All of the parcelers have their own type prefix. So we have to actually
use the legit methods. This is a bit annoying, as there's no fully
compatible way across all API versions, so we have to branch.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Unfortunately, we need to fix this up in both the Makefile and in CMake,
because add_link_options doesn't immediately reflect on the CMake
variables in scope.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Exception java.lang.IllegalStateException:
at androidx.fragment.app.FragmentStore.addFragment (FragmentStore.java:92)
at androidx.fragment.app.FragmentManager.addFragment (FragmentManager.java:1481)
at androidx.fragment.app.BackStackRecord.executeOps (BackStackRecord.java:387)
at androidx.fragment.app.FragmentManager.executeOps (FragmentManager.java:1965)
at androidx.fragment.app.FragmentManager.executeOpsTogether (FragmentManager.java:1873)
at androidx.fragment.app.FragmentManager.removeRedundantOperationsAndExecute (FragmentManager.java:1823)
at androidx.fragment.app.FragmentManager.execPendingActions (FragmentManager.java:1760)
at androidx.fragment.app.FragmentManager$5.run (FragmentManager.java:547)
at android.os.Handler.handleCallback (Handler.java:938)
at android.os.Handler.dispatchMessage (Handler.java:99)
at android.os.Looper.loop (Looper.java:268)
at android.app.ActivityThread.main (ActivityThread.java:8101)
at java.lang.reflect.Method.invoke
at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run (RuntimeInit.java:627)
at com.android.internal.os.ZygoteInit.main (ZygoteInit.java:997)
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Exception java.lang.IllegalStateException:
at androidx.fragment.app.Fragment.requireContext (Fragment.java:967)
at com.wireguard.android.fragment.TunnelListFragment$tunnelFileImportResultLauncher$1$1.invokeSuspend (TunnelListFragment.kt:64)
at kotlin.coroutines.jvm.internal.BaseContinuationImpl.resumeWith (ContinuationImpl.kt:33)
at kotlinx.coroutines.DispatchedTask.run (DispatchedTask.kt:104)
at android.os.Handler.handleCallback (Handler.java:761)
at android.os.Handler.dispatchMessage (Handler.java:98)
at android.os.Looper.loop (Looper.java:156)
at android.app.ActivityThread.main (ActivityThread.java:6617)
at java.lang.reflect.Method.invoke
at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run (ZygoteInit.java:942)
at com.android.internal.os.ZygoteInit.main (ZygoteInit.java:832)
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Harsh - feel free to replace this commit with something better. I'm sure
it's the most terrible way to do it.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This required some weird changes to prevent clipping on the top, because
apparently the new switch is a bit fatter.
I think this actually looks a bit uglier than before, but it seems like
that's what Material design wants. Maybe we can improve it?
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
I'm not sure why that comment (Samuel's) was there saying it was
necessary. Given it's been async for a long while, this wasn't
guaranteed anyway. So let's get rid of it and see what happens. Screen
rotation seems fine thus far.
Cc: Samuel Holland <samuel@sholland.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Otherwise, for lots of apps, the dialog shows before they're enumerated,
and the button text never gets set.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Nathan Chance dropped the ball repeatedly and never maintained this in a
consistent way that anybody could use. With Android 12 out now, just
drop it all together. A bummer, but I don't see much of a choice.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
wg-quick has supported this for a while, but not the config layer, and
not the Go backend, so wire this all up.
Requested-by: Alexis Geoffrey <alexis.geoffrey97@gmail.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This has several problems: 1) it blocks the main thread; 2) it doesn't
distinguish between a permanent error and a transient one; 3) the 10
seconds is hard coded; 4) there's no way for the user to cancel it.
We'll have to improve this.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
wgTurnOff can block for a while, in which case, calling getStatistics
will use a stale handle and stale tunnel. Not only that, but wgGetConfig
might return null, in which case string.split throws.
java.lang.NullPointerException: Attempt to invoke virtual method
'java.lang.String[] java.lang.String.split(java.lang.String)' on a null
at com.wireguard.android.backend.GoBackend.getStatistics
Reported-by: tomt@adslweb.co.uk
Link: https://lists.zx2c4.com/pipermail/wireguard/2021-May/006709.html
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
The hash was wrong, causing crashes. Android studio told me to remove it
so I did. I don't quite like that, but I don't know how Harsh came up
with the previous hash.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Modern Android likes to kill processes to free ram and resources. When
kernel-mode WireGuard is in use, this is quite alright with us, since
the app doesn't actually need to consume any resources at all in order
for the tunnel to run. So, we want to allow and encourage this resource
frugality. However, when the quick settings tile is being used or when
the app is referenced otherwise, the app will occasionally be restarted,
to, for example, repaint the quick settings tile. This is also fine, as
the process winds up being short-lived again. But, since process
initialization means asking for a new root shell in order to check on
kernel-mode WireGuard, this means that Magisk raises a systemwide toast.
On some phones, this happens each and every time that the notification
shade is pulled down. It's not only annoying but it sometimes obscures
other notifications that users want to see, prompting their pulling down
of the notification shade in the first place. In order to get rid of
this nuisance, just disable these notifications and extraneous logs, so
that we don't clutter the system every time that the process is
opportunistically killed and restarted.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
If we're horizontally scrolling, it makes sense to fill rows before
columns. But if it all fits in one page and we don't need to scroll
horizontally, it looks ridiculous. So, in this case, rearrange the tiles
so that it appears to fill columns before rows. But we don't want things
suddenly jumping around, so actually, keep the same ordering as
rows-before-columns, but add invisible spaces after certain items, so
that the fill area makes it look as though it's columns-before-rows.
This winds up being much more visually pleasing.
We do this by figuring out this kind of transformation:
If we convert this matrix:
0 3 6
1 4 _
2 5 _
To this one:
0 2 4 6
1 3 5 _
_ _ _ _
For a given index, how many spaces are under it? This changes depending
on how many total are in a grid. Going from 3x3 to 4x3, for example, we
have:
count == 12, index =
count == 11, index = 10
count == 10, index = 7,9
count == 9, index = 4,6,8
count == 8, index = 1,3,5,7
count == 7, index = 1,3,5,6!
count == 6, index = 1,3,4!,5!
count == 5, index = 1,2!,3!,4!
count == 4, index = 0!,1!,2!,3!
count == 3, index = 0!,1!,2!
count == 2, index = 0!,1!
count == 1, index = 0!
count == 0, index =
The '!' means two blanks below, no '!' means one blank below, and no
mention means no blanks below.
This commit adds code to compute such a table on the fly.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Fragment scopes get cancelled when the fragment goes away, but we don't
actually want to cancel an in-flight transition in that case. Also,
before when the fragment would cancel, there'd be an exception, and the
exception handler would call Fragment::getString, which in turn called
requireContext, which caused an exception. Work around this by using the
`activity ?: Application.get()` idiom to always have a context for
strings and toasts.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld