Package Details: anbox-image-gapps 2018.07.19-15

Git Clone URL: (read-only, click to copy)
Package Base: anbox-image-gapps
Description: Android image for running in Anbox, with OpenGApps and Houdini
Upstream URL:
Keywords: anbox gapps image
Licenses: custom
Conflicts: anbox-image
Provides: anbox-image
Submitter: None
Maintainer: redchenjs
Last Packager: redchenjs
Votes: 17
Popularity: 0.018537
First Submitted: 2019-02-05 09:31
Last Updated: 2020-11-26 10:56

Latest Comments

« First ‹ Previous 1 2 3 4 5 Next › Last »

csolisr commented on 2019-12-02 15:03

That reminds me - is it possible to modify this script to add Magisk to the Anbox image, and if so, how would it be integrated? Magisk (or more specifically, Magisk Hide) would allow Anbox to run apps that require SafetyNet to work.

yochananmarqos commented on 2019-10-22 18:04

@redchenjs: I see what you mean now, I've made the changes.

redchenjs commented on 2019-10-22 04:06

@yochananmarqos: All apps with armv7 or arm64 libs will not start, they need "houdini" to run on x86_64 platform, but it is not properly configured in this package.

yochananmarqos commented on 2019-10-21 18:59

@redchenjs: What's not working?

redchenjs commented on 2019-10-21 03:54

Hi @yochananmarqos, the houdini in this package is not working properly, modify the PKGBUILD as follows could solve this problem for both armv7 and arm64 libs, would you like to take a look at this?

    # load houdini_y
    mkdir -p houdini_y
    rm -rf ./houdini_y/*
    unsquashfs -f -d ./houdini_y ./houdini_y.sfs

    mkdir -p ./squashfs-root/system/lib/arm
    cp -r ./houdini_y/* ./squashfs-root/system/lib/arm
    mv ./squashfs-root/system/lib/arm/ ./squashfs-root/system/lib/

    # load houdini_z
    mkdir -p houdini_z
    rm -rf ./houdini_z/*
    unsquashfs -f -d ./houdini_z ./houdini_z.sfs

    mkdir -p ./squashfs-root/system/lib64/arm64
    cp -r ./houdini_z/* ./squashfs-root/system/lib64/arm64
    mv ./squashfs-root/system/lib64/arm64/ ./squashfs-root/system/lib64/

    # add houdini parser
    mkdir -p ./squashfs-root/system/etc/binfmt_misc
    echo ':arm_exe:M::\x7f\x45\x4c\x46\x01\x01\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x02\x00\x28::/system/lib/arm/houdini:P' >> ./squashfs-root/system/etc/binfmt_misc/arm_exe
    echo ':arm_dyn:M::\x7f\x45\x4c\x46\x01\x01\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x03\x00\x28::/system/lib/arm/houdini:P' >> ./squashfs-root/system/etc/binfmt_misc/arm_dyn
    echo ':arm64_exe:M::\x7f\x45\x4c\x46\x02\x01\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x02\x00\xb7::/system/lib64/arm64/houdini64:P' >> ./squashfs-root/system/etc/binfmt_misc/arm64_exe
    echo ':arm64_dyn:M::\x7f\x45\x4c\x46\x02\x01\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x03\x00\xb7::/system/lib64/arm64/houdini64:P' >> ./squashfs-root/system/etc/binfmt_misc/arm64_dyn

    # set processors
    sed -i '/^ro.product.cpu.abilist=x86_64,x86/ s/$/,arm64-v8a,armeabi-v7a,armeabi/' ./squashfs-root/system/build.prop
    sed -i '/^ro.product.cpu.abilist32=x86/ s/$/,armeabi-v7a,armeabi/' ./squashfs-root/system/build.prop
    sed -i '/^ro.product.cpu.abilist64=x86_64/ s/$/,arm64-v8a/' ./squashfs-root/system/build.prop

    # enable nativebridge
    echo 'persist.sys.nativebridge=1' >> ./squashfs-root/system/build.prop
    sed -i 's/ro.dalvik.vm.native.bridge=0/' ./squashfs-root/default.prop

Anonymous comment on 2019-09-25 23:09

@yochananmarqos: just add exit 1 between line 45 and line 46 should let makepkg stop while md5sums failed.

yochananmarqos commented on 2019-09-25 17:42

@Robosky: That command on it's own does not print anything whether they match or not, so I added an if then else statement. I'm not sure how to stop if they fail, though.

Anonymous comment on 2019-09-25 16:46

How about add a prepare() before build() like:

prepare() {
    [ `md5sum "${_OPENGAPPS_FILE}" | awk '{print $1}'` = `cat "${_OPENGAPPS_FILE}".md5 | awk '{print $1}'` ] || exit 1

and change the md5sum() of the zip and the .md5 to 'SKIP' so that you wouldn't needed to modify the PKGBUILD for md5sum? (Sorry for my poor English

yochananmarqos commented on 2019-09-25 16:41

@high-voltage: Good point. I switched to md5sums and included the OpenGApps md5 file.

high-voltage commented on 2019-09-23 12:13

Why to check for SHA256sum can't we check the md5 taken from same repo provided in sourceforge ?