**** BEGIN LOGGING AT Sun Jan 25 02:59:58 2009
Jan 25 03:00:03 <KNY> there aren't any differences in the system calls, though
Jan 25 03:00:09 <KNY> let me verify that with a diff, hold on
Jan 25 03:00:41 <Gary|tp> KNY: ... usb debugging is on?
Jan 25 03:00:43 <Gary|tp> lol
Jan 25 03:00:48 <CardinalFang> :)
Jan 25 03:00:50 <KNY> Gary|tp, haha yes
Jan 25 03:01:09 <KNY> Gary|tp, the weird thing is that it sees emulator(s) just fine
Jan 25 03:04:51 <herriojr_> this may be a silly question, but if my application is granted read/write permissions to an android content provider, am I also given direct read/write access to the database (so when a field in my database references a field in the android content provider, I can make triggers to handle deletions properly)
Jan 25 03:05:42 <Gary|tp> KNY: did you do anything weird to the desktop lately? if you mount the sdcard, does it show up?
Jan 25 03:05:55 <Gary|tp> is it a 2.0 usb port?
Jan 25 03:06:11 <KNY> Gary|tp, other than rebooting (which is odd for me), no. I can mount the sdcard and read/write. Yes, it's usb 2.0
Jan 25 03:06:24 <Gary|tp> hmm
Jan 25 03:06:34 <KNY> yeah, I'm stumped
Jan 25 03:07:14 <Gary|tp> try unchecking and checking the usb debug box, reboot the phone?
Jan 25 03:07:27 <KNY> tried that, and that.
Jan 25 03:07:39 <KNY> and tried rebooting the PC (again)
Jan 25 03:07:49 <Gary|tp> I'd say it was a PC problem if it wouldn't mount the sdcard, but since it will that's odd.
Jan 25 03:07:57 <KNY> hmm, there's an r2 of the SDK out now
Jan 25 03:07:58 <KNY> let me update
Jan 25 03:08:14 <Gary|tp> shouldn't matter, nothing changed on the phone so old adb should still work
Jan 25 03:08:24 <Gary|tp> (unless you're using some internal version lol)
Jan 25 03:08:29 <solca> hmm didnt know about this channel...
Jan 25 03:08:31 <KNY> nope, stock RC30
Jan 25 03:09:40 <KNY> r2 didn't change anything :-/
Jan 25 03:10:42 <Gary|tp> whats in the rules file? you do run ibex, right? /etc/udev/rules.d/
Jan 25 03:11:09 <KNY> yeah, followed the instructions, but... SUBSYSTEM=="usb", SYSFS{idVendor}=="0bb4", MODE="0666"
Jan 25 03:11:39 <Gary|tp> Same as mine :/
Jan 25 03:11:46 <KNY> yeah
Jan 25 03:11:52 <KNY> that was the first thing I checked
Jan 25 03:12:08 <Gary|tp> does it show up under adb devices?
Jan 25 03:12:21 <KNY> no, that's the problem
Jan 25 03:12:26 <KNY> but the emulator does
Jan 25 03:13:19 <Gary|tp> hmm
Jan 25 03:15:26 <Gary|tp> I... don't know :/.
Jan 25 03:15:47 <KNY> yeah, me neither
Jan 25 03:15:50 <KNY> sure sucks, though
Jan 25 03:16:13 <KNY> I really don't want to do a factory reset but I'll try it if I can't get this working
Jan 25 03:16:44 <Gary|tp> if you do a factory reset, you really should switch to JFs mod :P
Jan 25 03:16:59 <KNY> yeah
Jan 25 03:17:13 <KNY> though I think I can sucker T-Mobile into letting me return this
Jan 25 03:17:17 <KNY> meaning I can buy an ADP1
Jan 25 03:17:21 <Gary|tp> haha
Jan 25 03:17:29 <KNY> and by "sucker" i mean I do have a somewhat valid reason
Jan 25 03:17:45 <KNY> if I open or close the screen with force and do it just right, the phone shuts odd
Jan 25 03:17:46 <KNY> off
Jan 25 03:17:49 <KNY> just, dies
Jan 25 03:18:04 <Gary|tp> I can see that going well. "yeah hey, adb refuses to work on this phone, I'd like to return it", "uhh what's adb?"
Jan 25 03:18:13 <Gary|tp> haha
Jan 25 03:18:19 <KNY> haha
Jan 25 03:18:21 <Gary|tp> that would work quite a bit better
Jan 25 03:18:51 <KNY> but it's annoying because sometimes (like, multiple times a day) I pull the phone out of my pocket and it's been dead for god knows how long
Jan 25 03:18:57 <KNY> the battery isn't dead, just the phone is off
Jan 25 03:19:01 <Gary|tp> hmm
Jan 25 03:19:05 <Gary|tp> they'd probably just replace it
Jan 25 03:19:14 <Gary|tp> unfortunately :(
Jan 25 03:19:21 <Gary|tp> but there's no real reason to switch to ADP1 anyway?
Jan 25 03:19:27 <KNY> yeah, that's what I'm counting on
Jan 25 03:19:49 <KNY> and I'll tell them to note that if it still has the problem, I'll return the exchange in 14 days because it's obviously a design defect
Jan 25 03:20:01 <KNY> get a refund, buy an adp1
Jan 25 03:20:17 <KNY> and there's a good reason: the adp1 has the sweet backplate
Jan 25 03:20:18 <KNY> haha
Jan 25 03:20:25 <Gary|tp> I mean, the only difference is it's unlocked and a cool backplate. Not worth the $200+, and no warranty
Jan 25 03:20:47 <Gary|tp> I mean even if you screw up your phone with root, as evil as it is t-mobile would replace it
Jan 25 03:20:54 <Gary|tp> with ADP you're kinda screwed
Jan 25 03:21:05 <KNY> true
Jan 25 03:21:23 <Gary|tp> well
Jan 25 03:21:29 <Gary|tp> fastboot kinda makes that moot anyway. nvm lol.
Jan 25 03:21:51 <KNY> haha
Jan 25 03:21:54 <Gary|tp> I suppose it's possible to destroy the bootloader, but you'd really have to TRY to do that.
Jan 25 03:22:21 <KNY> heh
Jan 25 03:22:31 * KNY sighs
Jan 25 03:22:34 <KNY> stupid adb
Jan 25 03:22:40 <KNY> or stupid phone, I don't know which
Jan 25 03:22:43 <Gary|tp> just get a G1, lol.
Jan 25 03:22:50 <Gary|tp> oh, adb.
Jan 25 03:22:53 <KNY> haha
Jan 25 03:23:01 <Gary|tp> I really don't know :/
Jan 25 03:23:02 <KNY> yeah, upside-down p
Jan 25 03:23:46 <KNY> let me try the VM I used to use to do android dev...
Jan 25 03:25:02 <tmzt> KNY: did you get any error messages in dmesg?
Jan 25 03:25:12 <KNY> tmzt, no
Jan 25 03:25:53 <KNY> tmzt, http://pastebin.ca/1317671
Jan 25 03:27:44 <tmzt> sde is your g1?
Jan 25 03:27:56 <tmzt> why is it mounting the sd card?
Jan 25 03:28:21 <KNY> it's not mounted
Jan 25 03:28:26 <kcoury> hey guys, I'm using setMultiChoiceItems with an AlertDialog Builder and I was wondering if there was anyway to get a reference to the actual ListView within the dialog after the dialog is created?
Jan 25 03:28:29 <tmzt> dmesg -c and try again
Jan 25 03:29:04 <KNY> tmzt, http://pastebin.ca/1317672
Jan 25 03:29:42 <KNY> that is goofy about sde
Jan 25 03:30:35 <tmzt> same thing
Jan 25 03:31:21 <KNY> yep
Jan 25 03:31:39 <KNY> er, wait, I pasted the wrong one
Jan 25 03:31:46 <KNY> though the right one is the same, just with different timestamps
Jan 25 03:32:17 <KNY> http://pastebin.ca/1317675
Jan 25 03:32:19 <KNY> there you go
Jan 25 03:35:37 <tmzt> it's not asking you if you want to mount the card when you plug the usb?
Jan 25 03:36:00 <KNY> tmzt, it shows the thing in the notifications bar and if I drag it down, I can choose to mount it or not
Jan 25 03:36:13 <KNY> this dmesg is doing nothing other than just plugging it in
Jan 25 03:36:28 <KNY> it's not actually mounted, though
Jan 25 03:36:49 <KNY> it doesn't show up under fdisk
Jan 25 03:38:33 <tmzt> don't mount it
Jan 25 03:38:39 <tmzt> what does fdisk -l /dev/sde show?
Jan 25 03:39:06 <KNY> I haven't mounted it
Jan 25 03:39:29 <KNY> kny@kny-desktop:~$ sudo fdisk -l /dev/sde
Jan 25 03:39:29 <KNY> kny@kny-desktop:~$
Jan 25 03:40:04 <tmzt> what kernel/version of android are you using?
Jan 25 03:40:30 <KNY> the SDK?
Jan 25 03:40:41 <tmzt> on the g1
Jan 25 03:40:44 <KNY> the SDK is what's on the site... 1.0_r2
Jan 25 03:40:49 <KNY> stock OTA RC30
Jan 25 03:41:03 <KNY> it worked literally last night, but not today
Jan 25 03:41:07 <KNY> with updates to neither side
Jan 25 03:41:55 <tmzt> lsusb by itself works, the device shows up there or not?
Jan 25 03:42:43 <KNY> Bus 002 Device 011: ID 0bb4:0c02 High Tech Computer Corp.
Jan 25 03:44:36 <tmzt> what user are you running adb (the client) as?
Jan 25 03:45:21 <KNY> user, like I always have
Jan 25 03:45:30 <KNY> (not sudo/root)
Jan 25 03:45:30 <tmzt> also ls -lh /dev/usbdevs.11_*
Jan 25 03:45:39 <tmzt> also ls -lh /dev/usbdev2.11_*
Jan 25 03:46:04 <KNY> ls: cannot access /dev/usbdevs.11_*: No such file or directory
Jan 25 03:46:21 <KNY> crw-rw---- 1 root root 253, 19 2009-01-24 22:36 /dev/usbdev2.11_ep00
Jan 25 03:46:37 <KNY> and that, but repeated for ep01, ep02, ep81, ep82
Jan 25 03:46:52 <tmzt> so you don't have the correct udev settings
Jan 25 03:47:19 <KNY> ?
Jan 25 03:47:38 <KNY> SUBSYSTEM=="usb", SYSFS{idVendor}=="0bb4", MODE="0666"
Jan 25 03:47:53 <SanMehat> hmmm.
Jan 25 03:48:07 <CardinalFang> Just chmod to see if it's working with permissions changed.
Jan 25 03:48:51 <KNY> you mean chmod a+rx /etc/udev/rules.d/50-android.rules?
Jan 25 03:48:52 <KNY> I tried that
Jan 25 03:49:04 <tmzt> SanMehat: should the usb storage device show up when not mounted in the pull-down on android?
Jan 25 03:49:09 <CardinalFang> No.  the device, siclly.
Jan 25 03:50:12 <SanMehat> tmzt: it will show up as a removable device with no media
Jan 25 03:50:12 <CardinalFang> chmod 666 /dev/usbdev2.11_ep00
Jan 25 03:50:41 <tmzt> SanMehat: so shouldn't fdisk -l say "No media present"?
Jan 25 03:50:55 <tmzt> ah, fdisk just ignores things, very annoying at times
Jan 25 03:51:03 <SanMehat> tmzt: that depends on fdisk i guess :)
Jan 25 03:51:36 <tmzt> yeah, I usually use xxd
Jan 25 03:51:53 <KNY> CardinalFang, no change after the chmod
Jan 25 03:52:08 <tmzt> chmod 666 /dev/usbdev*
Jan 25 03:52:14 <tmzt> you only have one endpoint
Jan 25 03:52:19 <CardinalFang> KNY, no change to "adb devices"?
Jan 25 03:52:34 <KNY> CardinalFang, tmzt, no change
Jan 25 03:52:44 <KNY> I did adb kill-server && adb devices
Jan 25 03:52:49 <tmzt> I think you also need chmod -R 666 /dev/bus/usb to allow ioctl's
Jan 25 03:53:53 <KNY> tmzt, no change
Jan 25 03:54:15 <tmzt> does adb server use a socket on the pc side?
Jan 25 03:54:28 <CardinalFang> KNY, strace should tell you of ioctl and file operations.
Jan 25 03:54:59 <CardinalFang> Sorry.  My only tool is that hammer.
Jan 25 03:55:07 <KNY> CardinalFang, haha
Jan 25 04:00:09 <KNY> tmzt, was that question for me?
Jan 25 04:00:52 <KNY> if so, I'm pretty sure, yeah.. socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
Jan 25 04:00:52 <KNY> connect(3, {sa_family=AF_INET, sin_port=htons(5037), sin_addr=inet_addr("127.0.0.1")}, 16) = 0
Jan 25 04:03:53 <tmzt> is adb server running?
Jan 25 04:04:02 <tmzt> nc 127.0.0.1 5037
Jan 25 04:06:10 <KNY> yes, it's running
Jan 25 04:06:24 <KNY> well, it's running in that I don't get "Connection refused"
Jan 25 04:09:42 <tmzt> I don't know the adb protocol
Jan 25 04:10:55 <KNY> I assumed it from strace
Jan 25 04:11:10 <KNY> and I get FAIL0012device offline (x)
Jan 25 04:11:36 <tmzt> what does it write to the adb server to get the list of devices?
Jan 25 04:11:56 <KNY> "000c" and then "host:devices"
Jan 25 04:12:37 <tmzt> 000c as bytes or a string?
Jan 25 04:13:31 <KNY> the strace log says write(3, "000c", 4)
Jan 25 04:13:34 <KNY> so, string?
Jan 25 04:14:27 <tmzt> yeah
Jan 25 04:14:43 <tmzt> so it's 000chost:devices like that?
Jan 25 04:15:11 <tmzt> try that on nc then
Jan 25 04:15:49 <KNY> I get back OKAY0000
Jan 25 04:16:02 <KNY> which is what strace says it's trying to read
Jan 25 04:16:22 <tmzt> strace won't tell you what your program is trying to read, only what it has read
Jan 25 04:16:31 <KNY> ah, okay
Jan 25 04:16:32 <KNY> sorry
Jan 25 04:16:45 <tmzt> I mean it won't tell you what the program expects, it will tell you if it failed to read and why
Jan 25 04:17:03 <tmzt> we need to strace the adb server though to see why it cannot access the usb device
Jan 25 04:17:22 <tmzt> does adb launch the server if it's not running?
Jan 25 04:17:23 <herriojr_> so, when I do a delete on a table in a content provider, do I need to also need to do a contentResolver.notifyChange(uri, null) for every table that gets cascaded changes?
Jan 25 04:17:42 <tmzt> strace -o adblog -ff adb devices
Jan 25 04:17:45 <herriojr_> or not just delete, but any modification
Jan 25 04:17:49 <tmzt> or list or whatever the command is
Jan 25 04:18:06 <KNY> it's adb devices
Jan 25 04:18:07 <KNY> hold on
Jan 25 04:19:36 <KNY> tmzt, how long should I let this run?
Jan 25 04:20:15 <tmzt> let it run
Jan 25 04:20:23 <tmzt> ls |grep adb
Jan 25 04:20:28 <tmzt> in another shell
Jan 25 04:20:28 <KNY> read(10, 0x8e5e6c4, 4080)               = -1 EAGAIN (Resource temporarily unavailable)
Jan 25 04:20:31 <KNY> interesting line
Jan 25 04:23:39 <CardinalFang> Yah.  maybe.
Jan 25 04:23:52 <CardinalFang> What is fd 10?
Jan 25 04:24:07 <tmzt> KNY: you didn't use -0 ?
Jan 25 04:24:08 <CardinalFang> (ls -l /dev/PID/fd/10
Jan 25 04:24:10 <tmzt> -o
Jan 25 04:24:25 <KNY> tmzt, I used -o, yes
Jan 25 04:24:36 <KNY> it made adblog.* files
Jan 25 04:24:37 <tmzt> CardinalFang: /proc?
Jan 25 04:24:37 <KNY> 4 of them
Jan 25 04:24:58 <KNY> tmzt, and I was tail-ing through one of them
Jan 25 04:25:05 <CardinalFang> Yep.  Sorry.
Jan 25 04:26:47 <KNY> CardinalFang, what did you want me to ls?
Jan 25 04:27:12 <tmzt> ls -l /proc/<pid>/fd/10 the same pid as the adblog file you tailed
Jan 25 04:27:37 <CardinalFang> KNY, I wanted to know what fd read()ing caused EAGAIN.
Jan 25 04:27:57 <CardinalFang> EAGAIN isn't an error, btw.  It's probably just asych IO.
Jan 25 04:27:59 <KNY> ls: cannot access /proc/10329/fd/10: No such file or directory
Jan 25 04:28:18 <tmzt> it must be the adb devices process which should have exited
Jan 25 04:28:26 <tmzt> we need the pid of the adb server process
Jan 25 04:28:32 <tmzt> grep fork adblog*
Jan 25 04:28:57 <KNY> tmzt, returns nothing
Jan 25 04:30:39 <tmzt> SanMehat: how does adb fork the server process?
Jan 25 04:31:38 <KNY> clone()?
Jan 25 04:32:17 <SanMehat> tmzt: have you looked at the code?
Jan 25 04:32:29 <tmzt> only to see what it did on device
Jan 25 04:33:45 <tmzt> what .git is it?
Jan 25 04:39:01 <KNY> tmzt, I think it's in system/core/adb
Jan 25 04:40:43 <tmzt> yeah, core.git
Jan 25 04:42:10 <tmzt> ADB_TRACE=usb adb devices
Jan 25 04:42:13 <tmzt> no strace
Jan 25 04:43:16 <KNY> * daemon not running. starting it now *
Jan 25 04:43:16 <KNY> ADB server didn't ACK
Jan 25 04:43:16 <KNY> * failed to start daemon *
Jan 25 04:43:16 <KNY> error: cannot connect to daemon
Jan 25 04:43:56 <KNY> if I empty out ADB_TRACE, it starts (or says it does) successfully
Jan 25 04:47:04 <tmzt> grep "usb located" adblog*
Jan 25 04:47:22 <KNY> nothing
Jan 25 04:49:31 <tmzt> it's not using libusb, it's doing everything directly with ioctl's in a very abstracted way (transports for instance)
Jan 25 04:49:59 <tmzt> it's not even clear how to debug it if ADB_TRACE doesn't work, and the forked server cannot be started in foreground
Jan 25 04:51:05 <tmzt> SanMehat: lsusb works as the same user, ADB_TRACE=usb causes the server process to fail (not ACK), do you have any ideas how to debug this?
Jan 25 04:51:34 <herriojr_> so, when I create a ContentProvider, and after an add/update/delete to a uri which may make modifications to other uri's via triggers, I also need to notify the observers associated with those other uris.  My question is that if I just notify the base uri, will any  listener on a specific row in a URI also get notified?
Jan 25 04:59:17 <KNY> tmzt, SanMehat, Gary|tp -- thanks for all your help, but I have to go to bed before I pass out. If you find anything else, /msg me or just highlight me and I'll see it in the morning. Thanks again!
Jan 25 05:58:22 <Gary|tp> KNY: .... IS THE PHONE ON?
Jan 25 05:58:32 <Gary|tp> Honestly, I dunno what's wrong.
Jan 25 06:13:01 <Newbie> Have anyone thought of Android on iPhone ??
Jan 25 06:13:47 <SanMehat> nope
Jan 25 06:13:50 <SanMehat> nobody has
Jan 25 06:14:16 <Newbie> SanMehat, Was that suppose to mean a Sarcasm??
Jan 25 06:14:48 <SanMehat> Newbie: why would you think such a thing?
Jan 25 06:15:24 <unix_infidel> Newbie: it's not viable with the current input model. And those that have thought about it probably are working on something that warrants their effort.
Jan 25 06:17:08 <Newbie> Hey ppl, but just think, there is a lot more iPhones there, so if Android can be brought out to that platform then it will be the greatest achievement, open source promotoers have faced in their entire life
Jan 25 06:17:48 <Newbie> Is there a law that states, iPhone cannot boot Andriod or anyother OS for that matter ??
Jan 25 06:17:59 <Gary|tp> Greatest achievement? Sorry, but atm iPhone's OS and programs > Android.
Jan 25 06:18:04 <tmzt> unix_infidel: why isn't it viable with the current input model, you mean lack of on-screen keyboard or something in the kernel?
Jan 25 06:18:34 <Gary|tp> tmzt: I'm sure it would be pretty fun getting the radio to work, on top of OSK.
Jan 25 06:18:55 <tmzt> but how is that specific to android?
Jan 25 06:19:09 <unix_infidel> Ya know I'm a firm believer in the fact that a projects reasonableness to the outside world shouldnt stop a bunch of hackers from getting together and doing the impossible / unreasonable things that need doin'
Jan 25 06:19:35 <unix_infidel> So if you feel like you have the sheer will to accomplish the "greatest achievment" in mobile history...more power to ya.
Jan 25 06:20:01 <Gary|tp> Well, it's not. But it's still something that's not possible atm.
Jan 25 06:20:26 <Newbie> Gary|tp, u mean it is not possible technically? or by law ??
Jan 25 06:20:41 <unix_infidel> project's*
Jan 25 06:23:51 <Newbie> I believe it can be the next big thing as it will really take out the slow nature of UI of G1 ..
Jan 25 06:24:32 <tmzt> Newbie: moving everything that isn't on gpu to gpu should help there shouldn't it?
Jan 25 06:24:51 <tmzt> at least everything that should be
Jan 25 06:28:02 <Newbie> tmzt, I dint get that ..
Jan 25 06:28:27 <tmzt> what's slow about UI on the G1
Jan 25 06:29:02 <Newbie> tmzt, Ohh fine ... now i get it !
Jan 25 06:29:26 <Newbie> Is there any effort or even a possibility of extending Android on poular old phones??
Jan 25 06:29:50 <Newbie> Like for example the Moto razr v3i ??
Jan 25 06:29:52 <tmzt> yeah, what phones?
Jan 25 06:30:00 <tmzt> not that one, too simple
Jan 25 06:30:31 <Newbie> I knw it is painfully slow with its native software but why can't we think of an extremly tripped down version of Android ?
Jan 25 06:30:47 <Newbie> *stripped
Jan 25 06:31:04 <tmzt> the radio features run on the same core as the ui on that phone (from what I know, I haven't studied that particular phone)
Jan 25 06:31:55 <Newbie> Oh ok .. but will it be very difficult ??
Jan 25 06:32:06 <Newbie> what do u think abt it ??
Jan 25 06:33:32 <tmzt> it would be very difficult, you would have to be able to build a complete real-time os with gsm stack as well as android, you really couldn't use the linux parts of android so you would have to replace them
Jan 25 06:34:18 <tmzt> there are other phones with android ports in progress, but they are all smartphones (with seperate radio cores)
Jan 25 06:36:31 <Newbie> tmzt, so its better to leave it .. :) ?
Jan 25 06:47:51 <romainguy> tmzt: Newbie: moving everything that isn't on gpu to gpu should help there shouldn't it? << not necessarily actually
Jan 25 06:47:56 <romainguy> what would help is a much faster memory bus
Jan 25 06:52:54 <tmzt> romainguy: the sgx on the iPhone isn't a significant improvement for ui rendering speed?
Jan 25 06:54:22 <romainguy> in the case of Android it would not make a huge difference
Jan 25 06:54:45 <romainguy> I actually built a prototype (that was available in m5 I think, or maybe 0.9) of our UI rendered entirely in OpenGL
Jan 25 06:55:03 <romainguy> it opens the door to a bunch of new issues that can actually defeat the gain you get by rendering on the GPU
Jan 25 06:55:15 <romainguy> the most interesting thing about using the GPU is that you offload the CPU for background apps
Jan 25 10:45:59 <f00bar80> ppl i'm asking if using xml for screen design is better, also i want to know if any description for how UI Elements are related to each other and in which way, i checked http://code.google.com/android/devel/ui/glossary.html but this is only a glossary, and if any sample code for both xml and normal way of using and implementing GUI, any comment ?
Jan 25 12:57:48 <digitalspaghetti> anyone any ideas why my ImageAdapter using the MediaStore isn't working: http://paste.ifies.org/284
Jan 25 12:58:14 <digitalspaghetti> when i try return the count of the cursor, it's null and when i pass a fake value the view works, but shows nothing
Jan 25 13:18:05 <TD___> maybe try moving your cursor instantiation above the g.setAdapter line?
Jan 25 13:27:02 <digitalspaghetti> hrrm, nope - the cursor is instansiated, but still returns null on return cursor.getCount();  :|
Jan 25 13:30:25 <digitalspaghetti> hrrm, i do keep seeing this in the catlog: 01-25 13:30:01.380: ERROR/mountd(30): KillProcessesWithOpenFiles /sdcard
Jan 25 13:31:55 <TD> getCount returns an int though right, so it can't return null. do you mean that cursor itself is null?
Jan 25 13:32:29 <digitalspaghetti> when i hover over, it says the cursor has been created
Jan 25 13:33:00 <TD> so what is the exception you see ?
Jan 25 13:34:25 <digitalspaghetti> hold on, need to reboot - the debugger won't shut up
Jan 25 13:35:32 <TD> i usually try and work without a debugger.
Jan 25 13:35:49 <TD> unfortunately i haven't figured out how to disable the eclipse debugger yet :( if i could, i would, it's not very helpful
Jan 25 13:47:49 <digitalspaghetti> baahh - I had accidently removed the android.permission.MOUNT_UNMOUNT_FILESYSTEMS permission :|
Jan 25 13:48:03 <TD> doh
Jan 25 13:48:16 <TD> digitalspaghetti: have you used contact photos on your phone at all?
Jan 25 13:48:22 <digitalspaghetti> i've added it back in now, and i'm not getting a NullPointerException error, but i'm still getting an empty result from the cursor
Jan 25 13:48:31 <TD> i had sort of assumed they'd work, as they work fine on the emulator. but they don't appear on my actual phone :(
Jan 25 13:48:34 <digitalspaghetti> TD: i have a couple of contacts with photos, but not really
Jan 25 13:48:41 <TD> do they actually show up when they call you?
Jan 25 13:48:48 <digitalspaghetti> i'm trying to get the users pictures from their DCIM folder
Jan 25 13:48:50 <digitalspaghetti> TD: yea
Jan 25 13:48:54 <TD> hrm
Jan 25 13:49:01 <TD> there must be something about the way i'm setting them then. thanks
Jan 25 13:49:57 <TD> digitalspaghetti: is there anything helpful in logcat? i guess not
Jan 25 13:49:59 <digitalspaghetti> just going to double check .getCount actually returns 0
Jan 25 13:50:20 <TD> yeah, just whack a log statement in there
Jan 25 13:50:34 <digitalspaghetti> hah! getCount returns 38
Jan 25 13:51:08 <digitalspaghetti> so that is working now, but nothing is showing in my gallery view now
Jan 25 13:51:14 <digitalspaghetti> but at least it's the right track now
Jan 25 13:51:24 <TD> maybe check that the if != null statement is being taken properly ...
Jan 25 13:51:32 <TD> ie convertView != null
Jan 25 14:05:05 <digitalspaghetti> ok, should have been == null
Jan 25 14:06:06 <digitalspaghetti> hrrm, now it shows them but gives me a OutOfMemoryError :|
Jan 25 14:06:12 <digitalspaghetti> http://paste.ifies.org/287
Jan 25 14:06:34 <TD> uh oh :) you're probably trying to load/decompress every image at full size from the sdcard
Jan 25 14:06:42 <digitalspaghetti> yea, probably
Jan 25 14:06:53 <digitalspaghetti> i probably want the thumbnail, not the full image
Jan 25 14:12:48 <TD> right
Jan 25 14:23:07 <digitalspaghetti> hrmm, when i do MediaStore.Images.Media._ID i get 38 back
Jan 25 14:23:12 <digitalspaghetti> when i try MediaStore.Images.Thumbnails._ID I get 0 :(
Jan 25 14:24:21 <TD> i guess it's looking for thumbnails that were pre-calculated by some other app
Jan 25 14:24:32 <TD> isn't there a standard for storage of thumbnails ?
Jan 25 14:28:57 <digitalspaghetti> ahh, that sucks
Jan 25 14:29:30 <digitalspaghetti> it was a new Sd card i had put in, but i had copied over the files
Jan 25 14:29:42 <digitalspaghetti> i needed to go into pictures and let it generate the thumbnails
Jan 25 14:29:44 <digitalspaghetti> now it works
Jan 25 14:30:34 <digitalspaghetti> there must be something i can do before the view loads to check that and if not generate them?
Jan 25 14:31:59 <TD> maybe the pictures activity has a thumbnailer service? or you could generate them yourself (copy/paste the thumbnailer code). or if you're feeling adventurous contribute an improvement to the pictures app to make thumbnailing a service, if it's not already got one
Jan 25 14:32:37 <digitalspaghetti> i should really grab the android source code
Jan 25 14:32:43 <TD> you can browse it online
Jan 25 14:32:49 <TD> that's what i do - android.git.kernel.org
Jan 25 14:33:36 <digitalspaghetti> where;s the gitview?
Jan 25 14:33:51 <digitalspaghetti> just showing empty just now
Jan 25 14:34:00 <digitalspaghetti> must be down
Jan 25 14:35:21 <TD> hmm
Jan 25 14:35:24 <TD> http://android.git.kernel.org/ works for me
Jan 25 14:35:27 <TD> shows me all the projects
Jan 25 14:36:08 <digitalspaghetti> wierd, had to end up refreshing it like 20 times before gitview showed
Jan 25 14:36:27 <TD> maybe it's load-balanced
Jan 25 14:40:31 <digitalspaghetti> i'll check this out: http://android.git.kernel.org/?p=platform/packages/apps/Camera.git;a=blob;f=src/com/android/camera/ImageGallery2.java;h=c8abdaebe878944eb91e67c36d55e7bc3b5e4c17;hb=1d4c75065966c4f6f56900e31f655bfd1b334435
Jan 25 15:38:23 <CardinalFang> digitalspaghetti, ya, kernel.org is wonky lately.
Jan 25 15:38:40 <CardinalFang> Use "android2.git..."
Jan 25 16:19:49 <kcoury> anyone know how to get a reference to a ListView inside an AlertDialog that was created using setMultiChoiceItems?
Jan 25 16:32:22 <digitalspaghetti> any ideas folks?
Jan 25 16:32:24 <digitalspaghetti> http://www.anddev.org/viewtopic.php?p=17042
Jan 25 16:46:41 <TD> digitalspaghetti: not sure what you're asking, sorry ?
Jan 25 16:46:50 <TD> the location of the full image?
Jan 25 17:12:08 <KNY> Gary|tp, haha, yes, the phone is on ;)
Jan 25 18:21:25 <savid> When the "back" button on the phone is pressed,  what event(s) does that trigger on the activity?
Jan 25 18:26:38 <savid> oh nm, figured it out :-)
Jan 25 18:27:10 <savid> I guess it goes all the way down to onDestroy(),  which is what I expected.
Jan 25 20:27:23 <digitalspaghetti> TD: yea - basically i need to make a multi-part post request
Jan 25 20:27:39 <digitalspaghetti> the OAuth library uses HTTPClient4, so I can do the header stuff
Jan 25 20:27:50 <digitalspaghetti> but i don't know how to pass the image in android as a parameter
Jan 25 22:26:09 <android51> I have a question regarding CompareEverywhere ?
Jan 25 22:26:33 <android51> any help would be appreciated
Jan 25 22:32:36 <KNY> android51, don't ask to ask. Just ask
Jan 25 22:33:02 <KNY> we can't answer questions you don't ask. Also, this might be better served in #android, depending on the question
Jan 26 00:22:22 <KNY> tmzt, Gary|tp, if I adb kill-server, then sudo adb devices, it sees the attached G1. Then any subsequent `adb devices` calls as a user work fine. Go figure.
Jan 26 00:25:19 <KNY> So, it's "working" now
Jan 26 00:25:56 <tmzt> if it uses a local ip socket there are no user protections
Jan 26 00:26:31 <tmzt> but the fact that adb server apparently has to run as root makes no sense if the permissions from yesterday are still in place
Jan 26 00:30:51 <KNY> yeah
Jan 26 00:31:02 <KNY> I have no clue what would have changed since yesterday
Jan 26 00:31:06 <KNY> but I'm psyched that it's working now
Jan 26 00:31:30 <kcoury> is there no easy way of getting access to a ListView created in an AlertDialog Builder?
Jan 26 01:05:23 <morgan> kcoury: i was trying to figure that out yesterday but i gave up.
Jan 26 01:05:38 <esses> hi all, I'm coding a downloading image file method. The code works for the downloading but the file downloaded is either corrupted or a sum of coloured lines.
Jan 26 01:05:53 <esses> can you help me?
Jan 26 01:08:47 <morgan> kcoury: i ended up just having the ListView's callback store the selected index in a variable
Jan 26 01:11:02 <esses> hi morgan
Jan 26 01:11:10 <esses> can you see my code?
Jan 26 01:19:14 <kcoury> morgan: yeah I'm still trying to figure something out. I'm using a multiple choice dialog but I want to use getCheckedItemPositions() and I have no access to the ListView to do that
Jan 26 01:28:04 <esses> the code works if i only one byte each time
Jan 26 01:28:12 <esses> i tested it now
Jan 26 01:28:43 <esses> but the file creation is too slow in this way
Jan 26 01:44:56 <CardinalFang> I accidentally the whole byte.
Jan 26 01:46:54 <morgan> hehe
Jan 26 01:50:31 <morgan> esses: what code?
Jan 26 01:50:54 <esses> i paste it
Jan 26 01:50:58 <esses> into pastebin
Jan 26 01:51:02 <esses> okay?
Jan 26 01:52:21 <morgan> pastebin is good. what are you trying to do?
Jan 26 01:53:39 <esses> to download a jpeg file, but I obtain a corrupted or strange image file
Jan 26 01:53:42 <esses> http://pastebin.com/d262c5c6a
Jan 26 01:54:14 <esses> I need to solve the problem because tomorrow i have to finish the app
Jan 26 01:56:03 <esses> the code works only if i set the array data to 1 byte size
Jan 26 01:57:40 <morgan> your code looks reasonable, is the resulting image 100% garbage?
Jan 26 01:58:34 <morgan> have you tried logging some of the data as you download it? that might be a quick sanity check
Jan 26 01:59:00 <esses> sometimes corrupted other times garbage
Jan 26 02:00:46 <morgan> your code looks just like the file() method here: http://www.eboga.org/java/open-source/httpclient-demo.html
Jan 26 02:01:22 <morgan> do you get the same results when you run outside of android?
Jan 26 02:02:00 <morgan> oh,
Jan 26 02:02:11 <morgan>     	outFile.write(b);
Jan 26 02:02:14 <morgan> that should be:
Jan 26 02:02:19 <morgan> outFile.write(b, 0, read);
Jan 26 02:02:48 <morgan> currently you are always writing 4k blocks even if you just read less then 4k
Jan 26 02:07:13 <esses> :-) taht is the problem
Jan 26 02:07:23 <esses> I'm tired very tired
Jan 26 02:08:33 <esses> here is 3 o'clock
Jan 26 02:10:04 <morgan> yeah, it gets hard to spot the small things that late
Jan 26 02:31:33 <esses> morgan do you know how to set the bitmapfactory options
Jan 26 02:31:34 <esses> ?
Jan 26 02:33:34 <morgan> o = new BitmapFactory.Options()
Jan 26 02:33:52 <morgan> o.inDither = true
Jan 26 02:33:55 <esses> okay
Jan 26 02:33:58 <esses> thanks
Jan 26 02:33:58 <morgan> is that what you mean?
Jan 26 02:34:01 <esses> yes
Jan 26 02:34:03 <morgan> ok
**** ENDING LOGGING AT Mon Jan 26 02:59:57 2009