**** BEGIN LOGGING AT Mon Dec 15 02:59:57 2008 Dec 15 04:04:41 In the Google Market, what determines the category ("Lifestyle", "Entertainment") that an app is filed under? Does the developer get to pick? (I am updating the Android FAQ: http://android-dls.com/wiki/index.php?title=Android_FAQ ) Dec 15 04:06:07 bgupta_: yeah Dec 15 04:06:18 developer selects it upon upload Dec 15 04:06:49 Thanks! Dec 15 04:07:50 Added to Market section of FAQ: http://android-dls.com/wiki/index.php?title=Android_FAQ#Market Dec 15 05:37:34 with android apps that i install from the app store, are they "all inclusive" such that if i were to uninstall one from the app manager it won't leave behind any random dependencies or libraries? Dec 15 06:00:51 can anyone help with debugging an AT&T data connection? Dec 15 06:05:15 bparker06: You'd prolly be better off on #android I think. Dec 15 06:06:50 bgupta_: I know what the settings are... I doubt they can help Dec 15 06:06:55 and I've been trying in there for days Dec 15 06:06:57 (and in here) Dec 15 06:07:09 it just doesn't work Dec 15 06:07:19 and I tried posting on the groups Dec 15 06:07:26 all I got was "call at&t" Dec 15 06:07:26 Mmm.. what settings are you using? Dec 15 06:07:29 for APN? Dec 15 06:07:30 which, they won't help Dec 15 06:07:33 because it's not their phone Dec 15 06:07:39 and they don't sell it Dec 15 06:07:47 bgupta_: wap.cingular Dec 15 06:07:48 APN is the main thing you have to set.. Dec 15 06:07:58 trust me I've tried every combination of setting possible Dec 15 06:07:59 10 times over Dec 15 06:08:11 something else is wrong Dec 15 06:09:17 K.. damn, I should be able to borrow an AT&T SIM on Tuesday... if you haven't figured it out by then.. I could have sworn I've seen instructions posted... did you find any instructions? Dec 15 06:09:35 lots Dec 15 06:09:39 and I've tried them all Dec 15 06:09:50 I'm not an moron I know how to set AT&T apn lol Dec 15 06:09:50 it just doesn't work Dec 15 06:10:01 s/an/a/ Dec 15 06:11:18 I assumed it would just work... WOndering if something is up with your phone.. What phone did you have before on AT&T? Dec 15 06:13:15 lots Dec 15 06:13:20 WOndering if you need the ADP1 firmware to do it? IE: Is there something in the T-Mobile firmware that is hosing data connection. Dec 15 06:13:20 heh Dec 15 06:13:30 I have the dev phone Dec 15 06:13:59 How did you get teh unlock code BTW? From T-Mobile? Dec 15 06:14:09 the dev phone is already unlocked Dec 15 06:14:27 oh really. And the dev phone isn't working on AT&T data plan? Wow.. that's a bummer.. Dec 15 06:14:31 yea Dec 15 06:14:40 actually at first it wasn't working at all Dec 15 06:14:46 and I had to go to at&t and get a new sim Dec 15 06:14:50 then it would actually register to the network Dec 15 06:14:53 but no data Dec 15 06:15:06 I have one also... (And an unlocked/rooted G1) my buddy has AT&T, but I won't see him till tuesday.. Dec 15 06:16:53 WHen you go to the APN's section does it show the AT&T APN list? Dec 15 06:17:19 the one I created yes Dec 15 06:18:52 Mmm. powering up my ADP1.. checking a few things. Dec 15 06:19:49 My sense is that the device should be able to query your network for a list of APNs...(At least that's what it does for T-Mobile) Dec 15 06:20:08 I've never heard of that... Dec 15 06:20:28 and all phones 've used in the past (probably 100+) do not have that option Dec 15 06:20:32 I've used* Dec 15 06:21:42 Well on my Nokia 9300 I was able to do so, as well as the ADP1... I didn't have to configure anything for either device... (And neither are T-Mobile phones).. They just popped up with a list of known data networks.. in the APN list. and somehow poppulated them Dec 15 06:22:05 NOt sure it it is a T-Mobiel feature that AT&T doesn't support. Dec 15 06:25:11 Did you watch this video? http://www.youtube.com/watch?v=Y28EnYDOHrU Dec 15 06:25:26 yep I did that already Dec 15 06:25:27 exactly Dec 15 06:25:38 watching it now Dec 15 06:29:02 Mmm. I'm gonna have to play with it, and see if I can replicate the problem.. Wont be able to borrow a SIM till Tuesday unfortunately. Dec 15 06:29:48 One question, can you take the existing/new SIM and get it to work in another phone? Dec 15 06:29:55 yep Dec 15 06:31:02 Ok so then theoretically, if I can get it working, you should be able to as well.. I'll take notes... Dec 15 06:31:16 SOrry I can't solve it right now Dec 15 06:31:55 WOndering if the ADP1 has any support? Dec 15 06:33:16 it should Dec 15 06:33:38 IE: Can you call Google. Dec 15 06:38:31 how lol Dec 15 06:38:52 not like they have a phone number posted anywhere Dec 15 06:47:06 Right. Dec 15 11:44:49 is there an easy way to disable a button? Dec 15 11:53:02 jammr, setEnabled(false)? just guessing Dec 15 11:53:04 jammr: button.setEnabled(false) or android:enabled="false" Dec 15 11:54:36 Thanks, I searched the docs and got nothing :/ Dec 15 11:55:21 jammr, true, if you search for `disable' on the Button doc it's nil :) Dec 15 11:56:41 ahh, finally. thanks again Dec 15 13:39:13 Anybody ever successfully used a CursorWrapper? Dec 15 13:40:59 The behaviour is not as documented, or at least ambiguous Dec 15 13:41:52 Also, I'm seeing some random behaviour, I think it's concurrency/threading related Dec 15 13:49:11 i havent used it Dec 15 13:49:16 i dont get it though Dec 15 13:49:41 whats it used for? Dec 15 13:49:57 it looks almost identical to cursor Dec 15 13:50:53 it's for wrapping a cursor Dec 15 13:51:08 though I'm using it to make a (limited) join possible on cursors Dec 15 13:51:12 more like filtering Dec 15 13:52:20 a first problem that I have with it is that the docs say "This method will return false if the cursor is already past the last entry in the result set." for moveToNext()... Dec 15 13:52:42 how is that an issue? Dec 15 13:52:47 while it doesn't behave that way Dec 15 13:53:23 CursorWrapper just delegates the calls to the underlying Cursor Dec 15 13:53:28 it returns false if the cursor is already on the last entry Dec 15 13:53:42 ah Dec 15 13:53:47 then file a bug against the documentation Dec 15 13:54:19 sure, I will Dec 15 14:03:05 it's filed... **** BEGIN LOGGING AT Mon Dec 15 14:38:29 2008 Dec 15 15:32:42 has anyone successfully implemented a custom filter to an ArrayAdapter? Dec 15 15:34:09 eldenz: ummm, successfully vs properly hmmm Dec 15 15:34:47 =) any hints? Dec 15 15:34:50 ive played with filtering an arrayadapter Dec 15 15:34:59 i thought i just had to override `public Filter.FilterResults performFiltering(CharSequence constraint)` Dec 15 15:35:07 but `android.widget.Filter.FilterResults` seems to be not visible inside my ArrayAdapter class Dec 15 15:35:40 i had no problem with FilterResults hmm Dec 15 15:36:17 i have the filter as an inner class to a cusotm arrayadapter Dec 15 15:36:42 ah Dec 15 15:37:15 and then override getFilter() to return my filter i assume Dec 15 15:37:38 yeah Dec 15 15:37:47 i was having trouble Dec 15 15:38:02 and i found a convoluted way from someone who seemed to be having similar trouble Dec 15 15:38:03 so i used it Dec 15 15:39:24 why don't you simply look at the source code of ArrayAdapter to see how it's done? Dec 15 15:39:31 lazyness Dec 15 15:39:35 :) Dec 15 15:39:41 no stupidity ;) Dec 15 15:39:54 i recently reformatted my macbook Dec 15 15:40:00 on my part :p didn't think about that :x Dec 15 15:40:03 and never copied the source back Dec 15 15:40:09 blau-MikeDG: you can browse the source online Dec 15 15:40:20 yeah, but its a pain to search it Dec 15 15:40:28 unless you know better than i do? Dec 15 15:40:31 Dumb GPS question Dec 15 15:40:41 http://android.git.kernel.org/ right? Dec 15 15:40:48 blau-MikeDG: I would think so I wrote the filtering API :0 Dec 15 15:40:51 on the LocationListener interface, will onStatusChanged be called when the device loses a GPS fix? Dec 15 15:40:52 is there any search at that level? Dec 15 15:41:17 blau-MikeDG, yeah i also have trouble finding things :) Dec 15 15:41:51 http://android.git.kernel.org/?p=platform/frameworks/base.git;a=blob;f=core/java/android/widget/ArrayAdapter.java;h=fe50a01b07a5159437ed0fca9ec3b77e0911ebb9;hb=HEAD Dec 15 15:42:05 thanks romainguy Dec 15 15:42:25 beat me :p Dec 15 15:43:07 So, anyone? Dec 15 15:43:18 or do GPS problems disappear into the aether of Android? Dec 15 15:43:19 eldenz: there are mostly two directories to know about Dec 15 15:43:30 frameworks/base/core contains the framework and UI toolkit + all resources Dec 15 15:43:35 packages/ contains the apps Dec 15 15:44:40 ok thanks Dec 15 15:44:44 i didnt make the frameworks/base/c connection thanks Dec 15 15:47:19 ugh of course amazon only has one of the oven mitts i need Dec 15 15:47:34 stupid java question, i've read that you can change visibility of methods in inherited classes.. does this also count for attributes? like if i redeclare a (private in parent) attribute, can i access the same contents as well? or will it create a new variable? Dec 15 15:48:07 you can only change the visibility of non-private methods Dec 15 15:48:13 attributes cannot be overriden, so no Dec 15 15:48:22 (and you can only change the visibility to make it more public, not less) Dec 15 15:48:43 ah okay, so i can't use your mOriginalValues ;| Dec 15 15:48:51 nope Dec 15 16:08:41 my list somehow does not update... http://pastie.org/339661 Dec 15 16:08:55 the log shows that things were filtered, one item passed Dec 15 16:09:37 i wound up actually updating the arraylist that the adapter was on Dec 15 16:09:45 im sure thats wrong though :/ Dec 15 16:10:07 wel Dec 15 16:10:09 well Dec 15 16:10:09 like, i do the filter, send the filtered stuff back, but basically have the original arraylist mirror the filter Dec 15 16:10:15 you forgot one big thing in publishResults Dec 15 16:10:22 you forgot to actually set the results on the adapter Dec 15 16:10:28 now that i think about it... yeah Dec 15 16:10:34 look at the example I linked to Dec 15 16:10:36 there's a mObjects = (List) results.values; Dec 15 16:10:42 i call adapter.getfilter().filter(); and it doens' ttalk to the adapter Dec 15 16:10:56 it's your publishResults() implementation that's wrong Dec 15 16:11:45 so the only way is to call clear(); and then add each item again? :x Dec 15 16:12:09 pretty much Dec 15 16:12:37 ArrayAdapter::setItems() would be nice :) Dec 15 16:13:36 for f in `adb shell ls /system/lib`; do echo adb pull /system/lib/$f /cygdrive/c/linuxShare/android_windows/system/lib/$f; done Dec 15 16:13:39 what is wrong with this Dec 15 16:13:42 : ( Dec 15 16:13:51 ' does not existsystem/lib/libpixelflinger.so Dec 15 16:13:51 ' does not existsystem/lib/libagl.so Dec 15 16:13:52 ' does not existsystem/lib/libnativehelper.so Dec 15 16:14:02 Delightful! Dec 15 17:31:52 vol: you can "adb pull" a directory Dec 15 17:32:02 fadden: I know. Dec 15 17:32:14 The problem is, adb pull/push on a long list of items has a tendancy to just die in the middle Dec 15 17:32:18 it will just stop and...... nothing. Dec 15 17:32:25 so, I have to ctrl-c it Dec 15 17:32:29 and that makes it stop midway Dec 15 17:32:46 by specifying a list of files, I can hit ctrl-c, note which file died, and my script continues to the next file Dec 15 17:32:57 then, when finished, I can just type by hand adb pull /system/lib/foo.so Dec 15 17:33:24 speaking of which, is the fact that adb just kind of craps out in the middle of file transfers a known issue? Dec 15 17:33:34 I don't think so. Dec 15 17:33:53 it only seems to happen whenever I'm doing an entire directory, or specify a list of files. Dec 15 17:34:16 and by "specify a list of files" I mean "use a shell script to do adb push/pull over a lot of files" Dec 15 17:35:01 Interesting... your "for" loop maps to "adb pull /system/lib/libsqlite.so^M ./libsqlite.so" Dec 15 17:35:10 The "ls" is coming back with a ctrl-M that isn't getting stripped. Dec 15 17:35:17 fadden: I discovered that the other day :( Dec 15 17:36:05 yeah, my "solution" is to do adb shell ls /system/lib, copy the list from the console into a text file, do a regex find/replace to insert "adb pull /system/lib/\1 ." and then pasting that into the console Dec 15 17:36:25 as I said, it will occasionally go braindead and I have to hit ctrl-c, but it works on most of the files, and I can handle typing in a few names by hand Dec 15 17:36:47 it is, however, pretty annoying to go through all that trouble though :\ Dec 15 17:38:15 for f in `adb shell ls /system/lib | tr '\015' ' '` ; do ... Dec 15 17:38:52 looks like that would work. Dec 15 17:41:25 Actually, while I have either of you here, do either of you know the answer to my question in #android? Dec 15 17:41:52 Is there a programatic equivolent to the user pressing the "home" key (or even returning from the activity to the previous activity) without pressing "finish" Dec 15 17:41:58 er Dec 15 17:42:01 without calling "finish" Dec 15 17:48:47 I have no idea. Dec 15 17:50:48 : ( Dec 15 19:31:22 why not all rectangles are drawn correctly? http://nopaste.org/p/audZRlCG0 Dec 15 19:48:40 for one thing, you're drawing way out of bounds, but that's not the issue Dec 15 19:49:18 for another, it's entirely possible you're hitting that "Array Out of Bounds" exception Dec 15 19:49:38 have you stepped through this with a debugger? Dec 15 19:59:24 no Dec 15 20:00:11 you donet mean LogCat right? Dec 15 20:00:40 fadden: adb shell ls /system/lib | while read a; do ... -- that is substantially faster and more reliable. Dec 15 20:02:34 oinkboink: are you using eclipse? Dec 15 20:02:41 try hitting the little "bug" in your menu buttons Dec 15 20:02:44 that's debug, instead of run Dec 15 20:02:48 remember to set breakpoints! Dec 15 20:03:07 jasta: where were you hours ago : ( Dec 15 20:03:22 well, is there an better way to draw all already clicked rectangles than using that for-loop? Dec 15 20:03:36 vol: telling my current company I QUIT :) Dec 15 20:03:44 wooooo Dec 15 20:04:11 haha Dec 15 20:04:15 you got the job with t-mobile? Dec 15 20:04:28 jasta: Oh, did you?! Dec 15 20:04:32 * jasta nods Dec 15 20:04:34 jasta: NICE! Dec 15 20:04:35 congratsadoodle Dec 15 20:04:36 Heaven help us. :-) Dec 15 20:04:41 fadden: lol Dec 15 20:04:44 i want a job with google Dec 15 20:04:44 jasta: plz 2 send unlock codes? Dec 15 20:04:53 jasta: Yeah you better make with some unlock codez :P Dec 15 20:05:12 they have cool perks for mobile devices... Dec 15 20:05:19 jasta: I would assume. Dec 15 20:05:26 50% off any WinMo phone! Dec 15 20:05:26 i can get 3 lines of service with 2000 minutes, unlimited data, etc for $10/mo Dec 15 20:05:30 each Dec 15 20:05:32 wow Dec 15 20:05:41 jasta, grats :) what's the job title? Dec 15 20:05:48 yeah, phone companies really take care of their employees that way Dec 15 20:06:10 eldenz: they apparently don't like to give fancy titles, so it's just Software Engineer, Mobile officially Dec 15 20:06:23 but internally, its at what is their 3rd level Dec 15 20:06:29 Should dhcpcd always be using around 3% CPU? Dec 15 20:06:32 cool Dec 15 20:06:52 savid: no Dec 15 20:06:56 jasta: where at? Dec 15 20:06:58 vol, thats just for viewing the exceptions? Dec 15 20:07:01 location wise Dec 15 20:07:02 running "top" in a shell shows 2% to 3% usage by dhcpcd. Seems kinda high for a dhcp client Dec 15 20:07:19 oinkboink: no. Do you know what debugging is? Dec 15 20:07:23 blau-MikeDG: T-Mobile Dec 15 20:07:36 the idea is that you can step through the code, line by line. Dec 15 20:07:47 and see what each value is, and what the program is doing Dec 15 20:07:57 look up a quick tutorial on "eclipse debugging" Dec 15 20:08:12 ok Dec 15 20:09:09 you might want to make a simple, non-android java app to test this out with Dec 15 20:12:39 Also, system_server is consistently using 5% to 7% cpu. Is that normal? Dec 15 20:13:34 savid: does "adb logcat" show anything interesting? Dec 15 20:13:43 Like constant activity from some part of the system. Dec 15 20:15:50 vol: from #android, why do you feel the need to inject key events...? Dec 15 20:15:52 fadden, seems like a lot of activity from D/RC_WifiService and I/dhcpcd, but hard to tell how frequent as the log doesn't show times Dec 15 20:16:06 ctate: because I want to stop an app without destroying it Dec 15 20:16:21 basically, I want to be able to programatically "minimize" the app Dec 15 20:16:30 savid: adb logcat -v time Dec 15 20:16:36 vol: is it your own app? ;) Dec 15 20:16:39 yes Dec 15 20:16:59 ah, thx Dec 15 20:17:04 okay... hm. so you want to simulate the user hitting ... what? Back? Dec 15 20:17:07 home Dec 15 20:17:13 back I guess would work Dec 15 20:17:17 just start the Home activity Dec 15 20:17:18 fadden, yeah, definately frequent Dec 15 20:17:22 mm. Dec 15 20:17:23 and let the framework manage the lifecycle Dec 15 20:17:27 that might be simpler. Dec 15 20:17:29 (please) :) Dec 15 20:17:30 * savid can't spell definitely Dec 15 20:17:33 savid: can you pastebin that Dec 15 20:17:36 thank you! Dec 15 20:17:39 Sure Dec 15 20:17:40 oh no prob Dec 15 20:17:50 I only did the keypress injection because Disconnect was the only one to reply, and he handed me a perl script that did keystroke injection :\ Dec 15 20:17:57 vol: mostly, this is a lot less likely to cause havoc with history tracking/management Dec 15 20:18:05 that's the issue that sprang to mind when you said it Dec 15 20:18:10 ha ha Dec 15 20:18:12 possibly : ) Dec 15 20:18:16 savid: a full "adb bugreport" would potentially be useful, though you may want to pick through that for any personal info (e.g. numbers you've called) Dec 15 20:19:34 fadden, here's the log http://pastebin.com/m438a1a9f Dec 15 20:19:44 fadden, that's about 10 min. worth Dec 15 20:25:36 ctate: yes, that does it, and it's significantly simpler too (no need for an extra thread) Dec 15 20:25:44 thank you so much for suggesting this : ) Dec 15 20:25:45 voila! Dec 15 20:25:49 you're most welcome Dec 15 20:26:03 I would have stuck with this cruddy method otherwise Dec 15 20:26:07 eeep Dec 15 20:26:29 the sending of keypresses via instrumentation on another thread that exists just to take these keypresses Dec 15 20:26:30 if things start to feel kludgy, it's a pretty good sign that there's probably a better way. :) Dec 15 20:26:38 well, I asked : ( Dec 15 20:26:41 and I got blank stares Dec 15 20:26:58 yeah. the mailing lists are probably higher signal-to-noise for that kind of question, but slower. Dec 15 20:27:03 that's true Dec 15 20:27:38 savid: not sure what to make of the log, but I've forwarded to someone who hopefully does. Dec 15 21:39:56 something I've been wondering about for a long time now: what is the meaning of the boolean mNoteOnly in the (full) NotePad application example? Dec 15 21:40:09 in the NoteEditor class Dec 15 21:40:53 it's never true, always false Dec 15 21:41:08 :-D Dec 15 21:41:14 probably does nothing, someone just forgot about it Dec 15 21:41:32 I agree Dec 15 21:41:43 but my curiosity is aroused, what could it have been for? Dec 15 21:41:50 multimedia notes Dec 15 21:41:59 root access Dec 15 21:42:15 you mean musical notes? Dec 15 21:42:44 as in "do a deer, a female deer" ;-)? Dec 15 22:37:52 I meant more like "embedded photos" **** ENDING LOGGING AT Tue Dec 16 02:59:57 2008