infinite recursion was due to autoPatchelfHook being in buildInputs of
platform-tools, i will add a lint for it in nix-community/nixpkgs-lint.
```
$ nix build ".#pkgsCross.aarch64-android-prebuilt.hello" --show-trace 2>&1 | rg 'while evaluating the attr.+deriv'
… while evaluating the attribute 'stdenv' of the derivation 'zlib-aarch64-unknown-linux-android-1.2.13'
… while evaluating the attribute 'CPPFLAGS' of the derivation 'python3-aarch64-unknown-linux-android-3.10.8'
… while evaluating the attribute 'setuptools' of the derivation 'python-catch-conflicts-hook'
… while evaluating the attribute 'nativeBuildInputs' of the derivation 'python3.10-pyelftools-0.28'
… while evaluating the attribute 'passAsFile' of the derivation 'python3-3.10.8-env'
… while evaluating the attribute 'pythonInterpreter' of the derivation 'auto-patchelf-hook'
… while evaluating the attribute 'buildInputs' of the derivation 'platform-tools-33.0.2'
… while evaluating the attribute 'installPhase' of the derivation 'ndk-24.0.8215888'
… while evaluating the attribute 'installPhase' of the derivation 'aarch64-unknown-linux-android-ndk-toolchain-24.0.8215888'
… while evaluating the attribute 'bintools_bin' of the derivation 'aarch64-unknown-linux-android-ndk-toolchain-wrapper-24.0.8215888'
… while evaluating the attribute 'bintools' of the derivation 'aarch64-unknown-linux-android-ndk-toolchain-wrapper-24.0.8215888'
… while evaluating the attribute 'defaultNativeBuildInputs' of the derivation 'stdenv-linux'
… while evaluating the attribute 'stdenv' of the derivation 'hello-aarch64-unknown-linux-android-2.12.1'
```
stdenv -> stdenv.cc -> bintools -> android-ndk-toolchain -> ndk -> platform-tools -> auto-patchelf-hook -> python3 -> zlib -> stdenv -> stdenv.cc -> ...
autoPatchelfHook was in buildInputs of platform-tools so we needed the host tools to build
it but platform-tools was a required tool
Note: I DO NOT resign from nixpkgs, not at all!
However, I like a clean notification inbox and I get a lot of stuff for
packages where I'm only an end-user or don't use them anymore and thus
can't help out that much.
So please consider it a measure to reduce the mental load for me when
going through my notifications ;-)
Workaround build failure on -fno-common toolchains like upstream
gcc-10. Otherwise build fails as:
ld: src/host/usb-linux.c:82: multiple definition of `t_recovery_queue';
src/host/recovery.c:45: first defined here
Workaround build failure on -fno-common toolchains like upstream
gcc-10. Otherwise build fails as:
ld: ../ipsw-patch/libxpwn.a(libxpwn.c.o):(.bss+0x4): multiple definition of
`endianness'; CMakeFiles/xpwn-bin.dir/src/xpwn.cpp.o:(.bss+0x0): first defined here
This is supposed to fix an issue caused by this PR:
https://github.com/NixOS/nixpkgs/pull/163924
Which made `autoPatchelfHook` available only on Linux, resulting in
builds of Android packages failing with:
```
error: Package ‘auto-patchelf-hook’ in /nix/store/...-nixpkgs-source/pkgs/build-support/trivial-builders.nix:73
is not supported on ‘x86_64-darwin’, refusing to evaluate.
```
Signed-off-by: Jakub Sokołowski <jakub@status.im>
Currently trying to run Genymotion on Plasma 5 fails at all, Genymotion
itself complaining about libqtquickcontrols2materialstyleplugin.so using
"incompatible Qt library".
As it turns out, this package ships its own
version of Qt but does not ignore any environment variables related to
Qt, which results in Genymotion's Qt using (apparently incompatible)
QML plugins from user's system. This can be fixed quite easily by
unsetting `QML2_IMPORT_PATH` in a wrapper, which this patch does.
There might be more such problems, but I haven't encountered them yet,
so fixing those will be up to someone else ;)
The last repo.json update in a0f6a8af81 removed the default emulator version, so it had to be changed (or the repo.json had to be overwritten) for it to work.
Instead use the most recent available emulator version
Currently there are no `aarch64-darwin` builds of Android SDK available.
For this reason attempts to build `gomobile` on that platform fail with:
```
No Android SDK tarballs are available for system architecture: aarch64-darwin
```
Signed-off-by: Jakub Sokołowski <jakub@status.im>
I noticed this minor grammar mistake when running update.nix, and then
while grepping to find the source I noticed we had it a few times in
Nixpkgs. Just as easy to fix treewide as it was to fix the one
occurrence I noticed.
The distinction between the inputs doesn't really make sense in the
mkShell context. Technically speaking, we should be using the
nativeBuildInputs most of the time.
So in order to make this function more beginner-friendly, add "packages"
as an attribute, that maps to nativeBuildInputs.
This commit also updates all the uses in nixpkgs.
Android is deprecating ndk.dir in favor of specifying exact NDK
version in Gradle configuration. Ensure that we can support multiple
NDKs, and link them into the location the Android Gradle Plugin expects.
The script make_standalone_toolchain.py autogenerates some shell scripts
when invoked. These scripts have a hardcoded shebang of '#!/bin/bash'.
The generated scripts obviously do not work in a Nix environment. This
commit makes sure the scripts have the correct shebang.
continuation of #109595
pkgconfig was aliased in 2018, however, it remained in
all-packages.nix due to its wide usage. This cleans
up the remaining references to pkgs.pkgsconfig and
moves the entry to aliases.nix.
python3Packages.pkgconfig remained unchanged because
it's the canonical name of the upstream package
on pypi.
androidenv did not previously write license files, which caused certain
gradle-based Android tools to fail. Restructure androidenv's list of
Android packages into a single repo.json file to prevent duplication
and enable us to extract the EULA texts, which we then hash with
builtins.hashString to produce the license files that Android gradle
tools look for.
Remove includeDocs and lldbVersions, as these have been removed
from the Android package repositories.
Improve documentation and examples.
We just need jdk for this script that helps set up toolchains. The
script might need to know about the target platform, but the interpreter
that runs it (or part of it) doesn't.
Simple tool to synchronize a directory-tree between a local machine and
an Android device using `adb` and `rsync`.
Since this repo doesn't have any releases atm, I added the `-unstable`
suffix to `pname`. Since 2016-08-31 the following things changed:
7fc48ad1e1...fb7c549753
Also applied the following changes to the derivation:
* Removed the `phases` hack to enable important things like `fixupPhase`
again (amongst other things this is needed for shebang-patching and
reference-checks).
* Fixed the wrapper to make sure that every binary needed by those
scripts is available.
* Added myself as additional maintainer.
