Status Update
Comments
da...@google.com <da...@google.com> #2
Project: platform/frameworks/support
Branch: androidx-main
Author: Louis Pullen-Freilich <
Link:
Fixes NPE when recording drawContent to a GraphicsLayer
Expand for full commit details
Fixes NPE when recording drawContent to a GraphicsLayer
In some cases (for example when software rendering the graphics layer is required on API levels where the graphics layer implementation doesn't support software rendering) drawing a graphics layer requires drawing the drawBlock passed to record, outside of the draw pass. Since we use a shared draw scope for all draw modifier nodes, this means that we capture a reference to the scope but we don't have the node information anymore to be able to draw it. This CL fixes this issue by making sure that our ContentDrawScope implementation properly restores the node state when drawing a previously recorded drawBlock.
To avoid this issue it is important to use the record function inside DrawScope for this specific use case, and avoid capturing the draw scope in a lambda otherwise.
This CL also updates the CacheDrawScope to use the record function provided by the DrawScope to avoid this issue from occuring inside cache drawing nodes.
Test: DrawModifierTest
Fixes: b/389046242
Relnote: "Fixed an occasional NPE when using `GraphicsLayer.record { this@ContentDrawScope.drawContent() }`. If you are recording drawContent() in this way, make sure to use the GraphicsLayer#record extension function inside DrawScope, and not the member function on GraphicsLayer."
Change-Id: I75fc026069666b61cdd7e5c377fa4da706f985d8
Files:
- M
compose/ui/ui/src/androidInstrumentedTest/kotlin/androidx/compose/ui/draw/DrawModifierTest.kt
- M
compose/ui/ui/src/commonMain/kotlin/androidx/compose/ui/draw/DrawModifier.kt
- M
compose/ui/ui/src/commonMain/kotlin/androidx/compose/ui/node/LayoutNodeDrawScope.kt
Hash: 8025e8c6128dd3cf148684111953fbac71b55ac2
Date: Fri Jan 10 16:53:20 2025
m....@gmail.com <m....@gmail.com> #3
Project: platform/frameworks/support
Branch: androidx-main
Author: Louis Pullen-Freilich <
Link:
Revert^2 "Fixes issue with dragAndDropSource preventing child invalidations"
Expand for full commit details
Revert^2 "Fixes issue with dragAndDropSource preventing child invalidations"
This reverts commit 3d5fd7d04bdd7f02ee2b215af68b2832e06134bf.
Reason for revert: Underlying bug b/389046242 is now resolved
Fixes: b/379682458
Test: DragAndDropSourceTest
Test: AndroidDragAndDropIntegrationTest
Change-Id: I07cb2a4091b46d862628bf17625ae9d9eb087641
Files:
- A
compose/foundation/foundation/src/androidInstrumentedTest/kotlin/androidx/compose/foundation/draganddrop/DragAndDropSourceTest.kt
- M
compose/foundation/foundation/src/androidMain/kotlin/androidx/compose/foundation/draganddrop/AndroidDragAndDropSource.android.kt
Hash: 3c48a0fbd3c2902a6a7acb6c8b474a4928b1be41
Date: Mon Jan 13 09:46:38 2025
da...@google.com <da...@google.com> #4
The following release(s) address this bug.It is possible this bug has only been partially addressed:
androidx.compose.foundation:foundation:1.8.0-beta01
androidx.compose.foundation:foundation-android:1.8.0-beta01
androidx.compose.foundation:foundation-jvmstubs:1.8.0-beta01
androidx.compose.foundation:foundation-linuxx64stubs:1.8.0-beta01
androidx.compose.ui:ui-android:1.8.0-beta01
androidx.compose.ui:ui-jvmstubs:1.8.0-beta01
androidx.compose.ui:ui-linuxx64stubs:1.8.0-beta01
m....@gmail.com <m....@gmail.com> #5
Great ! .acknowledged. Ty
da...@google.com <da...@google.com>
da...@google.com <da...@google.com>
ap...@google.com <ap...@google.com> #6
Project: platform/frameworks/support
Branch: androidx-main
Author: Daniel Santiago Rivera <
Link:
Lazily load SQLite's native library when a connection is opened.
Expand for full commit details
Lazily load SQLite's native library when a connection is opened.
Moving the load call into an inner object effectively makes it so that the library is loaded lazily when the object's class is loaded, i.e. when instantiated and accessed during the open call.
Bug: 363985585
Test: BundledSQLiteDriverTest.kt
Change-Id: I72fa01618169cb3225ea9d5d5e9f9ae59c6d4b09
Files:
- M
sqlite/integration-tests/driver-conformance-test/src/androidInstrumentedTest/kotlin/androidx/sqlite/driver/test/BundledSQLiteDriverTest.kt
- M
sqlite/sqlite-bundled/src/jvmAndroidMain/kotlin/androidx/sqlite/driver/bundled/BundledSQLiteDriver.jvmAndroid.kt
Hash: 4f69d19088dae47b6a7bcf454ea22f7e2da78e92
Date: Tue Jan 21 17:51:32 2025
da...@google.com <da...@google.com>
na...@google.com <na...@google.com> #7
The following release(s) address this bug.It is possible this bug has only been partially addressed:
androidx.sqlite:sqlite-bundled:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-android:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-iosarm64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-iossimulatorarm64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-iosx64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-jvm:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-linuxarm64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-linuxx64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-macosarm64:2.5.0-alpha13
androidx.sqlite:sqlite-bundled-macosx64:2.5.0-alpha13
Description
Following the steps outlined in the official KMP Room docs (https://developer.android.com/kotlin/multiplatform/room ) , I see 5 strict mode violations in my code.
They dont seem to cause any noticable lag but i was wondering if this violation is expected?