Chris Renshaw

Results 239 comments of Chris Renshaw

I'm seeing this too. Even weirder, /data/data/com.android.systemui exists, so something isn't working quite right.

Doesn't look too early in the log, it's at late_start and boot_competed even. As for user_de, it doesn't exist in all Zygisk-supported Android versions, I think. But anyway, there must...

We don't do anything with the config file, no, we just check the boot partition ramdisk for backup/.magisk and extract it to repatch kernel binary/dtb if needed... I guess if...

Ah, I see, /dev/random can kick out some crazy unprintable characters, so really shouldn't ever be in a file like that. I just ran `echo "$(dd if=/dev/random bs=8 count=1 2>/dev/null)"`...

Maybe would be better just to send it `> randomseed` so it's contained in its own file (with however many lines of characters /dev/random spits out) and doesn't affect config...

I don't pull specific vars, I just export the entire contents, less upkeep for Magisk changes that way, and I think the reason why the config/.magisk file was formatted this...

I can't just start blanket skipping 10 bytes since AK3 can't know what version of Magisk the boot partition was patched by in advance.

@hanspark74 Can you test again on latest builds and report back?

Please go ahead and make a Pull Request and it will be considered.

Reopen since https://github.com/opengapps/opengapps/issues/48#issuecomment-138077508 is still a solid idea :+1: