhobbs_> but, don't really know
Feb 12 04:30:11 <DJTachyon> hmm okay
Feb 12 04:30:26 <zhobbs_> even if you call startService() again for a service it'll just call onStart() for the already running service
Feb 12 04:30:31 <DJTachyon> cool
Feb 12 04:30:34 <DJTachyon> okay thanks man
Feb 12 04:30:35 <DJTachyon> goodnight
Feb 12 04:30:38 <zhobbs_> :)
Feb 12 04:43:23 <KNY> how does one go about debugging through a testcase?
Feb 12 08:29:30 <herriojr> RomainGuy on per chance?
Feb 12 08:30:25 <herriojr> if I extend CursorAdapter and override getView, if I don't want to use the convertView passed in, can I just ignore it?
Feb 12 08:52:25 <anno^da_> Could someone help me please? When I have checkboxes in a ListView I can't press the single rows anymore.
Feb 12 08:52:30 <anno^da_> What can I do to change that ?
Feb 12 09:10:06 <grantmidwinter> check focus?
Feb 12 09:13:13 <anno^da_> Hmm when I scroll with the wheel the rows get the focus
Feb 12 09:13:24 <grantmidwinter> fat fingers?
Feb 12 09:13:27 <grantmidwinter> hate to ask
Feb 12 09:13:34 <grantmidwinter> seriously, when you click on the row
Feb 12 09:13:44 <grantmidwinter> you should be able to check the listener
Feb 12 09:13:57 <grantmidwinter> check 'in' the listener even
Feb 12 09:14:10 <grantmidwinter> and return for either the checkboxes or the row itself
Feb 12 09:15:20 <anno^da_> http://dl.getdropbox.com/u/287197/device.png <- thats how it looks. (UI doesnt matter at the moment)
Feb 12 09:16:01 <anno^da_> Hmm ok let me try to get that out
Feb 12 09:24:01 <anno^da_> Well
Feb 12 09:24:26 <anno^da_> the listener of the ListView (onListeItemClick) doesnt get called at the moment
Feb 12 09:24:33 <anno^da_> when I touch a row
Feb 12 09:24:48 <anno^da_> This happens since I added the checkboxes to each row :/
Feb 12 09:24:54 <anno^da_> So where should I check the focus ?
Feb 12 09:27:10 <anno^da_> Ok
Feb 12 09:27:13 <anno^da_> got the solution
Feb 12 09:27:27 <anno^da_> I have to setup a onClick listener for the items itself
Feb 12 09:27:34 <anno^da_> http://groups.google.com/group/android-developers/browse_thread/thread/7482624a20e2473c/1d2f0cace5b72629?lnk=gst&q=Checkbox+in+ListView#1d2f0cace5b72629
Feb 12 09:27:45 <anno^da_> romainguy__ explained it very nice
Feb 12 10:22:58 <rzr> hi
Feb 12 10:23:21 <rzr> is there a page listing build option , i need to enable developpers tools
Feb 12 11:04:11 <tauno> How do I load system resources? I want to load android.R.id.message and I do it like this: a) I get the inflater: getSystemService(Context.LAYOUT_INFLATER_SERVICE) b) inflater.inflate(android.R.id.message, myFancyViewGroup);
Feb 12 11:04:32 <tauno> and I get android.content.res.Resources$NotFoundException: File  from xml type layout resource ID #0x102000b
Feb 12 11:06:32 <tauno> I'm doing something stupid, right? :)
Feb 12 11:18:32 <anno^da_> I'm getting crazy with that sms provider
Feb 12 11:18:46 <anno^da_> I'm listening with an observer to content://sms
Feb 12 11:19:40 <anno^da_> when a sms is incoming everything works fine. But when I read the SMS and they change their status to "read = 1" nothing gets notified
Feb 12 11:19:42 <anno^da_> grr
Feb 12 11:38:45 <tauno> ah, sure I'm doing something stupid.. it says "..for inflating layout resources in this context." and not system contex :)
Feb 12 12:37:31 <bilboed-pi> does the g1 support rtsp natively ?
Feb 12 12:39:24 <dared> actually found that - maybe helps -> http://code.google.com/p/android/issues/detail?id=1513
Feb 12 13:22:37 <DJTachyon> services, binders, aidl, whew
Feb 12 13:22:40 <DJTachyon> this is getting heavy :P
Feb 12 13:23:12 <digitalspaghetti> DJTachyon: i know, thats where i am just now with my development
Feb 12 13:28:53 <DJTachyon> :P
Feb 12 13:29:11 <DJTachyon> damnit, how do you get eclipse to actually insert spaces for tabs, i set it to, but its still doing it
Feb 12 13:30:18 <DJTachyon> guess i have to define a custom style
Feb 12 13:39:07 <digitalspaghetti> DJTachyon: in general preferences under text editors
Feb 12 13:47:01 <DJTachyon> yeah it wasn't doing it for me ..
Feb 12 13:47:09 <DJTachyon> i had to set a custom style for the project
Feb 12 14:04:36 <DJTachyon> http://osdir.com/ml/AndroidBeginners/2009-01/msg00171.html  .. how do i get the application context correctly?
Feb 12 14:36:19 <tauno> oh $deity - I just realised what I tried to do before.. I'm stupid and blind... and stupid
Feb 12 14:43:21 <anno^da_> digitalspaghetti: How much spaces should be used normally ?
Feb 12 14:43:27 <anno^da_> for an indent
Feb 12 15:35:24 <tauno> If I set a 9patch as a theme bacground, it is painted in some weird position (eg the upper side of the 9patch is painted behind the "notification bar" or however it is called - 0:0 of the image is painted to 0:0 of the physical screen)
Feb 12 15:35:43 <tauno> I'm using android:windowNoTitle
Feb 12 15:37:06 <tauno> shouldn't it be painted to the below the notification bar not behind it?
Feb 12 15:43:02 <tauno> romainguy__ are you around by any chance?
Feb 12 15:43:30 <tauno> guess it's too early in the US :)
Feb 12 15:47:02 <jt436> 7.46 on the west coast so possibly :)
Feb 12 16:04:56 <verxion> I need some stupid help
Feb 12 16:05:05 <verxion> I apologize, this seems pretty straight forward
Feb 12 16:05:08 <verxion> but I can't figure it out
Feb 12 16:05:21 <verxion> I have installed the android sdk
Feb 12 16:05:42 <verxion> and eclipse-jee-ganymede-sr1-win32
Feb 12 16:05:42 <vol> you probably want intelligent help
Feb 12 16:05:46 <vol> sigh
Feb 12 16:05:56 <verxion> when I go to run eclipse, it tells me there isn't any jre installed
Feb 12 16:06:06 <vol> do you HAVE a java runtime installed?
Feb 12 16:06:18 <verxion> well, that is why I am apologizing
Feb 12 16:06:34 <verxion> is the jre NOT included with the jdk?
Feb 12 16:06:42 <vol> no, the jre is a subset of the jdk
Feb 12 16:06:44 <verxion> I have jdk-6u12-windows-x64 installed
Feb 12 16:06:46 <vol> did you install a jdk?
Feb 12 16:06:59 <vol> mmm, not sure if 64 bit will cause trouble, but ok
Feb 12 16:07:07 <verxion> ok, then perhaps this is simple - I just need a jre in addition to the jdk?
Feb 12 16:07:15 <vol> no, the jdk is just fine
Feb 12 16:07:19 <vol> the jdk is the jre with dev stuff
Feb 12 16:07:21 <verxion> (I would have expected the jdk to include a jre)
Feb 12 16:07:24 <verxion> ok, then darn
Feb 12 16:07:27 <verxion> I am at a loss here
Feb 12 16:07:37 <vol> in eclipse, go to window -> preferences
Feb 12 16:07:46 <verxion> it won't even LAUNCH
Feb 12 16:07:53 <verxion> immediately claims I have no jre
Feb 12 16:07:54 <vol> interesting
Feb 12 16:08:10 <vol> http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/classpath.html
Feb 12 16:08:17 <vol> try setting up your classpaths, see what that does
Feb 12 16:08:22 <verxion> I have a Program Files\Java\JRE6 directory
Feb 12 16:08:23 <vol> eclipse is a java app
Feb 12 16:08:32 <vol> so if you don't have java set up such that it can be launched, you can't run eclipse
Feb 12 16:08:37 <verxion> as well as a Program Files\Java\jdk1.6.0.12 directory
Feb 12 16:08:45 <tauno> if you type "java -version" from the commandline - what happens?
Feb 12 16:08:53 <vol> you want to use the jdk1.6.0.12 dir.
Feb 12 16:09:00 <vol> also, what tauno said
Feb 12 16:09:08 <vol> if you get "bad command or file name" then go to my link
Feb 12 16:09:10 <verxion> tauno: I get "java version '1.6.0_12'"
Feb 12 16:09:11 <vol> and set up your paths
Feb 12 16:09:14 <vol> orly?
Feb 12 16:09:17 <verxion> yeah
Feb 12 16:09:23 <verxion> the install SEEMED to go fine
Feb 12 16:09:28 <verxion> but I am new to the dev side of java
Feb 12 16:09:37 <vol> that's very interesting.
Feb 12 16:09:37 <verxion> I have only supported java apps in the past at work
Feb 12 16:09:40 <verxion> not developed any
Feb 12 16:10:22 <vol> hmm.
Feb 12 16:10:40 <vol> it might be complaining that you don't have a j2ee instance since you got the j2ee version of eclipse, but I consider that very unlikely.
Feb 12 16:11:09 <vol> you got the 64 bit version of java?
Feb 12 16:11:15 <vol> try uninstalling it and using 32 bit instead
Feb 12 16:11:21 <vol> that's about the only thing I can think of
Feb 12 16:11:28 <vol> past that, #eclipse I guess
Feb 12 16:13:34 <andrewoid> How do I detect long presses in a ListActivity?
Feb 12 16:15:57 <verxion> hrm, ok
Feb 12 16:15:59 <verxion> thanks vol
Feb 12 16:16:24 <nemyhlovecraft> hey all, I'm having a bizzare issue
Feb 12 16:18:56 <nemyhlovecraft> in my app, I have a rigged subclass of ViewAnimator which takes an arbitrary layout resource id, inflates it on a child of the ViewAnimator, and then animates it using the animation passed into the ViewAnimator
Feb 12 16:20:08 <nemyhlovecraft> in one of my views, I have a slightly modified version of the GLSurfaceView from the API Demos
Feb 12 16:20:42 <nemyhlovecraft> When this view is presented as the first view for the main activity (via the ViewAnimator I described)
Feb 12 16:20:58 <nemyhlovecraft> the GL commands draw
Feb 12 16:21:08 <verxion> ok, it looks like I do NOT have any classpath variable in my environmental variables
Feb 12 16:21:34 <nemyhlovecraft> when it is not the first view presented, the GL commands do not draw
Feb 12 16:22:37 <nemyhlovecraft> I have hooked up adb and clicked through every line of code that initializes the GLView and as far as I can tell the situation surrounding initialization is the same for either case
Feb 12 16:23:17 <vol> verxion: that probably is the problem
Feb 12 16:23:38 <nemyhlovecraft> I can provide the source, but I wanted to see if anyone had an initial reaction
Feb 12 16:24:10 <vol> nemyhlovecraft: cool?
Feb 12 16:24:28 <vol> for the record, I think I discovered a similar bug with SurfaceView
Feb 12 16:24:28 <romainguy__> tauno: shouldn't it be painted to the below the notification bar not behind it? < not, that's expected
Feb 12 16:24:40 <vol> if the surfaceview isn't the first thing that has setContentView on it, it will draw black.
Feb 12 16:24:45 <vol> so, this might be the same problem
Feb 12 16:25:33 <nemyhlovecraft> weird, so is this regardless of content?
Feb 12 16:26:13 <nemyhlovecraft> like GL, Canvas, other views on a Surface?
Feb 12 16:27:23 <vol> I think so.
Feb 12 16:27:31 <vol> This was happening for me with OpenGL and a normal SurfaceView too
Feb 12 16:27:51 <nemyhlovecraft> well, then, can you think of another way to have a view with a threaded GL renderer besides using surface?
Feb 12 16:28:31 <vol> no, you're stuck with surfaceview as far as I know.
Feb 12 16:28:31 <tauno> romainguy__, why so?
Feb 12 16:28:47 <romainguy__> because the window extends behind the status bar
Feb 12 16:28:59 <vol> My workaround was to call setContentView on the SurfaceView first thing, and then when surfaceCreated is called, have it call something else which will setContentView on whatever you want seen instead
Feb 12 16:29:32 <nemyhlovecraft> ok, then what about a workaround where the GLView is added to activity before everything else but not shown until the view I need it in is showing?
Feb 12 16:29:43 <vol> so, when surfaceCreated is called on my SurfaceView, I tell it to show the splash screen. Bear in mind that you might want to have a one-time-only boolean sort of thing to only have this happen once, and have normal behavior every other time.
Feb 12 16:29:53 <vol> I don't know, you'll have to work that one out yourself : (
Feb 12 16:30:02 <vol> I can only tell you what I've found
Feb 12 16:30:28 <bilboed-pi> anyone knows of a media player that could actually give more info on why opencore (sic) can't play a certain media file ?
Feb 12 16:31:47 <nemyhlovecraft> thanks
Feb 12 16:32:20 <nemyhlovecraft> I think that solution is about what I feared
Feb 12 16:32:39 <nemyhlovecraft> but its cool
Feb 12 16:32:54 <tauno> romainguy__, why I'm asking is because I want the user to "feel" the app is starting faster (like you are doing in Shelves). I set the theme background the same as I'm using in my "first activity" - that is a LinearLayout with android:background="@drawable/fancyImage". Now the LinearLayout paints the image starting from below the notification bar but the theme background is painted from 0:0 so they don't match and you'll see that the image
Feb 12 16:33:38 <tauno> romainguy__, is there any sane way I could force it to display below the notification bar.. not underneath it?
Feb 12 16:40:05 <tauno> or are there only two options like a) don't use a theme bacground at all ans stick with the black one b) add some whitespace to the top of the splash background so it will look like it's painten below the notification bar and not underneath it? (No, I don't plan on doing b) :P)
Feb 12 16:40:33 <romainguy__> tauno: no, there's no workaround this
Feb 12 16:40:47 <romainguy__> I plan on fixing this though
Feb 12 16:42:04 <tauno> ok, I'll scrap the idea of a theme background till a future date then :)
Feb 12 16:42:36 <romainguy__> you should do it anyway
Feb 12 16:42:46 <romainguy__> it's better to see it shift a little bit than nothing
Feb 12 16:42:47 <tauno> ..and thanks for fixing this (or.. improving.. since it's not a bug as you said but it's designed so)
Feb 12 16:42:56 <romainguy__> that's what I did in Shelves
Feb 12 16:43:46 <KNY> romainguy__, is Shelves out yet?
Feb 12 16:43:54 <romainguy__> nope
Feb 12 16:43:59 <romainguy__> almost done!
Feb 12 16:44:03 <KNY> well hurry up! My Collection is all kinds of buggy
Feb 12 16:44:04 <romainguy__> you can download the source code though
Feb 12 16:44:05 <tauno> romainguy__, I'll try to offer the shifting solution to the UI guys.. but I won't hold my breath that they think it's the better solution :)
Feb 12 16:44:10 <KNY> yeah, I did a while back
Feb 12 16:44:11 <romainguy__> KNY: and ugly :0
Feb 12 16:44:20 <KNY> but I have my own projects to finish first
Feb 12 16:44:24 <KNY> romainguy__, you got that right
Feb 12 16:44:35 <romainguy__> that said My Collections has more features
Feb 12 16:44:37 <romainguy__> but eh
Feb 12 16:45:12 <KNY> I don't care about anything but books
Feb 12 16:46:43 <anno^da_> Does someone know a solution to get the URI for a contact photo of a specific contact? Or a way to show the photo in the extended statur bar? (I'm fighting with the RemoteView which needs a URI or ID to set an image on an imageview)
Feb 12 17:16:40 <jasta> anno^da_: RemoteView needs that because the view is actually constructed in another process
Feb 12 17:16:50 <jasta> so it can only use things that can be easily communicated across processes
Feb 12 17:16:54 <jasta> like package resources or content uris
Feb 12 17:19:21 <jasta> that said you should be able to get what you need quite easily.  how does the contacts app do it?
Feb 12 17:20:02 <grantmidwinter> what's shelves?
Feb 12 17:20:19 <jasta> grantmidwinter: an application that romain wrote and published recently
Feb 12 17:20:36 <grantmidwinter> oh i've seen the demo actually
Feb 12 17:22:33 <grantmidwinter> that progress bar overlay at the bottom is slick
Feb 12 17:22:42 <grantmidwinter> i'm such a nerd :(
Feb 12 17:22:52 <anno^da_> jasta: Ok. Well I can get the contacts photo with a "loadContactPhoto()" method but this one drops me back a Bitmap which I can't use for a remote view right?
Feb 12 17:24:05 <jasta> anno^da_: look at the implementation of Contacts.java$People#openContactPhotoInputStream
Feb 12 17:24:13 <jasta> the uri it creates is the uri you want...
Feb 12 17:24:50 <jasta> actually no, they aren't using their own convention here.
Feb 12 17:25:01 <jasta> its not possible using their provider as it is designed it seems.  shame on them.
Feb 12 17:25:21 <jasta> so, you'll want to implement your own content provider which "maps" to this via openContactPhotoInputStream
Feb 12 17:26:24 <jasta> unfortunately you may also need to store the photo on disk.
Feb 12 17:26:42 <jasta> so you've got some operating overhead there.  this is all because they didnt follow their own pattern for storing binary data.
Feb 12 17:27:09 <jasta> they explicitly tell you not to store image data and such in a blob column, but rather to store a uri reference which maps to a file on disk through a ContentProvider
Feb 12 17:27:28 <jasta> to confirm that they dont have something like this already though, i'd look at the Contacts provider.
Feb 12 17:27:51 <jasta> specificlaly at openFile() to see what they do with it
Feb 12 17:28:04 <anno^da_> yeah ok thank you.
Feb 12 17:28:04 <anno^da_> :)
Feb 12 17:28:20 <anno^da_> I was wondering why they didnt store the URI
Feb 12 17:28:34 <jasta> probably becuse that code is older than the convention ;)
Feb 12 17:28:36 <jasta> because*
Feb 12 17:28:40 <anno^da_> In every documentation you are told "store the uri in the provider" :)
Feb 12 17:28:58 <jasta> any way, with some operating overhead you should be able to proxy this.
Feb 12 17:29:09 <anno^da_> But yeah ok. At the moment I will drop that feature but I will get back to that.
Feb 12 17:29:12 <anno^da_> Yeah sure
Feb 12 18:58:53 <jt436> .........
Feb 12 19:51:48 <jt436> busy day in here
Feb 12 20:04:40 <morrildl> jt436: sorry I missed your comment what with all the noise
Feb 12 20:05:04 <morrildl> I've never seen a channel with 82 people be this quiet :)
Feb 12 20:05:18 <morrildl> it's cool now, though, pjv is here
Feb 12 20:05:53 <jt436> I just found out Monday is a public holiday
Feb 12 20:05:59 <jt436> otherwise I would have turned up for work
Feb 12 20:09:45 <herriojr> romainguy: I remember seeing something about disabling the onScroll and ListViews and how to make it so the background isn't black during it...I can't seem to find it again...do you know what I'm talking about (I'm sure you do :) )
Feb 12 20:10:02 <herriojr> *disabling the black background
Feb 12 20:10:07 <herriojr> or setting it to something else
Feb 12 20:19:43 <zhobbs> herriojr: http://android-developers.blogspot.com/2009/01/why-is-my-list-black-android.html
Feb 12 20:48:50 <herriojr> what's the default listSelector color in #XXXXXXXX?
Feb 12 20:51:45 <herriojr> or better yet, how can I get the listSelector to show when my item has a background set without placing it in front and without putting in a "hack" to remove the background?
Feb 12 21:16:58 <herriojr> did anyone else have the issue where when they upgraded to 1.1SDK in eclipse, the javadoc no longer shows when you hover over an SDK class/function?
Feb 12 21:47:50 <xavd> herriojr: yeah we're looking at it now
Feb 12 22:30:58 <l0gis1c> i just pulled the whole source today and i'm trying to build the kernel for the emulator. can you point me as to where to get the right config file? pardon the cross-post, just hoping im at the right place this time.
Feb 12 22:37:29 <jt436> l0gis1c: wrong channel, try #android
Feb 12 22:37:35 <Disconnect> you might get better answers on the android-platform list
Feb 12 22:37:52 <Disconnect> and its only the wrong channel cuz its always idle :)
Feb 12 22:38:01 <l0gis1c> jt436: posted there first
Feb 12 22:38:18 <l0gis1c> Disconnect: ty
Feb 12 22:40:24 <m0z_work> does anyone know of any ongoing project to get Android running on the HTC Touch Pro (Raphael)?
Feb 12 22:40:53 <m0z_work> the touch pro is just so much sexier than the g1 :p
Feb 12 22:48:31 <SanMehat> m0z_work: thanks.
Feb 12 22:50:17 <m0z_work> SanMehat?
Feb 12 22:53:55 <[RiS]> FileObserver should detect when I upload a new file to the sdcard using adb push?
Feb 12 23:02:39 <SanMehat> no
Feb 12 23:07:36 <grey-> on
Feb 12 23:15:46 <herriojr> hey, so if I want to load extra information to the views after onCreate of a listActivity using a CursorAdapter, I'm trying to use a handler to load that information into the views.  When I start the application, the handler doesn't seem to draw to the views...what exactly do I need to do to get it to work?
Feb 12 23:16:16 <herriojr> Essentially, I'm trying to load images to the list
Feb 12 23:16:27 <herriojr> after the listView has been populated
Feb 12 23:18:01 <herriojr> so the contentView isn't loaded by the time onResume runs...which is the last time I can try and force it to load the extra data that I know of
Feb 12 23:32:40 <romainguy__> herriojr: seriously, you file bugs based on how you think the framework works? without even testing if that's the case??
Feb 12 23:39:46 <herriojr> romainguy: I look at it this way, I'm a user of the framework, so if things don't work how I expect, from a user-design perspective, or the documentation doesn't explain it in enough detail, there's something wrong
Feb 12 23:41:00 <herriojr> it's essentially user-feedback
Feb 12 23:42:05 <romainguy> how is it not working like you expect?
Feb 12 23:42:29 <herriojr> which issue are you talking about at this point?
Feb 12 23:43:31 <herriojr> if there was a way to report feedback without going the the issue tracker, I'd do that
Feb 12 23:54:31 <herriojr> romainguy: so if you look at the documentation under the header Drawing, you will notice it says that invalidate() forces it to redraw
Feb 12 23:54:53 <herriojr> so, when I look through the source code of a view and see that removeView() and addView() both call invalidate
Feb 12 23:55:45 <herriojr> that's a documentation error if what you said is true, which I'm assuming it is
Feb 12 23:55:57 <romainguy> no, that's not a documentation error
Feb 12 23:56:03 <romainguy> it does forces the view to redraw
Feb 12 23:56:12 <romainguy> but the documentation doesn't have to explain the implementation
Feb 12 23:56:32 <romainguy> especially when it's an optimization where we don't draw twice in the same event
Feb 13 00:01:30 <herriojr> romainguy: ok, I will take that into account in the future, documentation doesn't reflect optimizations
Feb 13 00:01:54 <romainguy> the thing is our documentation defines the public API
Feb 13 00:02:07 <romainguy> which means that any information given about the implementation means that it has to be implemented this way
Feb 13 00:02:33 <romainguy> not only can this be a hindrance for ourselves, but it prevents anyone from writing a different implementation
Feb 13 00:03:19 <herriojr> I understand, and I'll cool it on the bug reports ;)  Although there is one with BitmapFactory, but someone had already reported it :)
Feb 13 00:03:32 <romainguy> what's the bug?
Feb 13 00:03:36 <herriojr> memory leaks
Feb 13 00:04:03 <herriojr> I ready your replies on it
Feb 13 00:04:14 <herriojr> so don't worry, I don't have questions on it :)
Feb 13 00:04:29 <zhobbs_> herriojr: link?
Feb 13 00:04:42 <herriojr> I'd have to find it again
Feb 13 00:05:41 <romainguy> I find it weird that we never hit any memory leak with BitmapFactory in any of our apps... :)
Feb 13 00:06:25 <herriojr> well, it took a lot of scrolling to get it to happen on the phone, like 5+minutes
Feb 13 00:07:18 <herriojr> essentially, we load an image via http and display it in a list using an ArrayList (you don't have to worry about the list getting too big in this case)
Feb 13 00:07:41 <herriojr> when we scroll up and down enough, it crashes
Feb 13 00:08:13 <herriojr> I was trying to find out how to turn the heap display on/off, but couldn't find it....I just assumed it was associated with some of the reports I saw earlier
Feb 13 00:10:28 <herriojr> I didn't write the code, but it could potentially be something else, but we check that all cursors were closed
Feb 13 00:10:50 <romainguy> what about the http code?
Feb 13 00:11:13 <herriojr> he loads it all up front via http, so it shouldn't effect scrolling
Feb 13 00:11:25 <herriojr> or shouldn't be effected by scrolling
Feb 13 00:11:43 <zhobbs_> and saves it to the filesystem?
Feb 13 00:12:28 <herriojr> some portion of the data is saved to a database, but as I said, we checked the cursors to make sure he closed them all
Feb 13 00:12:41 <herriojr> and the list is using an ArrayAdapter, not a CursorAdapter
Feb 13 00:13:23 <romainguy> leaking views? :)
Feb 13 00:16:42 <herriojr> so in getView().....if (convertView!=null)view=convertView;else view=inflater.inflate(R.layout.somelayout, null);
Feb 13 00:16:47 <herriojr> That's the only inflate statement
Feb 13 00:17:17 <herriojr> and view is returned by getView
Feb 13 00:19:31 <herriojr> so, what can cause a view to leak?
Feb 13 00:27:23 <herriojr> I'm looking into it, so nevermind
Feb 13 00:34:52 <herriojr> oh wow, this view memory leak stuff is interesting
Feb 13 01:02:11 <eldenz> is itpossible to shut the phone down?
Feb 13 01:02:19 <eldenz> programmatically
Feb 13 01:03:11 <herriojr> check Permissions
Feb 13 01:03:28 <herriojr> if it was, I'd assume it would be a permission
Feb 13 01:03:55 * eldenz opens browser
Feb 13 01:05:31 <herriojr> there's REBOOT
Feb 13 01:06:07 <herriojr> but you also have to make sure that's not a permission that requires a sharedUserId or an OS level signature
Feb 13 01:10:05 <eldenz> it would be nice to have a timer that shuts the pphone off... so i could listen too some music when going to bed :)
Feb 13 01:10:15 <herriojr> try BRICK ;)
Feb 13 01:10:35 <herriojr> that was just a joke just in case you didn't catch it :)
Feb 13 01:10:42 <eldenz> lol
Feb 13 01:11:06 <eldenz> i'm sloe here... switching to browser&co ;)
Feb 13 01:11:10 <eldenz> slow
Feb 13 01:12:03 <jbq_> Amusingly, BRICK used to actually work as advertised. When we removed the code, we left the permission around.
Feb 13 01:14:01 <eldenz> heh
Feb 13 01:14:48 <eldenz> as long it is the permission and not the code that is left around.... ;)
Feb 13 01:15:29 <eldenz> nn!
Feb 13 01:16:01 <ctate> isDeviceOnFire()
Feb 13 01:18:03 <jbq_> boolean android.os.getDeviceOn(); void android.os.setDeviceOn(boolean on); (where calling the latter with false does a shutdown).
Feb 13 01:18:32 <herriojr> romainguy: sorry if I don't make sense a lot, I really haven't gotten much sleep in the last couple months
Feb 13 01:19:07 <jbq_> welcome to our world (sorry, couldn't resist)
Feb 13 01:19:13 <herriojr> :)
Feb 13 01:19:27 <herriojr> I've been getting around 3-4 hours of sleep
Feb 13 02:29:46 <herriojr> out of curiosity, why is onScrollStateChanged called after onScroll?
Feb 13 02:30:57 <herriojr> it makes it a tad bit more complex to load extra data into lists upon scrolling since you don't know what the new state is when onScroll is called
Feb 13 02:31:35 <herriojr> and I don't want to load them into the list on getView since that slows it down significantly
Feb 13 02:32:06 <herriojr> I already have a work-a-round, I was just curious as to why
Feb 13 02:35:49 <romainguy> read code.google.com/p/shelves, I've implemented exactly that kind of stuff
**** ENDING LOGGING AT Fri Feb 13 02:59:57 2009