DynamicLayout did not include the line spacing for the newly added
lines. This CL updates StaticLayout with a parameter to force addition
of line spacing to the last line, and updates DynamicLayout to use this
parameter. With this change the last line will have the line spacing
applied, and it will provide a consistent UX.
Test: Manual/Visual test
Test: bit -t FrameworksCoreTests:android.text.StaticLayoutTest
Test: bit -t FrameworksCoreTests:android.text.StaticLayoutTextMeasuringTest
Test: bit -t FrameworksCoreTests:android.text.StaticLayoutLineBreakingTest
Test: bit -t FrameworksCoreTests:android.text.DynamicLayoutTest
Test: bit -t FrameworksCoreTests:android.text.DynamicLayoutBlocksTest
Test: bit -t CtsTextTestCases:android.text.cts.StaticLayoutTest
Test: bit -t CtsTextTestCases:android.text.cts.DynamicLayoutTest
Bug: 25194907
Change-Id: I52497fbd108538729758e842072f3ce87c6bbfdd
mLineDirections variable needs to keep a Directions object for each line
of text. Therefore at any time array of line count is required. However
currently mLineDirections is initialized and grown in the size of mLines
array, which makes the size around mColumns*line count.
This CL removes that extra allocation.
Test: Manual test to check if device boots and displays
Test: bit -t CtsTextTestCases:*
Test: bit -t FrameworksCoreTests:android.text.
Test: bit -t FrameworksCoreTests:android.widget.TextViewTest
Test: bit -t FrameworksCoreTests:android.widget.EditorCursorTest
Test: bit -t CtsWidgetTestCases:android.widget.cts.TextViewTest
Test: bit -t CtsWidgetTestCases:android.widget.cts.EditTextTest
Bug: 64038460
Bug: 62889232
Change-Id: If9bda040b98efcf16c3cdcba5bc27667775ce1fc
StaticLayout and Dynamic layout changes the values of line bottom and
descent according to linespacing modification attributes. This loses
information of the real line descent and bottom, which causes some
utilities not to be able to draw themselves in correct positions.
Examples are cursor drawing to the empty space, BulletSpan not being
able to vertically center itself in the line, and DynamicLayout not
being able to calculate the last line height properly.
This CL stores the extra added to bottom and descent values so that it
can be used later.
Test: bit CtsTextTestCases:android.text.cts.LayoutTest
Test: bit CtsTextTestCases:android.text.cts.StaticLayoutTest
Test: bit CtsTextTestCases:android.text.cts.DynamicLayoutTest
Test: bit FrameworksCoreTests:android.text.LayoutTest
Test: bit FrameworksCoreTests:android.text.StaticLayoutTest
Test: bit FrameworksCoreTests:android.text.DynamicLayoutTest
Test: bit FrameworksCoreTests:android.text.DynamicLayoutBlocksTest
Bug: 30870806
Bug: 33138492
Bug: 25194907
Change-Id: I7a1039e71e4999c75b5f26122fe6239e3ee24868
Mostly changed to avoid repetition of code and remove the assumptions
about ellipsis being one code unit. The code for multi-code unit
ellipsis is not triggered yet, but is done in preparation for
potential future locale-dependent cases.
Test: bit CtsTextTestCases:*
Test: bit CtsWidgetTestCases:android.widget.cts.TextViewTest
Test: bit CtsWidgetTestCases:android.widget.cts.EditTextTest
Test: bit CtsWidgetTestCases:android.widget.cts.CheckedTextViewTest
Test: bit CtsWidgetTestCases:android.widget.cts.AutoCompleteTextViewTest
Test: bit CtsWidgetTestCases:android.widget.cts.MultiAutoCompleteTextViewTest
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Change-Id: Id1dfdc503f87fabed2447d55ab2107eee0eccd08
Added documentation to clarify what happens in the case of null or
empty arguments.
Change-Id: I694dc0f6a6d95ab6ca7ed95b51d81938618eb75f
Fixes: 36861796
Test: bit CtsTextTestCases:*
Test: bit CtsGraphicsTestCases:android.graphics.cts.FontVariationAxisTest
When TextView or EditText contains text that is larger than the
total parcelable limit, some of the FloatingToolbar operations would
crash.
This CL changes the behavior as follows:
- Show a toast message if cut or copy operation fails because it cannot
set the primary clip.
- Trim the text for share and process_text actions
- A simple app with an EditText and a long text in it, would not open
since Autofill value was being sent over IPC. Trimmed the text that is
sent for Autofill feature.
- Trim the value send to accessibility services
Test: bit CtsWidgetTestCases:.TextViewTest
Test: bit FrameworksCoreTests:android.widget.TextViewTest
Test: bit FrameworksCoreTests:android.text.TextUtilsTest
Test: Manual sample app test
Bug: 8013261
Change-Id: Ia0df6b4eb4c13071a1bf75cedac7241c7239663c
Previously, a mix and match of the default locale and the context's
locale were used in DateFormat methods that took a Context as a
parameter. Now we consistently use the Context's locale.
Also refactored the methods is24HourFormat() and hasDesignator() for
simpler and more readable code.
Change-Id: I418c1dcb710cdf2daf14684b258b97bfabe61d8c
Fixes: 62871956
Test: bit CtsTextTestCases:android.text.format.cts.DateFormatTest
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
To be able to parcel/unparcel Typeface, keep it in static context.
Bug: 62850669
Test: Manually done
Test: bit CtsTextTestCases:android.text.style.cts.TextAppearanceSpanTest
Change-Id: I408cd33b98d8bb13776560231d1eeaac0a7c6bf8
Also clean up lower-level drawing of strike-through in Cavas.cpp.
We still cannot use the strike-through information from the font
since Skia doesn't provide it, so we are going with the default
values.
Test: Manual
Test: bit CtsTextTestCases:*
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Bug: 32907446
Change-Id: Iee6f8102a35eba0ff4127dbbef189529ab573e6d
Use ICU's MeasureFormat to the degree possible for formatting file
sizes.
Bug: 36994779
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: bit CtsTextTestCases:*
Change-Id: I4ad3b568ae7585b6ff56fddb79ded7c5b2118176
Introduce new attribute "fallbackFor" to font element.
By specifying name of the family to this attribute, that font is used
when the developer specifies the font family.
For example, if fonts.xml has the following family entry,
<family lang="ja">
<font fallbackFor="serif">NotoSerifJP-Regular.ttf</font>
<font>NotoSansJP-Regular.ttf</font>
</family>
the Japanese text is rendered by NotoSansJP-Regular.ttf by default.
Then, if developer specifies fontFamily="serif" in TextView, the Japanese
text is rendered by NotoSerifJP-Regular.ttf.
Bug: 37328609
Bug: 31491668
Test: bit FrameworksCoreTests:android.graphics.TypefaceSystemFallbackTest
Change-Id: I2744db7384c8056795e841c88b387545434131f4
Previously, localizable strings were used instead, causing various
difficulties and inconsistencies. The formatting code also
hard-coded the pluralization rules for English and used it for all
locales.
Now we use ICU's MeasureFormat. There are some small formatting
differences with the previous code, but they are not important,
especially since the old strings were not necessarily well-thought.
Bug: 36994779
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Change-Id: Ie8cb119a76e89e2ee971353761fbd54e0d9a556c
Previously, locale lists and LocaleSpans were ignored in deciding line
breaking and hyphenation in layouts.
Now, we pass both changing locale information and the whole locale
list to Minikin.
Fixes: 21571721
Bug: 25122192
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Change-Id: Ia93b0b7cbe54de65bb0ec4937e65cc7993559a6f
Previously, it was using the locale's country code, which doesn't
make sense, since the locale's country code means which dialect of
the language the system should function in. We should use the SIM
card's country code instead.
Change-Id: Ib942f92f3218c943669fc6150d6cc6b4e340b451
Fixes: 30000652
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: Manual
The new code support non-BMP characters, as well as locale-specific
uppercasing and fine-grained span copying.
The modern capitalization code in AllCapsTransformationMethod is
moved to TextUtils now, so InputFilter.AllCaps can share it.
Fixes: 37222101
Test: New CTS and core tests are added.
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Change-Id: I021ff2a97a60396fb1b6e4940d91d3cd6ccb6196
Previously, font underline position and thickness were fixed in Android.
Although a custom font may have specified a different value, such a value
would never be used. Now we use the values from the font if they are
provided and fall back to the old default values only if needed.
Bug: 62353930
Test: adb shell am instrument -w -e package android.graphics com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: Manual
Change-Id: I6bf21000dd69a2780c894b231638bc0c122e41f4
In TextLine#handleRun(), exit early and shape the text as one whole
if there were spans, but none of them were a subclass of
MetricAffectingSpan or CharacterStyle.
Change-Id: I859c896ddb9c4828bfddf200e913e98262da6ea3
Fixes: 30985144
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: adb shell am instrument -w -e package android.text com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Also, in TextLine, fix underline color and thickness to come from the
underline info rather than the TextPaint object.
Bug: 32907446
Bug: 62353930
Test: Manual
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: cts-tradefed run cts-dev --module CtsWidgetTestCases
Change-Id: I259e9cf635a7056ccd367baa77c03100c69e3b98
Previously, we had singled out UnderlineSpan for special treatment
for proper kerning at span boundaries. That had the limitation that
any other span that used underlining, or even subclasses of
UnderlineSpan were not getting the fix.
With this CL, we actually look at all the effects of applying a
span's updateDrawState() and if the changes were only in underlining,
we merge the pieces and call handleText() on the merged chunk of
text.
Change-Id: I829162c758192b80594f164d1105699224b627ba
Bug: 32907446
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: cts-tradefed run cts-dev --module CtsWidgetTestCases
Test: manual
Previously, text that was partially annotated with an UnderlineSpan
could have a different width than text without it, since underlined
text was separated into a separate piece of text and was measured
separately, causing kerning at the span boundaries to disappear.
Change-Id: I118de8524c500fb6a4b05b1bf65fe93dc67f204c
Bug: 32907446
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Test: manual
The changes in I666da5eeb39f780c341597f347bfcba21eb34295 introduced
the concept of lines that need to be always redrawn, but it broke
tests for DynamicLayout#updateBlocks() that didn't need to set up the
actual lines.
Now, during the test setup, we insert empty lines to match the number
of lines needed.
Test: adb shell am instrument -w -e class android.text.DynamicLayoutBlocksTest com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: adb shell am instrument -w -e class android.text.DynamicLayoutTest#testGetBlocksAlwaysNeedToBeRedrawn_thai com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: adb shell am instrument -w -e class android.text.DynamicLayoutTest#testGetBlocksAlwaysNeedToBeRedrawn_en com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Test: cts-tradefed run cts-dev --module CtsTextTestCases
Change-Id: I4bfc82db4beef1b740433a25a3ae29201f54d9d3
Fixes: 38029583
Text classification is now being done entirely in the app's proccess.
See android.view.textclassifier.TextClassificationManager
Reverts Ic428d00f291e268211866b3fc6b7acbc3eb04e1e
Test: bit FrameworksCoreTests:android.view.textclassifier.TextClassificationManagerTest
Fixes: 36852122
Change-Id: Iffff3ab6e7d3b761576eb6bdcd5032db34b7a1b2
This switches TextUtils.concat() to use SpannableStringBuilder
if there are any spans in the input strings. As a result, we can
avoid crashes when concatenating two CharSequences can result in
paragraph spans that now don't happen to be at paragraph boundaries.
Also document the exact behavior of the method for edge cases based
on the previous implementation.
Change-Id: I9caffbe95ed729b9c10d63a9e7d22b1e8c7185a3
Fixes: 28271770
Test: CTS tests added and old CTS tests continue to pass.
Test: cts-tradefed run cts-dev --module CtsTextTestCases --test android.text.cts.TextUtilsTest
For AssistStructure, and AutofillValue, create a copy of the text in
setText and forText methods if it is a Spanned.
Fixes: 36838999
Test: cts-tradefed run cts -m CtsAutoFillServiceTestCases
Test: cts-tradefed run cts -m CtsAssistTestCases
Test: cts-tradefed run cts --test android.widget.cts.TextViewTest -m \
CtsWidgetTestCases
Change-Id: I52e780fa9baa17c375d3945dc714171f41fd7db5
Reverts I772961bd20bff4817a060f14a843abeceb55ac92
Until we bring back TextClassifier.getLinks
See I275a9d055ef0ab68f3ca339c37ee939257c4bd54
Test: none
Bug: 22362008
Bug: 37565246
Change-Id: I2948f22cf4c3462491f47376af48624697703969
Services recycle nodes when gathering possible actions
for users to take. It's very confusing for them if they
can't recycle the node without breaking their ability
to activate a clickable span inside it.
Making ClickableSpans used for accessibility independent
of their parent nodes.
Also adjusting the value used for invalid window ids to
come from AccessibilityWindowInfo. I must have missed this
in an earlier cleanup. I needed the value for the spans,
so I figured I might as well use the correct one.
Bug: 37004527
Test: Now recycling the node in the a11y cts test.
Change-Id: I6de4e98a182dd43c4fcd0430a3c082fcc8e458c7
