**** BEGIN LOGGING AT Wed Nov 07 03:00:02 2012
Nov 07 03:23:49 <regedit> does this channel have a log(ging bot)?
Nov 07 03:24:06 <Leeds> nope
Nov 07 03:24:58 <regedit> oh well
Nov 07 03:25:12 <regedit> argh must get new laptop, can't deal with this hunk of a rig anymore
Nov 07 03:25:21 <regedit> any recommendations?
Nov 07 03:25:29 <regedit> sorry for the off topic, ignore me
Nov 07 03:28:17 <dragorn> regedit: since it's quiet...  I love lenovo
Nov 07 03:29:02 <dragorn> regedit: never treated me wrong.  T-series is the best usually.
Nov 07 03:29:13 <Leeds> get an ARM Chromebook
Nov 07 03:29:24 <regedit> my criteria is; lightweight (ultrabook?), powerful (>= i5 ?), affordable (<= $800 / $900 or so)
Nov 07 03:30:07 <dragorn> regedit: check out the lenovo uhhh 510 or 520 ultra
Nov 07 03:30:13 <dragorn> regedit: Might be my next one
Nov 07 03:30:17 <regedit> mmm ok leme see
Nov 07 03:31:30 <regedit> i lost touch with CPUs nowadays. used to be all about the Ghz, like the last CPU i bought intel q6600 (quad 2.4 ghz)
Nov 07 03:31:32 <regedit> but now i'm seeng these i3/i5/i7 quads qith just 1.6 or 1.8 ghz wtf
Nov 07 03:31:39 <dragorn> base price is in the 600-700 range, slightly heavy but classified as 'ultrabook', ivy bridge
Nov 07 03:31:48 <regedit> *with
Nov 07 03:31:50 <dragorn> i figure "it's all fast" at this point
Nov 07 03:32:34 <regedit> so we're "there" already huh
Nov 07 03:32:52 <regedit> at that point in moore's law or wtvr
Nov 07 03:33:06 <g00s> isnt the 4 digit qualifier generally regarded as a performance thing, so the e5-2690 > e5-2670 ?
Nov 07 03:33:16 <g00s> either better cache, whatever
Nov 07 03:33:28 <g00s> makes it a little easier to compare cpus
Nov 07 03:33:56 <regedit> never fear - they always must make it hard and confusing, so that everything is marketable
Nov 07 03:34:33 <g00s> yeah, i look forward to buying a new pc like buying a new car
Nov 07 03:34:43 <g00s> apple makes it so easy.
Nov 07 03:34:51 * g00s <3 less choice
Nov 07 03:35:02 <dragorn> g00s: I haven't been able to track down BTLE APIs btw
Nov 07 03:35:09 <dragorn> g00s: but i was out of town yesterday
Nov 07 03:35:16 <Leeds> haven't actually bought 'a PC' for many years - just rolling upgrades
Nov 07 03:35:41 <g00s> dragorn: ok, cool thanks … yeah, i guess we'll find out when the 4.2 sources are out
Nov 07 03:35:51 <g00s> thought that would have happened on the 5th bday of android
Nov 07 03:35:59 <g00s> as a cool thing to do
Nov 07 03:36:34 <Leeds> it's a bit of a weird birthday, because it's the date of the announcement - G1/1.0 release is more significant
Nov 07 03:37:41 <Leeds> in the future, if anyone remembers at all, nobody will remember when WP8 or BB10 were announced, just when they released
Nov 07 03:38:49 <g00s> i'll always remember when wp8 was announced. it was the day google wanted to rain on microsoft's parade, but irony ruled the day and sandy rained on google's parade so they cancelled their announcement
Nov 07 03:38:54 <regedit> should 5400 RPM hdd bother me much?
Nov 07 03:39:24 <gutano> not if your giving it as a gift to someone you don't like
Nov 07 03:39:32 <regedit> rofl
Nov 07 03:39:39 <g00s> i have a 5400 and it is fine
Nov 07 03:39:54 <g00s> runs much lower temp than the 7200
Nov 07 03:40:07 <gutano> aftermarket SSD win?
Nov 07 03:40:29 <g00s> is there any smart status on ssd's that tell how much wear they have ?
Nov 07 03:40:45 <sampullman> just replace it every 2 years and you'll be fine
Nov 07 03:40:46 <g00s> not sure i would every get a used ssd
Nov 07 03:40:48 <gutano> no idea
Nov 07 03:40:51 <sampullman> :D
Nov 07 03:41:04 <gutano> new ones are bound to be faster, I wouldn't get used either
Nov 07 03:41:21 <gutano> I don't mean to mock traditional drives, just taht if your building a new machine, I wouldn't consider anything but ssd at this pt
Nov 07 03:41:29 <dragorn> regedit: watch out for 1.8 vs 2.5 inch drive bays
Nov 07 03:41:37 <dragorn> regedit: 2.5 is dirt cheap.  1.8 is not.
Nov 07 03:41:37 <maslen> g00s: Ever heard seth godin's talk on that? It's quite good
Nov 07 03:41:47 <maslen> g00s: (About more choices not necessarily better)
Nov 07 03:41:49 <regedit> i see
Nov 07 03:41:49 <Leeds> gutano: for a desktop, SSD for software, HDD for media is still fine
Nov 07 03:41:56 <gutano> +1
Nov 07 03:42:34 <g00s> maslen: not from seth … from Schwartz http://www.amazon.com/The-Paradox-of-Choice-ebook/dp/B000TDGGVU
Nov 07 03:44:13 <maslen> g00s: You're right. I was confusing the names.
Nov 07 03:44:29 <maslen> Referring to https://www.youtube.com/watch?v=Iy8R5TZNV1A
Nov 07 03:46:58 <AddisonE> Hello, I have created a custom DialogPreference for my settings screen that contains two textboxes. I am wanting to set attributes for the textboxes when I setup the DialogPreference in the settings.xml file. The problem is that the constructor has the AttributeSet variable, and I don't have access to the textboxes until the onCreateDialogView function. I currently have a variable for each attribute in the DialogPreference, bu
Nov 07 03:46:58 <AddisonE> t that's double the memory than required.
Nov 07 03:49:52 <g00s> btw, intel came out with their 3rd gen ssd controller http://www.anandtech.com/show/6432/the-intel-ssd-dc-s3700-intels-3rd-generation-controller-analyzed
Nov 07 03:50:10 <regedit> how does this look http://www.microsoftstore.com/store/msstore/en_US/pd/VIZIO-14Inch-Thin-Light-CT14A1/productID.252852900?ClickID=avvsla5layr5vn5zk0tz0oyntrkso0p5sllr dragorn
Nov 07 03:51:19 <dragorn> regedit: if you can't expand the memory, 4 gig is a little low for eclipse + any sort of VM
Nov 07 03:52:03 <regedit> dragorn: good point, but what if i could? chatting with rep now..
Nov 07 03:52:15 <Leeds> or to put it more generally, I wouldn't touch a PC which I didn't know I could easily upgrade to 16GB
Nov 07 03:52:29 <dragorn> regedit: I'm happy with 8 in a laptop
Nov 07 03:54:23 <regedit> is vizio good generally? never heard of 'em
Nov 07 03:54:47 <regedit> is it the 80s/90s again where everyone and their uncle is manufacturing PCs?
Nov 07 03:54:48 <dragorn> heard of them, never used them.  Also if you care about linux, do some searching to see if anyone has gotten it running cleanly
Nov 07 03:55:15 <g00s> i would just avoid hp
Nov 07 03:55:33 <g00s> asus, lenovo seem like good bets
Nov 07 03:56:42 <regedit> dragorn: don't need to search, the answer is no
Nov 07 03:56:42 <regedit> :P
Nov 07 03:56:59 <regedit> tried linux mint for a while recently, let me down...
Nov 07 03:59:56 <regedit> so on the one hand, it doesn't seem like this laptop can do > 4gb ram.... on the other hand, i dunno how many other laptops with this specs/price ratio can
Nov 07 04:00:15 <dragorn> regedit: that's fine, didn't know your requirements
Nov 07 04:00:18 <g00s> regedit: http://www.amazon.com/b/ref=sv_pc_2?ie=UTF8&node=2956501011
Nov 07 04:01:16 <g00s> what the hell is with movies costing $9.99 to rent on iTunes these days
Nov 07 04:01:24 <regedit> i am indeed worried about eclipse + emulator though, so unsure... otherwise i might have added to cart by now
Nov 07 04:01:26 <g00s> there is /no fucking way/ i'm paying that :P
Nov 07 04:01:55 <regedit> g00s: jobs still after you from the grave :D
Nov 07 04:02:03 <g00s> hehe
Nov 07 04:02:10 <AddisonE> Hm, quick question. Say I have a dialogpreference with two textboxes inside of it. Is there a simple way to pass custom attributes from the dialogpreference to the textboxes? (I don't think there is, but maybe there is a simple way)
Nov 07 04:02:14 <dragorn>  5169 dragorn   20   0 5258m 923m  32m S   0.3 11.9  22:41.54 java
Nov 07 04:02:23 <dragorn> my eclipse is using 5.2 gig
Nov 07 04:02:31 <g00s> dragorn: WHAT!!!!
Nov 07 04:02:39 <g00s> my eclipse usually stays around 500 MB
Nov 07 04:02:48 <g00s> maybe 600
Nov 07 04:02:55 <canadiancow> g00s, ive never seen eclipse under a gig :P
Nov 07 04:03:01 <canadiancow> after having a project open for an hour
Nov 07 04:03:09 <Leeds> my Chrome is usually using 5GB-6GB at least, but then I rarely have less than 150 tabs open
Nov 07 04:03:23 <g00s> canadiancow: i install the platform runtime, JDT, and ADT. it starts at about 350mb and climbs on usage to 500-600 for me
Nov 07 04:03:43 <canadiancow> Leeds, lol
Nov 07 04:04:01 <g00s> Leeds: heh, yeah thats why i had to stop using chrome :| ff seems to be a little better with lots of tabs
Nov 07 04:04:23 <maslen> g00s: That's their architecturing at work
Nov 07 04:04:57 <regedit> Leeds: how the f*** do you even navigate about tabs
Nov 07 04:05:07 <Leeds> regedit: lots of windows
Nov 07 04:05:13 <regedit> ah
Nov 07 04:05:31 <g00s> doesnt FF have tab groups or something
Nov 07 04:06:02 <regedit> my eclipse currently at 500 MB (peak working set, 403mb private working set))
Nov 07 04:06:16 <regedit> javaw.exe
Nov 07 04:06:41 <regedit> crashplan / mozy also eating up a bunch
Nov 07 04:07:23 <Leeds> I do like crashplan, but it eats a lot of memory... I do have it backing up about 1.3TB of data though, so that's a lot to track :)
Nov 07 04:07:45 <g00s> yeah, crashplan …
Nov 07 04:07:51 <regedit> me too 1.3 tb lol
Nov 07 04:08:05 <regedit> we have same pron tastes perhaps?
Nov 07 04:08:21 <Leeds> doubt it...
Nov 07 04:10:20 <regedit> rofl
Nov 07 04:13:19 <Tyrope> How would one go about having 2 buttons be exactly 50% width next to eachother in a RelativeLayout? I tried the weight trick but that doesn't seem to be a valid thing.
Nov 07 04:14:29 <g00s> weights are for LL
Nov 07 04:15:15 <Tyrope> Yeah, that's what eclipse told me. Though it seems kinda silly to need a Linear in a Relative, since the buttons are all in place, just not sized right. (they're on wrap_content at the moment)
Nov 07 04:18:36 <maxlamer> Tyrope add an  android.widget.Space object with 0dp height and width & layout_centerHorizontal="true"
Nov 07 04:18:49 <maxlamer> then line up the left and right margins with that
Nov 07 04:19:19 <Tyrope> Space objects aren't android 2.2 though, are they?
Nov 07 04:19:38 <maxlamer> ah, true
Nov 07 04:19:52 <maxlamer> any boject will do, i only chose that because its light
Nov 07 04:20:09 <Tyrope> Empty TextView?
Nov 07 04:20:14 <maxlamer> viewstub is api 1
Nov 07 04:20:26 <Leeds> maybe a generic View
Nov 07 04:20:38 <maxlamer> any really
Nov 07 04:20:40 <Tyrope> Leeds: Can those be generated with XML?
Nov 07 04:20:49 <Leeds> <View />
Nov 07 04:20:59 <Tyrope> Ofcourse.
Nov 07 04:22:13 <Tyrope> is there any performance difference between layout_centerInParent and layout_centerHorizontal?
Nov 07 04:58:23 <brickhead> Hey Guys I need to encrypt a file but also use randomaccessfile
Nov 07 04:58:52 <brickhead> As Cryptography is not really my strong area, is there any library I can refer to ?
Nov 07 05:28:36 <lasserix> Has anybody been able to import that latest release of action bar sherlock into the eclipse?
Nov 07 05:29:36 <g00s> lasserix, "they" told me we should be doing minsdk=17 now
Nov 07 05:29:45 <g00s> ;)
Nov 07 05:31:12 <lasserix> Not sure what you mean, I can import existing project on previous version, but the new version doesn't have any projects show up in the list to import
Nov 07 05:32:39 <lasserix> But thanks? :/
Nov 07 05:38:15 <lasserix> does the standard content view container have a top bar? I'm trying to set my ABS background to transparent but nothing I do gets rid of the vertical divider
Nov 07 05:40:38 <lasserix> Most people must be caught up with the elections, sigh.
Nov 07 05:40:51 <g00s> obama won. nothing to be caught up with :)
Nov 07 05:40:52 <tdignan> what election
Nov 07 05:41:16 <tdignan> are they electing a new king of Android?
Nov 07 05:41:27 <tdignan> if no then IDGAF
Nov 07 05:43:35 <lasserix> he has not technically won, and people get caught up for all sorts of reason most of all either in the excitment of having won or the in the belittling sadness of losing ;p
Nov 07 05:43:50 <lasserix> the android king, hmm...
Nov 07 05:44:01 <lasserix> makes me think of center command in tron
Nov 07 05:44:12 <tdignan> Pepsi and Coke are running for president
Nov 07 05:44:13 <lasserix> somewhere deep within the google server complex
Nov 07 05:44:27 <tdignan> I'll pass. Water is much healthier.
Nov 07 05:44:49 <lasserix> too bad all the world's water was used in the last few bottles of coke & pepsi co ;p
Nov 07 05:45:42 <tdignan> I read a study of the different PH of all sorts of drinks
Nov 07 05:45:51 <tdignan> and dasani water by coke is pretty bad
Nov 07 05:46:17 <tdignan> makes sense because I always thought it tasted slightly different
Nov 07 05:46:26 <tdignan> the worst drink for your teeth, is grape juice
Nov 07 05:47:10 <regedit> so i did buy this in the end http://www.microsoftstore.com/store/msstore/en_US/pd/ThemeID.27509700/VIZIO-14Inch-Thin-Light-CT14A1/productID.252852900
Nov 07 05:47:41 <regedit> 128 ssd, i5 up to 2.60, 4gb ram, lightweight, amawesome price,
Nov 07 05:47:56 <g00s> regedit: is that a ms certified thing; something w/o crapware installed
Nov 07 05:48:01 <ron_frown> I love that there is no politics talk going on here
Nov 07 05:48:01 <tdignan> i had a vizio tablet
Nov 07 05:48:09 <tdignan> cann't say i'd buy vizio again
Nov 07 05:48:19 <regedit> who cares? i always reformat to whatever i care for
Nov 07 05:48:24 <tdignan> if someone talks about politics, pour scalding oil on them
Nov 07 05:48:41 <lasserix> JackWharton, are you here? For the life of me I cannot get the sherlock action bar to be transparent.
Nov 07 05:48:42 <g00s> regedit: yeah … unless you dont get the CD :P
Nov 07 05:49:32 <regedit> g00s: they usually have a cd option, either to burn or comes with laptop or free shipped. worst case, it's retail win7 for me
Nov 07 05:49:33 <lasserix> *JakeWharton i mean
Nov 07 05:49:38 <regedit> or who knows - linux perhaps?
Nov 07 05:49:45 <JakeWharton> wut
Nov 07 05:49:57 <JakeWharton> it's transparent by default
Nov 07 05:50:01 <g00s> regedit: cool, i hope you enjoy it
Nov 07 05:50:16 <regedit> thanks, hope so too
Nov 07 05:50:27 <regedit> had enough with my custom rig
Nov 07 05:50:36 <regedit> dotn play as much crysis anymore
Nov 07 05:50:38 <lasserix> JakeWharton: how can I hide the vertical divider? I am trying to go for a white minimalistic background
Nov 07 05:50:40 <regedit> *don't
Nov 07 05:50:41 <tdignan> i won't be buying any more desktops
Nov 07 05:50:46 <JakeWharton> what divider?
Nov 07 05:50:53 <tdignan> need to get myself a mac as well
Nov 07 05:51:05 <lasserix> On the actionbar between the actionbar and the "content"
Nov 07 05:51:15 <JakeWharton> oh, set the background to null
Nov 07 05:51:17 <lasserix> Err between the activity's content and the actionbar's content
Nov 07 05:51:17 <regedit> too much technical trouble with desktop components. i very much needed a computer i can RMA the hell out of
Nov 07 05:51:28 <lasserix> I tried that, still I get the horizontal divider
Nov 07 05:51:41 <lasserix> getSupportActionBar().setBackgroundDrawable(null); you mean?
Nov 07 05:51:47 <JakeWharton> maybe
Nov 07 05:52:04 <JakeWharton> or in the theme
Nov 07 05:52:09 <JakeWharton> or set it to a transparent color drawable
Nov 07 05:52:17 <regedit> now i need to begin selling my desktop rig piece by piece... anyone need some ram? gpu? mobo? :)
Nov 07 05:52:25 <lasserix> I tried that too, and tried transparent color
Nov 07 05:52:52 <lasserix> <style name="AppTheme.ActionBar.Dark" parent="@style/Widget.Sherlock.ActionBar"> <item name="android:background">@color/clear</item> <item name="background">@color/clear</item> </style>	
Nov 07 05:53:29 <lasserix> where @color/clear has been #00000000 or has been a transparent drawable or has even just been same color as window background
Nov 07 05:53:30 <JakeWharton> what's color/clear?
Nov 07 05:53:41 <lasserix> #00xxxxxx
Nov 07 05:54:05 <lasserix> tried all three, and tried setting the getSupportActionBar to null in my oncreate too
Nov 07 05:59:01 <lasserix> Doh! I figured it out
Nov 07 05:59:23 <lasserix> Had wrong derived parent class set.
Nov 07 05:59:47 <lasserix> Thanks for responding though
Nov 07 06:01:30 <lasserix> Oh no wonder, I had this problem before. Is there a way to override parent themes? The problem is if I use Theme.Sherlock.Light it forces the horizontal divider, but if I just use Theme.Sherlock everything else goes out the window (on a white background)
Nov 07 06:07:50 <lasserix> How can I override the actionbar style theme while using a different parent theme?
Nov 07 06:15:27 <karakuri> anybody here?
Nov 07 06:15:33 <lasserix> no
Nov 07 06:15:37 <karakuri> darn
Nov 07 06:16:16 <karakuri> if i were curious as to how Android applies manifest properties to activities at run time, where might i look?
Nov 07 06:16:19 <lasserix> though if you ask maybe your question will echo enough to tunnel through to the ghost in the machine
Nov 07 06:16:26 <karakuri> as in, actual code that runs
Nov 07 06:17:02 <karakuri> like, where/when is it parsed from the xml, what calls are made, etc
Nov 07 06:17:40 <jmfurlott> karakuri, pretty sure from R
Nov 07 06:18:07 <jmfurlott> at least the parsing
Nov 07 06:19:07 <karakuri> its hard to explain, but i'm curious about things that might be happening to the activity instance before i get the onCreate callback
Nov 07 06:23:04 <lasserix> JakeWharton what is the background ninepatch image name of the ActionBar.Sherlock.Light theme named?
Nov 07 06:23:18 <g00s> karakuri: you probably want to look for something like 'attachContext()'
Nov 07 06:23:18 <JakeWharton> action bar background?
Nov 07 06:23:20 <JakeWharton> window background?
Nov 07 06:23:42 <JakeWharton> you can just look into abs__themes.xml and abs__styles.xml
Nov 07 06:25:52 <jmfurlott> JakeWharton I know you from ActionBarSherlock and I want to say thank for you for saving my Android project in class last semester. I'll leave it at that
Nov 07 06:26:07 <JakeWharton> excellent. hope you got an A
Nov 07 06:27:03 <jmfurlott> We did, and ViewPage Indicator is incredible as well.  Makes the theming too easy
Nov 07 06:29:10 <JakeWharton> awesome. If I ever find some free time I'd love to update those projects
Nov 07 06:29:11 <JakeWharton> they need some love
Nov 07 06:29:18 <lasserix> JakeWharton: sorry to bug you. I am using Theme.Sherlock.Light as parent, but I want a transparent actionbar background. If I use Theme.Sherlock and set the style of the ab to have a transparent background it works, but when I set the parent theme to Theme.Sherlock.Light it always shows the vertical divider. What am I missing to override the Theme.Sherlock.Light to make the background of
Nov 07 06:29:19 <lasserix> the actionbar not have the vertical divider?
Nov 07 06:29:41 <JakeWharton> you mean horizontal divider?
Nov 07 06:29:45 <lasserix> yes
Nov 07 06:29:51 <jmfurlott> JakeWharton, what are you working on nowadays?
Nov 07 06:30:15 <JakeWharton> jmfurlott: in the open source world? mostly Square's open source projects as well as some yet-unreleased ones
Nov 07 06:30:26 <JakeWharton> and, of course, our apps itself suck a lot of time
Nov 07 06:31:18 <jmfurlott> Square the credit card company?
Nov 07 06:31:34 <JakeWharton> indeed
Nov 07 06:32:25 <lasserix> JakeWharton, any ideas?
Nov 07 06:33:17 <jmfurlott> JakeWharton, awesome man, thanks again
Nov 07 06:34:28 <lasserix> How can I make the horizontal divider null for the action bar in a theme if I am using Theme.Sherlock.Light as the parent theme?
Nov 07 06:35:25 <f2prateek1> lasserix: i'm not sure about this, but it could be the last attribute on this samples page
Nov 07 06:35:26 <f2prateek1> http://actionbarsherlock.com/theming.html
Nov 07 06:35:33 <f2prateek1>     <item name="android:backgroundSplit">@drawable/bg_striped_split</item>
Nov 07 06:39:01 <lasserix> f2prateek1 Uhh, I'm not sure I see your point. I am declaring both android / background items but I don't use a split
Nov 07 06:40:05 <f2prateek> i thought the splitBackground was the divider between the action bar and the content, so if you're not overriding that it will use the default one form the parent
Nov 07 06:40:21 <f2prateek> *from
Nov 07 06:40:38 <lasserix> splitBackground doesn't refer to having a split action bar?
Nov 07 06:41:09 <f2prateek> lasserix: oh that makes more sense. didn't occur to me
Nov 07 06:41:58 <lasserix> thanks though, yeah just tried it no luck
Nov 07 06:42:57 <f2prateek> what about actionBarDivider
Nov 07 06:43:31 <lasserix> could you quality that?
Nov 07 06:43:53 <karakuri> does any one know how an activity actually starts?
Nov 07 06:44:59 <f2prateek> lasserix: yeah you would put it in you app theme as  <item name="actionBarDivider">#00000000</item>
Nov 07 06:45:55 <f2prateek> lasserix: it might be the vertical deiveder though
Nov 07 06:46:02 <lasserix> hehe
Nov 07 06:46:07 <lasserix> there is no hope.
Nov 07 06:46:47 <lasserix> didn't even change the vertical divider, i feel like i am ovveriding the parent theme incorrectly somehow
Nov 07 06:49:00 <lasserix> i guess at this point it'll just be easier to go through item by item and make the other things visibile, thanks for the help f2
Nov 07 06:53:12 <f2prateek> lasserix: np. one last potential solution - https://github.com/JakeWharton/ActionBarSherlock/issues/379. attribute is android:backgroundStacked
Nov 07 06:55:42 <lasserix> heh ive been going through the drawables adding colored dots can't seem to cause any change so i think there is something wrong in the inheritance
Nov 07 07:14:58 <karakuri> whats a good resource for learning about IBinders?
Nov 07 07:15:10 <karakuri> inb4 IBinders full of women
Nov 07 07:17:01 <Ivru> IBinders full of women
Nov 07 07:17:05 <Ivru> oh wait...
Nov 07 07:50:42 <superlinux-hp> hi. i have downloaded a static build of rtmpdump for android, my question is how to run it in the application context?
Nov 07 08:19:00 <RandomSort> Good morning
Nov 07 08:19:32 <RandomSort> is there a way to figure out differences between a specific function in different API levels?
Nov 07 08:22:14 <ron_frown> the api has a api level drop down
Nov 07 08:22:15 * ron_frown shrugs
Nov 07 08:22:20 <RandomSort> thanks
Nov 07 08:22:22 <ron_frown> also generally says since verson blah at the top
Nov 07 08:22:26 <g00s> lol
Nov 07 08:22:39 <tachikoma> moin
Nov 07 08:22:49 <RandomSort> yeah, the function I'm looking at was added in level 1 but acts differently in 10 and 15 :(
Nov 07 08:23:09 <g00s> RandomSort: surprise surprise !
Nov 07 08:23:14 <RandomSort> WebView.loadDataWithBaseUrl()
Nov 07 08:23:17 <RandomSort> still makes me sad :(
Nov 07 08:23:18 <appel1> if it isn't documented, you can always go read the source
Nov 07 08:23:39 <g00s> webview is … a world of hurt
Nov 07 08:23:41 <RandomSort> true
Nov 07 08:23:44 <RandomSort> indeed
Nov 07 08:54:38 <hrzhu> hello all, I have a problems with eclipse dev env setting up. I have eclipse 4.2, android sdk, android plugin for eclipse adt installed, followed with the instructions on android developer site, but I cant see android sdk manager in eclipse-windows menu, nor can I create a new android project. everything was fine with eclipse 3.7
Nov 07 08:57:45 <lasserix> If I am making a ninepatch file for the actionbar background (to replace the horizontal divider) what are recomended dimensions?
Nov 07 08:58:14 <baptistem> hrzhu: on eclipse check ctrl + 3 + log for plugin error
Nov 07 08:59:05 <baptistem> is there a way to display the home screen from an application? I mean if I've done my task let the user go back to his home
Nov 07 09:00:34 <Leeds> baptistem: you can launch the launcher, as it were, by startActivity-ing the right Intent
Nov 07 09:02:41 <baptistem> Leeds: mmmh, I'm not sure it's what I'm looking for, I want to act as the 'home' button act
Nov 07 09:02:55 <Leeds> yes, all the home button does is issue the appropriate Intent
Nov 07 09:03:02 <Leeds> nothing magic about it
Nov 07 09:03:44 <baptistem> oh right ok
Nov 07 09:03:47 <baptistem> thx Le
Nov 07 09:03:55 <baptistem> thx Leeds
Nov 07 09:04:03 <Leeds> np
Nov 07 09:04:27 <baptistem> ( found the good intent here http://stackoverflow.com/questions/2752319/android-simulate-home-click if anyone got the same issue )
Nov 07 09:05:10 <Leeds> right...
Nov 07 09:05:49 <Leeds> baptistem: as a general piece of advice, whenever you see romainguy's name attached to something, it's probably going to be as good an answer as you'll find anywhere
Nov 07 09:08:51 <baptistem> Leeds: I'm not sure I understood, you mean that it won't be 'The' perfect answer ? As far as I can see, he is a answer eater on SO
Nov 07 09:09:12 <Leeds> baptistem: he's one of the world's top experts on Android...
Nov 07 09:09:27 <baptistem> oh ok :)
Nov 07 09:09:52 <Leeds> and he's not here right now to disagree :)
Nov 07 09:10:46 <hrzhu> baptistem: When I open the preference dialog, there is a DDMS and the error log prints a line says "Invalid preference category path: com.android.ide.eclipse.preferences.main (bundle: com.android.ide.eclipse.ddms, page: com.android.ide.eclipse.ddms.preferences.LogCatPreferencePage)" what does it imply?
Nov 07 09:11:22 <[deXter]> Leeds, you forgot to mention he's also a very good photographer. ;)
Nov 07 09:11:40 <Leeds> that too
Nov 07 09:13:24 <baptistem> hrzhu: did you already download a sdk?
Nov 07 09:13:36 <hrzhu> yeah I'm pretty sure
Nov 07 09:13:58 <baptistem> ok, windows? linux?
Nov 07 09:14:11 <hrzhu> baptistem: xubuntu 12.04
Nov 07 09:14:30 <g00s> xubuntu ftw
Nov 07 09:15:21 <baptistem> hrzhu: ok I had 2 main issues with eclipse/android/adt first was path issue, eclipse is waiting for a ~/android-sdks by default iirc, the second one was about 32/64bits libs
Nov 07 09:17:26 <hrzhu> baptistem: my sdk is located in ~ and I've put it in PATH I cant run adb from terminal and I am running a 32bit linux with a pae kernal
Nov 07 09:18:30 <baptistem> hrzhu: you can't run adb? what happend?
Nov 07 09:18:57 <hrzhu> baptistem: sorry just typo I mean I can
Nov 07 09:19:04 <baptistem> ok
Nov 07 09:19:40 <g00s> bwah, my buddy just got a job at google :)
Nov 07 09:20:05 <baptistem> hrzhu: so the problem come from the plugin not from your conf, did you try the ctrl+3 stuff?
Nov 07 09:24:20 <hrzhu> baptistem: why do yo mean by ctrl+3, I press ctrl 3 typed log and there is a error says Invalid preference category path, by the way I can open the android logcat
Nov 07 09:27:31 <baptistem> hrzhu: this seem to be very specific to eclipse, you may have better answer on #eclipse you may find a way to reset your preference on this plugins
Nov 07 09:29:13 <hrzhu> baptistem: ok, thank you for your help. i'll try it later
Nov 07 09:29:19 <sNore> has anyone an idea why i get sound glitches since android 4.0, when setting the playbackrate of audiotrack?
Nov 07 09:45:40 <hrzhu> the android sdk works fine with idea 11
Nov 07 09:47:27 <vascobnunes> good morning. Any one knows how to configure SVN on eclipse for an existing project?
Nov 07 09:48:43 <tdignan> dont.. just use git svn on the command line :)
Nov 07 09:49:25 <vascobnunes> tdignan: hmm?
Nov 07 09:49:30 <Tee_Pee> 'ello
Nov 07 09:54:11 <Tee_Pee> http://pastebin.com/G1jvZ7yu  <---That is my gridview adapter... there are a few performance hiccups, and I am trying to figure out what I can change for it to improve performance.
Nov 07 09:54:35 <Tee_Pee> If anyone is bored enough to spend a few minutes on it, fresh thoughts might be helpful
Nov 07 09:56:07 <tdignan> Tee_Pee: see how you're calling findViewById every time, even if convertView is non-null?
Nov 07 09:56:11 <muszek> hi... I've had my app and its play store description translated to Spanish.  Play Store lets me translate to Spanish (Spain), S... (Latin America) and S... (USA). If I only add one translation (say Spain), will it also "count" for people with the other two Spanish variations?
Nov 07 09:56:43 <muszek> or should I create 3 translations and paste the same text? (I will not translate it again)
Nov 07 09:56:48 <tdignan> Tee_Pee: don't do that. Make a static class calle ViewHolder that contains the views that are in your layout. Then save it in the convertView, with setTag()
Nov 07 09:56:51 <sNore> has anyone an idea why i get sound glitches since android 4.0, when setting the playbackrate of audiotrack?
Nov 07 09:57:27 <superlinux-hp> erm, my phone, after being reformated, it's not showing up on eclipse. i am on linux. what should i do?
Nov 07 09:57:29 <tdignan> Tee_Pee: Then, instead of just hitting the findViewById() code (which is your bottleneck) you say if (tag = getTag()) { use views from tag } else  { fill in tag from findViewById and store tag }
Nov 07 09:57:37 <tdignan> for more info google "viewholder pattern android"
Nov 07 09:58:20 <baptistem> superlinux-hp: did you set up on your computer as said here http://developer.android.com/tools/device.html#VendorIds ?
Nov 07 09:58:22 <tdignan> findViewById is not that performant of a function. You'll find it's much faster after you make these changes.
Nov 07 09:58:52 <superlinux-hp> baptistem, the udev rule is there already
Nov 07 09:58:52 <Tee_Pee> That's a start, albeit I don't think it's the findViewById alone, as I have a similar adapter that doesn't look for views by ID and still has the same issue. Thanks though, let's fix this first then move on...
Nov 07 09:59:53 <tdignan> Tee_Pee: do I see a downloader?
Nov 07 10:00:01 <tdignan> downloading images from somewhere in there too...?
Nov 07 10:00:21 <Tee_Pee> aye, but I'm testing this on an already cached data set, so isCached(position) returns true every time.
Nov 07 10:00:29 <Tee_Pee> the else if isn't being called
Nov 07 10:00:49 <baptistem> superlinux-hp: err I guess that your dev option is on but I add the same problem when I use MTP instead of MSC as usb connectivity maybe it will solve your problem
Nov 07 10:01:25 <vascobnunes> Anyone want to team up to help me develop a small sports club app? :D
Nov 07 10:01:29 <tdignan> findAlbumById can't be that fast
Nov 07 10:01:50 <superlinux-hp> baptistem, yes there is something about MTP but i don't understand what it is and how to deal with it
Nov 07 10:01:55 <tdignan> I also question why you are using 5 array lists to hold entities in relation dependent on their position therein, instead of defining a custom object type and making one array list of that
Nov 07 10:02:47 <Tee_Pee> I don't know honestly, most of that was just "let's get it to display stuff on the screen, worry about performance later"
Nov 07 10:02:50 <Tee_Pee> well, now is later :p
Nov 07 10:02:52 <superlinux-hp> baptistem, it was off in the settings
Nov 07 10:02:59 <tdignan> this is more about clean code
Nov 07 10:03:10 <sNore> whos experienced with audiotrack?
Nov 07 10:03:12 <baptistem> superlinux-hp: MTP mean that your phone is a kind of gateway between computer and phone's memory card, MSC will just mount your memory card, sometime adb complain about installing if it's in MTP or MSC
Nov 07 10:03:13 <tdignan> object { int a; int b; int c; } arraylist<object> you get it
Nov 07 10:03:18 <tdignan> I'm going to get coffee and do my work
Nov 07 10:03:21 <Tee_Pee> anyway, let me implement the tag solution instead of findViewById and see what happens then
Nov 07 10:03:26 <Tee_Pee> aye, thanks for help
Nov 07 10:03:26 <tdignan> i'll be on tonight
Nov 07 10:03:41 <superlinux-hp> baptistem, it works now
Nov 07 10:04:04 <baptistem> superlinux-hp: perfect! did you change it?
Nov 07 10:04:23 <superlinux-hp> yes i did
Nov 07 10:04:38 <superlinux-hp> in the applications development settings
Nov 07 10:04:51 <superlinux-hp> they were off
Nov 07 10:04:57 <baptistem> glad that it's working for you now :)
Nov 07 10:05:04 <superlinux-hp> thanks mate.
Nov 07 10:05:35 <baptistem> your welcome!
Nov 07 10:06:32 <superlinux-hp> baptistem, do you have some experience in live video streaming?
Nov 07 10:07:23 <superlinux-hp> there are some sites are not using rtsp. instead they are using rtmp and rtpmpe protocols.
Nov 07 10:07:38 <baptistem> superlinux-hp: no sorry but many poeple here had issues with video streaming, I'm sure you will find what you want here
Nov 07 10:07:45 <superlinux-hp> so what should i do?
Nov 07 10:07:56 <superlinux-hp> baptistem, i understand...
Nov 07 10:08:40 <superlinux-hp> baptistem, do you ve experience then in running things from command line? "rtmpdump" commend can do me what i want.
Nov 07 10:08:46 <superlinux-hp> baptistem, do you ve experience then in running things from command line? "rtmpdump" command can do me what i want.
Nov 07 10:09:13 <baptistem> err I never heard of rtmp :)
Nov 07 10:09:20 <superlinux-hp> ah ok
Nov 07 10:09:57 <superlinux-hp> baptistem, thanks
Nov 07 10:18:16 <vascobnunes> How would I share a SVN repository with someone?
Nov 07 10:19:40 <Snuffel> rtfm?
Nov 07 10:19:50 <appel1> give them the url to the server and an account
Nov 07 10:22:25 <g00s> vascobnunes: maybe #svn :)
Nov 07 10:24:27 <vascobnunes> appel1: How do I give them an account?
Nov 07 10:25:17 <appel1> vascobnunes: http://svnbook.red-bean.com/
Nov 07 10:25:41 <miha> does anyone use paypal API with android? is there a way to disable asking for damn PIN?
Nov 07 10:29:15 <vascobnunes> appel1: thank you
Nov 07 10:38:22 <vascobnunes> appel1: but tell me, does it make sense to use svn for an android app development?
Nov 07 10:42:03 <root66> http://i.imgur.com/H0pMt.jpg
Nov 07 10:42:09 <root66> ha haaaaaaaaaaaaaa
Nov 07 10:44:20 <Zider> root66: they did board games from mobile games based on board games? :P
Nov 07 10:48:30 <root66> I can't wait to adapt my super mario board game to the video game platform
Nov 07 10:55:17 <Zider> isn't there even an "IRL" angry birds game?
Nov 07 10:55:51 <Zider> http://static.gamesradar.com/images/mb/GamesRadar/us/Daily/2011/01-Jan/07/angry%20birds%20board%20game/angry_birds_boardgame_1--article_image.jpg yeah.. :P
Nov 07 10:56:03 <root66> well, there was an episode of Tosh.0 where he slingshots actual dead birds at his staff
Nov 07 10:56:47 <root66> my god
Nov 07 10:56:56 <timroes> What is the best REST client for android?
Nov 07 10:57:11 <root66> you mean a library for rest communications? or...
Nov 07 10:57:14 <timroes> jap
Nov 07 10:58:12 <root66> check out Jersey
Nov 07 10:58:18 <root66> I havn't used it though
Nov 07 10:58:22 <timroes> doesnt work under android as far as i read
Nov 07 10:58:41 <root66> oh
Nov 07 10:58:48 <timroes> otherwise i would use it, since my serverside is jersey
Nov 07 10:59:03 <root66> I mean, to be completely honest, I would write my own
Nov 07 10:59:11 <root66> because all of the libraries I have seen for REST were bloated
Nov 07 10:59:31 <root66> it's so simple it hardly requires using a library
Nov 07 11:00:08 <root66> same goes for most web-based api's and crap I see
Nov 07 11:00:44 <miha> REST is supposed to be simple anyway :)
Nov 07 11:01:11 <timroes> thats why i thought there might be some simple rest lib out there
Nov 07 11:01:12 <miha> timroes: just learn to use HttpClient :)
Nov 07 11:01:22 <timroes> i know how to use httpclient...
Nov 07 11:01:32 <Sera6x7> my android game supports all sceen sizes, still the 'stretch/zoom to fill screen" button appears
Nov 07 11:01:42 <Sera6x7> is there any way to prevent this button from appearing?
Nov 07 11:01:47 <root66> I agree with miha.. and in my experience, any time someone writes a library to do make something already-simple "more simple" they bloat it and screw shit up
Nov 07 11:02:19 <root66> Sera6x7: I have never even seen such a button in my life
Nov 07 11:02:41 <root66> games have a "stretch/zoom to fill screen" button?
Nov 07 11:02:45 <miha> Sera6x7: i think you need to change targetSdkVersion to android 3.x+ http://developer.android.com/guide/topics/manifest/uses-sdk-element.html
Nov 07 11:02:54 <miha> i remember something like that
Nov 07 11:03:05 <root66> oh, a new feature
Nov 07 11:03:18 <root66> I should really get a new phone
Nov 07 11:03:29 <root66> still rocking the vivow
Nov 07 11:03:46 <miha> root66: it's compatibility for "old apps"
Nov 07 11:03:52 <Sera6x7> root66: this button is available for all apps that don't explicitly supports large screens
Nov 07 11:03:54 <root66> yeah, makes sense
Nov 07 11:04:26 <Sera6x7> miha: I already target SDK 13, and explicitly support all screen sizes with a  width > 320dp
Nov 07 11:05:54 <SpecialDragonfly> Morning all
Nov 07 11:07:15 <miha> Sera6x7: beats me, sorry
Nov 07 11:07:38 <Sera6x7> miha: thanks anyway
Nov 07 11:07:44 <SpecialDragonfly> Under eclipse, is it possible to test changes in signal strength? I've looked online and only found people mocking the connectivity manager - is there not a way to say, "Assume I'm on a GSM connection"?
Nov 07 11:08:36 <pragma-> https://www.youtube.com/watch?v=mp07dPusJNA&feature=youtube_gdata_player
Nov 07 11:08:41 <pragma-> 7 taps on about phone -> build version to unlock developer settings
Nov 07 11:08:44 <Estel> Sera6x7: have you read http://developer.android.com/guide/practices/screen-compat-mode.html ?
Nov 07 11:09:26 <pragma-> though I think it's embarrassing that they didn't code in a simple plural toggle so that it says "1 tap" instead of "1 taps"
Nov 07 11:09:40 <pragma-> seriously amateurish mistake
Nov 07 11:09:46 <Chris> definitely
Nov 07 11:10:21 <Sera6x7> Estel: I have, and I believe I have all the required settings in my manifest (including the deprecated varieties)
Nov 07 11:13:14 <Sera6x7> ah, apparently if your manifest contains <supports-screens>, even though the app runs on one of the supported screens, this compatibility button appears
Nov 07 11:14:02 <Sera6x7> so, the compatibility button disappears by not explicitly declaring that I support all screen sizes
Nov 07 11:14:16 <Estel> That sounds... wrong
Nov 07 11:15:20 <Sera6x7> well, the doc you linked to states, in the last section: "you can affect whether compatibility mode is enabled for certain screens by using the <supports-screens> element."
Nov 07 11:15:22 <MDijkstra> that makes sense
Nov 07 11:15:41 <MDijkstra> ie, if android knows the app doesn't support the particular screen size of the device it is running on
Nov 07 11:15:47 <MDijkstra> it shows the zoom button
Nov 07 11:16:09 <MDijkstra> if it claims to support all sizes, there's no point in showing the zoom button
Nov 07 11:17:05 <Sera6x7> MDijkstra: in  <supports-screens> I had listed all screens wider than 320dp, and I run the app on a 10" tablet. I don't think it makes sense to show the zoom button in that case
Nov 07 11:17:45 <Sera6x7> a follow-up question: assuming the compatibility button is there, can I handle the configuration change my self?
Nov 07 11:17:55 <MDijkstra> wider in portrait?
Nov 07 11:18:08 <Sera6x7> I already handle android:configChanges="keyboard|keyboardHidden|orientation|screenSize|smallestScreenSize|screenLayout|uiMode" but apparently the compatibility zoom is not included in that list
Nov 07 11:19:05 <MDijkstra> Sera6x7: in the docs:
Nov 07 11:19:19 <MDijkstra> "a typical handset screen has a smallestWidth of 320dp, a 7" tablet has a smallestWidth of 600dp, and a 10" tablet has a smallestWidth of 720dp."
Nov 07 11:19:39 <MDijkstra> so it makes perfect sense to show the zoom button in such a case
Nov 07 11:20:02 <MDijkstra> or wait
Nov 07 11:20:05 <MDijkstra> disregard that
Nov 07 11:20:29 <MDijkstra> what exactly do you have there?
Nov 07 11:20:50 <Estel> o_O
Nov 07 11:20:59 <MDijkstra> specifically, do you have android:compatibleWidthLimitDp specified?
Nov 07 11:24:04 <Sera6x7> the docs for ""requiresSmallestWidthDp" says  "the device's smallestWidth must be equal to or greater than this value" but it concludes with: "If your application properly resizes for smaller screen sizes (down to the small size or a minimum width of 320dp), you do not need to use this attribute."
Nov 07 11:24:22 <Sera6x7> I guess it pays to not include unnecessary stuff in your manifest
Nov 07 11:25:11 <MDijkstra> yeah, that's the wrong attribute, you want android:compatibleWidthLimitDp
Nov 07 11:26:13 <Sera6x7> not according to my reading of the docs: "the maximum "smallest screen width" for which your application is designed", while my application is designed to support all screen sizes
Nov 07 11:26:32 <MDijkstra> it can't be designed for 'all' screen sizes
Nov 07 11:26:57 <MDijkstra> you can't predict an android device won't come out with a 200000dp width
Nov 07 11:27:23 <MDijkstra> there's always some subset of screens you target
Nov 07 11:27:42 <MDijkstra> in your case it's everything from a 3" phone screen to a 10" tablet
Nov 07 11:27:46 <MDijkstra> but it's not every screen
Nov 07 11:28:08 <miha> yeah need to support 1" android watch and 120" android tv too
Nov 07 11:29:08 <Sera6x7> my game uses gles2 with its own ui scaling, so yes it 'supports' all screen sizes
Nov 07 11:29:44 <feesh> hi guys
Nov 07 11:29:54 <feesh> (long time no see), how do you quit a HandlerThread?
Nov 07 11:30:05 <feesh> mHandlerThread.quit() is just hanging indefinitely
Nov 07 11:32:16 <MDijkstra> Sera6x7: again, that doesn't support all screen sizes
Nov 07 11:32:43 <MDijkstra> android device with 2dpx1000000dp
Nov 07 11:34:07 <MDijkstra> some kind of android powered ticker display
Nov 07 11:34:13 <Sera6x7> MDijkstra: that's why I included the minimal width of 320dp
Nov 07 11:34:40 <MDijkstra> minimal width says nothing about the ratio
Nov 07 11:38:12 <Sera6x7> MDijkstra: from those same docs: "The smallestWidth is the shortest dimension of the screen space (in dp units) that must be available to your application UI—that is, the shortest of the available screen's two dimensions. "
Nov 07 11:38:57 <Sera6x7> so no, it says nothing about the ratio but it clearly defines which dimension is the 'width' of the screen, and my app has no issue with varying ratios
Nov 07 11:41:23 <MDijkstra> Sera6x7: again, if you just declare support for larger screens the issue will go away
Nov 07 11:41:30 <MDijkstra> so I don't know what exactly you want
Nov 07 11:41:53 <Sera6x7> MDijkstra: yes, that's what I said ;)
Nov 07 11:42:44 <MDijkstra> O_o
Nov 07 11:42:48 <MDijkstra> ok, well, good luck
Nov 07 11:43:58 <Tee_Pee> http://pastebin.com/gCSY7DyE  <----This is driving me insane... any idea what might cause this?
Nov 07 11:44:30 <Tee_Pee> Line 42 is the inflater.inflate line in the first constructor
Nov 07 11:45:00 <Tee_Pee> the albumadapter line 119 is just albumView = new PycsellAlbumView(mContext);
Nov 07 11:45:39 <Tee_Pee> if I comment out the two inflater lines, and uncomment the Layoutinflater.from(context) line, the error changes
Nov 07 11:48:36 <vascobnunes> has anyone used this: http://caughtinthemobileweb.wordpress.com/2011/06/20/how-to-implement-calendarview-in-android/
Nov 07 11:48:37 <vascobnunes> ?
Nov 07 11:50:18 <Tee_Pee> 'fraid not
Nov 07 11:51:09 * miha bugging paypal forums.. i feel bad
Nov 07 11:51:25 <miha> what's in their samples works great, what's not is mystery :D
Nov 07 11:51:32 <miha> no javadoc or anything
Nov 07 11:52:03 <miha> pdf has even less details than Simple Sample
Nov 07 11:52:17 * miha spam :p
Nov 07 11:54:58 <Tee_Pee> Trying to figure out what the hell is bottlenecking my gridview scrolling performance
Nov 07 11:58:47 <Gaz`> getting a weird error, fixed this once before but i cant remember how, noClassDefFounError org.apache.http.entity.mime.MultipartEntity, thing is its definitely there
Nov 07 11:59:08 <SpecialDragonfly> Trying to set the Android emulator to use GPRS by running the emulator executable from the command line and using -netspeed gprs, it's not actually doing anything. Can anyone offer any advice?
Nov 07 12:00:14 <SpecialDragonfly> Gaz`: Sounds like a build path issue.
Nov 07 12:03:49 <Gaz`> Yeh,
Nov 07 12:03:53 <Gaz`> and bloody eclipse
Nov 07 12:06:04 <Gaz`> dalvik errors now,
Nov 07 12:10:17 <SpecialDragonfly> Glad I'm not the only one with Dalvik errors.
Nov 07 12:10:39 <Hodapp> Try blaming it on Obama.
Nov 07 12:11:01 <SpecialDragonfly> I might, but I'm British, so blaming it on Romney is more my thing ;)
Nov 07 12:11:28 <Hodapp> I find that it's effective to simply add "THANKS, Obama." sarcastically to the end of anything angry or irritated that you say.
Nov 07 12:12:24 <SpecialDragonfly> Gaz`: Any luck with the dalvik errors?
Nov 07 12:22:43 <Gaz`> yes
Nov 07 12:22:53 <Gaz`> some order and export crap
Nov 07 12:23:00 <Gaz`> and when i was ticking some i hit android.jar
Nov 07 12:23:05 <Gaz`> so it must have been trying to use it twice
Nov 07 12:23:15 <Gaz`> solved all my issues now :) finally
Nov 07 12:23:21 <Gaz`> now im trying to talk to my damn servlet
Nov 07 12:26:07 <Gaz`> altho getting a servlet to work is like taming a shrew
Nov 07 12:26:11 <Tee_Pee> so.... finally traced my gridview performance dip to my scroll listener... sigh.
Nov 07 12:27:38 <Gaz`> ;/
Nov 07 12:27:58 <Tee_Pee> or rather
Nov 07 12:28:10 <Tee_Pee> to the notifyDataSetChanged() called on the adapter from the scroll listener.
Nov 07 12:28:52 <muszek> hi... I'm trying to localize my app's label below the icon on the home screen. it appears android:label (in AndroidManifest.xml) is not responsible for this. It appears only the label in settings/applications is affected by android:label... how do I change it?
Nov 07 12:28:55 <Tee_Pee> What is essentially happening is this - I have an image gallery, and if the user is just zooming past, I don't want the images to load at all... so I track the speed of the scroll via listener... if the speed dips below a certain value, and the user was flinging earlier, I tell the adapter to load shit.
Nov 07 12:42:27 <brickhead> Hey Guys I'm encrypting my files in Android
Nov 07 12:42:31 <brickhead> it works well with CipherInputStream/CipherOutputStream, but I'm running in issues with RandomAccessFile
Nov 07 12:42:38 <brickhead> here's what I'm doing to encrypt the data using RAF http://pastebin.com/hzZTN9zJ
Nov 07 12:43:30 <brickhead> basically I'm using AES/CBC/PKCS5Padding so  I'm the blocks are 16 byte aligned
Nov 07 12:44:31 <adq> do you convert in base64 your binary data?
Nov 07 12:44:47 <brickhead>  adq  no
Nov 07 12:45:06 <adq> hmmm not sure (has to be verified by more ppl from here), but it might be an issue
Nov 07 12:45:19 <adq> because if you end up with a \0 in binary
Nov 07 12:45:27 <adq> s/if/you/
Nov 07 12:45:44 <adq> errr you might end up* i meant
Nov 07 12:46:04 <Zharf> so?
Nov 07 12:46:22 <brickhead> hmm... so tried couple of files with Cipher*Stream functions they seems to be working
Nov 07 12:46:54 <Zharf> adq, bytes don't care if it's 0
Nov 07 12:46:57 <adq> i'm probably wrong :|
Nov 07 12:47:05 <adq> or is it a "padding" problem?
Nov 07 12:47:20 <brickhead> its most likely a padding issue,
Nov 07 12:47:51 <brickhead> because when I seek, the rf.length() is always in multiple of 16 where the offset server ask me to write at varies
Nov 07 12:48:13 <adq> yeah, so you have to pad it to ensure the blocks is always the correct length
Nov 07 12:48:19 <brickhead> so I guess PKCs5Padding is mandating a certain padding
Nov 07 12:48:47 <brickhead> so that's what I thought I did http://pastebin.com/hzZTN9zJ
Nov 07 12:50:27 <adq> brickhead, is your test on "iif (offset != 0)" should not be against the value of padding instead?
Nov 07 12:51:21 * miha thinks it's brave to do such code
Nov 07 12:51:23 <adq> and then add the necessary remaining byte as padding to fill the correct length of the block (so 16 in this case)
Nov 07 12:52:41 <brickhead> adq: yeap that could be it for 16 byte aligned I would be adding 16 to it
Nov 07 12:54:23 <adq> what i meant is, if you have per example 2 blocks which in total is per example 20 bytes, the first block is 16bytes (no need padding here) and the second block is 4 bytes and need 12 bytes of padding
Nov 07 12:55:05 <miha> isnt that padding 16 bytes block, makes 32 bytes block? cause you need at least 1 byte for padding?
Nov 07 12:55:13 <miha> :p
Nov 07 12:55:22 <miha> two 16 bytes blocks, i mean
Nov 07 12:56:02 <adq> no i don't think so, if it is exactly 32 bytes in total, there is no need of padding
Nov 07 12:56:40 <brickhead> adq: Thanks that did it, stupid coding by me
Nov 07 12:56:43 <adq> cool
Nov 07 12:57:03 <adq> i played a lot with ECB/CBC long time ago, so it's not fresh in my head anymore :)
Nov 07 12:57:18 <miha> adq: ECB is bad m'key
Nov 07 12:57:19 <adq> worst i run into was the GP (global platform)
Nov 07 12:59:39 <brickhead> I was earlier using default encryption scheme, but was later suggested to explicitly define mode and padding and then IV
Nov 07 12:59:57 <brickhead> but now it looks like the whole encryption thing is darn slow
Nov 07 13:01:49 <adq> IV chaining improves a bit the security by preventing decryption if someone eavesdrops in the middle of the "conversation"
Nov 07 13:03:28 <adq> brickhead, FYI: http://en.wikipedia.org/wiki/Weak_key choose carefully your key ;)
Nov 07 13:05:50 <root66> anyone here interested in producing raw sound waves? If so I could fill your mind with awesome right now
Nov 07 13:06:24 <root66> I just wrote a synthesizer and it was so incredibly simple.. nothing like "they" try to teach you
Nov 07 13:06:33 <root66> when you google for examples, that is
Nov 07 13:06:54 <root66> in fact I might just write a tutorial
Nov 07 13:07:02 <root66> screw all you
Nov 07 13:12:13 <lasserix> 100% functionality implemented booya!
Nov 07 13:12:35 <adq> root66, I did one too :) I use like a software VCF, VCA and VCO
Nov 07 13:12:49 <adq> which produces sound clearly close (and also so  far) to a bassline TB303 :))
Nov 07 13:13:23 <adq> it is not much resource/cpu consuming compared to applying effect like chorus/flanger (because of some expensive maths functions)
Nov 07 13:14:06 <adq> root66, I'm looking forward to read your tutorial
Nov 07 13:23:59 <j4hr0m> hi guys, i have an Application Object, that's being created when the Application starts. I want it to call onTerminate when the main Activity stops/terminates. How can i do this?
Nov 07 13:24:16 <j4hr0m> it is never being calles
Nov 07 13:24:19 <j4hr0m> called
Nov 07 13:24:31 <alankila> yes, that's what the documentation says. It will never be called.
Nov 07 13:24:46 <alankila> well, there's some circumstance with emulator or something where it is, apparently, but not on real device.
Nov 07 13:25:07 <alankila> perhaps you'd be interested in the activity lifecycle methods?
Nov 07 13:25:19 <j4hr0m> should i use onStop?
Nov 07 13:25:41 <alankila> if onStop fits what you are doing, then you should use onStop. Also tautological answer is tautological.
Nov 07 13:26:03 <adq> true is true, it's a tautology
Nov 07 13:26:08 <j4hr0m> makes sense
Nov 07 13:26:10 <j4hr0m> ^^
Nov 07 13:26:21 <j4hr0m> but gives no information gain
Nov 07 13:26:30 <adq> wrong :)
Nov 07 13:27:07 <j4hr0m> thx btw
Nov 07 13:28:24 <alankila> the general sentiment about Activity subclasses is "don't". In general google's engineers would like you to use the lifecycle methods on the android-managed objects such as fragment, activity, and service
Nov 07 13:28:51 <j4hr0m> ok, again. the Application object has nothing to do with Activity lifecycle
Nov 07 13:28:56 <ron_frown> alankila - awefully broad statement
Nov 07 13:29:16 <j4hr0m> Application dosn't implement onStop, etc. only onTerminate()
Nov 07 13:29:48 <alankila> j4hr0m: indeed. And onTerminate is not called, probably to not allow applications to hold up termination of their processes. So...
Nov 07 13:30:44 <j4hr0m> so how can i tell if the application is closed?
Nov 07 13:31:15 <alankila> there's no such thing on android. The general illusion is that applications are always running, and can be programmed to act as if this were true. Whether the application process is currently in memory shouldn't matter to end user, or even to the programmer in many cases.
Nov 07 13:32:04 <j4hr0m> this is not satisfyinfg. i want to clean up temporary files, where can i do this?
Nov 07 13:32:37 <qkzoo1978> Hey. I'm having difficulty creating an intent properly, my http://stackoverflow.com/questions/13260592/android-unable-to-properly-create-intent-to-share-kmlxml-file any thoughts from you folks what I'm doing wrong?
Nov 07 13:32:39 <adq> j4hr0m,  you override "onDestroy", maybe?
Nov 07 13:32:43 <alankila> perhaps you can remove temporary files on the creation of the activity that made them previous time.
Nov 07 13:33:02 <adq> or at creation as alankila said
Nov 07 13:33:06 <alankila> it all depends on how long the temporary files are useful
Nov 07 13:33:35 <appel1> adq: onDestroy isn't guaranteed to get called..
Nov 07 13:33:41 <aksu> is there a way to take programmatically screenshots of open gl view that is on native side. I am seeing only black screenshots
Nov 07 13:33:42 <alankila> also there's File#deleteAtExit() or some such method, which may or may not work on android. I imagine it does, but in case the application terminates abruptly you still need a back-up plan
Nov 07 13:33:47 <adq> appel1, ah thx :/
Nov 07 13:34:06 <ron_frown> whats wrong with activity subclasses, or were you alking about a particular very specific cse
Nov 07 13:34:07 <ron_frown> case
Nov 07 13:34:13 <j4hr0m> ok, i will need an extra Boolean variable, that tells me whether onDestroy is called to move to another Activity in the App or back to Android... this could helpo. thanks
Nov 07 13:34:15 <alankila> ron_frown: application subclasses?
Nov 07 13:34:32 <tdignan> wow. facebook's android sdk login flow is absolutely stupid
Nov 07 13:34:45 <ron_frown> oh
Nov 07 13:34:48 <adq> aksu, i think you need to draw the opengl buffer in a bitmap :/
Nov 07 13:34:54 <tdignan> "LoginButton" just drop it in! Then it spawns an ugly as hell webview over the current activity that doesn't survive orientation changes.
Nov 07 13:34:57 <adq> then display and screenshot the bitmap
Nov 07 13:35:06 <ron_frown> yeah I can see that
Nov 07 13:35:34 <ron_frown> not much about application subclass is actually all that useful
Nov 07 13:35:38 <j4hr0m> appel1: then what is guaranteed to be called?
Nov 07 13:35:45 <appel1> j4hr0m: onPause
Nov 07 13:35:57 <j4hr0m> ok, thanks
Nov 07 13:36:08 <alankila> I thought onDestroy is guaranteed though
Nov 07 13:36:21 <appel1> see Figure 1 on http://developer.android.com/guide/components/activities.html#Lifecycle
Nov 07 13:36:27 <adq> if someone kill your pid, ondestroy is probably not called
Nov 07 13:36:32 <adq> but this should not happen often
Nov 07 13:36:33 <appel1> after onPause the process can get killed at any time
Nov 07 13:36:35 <alankila> I see, it's not
Nov 07 13:36:39 <j4hr0m> of course ;) see this so many times, but still relevant
Nov 07 13:37:32 <alankila> the lifecycle chart is kinda vague but it suggests that after onPause or onStop, there can be death of activity without onDestroy
Nov 07 13:37:36 <appel1> adq: it is enough that memory pressure is high on the device for it to happen, no manuall killing necessary
Nov 07 13:37:58 <miha> qkzoo1978: looking at your example, i do Intent share = new Intent( Intent.ACTION_SEND);
Nov 07 13:38:10 <miha> qkzoo1978: other than that, you do much the same
Nov 07 13:38:28 <j4hr0m> kinda OT, but is this the same, where windows 8 is heading? assuming, applications never get killed? :D
Nov 07 13:39:00 <alankila> I kinda thought that there is always onDestroy call if there is onCreate call. This is unsatisfying to me, I kinda hoped they'd form a logical pair.
Nov 07 13:39:03 <qkzoo1978> Miha tu, I'll check it out!
Nov 07 13:39:05 <adq> appel1, yes, totally agreed
Nov 07 13:43:21 <qkzoo1978> Miha, that was it!
Nov 07 13:43:33 <miha> qkzoo1978: cool
Nov 07 13:44:17 <qkzoo1978> Why isn't by listed for sharing?
Nov 07 13:44:27 <qkzoo1978> Bluetooth
Nov 07 13:47:22 <miha> qkzoo1978: don't know. i have bluetooth too, but i dont set mail, subject and body, just attach file
Nov 07 13:47:39 <miha> also dropbox, skype, andftp :)
Nov 07 13:48:17 <qkzoo1978> I'll give it a shot, thanks!
Nov 07 13:49:28 <miha> qkzoo1978: and if you want to attach multiple files, i think http://stackoverflow.com/questions/2264622/android-multiple-email-attachments-using-intent this works
Nov 07 14:01:23 <Tee_Pee> Anyone here had the pleasure of working with gridviews that display cached images?
Nov 07 14:01:28 <Tee_Pee> from the sdcard that is
Nov 07 14:01:31 <Tee_Pee> i.e. a gallery
Nov 07 14:03:06 <qkzoo1978> Thanks miha
Nov 07 14:04:24 <Tee_Pee> I'm trying to get rid of scrolling performance issues the entire day now, and I can't seem to do it... my adapter is as minimal as can be - this is the getView() function
Nov 07 14:04:24 <Tee_Pee> http://pastebin.com/TXdGwrJK
Nov 07 14:05:08 <Tee_Pee> The line that causes the performance hiccup seems to be within the PycsellAlbumView class, namely the setImage(String imagepath) method, which just does this.imageView.setImageBitmap(BitmapFactory.decodeFile(imagepath));
Nov 07 14:05:29 <Tee_Pee> so the problem lies within the setting and / or decoding of the bitmap
Nov 07 14:05:47 <Tee_Pee> I've tried using the official guide and implement it that way - http://developer.android.com/training/displaying-bitmaps/process-bitmap.html
Nov 07 14:06:00 <Tee_Pee> (See Handle Concurrency)
Nov 07 14:06:03 <j0hnsm1th> image processing is always gonna be expensive...
Nov 07 14:06:06 <Tee_Pee> But that frankly just made it worst.
Nov 07 14:06:18 <Tee_Pee> worse*
Nov 07 14:08:11 <Denommus> hi
Nov 07 14:08:26 <j0hnsm1th> Tee_Pee, you can try using the ViewHolder pattern too
Nov 07 14:08:32 <miha> Tee_Pee: what do you mean worst? it shouldnt block? :)
Nov 07 14:08:37 <Gaz`> hmm the result im getting back from submitting an image to my servlet is the html of the page, anyone know what im doing wrong, should be receiving a jpeg, it works from browser
Nov 07 14:08:44 <Tee_Pee> miha - typo, I corrected to worse.
Nov 07 14:08:48 <j0hnsm1th> Tee_Pee, oh u are, nvm
Nov 07 14:08:52 <Tee_Pee> johnsmith - I am already doing that?
Nov 07 14:09:22 <Tee_Pee> checking the source code of the gallery now lol
Nov 07 14:09:25 <Tee_Pee> to see how it did it
Nov 07 14:09:41 <root66> adq: getting to the good bits now... but here it is so far: http://johnnystreet.com/synthesizer_tutorial.html
Nov 07 14:09:59 <root66> I am trying to explain it to you like you are five years old
Nov 07 14:10:07 <adq> add pictures, if so :)
Nov 07 14:10:08 <miha> Tee_Pee: i dont know how bad my idea is, but  http://developer.android.com/guide/topics/ui/layout/gridview.html you could spawn your bitmap loading task in getView()
Nov 07 14:10:13 <root66> yeah, definitely
Nov 07 14:10:26 <root66> let me know what you think so far though
Nov 07 14:10:31 <root66> since that is most of the nontechnical stuff
Nov 07 14:10:48 <Tee_Pee> miha - not sure what you mean?
Nov 07 14:10:51 <root66> especially let me know if any of the wording seems confusing
Nov 07 14:10:52 <Tee_Pee> -?
Nov 07 14:11:07 <miha> Tee_Pee: in that page, instead of imageView.setImageResource(mThumbIds[position]);
Nov 07 14:11:32 <miha> you would check if you have bitmap in memory, or start task and set bitmap later?
Nov 07 14:12:09 <miha> and just return empty imageView
Nov 07 14:12:18 <Tee_Pee> I've had a similar setup earlier, didn't make much of a difference... either way, I forgot to mention that, if I replace this.imageView.setImageBitmap(BitmapFactory.decodeFile(imagepath));    with this.imageView.setImageResource(R.drawable.SOME_IMAGE); it works flawlessly.
Nov 07 14:12:31 <Tee_Pee> So it seems to be an issue with the decoding of the file.
Nov 07 14:14:43 <j0hnsm1th> Tee_Pee, I'm no expert here, but maybe decode the file off the UI thread and load the images when they're ready.. ?
Nov 07 14:14:47 <root66> adq: I am working on images and a header breakdown now
Nov 07 14:14:58 <j0hnsm1th> Tee_Pee, image processing is always going to be expensive
Nov 07 14:15:16 <root66> I feel like I should have an actual sine wave and not something crudely bent in photoshop
Nov 07 14:15:37 <root66> but I am not writing a program just for that so I am stealing something off google
Nov 07 14:15:38 <Tee_Pee> johnsmith - it would seem that setImageBitmap works off the UI thread anyway
Nov 07 14:15:51 <root66> oh wait, wiki commons!
Nov 07 14:15:52 <adq> root66, there are lot of similar tutorial but it's good you show some altruism. Think of a pedagocial pov in what you explain, what are the basics, what is advanced, etc.. Also to entertain your readers, maybe add some funny anecdote like why a CD was 650MB ( Beethoven's Ninth Symphony on one disc in 16bits PCM fit exactly in it), etc..
Nov 07 14:15:54 <Tee_Pee> Let's see if I can do anything with bitmapoptions...
Nov 07 14:16:43 <j0hnsm1th> Tee_Pee, this is true, but doing the decoding on the UI thread is what is giving you perf issues
Nov 07 14:16:44 <adq> s/pedagocial/pedagogical/
Nov 07 14:17:06 <Tee_Pee> sorry, I meant decodeFile doesn't work on the UI thread
Nov 07 14:17:17 <j0hnsm1th> if you dont mind some delay for the images popping up, decoding them off the UI thread or doing some other less expensive procesing is going to be the option I think...
Nov 07 14:18:24 <root66> most stuff I do like this is broken down into ONLY the information you need to do it with no distractions. I will take that into consideration, though.
Nov 07 14:18:36 <alankila> root66: it's a pity that wave editors as a rule do not visualize waveforms correctly. The waveform editors usually draw a straight connecting line between the samples in the waveform, but that's not actually what hardware systems generate. They do something more complicated in between those points.
Nov 07 14:19:12 <alankila> you refer to the waveform as a line somewhere, and much confusion about digital audio arises from this misconception when people begin to argue why, for instance, true analog waveforms are better.
Nov 07 14:19:27 <alankila> and invariably, the lines-connecting-points-of-digital-waveform model is there
Nov 07 14:19:48 <adq> cannot agree more
Nov 07 14:20:37 <root66> yeah, it is just a series of values going from left to right. the fixed sample rate is the limitation of digital audio
Nov 07 14:20:38 <alankila> however, for low-frequency sounds the lines model is very accurate.
Nov 07 14:20:51 <Titi> Hi
Nov 07 14:21:11 <root66> i mean, a sine wave of digital audio is not a true sine wave
Nov 07 14:21:30 <alankila> for high frequency sounds, the reconstruction filter comes into play and it's probably a sinc of some type
Nov 07 14:21:32 <root66> in between the two 1/44100 samples is a tiny amount of missing data
Nov 07 14:21:49 <alankila> only if the source audio violates the nyquist theorem
Nov 07 14:21:53 <root66> and the data becomes more important as you get into higher freqs
Nov 07 14:22:02 <alankila> if not, then the discrete sampling can reconstruct the original waveform perfectly.
Nov 07 14:22:19 <root66> but isn't that done by the thing rendering the sound?
Nov 07 14:22:24 <root66> the data is still missing
Nov 07 14:22:36 <root66> it is just predicted by the player
Nov 07 14:22:47 <alankila> it's only "missing" if it existed in the first place. Digital waveforms can represent all sound data of frequencies less than the nyquist frequency perfectly.
Nov 07 14:22:48 <root66> and with such a tiny sample size it is going to be damn close
Nov 07 14:22:48 <miha> root66: "sampling frequency must be at least 2 times highest signal frequency" :)
Nov 07 14:22:58 <alankila> anything above it is aliased or missing, depending on the hardware's capability
Nov 07 14:23:08 <root66> miha: I did not know that
Nov 07 14:23:11 <root66> that makes total sense
Nov 07 14:23:15 <miha> alankila told you
Nov 07 14:23:16 <Titi> In a idea of performance I would like to know how it is the best practice of using HttpURLConnection
Nov 07 14:23:29 <feesh> nyquist frequencies and all that
Nov 07 14:23:38 <root66> this is why I love this channel
Nov 07 14:23:42 <root66> and not so much #math
Nov 07 14:23:58 <alankila> unfortunately I need to get going about now, so I can't continue discussion, but remember the nyquist's theorem. It's no idle talk. A lot of people seem to not understand it.
Nov 07 14:24:33 <root66> thanks
Nov 07 14:24:48 <Titi> my phone take 154ms on this line : final HttpURLConnection res = (HttpURLConnection) (url.openConnection()); is it normal ?
Nov 07 14:25:23 <feesh> Titi: when you're debugging it's going to go slower
Nov 07 14:25:43 <Titi> oh, ok :)
Nov 07 14:25:55 <feesh> Titi: compared to actually retrieving that data anyway?
Nov 07 14:26:04 <feesh> as in making the connection etc ?
Nov 07 14:26:08 <feesh> pocket change!
Nov 07 14:26:20 <root66> so am I essentially wrong that a wav is a single line from left to right?
Nov 07 14:26:31 <root66> should I avoid referring to it as a line?.
Nov 07 14:26:42 <feesh> root66: a wav is a list of amplitudes
Nov 07 14:26:46 <Titi> no just in this new
Nov 07 14:27:18 <Titi> for the connection, and the exchange of data, he took between 350 and 7000ms
Nov 07 14:27:27 <feesh> at every sample point, you have an amplitude, if you plot those amplitudes on a graph, you get a line
Nov 07 14:27:55 <Titi> depending a lot on the connection line : _code = urlConnection.getResponseCode();
Nov 07 14:28:08 <root66> Well duh, I just meant in my explanation there, but yeah I am going to change the wording
Nov 07 14:30:52 <Tee_Pee> whoo, that actually improved stuff.
Nov 07 14:31:00 <Tee_Pee> Moving the decoding into an asynctask
Nov 07 14:32:17 <j0hnsm1th> Tee_Pee, great success! I was going to encounter this problem in a few days so glad to here its working
Nov 07 14:32:54 <Tee_Pee> To tell you the truth, it's still not PERFECTLY smooth
Nov 07 14:33:06 <j0hnsm1th> compared to a vanilla listview?
Nov 07 14:33:16 <Tee_Pee> and considering I'm on a galaxy s3
Nov 07 14:33:24 <Tee_Pee> it should not stutter :p
Nov 07 14:33:30 <Tee_Pee> anyway, need to run to the shop, be back in a bit
Nov 07 14:33:36 <Tee_Pee> I can show you the entire implementation then.
Nov 07 14:33:42 <j0hnsm1th> im keen
Nov 07 14:39:25 <Titi> Is that the best way to read a JSON ? final JSONObject obj = (JSONObject) new JSONTokener(data).nextValue(); ?
Nov 07 14:43:39 <lov> I don't know, is it?
Nov 07 14:47:18 <regedit> the docs recommend using the NotificationCompat.builder, however this does not have setters for all aspects of the notification - for example Notification.flags
Nov 07 14:49:27 <regedit> what is one expected to do?
Nov 07 14:53:13 <lov> set the flags on the Notification after you build it
Nov 07 14:53:59 <regedit> lov: will that not be circumventing/breaking the compatibility?
Nov 07 14:55:25 <Tee_Pee> j0hnsm1th - still here?
Nov 07 14:55:36 <j0hnsm1th> Tee_Pee, yup
Nov 07 14:56:12 <lov> no.
Nov 07 14:56:25 <lov> The NotificationCompat.builder is a convenience class
Nov 07 14:56:43 <regedit> i see
Nov 07 14:56:46 <regedit> ok thanks
Nov 07 14:56:57 <regedit> rofl http://developer.android.com/reference/android/support/v4/app/NotificationCompat.Builder.html#setPriority(int) read end of paragraph
Nov 07 14:57:16 <regedit> (make sure you get full URL with parentheses)
Nov 07 14:58:24 <lov> yeah I love that line
Nov 07 14:58:37 <regedit> i mean wtf
Nov 07 14:58:40 <regedit> http://www.google.com/search?hl=en&safe=off&tbo=d&output=search&sclient=psy-ab&q=site%3Adeveloper.android.com+mumble&btnK=
Nov 07 14:59:28 <Tee_Pee> alright, here's the implementation of the gridview... what I have is every view showing an image, a title and an image count on top. So I inflate a custom layout and pass all the relevant parameters
Nov 07 15:00:36 <Tee_Pee> http://pastebin.com/x6rhHSut  <----That's the adapter getView() method
Nov 07 15:01:52 <Tee_Pee> http://pastebin.com/NzVpL4kC   and the custom view
Nov 07 15:02:04 <Tee_Pee> look for //RELEVANT for the code setting the image
Nov 07 15:04:28 <regedit> when making a custom notification RemoteViews, how do i set the icon for the status bar? the normal NotificationCompat.builder.setSmallIcon() ? or is it something in the layout.xml
Nov 07 15:04:40 <Tee_Pee> I am still far from happy with the scrolling performance... but frankly I don't know what else I can do.
Nov 07 15:07:56 <Tee_Pee> Ok so.... I've figured out the biggest culprint in my scrolling performance
Nov 07 15:08:00 <Tee_Pee> garbage collection...
Nov 07 15:08:16 <Tee_Pee> I keep getting messages like GC_CONCURRENT freed 5984K, 57% free 6813K/15555K, paused 6ms+8ms, total 52ms
Nov 07 15:08:29 <Tee_Pee> so... back to reducing memory usage again...
Nov 07 15:11:08 <tdignan> Tee_Pee: line 14-16 you call albuminfo.get() 3 times on the same value
Nov 07 15:11:21 <tdignan> just keep a temporary reference to it.
Nov 07 15:11:25 <adq> Tee_Pee, try to add android:scrollingCache="false"
Nov 07 15:11:25 <adq> in xml
Nov 07 15:11:39 <adq> but i don't know which view you are using, works well on listview & gridview, iirc
Nov 07 15:12:05 <adq> it made a huggge diff on one of my app
Nov 07 15:13:03 <Tee_Pee> alright, let's see....
Nov 07 15:13:08 <evancharlton> Tee_Pee: there's some weird crap going on with your asynctask
Nov 07 15:13:22 <evancharlton> You never cancel the old BackgroundDecode
Nov 07 15:14:06 <evancharlton> Tee_Pee: also, in your constructors, just call this(context, attrs, 0) and this(context, null, 0)
Nov 07 15:14:14 <Tee_Pee> tdingan - you're right, I was going to change that, slipped my mind.
Nov 07 15:14:21 <Tee_Pee> adq - gonna try that in a minute
Nov 07 15:14:49 <Tee_Pee> evan - not sure what you mean by weird crap really
Nov 07 15:15:37 <evancharlton> Tee_Pee: well you schedule the task
Nov 07 15:16:10 <evancharlton> then the view gets recycled and and another decode gets queued. When the first task finishes, you have this bitmap that is never used and just needs to be thrown away
Nov 07 15:16:31 <evancharlton> you need to cancel the old task
Nov 07 15:16:54 <evancharlton> and for bonus points, just return null from doInBackground if isCancelled() is true
Nov 07 15:17:29 <Tee_Pee> ah
Nov 07 15:17:37 <Tee_Pee> right-o, gonna fix that
Nov 07 15:17:38 <evancharlton> also, what does getCursor() do?
Nov 07 15:17:45 <evancharlton> that seems extremely suspect
Nov 07 15:18:29 <Tee_Pee> it's a built in method of the simple cursor adapter... gets the active cursor... not my method
Nov 07 15:18:37 <evancharlton> ah okay
Nov 07 15:20:35 <evancharlton> but yeah, make the decode task a member variable and when setImage(String) is called, cancel the old and create a new one
Nov 07 15:23:42 <evancharlton> (and get rid of that WeakReference nonsense; totally unnecessary here)
Nov 07 15:27:59 <shmooz> when I use myProgressBar.startAnimation(alpha); it sets the foreground transparent but I need background tranparency on the progress bar
Nov 07 15:29:04 <shmooz> so that it will dim the background and the paused video behind doesn't cause the viewer to not be able to view progress bar
Nov 07 15:31:14 <Tee_Pee> evancharlton - Like this? http://pastebin.com/uP9RJ8gk
Nov 07 15:31:23 <Sculptor> hello fellow developers
Nov 07 15:31:29 <Tee_Pee> also, the weak reference stuff is from the google documentation http://developer.android.com/training/displaying-bitmaps/process-bitmap.html
Nov 07 15:36:35 <Jokr> hello, i have read about this little issue i have have with listview custom adapter and asynctask, this is my activity class that packs everything together http://pastebin.com/AHCe2pbL  this is my custom adapter http://pastebin.com/axNDJQhE. The problem is that it only shows the first item in the listview all the other items are blank like this: http://img525.imageshack.us/img525/1035/androidimage.png.
Nov 07 15:36:48 <Jokr>  So Initially, the first item is shown from the data i retrieved from the xml datastore, but all the other items are not showing. But if i change the orientation to landscape the last two rows are not shown immediately in the view, if you scroll down to the last items in the list the values are there and then the two top rows are not shown. When you scroll up, the 2nd row in the list is now shown.
Nov 07 15:39:58 <evancharlton> Tee_Pee: first, don't use != with strings
Nov 07 15:40:12 <evancharlton> second, that will NPE the first time through
Nov 07 15:40:25 <Tee_Pee> evan - already fixed both of those :P
Nov 07 15:40:27 <evancharlton> third, why aren't you using generics?
Nov 07 15:40:52 <Tee_Pee> probably because I have no idea what you're referring to :(
Nov 07 15:41:25 <evancharlton> fourth, I'd probably add in something like if (!isCancelled()) { Bitmap b = BitFac.decodeFile(path); return isCancelled() ? null : b; } return null; but I don't know if that will improve things
Nov 07 15:41:59 <evancharlton> fifth, cancel the task in onDetachedFromWindow as well
Nov 07 15:42:03 <Tee_Pee> well, right now I'm about to log stuff to see why I'm not getting any images at all whatsoever
Nov 07 15:42:10 <Tee_Pee> gimme a min
Nov 07 15:44:58 <Jokr> I tried adapter.notifyDataSetChanged(); , listView.refreshDrawableState(); & listView.requestLayout(); individually and together none of which worked.  I have also tried the runOnUiThread function in both doingInbackground and onPostExecute of my asynctask none of which also work.
Nov 07 15:45:19 <Jokr> :(
Nov 07 15:51:55 <kamoricks> Jokr: In your Activity class, remove lines 101 and 102. Next up, you really *should* be using for(Rate y : val) there, because that for loop should show the first two items.
Nov 07 15:53:10 <Jokr> Oh i added the "hardcoded value 2" to see what will happen and surprisingly it only show the image of one row
Nov 07 15:53:20 <kamoricks> Jokr: That is the opposite of suprising.
Nov 07 15:53:56 <Jokr> oh =/ how so
Nov 07 15:54:49 <kamoricks> Jokr: You're inflating the same view twice.
Nov 07 15:55:10 <kamoricks> Jokr: Plus limiting the number of items will show fewer items. Plus you seem to be treating item 0 of the List special for some reason.
Nov 07 15:56:29 <Jokr> how so i am confused :s i have been searching for the answer for days
Nov 07 15:57:22 <kamoricks> Glad to have helped.
Nov 07 15:57:36 <Jokr> wait... im still stuck
Nov 07 15:57:38 <Jokr> :(
Nov 07 15:57:53 <Jokr> what do u mean i am iflating the same thing twice
Nov 07 15:58:11 <Jokr> and im treating the first one in the list differently
Nov 07 15:58:28 <Jokr> i had the foreach before and that didn't fix the problem
Nov 07 15:58:46 <kamoricks> Jokr: If you're handed the same null view twice, you'll inflate the same view twice. Or rather, you won't.
Nov 07 15:58:48 <Tee_Pee> evancharlton - thanks for all the help, I think I got the scrolling to as smooth as it can be.
Nov 07 15:59:02 <kamoricks> Jokr: If you're handed a List of one element, you're treating it as if it were empty.
Nov 07 15:59:20 <kamoricks> Jokr: And iterating over a dynamically-sized list for a fixed number of iterations is a bad idea.
Nov 07 15:59:24 <Tee_Pee> the rest will depend on memory handling while downloading the images initially
Nov 07 15:59:28 <Tee_Pee> which I'll have to get to as well
Nov 07 16:00:43 <evancharlton> Tee_Pee: here's what I just whipped up (pulls from a contentprovider, but the principle is the same): http://pastebin.com/4strwA0s
Nov 07 16:02:00 <Jokr> hmmm... the weirdest part is that when i complied the code awhile ago and ran it on my android device it showed 5/6 items. When i complied it again it showed only the first item
Nov 07 16:03:52 <Jokr> would you like to see an image of what i am talking about?
Nov 07 16:04:15 <docmur> What is the right way to call methods from a seperate class? http://pastebin.com/EiqRf9SJ  Seems to cause a null point exception
Nov 07 16:04:37 <kamoricks> Jokr: Not particularly. I'd like to see the code differences.
Nov 07 16:05:25 <kamoricks> docmur: What line does the NPE trigger on?
Nov 07 16:05:25 <freannrak> Hey. I'm trying to figure out why my constructor isn't working. I'm kind of confused here. Why isn't it accepting 'searchItems'? There's the 2 lines of code I'm trying to run and below is the constructor. http://pastebin.com/qNdGjXKe
Nov 07 16:05:26 <Jokr> i gives the same results unless i could clear some cache that it might have saved.
Nov 07 16:05:34 <Tee_Pee> evancharlton - yeah, this is essentially what I got as well now, except I load images from a file on the SD card rather than the gallery, but yeah, same principle.
Nov 07 16:05:57 <docmur> in CVPSearch this line:     public class GetURLListData extends AsyncTask<String, Void, JSONObject>
Nov 07 16:06:28 <kamoricks> docmur: That's not a line of code. Try again.
Nov 07 16:06:59 <kamoricks> freannrak: What error do you get from the compiler?
Nov 07 16:07:09 <docmur> sorry it's this function, http://pastebin.com/YRmfYw52
Nov 07 16:07:23 <freannrak> kamoricks: The constructor TweetListAdaptor(SearchResultsActivity, int, ArrayList<Tweet>) is undefined
Nov 07 16:07:25 <kamoricks> docmur: What line?
Nov 07 16:07:46 <docmur> setProgressBarIndeterminateVisibility(true);
Nov 07 16:07:48 <kamoricks> freannrak: I'll need a more complete code sample.
Nov 07 16:08:12 <kamoricks> docmur: Post your entire stacktrace. You're guessing.
Nov 07 16:09:23 <freannrak> kamoricks: hold on
Nov 07 16:09:25 <docmur> I'm not guess I'm reading the stack trace, CVP line 133: search.doSearch(query , 1);  Then CVPSearch line 246: new GetURLListData().execute(query, page.toString());  then CVPSearch: 309 setProgressBarIndeterminateVisibility(true);
Nov 07 16:10:03 <docmur> but here is the trace: http://pastebin.com/32KRWByT
Nov 07 16:10:19 <Jokr> if i change the phone's orientation and the last two rows that arent showing after you scroll down to them. They have values.
Nov 07 16:10:46 <Jokr> This in term hides the top two rows and after you scroll back up they have values
Nov 07 16:11:39 <Jokr> and i tried the to use adapter.notifyDataSetChanged(); , listView.refreshDrawableState(); & listView.requestLayout(); individually and together none of which worked.
Nov 07 16:11:43 <freannrak> kamoricks: http://pastebin.com/Sd3nA5Pw and http://pastebin.com/gkcLnp0R
Nov 07 16:11:46 <Jokr> to try and refresh the ui
Nov 07 16:12:16 <kamoricks> docmur: Your target intent for progress bar visibility is null. You'll want to call super.onPreExecute() first.
Nov 07 16:13:02 <docmur> Okay
Nov 07 16:13:30 <regedit> please halp http://codepad.org/W8TX1vD4
Nov 07 16:14:02 <kamoricks> docmur: The NPE line in question was Activity.java:3614, by the way.
Nov 07 16:14:18 <regedit> bit clearer http://codepad.org/QnZkoFEQ
Nov 07 16:14:31 <docmur> well yes but my code has to be causing it.
Nov 07 16:16:33 <pfn> docmur, you're calling it before onCreate
Nov 07 16:17:21 <pfn> or it's happening after the activity is no longer visible
Nov 07 16:17:24 <freannrak> kamoricks: i've no idea why it isn't accepting searchItems as ArrayList<Tweet>... it should, shouldn't it?
Nov 07 16:17:27 <regedit> clearer yet http://codepad.org/rPf1yZc1 what am i doing wrong
Nov 07 16:17:29 <Ivru> regedit: what's your remove view ?
Nov 07 16:17:33 <kamoricks> freannrak: The error message says it is.
Nov 07 16:17:43 <Ivru> oh sorry
Nov 07 16:18:04 <Ivru> regedit: you can not use a toggle button in a remote view
Nov 07 16:18:08 <Ivru> ...
Nov 07 16:18:15 <regedit> really?...
Nov 07 16:18:42 <evancharlton> it wouldn't do you much good
Nov 07 16:18:45 <Ivru> yeah... read the doc
Nov 07 16:18:50 <Titi> Hi, if I have an horizontalscrollview on a viewpager,; how can I disable Viewpager touchevent on the horizontalscrollview area ? I tried, requestDisallowInterceptTouchEvent but nothing happens
Nov 07 16:19:48 <regedit> Ivru: http://developer.android.com/reference/android/widget/RemoteViews.html searching...
Nov 07 16:19:55 <freannrak> kamoricks: interesting that line 30 here works perfectly... http://pastebin.com/0J052MTY
Nov 07 16:19:57 <docmur> I don't want to start the activity I just want to access methods from the CVPSearch Class so I shouldn't have to call onCreate();
Nov 07 16:20:04 <Ivru> regedit: http://developer.android.com/guide/topics/appwidgets/index.html#CreatingLayout
Nov 07 16:20:21 <evancharlton> regedit: if a class doesn't have the @RemoteView annotation, you can't use it in removeviews
Nov 07 16:20:21 <Ivru> yeah It's in the appwidget doc
Nov 07 16:20:25 <evancharlton> remoteviews, too
Nov 07 16:20:38 <kamoricks> freannrak: Are you sure that's the first error?
Nov 07 16:20:41 <theneb> Is there any way to get access to API Level 17 and a Nexus 10 emulator?
Nov 07 16:21:08 <regedit> how does the WiFi notification have a fancy on/off switch button? i want a switch like that too
Nov 07 16:21:21 <evancharlton> regedit: redraw the notification with the new state
Nov 07 16:21:42 <evancharlton> also, I suspect this "wifi notification" you keep talking about is some OEM-provided thing? like Samsung or something?
Nov 07 16:21:59 <regedit> nope, default on my ICS install
Nov 07 16:22:19 <evancharlton> er
Nov 07 16:22:20 <regedit> same as all on/off switches in Settings etc
Nov 07 16:22:29 <evancharlton> got a screenshot?
Nov 07 16:22:49 <regedit> http://www.google.com/search?q=android+on+off+switch&sugexp=chrome,mod%3D0&safe=strict&sout=1&um=1&ie=UTF-8&hl=en&tbm=isch&source=og&sa=N&tab=wi&ei=zoqaUN_lFuqO0QHC8IDIDw
Nov 07 16:22:54 <regedit> first images
Nov 07 16:23:10 <Ivru> this is a setting
Nov 07 16:23:13 <Ivru> not a notification
Nov 07 16:23:17 <Titi> Hi, if I have an horizontalscrollview on a viewpager,; how can I disable Viewpager touchevent on the horizontalscrollview area ? I tried, requestDisallowInterceptTouchEvent but nothing happens ??
Nov 07 16:23:25 <Jokr> kamoricks: don't forget me :(
Nov 07 16:23:31 <regedit> yes but it exists as a notification too
Nov 07 16:23:39 <evancharlton> regedit: I want to see a screenshot of this notification
Nov 07 16:23:41 <Ivru> regedit: I've never seen it
Nov 07 16:23:45 <evancharlton> because that's definitely not a stock android thing
Nov 07 16:23:52 <freannrak> kamoricks: pretty sure. :P
Nov 07 16:23:53 <regedit> oh, hang on
Nov 07 16:24:20 <Leeds> regedit: hint - a screenshot from *your phone* would probably be simplest
Nov 07 16:24:31 <regedit> oh ok
Nov 07 16:24:35 <regedit> how do i snap
Nov 07 16:24:53 <freannrak> kamoricks: I was using that code to display twitter's timeline in a listview. Now I'm trying to do the same thing for search results and it doesn't accept my ArrayList<Tweet> searchItems and wants to change the constructor in TweetListAdaptor
Nov 07 16:24:59 <Leeds> power+voldown - or could be volup
Nov 07 16:24:59 <pfn> regedit, power+voldown
Nov 07 16:25:01 <Ivru> regedit: power + voldown
Nov 07 16:25:13 <regedit> k hang on..
Nov 07 16:25:43 <kamoricks> freannrak: Is it the same Tweet referred to in both places?
Nov 07 16:26:09 <kamoricks> freannrak: Change the constructor. See what it changes it to. It looks like you have two classes, both named 'Tweet'.
Nov 07 16:26:26 <karakuri> can anyone help with this?: http://stackoverflow.com/questions/13254758/edittext-cursor-overlapping-last-letter
Nov 07 16:26:34 <Titi> hi
Nov 07 16:26:41 <Titi> :(
Nov 07 16:26:56 <BinaryMan> is there a way to call a private void in another class without exposing it directly to external methods?
Nov 07 16:27:19 <f2prateek> BinaryMan: no
Nov 07 16:27:24 <evancharlton> BinaryMan: I feel like I'm going to regret telling you this, but you can use reflection
Nov 07 16:27:28 <freannrak> kamoricks: Eclipse autochanged it to ArrayList<twitter4j.Tweet> searchItems. Oh well, now I feel stupid. Happy, but stupid.
Nov 07 16:27:36 <evancharlton> BinaryMan: but don't pretend it's secure
Nov 07 16:28:02 <BinaryMan> evancharlton: is there a better, more secure way?
Nov 07 16:28:21 <regedit> argh as soon as i try screenshot, the notification drawer folds away...
Nov 07 16:28:38 <freannrak> kamoricks: Thanks for taking the time to look at my code.
Nov 07 16:28:48 <Leeds> regedit: camera or spare phone, take a photo... what phone do you have, anyway?
Nov 07 16:30:02 <Ivru> I'm just waiting for regedit to post a screen before leaving
Nov 07 16:30:24 <regedit> k leme try snapping from my iphone or somethin..
Nov 07 16:30:58 <shmooz> I thought he's talking about the feature that was included with ICS that lets quick settings access from notification slide down
Nov 07 16:31:06 <Leeds> I can definitely take a screenshot with the notification panel down on my phone
Nov 07 16:31:46 <Ivru> same here :)
Nov 07 16:32:09 <BinaryMan> wait
Nov 07 16:32:09 <regedit> maybe different combo of buttonpresses
Nov 07 16:32:21 <regedit> its a touchpad with ICS on it (dont ban me)
Nov 07 16:32:21 <BinaryMan> nevermind, i think i found a public void that'll accomplish what i'm looking for
Nov 07 16:32:39 <regedit> k took iphone snaps, how to upload hmmm...
Nov 07 16:33:10 <Leeds> regedit: so when you say 'ICS', you mean a cyanogen port or similar?
Nov 07 16:33:18 <regedit> ye
Nov 07 16:33:56 <TachyonDev2> zomg beta release emminent
Nov 07 16:34:13 <SimonVT> Is ToggleButton even supported in notifications?
Nov 07 16:34:20 <Ivru> it is not
Nov 07 16:34:27 <SimonVT> There you go
Nov 07 16:34:33 <Jokr> kamoricks: i know i am using one view to reiterate to form the list, i add the values in the getView function in my custom adapter class
Nov 07 16:34:42 <regedit> imageshack is blocked by my workplace, where else can i post
Nov 07 16:34:46 <erasmospunk> I am trying to debug an app but I get INSTALL_FAILED_MISSING_SHARED_LIBRARY. Any ideas?
Nov 07 16:34:55 <Leeds> regedit: so not, actually, stock Android
Nov 07 16:35:00 <Ivru> regedit: http://imgur.com/
Nov 07 16:35:03 <Leeds> erasmospunk: install the library?
Nov 07 16:35:12 <regedit> Ivru: also blocked... :(
Nov 07 16:35:18 <erasmospunk> Leeds: any tip?
Nov 07 16:35:43 <Leeds> regedit: picasa
Nov 07 16:35:45 <regedit> Leeds: you'll see, it's definitely the stock on off switch
Nov 07 16:35:53 <erasmospunk> on the emulator it works ok
Nov 07 16:36:04 <Leeds> regedit: have you actually *used* stock Android?
Nov 07 16:36:05 <regedit> Leeds: no go..
Nov 07 16:36:24 <regedit> Leeds: not much, bit
Nov 07 16:36:32 <erasmospunk> I have a device with cyanogenmod
Nov 07 16:36:44 <Ivru> but regedit like ... Leeds said (or someone else) it must be just a tow states imagebutton)
Nov 07 16:36:44 <Leeds> I mean not whatever port you have on your touchpad, actual vanilla AOSP Nexus Android?
Nov 07 16:37:06 <shmooz> stock ICS should have that quick settings feature
Nov 07 16:37:25 <regedit> email anyone?
Nov 07 16:37:30 <shmooz> I thought it did
Nov 07 16:38:10 <regedit> PM me if you want email snapshot
Nov 07 16:38:23 <Leeds> regedit: vmlinuz at gmail, I'll post it
Nov 07 16:38:35 <regedit> ah ok thanks, here goes
Nov 07 16:40:08 <Leeds> right... that's the HC/ICS 'tablet-mode' quick settings
Nov 07 16:40:10 <regedit> note that what you are looking at is the notifications drawer open (Settings app in the background)
Nov 07 16:40:12 <Leeds> it's not a notification at all
Nov 07 16:40:30 <regedit> oh?
Nov 07 16:41:18 <regedit> i guess you can figure out why i'd confuse those 2... how would i know
Nov 07 16:41:36 <Leeds> https://lh4.googleusercontent.com/-0av2bEd3Crs/UJqPGfhg67I/AAAAAAAAHDU/YjcO7N9hYy8/s800/photo.JPG
Nov 07 16:42:05 <Titi> Hi, if I have an horizontalscrollview on a viewpager,; how can I disable Viewpager touchevent on the horizontalscrollview area ? I tried, requestDisallowInterceptTouchEvent but nothing happens ?? :( :(
Nov 07 16:42:27 <SimonVT> The fact that they are not visible along with the notifications would be the first hint
Nov 07 16:42:35 <regedit> it has a status bar icon and a drawer entry, seemed like that was Notification's turf
Nov 07 16:43:02 <Titi> you speak to me ?
Nov 07 16:43:25 <regedit> who me? no
Nov 07 16:43:32 <hkhalid> I have a swipe+tabs setup, and its setup to use fragments, which I got working fine, but now I have a listview and so far it seems I have to use a listfragment, is there anything else I have to do to implement it?
Nov 07 16:43:34 <Titi> ok
Nov 07 16:43:43 <SimonVT> No, Titi.. You should just stop doing what you're doing
Nov 07 16:44:46 <pfn> you don't have to use a listfragment in order to use a listview
Nov 07 16:44:52 <pfn> it's "easier" if you do, is all
Nov 07 16:45:12 <regedit> so if i want a foreground service with 1) a persistent status bar icon 2) a quick settings entry to switch this service of mine on/ff -- similar to the wifi -- do i need to use Notification apis at all? or just quick settings stuff
Nov 07 16:48:01 <SimonVT> You can't add your own quick settings
Nov 07 16:48:34 <SimonVT> Supported remoteviews widgets are here: http://developer.android.com/guide/topics/appwidgets/index.html#CreatingLayout
Nov 07 16:48:49 <Leeds> plus, that panel has pretty much gone away in 4.1, and been replaced in 4.2
Nov 07 16:48:54 <SimonVT> They should really add API level to that list
Nov 07 16:49:58 <regedit> so what would be the recommended way to offer the user a way to quickly switch an on/off setting of this foreground service?
Nov 07 16:50:27 <regedit> should it perhaps be in a persistent non-cancelable notification?
Nov 07 16:50:32 <SimonVT> An app widget and not make it foreground
Nov 07 16:51:16 <Jokr> maybe it's my activity that is doing something wrong. Can you help kamoricks gave me some information that i am inflating the same menu thats y i am getting these results : http://img525.imageshack.us/img525/1035/androidimage.png
Nov 07 16:52:14 <regedit> SimonVT: not sure i understand. im making a service that will be actively doing stuff in "foreground-land" (i.e. user is to be actively aware of it) so what do you mean not make it foreground
Nov 07 16:52:54 <Jokr> here is my activity http://pastebin.com/BgBaFAEE and my adapter http://pastebin.com/BC3fdEuN
Nov 07 16:52:57 <hkhalid> pfn, in my main class, it has a function for Fragment, getitem, I cant use a listfragment with this can I? I am having trouble getting this switch/case function to call my Files class which is a listview
Nov 07 16:53:09 <SimonVT> You want it to remain foreground from boot til shutdown.. And basically taking up memory, and priority over other apps
Nov 07 16:53:19 <SimonVT> Not really recommended to do that
Nov 07 16:53:23 <regedit> SimonVT: i wish to have an icon in the status bar that can be touched to switch on/off
Nov 07 16:53:32 <regedit> similar to wifi
Nov 07 16:54:00 <Leeds> SimonVT: I wish for a pony
Nov 07 16:54:03 <SimonVT> I'd argue that wifi is more important than your app
Nov 07 16:54:34 <regedit> SimonVT: ongoing speech recognition
Nov 07 16:54:45 <SimonVT> That can be turned off
Nov 07 16:54:50 <kamoricks> regedit: Here's what you can do. You can have a sticky notification that displays the current state, and when triggered, have it toggle the state.
Nov 07 16:55:21 <regedit> SimonVT: you are asking or stating?
Nov 07 16:55:41 <regedit> kamoricks: ye i guess that was kind of what i was doing up until this point, but the on/off switch is not supported apparently
Nov 07 16:55:46 <SimonVT> Just stating
Nov 07 16:55:50 <SimonVT> You're asking for recommendations
Nov 07 16:55:57 <SimonVT> I'm recommending you don't do that
Nov 07 16:56:08 <Robbilie> http://on.fb.me/SqxBgG & http://bit.ly/TI8Rnq please take a look at it and give me feedback :(
Nov 07 16:56:30 <regedit> SimonVT: what do you mean, how can it be turned off, and what if the user doesnt want to turn off general speech recognition - just turn off my service of continuous speech recognition
Nov 07 16:57:33 <regedit> SimonVT: this is to be a Voice Actions-like service, where you can command the device by voice to do stuff
Nov 07 16:58:50 <regedit> SimonVT: if the user so chooses, the service would be able to listen continuously
Nov 07 16:59:20 <regedit> in which case i need a status bar icon and quick way of switching it off
Nov 07 16:59:28 <regedit> and on
Nov 07 16:59:49 <SimonVT> oh well
Nov 07 17:00:19 <Leeds> regedit: are you aware of the concept of something which you shouldn't do unless you know why you shouldn't?
Nov 07 17:00:56 <regedit> what are you trying to say, please clarify
Nov 07 17:01:31 <regedit> is it too much to ask to make a service that's kinda like wifi, with status bar icon and a quick way of switching on/off
Nov 07 17:01:39 <regedit> what if i were making a music app
Nov 07 17:01:54 <SimonVT> Then you'd put up a notification while it's playing music
Nov 07 17:02:05 <SimonVT> Not when the user stopped the player
Nov 07 17:02:07 <Leeds> I'm trying to say that there may be ways to do what you're asking, but you shouldn't be considering them until and unless you're fully aware of why you shouldn't be doing it, and why you're a special case
Nov 07 17:02:54 <Leeds> also, have you actually tested speech recognition from a Service yet?
Nov 07 17:02:57 <karakuri> does DashPathEffect not work on some devices?
Nov 07 17:03:06 <regedit> SimonVT: fair enough, i'll do that. so, how is the icon & button done
Nov 07 17:03:47 <SimonVT> On old platforms, just put a button in your layout that says "stop" or whatever
Nov 07 17:03:48 <regedit> Leeds: i told you already yes, you are starting to annoy me with that repeated question
Nov 07 17:04:01 <Leeds> pretty sure you didn't tell me yes
Nov 07 17:04:02 <SimonVT> On jb+ you can expand it, and put the stop button on one of the built in buttons
Nov 07 17:04:18 <regedit> Leeds: yesterday
Nov 07 17:04:28 <Leeds> all my troubles seemed so far away!
Nov 07 17:04:44 <evancharlton> Leeds++
Nov 07 17:05:30 <shmooz> HEY TEACHER!
Nov 07 17:05:31 <regedit> SimonVT: which layout are you referring to, a remoteview in a sticky Notification?
Nov 07 17:05:35 <shmooz> LEAVE THE KIDS ALONE!
Nov 07 17:05:49 <SimonVT> yes
Nov 07 17:06:10 <regedit> SimonVT: is that how like music services do it
Nov 07 17:06:27 <SimonVT> They probably use imagebutton
Nov 07 17:06:39 <SimonVT> But yes, those are notifications as well
Nov 07 17:07:07 <regedit> ah then i was on the right track after all, what a releif
Nov 07 17:07:26 <regedit> *relief
Nov 07 17:07:35 <regedit> thanks SimonVT
Nov 07 17:07:50 <regedit> Leeds: but yeah it works like a charm and then some
Nov 07 17:08:01 <regedit> Leeds: even while in other apps
Nov 07 17:08:11 <SimonVT> Right track, just remember to be a good android citizen
Nov 07 17:08:20 <SimonVT> Leaving a service foreground forever is not
Nov 07 17:08:35 <regedit> SimonVT: trying my best, worst case don't accept my app :P
Nov 07 17:09:35 <SimonVT> Worst case you'd link it here some day and I'd give it 1 star.. But if you're going to change your ways :)
Nov 07 17:10:10 <regedit> heh
Nov 07 17:10:46 <regedit> somehow i dont find it impossible to believe that after a user pauses music playing, that the notification/status-icon still lingers
Nov 07 17:11:11 <regedit> pauses or even stops
Nov 07 17:11:17 <regedit> am i wrong?
Nov 07 17:11:17 <SimonVT> It does after pausing
Nov 07 17:11:22 <SimonVT> That's just "hold on, just a minute"
Nov 07 17:11:29 <SimonVT> After stopping it goes away and the service is stopped
Nov 07 17:11:47 <Reaga> why is there no damn
Nov 07 17:11:49 <Reaga> flow layout
Nov 07 17:11:59 <regedit> i see, and then to make it show up again you need to open the app and start playing again?
Nov 07 17:12:07 <kamoricks> Reaga: There is. It's called LinearLayout.
Nov 07 17:13:21 <SimonVT> yes
Nov 07 17:14:07 <regedit> i see now, so yes i was possibly gonna do something evil :D at least now i know
Nov 07 17:15:31 <canadiancow> "flow layout" ?
Nov 07 17:15:55 <Amex> hello everyone
Nov 07 17:16:05 * monsti loves to flow
Nov 07 17:16:11 <Amex> anyone who can help me something with webview
Nov 07 17:16:13 <Amex> ?
Nov 07 17:16:38 <regedit> Amex: see topic "Ask your question, don't ask to ask"
Nov 07 17:16:45 <dragorn> ask your question about it and if someone knows they'll respond, most likely.
Nov 07 17:16:57 <Amex> ok, sorry
Nov 07 17:17:15 <Amex> how to prevent webview to reload after orientation change
Nov 07 17:18:11 <Reaga> kamoricks: no thats just a grid layout with 1 column or 1 row
Nov 07 17:18:20 <Reaga> how the fuck do i get rid of this header
Nov 07 17:18:21 <Reaga> on my app
Nov 07 17:18:29 <kamoricks> Reaga: What "header"?
Nov 07 17:18:41 <Reaga> http://developer.android.com/images/ui/sample-linearlayout.png
Nov 07 17:18:46 <Reaga> the bit at the top that says relative layout
Nov 07 17:19:01 <kamoricks> Reaga: You mean the actionbar?
Nov 07 17:19:10 <dragorn> Amex: you'll probably have to record the scroll position then re-set it after the orientation
Nov 07 17:19:17 <Reaga> kamoricks: yeah?
Nov 07 17:19:19 <dragorn> Amex: what the API to do that is, I don't know
Nov 07 17:19:20 <Reaga> if thats what its called
Nov 07 17:19:29 <MastaFlash> hey all. for some reason, every now and again when I run my app through the emulator, the emulator decides to reboot itself. not shut down, just restart. any ideas on how I get it to stop doing that?
Nov 07 17:19:43 <kamoricks> Reaga: .getActionBar().hide() on API 11+.
Nov 07 17:20:00 <kamoricks> Reaga: Also, if you're developing an application for a platform, it may be helpful to learn what things are called.
Nov 07 17:21:00 <Reaga> kamoricks: i know. i just learned what one thing is called
Nov 07 17:21:04 <dragorn> Amex: if I don't know the API, privmsg'ing me isn't going to make me suddenly know it.  developer.google.com should have a whole page on it
Nov 07 17:21:15 <dragorn> er, .android.com
Nov 07 17:21:16 <Amex> dragorn: ok
Nov 07 17:22:03 <Amex> dragorn: thing is that i tried all solutions, and still it's not working, it reloads every time :S
Nov 07 17:36:27 <regedit> i'm designing a Notification layout in eclipse, but it renders everything as if it were a fullblown app (with status bar and app header etc) is there any way to make the GUI preview a specific width/height and eliminate the extras, 'cuz this is for a notification (64dp i think)
Nov 07 17:37:19 <regedit> the WYSIWYG editor
Nov 07 17:37:46 <zinx> dp doesn't change in a notification, just the size
Nov 07 17:38:15 <zinx> i think it does let you view it in different sizes/configurations, but i'm not sure if it lets you do custom ones
Nov 07 17:38:25 <zinx> been a while since i used it >_<
Nov 07 17:38:36 <regedit> i was trying to find a configuration for Notification
Nov 07 17:38:54 <zinx> that's definitely something i'd suggest to the ADT devs
Nov 07 17:46:36 <canadiancow> regedit, can you not specify exactly what dimensions you want?
Nov 07 17:48:06 <regedit> trying different things, wouldve been nice if there was a ready config for different kinds of layouts - everything seems to be geared for fullblown app
Nov 07 17:50:51 <Reaga> guys?
Nov 07 17:50:55 <Reaga> my style ISNT WORKING
Nov 07 17:51:00 <Reaga> ITS NOT DOING ANYTHING WHEN I CHANGE THE STYLE
Nov 07 17:51:12 <Leeds> Reaga: then go down to the store and get some new clothes!  Maybe a haircut as well
Nov 07 17:51:16 <evancharlton> WHOA CALM DOWN
Nov 07 17:51:53 <regedit> all caps is capital offence
Nov 07 17:52:02 <regedit> no pun intended whatsoever
Nov 07 17:52:09 <Leeds> regedit: custom notifications is pretty advanced...
Nov 07 17:52:53 <Leeds> if you need drag'n'drop handholding...
Nov 07 17:53:45 <regedit> Leeds: how advanced should this be already? cant a guy make a music app around here, or some service which needs some quick setting/config
Nov 07 17:54:12 <Leeds> it's. not. quick. setting.
Nov 07 17:54:36 <regedit> wasnt referring to Quick Settings
Nov 07 17:54:41 <Leeds> and no, people who've been playing with Android for a few days aren't usually trying to write full-featured music apps
Nov 07 17:54:51 <regedit> well excuse me if i do
Nov 07 17:55:05 <evancharlton> regedit: sorry if I missed it, but did you ever post a screenshot?
Nov 07 17:55:25 <Leeds> evancharlton: https://lh4.googleusercontent.com/-0av2bEd3Crs/UJqPGfhg67I/AAAAAAAAHDU/YjcO7N9hYy8/s800/photo.JPG
Nov 07 17:55:27 <regedit> evancharlton: Leeds posted it for me
Nov 07 17:55:36 <Leeds> he was referring to the HC/ICS 'tablet-mode' quick settings
Nov 07 17:55:58 <evancharlton> oh, yeah, that's not a notification.
Nov 07 17:56:10 <regedit> Leeds: talking of which, link to statement that it's being pulled?
Nov 07 17:56:24 <Leeds> nope
Nov 07 17:56:27 <Leeds> just trust me
Nov 07 17:56:38 <regedit> i thought you said in 4.2 they're removing it
Nov 07 17:56:42 <regedit> oh
Nov 07 17:56:44 <Leeds> but it doesn't matter, because you can't touch it, can't integrate with it, can't make any use of it
Nov 07 17:56:46 <regedit> no way man
Nov 07 17:56:57 <Leeds> it is not something which concerns you as an app author
Nov 07 17:57:31 <regedit> also your mention of a concept "tablet-mode" intrigues me, i shall google on that
Nov 07 17:57:42 <Leeds> yeah, that? also going away
Nov 07 17:57:48 <regedit> oh
Nov 07 17:58:23 <SimonVT> Leeds: Are they adding some new quick settings thing?
Nov 07 17:58:29 <SimonVT> I haven't looked up 4.2 changes at all
Nov 07 17:58:52 <Leeds> SimonVT: yeah... it's a two-finger drag down, or a button on the notification panel... it's sort of big toggle buttons, looks a bit like the car mode launchers
Nov 07 17:59:13 <SimonVT> ah
Nov 07 17:59:50 <Leeds> http://androidheadlines.com/wp-content/uploads/2012/11/tablets-pulldown.png on tablets, http://forums.androidcentral.com/attachments/google-nexus-4/43658d1351525210t-android-4-2-quick-settings-settings.jpeg for the phone
Nov 07 18:00:48 <SimonVT> Also two-finger drag on tablets?
Nov 07 18:00:52 <SimonVT> I kinda liked that it was just a tap
Nov 07 18:01:08 <Leeds> not sure about that - most of the reports have been for the N4, not N10
Nov 07 18:01:25 <SimonVT> Alright
Nov 07 18:01:35 <Leeds> we will know more next week...
Nov 07 18:01:53 <SimonVT> yeah
Nov 07 18:02:06 <Leeds> been lots of people wondering if there might be any way for apps to integrate with it, now it's AOSP-standard... not heard anything from any Googlers
Nov 07 18:06:13 <hkhalid> is it possible to use ListFragment with newinstance
Nov 07 18:06:46 <SimonVT> I don't really see it making 4.2.. It's kind of like they haven't really completely decided on an approach yet
Nov 07 18:07:06 <SimonVT> Extend ListFragment, make a newInstance method
Nov 07 18:09:37 <trytrytry> hi, I'm an android beginner. I want to create an app that lets you order food from a restaurant. The restaurant itself has to get notifications about orders somehow (not necessesarily in an android device). What is the prefered android way of doing this app. I'm thinking about sending http requests with JSON to a web server running php and storing data about orders in MySQL on the remote server. Any other ways to think about this? What's common in andro
Nov 07 18:10:05 <xroberx> hi
Nov 07 18:10:19 <tophyr> trytrytry: that's how i'd build it
Nov 07 18:13:01 <canadiancow> oh wow you can pull down both notifs and quick settings at the same time?
Nov 07 18:13:04 <hkhalid> ill pay a small fee if someone can help get a listview working with a fragment
Nov 07 18:13:05 * canadiancow learns something new every day
Nov 07 18:13:26 <monsti> canadiancow: me, too
Nov 07 18:14:30 <Leeds> http://arstechnica.com/gadgets/2012/11/review-android-4-2-is-a-sweeter-tasting-jelly-bean/ may be of assistance - also, http://www.android.com/whatsnew/
Nov 07 18:14:41 <Leeds> don't you all check android.com for new stuff every morning over breakfast?
Nov 07 18:15:11 <canadiancow> android.com? not really
Nov 07 18:15:14 <canadiancow> androidpolice.com maybe
Nov 07 18:15:57 <Leeds> bah
Nov 07 18:15:59 <Leeds> lightweight :P
Nov 07 18:16:21 <canadiancow> if i got all my information from android.com i'd never know anything
Nov 07 18:16:47 <Leeds> are you trying to say that it's not an info-rich site of knowledge?
Nov 07 18:17:02 <canadiancow> im trying to say that it's updated a lot less frequently than other places i go for my information
Nov 07 18:17:10 <Leeds> aka "yes" :)
Nov 07 18:17:14 * Leeds -> to shower
Nov 07 18:25:41 <Reaga> im new to android
Nov 07 18:25:57 <Reaga> im trying to get this theme working but its not doing anything
Nov 07 18:28:13 <gutano> it's like the inverse of Jeopardy.. your question must be in the form of a question
Nov 07 18:29:27 <Reaga> whats the difference between styles and themes
Nov 07 18:30:33 <gutano> http://developer.android.com/guide/topics/ui/themes.html
Nov 07 18:30:42 <gutano> "A style is a collection of properties that specify the look and format for a View or window."
Nov 07 18:30:51 <gutano> "A theme is a style applied to an entire Activity or application, rather than an individual View (as in the example above)."
Nov 07 18:32:41 <Reaga> hmmm
Nov 07 18:33:01 <Reaga> maybe its because im running cyanogenmod
Nov 07 18:33:27 <Reaga> theres only 1 theme under Theme Chooser
Nov 07 18:34:21 <tdignan> are you asking a development question?
Nov 07 18:34:28 <tdignan> this channel is about writing apps
Nov 07 18:34:50 <Leeds> er... yeah... are you trying to theme your app, or the system?
Nov 07 18:35:03 <tdignan> i think just choose the theme :)
Nov 07 18:35:17 <Reaga>  . . . im trying to write an app
Nov 07 18:35:25 <Reaga> and im not succesfully changing the theme on it
Nov 07 18:35:45 <Leeds> then that has absolutely nothing to do with Cyanogenmod's theme chooser
Nov 07 18:36:02 <Reaga> alright
Nov 07 18:39:23 <preds> anybody know of some sort of hack force a device that reports itself as xlarge to use non xlarge resources?
Nov 07 18:39:56 <Leeds> preds: for your app?
Nov 07 18:40:07 <preds> I have to get this app running nicely on the kindle which reports itself as an xlarge, even though it falls below google's official xlarge recommendations
Nov 07 18:40:12 <preds> Leeds: yeah
Nov 07 18:40:16 <sleepster> is there a way to create custom AVDs that match an *exact* hardware spec?
Nov 07 18:40:21 <Leeds> then fix your resources
Nov 07 18:40:24 <sleepster> assuming I know the hardware spec before hand
Nov 07 18:40:28 <zinx> sleepster: no
Nov 07 18:40:39 <zinx> sleepster: not without writing the code for qemu to match your exact hardware, anyway
Nov 07 18:40:40 <Reaga> cant solve this
Nov 07 18:40:42 <Reaga> its too hadrt
Nov 07 18:41:02 <Leeds> also, kindle is not considered a 'compatible' device, so it's perfectly normal to have to hack your app to support it
Nov 07 18:41:05 <sleepster> I see.. thanks zinx .  Do you know if people have done this by modifying qemu?
Nov 07 18:41:06 <zinx> sleepster: you can match things like screen size/density, accelerometer, etc
Nov 07 18:41:23 <tdignan> Reaga: start by saying what you are trying to do, without telling us about what you tried.
Nov 07 18:41:26 <zinx> sleepster: but at the hardware level, you can't specify which screen chips are used, which accelerometer chip is used, etc
Nov 07 18:41:42 <sleepster> I gotcha.. that makes sense
Nov 07 18:42:02 <zinx> sleepster: i would be very surprised if anyone had modified qemu for that support, it's ridiculously hard to do, and most people don't have the necessary specifications
Nov 07 18:42:11 <kevinb> also a fair amount of compatibility issues, for me at least, come from vendor software rather than hardware. so you really need the real device to fully test against it
Nov 07 18:42:38 <sleepster> zinx: I see.  I am just wondering if given the exact phone and model (e.g. Samsung SGH-959T, etc.), is there a way to test that exact phone model
Nov 07 18:42:51 <sleepster> without buying a 300$ phone of course
Nov 07 18:42:52 <zinx> sleepster: yes, get that exact phone and try your software on it
Nov 07 18:43:00 <sleepster> hah
Nov 07 18:43:01 <zinx> sleepster: i think you're asking the wrong questions though
Nov 07 18:43:20 <zinx> sleepster: if you're having a problem on specific devices, you're probably not using the standard Android API
Nov 07 18:43:42 <zinx> sleepster: if you /are/ using the standard API, and are still having problems, you should make a test case and submit it so that it can be added to the Compatibility Test Suite
Nov 07 18:43:51 <sleepster> I am making some modifications to the android OS, and I was hoping to create a test suite that runs all across the most popular phones and models to ensure that things are not broken
Nov 07 18:43:55 <zinx> sleepster: because it's a bug in the phone
Nov 07 18:43:59 <zinx> ah
Nov 07 18:44:07 <tdignan> sleepster: are there any android meetups near you?
Nov 07 18:44:15 <tdignan> some of them might have device sharing programs
Nov 07 18:44:23 <Reaga> tdignan: im just trying to change the theme of my very basic application but nothing is happeneing
Nov 07 18:44:24 <sleepster> tdignan: ah interesting..
Nov 07 18:44:24 <tdignan> like, loaners.
Nov 07 18:44:29 <zinx> sleepster: embedded hardware is nowhere near that simple, and almost all devices have customized versions of Android
Nov 07 18:44:33 <tdignan> everyone groups their devices together
Nov 07 18:44:50 <tdignan> Reaga: are you setting the theme in the manifest?
Nov 07 18:45:00 <sleepster> I see. thanks guys for the input.. much appreciated
Nov 07 18:45:01 <tdignan> paste your manifest
Nov 07 18:45:49 <sleepster> on to question #2 :)  I need a way to secure communication between my Linux service and a set of specific apps that I am loading onto the appliance... is there a simple, fast, secure way to do this?
Nov 07 18:46:02 <sleepster> I looked into BSD sockets, but I don't like the idea of having to encrypt/decrypt
Nov 07 18:46:05 <sleepster> to secure the channel
Nov 07 18:46:30 <sleepster> I also noticed that Android OS supports DBus
Nov 07 18:46:53 <sleepster> I was thinking of using DBus since I am familiar with it, but not sure if there is a better solution
Nov 07 18:47:30 <Reaga> tdignan: yeah
Nov 07 18:48:05 <tdignan> sleepster: whats wrong with intents
Nov 07 18:48:35 <Reaga> tdignan: http://pastebin.com/UGCJEggs
Nov 07 18:48:39 <tdignan> oh, "Linux service"
Nov 07 18:48:44 <sleepster> tdignan: intents are global and I am worried about spoofing possibilities
Nov 07 18:48:44 <tdignan> not android service :)
Nov 07 18:48:44 <Reaga> i set my theme to apptheme
Nov 07 18:48:55 <tdignan> sleepster: I am not sure you are 100% right about that.
Nov 07 18:49:05 <tdignan> intents have security facilities, permissions, and all that
Nov 07 18:50:00 <tdignan> Reaga: lets see GoalTheme
Nov 07 18:50:06 <sleepster> tdignan: do you know if intents can be hijacked?
Nov 07 18:50:06 <tdignan> er AppTheme
Nov 07 18:50:18 <sleepster> there will be very important information going through this channel
Nov 07 18:51:17 <kevinb> on a non-rooted phone, an intent protected by a permission that is signature or system only is safe. on a rooted phone  nothing is safe
Nov 07 18:51:51 <Reaga> tdignan: whats GoalTheme?
Nov 07 18:52:28 <kevinb> you can also use IBinders from linux processes, but things can be awkward because you can't easily get a context to like bind to a service
Nov 07 18:52:38 <tdignan> sleepster: read the BroadcastReceiver doc page
Nov 07 18:52:51 <tdignan> access control can be imposed on both the send and receive end of a broadcast
Nov 07 18:52:59 <tdignan> so yeah, you can supposedly secure it.
Nov 07 18:53:00 <Reaga> did you mean AppTheme?
Nov 07 18:53:08 <tdignan> Reaga: 13:50 < tdignan> er AppTheme
Nov 07 18:53:21 <Reaga> oh right, i dint see that message
Nov 07 18:53:25 <Reaga> im not very observant
Nov 07 18:53:51 <evident> hi everybody... On my app I sometimes get an IllegalStateException, that the contents of a ListView have changed without notifying the ListView that it has changed. Unfortunately I have no idea where this comes from. This only happens sometimes, I think after the app in background has been deleted when there is not enough RAM. Does anybody have an idea, what method exactly is causing the exception? http://pastebin.com/N7mKik8v
Nov 07 18:54:04 <Reaga> tdignan: http://pastebin.com/Z67SfDbK
Nov 07 18:54:37 <tdignan> evident: you need to call notifyDataSetChanged() when you add or remove items from the data that backs your ListAdapter
Nov 07 18:56:21 <Jokr> i fixed it finally
Nov 07 18:56:32 <Jokr> such a breath of fresh air
Nov 07 18:56:38 <evident> tdignan: I know... I do this when I add or delete items of the list every time... Most of the time works fine. As I said this is only sometimes, after the app has been in background for a while...
Nov 07 18:56:53 <Raimonds> Hello. Can anyone say why this is charshing the app https://gist.github.com/4033611 Its the same as in Documentation.
Nov 07 18:57:00 <tdignan> Reaga: looks fine -- so what do you expect that to do?
Nov 07 18:57:54 <tdignan> evident: well, what's in the data when it comes back?
Nov 07 18:58:10 <tdignan> evident: set a breakpoint and look or run some code from onResume to print it out
Nov 07 18:58:23 <Reaga> tdignan: to show a theme other than the default one
Nov 07 18:58:39 <tdignan> Reaga: do you know what the default theme is on your device?
Nov 07 18:58:47 <tdignan> what device are you using and what is its default
Nov 07 18:59:07 <Reaga> i dont know what the default is
Nov 07 18:59:11 <tdignan> try Theme.Holo.Light.. That will probably change it for you :)
Nov 07 18:59:17 <Reaga> ive just been trying all the different themes and they all look the same
Nov 07 18:59:17 <tdignan> I think your default *is* holo.
Nov 07 18:59:38 <Reaga> no its still the same
Nov 07 19:00:08 <Reaga> should i just give up
Nov 07 19:00:09 <tdignan> I think your next step should be to do the research to find out what your default theme is
Nov 07 19:00:17 <tdignan> how about on the emulator?
Nov 07 19:00:31 <evident> tdignan: hmm well I don't know what is in the data. It just crashes the app, when I restart it it works fine again... And the error doesn't happen very often, so it's hard to reconstruct... I guess I will have to put in some debug information and output and hope I will see then where exactly this is caused...
Nov 07 19:00:35 <Reaga> o god not the emulator
Nov 07 19:00:55 <tdignan> evident: it's probably happening only if your activity is recycled
Nov 07 19:01:13 <tdignan> er, no
Nov 07 19:01:20 <onigunn> I'm trying to get the current users name, mail and phone number. ContactsContract.Profile provides me with user_display_name how do I get the other requested values? Probably someone has a snippet for this?
Nov 07 19:01:25 <tdignan> Just run some code from onResume()
Nov 07 19:01:29 <tdignan> to print out the data
Nov 07 19:01:43 <root66> http://johnnystreet.com/synthesizer/ <-- finished the basics, working on jazzing it up still
Nov 07 19:02:15 <root66> it's a fun project to play with, so I am just sort of journaling what I learn as a how-to
Nov 07 19:02:25 <evident> tignan: that's what I am thinking, too... that maybe the data is getting garbage collected somehow...
Nov 07 19:03:22 <Reaga> hey guys
Nov 07 19:03:32 <g00s> root66: cool. i think moog has a synth engine for iStuff
Nov 07 19:03:41 <tdignan> evident: where does the data come from/
Nov 07 19:03:52 <TachyonDev2> Hey everybody! Go grab the new Android GroupMe Beta! :)
Nov 07 19:03:54 <TachyonDev2> http://goo.gl/iW3hw
Nov 07 19:04:11 <TachyonDev2> facebook stole our deck idea >_>
Nov 07 19:04:24 <preds> what happens if you use res/layout-sw600dp/* on a device <3.2, will it just ignore that directory entirely?
Nov 07 19:05:43 <Raimonds> If I get IllegalStateException do I need to just trycatch it?
Nov 07 19:06:18 <preds> if you can't prevent it then sure..
Nov 07 19:06:32 <tdignan> "Cool app, would totally use it if I had friends" 5 stars
Nov 07 19:06:55 <Raimonds> preds I get it when trying to launch new Activity, so I guess. Thanks (:
Nov 07 19:07:49 <pfn> preds, it gets ignored
Nov 07 19:07:58 <preds> thanks pfn
Nov 07 19:08:44 <preds> Raimonds, the exception should give you more info as to why it's being thrown though, starting a new activity is unlikely to be a case where you should try/catch it
Nov 07 19:10:36 <canadiancow> preds, yes, it ignores it
Nov 07 19:11:34 <AddisonE> Hello, I have a simple settings preference screen that has two checkboxes currently. Whenever I initially open the settings activity, it says I used 4.0KB data, and then when I check one or the other, it increases to 8.0KB. How come this is so high?
Nov 07 19:11:51 <AddisonE> Two simple booleans shouldn't be 8KB.
Nov 07 19:12:27 <canadiancow> im going to go out on a limb and say it has something to do with the filesystem
Nov 07 19:12:43 <canadiancow> and how all files are multiples of 4KB
Nov 07 19:12:54 <canadiancow> so you open the app, and you get some kind of default file, taking 4K
Nov 07 19:13:01 <Jokr> the constant change in orientation crashed my app... i tried to add android:configChanges="orientation|keyboardHidden" in my manifest to stop my application from calling OnCreate again but it still gets called (i added onConfigurationChanged) and its not working.
Nov 07 19:13:04 <canadiancow> then you change a pref, and it creates the pref XML, which is another 4k
Nov 07 19:13:11 <canadiancow> and why are you worried about 8k?
Nov 07 19:13:11 <Raimonds> preds https://gist.github.com/4033611 the thing is, maybe my eclipse is bugging out, I dont get any info from logcat or console it just crashes anyway, but Thread is only place that shows its suspended. http://i.imgur.com/9tNGr.png It just opens the View.java file on crash.
Nov 07 19:13:20 <tdignan> brilliant deductions
Nov 07 19:13:50 <AddisonE> My android internal storage memory is precious, and I can create a XML file that stores preferences in under a KB. It could start to add up easily.
Nov 07 19:14:05 <canadiancow> upgrade your device
Nov 07 19:14:15 <canadiancow> this is how filesystems work
Nov 07 19:14:31 <AddisonE> What do you mean?
Nov 07 19:14:39 <canadiancow> pretty sure ijust explained it
Nov 07 19:15:19 <AddisonE> Yeah, but that's not how all filesystems work? I'm not sure where you're going with it.
Nov 07 19:15:46 <canadiancow> well, if your device is exhibiting that behaviour...
Nov 07 19:15:54 <tdignan> perhaps instead of quarreling you should research what filesystem you have and whether this is true of it.
Nov 07 19:15:58 <tdignan> i would bet it is
Nov 07 19:17:00 <AddisonE> Ahh, I understand now. I missed the 'how all files are multiples of 4KB' sentence. It makes sense now.
Nov 07 19:17:32 <AddisonE> Thanks for your help.
Nov 07 19:17:47 <canadiancow> regardless, dotn worry about a few KB
Nov 07 19:19:05 <angelos> you can choose smaller block sizes too, but it's always a trade off between file size and fragmentation
Nov 07 19:19:14 <angelos> (which wouldn't matter much on flash memory anyway)
Nov 07 19:19:25 <dragorn> unless your filesystem does tail packing (giggity) you're going to be dealing w/ the minimum block size
Nov 07 19:19:32 <canadiancow> yea... i think most modern filesystems dont make sense for flash memory
Nov 07 19:19:41 <angelos> indeed
Nov 07 19:20:01 <dragorn> adding another preference item isn't going to add another 4k unless you make another preference file.  Adding another 4.01k of preference options will make it 8k
Nov 07 19:20:04 <angelos> not sure if btrfs has a minimum block size, ext4 has 1024 according to man
Nov 07 19:20:32 <canadiancow> i remember back in the day setting it to whatever the maximum was on NTFS
Nov 07 19:20:37 <canadiancow> to try to reduce fragmentation
Nov 07 19:20:42 <Reaga> tdignan: i ran it in the emulator, looks same
Nov 07 19:25:36 <g00s> canadiancow: yeah, it was common to tune filsystem block size by workload; for a mail server you would use small blocks
Nov 07 19:26:04 <canadiancow> this was on my desktop
Nov 07 19:26:11 <canadiancow> storage space was cheap, speed was expensive
Nov 07 19:26:26 * hackkitten waves at g00s o/
Nov 07 19:26:32 <g00s> hey hackkitten :D
Nov 07 19:26:48 <hackkitten> :D
Nov 07 19:26:54 * hackkitten hi-paws g00s ^--^
Nov 07 19:26:57 <hackkitten> how's things?
Nov 07 19:27:03 <hackkitten> still lovin' the Android?
Nov 07 19:27:18 <g00s> hehe, as loving as i get, heh
Nov 07 19:27:34 <hackkitten> :)
Nov 07 19:27:43 <g00s> hackkitten: how are the games doing :)
Nov 07 19:27:43 * hackkitten is focusing more on cross-platform these days :o
Nov 07 19:27:46 <hackkitten> doing alright
Nov 07 19:27:50 <docmur> when detecting touch events how do I figure out if the user has touched and moved.  I thought it was MotionEvent.ActionMove but the break point never triggers when I swipe on the screen
Nov 07 19:27:52 <hackkitten> about to release the first Windows game
Nov 07 19:27:59 <hackkitten> and working on a new cross-platform mobile game
Nov 07 19:28:08 <hackkitten> WP7 and Android for now
Nov 07 19:28:09 <g00s> hackkitten: use a cross platform game engine, or natice stuff ?
Nov 07 19:28:16 <hackkitten> former
Nov 07 19:28:23 <hackkitten> no idea how to get it on iOS, though
Nov 07 19:28:26 * hackkitten kicks stupid Apple
Nov 07 19:28:55 <g00s> hackkitten: which engine ?
Nov 07 19:30:46 <jasta> god damnit, the ADT's dependency resolution stuff is such a huge mystery to me.
Nov 07 19:30:50 <jasta> it's such magic.
Nov 07 19:31:00 <hackkitten> g00s > Cocos2D
Nov 07 19:31:04 <lov> presto chango your project is broken
Nov 07 19:31:04 <jasta> randomly closing projects and opening them up again seems to actually be a necessary step in a lot of cases
Nov 07 19:31:08 * lov waves a wand
Nov 07 19:31:53 <hackkitten> as well as cleaning projects
Nov 07 19:32:00 <hackkitten> I clean every time before I run :D
Nov 07 19:34:23 <Reaga> i tired to learn more about themes here http://developer.android.com/guide/topics/ui/themes.html#PlatformStyles
Nov 07 19:34:30 <canadiancow> hackkitten, cocos2d-x?
Nov 07 19:34:45 <Reaga> but the link to here is dead http://android.git.kernel.org/?p=platform/frameworks/base.git;a=blob;f=core/res/res/values/styles.xml;h=d7b654e49809cb97a35682754b1394af5c8bc88b;hb=HEAD
Nov 07 19:35:05 <Reaga> its Android Styles (styles.xml) on the first web pge
Nov 07 19:35:08 <Reaga> so now i cant learn
Nov 07 19:35:44 <Reaga> stupid thing sucks
Nov 07 19:36:07 <hackkitten> canadiancow > yup
Nov 07 19:36:41 <canadiancow> :(
Nov 07 19:36:44 <canadiancow> when i tried it i hated it
Nov 07 19:36:49 <hackkitten> why is that?
Nov 07 19:36:50 <canadiancow> it was the worst
Nov 07 19:36:56 <Reaga> if the name of the file is styles.xml, whya re styles referred to with @style rather than @styles
Nov 07 19:37:21 <SimonVT> Because file names don't matter in values/
Nov 07 19:37:23 <jasta> "Warn about package private methods in subclasses with the same signatures as package private methods in a different package: Dalvik will treat these as overridden methods." -- whoa, it will?
Nov 07 19:37:27 <SimonVT> And you defined a <style />
Nov 07 19:37:55 <jasta> oh, nevermind, i misread as one of them being private.  carry on.
Nov 07 19:37:59 <canadiancow> note to self: never use package private
Nov 07 19:38:47 <hackkitten> :3
Nov 07 19:38:58 <hackkitten> canadiancow > any specific issues with cocos2d-x?
Nov 07 19:39:14 <canadiancow> i wrote a document about it, but i no longer have access
Nov 07 19:39:23 <canadiancow> and it was like 17 months ago
Nov 07 19:39:26 <hackkitten> :o
Nov 07 19:39:30 <canadiancow> so 1. i dont remember, and 2. things could have changed
Nov 07 19:39:35 <canadiancow> :)
Nov 07 19:39:45 <canadiancow> that was before i even worked for zynga
Nov 07 19:40:03 <hackkitten> ah :3
Nov 07 19:40:09 <TachyonDev2> canadiancow: yo, did you grab our new beta?
Nov 07 19:40:19 <hackkitten> canadiancow > which engine would you recommend otherwise?
Nov 07 19:40:21 <canadiancow> TachyonDev2, no
Nov 07 19:40:29 <canadiancow> hackkitten, i have never used a cross platform that i liked
Nov 07 19:40:40 <canadiancow> and im no longer in games at all, so i havent been ekeping up
Nov 07 19:40:40 <hackkitten> ah
Nov 07 19:40:44 * hackkitten nods
Nov 07 19:40:45 <hackkitten> okay :)
Nov 07 19:43:15 <TachyonDev2> Need more good android peeps to try out the beta of groupme
Nov 07 19:44:56 <g00s> TachyonDev2: maybe, it would be a good idea to do some usability testing … less haphazardly
Nov 07 19:45:29 <g00s> its not that hard setting up a study to get some basic qualitative data
Nov 07 19:45:39 <TachyonDev2> g00s: lol, we do buddy
Nov 07 19:46:05 <TachyonDev2> g00s: but predicting the android ecosystem before a launch is always a trick
Nov 07 19:46:09 <g00s> what was your methodology ?
Nov 07 19:46:56 <TachyonDev2> g00s: don't have time to explain it at the moment, but just give it a whirl for yourself :)
Nov 07 19:47:04 <TachyonDev2> its mostly about compatibility testing now
Nov 07 19:47:09 <codebutler> TachyonDev2: you work at groupme? can I join your beta?
Nov 07 19:47:09 <TachyonDev2> and iOS just went live today too
Nov 07 19:47:22 <TachyonDev2> codebutler, its public beta! http://goo.gl/iW3hw
Nov 07 19:47:30 <codebutler> ah!
Nov 07 19:47:37 <TachyonDev2> :)
Nov 07 19:47:40 <Reaga> tdignan: i did it by using  this.setTheme(R.style.AppTheme;
Nov 07 19:50:10 <kevinb> TachyonDev2: i highly recommend posting to http://reddit.com/r/Android and http://forum.xda-developers.com/ if you haven't already. those guys love betas.
Nov 07 19:51:03 <canadiancow> TachyonDev2, or you could try not using #android-dev as your corporate advertising medium
Nov 07 19:51:37 <calvinchan> hi
Nov 07 19:51:58 <calvinchan> has anyone tried reading a piece of xml code in java and create an object?
Nov 07 19:52:19 <g00s> xmlpull works for that :)
Nov 07 19:52:26 <calvinchan> I want to create a button from xml code but not using the traditional way of putting the code in a layout xml file
Nov 07 19:52:38 <TachyonDev2> kevinb: thanks man
Nov 07 19:52:50 <TachyonDev2> canadiancow: dude, its a beta, im curious what you guys think
Nov 07 19:52:59 <codebutler> TachyonDev2: is there supposed to be space below the text entry bar? looks a bit off
Nov 07 19:53:12 <TachyonDev2> codebutler: it's for typing notifications
Nov 07 19:53:39 <canadiancow> and you've done nothing but talk about it non-stop for like 24 hours
Nov 07 19:53:40 <codebutler> hm
Nov 07 19:53:54 <calvinchan> g00s: but after I read the xml, how can I "feed" into the button object?
Nov 07 19:54:12 <g00s> do you want to just manually inflate it then ?
Nov 07 19:54:42 <codebutler> TachyonDev2: if you push the button that looks like an EKG display, then push the back button, it closes the app instead of going back to where you were
Nov 07 19:54:47 <g00s> also, never feed buttons … bad things can happen
Nov 07 19:54:58 <calvinchan> g00s: why so?
Nov 07 19:55:03 <codebutler> TachyonDev2:  actually "Start Group" has the same problem
Nov 07 19:55:05 <g00s> :P just joking
Nov 07 19:55:55 <calvinchan> haha, so I don't have deep understanding on inflate
Nov 07 19:56:08 <calvinchan> basically I want to read a xml code and create a button from it
Nov 07 19:56:09 <TachyonDev2> yeah there was a big discussion on if you should go to groups or the chat before exiting
Nov 07 19:56:18 <TachyonDev2> codebutler: what phone? screenshot?
Nov 07 19:56:37 <codebutler> galaxy nexus
Nov 07 19:58:05 <TachyonDev2> hmmm
Nov 07 19:58:11 <TachyonDev2> yeah thats a design thing that they wanted
Nov 07 19:58:19 <TachyonDev2> most of the spacing and graphics are all them
Nov 07 19:59:19 <codebutler> the grey space below the white text entry bar looks odd to me, and the EditText doesn't look like it has enough padding below it.
Nov 07 19:59:25 <codebutler> but the back button is a bigger issue
Nov 07 20:01:01 <TachyonDev2> codebutler: the idea is most people only have a couple groups, so they want to go back to their chat before leaving
Nov 07 20:01:12 <hkhalid> when I try to change my activity to a fragmentactivity, its just blank, no errors, I think it has something to do with how I create a new instance of it, anyone know?
Nov 07 20:02:21 <calvinchan> g00s: so I have an xml file that only contains the button xml code
Nov 07 20:02:23 <codebutler> TachyonDev2: it doesn't return to the chat though.
Nov 07 20:02:34 <calvinchan> and then I want to read from it and create a button from there
Nov 07 20:03:09 <g00s> calvinchan: look up layout inflation
Nov 07 20:09:33 <calvinchan> g00s: http://stackoverflow.com/questions/7803771/call-to-getlayoutinflater-in-places-not-in-activity
Nov 07 20:09:57 <calvinchan> what would be R.layout.myNewInflatedLayout??
Nov 07 20:10:35 <docmur> If I want to pick up a touch event with a move I can grab the down touch with motionEvent.getX() / getY() but how do I get the point I ended up at?
Nov 07 20:11:36 <Ge0rG> is it possible to un-format an NFC tag?
Nov 07 20:21:22 <Reaga> hey guys,  i solved my problems
Nov 07 20:21:23 <Reaga> :)
Nov 07 20:21:30 <Reaga> i need tow rite my own custom gui at some point though
Nov 07 20:24:00 <ultra-> if i declare a variable as static, that means it can't be changed, right?
Nov 07 20:24:12 <p_l> no
Nov 07 20:24:13 <Mavrik> em, no.
Nov 07 20:24:19 <Mavrik> ultra-, that's "final"
Nov 07 20:24:27 <Mavrik> there IS a language manual you know? :)
Nov 07 20:24:27 <ultra-> ah!
Nov 07 20:24:32 <p_l> it's a binding-related modifier, not value-modifier
Nov 07 20:24:45 <ultra-> i know, and i could have googled it too… but i was also wondering why the compiler wasn't warning me
Nov 07 20:24:56 <p_l> ultra-: because it probably was still correct code
Nov 07 20:25:03 <ultra-> thanks
Nov 07 20:25:11 <p_l> the compiler is not sufficiently smart to call out bullshit ;)
Nov 07 20:25:32 <hkhalid> I cant figure out how to convert my activity to fragment activity...all it has is text and a button, but its just blank when I run the app....
Nov 07 20:30:17 <Gsingh> In a WebView I am trying to load a url via loadUrl() my problem is that it requires a fully qualified url otherwise it will not work. Is there a way to load it with only a partail url like google.com or do I have to manually convert the string to a fully qualified url before i pass it into loadUrl() ?
Nov 07 20:31:07 <p_l> convert it
Nov 07 20:31:14 <g00s> WebView doesnt work well no matter what you give it :)
Nov 07 20:32:14 <Gsingh> ok p_l thanks that's what I thought I would have to do...
Nov 07 20:32:48 <hkhalid> technically can I just extent my Activity to FragmentActivity and add a newinstance method?
Nov 07 20:34:22 <docmur> My section of code: pastebin.com/UP5mgRay  stoppe working when I made startX, startY, finishX and finishY local to the class and not just the onTouch function.  Anyone know why?
Nov 07 20:54:42 <kakazza> g00s: How does it not work well exactly?
Nov 07 20:55:43 <kakazza> Also, a question. I began a new App using Fragments from the beginning. However I am not sure if I should re-use the Activity as some sort of container and just load the different Fragments in there or separate them. But having one Activity per Fragment seems to defeat the purpose.
Nov 07 20:56:09 <kakazza> So far, the App only has two Fragments and one PreferenceActivity.
Nov 07 20:56:11 <g00s> well, for a while WebView had some memory leaks - dunno if those have been fixed. among other things
Nov 07 20:56:24 <kakazza> Is there some kind of guideline when to use a new Activity and when to just reuse it?
Nov 07 20:56:26 <g00s> uh, and it wasnt very html compliant :|
Nov 07 20:56:33 <kakazza> g00s: :\
Nov 07 20:57:26 <mrmaffen> is there any way to use a SupportLoaderManager with a SherlockActivity ? I'm currently trying to port my SherlockFragment over to a SherlockActivity, but the method getSupportLoaderManager doesn't seem to be supported
Nov 07 20:58:38 <mrmaffen> from what i can see only a sherlockFragmentActivity would support a SupportLoaderManager. however when using the non-compat LoaderManager with getLoaderManager it does seem to work :S
Nov 07 21:03:09 <Gsingh> exit
Nov 07 21:03:11 <Gsingh> :exit
Nov 07 21:09:13 <SimonVT> mrmaffen: No, Activity doesn't have support loadermanager
Nov 07 21:09:39 <RazielZ> Hmm. Quick way to get all my source back from an apk?
Nov 07 21:09:50 <RazielZ> It wasn't compiled with proguard so it should be easy.
Nov 07 21:16:55 <TheTrash> RazielZ: apktool, then dex2jar, then that java decompiler thingy
Nov 07 21:20:44 <SimonVT> s/apktool/unzip/
Nov 07 21:30:53 <dark3n> RazielZ, HDD gone?
Nov 07 21:31:02 <RazielZ> Yup
Nov 07 21:31:13 <dark3n> No Versioncontrol ?
Nov 07 21:31:19 <RazielZ> Nope
Nov 07 21:31:27 <dark3n> github/bitbucket? , better next time
Nov 07 21:31:28 <dark3n> ;)
Nov 07 21:31:29 <RazielZ> I should probably do that though.
Nov 07 21:31:29 <RazielZ> Yeah.
Nov 07 21:31:39 <RazielZ> I've got the latest version on my phone in apk form anyway
Nov 07 21:31:51 <RazielZ> And since it's not proguarded or anything it shouldn't be a problem
Nov 07 21:32:02 <dark3n> Thats not an enjoyable method of restoring your project though ;)
Nov 07 21:32:05 <RazielZ> True
Nov 07 21:32:24 <dark3n> git clone, :)
Nov 07 21:32:34 <dark3n> bitbucket offers unlimited private repos for free too btw
Nov 07 21:32:51 <TheTrash> Bitbucket is awesome for us evil guys wanting to make money
Nov 07 21:33:09 <dark3n> TheTrash, go on...
Nov 07 21:33:26 <TheTrash> Well, because of the unlimited private repo thing :P
Nov 07 21:33:38 <TheTrash> github -> private? pay up.
Nov 07 21:33:43 <dark3n> detail more on the "evil" part
Nov 07 21:34:06 <RazielZ> Gonna use bitbucket for that then
Nov 07 21:34:12 <g00s> RazielZ: uh, you're not going to get your exact code back
Nov 07 21:34:22 <RazielZ> Why not? :[
Nov 07 21:34:30 <dark3n> cuz its been compiled
Nov 07 21:34:30 <g00s> are you kidding me ?
Nov 07 21:34:34 <RazielZ> Well.
Nov 07 21:34:35 <TheTrash> Some consider it evil to keep code for yourself, that's it :P
Nov 07 21:34:56 <dark3n> i.e. all comments are not in the apk
Nov 07 21:35:10 <RazielZ> Oh, that's okay, I didn't use comments yet anyway.
Nov 07 21:35:15 <RazielZ> Damn my horrible coding style.
Nov 07 21:35:17 <g00s> also variables will probably have lost their names, among a whole bunch of other things
Nov 07 21:35:22 <dark3n> its not a lossless conversion
Nov 07 21:35:24 <RazielZ> Even without proguard?
Nov 07 21:35:26 <RazielZ> Hm
Nov 07 21:35:41 <dark3n> like a story can be told with many different words
Nov 07 21:35:42 <RazielZ> I'll just refactor them, hopefully won't take long
Nov 07 21:35:44 <Drome> guys, anyone can help on this? http://stackoverflow.com/questions/12843119/android-bad-notification-posted-only-on-android-2-3
Nov 07 21:35:53 <dark3n> RazielZ, how big is the project?
Nov 07 21:36:00 <RazielZ> Not that big, I guess
Nov 07 21:36:04 <RazielZ> Not sure what counts as big
Nov 07 21:36:32 <dark3n> well decompile and see
Nov 07 21:36:56 <j0p> hello. i'd like to debug over wifi with a no root phone.
Nov 07 21:37:05 <dark3n> nexttime, backup :) dropbox/hdd/git repo, anything is better than getting your code back from an apk
Nov 07 21:37:08 <j0p> i've read its possible (http://stackoverflow.com/questions/4893953/android-run-install-debug-applications-over-wifi) but this doesnt work here..
Nov 07 21:37:19 <RazielZ> dark3n, will do XD
Nov 07 21:38:00 <docmur> is ther a scroll method that would get called when you page is bigger then the screen?
Nov 07 21:46:53 <roler> i am subclassing Drawable to create text based drawables on the fly - where is it appropriate to set the bounds? on the onDraw?
Nov 07 21:47:34 <roler> it seems to me the drawable is defaulting to 540 pixels wide which is way too big
Nov 07 21:53:23 <romainguy> roler: drawables don't have default bounds
Nov 07 21:53:35 <roler> what about paint?
Nov 07 21:53:40 <romainguy> what about Paint?
Nov 07 21:55:51 <pragma-> romainguy: aren't you embarrassed that the 7 tabs on about phone -> version doesn't have a plural toggle such that it shows "you are 1 steps away from being a developer" instead of "1 step away"?
Nov 07 21:55:59 <pragma-> s/tabs/taps/
Nov 07 21:56:23 <roler> I am using this class, - http://stackoverflow.com/questions/3972445/how-to-put-text-in-a-drawable - but the drawable it creates is way too big. just trying to figure out the appropriate place to set it's size
Nov 07 21:56:39 <romainguy> pragma-: I have more important things to be embarrassed about
Nov 07 21:56:58 <adq> lol
Nov 07 21:57:43 <evancharlton> pragma-: sounds like a perfect AOSP patch once the source drops
Nov 07 22:00:39 <g00s> pragma-: if you see a shadow around you getting bigger, better run - that source could drop and smite you
Nov 07 22:02:22 <canadiancow> hah pragma- i hadnt noticed that
Nov 07 22:02:31 <canadiancow> but yea, as evancharlton said...
Nov 07 22:12:48 <JakeWharton> i heard Google was closing Android's source again... #trollin'
Nov 07 22:12:57 <Drome> I'm having a storage crash report for my app, it uses DialogFragment with setTargetFragement, here is the stack trace: http://pastebin.com/qteHePx4
Nov 07 22:13:17 <Drome> s/storage/strange
Nov 07 22:13:23 <pragma-> JakeWharton: 4.2 is going to be another honeycomb, you say?
Nov 07 22:14:28 <Drome> I've read around that it was a bug in older version of support library, but I'm using the latest revision (r10)
Nov 07 22:24:08 <Sicp> JakeWharton, you're my only hope.
Nov 07 22:24:17 <Sicp> Taking a screenshot of a certain layout element, i.e. you have a square with some text inside of it, transform this square to a static image, similar to taking a screenshot of the current state of the phone.
Nov 07 22:24:18 <JakeWharton> only?
Nov 07 22:24:22 <Sicp> is, this, doable.
Nov 07 22:24:30 <evancharlton> yes
Nov 07 22:24:48 <Sicp> alright, SO has no idea though
Nov 07 22:24:55 <JakeWharton> create a Bitmap, wrap it in a Canvas, call view.draw(canvas)
Nov 07 22:24:58 <evancharlton> yep
Nov 07 22:25:00 <evancharlton> what JakeWharton said
Nov 07 22:25:26 <Sicp> and inside that Bitmap is my stuff, meaning that I have to have that element be screenshottable beforehand
Nov 07 22:25:50 <evancharlton> what?
Nov 07 22:26:10 <evancharlton> http://stackoverflow.com/questions/5536066/convert-view-to-bitmap-on-android
Nov 07 22:26:13 <Sicp> say I have an EditText and there's a photo beside it, all of that is in some layout
Nov 07 22:26:21 <Sicp> I want, on demand, to be able to screenshot that element
Nov 07 22:26:23 * evancharlton hasn't read it but "view to bitmap" sounds about right
Nov 07 22:26:26 <Sicp> that ...layout
Nov 07 22:26:42 <dark3n> evancharlton, i think i can give you a code snippet
Nov 07 22:26:51 <dark3n> made that once, to screenshot my own app
Nov 07 22:26:51 <evancharlton> dark3n: of what?
Nov 07 22:26:58 <dark3n> ah
Nov 07 22:26:59 <Sicp> it's me who's asking, dark3n
Nov 07 22:27:00 <dark3n> wrong user
Nov 07 22:28:19 <ctate> Sicp:  like they said; create a Canvas backed by a bitmap, and explicitly tell the view you want to snapshot to draw itself into that canvas
Nov 07 22:28:36 <ctate> poof! the bitmap now has whatever contents the view would currently draw.
Nov 07 22:28:53 <Sicp> very smart
Nov 07 22:29:37 <evancharlton> I see how it is
Nov 07 22:29:41 <evancharlton> jake and I explain it
Nov 07 22:29:46 <Sicp> ctate gets credit
Nov 07 22:29:48 <evancharlton> ctate swoops in with the repost, gets the credit
Nov 07 22:29:51 <dark3n> lol
Nov 07 22:29:55 <Sicp> it's what you get for oping
Nov 07 22:29:58 <JakeWharton> karma stealer
Nov 07 22:30:02 <Sicp> if you were op.....things would be different
Nov 07 22:30:09 <dark3n> Sicp, something like this
Nov 07 22:30:10 <evancharlton> k
Nov 07 22:30:10 <dark3n> http://pastie.org/5342814
Nov 07 22:30:13 <Sicp> well you are, but not NOW :P
Nov 07 22:30:16 * ctate notes that he opened with "like they said...."
Nov 07 22:30:18 <birbeck> (02:28:19 PM) ctate: Sicp:  like they said;...
Nov 07 22:30:32 <Sicp> all hail eva..
Nov 07 22:30:46 <dark3n> takes a view and gives you a jpg
Nov 07 22:30:52 <Sicp> I feel like I'm a loser if I click that link
Nov 07 22:30:53 <Sicp> haha
Nov 07 22:30:55 <dark3n> well if it would be a function, that is
Nov 07 22:31:10 <dark3n> pff
Nov 07 22:31:13 <dark3n> we're all lazy
Nov 07 22:31:35 <Sicp> I did write a randomizer for sounds in R, once
Nov 07 22:31:45 <Sicp> that's the kind of stuff you "find on the Internet"
Nov 07 22:32:47 <dark3n> sooo why do you screenshot single layout elements :>?
Nov 07 22:33:46 <Reaga1> im writing my android app now
Nov 07 22:33:54 <Sicp> well I'm working on an app idea with a friend, developing the Android part of it, it's his idea..and somewhere down the line he wants to have e-business cards
Nov 07 22:33:58 <evancharlton> cool, thanks for the update Reaga1
Nov 07 22:34:09 <Reaga1> i have this method.          public void rawr(View view) {       setContentView(R.layout.newlog);    }
Nov 07 22:34:12 <Sicp> the exchange of said cards is going to be as a JPEG with the shape of that card
Nov 07 22:34:19 <Reaga1> when i use it, when the user presses back it exist the app
Nov 07 22:34:25 <Reaga1> instead of just going back to the previous view
Nov 07 22:34:36 <Sicp> so you want to be able to take a screenshot of that area where the user filled out some stuff (the business card)
Nov 07 22:34:37 <Reaga1> i find this odd.
Nov 07 22:34:39 <evancharlton> I'm calling "not it" on this one
Nov 07 22:34:43 <evancharlton> someone else can take it
Nov 07 22:34:55 <Reaga1> its strange
Nov 07 22:35:08 <Reaga1> seems like i ahve to change the whole program just to make back work proeprly
Nov 07 22:35:10 <dark3n> Sicp,  ok, better explanation than  i thought ;)
Nov 07 22:35:22 <Sicp> first why are you setting a function to set the layout?
Nov 07 22:35:37 <Sicp> never had to do this ever..
Nov 07 22:35:47 <evancharlton> Reaga1: the short version is that setContentView doesn't add an entry to the back stack. Nor should it.
Nov 07 22:35:51 <Sicp> you reach an activity, you set its contentView in onCreate..
Nov 07 22:36:08 <Sicp> it's as if it didn't occur
Nov 07 22:36:16 <Sicp> beautiful :o
Nov 07 22:36:20 <Reaga1> evancharlton: so, i need to do someting else then
Nov 07 22:36:33 <Reaga1> create a new activity?
Nov 07 22:36:51 <Reaga1> fucking android site is being slow as shit again
Nov 07 22:36:59 <dark3n> Sicp, hm onPause/onResume which content layout is used then? the newly set one?
Nov 07 22:37:21 <Sicp> I dont know
Nov 07 22:37:35 <dark3n> Reaga1, loads fine for me
Nov 07 22:37:51 <Reaga1> :/
Nov 07 22:37:53 <Sicp> logic says that yes, if you're talking about the functions related to the activity's lifecycle
Nov 07 22:37:58 <ctate> setContentView() just declares that your window's content should be that view.
Nov 07 22:38:02 <ctate> hence the name.  :)
Nov 07 22:38:04 <Reaga1> yeah i figured
Nov 07 22:38:16 <Reaga1> but i just didnt know how the back button worked
Nov 07 22:38:18 <ctate> doesnt' have anything to do with the activity lifecycle
Nov 07 22:38:20 <Sicp> onPause will read from whatever activity it belongs to
Nov 07 22:38:38 <Sicp> in answer to dark3n, I mean
Nov 07 22:38:50 <ctate> the back button, if not intercepted and handled explicitly by the foreground activity, calls finish() on that activity.
Nov 07 22:38:52 <ctate> that's all.
Nov 07 22:39:12 <Reaga1> i see
Nov 07 22:39:18 <dark3n> i wish there were an easy way to redefine the backbutton per fragment
Nov 07 22:39:18 <Sicp> and what happens usually when "LeakedView" exception or some other, occurs?
Nov 07 22:39:34 <Sicp> Leaked...something, I remember it, when pressing back all is fine but Logcat shouts at that
Nov 07 22:39:37 <ctate> that means you've failed to properly deallocate something that you should have known better.  :)
Nov 07 22:39:43 <Sicp> ahh
Nov 07 22:39:47 <ctate> typically it's a Cursor or some such
Nov 07 22:39:49 <Sicp> LeakedWindow it says
Nov 07 22:39:58 <Reaga1> i remember ages ago i installed a launcher that updated itself and hijacked the whole android system and held you to random "buy this app for $10 of you can't access any of your apps or prorgams or do anything ever"
Nov 07 22:40:02 <ctate> or that, right.  i'd have to see the exact message again to tell
Nov 07 22:40:04 <Reaga1> managed to work my way out of it though
Nov 07 22:40:16 <ctate> Reaga1:  ow
Nov 07 22:40:27 <ctate> also we've taken steps to prevent those kinds of shenanigans more recently
Nov 07 22:40:28 <ctate> tsk tsk
Nov 07 22:41:02 <dark3n> is there an ETA on the new dev console, for everyone? not just "top developers"
Nov 07 22:42:18 <dark3n> really want that "comment on comments" feature
Nov 07 22:42:39 <Sicp> ^+1
Nov 07 22:43:01 <Sicp> in my app, some guy reported a bug about a 10-digit number giving him a NumberFormatException
Nov 07 22:43:16 <Sicp> he was putting that number in the Port Number field
Nov 07 22:43:23 <Sicp> how do you reply to that
Nov 07 22:43:42 <dark3n> I've been told my app is the worst game someone played
Nov 07 22:43:46 <dark3n> its not a game lol...
Nov 07 22:43:54 <Sicp> haha
Nov 07 22:43:56 <Sicp> link?
Nov 07 22:44:17 <adq> !i would be more concerned about profile who have an incredible amount of 1 star rates
Nov 07 22:44:46 <adq> a system like stackoverflow, alas, is not possible
Nov 07 22:44:56 <adq> to prevent ppl abusing of bad rating just because they like to bad rate
Nov 07 22:44:56 <dark3n> lots of 1 star doesn't always mean bad app
Nov 07 22:45:22 <dark3n> adq, ya sometimes they rate bad, just out of spite
Nov 07 22:45:40 <adq> dark3n, oO it's just the worst rating, better having no rate than lot of 1 star, look the color code which is not green
Nov 07 22:46:37 <adq> responsbile raters is the key, meaning whatever the rate, it will/should be justified
Nov 07 22:47:03 <adq> responsible*
Nov 07 22:48:09 <Reaga1> http://developer.android.com/images/fundamentals/diagram_backstack.png   I like this diagram, especially the white cloud with "Activity 3" written in it to represent that it is destroyed
Nov 07 22:48:10 <adq> I'm also pretty sure non ethical groups can use bots to drown their competition(s)
Nov 07 22:48:20 <adq> (totally unfair, but it's possible)
Nov 07 22:49:27 <kbs> Is there a page that lists the apps written by the regulars here? curious...
Nov 07 22:50:42 <adq> kbs, not aware of
Nov 07 22:55:27 <Reaga1> reading about android is real interesting
Nov 07 22:55:52 <g00s> in what way ?
Nov 07 22:56:04 <Reaga1> it makes me feel intelligent
Nov 07 23:02:50 <Estel> Anyone here used Polaris? Is there a builtin function to centre on the user's location on first fix?
Nov 07 23:06:07 <Gsingh> Hi, I have this problem where everything on my app appears "
Nov 07 23:06:12 <Gsingh> "zoomed in"*
Nov 07 23:06:44 <g00s> hmm. so on WP8, an app can register custom voice actions to create a Voice User Interface. is this possible on android? my research indicates so far it is limited to the voice actions baked into android, am i correct ?
Nov 07 23:06:48 <Gsingh> Everything is large icons, text, etc any ideas on what is causing this?
Nov 07 23:06:51 <Reaga1> Gsingh: are you sure your face isnt just very close to the screen?
Nov 07 23:06:58 <Gsingh> Very funny...
Nov 07 23:07:34 <Gsingh> By comparison, the other apps' widgets are small...
Nov 07 23:08:17 <Estel> So it's an app you've developed?
Nov 07 23:08:23 <Estel> Uhh s/app/widget
Nov 07 23:08:24 <Estel> Words >_>
Nov 07 23:10:15 <dark3n> screenshot, or it didnt happen :p
Nov 07 23:11:00 <Reaga1> hey
Nov 07 23:11:40 <karakuri> how do i add a new window over my application?
Nov 07 23:11:57 <karakuri> like with hints for the user?
Nov 07 23:11:58 <evancharlton> karakuri: DialogFragment
Nov 07 23:12:02 <evancharlton> if that's what you mean
Nov 07 23:12:19 <karakuri> i can't use a fragment
Nov 07 23:13:12 <Estel> Why not?
Nov 07 23:13:31 <karakuri> project architecture conflicts
Nov 07 23:13:34 <karakuri> its complicated
Nov 07 23:13:42 <evancharlton> then use dialog?
Nov 07 23:13:52 <karakuri> can dialogs be full screen?
Nov 07 23:14:37 <Gsingh> Estel: Yes, it
Nov 07 23:14:44 <Gsingh> its my own app
Nov 07 23:15:09 <kbs> Gsingh: what units are you using to specify sizes of text, etc?
Nov 07 23:15:14 <Gsingh> sp
Nov 07 23:15:26 <Estel> sp for text, dp for layout?
Nov 07 23:16:21 <Gsingh> Its not just the text its everything I have a webview and the webpages are enlarged so is the edit text and every other standard android widget.
Nov 07 23:16:42 <Gsingh> sp for everything.
Nov 07 23:16:57 <Gsingh> layout is just match_parent all the way around
Nov 07 23:17:33 <Gsingh> its as if the resolution was reset to be really low or something
Nov 07 23:18:20 <kbs> just to ask a silly question -- this is true for non-webview controls as well?
Nov 07 23:19:16 <Gsingh> kbs: yes for everything
Nov 07 23:19:38 <Gsingh> Everything on the screen is bigger then it should be.
Nov 07 23:24:07 <kbs> weird. There isn't any chance that the sp is set to a different value than dp, is there? I guess you could dump out the displayMetrics values to see if tehre's anything strange there. Other weird possibility of course is that somewhere the canvas is getting scaled, but this is flailing for theories.
Nov 07 23:25:43 <kbs> To test the last theory, I guess you could dump out the transformation matrix for the canvas for one of your internal controls
Nov 07 23:27:28 <tophyr> anyone here who does extensive work on the emu - do you commonly get broken pipe errors using HttpsURLConnection?
Nov 07 23:27:37 <tophyr> not every time, just common
Nov 07 23:31:51 <Sicp> ^^ something deeper to your bug, dude
Nov 07 23:33:04 <evancharlton> Chainfire: hey, whatever happened to that aapt patch of yours that was floating around AOSP?
Nov 07 23:34:53 <g00s> evancharlton: i thought he pushed that to gerrit. what did you guys do with it :P
Nov 07 23:35:48 <evancharlton> that's what I was wondering
Nov 07 23:35:53 <evancharlton> I was too lazy to crawl gerrit and find it
Nov 07 23:36:13 <pragma-> good thing gerrit has a search
Nov 07 23:36:18 <g00s> FAAPT :)
Nov 07 23:39:08 <Chainfire> evancharlton I actually linked you to a post in the past that had links to the Gerrit but you went off on some rant on people who use other peoples binaries are stupid and such
Nov 07 23:43:10 <ctate> btw, aapt was being substantially optimized internally right around the same time you did that work, iirc, so i'm not sure the external contribution would have been paid much attention
Nov 07 23:43:27 <ctate> so if you have gerrit links to your patch, i'd like to see them
Nov 07 23:44:13 <ctate> (specifically, images are now processed in parallel, and a lot of the string handling has been made much faster and redundant sorts avoided)
Nov 07 23:45:21 <Manchu> Hey, I have a quick question. From a hardware/software perspective, is it possible to create a USB input devices, like a microphone for andriod? I know that there is support as of 4.1 for audio output.
Nov 07 23:45:33 <Manchu> device*
Nov 07 23:46:09 <Chainfire> ctate> https://android-review.googlesource.com/#/c/43680/
Nov 07 23:46:14 <Chainfire> its not exactly brilliant or even great code
Nov 07 23:46:25 <Chainfire> but it works like WTF?!
Nov 07 23:47:01 <Chainfire> undoubtedly, somebody who has better knowledge of AAPT itself can do a better job. I know nothing about AAPT internal, I just saw this speedbump and fixed it
Nov 07 23:49:27 <Chainfire> interestingly, nobody who was tagged has even looked at it, according to Gerrit ... :P
Nov 07 23:50:04 <evancharlton> Chainfire: no, I remember that, I just didn't have the link anymore
Nov 07 23:50:14 <evancharlton> and yes, grabbing some random binary from xda is still not a good idea
Nov 07 23:50:38 <Chainfire> grabbing some random binary from ANYONE is still not a good idea. But I doubt you never grabbed a binary from anyone :P
Nov 07 23:50:55 <Chainfire> and still, it had the links to the sources even regardless of those binaries.
Nov 07 23:51:15 <ctate> hm.  i like the idea; don't much like the actual cache implementation.
Nov 07 23:51:29 <ctate> order-created linear scan doesn't seem like the best approach.
Nov 07 23:51:32 <ctate> but!
Nov 07 23:51:36 <Chainfire> ctate> it isn't
Nov 07 23:51:42 <Chainfire> as I said it isn't great code, it gets the job done
Nov 07 23:51:53 <Chainfire> if you guys fix it with something better, I won't feel offended.
Nov 07 23:52:00 <ctate> heh
Nov 07 23:52:04 <ctate> :)
Nov 07 23:52:11 <Chainfire> in fact, I'd find it surprising if you'd include this code as-is
Nov 07 23:52:19 <ctate> it's good to know that even with it being linear scan, it's a noticeable speed improvement
Nov 07 23:52:27 <Chainfire> that just it
Nov 07 23:52:39 <Chainfire> its not just noticably, its ridiculously large speed improvement
Nov 07 23:52:42 <ctate> what were you using for test cases?  [heh]
Nov 07 23:52:44 <ctate> framework?
Nov 07 23:52:46 <Chainfire> note that average cache population will be 50-100 items
Nov 07 23:52:49 <ctate> just the whole thing over and over?
Nov 07 23:52:55 <ctate> ok good to know
Nov 07 23:53:23 <Chainfire> compiling many things. for myself it had significant improvements in incremental builds
Nov 07 23:53:35 <Chainfire> i.e. when you're in eclipse, change something, build, change something, build
Nov 07 23:53:46 <Chainfire> those times went down from minutes to seconds on some projects
Nov 07 23:53:56 <ctate> kk
Nov 07 23:54:27 <Chainfire> I've had users report the same dramatic speed increases for some projects when doing an export to APK, but I don't actually find that a noticable difference for myself
Nov 07 23:55:38 <Chainfire> also note that whats being looked up here are mostly XML attributes, like, whats the binary id for "android:layout_width"
Nov 07 23:57:14 <ctate> right
Nov 07 23:57:21 <ctate> okay, thanks for the guidance!
Nov 07 23:57:39 <Chainfire> np, hope you guys implement this or something that does the same thing but better :)
Nov 08 00:05:32 <RustyShackleford> Would it be possible to have an activity that displays 1 or 2 ListViews, depending on what Intent I pass it?
Nov 08 00:06:05 <canadiancow> ive made so many private inner DialogFragment classes, and every single time, i get an exception on configuration change, and have to make it public
Nov 08 00:06:06 <canadiancow> i never learn
Nov 08 00:06:11 <dark3n> RustyShackleford, sure, why not?
Nov 08 00:06:13 <canadiancow> yes RustyShackleford
Nov 08 00:06:21 <RustyShackleford> or would I be better off making two separate activites?
Nov 08 00:06:41 <dark3n> RustyShackleford, depends how different the code is
Nov 08 00:07:02 <dark3n> *how much less duplicate code it gives you
Nov 08 00:07:24 <dark3n> difficult to say without seeing the complete picture
Nov 08 00:07:30 <RustyShackleford> perhaps there is a better way to display short (<5 items) lists of data
Nov 08 00:08:57 <RustyShackleford> dark3n: the items will be very similar. I just want to sort them into two separate ListViews
Nov 08 00:10:07 <kbs> I've found the metro ui pattern to be nice for that sort of thing; but it's a bit of work to get it tweaked into nadroid
Nov 08 00:10:24 <kbs> ie, a horizontally pannable set of grouped tiles
Nov 08 00:12:25 <mrmaffen> i'm currently working on integrating a search dialog into my app project. I've managed to implement the standard behaviour, meaning that when the user presses the search actionButton a EditText pops up
Nov 08 00:12:30 <mrmaffen> however i want to customize the style of my search dialog
Nov 08 00:12:35 <mrmaffen>  is that possible? or do i need to create my own custom search dialog in order to be able to do this ?
Nov 08 00:12:40 <mrmaffen> by "style of my search dialog" i mean the colors basically
Nov 08 00:14:56 <adq> mrmaffen, yes it is possible. in your AutoCompleteTextView you can set android:background and android:popupBackground
Nov 08 00:15:32 <adq> be aware you might need to play with 9-patch drawable
Nov 08 00:16:51 <adq> and if you are using ABS, you can try http://jgilfelt.github.com/android-actionbarstylegenerator
Nov 08 00:16:54 <adq> which is really great
Nov 08 00:16:55 <mrmaffen> adq: thanks for answering. where exactly can i find the AutoCompleteTextView of the standard android search dialog ?
Nov 08 00:17:43 <adq> mrmaffen, ah I assumed you already mimic the standard search dialog, which is mainly an edittext and few layouts & views
Nov 08 00:17:51 <adq> so i don't know more:p
Nov 08 00:17:52 <mrmaffen> adq: yeah i'm using abs. and i'm already using the style generator. it really is a great tool. however this didn't help me customize my search dialog :S
Nov 08 00:18:04 <mrmaffen> kk :S
Nov 08 00:18:26 <mrmaffen> searched the web for quite some time now. i guess i'll mimic my own search dialog then
Nov 08 00:32:13 <Reaga1> im not too familiar with android, especially this line: Intent intent = new Intent(this, DisplayMessageActivity.class);
Nov 08 00:32:23 <Reaga1> what does it mean when i type DisplayMessageActivity.class
Nov 08 00:32:34 <Reaga1> .class
Nov 08 00:34:10 <sonOfRa> The Intent needs a Class, not an instantiated object/reference to an object of a class.
Nov 08 00:34:22 <Reaga1> i see
Nov 08 00:34:39 <sonOfRa> If you start an Activity with that Intent now, the DisplayMessageActivity class you defined, is the class that holds the activity
Nov 08 00:46:15 <RustyShackleford> should I use Fragments and why?
Nov 08 00:46:31 <RustyShackleford> rather than making several Activities
Nov 08 00:46:46 <sonOfRa> Well, you don't stuff everything in one activity with fragments
Nov 08 00:47:19 <sonOfRa> Combining the two is the best recipe. For example, you can even have Fragment Activities, which can basically be used as both
Nov 08 00:50:30 <sonOfRa> Which is quite convenient if you are layouting for tablets and phones at the same time
Nov 08 00:52:46 <ron_frown> RustyShackleford reuse
Nov 08 00:53:06 <ron_frown> and sharing layout etc for various screen sizes
Nov 08 00:53:19 <ron_frown> tablet can have a listview on the left and a details view fragment on the right
Nov 08 00:53:42 <ron_frown> then have diff activities housing thme so you can click on a list view and have details etc etc etc
Nov 08 00:53:47 <RustyShackleford> i think people with tablets are out of luck on this app
Nov 08 00:54:00 <RustyShackleford> I have not really planned for larger screen sizes
Nov 08 00:55:38 <cloudgroup> fragments are also very useful for reusing certain parts of your app...
Nov 08 00:56:27 <dark3n> they are the future
Nov 08 00:56:29 <dark3n> :o
Nov 08 01:01:14 <tophyr> wtf.... http://www.screencast.com/t/OoYy6X0klST
Nov 08 01:02:41 <ron_frown> tophyr I dont trut that shit
Nov 08 01:02:44 <ron_frown> what did the log spit out
Nov 08 01:02:51 <ron_frown> eclipse has major problems scoping variables
Nov 08 01:03:12 <tophyr> didn't spit anything out, i just kill -9'ed eclipse and restarted it
Nov 08 01:03:28 <tophyr> when something like that happens... eclipse is just horked
Nov 08 01:03:43 <ron_frown> also people can explcitly throw exceptions that dont include stack traces
Nov 08 01:04:01 <tophyr> still can't throw null; tho, lol
Nov 08 01:04:09 <g00s> also, its usually a good idea for your exceptions to take a Throwable cause
Nov 08 01:04:20 <g00s> well, uh, nm. sometimes
Nov 08 01:04:26 <tophyr> OR CAN YOU
Nov 08 01:04:29 <tophyr> throw null; compiles
Nov 08 01:09:24 <g00s> "Windows 8 has overtaken Android with regards to web traffic" … well that was quick
Nov 08 01:10:39 <canadiancow> i tend to browse a lot more on desktop than on android
Nov 08 01:10:43 <canadiancow> desktop/laptop
Nov 08 01:11:20 <g00s> iits hard hoof-optimizing touch targets for mobile websites, yeah
Nov 08 01:13:07 <ctate> interestingly, there are certain web-ish things i prefer to do on android devices
Nov 08 01:13:10 <ctate> Reader in particular
Nov 08 01:13:23 <ctate> so i do a lot of web page reading on Android as a consequence of finding the links etc in Reader
Nov 08 01:15:10 <_klk_> hi all.  I'm using NDK to build a C++ application.  it works great, but exception handling is a bit wonky.  basically, if i do catch(...) {} it works, if i do catch(std::exception &exc) {}, it segfaults.  here's a clearer example:  http://pastebin.com/F1WCGpPk
Nov 08 01:15:32 <_klk_> this was using NDK-r7.  I'm going to try NDK-r8b and see if it resolves the issue
Nov 08 01:15:45 <canadiancow> ctate, i do g+ reading primarily on android
Nov 08 01:15:48 <canadiancow> i ahve no idea why
Nov 08 01:18:54 <Hoosteen> heyoo
Nov 08 01:19:15 <BadMrsFrosty> What instructions are available for prefetching on Android?  Or does it depend on the device?
Nov 08 01:20:40 <canvas> hi, I need a very good tip to show a view with no user interaction, but updating, under a readerview class using also a canvas (???)
Nov 08 01:21:52 <canvas> (a surfaceview, I mean)
Nov 08 01:22:46 <canvas>  hi, I need a very good tip to show a view with no user interaction, but updating, under a running surfaceview having all updates, touch and input s from UI and user
Nov 08 01:26:18 <g00s> ron_frown: are you using a virus scanner on your macs ?
Nov 08 01:29:14 <ron_frown> no
Nov 08 01:29:23 <ron_frown> I use them on my fuckign vm inside of my mac
Nov 08 01:29:26 <ron_frown> which is retarded
Nov 08 01:29:28 <ron_frown> but windows needs it
Nov 08 01:29:35 <ron_frown> fucking look at it wrong and BAM
Nov 08 01:29:35 <ron_frown> virus
Nov 08 01:29:59 <Hoosteen> youre obviously doing something wrong
Nov 08 01:30:59 <g00s> i was going to try the free sophos one for mac, but that google guy had some things to say about that
Nov 08 01:31:23 <g00s> sucks when the virus scanner has critical vulnerabilities :|
Nov 08 01:31:46 <Hoosteen> Dont use mac?
Nov 08 01:31:49 <Hoosteen> problem solverd
Nov 08 01:32:00 <Hoosteen> LD
Nov 08 01:32:03 <Hoosteen> XD
Nov 08 01:37:56 <BadMrsFrosty> No one knows about prefetching?
Nov 08 01:38:06 <BadMrsFrosty> Everyone is Java programmer?
Nov 08 01:39:05 <ctate> prefetching of what?
Nov 08 01:39:08 <ESphynx> nope. eC programmer.
Nov 08 01:39:25 <BadMrsFrosty> Prefetching memory into the cache.
Nov 08 01:39:29 <ctate> cpu cache prefetch?  http request prefetch?
Nov 08 01:39:35 <ctate> aha
Nov 08 01:39:37 <BadMrsFrosty> CPU cache.
Nov 08 01:39:50 <ctate> well, nothing is in the public API for that level of control
Nov 08 01:40:11 <ctate> partly because Android apps run on three (and a half) different major cpu architectures
Nov 08 01:40:31 <BadMrsFrosty> That’s normal.  On any platform you have to use assembly directly.  But with the different CPU’s out there…
Nov 08 01:40:47 <ctate> if you want to do it yourself in your app, you'll need to test your runtime architecture, include the right asm-level sequence for the platform your'e on at runtime, etc.
Nov 08 01:40:51 <BadMrsFrosty> But I was just hoping there would be a common way.
Nov 08 01:40:54 <SpeedEvil> BadMrsFrosty: the CPU cache is tiny
Nov 08 01:40:54 <ctate> nope
Nov 08 01:41:05 <ctate> yeah, in general mobile cpus have tiny tiny caches
Nov 08 01:41:07 <SpeedEvil> why do you think you want to do this
Nov 08 01:41:11 <ctate> certainly compared to desktop/server cpus
Nov 08 01:41:15 <BadMrsFrosty> SpeedEvil, that is why prefetching can really help when done properly.
Nov 08 01:41:16 <SpeedEvil> as small as 32k
Nov 08 01:41:33 <SpeedEvil> prefwtching can really hurt
Nov 08 01:41:43 <BadMrsFrosty> When done improperly yes.
Nov 08 01:41:50 <SpeedEvil> as it causes cache thrashing
Nov 08 01:42:11 <BadMrsFrosty> But our engine runs on Xbox 360, PlayStation 3, PlayStation Vita, PlayStation Portable, Nintendo 3DS, and iOS.
Nov 08 01:42:12 <SpeedEvil> I assume you mean for ndk code?
Nov 08 01:42:16 <BadMrsFrosty> Yes.
Nov 08 01:42:34 <BadMrsFrosty> On iPhone 4, supplying the macro we use for prefetching increased the FPS by 1.
Nov 08 01:42:44 <SpeedEvil> you can only do it on a rooted phone
Nov 08 01:42:47 <BadMrsFrosty> That seems like little but for just prefetching, that is actually a lot.
Nov 08 01:42:52 <ctate> SpeedEvil: err, what?
Nov 08 01:43:13 <SpeedEvil> manipulating the cache requires root capabilities
Nov 08 01:43:25 <ctate> orly
Nov 08 01:43:35 <BadMrsFrosty> That seems suspicious.
Nov 08 01:43:43 <SpeedEvil> locking stuff in the CPU cache, of course
Nov 08 01:43:48 <ctate> that ... surprises me.  they're just cpu instructions, not syscalls
Nov 08 01:43:56 <ctate> he isn't talking about locking stuff into cache, he's talking about prefetching
Nov 08 01:43:56 <BadMrsFrosty> If so, that would be the first time I have heard of prefetching being so restricted.
Nov 08 01:44:04 <SpeedEvil> the CPU has priciledged instructions
Nov 08 01:44:16 <SpeedEvil> oh, prefetching,..
Nov 08 01:44:24 <SpeedEvil> misread
Nov 08 01:44:26 <BadMrsFrosty> __dcbt() and friends are usually not privileged.
Nov 08 01:46:33 <BadMrsFrosty> As a separate issue, we are having major performance issues with Mali-400MP devices (so this is changing to a GPU question now).
Nov 08 01:47:24 <ctate> my impression is that Mali gpus have a rather different performance *profile* than e.g. PVR gpus
Nov 08 01:47:39 <BadMrsFrosty> Is it tile-based?
Nov 08 01:48:01 <BadMrsFrosty> Are there certain operations that work well on PowerVR that actually hurt performance on Mali-400MP?
Nov 08 01:48:03 <ctate> Mali?  I believe so, yes
Nov 08 01:48:05 <ctate> (tile based)
Nov 08 01:48:07 <BadMrsFrosty> Opeations = optimizations.
Nov 08 01:48:20 <ctate> i'm not a graphics guy, unfortunately
Nov 08 01:48:43 <ctate> i would not be surprised if specific optimal approaches for one gpu archtecture turned out to be counterproductive on others
Nov 08 01:48:49 <BadMrsFrosty> There aren’t any good profiling tools for Android.
Nov 08 01:49:22 <ctate> well, depends on wha tyou're profiling and so on
Nov 08 01:49:24 <ctate> and they're working on it
Nov 08 01:49:42 <ctate> for what it's worth, there are no good profiling tools for nVidia or ARM gpus available for iOS, either :)
Nov 08 01:49:45 <BadMrsFrosty> You mean Google plans to provide better native support?
Nov 08 01:49:58 <ctate> no, i mean google plans to provide better profiling tools
Nov 08 01:50:04 <BadMrsFrosty> For iOS there are fairly good tools.
Nov 08 01:50:13 <BadMrsFrosty> Instruments, etc.
Nov 08 01:50:16 <ctate> yes, but not for ARM or nVidia GPUs :)
Nov 08 01:50:28 <ctate> [when you only have one hw architecture, it's a much easier problem.]
Nov 08 01:50:37 <BadMrsFrosty> Yes.
Nov 08 01:51:19 <BadMrsFrosty> Our only advantage is that the engine is basically the same between iOS and Android, so in general whatever I find via profiling on iOS is also a bottleneck on Android.
Nov 08 01:51:30 <BadMrsFrosty> But that doesn’t apply when the GPU is not PowerVR.
Nov 08 01:51:34 <ctate> anyway, you should probably read up on ARMs docs about how to optimize Mali output; e.g. http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0555a/CACDBIJF.html
Nov 08 01:52:14 <BadMrsFrosty> That could be very helpful thank you.
Nov 08 01:52:30 <ctate> the basics are always the same, of course.  For example, avoid overdraw as much as possible.
Nov 08 01:52:39 <ctate> that's cross-architecture :)
Nov 08 01:52:57 <BadMrsFrosty> Right.
Nov 08 01:53:19 <BadMrsFrosty> I am actually writing a book on some of these cross-platform topics.
Nov 08 01:53:26 <ctate> excellent!
Nov 08 01:54:33 <g00s> BadMrsFrosty: which publisher? or self published ?
Nov 08 01:54:59 <BadMrsFrosty> Addison-Wesley Professional.
Nov 08 01:55:05 <g00s> terrific !
Nov 08 01:55:21 <romainguy> Perfhud ES is very very good to profile and debug apps on Tegra
Nov 08 01:55:27 <BadMrsFrosty> The book aims to focus on advanced topics with a heavy focus on optimization.
Nov 08 01:56:11 <romainguy> BadMrsFrosty: Mali is tile based and working with the Mali 604 I haven't seen huge differences with SGX 543 (except raw performance of course)
Nov 08 01:56:15 <BadMrsFrosty> I need to double-check with my coworker but I think our games are not running on Tegra 2 (our only Tegra device) at this time due to poor shader compiler implementation.
Nov 08 01:56:18 <romainguy> it has the same performance cliffs
Nov 08 01:56:51 <romainguy> one I can remember: glCopyTexImage is even worse :)
Nov 08 01:57:34 <BadMrsFrosty> Okay I checked.  He just brought over a Tegra 2 and a Mali-400MP device.
Nov 08 01:57:45 <BadMrsFrosty> The Mali-400MP is running at (I guess) 5 FPS.
Nov 08 01:57:57 <BadMrsFrosty> The same game/scene runs at maybe 25 or 30 on the Tegra 2.
Nov 08 01:58:04 <BadMrsFrosty> Isn’t that fishy?
Nov 08 01:58:13 <romainguy> That's surprising :)
Nov 08 01:58:22 <BadMrsFrosty> He said the Mali-400MP is stronger than Tegra 2.
Nov 08 01:58:42 <romainguy> Apparently not for everything :)
Nov 08 01:59:14 <BadMrsFrosty> (I’ve been focusing on optimizing iOS until now so I am not yet up-to-date on Android devices.)  I am going a lot on what he says.
Nov 08 02:08:22 <ctate> the Tegra2 has crappy fill rate but iirc it can do polygons really fast
Nov 08 02:09:13 <ctate> but yeah, that sounds screwy.
Nov 08 02:12:50 <BadMrsFrosty> I am reading the PDF and will try its suggestion to determine if we are CPU or GPU limited.
Nov 08 02:13:10 <BadMrsFrosty> But our engine is fairly huge and takes about 20 minutes to build.
Nov 08 02:13:26 <BadMrsFrosty> We have another test building now so it will be a while before I can test that.
Nov 08 02:19:26 <andr0ide> hi
Nov 08 02:19:29 <andr0ide> i have an application with a frame animation
Nov 08 02:19:32 <andr0ide> but the same code works on a nexus s but not on a motorola defy
Nov 08 02:20:26 <BadMrsFrosty> I am reading that PDF about Mali-400MP and section 2.17 (“It is critical for application performance that your system is set up correctly. Even well  optimized applications run badly if your system settings are incorrect.”) stands out to me.
Nov 08 02:21:00 <BadMrsFrosty> “Check the caches are switched on”—Where do we make this check?  Code?  A device setting?
Nov 08 02:21:53 <ctate> i suspect that as an app author you can pretty much ignore the system-level recommendations
Nov 08 02:22:05 <ctate> either the device vendor will get them right, or they won't, and there's little to nothing you can do
Nov 08 02:25:24 <BadMrsFrosty> He changed the back-buffer resolution down to 160×something to test if we are GPU-bound or CPU-bound.  The FPS remained mostly the same, so we are CPU bound.
Nov 08 02:25:45 <BadMrsFrosty> That’s a major problem, we he have no good way to profile the CPU (do we)?
Nov 08 02:26:41 <BadMrsFrosty> I am pretty sure we don’t have per-frame 0-1000000 for-loops that do printf().
Nov 08 02:36:21 * g00s rages at mit ebooks 
Nov 08 02:53:19 <DrPoole> G'day
Nov 08 02:54:09 <DrPoole> I have an app that uses GCM to get notifications.  I've come across a problem where if I uninstall my app, and then days later install it again i will get a new register_id from GCM but ... the old register_id still works
Nov 08 02:54:27 <DrPoole> this causes duplicate messages to be sent (because the server doesn't know the first register_id is no longer valid)
Nov 08 02:54:33 <DrPoole> this seems like it shouldn't be possible.
**** ENDING LOGGING AT Thu Nov 08 03:00:01 2012