**** BEGIN LOGGING AT Wed Jan 14 02:59:57 2009
Jan 14 03:20:30 <wasabi> Howdy. So... trying to rebuild a kernel with a modified config. This is mostly a build system issue.
Jan 14 03:20:44 <wasabi> config is in a vendor/product, prebuilt image is too
Jan 14 03:21:17 <wasabi> kernel says no work to do.
Jan 14 03:30:34 <jasta> romainguy: well, that's because you run Diet UNIX.  with less sugar.
Jan 14 03:30:50 <romainguy> which is more than fine by me usually :)
Jan 14 03:30:52 <jasta> while I enjoy the full, rich flavor of Linux :)
Jan 14 03:31:13 <romainguy> well I also enjoy the full, rich flavor of decent tools for my photos :))
Jan 14 03:31:17 <jasta> well, you should still be able to do this easily with sed or perl
Jan 14 03:31:23 <jasta> filter NUL bytes out of the file, then run strings on it
Jan 14 03:31:40 <jasta> sed -e 's/\0//g;' < foo | strings
Jan 14 03:31:42 <jasta> should do it
Jan 14 03:31:57 <jasta> well, it works for me because i speak and write exclusively English
Jan 14 03:32:00 <jasta> might not work so well for you :)
Jan 14 03:32:09 <romainguy> turns out I did not put any French in there
Jan 14 03:32:24 <romainguy> also turns out that the accented French characters are part of the 8 bits ASCII set
Jan 14 03:32:32 <romainguy> so it's often fine :))
Jan 14 03:32:47 <jasta> theres no such thing as 8 bit ASCII
Jan 14 03:32:52 <jasta> thats a falacy :)
Jan 14 03:33:08 <jasta> 8 bit ASCII typically refers to ISO-8859-1, but since it technically does not exist, its gibberish :)
Jan 14 03:33:41 <romainguy> funny
Jan 14 03:33:53 <romainguy> you sound exactly like the first paragraph of the wikipedia on the subject
Jan 14 03:33:58 <romainguy> "The use of the term is sometimes criticized"
Jan 14 03:33:59 <romainguy> :)
Jan 14 03:34:02 <jasta> thats because i wrote it, foo :)
Jan 14 03:34:16 <jasta> no im kidding, but i have been lectured about this enough times ot know better :)
Jan 14 03:34:32 <romainguy> I'll keep calling it 8 bits ASCII or extended ASCII :)
Jan 14 03:34:51 <jasta> why not call it precisely what it is, iso-8859-1 or latin-1?
Jan 14 03:35:04 <romainguy> because iso-8859-1 annoying to type
Jan 14 03:35:32 <jasta> iso-8859-1 is the charset that every European country whose economy isn't in the toilet uses, and plus some. :)
Jan 14 03:35:37 <jasta> well latin1 isn't :)
Jan 14 03:35:37 <romainguy> I know
Jan 14 03:36:06 <jasta> anyway, im not a charset purist.  im just pretending to be hehe
Jan 14 03:36:31 <romainguy> and that's why I don't care your lecture about a charset I've used enough to be sick of it :)
Jan 14 03:37:04 <romainguy> (though I have to admit it was a sanity savior in the early days of XML when UTF-8 support in some parsers was... interesting sometimes)
Jan 14 03:37:17 <jasta> to reiterate, if you are using the single byte subset of UTF-16, then you're safe to just strip NUL bytes and use strings
Jan 14 03:38:56 <jasta> btw, i have learned that lesson many times about deleting sources to new projects.  i now have a subversion sandbox on my local network specifically for this purpose :)
Jan 14 03:39:16 <romainguy> yeah usually I do the same
Jan 14 03:39:23 <romainguy> sed -e 's/\0//g;' < foo | strings << doesn't work btw :)
Jan 14 03:39:36 <jasta> well i just made it up off the top of my head
Jan 14 03:39:39 <jasta> let me actually verify the command line
Jan 14 03:40:25 <jasta> oh right, all this crud is in resources.arsc
Jan 14 03:41:12 <romainguy> yeah
Jan 14 03:41:17 <jasta> hmm, my sed is rusty.  use perl -pe :)
Jan 14 03:41:30 <jasta> i think that sed does not understand the meaning of "\0".  perl does.
Jan 14 03:41:49 <romainguy> that works much better :))
Jan 14 03:42:15 <romainguy> thanks
Jan 14 03:42:30 <jasta> np.  i knwo its not a great solution but at least it will get you the literal strings out
Jan 14 03:42:38 <romainguy> that's good enough for me
Jan 14 03:42:49 <romainguy> at least I learned how to read the output of aapt dump
Jan 14 03:43:00 <romainguy> not very useful knowledge though ^^
Jan 14 03:43:03 <jasta> hehe
Jan 14 03:43:26 <romainguy> what sucks is that I still have no way to recreate my theme.xml and my attrs.xml
Jan 14 03:43:41 <jasta> yeah, that was my big frustration when i went through all the trouble to reverse engineer this in the first place
Jan 14 03:43:49 <jasta> only to realize there are two different formats.  the other one i didn't understand.
Jan 14 03:44:13 <romainguy> interesting, it doesn't show all the strings
Jan 14 03:44:49 <romainguy> but enough to be useful
Jan 14 03:45:01 <jasta> any chance those strings have multi-byte utf-16 representations?
Jan 14 03:45:24 <romainguy> I doubt it, they're just in English
Jan 14 03:45:33 <romainguy> no punctuation or any weird character
Jan 14 03:45:41 <jasta> make sure you search through the full output too: the order though seemingly in tact could be random
Jan 14 03:45:42 <romainguy> so unless aapt does weird stuff to the text
Jan 14 03:46:01 <romainguy> yeah I checked
Jan 14 03:47:15 <jasta> hmm, no idea then
Jan 14 03:49:32 <jasta> you could look at it with xxd to try to find the strings manually.  then you could see why stripping NUL missed them
Jan 14 03:50:04 <romainguy> I just opened it in an hex editor
Jan 14 03:50:31 <jasta> yeah, same thing :)
Jan 14 03:50:44 <romainguy> ah interesting
Jan 14 03:51:20 <romainguy> ooh I understand what's going on
Jan 14 03:51:25 <romainguy> I had string arrays defined
Jan 14 03:51:25 <jasta> ?
Jan 14 03:51:40 <jasta> where did they go?
Jan 14 03:51:40 <romainguy> and the entries are prefixed with an index
Jan 14 03:51:57 <romainguy> and when I use your shell command some entries of the arrays just disappear
Jan 14 03:52:37 <romainguy> wait no, those are not indices
Jan 14 03:52:44 <romainguy> stupid format
Jan 14 03:55:24 <romainguy> ah!
Jan 14 03:55:27 <romainguy> fixed the command
Jan 14 03:55:30 <jasta> ?
Jan 14 03:55:43 <romainguy> I just had to use s/\0+//g
Jan 14 03:56:41 <jasta> hmm.  why would that matter?
Jan 14 03:57:04 <romainguy> I just confused
Jan 14 03:57:05 <romainguy> :))
Jan 14 03:57:09 <romainguy> I should just go to bed :)
Jan 14 03:57:44 <jasta> my perl fu is pretty up there.  and i say s/\0+//g and s/\0//g are the same :)
Jan 14 03:58:14 <romainguy> isn't \0+ supposed to match only sequences of more than 1 \0?
Jan 14 03:58:28 <romainguy> argh
Jan 14 03:58:29 <romainguy> stupid me
Jan 14 03:58:31 <romainguy> inclusive
Jan 14 03:58:32 <jasta> yes.  but "g" means keep going: match forever.
Jan 14 03:58:33 <romainguy> lalala
Jan 14 03:59:11 <jasta> \0 will match exactly 1, but s///g will say to keep matching on the string, so itll get all future \0's even if they were right next to the one it just matched
Jan 14 03:59:39 <romainguy> I stupidly thought that \0+ would match {2,n}
Jan 14 03:59:42 <romainguy> which is very wrong
Jan 14 03:59:47 <romainguy> on many levels :)
Jan 14 03:59:56 <jasta> well, its close.  it's \0{1,} :)
Jan 14 04:00:12 <romainguy> like I said "romainguy: inclusive"
Jan 14 04:00:14 <romainguy> :)
Jan 14 04:00:35 * jasta likes that Java lifted Perl's re syntax btw.
Jan 14 04:00:50 <jasta> that was a very smart move.  For all the things Perl does wrong, it has produced a godly regexp standard.
Jan 14 04:01:15 <romainguy> godly, I don't know, but it's good
Jan 14 04:01:22 <romainguy> I actually really like Java SE's regex engine
Jan 14 04:01:31 <jasta> i believe it is the regexp specification that the gods themselves use. :)
Jan 14 04:02:16 <jasta> Unfortunately I think that Perl6 is when we will see just how insane Larry Wall really is.  He basically wants to make regexps a functional language :)
Jan 14 04:02:31 <romainguy> only if it ever comes out :)
Jan 14 04:02:47 <jasta> i hope that it does not, for his sake
Jan 14 04:06:53 <romainguy> what the...
Jan 14 04:07:03 <romainguy> Perl 6 uses - as the concatenation operator??
Jan 14 04:07:36 <unix_infidel> HAHA.
Jan 14 04:07:48 <jasta> romainguy: i told you
Jan 14 04:07:50 <romainguy> I was reading a sample script and I read:
Jan 14 04:07:56 <romainguy> $z -= "!"
Jan 14 04:08:08 <romainguy> and I thought it was a weird but cool way to remove part of the string
Jan 14 04:08:12 <KNY> romainguy, it looks like it's ~
Jan 14 04:08:16 <romainguy> no... it concatenates
Jan 14 04:08:22 <KNY> [Update: Concatenation is now ~.]
Jan 14 04:08:22 <romainguy> ah maybe what I'm reading is outdated
Jan 14 04:08:24 <KNY> http://dev.perl.org/perl6/doc/design/exe/E03.html
Jan 14 04:08:31 <KNY> or maybe mine is outdated
Jan 14 04:08:46 <KNY> either way, that's going to break a lot of stuff
Jan 14 04:09:05 <KNY> [Update: Since ~ is concatenation, numeric XOR is now +^ instead.]
Jan 14 04:09:09 <KNY> this just gets better and better
Jan 14 04:09:47 <romainguy> "And here we see the new Perl 6 string concatenation operator: underscore"
Jan 14 04:10:01 <KNY> so we have three candidates already
Jan 14 04:10:05 <romainguy> from http://dev.perl.org/perl6/doc/design/exe/E03.html#___top
Jan 14 04:10:23 <KNY> http://dev.perl.org/perl6/doc/design/exe/E03.html#___top
Jan 14 04:10:25 <KNY> damn
Jan 14 04:10:29 <romainguy> ah but then it says it changed to ~
Jan 14 04:10:30 <romainguy> wtf
Jan 14 04:10:34 <KNY> scroll down to "Don't break the chain"
Jan 14 04:10:39 <KNY> and then look at the line right above it
Jan 14 04:10:47 <romainguy> yes just saw it
Jan 14 04:10:48 <KNY> looks like _ caused problems with identifiers
Jan 14 04:11:20 <romainguy> [Update: & is now +& or ~&, while | is now +| or ~|.]
Jan 14 04:11:25 <romainguy> awesome
Jan 14 04:11:29 <KNY> [Update: No, that's now the smart-match operator, just to avoid the =~ confusion. High precedence XOR is ^^ instead.]
Jan 14 04:11:42 <KNY> haha
Jan 14 04:12:02 <romainguy> [Update: No, it becomes +^ or <~^>.]
Jan 14 04:12:15 <romainguy> speaking of the bitwise xor ^
Jan 14 04:12:20 <KNY> haha
Jan 14 04:12:44 <KNY> it's looking like you could write a valid perl script with nothing more than <shift> and the number row
Jan 14 04:13:02 <romainguy> feels like Brainfuck
Jan 14 04:13:19 <romainguy> I guess they wanted to let people make fun of Perl for being a language sometimes hard to read
Jan 14 04:13:46 <jasta> or maybe, just maybe...
Jan 14 04:13:48 <KNY> better than Whitespace, I guess
Jan 14 04:14:03 <jasta> Larry Wall played a prank on the world with Perl, and he just can't believe people bought it.
Jan 14 04:14:07 <KNY> I'll be sticking with perl 5 for a while, it seems
Jan 14 04:14:10 <jasta> Trying to see just how far he can go :)
Jan 14 04:14:16 <romainguy> jasta: that would be awesome :)
Jan 14 04:14:17 <KNY> jasta, haha
Jan 14 04:15:08 <romainguy> ok
Jan 14 04:15:12 <romainguy> enough tv shows
Jan 14 04:15:17 <romainguy> time to rewrite this @!$# strings.xml
Jan 14 04:15:21 <KNY> A subroutine's adverbs are specified as part of its normal parameter list, but separated from its regular parameters by a colon:
Jan 14 04:15:27 <KNY> my sub operator:&sum; is prec(\&operator:+($)) ( *@list : $filter //= undef) {...
Jan 14 04:15:32 <KNY> uh... what?
Jan 14 04:16:02 <romainguy> KNY: you know, after seeing this, I will never make fun of C++ ever again
Jan 14 04:16:07 <KNY> haha
Jan 14 04:16:27 <jasta> my theory might just be right
Jan 14 04:17:15 <KNY> they even changed ternary operators from ?: (to ??::)
Jan 14 04:17:43 <KNY> oh, wait, there was an update: [Update: We've changed :: to !! to reduce that confusion, and because of the ? vs ! symbology of true? vs false? that pervades the rest of Perl 6.]
Jan 14 04:17:56 <KNY> so, folks: ternary operators are now ??!!
Jan 14 04:19:31 <CardinalFang> What do people use in XML being loaded into a PreferenceActivity, for "android:key" values?  I think it makes sense to use @+id/Foo, but that doesn't seem to work later on reading from the SharedPreferences object with getTYPE(String, DEFAULT), where I generate the "String" with  Context.getString(R.id.Foo) .
Jan 14 04:19:48 <romainguy> CardinalFang: use any string you want
Jan 14 04:20:13 <CardinalFang> Yeah, but having a literal in two places makes me sad.
Jan 14 04:20:31 <romainguy> you can try @string/blah
Jan 14 04:21:24 <CardinalFang> Yeah, then I have to add entries in values  <string name="blah">Some unique name that should be generated with something like @+id, damnit</string> .
Jan 14 04:22:31 <CardinalFang> Ugh.  I'm annoyed because I'm tired.  Sorry.  G'night.
Jan 14 04:22:44 <romainguy> er
Jan 14 04:22:52 <CardinalFang> Er?
Jan 14 04:22:58 <romainguy> no the unique name doesn't have to be generated by an @+id
Jan 14 04:23:04 <romainguy> why can't you just use a string that you pick?
Jan 14 04:23:12 <romainguy> you know, like "myapp.mypref" :)
Jan 14 04:25:41 <CardinalFang> Right.  I could.   Put   android:key="myapp.mypref"  and later  sp.getLong("myapp.mypref", 42)  .  I just though there should be some way of using the  R  generation to enforce that exists, and not find out at run time that I really typed "myapp.pref" the second time.
Jan 14 04:26:53 <CardinalFang> romainguy, Am I making sense at all?  It *is* way past my bed time.
Jan 14 04:27:34 <romainguy> I just told you, use @string/blah instead
Jan 14 04:27:54 <romainguy> then in your code you just use getString(R.string.blah)
Jan 14 04:30:04 <CardinalFang> Okay.  At least that has the   name  attribute close to the text contents of the string XML node, so I can see they're similar.
Jan 14 04:30:44 * CardinalFang snoozes.
Jan 14 04:47:35 <romainguy> yay!
Jan 14 04:47:39 <romainguy> my app is alive again
Jan 14 04:53:30 <unix_infidel> ?
Jan 14 05:51:40 <romainguy> jasta: you around?
Jan 14 06:29:44 * romainguy pokes jasta
Jan 14 06:45:46 <romainguy_> jasta: <poke poke>
Jan 14 07:14:52 <jasta> yes?
Jan 14 07:16:36 <romainguy_> jasta: code.google/p/shelves
Jan 14 07:16:38 <romainguy_> enjoy
Jan 14 07:16:43 <jasta> yaaay
Jan 14 07:17:14 <romainguy_> what you wanted would be in ShelvesActivity and BooksAdapter
Jan 14 07:17:32 <romainguy_> note that the application is still far from being feature complete, etc.
Jan 14 07:17:42 <jasta> so, i was gonna ask you something actually
Jan 14 07:17:47 <romainguy_> ok?
Jan 14 07:17:47 <jasta> im having the issue where Home is dying on me a lot again
Jan 14 07:18:05 <romainguy_> doh
Jan 14 07:18:07 <jsharkey> ooh  =D
Jan 14 07:18:22 <jasta> or more specifically, I think Home is being killed.
Jan 14 07:18:29 <jasta> but it takes for bloody ever to restart
Jan 14 07:18:52 <jasta> looks like the phone is bitching a lot about:
Jan 14 07:18:54 <romainguy_> I recently made several changes in cupcake thathelp a lot
Jan 14 07:18:55 <jasta> I/ActivityManager(14880): Low Memory: No more background processes.
Jan 14 07:19:02 <jasta> so i mean, that's probably why
Jan 14 07:19:04 <romainguy_> but it's weird that you run into this situation
Jan 14 07:19:10 <romainguy_> and yes, that won't help Home :)
Jan 14 07:19:29 <jasta> i'm wondering if you have any useful insight into how to debug the low memory killer
Jan 14 07:19:36 <jasta> or why the phone has such little memory anyway
Jan 14 07:20:11 <romainguy_> all I can think of is to check the running processes :)
Jan 14 07:20:16 <romainguy_> other than that...
Jan 14 07:20:51 <jasta> right but top is kinda useless.  dumpsys meminfo is not especially revealing
Jan 14 07:21:01 <jasta> mostly because i dont know what's "normal" for any of these processes
Jan 14 07:21:19 <jasta> for instance, system's private dirty total is 9240.  that i would call high, but i dont know.
Jan 14 07:21:48 <jasta> 9240kB i mean.
Jan 14 07:21:49 <romainguy_> do you often end up in this situation?
Jan 14 07:22:02 <jasta> hmm, it's hard to say.  when it starts happening, it seems to do this all damn day
Jan 14 07:22:19 <jasta> almost every time i've opened the browser today, hitting home clearly triggers a restart of the home app
Jan 14 07:22:31 <romainguy_> ok so when you are in this situation, can you grab an adb bugreport and mail it to me?
Jan 14 07:22:32 <jasta> and several times when ive used my own application it did this
Jan 14 07:22:38 <jasta> and even once when i used calendar
Jan 14 07:22:41 <romainguy_> there's someone at Google who would now what to look for
Jan 14 07:22:47 <jasta> sure, i *just* saw it happen.  and it's done it lots.
Jan 14 07:23:36 <jasta> btw, please yell at the guy who logs the full text of your incoming text messages to logcat.
Jan 14 07:23:47 <jasta> i hate having to prune that crap when i send bugreports
Jan 14 07:24:05 <romainguy_> ouch that sucks indeed
Jan 14 07:24:30 <romainguy_> at least it's better than a year ago when we noticed that passwords were logged and present in several bug reports in our internal bugs database :)
Jan 14 07:26:07 <jasta> whoa.
Jan 14 07:26:13 <jasta> 01-13 23:12:38.522 E/JavaBinder(14880): !!! FAILED BINDER TRANSACTION !!!
Jan 14 07:26:13 <jasta> 01-13 23:12:38.592 I/ActivityManager(14880): Process com.android.browser (pid 21692) has died.
Jan 14 07:26:36 <romainguy_> dude, your phone is either haunted or cursed
Jan 14 07:26:38 <romainguy_> :p
Jan 14 07:27:31 <jasta> maybe :)
Jan 14 07:27:54 <jasta> bbl, gotta get some work done before bed.  i will post this bug report i just captured though.  check this out...
Jan 14 07:28:32 <jasta> $ cat acore-dies-a-lot.bugreport | grep Process | grep 'has died' | wc -l
Jan 14 07:28:33 <jasta> 38
Jan 14 07:28:35 <jasta> hehe
Jan 14 07:29:37 <romainguy_> ahah nice
Jan 14 07:29:46 <jasta> granted thats not all acore
Jan 14 07:29:51 <jasta> in fact, i think i only have captured acore death like twice in that
Jan 14 07:30:58 <jasta> that said, all of those daths are from 01-13 22:22 to 23:20
Jan 14 07:31:02 <jasta> deaths*
Jan 14 07:32:19 <jasta> maybe when i'm through tormenting t-mobile, google can hire me to find and fix obscure system bugs :)
Jan 14 07:32:45 <romainguy_> finding them is easy :p
Jan 14 07:33:41 <romainguy_> anyway, time to go to bed
Jan 14 07:33:43 <romainguy_> good night
Jan 14 12:45:31 <esses77> hi all
Jan 14 12:45:55 <esses77> hi all
Jan 14 12:46:10 <esses77> hi all
Jan 14 12:47:07 <esses77> Does anyone now where is the package IOUtils to use copy method copy(inputstream,outputstream) ?
Jan 14 12:47:29 <esses77> i don't find it
Jan 14 13:53:06 <unasi7> anyone tried WhiteTheme with a ListDialog? I got white on white list enties. changable?
Jan 14 14:03:38 <marklar> unasi7: change the text color in your list item row layout?
Jan 14 14:04:06 <unasi7> marklar,... not a ListActivity. it's a ListDialog... i only set Strings there.
Jan 14 14:04:12 <marklar> ah, sorry misread
Jan 14 14:05:00 <unasi7> setItems is the Methode
Jan 14 14:05:22 <unasi7> it's white on white... creazy White-Theme. Looks not that propper. Also when building Tab-Bars
Jan 14 14:17:50 <unasi7> any ideas how to get the default date format?
Jan 14 14:17:54 <unasi7> i'v tried: android.provider.Settings.System.DATE_FORMAT
Jan 14 14:18:00 <unasi7> but this onl returns "date_format"
Jan 14 14:18:23 <marklar> not specifically, but wouldn't that be a locale thing?
Jan 14 14:20:05 <unasi7> yes. somehow. but i have no idea how. and i could not found any docu
Jan 14 14:20:28 <marklar> yeah sorry going to need to tackle that soon myself
Jan 14 14:21:42 <unasi7> okay... im working on it. tell you when i got it
Jan 14 14:22:19 <marklar> thanks :)
Jan 14 14:38:39 <unasi7> any swiss guys here?
Jan 14 14:38:44 <unasi7> need beta testers
Jan 14 15:03:49 <eldenz> any idea why TextView.setTextColor(R.drawable.mah_color); leads to black textcolor (it changes compared to default) althought mah_color is dfined as '#af00'?
Jan 14 15:03:57 <eldenz> unasi7, you got my email :p
Jan 14 15:04:52 <unasi7> eldenz, query?
Jan 14 15:04:56 <eldenz> sure
Jan 14 15:05:50 <unasi7> is there a chance to give a EditText (TextView) the focus for editing text?
Jan 14 15:10:13 <anno^da_> Could someone tell me why I'm getting "My app name" is not responding directly after boot up. I am having a boot receiver registered which directly starts a service which does its hard work in a background thread.
Jan 14 15:32:27 <CardinalFang> Can anyone recommend a doc that explains the "@" decoding rules for XML attribute values?
Jan 14 15:36:25 <eldenz> unasi7, requestFocus()?
Jan 14 15:37:14 <anno^da_> Can someone here confirm that Android is restarting all the services that had been running on the system before I rebooted the system?
Jan 14 15:38:14 <unasi7> eldenz, okay.. let me try
Jan 14 15:38:44 <jbq> no, the notion that a service was started doesn't survive a reboot, you need to explicitly restart your services.
Jan 14 15:39:48 <anno^da_> jbq: Well I thought so before I recognized the following behaviour.
Jan 14 15:39:58 <anno^da_> http://groups.google.com/group/android-developers/browse_thread/thread/acead84414770013#
Jan 14 15:40:19 <CardinalFang> anno^da_, Services can't automatically be started unless they have that permission, and if they catch a specific Intent (something like "BOOT_FINISHED').  It's not automatic.
Jan 14 15:40:57 <anno^da_> Well I have no Broadcast Receiver waiting for BOOT_COMPLETED now
Jan 14 15:41:04 <anno^da_> but my service gets started
Jan 14 15:41:11 <anno^da_> automatically after reboot
Jan 14 15:41:48 <anno^da_> (I was confused because of getting the not responding message)
Jan 14 15:41:52 <jbq> Do you have something that starts automatically during or soon after boot and connects to your service?
Jan 14 15:42:31 <unasi7> eldenz, great! was requestFocus()
Jan 14 15:42:32 <jbq> or do you listen to other broadcasts and start your service on those? (e.g. network changes, which are typically sent soon after boot)
Jan 14 15:43:05 <anno^da_> No. I had one Broadcast Receivcer that was intended to catch the Boot completed action in order to start my background service. But I have deleted this receiver in order to test the whole thing. Since the not responding message made no sense to me.
Jan 14 15:44:31 <anno^da_> Let me check something different I will close the service now while the system is running
Jan 14 15:44:37 <anno^da_> and reboot.
Jan 14 15:46:52 <eldenz> i assume there isn't a way to access com.android.internal.widget.NumberPicker?
Jan 14 15:48:48 <jbq> eldenz: if it's com.android, you're not supposed to access it, it's an implementation detail and isn't supported.
Jan 14 15:49:57 <tauno> on a similar note - can I use the "native" dialer app in my own app? I don't want to actually dial out - I just want to launch the dialer and get back the number that was entered
Jan 14 15:51:52 <jbq> I don't know whether you can just use it to pick a phone#. It's not clear from the manifest http://android.git.kernel.org/?p=platform/packages/apps/Contacts.git;a=blob;f=AndroidManifest.xml;h=949bfc88b8898f2014f4af3f552a1a71c6849d6f;hb=master
Jan 14 15:51:58 <anno^da_> Hmm I'm getting crazy my app gets automatically restarted on boot time but I don't know why. I have no Bro.Receiver :(
Jan 14 16:12:17 <anno^da_> Hmm when I run exactly the same APK on the emulator (app without broadcast receiver waiting for BOOT_COMPLETED) my app doesn't get started
Jan 14 16:12:53 <anno^da_> When I run it on the phone it always starts up :(
Jan 14 16:13:35 <anno^da_> jbq: Is it possible that there are multiple versions of the same app on the phone ? (by a bug of the installer or sth like that)
Jan 14 16:13:52 <jbq> anno^da_: unlikely
Jan 14 16:17:09 <wasabi> Are there instructions on how to properly build the android kernel tree?
Jan 14 16:20:55 <anno^da_> jbq: I have done the following now. Uninstalled -> reboot -> install   . Opened the app (just the UI/activity). Went back to the home screen (I can see the myapp thread running via top) . Then I rebooted the system and voila after reboot the myapp thread is up again.
Jan 14 16:21:16 <anno^da_> (app with no broadcast receiver at all)
Jan 14 16:21:31 <jbq> Sounds like a good question for android-developers.
Jan 14 16:21:55 <anno^da_> Yeah it seems so. I will extend my question in the groups cause this is totally weird :)
Jan 14 16:22:19 <anno^da_> (all my other apps work pretty fine without that "bug" or whatever it is)
Jan 14 16:23:36 <wasabi> *sigh*   i do not comprehend repo
Jan 14 16:26:26 <jbq> what are you trying to accomplish?
Jan 14 16:34:34 <CardinalFang> Has anyone wrapped widget.TimePicker in preference.Preference?
Jan 14 16:34:46 <CardinalFang> I'm looking for an example
Jan 14 16:57:03 <unasi7> is there a chance to hide the title bar with the app name?
Jan 14 17:00:22 <uscheller> requestWindowFeature(Window.FEATURE_NO_TITLE);
Jan 14 17:03:17 <unasi7> uscheller, thx!
Jan 14 17:04:41 <unasi7> uscheller, is that not possible in the Manifest.xml? because with this you first will see the bar, than it goes away.
Jan 14 17:05:15 <uscheller> I dont know - maybe it is
Jan 14 17:09:45 <unasi7> guys, usheller, much better in the manifest: android:theme="@android:style/Theme.NoTitleBar"
Jan 14 17:23:44 <uscheller> ok thx
Jan 14 17:36:31 <famast> i'm trying to use the make-live script to boot android on my PC but I'm getting a "kernel panic not syncing VFS..." error on boot. Any ideas? Some sort of file support perhaps is missing?
Jan 14 17:37:35 <CardinalFang> famast, I don't know Android per se, but I do/did know Linux.  Those five words, exactly?
Jan 14 17:37:58 <famast> CardinalFang: Yes theres more after the ... let me see...
Jan 14 17:38:05 <CardinalFang> I'd expect more before that, too.
Jan 14 17:38:18 <CardinalFang> famast, Don't paste it here, if it's more than 2 lines.
Jan 14 17:39:28 <famast> CardinalFang: Not that much more... android kernel panic not syncing vfs unable to mount root fs on unknown block
Jan 14 17:39:41 <famast> this thing is essentially trying to boot on a USB disk
Jan 14 17:40:34 <CardinalFang> Ah, then it needs USB support and USB Mass Storage support, and whatever filesystem that is on the "disk", and so on.
Jan 14 17:41:09 <famast> fdisk -l is showing "Linux"
Jan 14 17:41:27 <famast> and FAT16 most likely for the boot
Jan 14 17:41:29 <CardinalFang> That's the partition type.  Probably ext2.
Jan 14 17:41:58 <CardinalFang> Partition type != filesystem type
Jan 14 17:42:05 <famast> hmm ok i will have to check my kernel config. I've never worked with linux before and linux+android is making it even more interesting
Jan 14 17:42:16 <CardinalFang> Convention means nothing, especially when there's a problenm.
Jan 14 17:42:36 <CardinalFang> famast, Good luck.
Jan 14 17:42:52 <CardinalFang> famast, Are you building the kernel for android?
Jan 14 17:43:37 <famast> CardinalFang: Yes, and I just checked the filesystem its ext3 as you said
Jan 14 17:44:03 <CardinalFang> famast, Ah.  Then, it's probably lower level, in support for the hardware.
Jan 14 17:46:02 <famast> recompiling takes too long =/
Jan 14 17:46:43 <CardinalFang> It's a lot of source.  :\  That's the cost of trying to do everything.
Jan 14 17:47:25 <famast> if I compile then compile again it goes really quick (unless I do a make clean), but I'm not sure if my kernel modules were getting added if I don't start from the very beginning again
Jan 14 17:47:58 <CardinalFang> At least there's far less "Linux doesn't work with my new Whizbang 5000!".  The cost is everyone has to wait another tenth of a second to compile whizbang5xxx.c .
Jan 14 17:48:15 <famast> lol
Jan 14 17:51:18 <famast> brb
Jan 14 18:19:24 <anno^da_> jbq: I have got it working now. I have restarted again. Uninstalled -> installed. reboot. And voila now it doesn't start up again any more.
Jan 14 18:20:12 <anno^da_> It only starts up with my BroadcastReceiver as intended. But I can't really say why this has happened. I haven't changed anything in code.
Jan 14 18:21:48 <anno^da_> (I heard the same bug from users using my app after installing an update. (Although I hadn't changed anything on the BroadcastReceiver for catching the boot. So well I would bet that there is a bug in the installer. But hmm well I haven't got the skills to proof this :) )
Jan 14 20:48:39 <jbermudes> hello
Jan 14 20:49:45 <vol> hi
Jan 14 20:52:10 <jbermudes> does anyone have any advice for making apps that don't drain the battery too much?
Jan 14 20:53:51 <Disconnect> make an app that turns off the phone
Jan 14 20:54:58 <jbermudes> Disconnect: I suppose that could work, haha
Jan 14 20:56:07 <jbermudes> I've heard things about intense CPU usage draining the battery faster, is this a linear correlation or is there some sort of threshold where you start to drain faster?
Jan 14 20:57:37 <Disconnect> if the cpu is going hard, its at full speed and not sleeping. so it requires more power.
Jan 14 20:58:34 <wasabi> doing stuff uses power.
Jan 14 20:58:45 <wasabi> i don't think this requires a degree in power management. don't do stuff.
Jan 14 21:01:29 <Disconnect> heh
Jan 14 21:02:22 <jbermudes> yeah, that's what I figured, but I guess as I'm developing a game, there's not much I can do to avoid doing stuff constantly, heh
Jan 14 21:05:14 <vol> have threads sleep if they don't need to do more work
Jan 14 21:05:27 <vol> don't have something constantly invalidate the ui if nothing has chagned
Jan 14 21:05:28 <vol> *changed
Jan 14 21:05:42 <vol> don't suck up the GPS if it's not needed
Jan 14 21:06:20 <jbermudes> I'm a little confused as to having a thread sleep versus yielding
Jan 14 21:07:09 <jbermudes> I'm guessing that if a thread yields and nobody else needs to do work, it goes back immediately to your control
Jan 14 21:07:09 <vol> java threads that yield let other java threads do work
Jan 14 21:07:16 <vol> yes exactly
Jan 14 21:07:33 <vol> yielding is great if you want to keep CPU time but this thread isn't the most critical
Jan 14 21:08:08 <vol> sleeping lets the thread give up its time, and if no other java threads need the time, Java relinquishes this time slice back to the kernel
Jan 14 21:08:27 <vol> and continues to relinquish it until the sleep time is up or it is interrupted
Jan 14 21:08:37 <jbermudes> that makes sense
Jan 14 21:08:51 <vol> So, if you have a thread that runs sporadically, have it use sleep or wait
Jan 14 21:10:20 <jbermudes> right now my game loop is synchronized to the ViewSurface's holder, which more or less makes it align with the screen's vsync, but I think I'll modify it to update at a set time interval and sleep in between
Jan 14 21:11:25 <vol> set the game loop to keep track of how long it should wait until the next update, and then sleep that long.
Jan 14 21:11:47 <CardinalFang> Does anyone know of web-browsable Android system source code?  I'm astonished that Google Search doesn't find any source
Jan 14 21:12:24 <CardinalFang> I tried GIT'ing the repo a while back, but I ran out of disk space where I was, and didn't care enough to find more.
Jan 14 21:13:09 <vol> I know one exists because I've been linked to it before
Jan 14 21:13:15 <vol> but I couldn't tell you the address, sorry
Jan 14 21:25:09 <andrewoid> you can browse the source from the main git page
Jan 14 21:25:32 <andrewoid> http://android.git.kernel.org/
Jan 14 21:25:35 <andrewoid> but it's not very nice
Jan 14 21:25:45 <CardinalFang> Hrm.
Jan 14 21:29:22 <CardinalFang> Hot damn!  That's just what I needed.  Problem solved.  Thank you, andrewoid.
Jan 14 21:53:17 <andrewoid> you're welcome
Jan 14 22:22:20 <digitalspaghetti> if i download the code from git, can i set up eclipse to load the source when there is an exception?
Jan 14 22:22:33 <digitalspaghetti> at the moment i just get a JAR file and it asks me to link to source
Jan 14 22:23:06 <morgan> i haven't used eclipse in a while, but that should work fine
Jan 14 22:23:35 <digitalspaghetti> might help with some debugging, i;m getting null pointer exceptions
Jan 14 22:23:50 <morgan> hmm
Jan 14 22:24:13 <morgan> if you are getting NullPointerExceptions in android.jar code then you are prolly passing in something bad
Jan 14 22:24:40 <digitalspaghetti> i'm not 100% sure, it was late last night an i need to revisit it
Jan 14 22:24:55 <digitalspaghetti> i'm pretty sure the error is in the OAuth library and not Android thouhg
Jan 14 22:25:03 <morgan> ahh
Jan 15 00:56:31 <morgan> i have a slow View.onDraw() but when I time that method it is fast. where should i put timing code to capture the total time taken to draw the screen?
Jan 15 00:58:29 <romainguy_> you can't really do this without modifying the source code of the platform and replacing the framework on your device
Jan 15 00:58:41 <romainguy_> what you can do though is use traceview
Jan 15 00:59:10 <romainguy_> it will show you the total time spent in View.draw()
Jan 15 00:59:26 <morgan> i'll check it out. thanks.
Jan 15 01:16:28 <morgan> traceview is nice
Jan 15 01:27:44 <danfuzz> RIP Patrick McGoohan
Jan 15 01:28:03 <danfuzz> oops; wrong channel (still sad)
Jan 15 01:37:15 <DannyB|away> PATRICCCCCCCCKKKKKKKKKKKKKK
Jan 15 02:17:40 <jbermudes> how can I check the CPU usage of my program?
Jan 15 02:34:40 <Disconnect> tp
Jan 15 02:34:41 <Disconnect> top
**** ENDING LOGGING AT Thu Jan 15 02:59:57 2009