Android 14 bietet tolle Funktionen und APIs für Entwickler. Im Folgenden erfahren Sie mehr über Funktionen für Ihre Apps und erhalten einen Einstieg in die zugehörigen APIs.
Eine detaillierte Liste der hinzugefügten, geänderten und entfernten APIs finden Sie im API-Diff-Bericht. Weitere Informationen zu den hinzugefügten APIs finden Sie in der Android API-Referenz. Suchen Sie bei Android 14 nach APIs, die in API-Level 34 hinzugefügt wurden. Informationen dazu, in welchen Bereichen sich Plattformänderungen auf Ihre Apps auswirken können, finden Sie in den Änderungen am Verhalten von Android 14 für Apps, die auf Android 14 ausgerichtet sind und in den Änderungen am Verhalten von Android 14 für alle Apps.
Lokalisierung
App-spezifische Spracheinstellungen
Android 14 expands on the per-app language features that were introduced in Android 13 (API level 33) with these additional capabilities:
Automatically generate an app's
localeConfig
: Starting with Android Studio Giraffe Canary 7 and AGP 8.1.0-alpha07, you can configure your app to support per-app language preferences automatically. Based on your project resources, the Android Gradle plugin generates theLocaleConfig
file and adds a reference to it in the final manifest file, so you no longer have to create or update the file manually. AGP uses the resources in theres
folders of your app modules and any library module dependencies to determine the locales to include in theLocaleConfig
file.Dynamic updates for an app's
localeConfig
: Use thesetOverrideLocaleConfig()
andgetOverrideLocaleConfig()
methods inLocaleManager
to dynamically update your app's list of supported languages in the device's system settings. Use this flexibility to customize the list of supported languages per region, run A/B experiments, or provide an updated list of locales if your app utilizes server-side pushes for localization.App language visibility for input method editors (IMEs): IMEs can utilize the
getApplicationLocales()
method to check the language of the current app and match the IME language to that language.
Grammatical Inflection API
3 Milliarden Menschen sprechen Geschlechtssprachen: Sprachen, in denen sich grammatische Kategorien wie Substantive, Verben, Adjektive und Präpositionen nach dem Geschlecht der Personen und Objekte widerspiegeln, mit denen Sie sprechen oder über die Sie sprechen. Traditionell wird in vielen geschlechtsspezifischen Sprachen das männliche grammatische Genus als Standardgenerierung oder generisches Geschlecht verwendet.
Wenn Sie Nutzer mit dem falschen grammatikalischen Geschlecht ansprechen, z. B. Frauen im männlichen grammatikalischen Geschlecht, kann sich dies negativ auf Leistung und Einstellung auswirken. Im Gegensatz dazu kann eine Benutzeroberfläche mit einer Sprache, die das grammatikalische Geschlecht des Nutzers korrekt wiedergibt, die Nutzerinteraktion verbessern und eine personalisiertere und natürlich klingende Nutzererfahrung bieten.
In Android 14 wird die Grammatical Inflection API eingeführt, um Ihnen beim Erstellen einer nutzerorientierten UI für geschlechtsspezifische Sprachen zu helfen. Damit können Sie die Genusssprache unterstützen, ohne Ihre App refaktorieren zu müssen.
Regionale Präferenzen
Regional preferences enable users to personalize temperature units, the first day of the week, and numbering systems. A European living in the United States might prefer temperature units to be in Celsius rather than Fahrenheit and for apps to treat Monday as the beginning of the week instead of the US default of Sunday.
New Android Settings menus for these preferences provide users with a
discoverable and centralized location to change app preferences. These
preferences also persist through backup and restore. Several APIs and
intents—such as
getTemperatureUnit
and
getFirstDayOfWeek
—
grant your app read access to user preferences, so your app can adjust how it
displays information. You can also register a
BroadcastReceiver
on
ACTION_LOCALE_CHANGED
to handle locale configuration changes when regional preferences change.
To find these settings, open the Settings app and navigate to System > Languages & input > Regional preferences.
Bedienungshilfen
Nicht lineare Schriftskalierung auf 200%
Starting in Android 14, the system supports font scaling up to 200%, providing low-vision users with additional accessibility options that align with Web Content Accessibility Guidelines (WCAG).
To prevent large text elements on screen from scaling too large, the system applies a nonlinear scaling curve. This scaling strategy means that large text doesn't scale at the same rate as smaller text. Nonlinear font scaling helps preserve the proportional hierarchy between elements of different sizes while mitigating issues with linear text scaling at high degrees (such as text being cut off or text that becomes harder to read due to an extremely large display sizes).
Test your app with nonlinear font scaling
If you already use scaled pixels (sp) units to define text sizing, then these additional options and scaling improvements are applied automatically to the text in your app. However, you should still perform UI testing with the maximum font size enabled (200%) to ensure that your app applies the font sizes correctly and can accommodate larger font sizes without impacting usability.
To enable 200% font size, follow these steps:
- Open the Settings app and navigate to Accessibility > Display size and text.
- For the Font size option, tap the plus (+) icon until the maximum font size setting is enabled, as shown in the image that accompanies this section.
Use scaled pixel (sp) units for text-sizes
Remember to always specify text sizes in sp units. When your app uses sp units, Android can apply the user's preferred text size and scale it appropriately.
Don't use sp units for padding or define view heights assuming implicit padding: with nonlinear font scaling sp dimensions might not be proportional, so 4sp + 20sp might not equal 24sp.
Convert scaled pixel (sp) units
Use TypedValue.applyDimension()
to convert from sp units
to pixels, and use TypedValue.deriveDimension()
to
convert pixels to sp. These methods apply the appropriate nonlinear scaling
curve automatically.
Avoid hardcoding equations using
Configuration.fontScale
or
DisplayMetrics.scaledDensity
. Because font scaling is
nonlinear, the scaledDensity
field is no longer accurate. The fontScale
field should be used for informational purposes only because fonts are no longer
scaled with a single scalar value.
Use sp units for lineHeight
Always define android:lineHeight
using sp units instead
of dp, so the line height scales along with your text. Otherwise, if your text
is sp but your lineHeight
is in dp or px, it doesn't scale and looks cramped.
TextView automatically corrects the lineHeight
so that your intended
proportions are preserved, but only if both textSize
and lineHeight
are
defined in sp units.
Kamera und Medien
Ultra HDR für Bilder
Unter Android 14 werden jetzt HDR-Bilder (High Dynamic Range) unterstützt, bei denen beim Aufnehmen eines Fotos mehr Informationen vom Sensor gespeichert werden. Dadurch sind leuchtende Farben und ein größerer Kontrast möglich. Android verwendet das Ultra HDR-Format, das vollständig abwärtskompatibel mit JPEG-Bildern ist. Dadurch können Apps nahtlos mit HDR-Bildern zusammenarbeiten und sie bei Bedarf in Standard Dynamic Range (SDR) anzeigen.
Das Rendern dieser Bilder in der HDR-Benutzeroberfläche erfolgt automatisch durch das Framework, wenn in Ihrer App die HDR-UI für das Aktivitätsfenster aktiviert wird. Dies erfolgt entweder über einen Manifesteintrag oder während der Laufzeit durch Aufruf von Window.setColorMode()
. Auf unterstützten Geräten können Sie auch komprimierte Ultra-HDR-Standbilder aufnehmen. Da der Sensor mehr Farben zurückgewinnt, lässt sich die Bearbeitung im Nachhinein flexibler. Die mit Ultra HDR-Bildern verbundenen Gainmap
können verwendet werden, um sie mit OpenGL oder Vulkan zu rendern.
Zoom, Fokus, Postview und mehr in Kameraerweiterungen
Android 14 upgrades and improves camera extensions, allowing apps to handle longer processing times, which enables improved images using compute-intensive algorithms like low-light photography on supported devices. These features give users an even more robust experience when using camera extension capabilities. Examples of these improvements include:
- Dynamic still capture processing latency estimation provides much more
accurate still capture latency estimates based on the current scene and
environment conditions. Call
CameraExtensionSession.getRealtimeStillCaptureLatency()
to get aStillCaptureLatency
object that has two latency estimation methods. ThegetCaptureLatency()
method returns the estimated latency betweenonCaptureStarted
andonCaptureProcessStarted()
, and thegetProcessingLatency()
method returns the estimated latency betweenonCaptureProcessStarted()
and the final processed frame being available. - Support for capture progress callbacks so that apps can display the current
progress of long-running, still-capture processing operations. You can check
if this feature is available with
CameraExtensionCharacteristics.isCaptureProcessProgressAvailable
, and if it is, you implement theonCaptureProcessProgressed()
callback, which has the progress (from 0 to 100) passed in as a parameter. Extension specific metadata, such as
CaptureRequest.EXTENSION_STRENGTH
for dialing in the amount of an extension effect, such as the amount of background blur withEXTENSION_BOKEH
.Postview Feature for Still Capture in camera extensions, which provides a less-processed image more quickly than the final image. If an extension has increased processing latency, a postview image could be provided as a placeholder to improve UX and switched out later for the final image. You can check if this feature is available with
CameraExtensionCharacteristics.isPostviewAvailable
. Then you can pass anOutputConfiguration
toExtensionSessionConfiguration.setPostviewOutputConfiguration
.Support for
SurfaceView
allowing for a more optimized and power-efficient preview render path.Support for tap to focus and zoom during extension usage.
Zoomen im Sensor
Wenn REQUEST_AVAILABLE_CAPABILITIES_STREAM_USE_CASE
in CameraCharacteristics
SCALER_AVAILABLE_STREAM_USE_CASES_CROPPED_RAW
enthält, kann Ihre App mithilfe erweiterter Sensorfunktionen für einen zugeschnittenen RAW-Stream dieselben Pixel wie das gesamte Sichtfeld ausgeben. Dazu wird ein CaptureRequest
mit einem RAW-Ziel verwendet, für das der Anwendungsfall für den Stream auf CameraMetadata.SCALER_AVAILABLE_STREAM_USE_CASES_CROPPED_RAW
festgelegt ist.
Durch Implementierung der Steuerelemente zum Überschreiben von Anfragen bietet die aktualisierte Kamera Nutzern eine Zoomsteuerung, noch bevor andere Kamerasteuerelemente bereit sind.
Verlustfreie USB-Audioübertragung
Android 14 gains support for lossless audio formats for audiophile-level
experiences over USB wired headsets. You can query a USB device for its
preferred mixer attributes, register a listener for changes in preferred mixer
attributes, and configure mixer attributes using the
AudioMixerAttributes
class. This class represents the
format, such as channel mask, sample rate, and behavior of the audio mixer. The
class allows for audio to be sent directly, without mixing,
volume adjustment, or processing effects.
Produktivität und Tools für Entwickler
Anmeldedaten-Manager
Android 14 adds Credential Manager as a platform API, with additional support back to Android 4.4 (API level 19) devices through a Jetpack Library using Google Play services. Credential Manager aims to make sign-in easier for users with APIs that retrieve and store credentials with user-configured credential providers. Credential Manager supports multiple sign-in methods, including username and password, passkeys, and federated sign-in solutions (such as Sign-in with Google) in a single API.
Passkeys provide many advantages. For example, passkeys are built on industry standards, can work across different operating systems and browser ecosystems, and can be used with both websites and apps.
For more information, see the Credential Manager and passkeys documentation and the blogpost about Credential Manager and passkeys.
Health Connect
Health Connect is an on-device repository for user health and fitness data. It allows users to share data between their favorite apps, with a single place to control what data they want to share with these apps.
On devices running Android versions prior to Android 14, Health Connect is available to download as an app on the Google Play store. Starting with Android 14, Health Connect is part of the platform and receives updates through Google Play system updates without requiring a separate download. With this, Health Connect can be updated frequently, and your apps can rely on Health Connect being available on devices running Android 14 or higher. Users can access Health Connect from the Settings in their device, with privacy controls integrated into the system settings.
Health Connect includes several new features in Android 14, such as exercise routes, allowing users to share a route of their workout which can be visualized on a map. A route is defined as a list of locations saved within a window of time, and your app can insert routes into exercise sessions, tying them together. To ensure that users have complete control over this sensitive data, users must allow sharing individual routes with other apps.
For more information, see the Health Connection documentation and the blogpost on What's new in Android Health.
OpenJDK 17-Updates
Android 14 continues the work of refreshing Android's core libraries to align with the features in the latest OpenJDK LTS releases, including both library updates and Java 17 language support for app and platform developers.
The following features and improvements are included:
- Updated approximately 300
java.base
classes to Java 17 support. - Text Blocks, which introduce multi-line string literals to the Java programming language.
- Pattern Matching for instanceof, which allows an object to
be treated as having a specific type in an
instanceof
without any additional variables. - Sealed classes, which allow you restrict which classes and interfaces can extend or implement them.
Thanks to Google Play system updates (Project Mainline), over 600 million devices are enabled to receive the latest Android Runtime (ART) updates that include these changes. This is part of our commitment to give apps a more consistent, secure environment across devices, and to deliver new features and capabilities to users independent of platform releases.
Java and OpenJDK are trademarks or registered trademarks of Oracle and/or its affiliates.
Verbesserungen für App-Shops
Android 14 introduces several PackageInstaller
APIs that
allow app stores to improve their user experience.
Request install approval before downloading
Installing or updating an app might require user approval.
For example, when an installer making use of the
REQUEST_INSTALL_PACKAGES
permission attempts to install a
new app. In prior Android versions, app stores can only request user approval
after APKs are written to the install session and the
session is committed.
Starting with Android 14, the requestUserPreapproval()
method lets installers request user approval before committing the install
session. This improvement lets an app store defer downloading any APKs until
after the installation has been approved by the user. Furthermore, once a user
has approved installation, the app store can download and install the app in the
background without interrupting the user.
Claim responsibility for future updates
The setRequestUpdateOwnership()
method allows an installer
to indicate to the system that it intends to be responsible for future updates
to an app it is installing. This capability enables update ownership
enforcement, meaning that only the update owner is permitted
to install automatic updates to the app. Update ownership enforcement helps to
ensure that users receive updates only from the expected app store.
Any other installer, including those making use of the
INSTALL_PACKAGES
permission, must receive explicit user
approval in order to install an update. If a user decides to proceed with an
update from another source, update ownership is lost.
Update apps at less-disruptive times
App stores typically want to avoid updating an app that is actively in use because this leads to the app's running processes being killed, which potentially interrupts what the user was doing.
Starting with Android 14, the InstallConstraints
API
gives installers a way to ensure that their app updates happen at an opportune
moment. For example, an app store can call the
commitSessionAfterInstallConstraintsAreMet()
method to
make sure that an update is only committed when the user is no longer
interacting with the app in question.
Seamlessly install optional splits
With split APKs, features of an app can be delivered in separate APK files,
rather than as a monolithic APK. Split APKs allow app stores to optimize the
delivery of different app components. For example, app stores might optimize
based on the properties of the target device. The
PackageInstaller
API has supported splits since its
introduction in API level 22.
In Android 14, the setDontKillApp()
method allows an
installer to indicate that the app's running processes shouldn't be killed when
new splits are installed. App stores can use this feature to seamlessly install
new features of an app while the user is using the app.
App-Metadaten-Bundles
Ab Android 14 kannst du mit dem Installationsprogramm für das Android-Paket App-Metadaten angeben, z. B. Praktiken zur Datensicherheit, die auf App-Store-Seiten wie Google Play enthalten sein sollen.
Erkennen, wenn Nutzer Screenshots auf dem Gerät erstellen
Für eine standardisierte Erkennung von Screenshots wird in Android 14 eine datenschutzfreundliche API zur Screenshoterkennung eingeführt. Mit dieser API können Apps Callbacks pro Aktivität registrieren. Diese Callbacks werden aufgerufen und der Nutzer wird benachrichtigt, wenn er einen Screenshot macht, während diese Aktivität sichtbar ist.
Nutzererfahrung
Benutzerdefinierte Aktionen für die Freigabeseite und verbessertes Ranking
Mit Android 14 wird das System-Sharesheet aktualisiert, um z. B. benutzerdefinierte App-Aktionen zu unterstützen für Nutzer informative Vorschauergebnisse.
Benutzerdefinierte Aktionen hinzufügen
Mit Android 14 kann Ihre App dem aufgerufenen System-Sharesheet benutzerdefinierte Aktionen hinzufügen.
Rang der Ziele für die direkte Freigabe verbessern
Unter Android 14 werden mehr Signale aus Apps verwendet, um das Ranking der Ziele für die direkte Freigabe zu bestimmen und so hilfreichere Ergebnisse für den Nutzer zu liefern. Um das aussagekräftigste Signal für das Ranking zu liefern, befolgen Sie die Richtlinien für Verbesserung des Rankings Ihrer Direct Share-Ziele Kommunikations-Apps können auch die Nutzung von Kurzbefehlen melden für ausgehende und eingehende Nachrichten.
Unterstützung für integrierte und benutzerdefinierte Animationen für die intelligente „Zurück“-Touch-Geste
Android 13 introduced the predictive back-to-home animation behind a developer option. When used in a supported app with the developer option enabled, swiping back shows an animation indicating that the back gesture exits the app back to the home screen.
Android 14 includes multiple improvements and new guidance for Predictive Back:
- You can set
android:enableOnBackInvokedCallback=true
to opt in to predictive back system animations per-Activity instead of for the entire app. - We've added new system animations to accompany the back-to-home animation from Android 13. The new system animations are cross-activity and cross-task, which you get automatically after migrating to Predictive Back.
- We've added new Material Component animations for Bottom sheets, Side sheets, and Search.
- We've created design guidance for creating custom in-app animations and transitions.
- We've added new APIs to support custom in-app transition animations:
handleOnBackStarted
,handleOnBackProgressed
,handleOnBackCancelled
in
OnBackPressedCallback
onBackStarted
,onBackProgressed
,onBackCancelled
in
OnBackAnimationCallback
- Use
overrideActivityTransition
instead ofoverridePendingTransition
for transitions that respond as the user swipes back.
With this Android 14 preview release, all features of Predictive Back remain behind a developer option. See the developer guide to migrate your app to predictive back, as well as the developer guide to creating custom in-app transitions.
App-spezifische Überschreibungen von Geräteherstellern mit großem Display
Per-app overrides enable device manufacturers to change the behavior of apps on large screen devices. For example, the FORCE_RESIZE_APP
override instructs the system to resize the app to fit display dimensions (avoiding size compatibility mode) even if resizeableActivity="false"
is set in the app manifest.
Overrides are intended to improve the user experience on large screens.
New manifest properties enable you to disable some device manufacturer overrides for your app.
Nutzer mit großem Bildschirm – App-spezifische Überschreibungen
Per-app overrides change the behavior of apps on large screen devices. For example, the OVERRIDE_MIN_ASPECT_RATIO_LARGE
device manufacturer override sets the app aspect ratio to 16:9 regardless of the app's configuration.
Android 14 QPR1 enables users to apply per‑app overrides by means of a new settings menu on large screen devices.
App-Bildschirmfreigabe
App screen sharing enables users to share an app window instead of the entire device screen during screen content recording.
With app screen sharing, the status bar, navigation bar, notifications, and other system UI elements are excluded from the shared display. Only the content of the selected app is shared.
App screen sharing improves productivity and privacy by enabling users to run multiple apps but limit content sharing to a single app.
LLM-basierte intelligente Antwort in Gboard auf dem Pixel 8 Pro
On Pixel 8 Pro devices with the December Feature Drop, developers can try out higher-quality smart replies in Gboard powered by on-device Large Language Models (LLMs) running on Google Tensor.
This feature is available as a limited preview for US English in WhatsApp, Line, and KakaoTalk. It requires using a Pixel 8 Pro device with Gboard as your keyboard.
To try it out, first enable the feature in Settings > Developer Options > AiCore Settings > Enable Aicore Persistent.
Next, open a conversation in a supported app to see LLM-powered Smart Reply in Gboard's suggestion strip in response to incoming messages.
Grafik
Pfade können abgefragt und interpoliert werden
Android's Path
API is a powerful and flexible mechanism for
creating and rendering vector graphics, with the ability to stroke or fill a
path, construct a path from line segments or quadratic or cubic curves, perform
boolean operations to get even more complex shapes, or all of these
simultaneously. One limitation is the ability to find out what is actually in a
Path object; the internals of the object are opaque to callers after creation.
To create a Path
, you call methods such as
moveTo()
, lineTo()
, and
cubicTo()
to add path segments. But there has been no way to
ask that path what the segments are, so you must retain that information at
creation time.
Starting in Android 14, you can query paths to find out what's inside of them.
First, you need to get a PathIterator
object using the
Path.getPathIterator
API:
Kotlin
val path = Path().apply { moveTo(1.0f, 1.0f) lineTo(2.0f, 2.0f) close() } val pathIterator = path.pathIterator
Java
Path path = new Path(); path.moveTo(1.0F, 1.0F); path.lineTo(2.0F, 2.0F); path.close(); PathIterator pathIterator = path.getPathIterator();
Next, you can call PathIterator
to iterate through the segments
one by one, retrieving all of the necessary data for each segment. This example
uses PathIterator.Segment
objects, which packages up the data
for you:
Kotlin
for (segment in pathIterator) { println("segment: ${segment.verb}, ${segment.points}") }
Java
while (pathIterator.hasNext()) { PathIterator.Segment segment = pathIterator.next(); Log.i(LOG_TAG, "segment: " + segment.getVerb() + ", " + segment.getPoints()); }
PathIterator
also has a non-allocating version of next()
where you can pass
in a buffer to hold the point data.
One of the important use cases of querying Path
data is interpolation. For
example, you might want to animate (or morph) between two different paths. To
further simplify that use case, Android 14 also includes the
interpolate()
method on Path
. Assuming the two paths have
the same internal structure, the interpolate()
method creates a new Path
with that interpolated result. This example returns a path whose shape is
halfway (a linear interpolation of .5) between path
and otherPath
:
Kotlin
val interpolatedResult = Path() if (path.isInterpolatable(otherPath)) { path.interpolate(otherPath, .5f, interpolatedResult) }
Java
Path interpolatedResult = new Path(); if (path.isInterpolatable(otherPath)) { path.interpolate(otherPath, 0.5F, interpolatedResult); }
The Jetpack graphics-path library enables similar APIs for earlier versions of Android as well.
Benutzerdefinierte Shader mit Vertex- und Fragment-Shadern
Android has long supported drawing triangle meshes with custom shading, but the input mesh format has been limited to a few predefined attribute combinations. Android 14 adds support for custom meshes, which can be defined as triangles or triangle strips, and can, optionally, be indexed. These meshes are specified with custom attributes, vertex strides, varying, and vertex and fragment shaders written in AGSL.
The vertex shader defines the varyings, such as position and color, while the
fragment shader can optionally define the color for the pixel, typically by
using the varyings created by the vertex shader. If color is provided by the
fragment shader, it is then blended with the current Paint
color using the blend mode selected when
drawing the mesh. Uniforms can be passed
into the fragment and vertex shaders for additional flexibility.
Hardware-Puffer-Renderer für Canvas
To assist in using Android's Canvas
API to draw with
hardware acceleration into a HardwareBuffer
, Android 14
introduces HardwareBufferRenderer
. This API is
particularly useful when your use case involves communication with the system
compositor through SurfaceControl
for low-latency
drawing.