This reverts commit c90f42c80f.
Reason for revert: Introduces the escaping *into the APKs*; as mentioned in b/140373430#comment3 the Right Way to dump XML is via textproto.
Fix: 222745626
Bug: 140373430
Change-Id: I775b8a304e76a83934c36886c17cf23560914d21
Merged-In: I775b8a304e76a83934c36886c17cf23560914d21
(cherry picked from commit 27e9a4a85f)
Make it executable, make it work with python3.
Bug: 228451704
Test: $ m api_txt dist
$ PREVIOUS_API=prebuilts/sdk/##
$ for api in $(ls -1 out/dist/apistubs/android); do
frameworks/base/tools/apilint/deprecated_at_birth.py \
out/dist/apistubs/android/$api/api/ \
${PREVIOUS_API}/$api/api/
done
Change-Id: Ib0eda541be68b0b21ae0dea3f51a82b44f7f9e7c
Would have been very useful to avoid http://b/227682097
Bug: 227682097
Test: m apexer apexer_test deapexer debugfs_static; atest --host apexer_test
Change-Id: I24234d053a06868493831d40f94f7e46956391a1
Soong's cc support for protos is inconsistent, it adds a dependency on
lite proto library is proto.type is unspecified, but builds as full.
Update aapt2 to explicitly build full protos rather than rely on
Soong's inconsistent defaulting of proto.type
Test: CI
Change-Id: I5600d04c7ac32a04fd2b230912e02280a3874812
Bug: 226676748
Test: Confirmed that the updated fontchain lint fails on the COLR font w/o PUA and passes once added
Change-Id: If831ed689ce80f26564279c6a6243cedddc56c36
For convenience, builds against musl libc currently use the
linux_glibc properties because they are almost always linux-specific
and not glibc-specific. In preparation for removing this hack,
tweak the linux_glibc properties by either moving them to host_linux,
which will apply to linux_glibc, linux_musl and linux_bionic, or
by setting appropriate musl or linux_musl properties. Properties
that must not be repeated while musl uses linux_musl and also still
uses the linux_glibc properties are moved to glibc properties, which
don't apply to musl. Whether these stay as glibc properties or get
moved back to linux_glibc later once the musl hack is removed is TBD.
Bug: 223257095
Test: m checkbuild
Test: m USE_HOST_MUSL=true host-native
Change-Id: I7058c8f1dadd7bbfd7e169bdf0a0441eb6d10ec5
This tool rewrites the bytecode in the designated JAR files to produce
tracing calls on enter and exit, while making sure to close the tracing
span even on exceptions being thrown.
The idea is mostly to reduce the amount of noise within methods when
just trying to add some tracing.
Test: atest --host TraceInjectionTests
Change-Id: If6acb72f34cbb83d9b041a62ee3d8c2abf74b69e
Merged-In: If6acb72f34cbb83d9b041a62ee3d8c2abf74b69e
Includes instructions on how to expose new public resource IDs
and how to "remove" unused resource IDs before finalization.
Temporarily restricts OWNERS to ensure T remains stable until
a long term static analysis solution is found.
Fixes the script for finalization to handle split files and
seed the next release's staging-public-group blocks.
Bug: 222537368
Test: manual, checked build before and after mock finalization
Change-Id: I901a80e487f5eaab9689f469196648535857e0e1
Ignores and increments the ID if the entry contains removed_, to prevent
shifting IDs by leaving a hole, allowing staging resources to be removed
before finalization.
This was broken out from the parent change to allow other CLs fixing the
resource IDs to be merged without splitting public.xml.
Bug: 222537368
Test: manual, see bug for other changes
Change-Id: I56f4fd70824bd42ec62f4fa0a2f067e2a3c094b5
JDK 17 no longer allows access to com.sun.tools.javac.code by default,
use the public APIs instead.
Bug: 219098645
Test: m checkbuild
Change-Id: If379935cc4ab9e5c809cc9fbfccbacfd8d87e8c8
It was available for unbundled builds when the build rules were in mk form and aapt2 is enabled for unbundled, so there's not a strong reason why aapt should be disabled for unbundled build
Change-Id: I6589eafbe8d2e423c9ac925b3614763fa6a511ac
Fixes: 217747798
Test: local compile
It seems the behavior of strdup is undefined when NULL is passed in.
Clang's static analyzer complains about this:
> frameworks/base/tools/bit/command.cpp:195:25: warning: Null pointer
passed to 1st parameter expecting 'nonnull'
[clang-analyzer-core.NonNullParamChecker]
Since we don't seem to care about malloc failures (which is a good thing
IMO), adding a second check for the successful completion of `strdup`
seems pointless.
Bug: 206470603
Test: TreeHugger
Change-Id: Ib621659e6fb600203694974f02d96ba0acb3362f
This type will be used as part of ResourceName to support types with
custom names inside aapt2.
DD: go/custom-resource-types-in-aapt2
Bug: b/215108200
Test: Resource_test.cpp
Change-Id: Ib09776e6b2f69fefb9724136a345dc25bea42573
