[01:09:45] <joakino>	 jdlrobson: we're on R35
[01:40:43] <phuedx>	 jdlrobson: yt?
[02:57:06] <josephine_l>	 Hi niedzielski  :)
[02:58:40] <niedzielski>	 josephine_l: hello :)
[02:59:35] <josephine_l>	 niedzielski, I think I'm stuck on fixing the image URI bug for real devices :/
[03:00:59] <niedzielski>	 josephine_l: so pull #38 doesn't fix it?
[03:01:30] <josephine_l>	 niedzielski, I thought it did. Then today it doesn't work again for my emulator, even though it works for my real device
[03:01:56] <josephine_l>	 niedzielski, I did put in a try catch block and such, but after it didn't work, I reverted the new commit
[03:02:10] <josephine_l>	 So I'm back to the old code again but it still doesn't work
[03:03:00] <josephine_l>	 Or maybe it shouldn't matter that it doesn't work for emulators? I dunno. :)
[03:04:05] <niedzielski>	 josephine_l: is this the relevant line in the patch? https://github.com/nicolas-raoul/apps-android-commons/pull/38/files#diff-72e3a9e6acdbd730cbc0d597e9d51c92R170
[03:04:33] <josephine_l>	 niedzielski, yeah, that's the one
[03:05:13] <niedzielski>	 josephine_l: and you're just trying to get a filesystem path?
[03:05:19] <josephine_l>	 niedzielski, yeah
[03:08:19] <josephine_l>	 I've been looking for a workaround to see if there's a way to get ExifInterface working with just a URI, but everything I've read on that just involves converting the URI to a real path anyway...
[03:08:40] <niedzielski>	 josephine_l: when Uri.getPath fails, is just returning null?
[03:08:45] <josephine_l>	 niedzielski, yeah
[03:09:00] <niedzielski>	 josephine_l: and you're sure the file it points to exists in these scenarios?
[03:09:23] <josephine_l>	 yep
[03:10:02] <josephine_l>	 niedzielski, can a user choose a file that doesn't exist for uploading?
[03:10:50] <niedzielski>	 josephine_l: i would think it would be possible but not a likely scenario. for example, if i delete a file over ADB, i'm not sure the media store is updated until reboot
[03:11:07] <josephine_l>	 niedzielski, oh, okay. But no, I haven't deleted any files.
[03:11:18] <josephine_l>	 I'll try rebooting my emulator and see
[03:11:40] <josephine_l>	 niedzielski, you think the code looks okay though? I'm not missing anything?
[03:11:51] <niedzielski>	 josephine_l: can you share the URI and file path again?
[03:12:00] <niedzielski>	 josephine_l: the URI patch is a little noisy :)
[03:12:48] <josephine_l>	 niedzielski, haha, yeah, sorry... :) How do you mean share them?
[03:12:57] <josephine_l>	 niedzielski, you mean the log output, or the code?
[03:14:03] <niedzielski>	 josephine_l: just the URI string like content://foo/bar/baz and filename like /sdcard/foo.jpg
[03:14:41] <josephine_l>	 Uri: content://com.android.providers.media.documents/document/image%3A116
[03:15:52] <josephine_l>	 Filename: sdcard/Download/something.jpg
[03:16:16] <josephine_l>	 niedzielski, I've tried with various images, I mean
[03:20:11] <niedzielski>	 josephine_l: hm. so even for the simple case of a filename without spaces or special characters, it fails?
[03:20:53] <josephine_l>	 niedzielski, just tried one, yeah. hukafalls.jpg is the filename
[03:21:30] <Nicolas>	 josephine_l, Hi!
[03:21:34] <josephine_l>	 Hi Nicolas! :)
[03:21:51] <Nicolas>	 How are you doing? It is snowing here in Tokyo.
[03:22:12] <josephine_l>	 Nicolas, I'm good, thanks :) niedzielski and I are trying to troubleshoot the URI bug
[03:22:34] <niedzielski>	 Nicolas: hello!
[03:23:41] <Nicolas>	 The latest version is on Play Store, and works correctly on phones, but not on emulators, right?
[03:24:18] <Nicolas>	 If yes, time to ask the whole world to test the app as much as they can :-)
[03:25:01] <josephine_l>	 Nicolas, I haven't uploaded the latest to Play Store yet, haha. Should I? I've only tested it on one device...
[03:27:06] <Nicolas>	 Yes, post it, lives are not in danger and rollbacking is easy anyway
[03:28:24] <josephine_l>	 Nicolas, hahah, okay :) I'll add a try-catch block so it doesn't crash at least, and do the versioning/tagging, then I'll push the release
[03:28:34] <Nicolas>	 Users are our testers, since the app can be considered in its infancy
[03:28:39] <Nicolas>	 OK!
[03:28:41] <josephine_l>	 Should I announce on the mailing lists etc too?
[03:29:44] <niedzielski>	 josephine_l: is this only an issue with sdcard files?
[03:30:23] <Nicolas>	 Yes, a few hours after publishing, so that everyone gets it immediately without waiting for Play Store propagation.
[03:30:53] <josephine_l>	 niedzielski, I'm not sure. Let me test that, not sure if I have any imgs on my phone sdcard
[03:31:06] <josephine_l>	 Nicolas, okay, sure :)
[03:33:34] <josephine_l>	 niedzielski, it works with my phone's SDcard
[03:33:49] <josephine_l>	 I just took a picture stored on the sdcard, and it got the filepath fine
[03:33:56] <niedzielski>	 josephine_l: yeesh, there's a bit of code around handling some api differences here: https://github.com/iPaulPro/aFileChooser/blob/master/aFileChooser/src/com/ipaulpro/afilechooser/utils/FileUtils.java#L257
[03:34:51] <josephine_l>	 niedzielski, oh, whoa. That looks pretty complex, but I could give it a shot...
[03:35:08] <Nicolas>	 The score with 10k radius is very similar to the score of method D, right. Actually, just a little lower since there are a few extra false positives, but it does not seem significant, I guess another sample could produce a score a little higher just as well.
[03:35:46] <josephine_l>	 Nicolas, yeah, I think so too
[03:37:31] <Nicolas>	 I suggest stealing that code :-) Apache 2.0 license, same as us.
[03:37:44] <josephine_l>	 Nicolas, haha, okay :)
[03:40:11] <josephine_l>	 Btw, Nicolas and niedzielski , where should I be going from here? I'll try and use the URI code from that page as best I can, then push a release and advertise for testers. What next?
[03:41:29] <Nicolas>	 Next would be starting the next phase, while waiting for feedback.
[03:41:56] <niedzielski>	 josephine_l: i'm not sure what the dependencies of the file are in isolation. if it isn't independent, it might make more sense to add it as a library
[03:42:11] <josephine_l>	 Should I try to refactor the MwVolleyApi code, or leave that for later?
[03:42:29] <Nicolas>	 Once feedback comes (and it might take a week before people actually try it in the wild for real), there might be some work to do, so better start Phase 2 early.
[03:42:58] <josephine_l>	 niedzielski, I don't need all of his code though, just one method. Is it still preferable to add it as a library?
[03:43:29] <niedzielski>	 josephine_l: i wasn't sure if the method would have dependencies on other files. if it lifts cleanly, then fine. just take it and add the licensing as per usual
[03:43:33] <josephine_l>	 Nicolas, ah, okay, gotcha.
[03:44:09] <Nicolas>	 I wrote Phase 1 and Phase 2 as separate, but in reality the two will most probably overlap, with Phase 1 fixes taking some time during Phase 2.
[03:45:28] <josephine_l>	 niedzielski, oh, hm, there is one import that I can't seem to resolve
[03:45:35] <josephine_l>	 import com.ianhanniballake.localstorage.LocalStorageProvider;
[03:47:09] <josephine_l>	 Nicolas, ah, okay. Wow, just realized that phase 2 implementation officially only takes one week according to our schedule... :)
[03:50:37] <Nicolas>	 One week is rather though, so better start as soon as possible.
[03:52:31] <josephine_l>	 Nicolas, yeah, definitely. I honestly doubt I could do it in a week.
[03:53:59] <niedzielski>	 josephine_l: er, am i missing something? i thought the schedule only had two phases
[03:54:22] <josephine_l>	 niedzielski, yeah, two phases.
[03:55:13] <josephine_l>	 Hmm, seems like I will have to import everything as a library after all, due to the dependencies
[03:55:30] <niedzielski>	 josephine_l: ah, sorry :) i misunderstood the statement
[03:55:59] <josephine_l>	 niedzielski, haha, s'all good :)
[03:57:49] <josephine_l>	 niedzielski, do you think the MwVolleyApi refactor is urgent? I could see if I can get that done before starting on Phase 2. But otherwise if it isn't, I'll do that after implementing Phase 2 if there's time
[03:59:08] <niedzielski>	 josephine_l: i'm not suggesting htis is an ideal implementation but here's another URI to filepath reference: https://github.com/wordpress-mobile/WordPress-Android/blob/97672e22c0a862296c0261e96d29e2a4c55975e9/libs/utils/WordPressUtils/src/main/java/org/wordpress/android/util/ImageUtils.java#L41
[04:00:15] <josephine_l>	 Thanks niedzielski :) I think we will probably need to at least add the check for KitKat if we use this one. But always good to have more examples at hand
[04:00:15] <niedzielski>	 josephine_l: i recommend refactoring it if you're getting NPEs. otherwise, the complications and bugs will build. if you've resolved the NPE issue, then it can wait
[04:00:24] <josephine_l>	 NPE?
[04:00:41] <josephine_l>	 niedzielski, oh, you mean NullPointerException?
[04:00:54] <niedzielski>	 josephine_l: right :)
[04:00:57] <josephine_l>	 niedzielski, I don't get NPEs with the original code. I got them during the attempt to refactor
[04:01:36] <niedzielski>	 josephine_l: oh, ok :)
[04:01:36] <josephine_l>	 But I agree that all the statics are pretty awful. Annoyingly though it seems that if you don't want to use statics with Volley you're stuck with instance loaders
[04:03:03] <josephine_l>	 niedzielski, Nicolas , anything else you guys would like to discuss? I'm done with my questions early today, amazingly... :)
[04:04:04] <niedzielski>	 josephine_l: i'm good :)
[04:05:07] <Nicolas>	 OK for me!
[04:06:17] <josephine_l>	 Oh, niedzielski , sorry - I just saw your note on my caching pull request
[04:06:54] <Nicolas>	 Contact me about Phase 2 whenever you want.
[04:07:17] <josephine_l>	 Thanks Nicolas ! :)
[04:07:43] <niedzielski>	 josephine_l: that can wait if need be
[04:08:32] <josephine_l>	 niedzielski, ah, okay. I'll put it in a future patch along with MwVolleyApi then?
[04:08:40] <niedzielski>	 josephine_l: sounds good
[04:09:10] <josephine_l>	 Thanks niedzielski :)
[04:09:21] <josephine_l>	 I'll get started on that filepath library now, and I'll release after testing
[04:11:02] <niedzielski>	 josephine_l: ok great! is caching ready to be merged or do you want to add to it?
[04:11:30] <josephine_l>	 niedzielski, I'm good with the merge :)
[04:19:06] <josephine_l>	 niedzielski, I need to AFK for a bit, if there's nothing else?
[04:19:34] <niedzielski>	 josephine_l: ah, nothing on my end. have a good one!
[04:19:42] <josephine_l>	 Bye! :)
[04:19:51] <niedzielski>	 bye!
[04:26:05] <Nicolas>	 josephine_l, you have passed brion in terms of commits to the Commons app :-) https://github.com/nicolas-raoul/apps-android-commons/graphs/contributors
[05:10:18] <josephine_l>	 Nicolas, hahah, wow! I hadn't realized I'd made that many commits, time does fly :)
[05:52:36] <josephine_l>	 Hi Nicolas, you still there?
[06:01:27] <Nicolas>	 josephine_l, yes!
[06:02:21] <josephine_l>	 Nicolas, I appear to be having problems with importing the filepath library
[06:02:23] <josephine_l>	 Unsupported Modules Detected: Compilation is not supported for following modules: commons, commons-parent, ~apklib-com.actionbarsherlock_actionbarsherlock_4.4.0, ~apklib-com.viewpagerindicator_library_2.4.1. Unfortunately you can't have non-Gradle Java modules and Android-Gradle modules in one project.
[06:02:37] <josephine_l>	 Because that library has a gradle build I think
[06:05:53] <Nicolas>	 It seems to be pulling far more than we need... Can you to just replicate the logic?
[06:07:11] <josephine_l>	 Nicolas, yeah, I'll try to do it code-wise. His FileUtils class has a lot of dependencies but we can do without some of them.
[06:07:35] <josephine_l>	 Nicolas, it seems like we might need com.ianhanniballake.localstorage.LocalStorageProvider to check if there is local storage, but I guess I could copy that too...
[06:07:49] <Nicolas>	 Is the problem different from http://stackoverflow.com/questions/3401579/get-filename-and-path-from-uri-from-mediastore ?
[06:09:15] <josephine_l>	 Nicolas, I tried that code. Works in real device, not in emulator
[06:10:00] <josephine_l>	 I will probably just try and use the KitKat checking code from that GitHub repo, and skip the rest?
[06:10:39] <Nicolas>	 Did you post a new question including the emulator problem? It sounds like a problem that many people would face.
[06:11:08] <josephine_l>	 Okay, I'll post one and see
[06:11:24] <josephine_l>	 Nicolas, I should probably still release as soon as I can, right? While waiting for the answers
[06:11:59] <Nicolas>	 yes, post, release, read answers :-)
[06:13:01] <josephine_l>	 Nicolas, okay, will do :)
[06:52:41] <josephine_l>	 Nicolas, the fix requires use of DocumentsContract, which requires API lvl 19, whereas our app targets API lvl 17
[06:52:48] <josephine_l>	 Can I change the target SDK in the manifest?
[06:53:14] <Nicolas>	 What Android version is lvl 19?
[06:55:16] <josephine_l>	 Nicolas, 4.4 kitkat
[07:00:48] <Nicolas>	 10% of our users are below 4.4
[07:01:18] <Nicolas>	 But if that simplifies development greatly, go for it
[07:01:45] <josephine_l>	 Nicolas, it should still support them right? The min sdk is still 9, just that the target sdk is being increased
[07:02:52] <Nicolas>	 oh if that's it of course! target SDK should be set to the latest released SDK :-)
[07:03:55] <josephine_l>	 Nicolas, oh, okay! Didn't know that, thanks. We have been going on the old 17 all this while... 
[08:32:42] <josephine_l>	 Hi Nicolas, I think the 1.6 update has gone through. It seems to work well on my phone. :) Could you try it if possible please?
[08:32:57] <Nicolas>	 I will try asap!
[08:33:20] <josephine_l>	 Thanks! :)
[18:45:33] <mbinder>	 bearND: quick meeting :)
[18:56:42] <bearND>	 mbinder: where?
[18:57:24] <bearND>	 nm, i found it
[20:51:33] <coreyfloyd>	 dbrant: do you have the link to all the mobile EL schemas?
[20:51:40] <coreyfloyd>	 it was on office or something…
[20:52:37] <bearND>	 coreyfloyd: https://meta.wikimedia.org/w/index.php?title=Special%3APrefixIndex&prefix=MobileWikiApp&namespace=470
[20:52:46] <dbrant>	 coreyfloyd: there's this: https://meta.wikimedia.org/wiki/Special:PrefixIndex/Schema:MobileWikiApp
[20:52:56] <dbrant>	 ah, bearND beat me
[20:53:10] <coreyfloyd>	 bearND: brant awesome thanks guys!
[21:06:03] <niedzielski>	 dbrant bearND: do either of you know why FetchEditTokenTask is necessary? i don't think i've ever used SharedPreferences off the main thread. for example, we put some pretty big payloads + marshalling / unmarshalling for tabs without any AsyncTask.
[21:06:36] <niedzielski>	 dbrant bearND: ah, nevermind :) i didn't realize this was fetching from the server
[21:06:44] <dbrant>	 :)
[21:06:45] <bearND>	 k
[22:56:13] <dr0ptp4kt>	 niedzielski: argh, lost track of time. will reschedule
[22:56:26] <dr0ptp4kt>	 need to note phone notifications
[22:56:45] <niedzielski>	 dr0ptp4kt: no worries! sounds good
[23:10:57] <dr0ptp4kt>	 niedzielski: rescheduled
[23:11:30] <niedzielski>	 dr0ptp4kt: thanks!