Change styled string whitespace processing to be like AAPT's was.
Main changes:
- whitespace around tags is preserved.
- tags start exactly where they are supposed to, not off by one.
Bug: 72406283
Test: make aapt2_tests
Change-Id: I4d12728c493efd8c978e2e3d2718b56534ff52ef
This section simply gzip a large file and stores result in GZippedFileProto
This greatly improves the size, before gzip, the last kmsg size ~500KB,
after gzip the proto size is ~60KB.
Bug: 73354384
Test: atest incidentd_test and manual on device test
Change-Id: I9bfc2cf07384487671edbffb5f0bd8495608fea6
Compiled Strings (previously not encountered) in an XML resource
were using a different StringPool than the one being referred to
in the XmlFlattener, and so the indices were all wrong.
Bug: 72700446
Test: make aapt2_tests
Change-Id: I663924f8fad50fd4c69cfa196318dc63fb641a25
+ A state change atom can have one exclusive state field, and any
number of primary key fields.
When there is primary key in the atom, it means the state belongs to the primary key.
For example,
message UidProcessStateChanged {
optional int32 uid = 1 [(stateFieldOption).option = PRIMARY];
optional android.app.ProcessStateEnum state = 2 [(stateFieldOption).option = EXCLUSIVE];
}
When there is no primary key fields in the atom, the state is global.
For example,
message ScreenStateChanged {
optional android.view.DisplayStateEnum state = 1 [(stateFieldOption).option = EXCLUSIVE];
}
+ The annotation is consumed by stats_log_api_gen to generate a static map from the state
atoms to its primary fields, and exclusive fields
+ stats_log.proto is splitted into 2 proto files, because statsd needs proto lite, and c++
lite proto library cannot properly ignore the field options which requires full proto.
This CL doesn't change any logic in the statsd yet. A separate CL will use the field option
information to correctly track the state.
Test: added unit tests in stats_log_api_gen_test. and statsd_test pases.
Change-Id: I9e8a979fe81ba60efd4d854bb7087ce4b2b147ec
When ICU has a better replacement class, recommend that APIs use
that instead.
Yell if someone implements clone(), and point them towards a copy
constructor instead.
Yell if someone exposes "Impl" classes.
Test: manual
Bug: 73392214, 71906588, 62627348
Change-Id: I75dcba2c4ab7ca426057eefb0335c935c3ced79a
When compiling straight to proto from source, the file types
were not set correctly (binary XML or proto XML).
Bug: 73406447
Test: manual
Change-Id: I68c1881a00a2ec1ad58f7ec485af2543d262d076
We assumed that a raw text value set for an attribute meant there
were no compiled values set either.
This would only really happen for attributes that did not belong to any
namespace (no prefix:), since we always kept their raw string values
in case some code relied on it.
Bug: 72700446
Test: make aapt2_tests
Change-Id: Icba40a1d4b181bfe7cad73131c4dbe5ba7f8b085
Fix an issue where the compatible-screens element was not being
populated correctly. The previous version was missing the screenSize
attribute which is mandatory. The attributes were also missing the
resource ID from the framework library as these are a part of the public
API.
Change-Id: I2f594c2259831dbbd96c58db4ba55e8288d4231e
Test: unit tests
Test: manually split an APK and dumped with aapt
Currently AAPT2 does not allow permissions which last piece contains the
"-" symbol (since it is an illegal character for a java identifier).
AAPT1 would normalize the last piece, therefore creating a valid java
identifier.
This CL makes AAPT2 behave in a similar way to AAPT1, but instead of
modifying the original value of the permission string, modifies only the
java identifier part, leaving the permission string unchanged.
Fixes: 72980877
Test: updated
Change-Id: Ie44317e07407341ba3e91a84d9b06980547b3448
Parcelable shouldn't be in the list of parcelables in framework.aidl.
Remove it to fix warning when running aidl:
framework.aidl:287 attempt to redefine built in class android.os.Parcelable
Also make the dependency on sdk_parcelables not be order-only so
framework.aidl gets rebuilt when sdk_parcelables changes.
Bug: 73135791
Test: ParcelableDetectorTest
Test: out/target/common/obj/framework.aidl does not contain android.os.Parcelable
Change-Id: If5222879be9ec1e5fa08810adc624ec526ddc0ec
Currently in the link step ContainerReader is used to read files but the
reader is not checked for errors prior to the read. This can lead to
silent failure when reading invalid resource files.
Change-Id: I6420699ad30b5b84b0b2c696af8eb5a6cbd4111e
Require explicit ordering of groups in the configuration file to ensure
that the correct version code is set. Ordering based on a single ABI is
straight forward to ensure Play Store delivers the correct APK, but when
an APK needs more than one ABI things get messy quickly. This also goes
for screen density etc. The only thing that is easily sorted without
this attribute is android-sdk since an artifact can only reference a
single SDK.
Test: unit tests
Test: manually split an APK with update config.xml
Change-Id: I37a2b8b8a8409d6d6ff27c7142d4c8c8065a7a51
Otherwise during a build, Locked Region Code Injection may encounter an
ASM Analyzer exception, and the build will not fail, even though
injection won't have succeeded.
Test: make
Change-Id: Iee69f1a78fde735db62c7f73405b8cec4e410793
It is totally valid if the font is missing even if the font is listed in
fonts.xml. Android Watch has less fonts and OEM vendor may reduce the
font files.
Bug: 72159040
Test: m fontchain_lint
Change-Id: I22bdf7bfeb6d80a2a338ebf7536c8eaf193b03c5
aapt2 currently looks-up packages only by package name and then verifies
whether the package ID has the expected value. For pre-L we need to be able
to handle resource tables having packages of same package name but
different IDs.
Note that this CL fixes only proto->binary conversion but many other aapt2
commands are still affected. This is because many transformations still
consider package name as sufficient identifier of a package.
Bug: 72143207
Test: Manual
Change-Id: Id8a920d6cd15bec747d3124270f5bcb7f48924cf
When converting between binary and proto APKs in either direction, the
'aapt2 convert' command did not adjust FileReference type for entries
inside the resource table. This fix makes in-place changes of the resource
table passed to ConvertApk, which is safe since the resource table is
thrown away after the conversion.
Bug: 71854843
Test: Manual
Change-Id: Ide6e141269392db15b399db3566f18175f5133d5
