« first day (1953 days earlier)      last day (2970 days later) » 

6:14 AM
All the best @Izzy and @Firelord
Permit me to brag :)
Got my bronze charging tag today. What makes it special for me is that I am the first to bag it (though it took more than double of expected questions).
Surprising I guess since charging is something that everybody had to do
 
@beeshyams Well done!
 
Thanks @MatthewRead
To you and also Izzy to spare some low hanging fruits for lesser mortals :)
 
Haha, we have hogged the rep a little I suppose
 
Early mover advantage, more like it
Knowledge and skill being the main thing
Aside, in line with discussion yesterday, charging had nearly the same number of questions compared to whatsapp :)
Thanks to all you guys who help me fit in @MatthewRead: You may possibly recollect, "deletion of posts", incident which you handled very well, to my advantage ..Thanks
 
6:35 AM
No problem :)
 
7:30 AM
@beeshyams Congrats!
@beeshyams LOL
@MatthewRead Just a little, maybe :)
 
7:54 AM
Seems like iRoot spammer...
 
Yeah I find it hard to tell if he's spamming or just enthused about the tool.
 
8:09 AM
@Izzy Proof or it didn't happen. :P
 
Thanks @izzy
 
@Firelord whistleblower protection, sorry :)
 
@Izzy not in this jurisdiction. :P
 
@Firelord permission denied :)
I am not rooted, so you've got no access #D
(always get a laugh when one cannot distinguish between a "rooted device" and a "rooted self" ;)
(and feel tempted to link Urban dictionary on the term "rooted" when used for people … #D)
 
@beeshyams nicely done. :)
 
8:15 AM
Thanks @Firelord..though honestly yesterday was a "charged day" (pun intended)..was surprised at number of upvotes...not complaining
Now I feel like I am making a BIG thing out of getting the tag
 
No worries, I remember being excited at getting my first tag badges
 
So do I :)
 
8:44 AM
@beeshyams Congrats for your tag badge! :)
 
0
Q: How can I take on-device screenshots of "secure" apps?

Matthew ReadAs per this Stack Overflow answer, Android apps using FLAG_SECURE in their activities are protected from screenshots being taken by the system. However, this is not a hard guarantee. In Android Studio, I can still take screenshots of such "secure" apps with the button pointed out in this SO ans...

 
@AndrewT. Thanks
 
 
1 hour later…
10:04 AM
@MatthewRead there you go (my Google-Fu is strong today :) DisableFlagSecure module for Xposed.
 
@MatthewRead I'm surprised that taking the screenshot from the debugger works.
They changed the way it all works since I was working on the Android graphics subsystem, but my understanding is that secure windows are rendered into a separate framebuffer, which is composited by the display controller hardware.
the "why are you taking my money" guy is back
I feel like the best thing to do is edit his question to say "how do I cancel a subscription in google play" and answer with a link to the support site. What do you think?
 
 
3 hours later…
1:19 PM
(I went ahead and did it anyway)
 
2:00 PM
Which was sensible IMO..+1
Is use of "spoiler" not encouraged here? Or is it disabled ? Method here didn't work.
I have seen it being used on meta
For instance readability of this answer Would be enhanced of I used a spoiler for background and Additional information
 
@beeshyams I don't think that's true, because you'd just get huge empty boxes instead of the text.
I hate on web fora where they use the spoiler feature just to hide large text blocks, because you have no idea whether the text is actually going to have a spoiler or not
 
@DanHulme: Visually unappealing maybe distracting also, but not everybody would be interested in the sections mentioned, those interested can click to read
Otherwise the huge length devalues the key element of answer and I don't favor omitting those sections for the completeness they provide
 
@beeshyams the SE spoiler thing is mouse-over, not click-to-read, so it's awful for following along text while doing something in another window
 
Is there an alternative?
Taking my answer linked as an example?
 
2:16 PM
I think if you used actual headings in your answer instead of just bold text, it would be a lot easier to see the structure
 
I don't think I follow you. Can you please elaborate?
 
look at the headings in my answer here to see
28
A: Google Backup: Multiple devices using the same account - what happens on Restore?

Dan HulmeLet's talk about sets, baby Android's backup service has a concept called a set: the set of all data backed-up from one device (on one transport, but that's a detail). Each set is identified by a unique string, such as the IMEI on the device. When an app (or the list of installed apps) is backed...

it's not just bold text, it's actually larger and better spaced
 
Thanks..Will have a look and edit later when I have a good connection. Now on a train with patchy connection
Yes definitely more readable
 
 
2 hours later…
4:27 PM
@Izzy Greenify has support for Tasker? Didn't know.
Raw concept (untested): Use tasker and Greenify. Profile something like "When app X is running › Greenify Playstore". When greenified, it won't do updates. — Izzy 18 hours ago
 
4:44 PM
@Firelord Sure. Google-Fu turned up this concept at Reddit :) Maybe you need the paid version of Greenify for that.
 
5:03 PM
@Firelord BTW: My lists closing up to 13k apps (12,963 currently). Still in the middle of my run :)
 
@Izzy Cool. Add some stuff searchable from baidu. It will shoot to 26k easily. :P
@Izzy any app you're aware of which I can use to test your answer?
1
A: How can I take on-device screenshots of "secure" apps?

IzzyAs your device is rooted, you could use the Xposed module DisableFlagSecure: An Xposed Framework module that disables FLAG_SECURE on windows system-wide. This lets you take screenshots and do screen capture in apps that normally won't let you. (That said, you probably don't want to leave this...

 
how appropriate!
 
6:15 PM
@Izzy Excellent suggestion, trying it out now
 
6:37 PM
@Firelord play.google.com/store/apps/details?id=com.radicalapps.cyberdust is one, apparently some banking apps have this protection as well
 
Nice. I'll try.
 
7:00 PM
@Firelord Huh? Guess not, I'm not after the Chinese market. Too dangerous :)
@Firelord Ask @MatthewRead – he had the issue, so he probably will have an app for that.
@Firelord LOL OP asked for downgrade but doesn't like the flag?
Oh, Matthew already mentioned one. Sorry, had to catch up first :)
 
 
2 hours later…
9:05 PM
Just in case anyone here has an idea: Strange permissions in several apps (I've posted that on SO as it better fits where people have the "dev background" ;)
And I can give numbers, if anyone is interested ;)
 
@Izzy Interesting copy-pasta. :D
@Izzy Do you know what happens during installation when a permission declared in manifest isn't found in OS?
 
@Firelord Yeah. More of that: permRead for example. And plenty of "guessed permissions" I assume.
@Firelord Yupp. Check logcat: "Not granting unknown permission com.bullshit.I_AM_STUPID" :)
Just google a few of that list after prefixing them with com.android., and you will find plenty of those logcat entries.
Btw: According to my (growing) stats, about 1/3 of all apps are somehow glued to the Google ecosystem. More than 10% are "hard glued" (i.e. either won't install at all or simply not run unless you've got GApps on the device). Shocking.
 
9:25 PM
@Izzy Shocking? nope! I'm actually speculating (pun) that by the time Android P or Q would be launched, almost everything in AOSP would have been moved to proprietary Play Services in the name of solving the problem of keeping OS of every device in the world updated. The only thing vendors would have to do for qualifying their device for Play Services would be to test stability of wallpapers in all orientations. I guess that ~10% shows the results for early experiments of this plan. :P
---
Okay. This is strange to me. I'm trying `fastboot oem lock` (yes, lock, not unlock) and it outputs:
> Device state transition will erase userdata.
> Are you sure you want to continue this transition?

I don't understand this. Why would locking the bootloader should wipe data?
 
Over-zealous developers?
 
9:45 PM
@Firelord Well, those "soft glues" don't bother me too much: I don't care e.g. whether an app can access my Gmail data, as I don't have Gmail installed. And OK, I can't use in-app payment via vending.BILLING. Other parts can be worked around, and so I e.g. can use GCM (thanks to microG) – and even was able to work around the "hard glue" of MAPS_RECEIVE via the maps API of the microG predecessor NOGAPPS. What really bothers me is vending.CHECK_LICENSE: no way around that except for LuckyP
 
@Izzy Does that LuckyP has any legit usage? I'm gonna try it someday though, out of curiosity.
 
@Firelord "primarily opinion-based" – but I say so, yes. If I bought an app, I've a right to use it, right? So what if, for privacy reasons, I don't want GApps on my device – but want to use "that app"?
Leave temptation out of the equation ;)
 
Yeah, the licence checking Amazon does for example is annoying as heck. I had an app called Gentle Alarm that turned out to not be so gentle -- it tried to do a licence check and failed for some reason, so I couldn't open the app or shut off the alarm but the service kept running. Had to do a battery pull
 
10:14 PM
@MatthewRead It appears that flipping the last byte of that partition can enable and disable FRP since it changes OEM Unlock bit. Well it doesn't hurt either if you zeroed the whole partition but have a backup of it, just to remain at safe side. // I spent hours in analysis only to find in the end that it is already covered here: forum.xda-developers.com/nexus-6/help/…
 
@Firelord Are you saying that unlocking the bootloader automatically disables FRP? Or just that you can do the latter after the former?
 
10:29 PM
That link begins with information which conflicts with my observations. I was able to unlock the bootloader without enabling "OEM unlock" from developer settings. I was able to do flashing from fastboot and flashed stock images multiple time. At one time, I tried to check about FRP and noticed that after a reset, Android required me to verify the ownership at *supposed* first boot. --- All of this info is in consonance with the Android Police article linked in your answer.

Now, the XDA article I linked says that OEM unlock must be enabled to have the bootloader unlock but that's contradict
 
Ah, OK. I think when FRP was added to the L dev branch, it wasn't disabled when you did that
And I wouldn't find it hard to believe that an OEM failed to make the bootloader check that OEM Unlock was enabled. They mess this stuff up all the time
 
@MatthewRead Since the production is automated (or I think so), I think a whole particular badge of devices might have this particular issue.
 
@Firelord Oh yes, definitely. I misread the mention of "particular device"
 
Oops. Typo. It's batch not badge.
 
11:27 PM
@Izzy Android source code and official reference says that their permissions are prefixed with android.permission, not com.android. Could you check once again?
 
11:46 PM
@AndrewT. Yupp, right – my fault. Corrected.
The example links I gave also show that clearly – e.g. "android.permission.READ_INTERNAL_STORAGE, android.permission.WRITE_INTERNAL_STORAGE" for RaspManager.
android.permission.SYSTEM_OVERLAY_WINDOW for GPS HUD, same thing.
Also, if those apps had declared them, that means their Manifests must have them as <permission …, not just <uses-permission …, right?
Not that I'd check some thousands APKs for it, but I could check a few examples.
GPS HUD: only uses-permission (not a single one declared on its own)
(using aapt d badging on the APKs)
Same for MyBackup.
 

« first day (1953 days earlier)      last day (2970 days later) »