**** BEGIN LOGGING AT Wed Jan 28 02:59:57 2009 Jan 28 03:38:40 so, if I have a database where TABLE_1 holds the ID of the "primary" item in TABLE_2, and I add an item to TABLE_2, how can I make it so the ID of the primary item in TABLE_1 holds the newly added ID? Jan 28 03:38:46 I can't seem to get the trigger to work Jan 28 03:41:14 I tried: Jan 28 03:43:49 create trigger TABLE_2Add insert on TABLE_2 begin update TABLE_1 set primary_id=new._id where (TABLE_2.shared_value = TABLE_1.shared_value); end; Jan 28 03:44:11 but the result is that primary_id is set to null Jan 28 03:54:37 quick question... is there a way to programatically provide arguments/parameters to an Intent as you call it? Jan 28 03:56:19 Intent.putExtra Jan 28 03:57:13 ah ha Jan 28 03:57:14 thanks Jan 28 09:31:19 so has anyone tried this admob advertising in their app? Jan 28 09:35:06 <_avatar> ugh, no, and i hope android apps don't start becoming ad infested Jan 28 09:48:54 Could someone help me with the layout. I got the following setup: http://paste.pocoo.org/show/101642/ Jan 28 09:49:12 How can I add 10px of space between the module1 and module2 Jan 28 09:49:31 I have tried to set layout_topMargin in module 2 which doesn't work Jan 28 09:51:45 <_avatar> how about android:paddingTop="10dp" instead of layout_topMargin? Jan 28 09:52:21 well padding ads me 10px inside my layout. But I have a border around each of the two modules Jan 28 09:52:22 http://dl.getdropbox.com/u/287197/Hierarchy%20Viewer.jpg Jan 28 09:52:34 thats how the bottom of module 1 and top of module 2 looks like Jan 28 09:53:01 you can see the blue and grey border Jan 28 09:53:12 <_avatar> i've never used layout_topMargin, but you could always just put an empty 10px tall view in between the modules. Jan 28 09:53:39 Well ok that could be possible Jan 28 09:53:53 I just thought that it might be possible with parameters Jan 28 09:54:17 <_avatar> probably, but i've never been able to figure it out. :) Jan 28 09:56:17 <_avatar> along the same lines, it'd be really nice to be able to specify a default "spacing" size between children in a linearlayout. Jan 28 10:00:25 yeah Jan 28 10:01:00 well I'm coming from a web backrground and have done a lot of CSS stuff and margins are fine if they would work here Jan 28 10:01:34 I apply marginTop but when I look at the layout in the Hierarchy Viewer the settings is gone Jan 28 10:01:41 and marginTop is set to 0 Jan 28 10:02:15 All the other settings like padding are set pretty fine :( Jan 28 10:29:41 anno^da_: android:layout_marginTop? (instead of topMargin) Jan 28 10:31:29 jsharkey: I got this one: android:layout_marginTop="10px" Jan 28 10:31:58 I have opened a thread since I can't find the error: http://groups.google.com/group/android-developers/browse_thread/thread/ce580ad6ae1fd92a Jan 28 10:33:09 Perhaps you could look at it. I have no idea how to achieve that except adding a spacer (dummy) view between my two modules Jan 28 10:36:35 works fine for me Jan 28 10:36:44 anno^da_: Are you setting layout height and layout width? Jan 28 10:37:36 Jan 28 10:37:39 Jan 28 10:41:37 yeah sure I have set the values Jan 28 10:41:39 http://paste.pocoo.org/show/101643/ Jan 28 10:41:45 this is module 2 Jan 28 10:42:00 the padding works fine Jan 28 10:42:05 the backrgound as well Jan 28 10:42:10 umm Jan 28 10:42:19 anno^da_: yea works fine for me here too Jan 28 10:42:21 (as you can see on the picture) Jan 28 10:42:33 (http://dl.getdropbox.com/u/287197/Hierarchy%20Viewer.jpg) Jan 28 10:42:43 oh, you're saying it doesn't work for the "stub" linearlayout? Jan 28 10:42:57 Well I'm doing the following Jan 28 10:43:16 inflating the FrameLayout and finding the stub LinearLayout Jan 28 10:43:31 anno^da_: this works for me http://paste.pocoo.org/show/101644/ Jan 28 10:43:40 after that I add the module one and module two LinearLayouts to the stub LinearLayout Jan 28 10:43:43 anno^da_: you might consider using the tag Jan 28 10:43:59 which works like #include in c++ Jan 28 10:44:44 anno^da_: So you're adding to a LinearLayout that doesn't have layout height and width set... Jan 28 10:45:12 Why? Jan 28 10:45:13 http://paste.pocoo.org/show/101643/ Jan 28 10:45:21 This is the LinearLayout for module 2 Jan 28 10:45:29 with these parameters set right? Jan 28 10:45:38 Yes, but you're not replacing the stub LinearLayout, you're adding to it? Jan 28 10:45:52 right Jan 28 10:46:09 stub.addView(module1) (pseudo code) Jan 28 10:46:25 and addView(module2) Jan 28 10:46:57 try setting the layout height and width to the stub code Jan 28 10:47:03 anno^da_: are you doing anything custom with those modules, or could you statically them? Jan 28 10:47:41 i think the the stub code already gets the width/height from the second layout he just posted Jan 28 10:47:54 he's not replacing the stub, though Jan 28 10:48:00 oh hmm true Jan 28 10:48:18 if you forget the height and width somewhere in the tree, it will go wonky. Jan 28 10:48:30 looking at other peoples layouts is so much fun :) Jan 28 10:48:49 But haven't I set the layout params in the stub? Jan 28 10:49:05 http://paste.pocoo.org/show/101642/ Jan 28 10:49:31 when you say "stub" do you mean that this code isnt real? Jan 28 10:49:34 shouldnt work Jan 28 10:49:44 because plain "id" was depricated i thought Jan 28 10:49:59 that was just dummy code Jan 28 10:50:02 Wait Jan 28 10:50:08 I will paste the whole thing :) Jan 28 10:50:15 otherwise it gets too much of guessing :D Jan 28 10:50:19 crap im about to fall over asleep Jan 28 10:50:58 So this is the empty layout without the modules: http://paste.pocoo.org/show/101645/ Jan 28 10:51:04 * kRutOn pushes jsharkey over Jan 28 10:51:24 aha, then you inflate the stuff and add them to .getViewById(stub) Jan 28 10:51:25 module one: http://paste.pocoo.org/show/101647/ Jan 28 10:51:44 module two: http://paste.pocoo.org/show/101643/ Jan 28 10:52:37 grr that seems like it should work Jan 28 10:52:48 adding: http://paste.pocoo.org/show/101649/ Jan 28 10:53:13 anno^da_: is there a really good reason you want to inflate those layouts in code? Jan 28 10:53:28 AHA Jan 28 10:53:40 Well. I have to reuse these modules all the time. Jan 28 10:53:41 this whole "getModule()" business Jan 28 10:54:08 And they slightly change depending on the "service" Jan 28 10:54:29 Well and I thought it would be a good idea to reuse them in this situation. Jan 28 10:54:38 hmm Jan 28 10:54:56 does ModuleOne also have a LayoutInflater of its own? Jan 28 10:55:11 (well I'm a beginner :). It seemed to be the best solution for me not ending up in writing all the code over and over again) Jan 28 10:56:06 hmm Jan 28 10:56:22 http://paste.pocoo.org/show/ZHpJoBrrv4ILrXGzHDAp/ Jan 28 10:56:25 moduleOne Jan 28 10:56:26 :) Jan 28 10:57:23 (well I'm just at the starting point of coding *g So well there could be something too complicated. But I'm open for every hint :-) ) Jan 28 10:57:31 hmm Jan 28 10:57:56 usually inflater.inflate() likes have the parent view, with attach=false Jan 28 10:58:01 isntead of null Jan 28 10:58:05 not sure if that matters in this case Jan 28 10:59:11 also, not sure if the background is somehow overriding any set margins Jan 28 10:59:17 since 9patches sometimes have their own content area Jan 28 10:59:30 The strange thing that really confused me was that the padding parameter set for the modules works pretty fine. Just the margin gets dropped. Jan 28 10:59:48 hmm ok I test it without the back Jan 28 10:59:52 yea everything else /looks/ like it /should/ work Jan 28 11:00:35 also, might try setting marginBottom too, and then copy both top/bottom values to the first module Jan 28 11:00:40 just to play with debugging it Jan 28 11:01:07 in hierarcyviewer, does the property viewer verify that the margintop is dropped to 0? Jan 28 11:01:18 Yeah. Jan 28 11:01:30 padding is set to 10 in the viewer and margin is 0. Jan 28 11:02:28 even without the background there is no margin :( Jan 28 11:08:17 Thats how it looks in the Viewer: http://dl.getdropbox.com/u/287197/android.widget.LinearLayout%404307a130.jpg Jan 28 11:20:17 I think it's because you're telling the inflater service that this is the root of the viewgroup for each module Jan 28 11:21:02 because if you do that, LayoutParams will be null Jan 28 11:22:13 So I have to change this one right: module = (LinearLayout) _inflater.inflate(R.layout.modul_one, null); Jan 28 11:29:32 yeah, maybe pass a reference to the holder and then remove the .addView Jan 28 11:30:05 There is a way to do it during the .addView, but it might be simpler to do it the other way Jan 28 11:30:27 so have it be inflater.inflate(R.layout.module_one, root) Jan 28 11:32:09 Otherwise you might have to do some stuff with XmlPullParser Jan 28 11:40:55 kRutOn Jan 28 11:40:58 I will try this Jan 28 11:41:05 thank you to jsharkey as well. Jan 28 11:41:53 I'm trying this now. jsharkey do you see a posibility to to the same thing (with reuse of the modules) with Jan 28 11:41:56 +e Jan 28 11:42:57 Take a look at this post Jan 28 11:42:59 http://www.mail-archive.com/android-developers@googlegroups.com/msg06247.html Jan 28 11:47:33 thanks. Jan 28 11:47:38 Now it works :) Jan 28 11:47:53 thank you guys. Jan 28 11:49:13 No problem Jan 28 12:15:37 I got another small question Jan 28 12:16:15 Why can't I add the same module two times? (this was possible before changin the inflate(....) ) Jan 28 12:17:24 Adding two of the same modules leads me to two modules in the layout but only one is set to the values which I set in the ModuleOne class. The other one just has the dumy values from my xml file. Jan 28 12:22:05 which looks like this: http://dl.getdropbox.com/u/287197/Hierarchy%20Viewer-1.jpg Jan 28 12:22:52 It is the same box but the second box has the default values. (created exactly the same like the first module) Jan 28 12:23:39 http://paste.pocoo.org/show/101657/ <- That's how I create them. Jan 28 12:25:09 Ok the first one overrides the second one. Hmmm the view hirarchy confuses me :) Jan 28 13:00:53 I know why this happens. But I don't know how to avoid it. I just want ro reuse the layout xml file which leads to the mentioned problems. :( Jan 28 13:10:06 how are you picking up the references to the things you're changing Jan 28 13:12:11 If you're doing findViewById on the container, it traverses and returns the first match Jan 28 13:21:38 http://paste.pocoo.org/show/101665/ Jan 28 13:21:44 Thats what I do Jan 28 13:21:57 I'm getting them from the inflated module Jan 28 13:24:13 how are you setting module Jan 28 13:24:59 module = (LinearLayout) _inflater.inflate(R.layout.modul_one, root); Jan 28 13:25:41 unfortunately .inflate() returns the root you passed in Jan 28 13:26:37 Yeah it seems so Jan 28 13:26:58 So there is no way to avoid that? Jan 28 13:27:22 No, you'll have to .getChildAt() Jan 28 13:27:23 bbl Jan 28 13:28:45 getChildAt on the module that gets returned ? Jan 28 13:29:06 er forgot to mention you coukd try Jan 28 13:29:18 (on G1 now) Jan 28 13:29:50 inflater.inflate(id, root, false) Jan 28 13:30:16 then take the return as module, the do root.addView(module) Jan 28 13:30:53 that will still set layout params but now attach the view yet Jan 28 13:30:57 bbl Jan 28 13:34:09 ahh ok. Jan 28 13:34:18 I will try that. Thank you for your help kRutOn Jan 28 13:36:21 Ok this works. Jan 28 13:37:15 But I have one last problem. When I switch to landscape mode onPause and onDestroy onResume and onCreate get called. Jan 28 13:37:43 Now the values for my second moduleOne override the values of the first moduleOne Jan 28 13:38:25 So I end up with two modules with the same values Jan 28 13:38:57 BUT (an this is the strange thing) when I press back and load up the same activity again all values are applied correct. Jan 28 13:39:37 Just switching to landscape mode leads me to this problem which makes no sense at all for me since it is pretty the same like opening up the activity Jan 28 15:03:01 hi all, Jan 28 15:04:33 I try put a simple scrollbar into an imageview but this throws a nullpointerexception on the method view.ondrawscrollbar I receive it when I set the horizontal and the vertical enabled Jan 28 15:05:05 the other code is okay because it always works correctly. I added only the setting of the scrollbars Jan 28 15:05:33 and now my famous phrase: any suggesions? Jan 28 15:09:00 put the imageview inside a scrollview? Jan 28 15:14:03 hi all, conection troubles as usual, no answers during my absence? Jan 28 15:15:35 (10:09:00 AM) digitalspaghetti: put the imageview inside a scrollview? Jan 28 15:16:07 iWas-Esses: That is the only thing said after you asked for suggestions Jan 28 15:51:34 i'm using to run a sample android application from eclipse, but always i got "can't find Hello.apk" neither i can run adb, at the time i did set the path for the android sdk/tools in the system PATH env. var. and in eclipse preferences , any comment ? Jan 28 15:54:03 hmm that is unusual Jan 28 15:54:10 f00bar80: From the terminal can you run adb? Jan 28 15:54:47 famast, no not recognnized as a command Jan 28 15:57:15 famast, i did find a way to build an apk using apkbuilder.bat, but i wonder why it's not working from eclipse Jan 28 15:58:17 f00bar80: I'm not sure about eclipse. But if you can't run adb from the term your environment isn't setup correctly Jan 28 15:58:42 f00bar80: your using Ubuntu? Jan 28 15:59:34 famast, no i'm on windows Jan 28 16:02:38 ah ok Jan 28 16:03:22 f00bar80: I know you set the path in eclipse preferences, but are you sure you set it correctly? That is really the only thing I could think of Jan 28 16:04:40 famast, i followed http://code.google.com/android/intro/installing.html Jan 28 16:05:46 What version of eclipse? Jan 28 16:06:12 3.4.1 Jan 28 16:07:49 arg Jan 28 16:07:51 i'm at a loss Jan 28 16:07:56 but I do know your path isn't setup right Jan 28 16:08:06 even though thats optional, you could try fixing that and see if it helps Jan 28 16:11:19 famast, is installing the android plugin or not could be a reason ? Jan 28 16:21:13 famast, any comment ? Jan 28 16:27:36 f00bar80: You don't have the eclipse android plugin? yea get that :) Jan 28 16:27:36 ppl any comment ? Jan 28 16:27:44 sorry about the delay Jan 28 16:32:26 famast, does the plugin has a role in creating the apk file Jan 28 16:37:06 f00bar80: pretty sure it has a role in just about everything if your using eclipse Jan 28 16:37:38 f00bar80: You coulld always compile though and debug without eclipse, but thats just a pain :) Jan 28 16:53:57 any idea who we report a market bug to? Jan 28 17:46:55 how can I tell when a ListView is done loading the initial screen of Views? Jan 28 17:47:53 hmm, maybe setListAdapter() will block until the first set of rows is loaded... Jan 28 17:48:27 hmm, maybe not Jan 28 17:53:04 zhobbs: after the first layout pass Jan 28 18:03:49 how do I know when that has been completed? Jan 28 18:22:17 jsharkey: Are you online? I got one last problem with my UI setup. Jan 28 18:39:19 zhobbs_: You can override onLayout and I think it's done after you call super.onLayout Jan 28 18:43:48 * kRutOn takes a look at the code Jan 28 18:54:10 or just use a ViewTreeObserver Jan 28 18:59:51 kRutOn: could you help me hopefully one last time? Jan 28 19:03:43 romainguy: Cool. Jan 28 19:03:51 anno^da-: What's up? Jan 28 19:06:55 Well could I link you an APK otherwise it would be too difficult to explain since I don't really understand why it happens. It is a problem that is caused perhaps by the same things we talked about this afternoon. Jan 28 19:07:07 ok Jan 28 19:16:20 kRutOn: thanks Jan 28 19:39:19 yawn Jan 28 19:56:44 hmm, i think someone broke the emulator. you can't "call" one emu to another using the console port anymore Jan 28 19:57:21 the problem seems to be a shallow bind or connect (only 127.0.0.1 is bound, but then the first address returned from localhost is used, which is ::1 if ipv6 is enabled) Jan 28 20:14:13 its broken for other reasons. damn Jan 28 20:30:38 can a linear layout have a sub-linearlayout? Jan 28 20:30:50 yes Jan 28 20:31:15 very common actually Jan 28 20:32:28 hrmm, doing something wrong then. This is my parent view: http://paste.ifies.org/295 and in the inner linearlayout i want to have this: http://paste.ifies.org/296, http://paste.ifies.org/297 is my code - line 58 Jan 28 20:33:44 and what's wrong? Jan 28 20:34:25 http://paste.ifies.org/298 Jan 28 20:34:31 oh Jan 28 20:34:35 you're calling getLayout() Jan 28 20:34:37 that won't work Jan 28 20:34:55 you need to inflate R.layout.user_home with a LayoutInflater Jan 28 20:35:00 ahh Jan 28 20:35:04 look at the sample code in the SDK there are examples of that Jan 28 20:35:18 thanks Jan 28 21:17:32 romainguy: I also got a problem with my app. Could I query you with some code snippets? I have searched the whole day for the error but well I don't get the answer. (some inflating of a layout problem) Jan 28 21:17:50 sure Jan 28 21:20:35 android's build system is so weird :) Jan 28 21:20:52 You should've seen it before. Jan 28 21:21:01 what if i was hacking on the framework, how would i just make it rebuild that efficiently? Jan 28 21:21:06 and then make a new system.img for me? Jan 28 21:21:08 jasta: here you go: Jan 28 21:21:11 The old recursive build system wasn't tremendous. :-) Jan 28 21:21:20 make core framework snod Jan 28 21:21:21 function fsnod { Jan 28 21:21:21 ONE_SHOT_MAKEFILE="$ANDROID_BUILD_TOP/frameworks/base/Android.mk" make -j8 framework snod && adb shell stop && adb sync && adb shell start Jan 28 21:21:21 } Jan 28 21:21:36 this will compile only the framework, make the system image, and sync framework.jar onto the device Jan 28 21:21:50 (which implies you did an adb remount first) Jan 28 21:22:07 Or you could always backport it to http://fadden.com/dl-misc/#fms Jan 28 21:22:37 is there any way to do % based widths? I would like a tableview with 4 cells in a row at 25% width each Jan 28 21:22:43 fsnod is something you wrote? Jan 28 21:22:54 I put that in my .profile Jan 28 21:23:05 i saw ONE_SHOT_MAKEFILE when i was poking around at tthe build setup shell script (makes an mmm function that uses it) Jan 28 21:25:00 digitalspaghetti: just use a linearlayout and set weight=1 to each child Jan 28 21:25:23 TableRow should support weight too Jan 28 21:25:45 romainguy: try "adb shell setprop dalvik.vm.dexopt-flags v=n" with today's build Jan 28 21:25:58 romainguy: what is stop/start btw? Jan 28 21:26:00 \o/ Jan 28 21:26:02 its on the shell? Jan 28 21:26:12 jasta: yes, it stops/starts the Android runtime Jan 28 21:26:18 oh very convenient Jan 28 21:26:20 it's a bit faster than rebooting the phone entirely Jan 28 21:26:29 thanks romain Jan 28 21:26:33 note that when you change framework.jar, the system will dexopt every apk Jan 28 21:26:44 hence fadden's command above :)) Jan 28 21:27:10 fadden: testing Jan 28 21:28:41 Seems to cut the dexopt-ing for core + apps to about 30 seconds Jan 28 21:29:03 wow indeed Jan 28 21:29:05 much much better Jan 28 21:29:39 Just note that you're running without verification and dex optimization. If you do something screwy, I reserve the right to crash. :-) Jan 28 21:30:24 I'm fine with it :)) Jan 28 21:30:25 Most of the time spent now is just unzipping jar -> dex. Jan 28 21:31:25 We've got about 20MB of stuff after uncompressing. Jan 28 21:39:57 * jasta ponders Jan 28 21:42:34 romainguy: do i have to explicitly clean or something? touching any of these java files doesn't cause make framework to respond. Jan 28 21:42:46 what files did you touch? Jan 28 21:43:02 in this case, base/services/java/com/android/server/status/StatusBarService.java Jan 28 21:43:13 ah then you can't do make framework for this Jan 28 21:43:20 oh, those aren't in the framework Jan 28 21:43:27 those are servers :) Jan 28 21:43:36 for these you do mmm frameworks/base/services/ snod Jan 28 21:43:53 and then adb shell stop && adb sync && adb shell start Jan 28 21:43:57 welcome to the wonderful world of Android teminology. we have servers, services, framework, platform, providers... Jan 28 21:44:03 ah ok. Jan 28 21:44:19 i had thought because the folder itself was "frameworks" that make framework was some uber one that caught all of core, services, etc. Jan 28 21:44:22 my fault :) Jan 28 21:44:40 that said Jan 28 21:44:47 you can modify the fsnod above to have additional makefiles Jan 28 21:44:53 including services, etc. Jan 28 21:45:26 the "framework" as that term is typically used is "the SDK implementation library" Jan 28 21:45:29 more or less Jan 28 21:45:44 ok, i think i understand how this works. thanks. Jan 28 21:45:44 ... as "statically" linked, right? Jan 28 21:46:06 (i.e. if it IPC's, it's not the framework, right?) Jan 28 21:46:07 except that in the sources it's grouped together in the same tree/package as the system-side servers that implement that end of things Jan 28 21:46:36 jbq: sort of, right. the binder IPCs are emphatically not part of the fixed API. Jan 28 21:46:58 they can and will change from one release to another Jan 28 21:47:27 Yeah, but the "other" side of a binder IPC that implements an SDK API (e.g. notifications) isn't considered part of the framework either, right? Jan 28 21:48:47 so, you guys don't use make sdk for your normal work, right? Jan 28 21:49:03 you use the ANDROID_PRODUCT_OUT stuff? Jan 28 21:49:14 for those of us not working on the SDK itself, we don't use make sdk, no. Jan 28 21:49:21 jasta: correct Jan 28 21:49:32 we just source build/envsetup.sh Jan 28 21:49:40 and use choosecombo? Jan 28 21:49:42 do a choosecombo or lunch Jan 28 21:49:48 ok, cool. Jan 28 21:49:52 and just use make/m/mm/mmm after that Jan 28 21:50:09 ok, ive been lookign through envsetup.sh Jan 28 21:51:07 why is there a special make target for framework btw? the others seem to be buildable using just ONE_SHOT_MAKEFILE and make? Jan 28 21:54:28 convenience, presumably Jan 28 21:55:01 also iirc in the old build system you could do "make framework" and it would build android.jar + all the servers, or some such package of work. Jan 28 21:55:28 what are the various products? emulator, generic, sim? Jan 28 21:56:01 right, stuff like that Jan 28 21:56:21 different set of binaries (and other files) being built. The difference becomes more meaningful if you think that e.g. the US G1 has the amazon MP3 app but the ADP1 doesn't. Jan 28 21:56:32 yes but what are they? why does generic seem to be what you use to build emulator images? and what is sim? Jan 28 21:56:40 jbq: oh, that makes way more sense now Jan 28 21:57:31 sim = run most of android mostly in a single process on a linux host. Not everything can run like that, so there's a product for it that builds less. Jan 28 21:59:01 ctate: the old system also need to do some weird stuff with the framework resources Jan 28 21:59:08 ah right Jan 28 21:59:10 whee :) Jan 28 21:59:25 ah, those framework guys, thinking they're special ;-) Jan 28 21:59:40 we just think that .apk files are boring :) Jan 28 21:59:58 we just love to spend 4 minutes waiting for our changes to appear on screen Jan 28 22:00:04 ... at least. Jan 28 22:00:07 :)) Jan 28 22:00:21 ctate: speaking of which, have you seen fadden's fancy new prop? Jan 28 22:00:27 i have not! Jan 28 22:00:51 fadden: romainguy: try "adb shell setprop dalvik.vm.dexopt-flags v=n" with today's build << makes a framework change must faster Jan 28 22:01:03 oh yeah, turning off dexpot Jan 28 22:01:11 Caveat coder Jan 28 22:01:23 dexpot? sheesh. dexOPT. Jan 28 22:01:36 * ctate cues up Tool, "The Pot" Jan 28 22:09:32 in order to use a given intent (such as Intent.ACTION_SEND), do I need to do anything in AndroidManifest (or anywhere else)? Jan 28 22:15:37 not normally, no. Jan 28 22:16:44 that's what I thought Jan 28 22:17:23 android.content.ActivityNotFoundException: No Activity found to handle Intent { action=android.intent.action.SENDTO type=text/plain (has extras) } Jan 28 22:17:35 I thought Gmail handled that Jan 28 22:18:00 you can list everything that has an intent receiver... Jan 28 22:18:16 one of the activity dumpsys targets does it Jan 28 22:18:29 I'll poke around Jan 28 22:19:06 adb shell dumpsys activity.broadcasts Jan 28 22:21:12 for the future, how do you get the list of what you can do in `adb shell`? Jan 28 22:21:23 `adb help shell` doesn't give me anything new Jan 28 22:21:37 what? Jan 28 22:22:30 oh, nevermind Jan 28 22:22:30 haha Jan 28 22:22:31 duh Jan 28 22:27:04 romainguy: if you edit framework core resources, the build system doesn't seem to care and rebuild the framework.jar? Jan 28 22:27:31 mmm frameworks/base/core/res/ Jan 28 22:27:43 ahh. Jan 28 22:27:49 we used to have a framework-res target but it's gone in cupcake for some reason Jan 28 22:28:12 i just need to pay closer attention to where Android.mk files are :) Jan 28 22:32:14 i want to have a ZoomControls displayed over my custom view. Would you suggest using an AbsoluteLayout to do the layering? Jan 28 22:32:50 no Jan 28 22:32:55 don't use AsboluteLayout Jan 28 22:32:59 FrameLayout will do just fine Jan 28 22:33:34 i thought a framelayout only could hold one child? Jan 28 22:33:48 nope Jan 28 22:33:54 oh, i see Jan 28 22:34:02 what would be the point? :) Jan 28 22:34:16 hard to say Jan 28 22:39:02 yeah, I thought FrameLayout could only hold one kid for a while too... Jan 28 22:39:25 maybe because the first line in the docs is "FrameLayout is designed to block out an area on the screen to display a single item. " Jan 28 22:39:42 stupid doc Jan 28 22:40:10 hehe, the next sentence starts "You can add multiple children to a FrameLayout" Jan 28 22:40:17 here you go Jan 28 22:40:22 now stop complaining and code correctly :) Jan 28 22:40:38 yeah, short attention span...read the first sentence of each doc Jan 28 23:08:39 damn, seems like you can either specify an email to send to, or an attachment, but not both. (ACTION_SEND / ACTION_SENDTO) Jan 28 23:08:40 weak Jan 28 23:22:09 wow, just discovered katapult...it's like quicksilver on mac, for linux... Jan 28 23:22:13 pretty cool Jan 28 23:30:07 is there a way to have repeating views in a LinearLayout/ScrollView? Almost like a static ListView that will just expand to be taller than it's parent instead of being scollable? Jan 28 23:36:00 man i like a lot of the new animations in cupcake Jan 28 23:36:43 the activity transition? Jan 28 23:38:03 I like the improved Japanese support; but am looking forward to having an IME. Jan 28 23:38:15 but yeah the animations are slick. Jan 28 23:39:36 jasta: you have it on a handset? Jan 28 23:44:35 * jasta looks around Jan 28 23:44:54 actually, no, i don't. Jan 28 23:44:56 hehe Jan 28 23:45:09 btw. what is the deal with the eclipse plugins only building on Linux? Jan 28 23:45:17 normally i'd be ok with that but i have to use a macbook at work Jan 28 23:45:32 I doubt it's the case Jan 28 23:45:40 the SDK guys work on Mac OS X and Windows Jan 28 23:47:16 no, it definitely is the case... Jan 28 23:47:39 you can build the plugins from eclipse, but the actual build system thats in here tells you it needs linux Jan 28 23:47:46 there are some scripts to automate it all Jan 28 23:48:10 then build the plugins from eclipse Jan 28 23:48:13 or ask xavd Jan 28 23:48:33 my teammates here are telling me that it's broken for some reason. hmm. Jan 28 23:48:43 what's broken? Jan 28 23:48:52 no idea, i'll try it myself first. Jan 28 23:49:14 it depends a lot on properly installing a specific version of Eclipse to build (and actually running it once manually so that it setup its workspace Jan 28 23:49:22 it's far from perfect really :( Jan 28 23:49:47 there should be comments in the scripts that explain more Jan 28 23:49:56 there are, but even those need lots of work Jan 28 23:50:07 that's possible :) Jan 28 23:50:26 so is the ideal way to build this using those scripts on a linux host? Jan 28 23:50:44 well that's what our build server do Jan 28 23:54:19 there seems to be some automated way of setting this up Jan 28 23:54:38 jasta: you should look at http://android.git.kernel.org/?p=platform/development.git;a=blob;f=tools/eclipse/scripts/build_plugins.sh;h=5f94ca060e986d6957266940a9121a2fca449766;hb=HEAD Jan 28 23:54:41 * jasta tinkers Jan 28 23:54:52 i had been looking at that Jan 28 23:55:00 i think setuP_eclipse.sh is what im after. Jan 28 23:55:18 this seems to be a cleaner way to do it Jan 28 23:55:18 oh yeah that one too Jan 28 23:56:36 is there a changelog for the 0.9.0 version? Jan 28 23:56:52 yes, changes.txt. nm :) Jan 28 23:57:14 you should look at the top level calls in build_server.sh too though. you can't just call build_plugins because the plugins depends on some other jars that need to be built first Jan 28 23:59:03 yeah, i am following a google thread which states to use build_server.sh Jan 29 00:29:55 sigh. Jan 29 00:30:04 my SDK build must be broken. Jan 29 00:53:23 romainguy: when you guys debug against the framework, do you just use DDMS manually and avoid the android plugin? Jan 29 00:53:50 it depends Jan 29 00:53:58 if you use Eclipse you can use the plugin's DDMS Jan 29 00:54:16 I use IntelliJ and therefore DDMS directly Jan 29 00:54:44 how do you even debug against the framework? is it in android.process.acore? Jan 29 00:55:08 no, the framework is in all processes that use the framework Jan 29 00:55:18 wait, i'm being dense. yes, i just realized that it's just a library. Jan 29 00:55:22 :)) Jan 29 00:55:29 jasta: the eclipse plugin will never really work for framework development. it relies on custom projects which only work with the SDK Jan 29 00:55:31 now, services are different Jan 29 00:55:34 what i meant was the android system services, though. Jan 29 00:55:41 like the StatusBarService Jan 29 00:55:45 then the system process Jan 29 00:56:00 ok. Jan 29 00:56:09 acore is Launcher + Contacts + a couple of providers Jan 29 00:58:08 thanks, youve been a big help today :) Jan 29 01:26:24 Would anyone know why my transparent .pngs are showing up with white backgrounds? Jan 29 01:39:27 marklar, i had that happen when loading the pngs via BitmapFactory, while at the same time using strange BitmapFactory.Options. (I wound up using null options to fix it) Jan 29 01:46:11 hm... this is just image views set in xml Jan 29 01:46:32 I found something saying it may be the bit depth, but I tried them with 24bit depth and same thing **** ENDING LOGGING AT Thu Jan 29 02:59:57 2009