The CL fixes a crash in Layout.primaryIsTrailingPreviousAllLineOffsets.
The crash was happening when the method was called for a line beginning
with an empty bidi run. This could happen, for example, for empty text -
I was unable to find any other case. The CL improves the existing test
for the method with this case, which was previously crashing.
The CL also fixes a potential crash in getLineHorizontals. However, this
bug could never happen as in the current code path clamped is always
false (and kept as parameter for parity with getHorizontal).
Bug: 135444178
Bug: 78464361
Test: atest FrameworksCoreTests:android.text.LayoutTest\#testPrimaryIsTrailingPrevious
Change-Id: I47157abe1d74675884734e3810628a566e40c1b4
(cherry picked from commit 7ad499d007)
Also don't show smart actions for selections in text with unsupported
characters.
Bug: 116321860
Test: runtest -x cts/tests/tests/text/src/android/text/util/cts/LinkifyTest.java
Change-Id: Id271cab8aef6b9b13ef17f1a8654c7616f75cf13
The crash was introduced by Ib66ef392c19c937718e7101f6d48fac3abe51ad0
The root cause of the crashing is requesting out-of-line access for the
horizontal width. This invalid access is silently ignored by
TextLine#measure() method but new implementation end up with out of
bounds access.
To makes behavior as old implementation, calling getHorizontal instead
of accessing measured result array.
Bug: 78464361, 111580019
Test: Manually done
Change-Id: I5c5778718f6b397adbb1e4f2cf95e9f635f6e5c8
(cherry picked from commit 960647d582)
Merged-In: I5c5778718f6b397adbb1e4f2cf95e9f635f6e5c8
Layout#getOffsetForHorizontal was running in O(n^2) time, where n is the
length of the current line. The method is used when a touch event
happens on a text line, to compute the cursor offset (and the character)
where it happened. Although this is not an issue in common usecases,
where the number of characters on a line is relatively small, this can
be very inefficient as a consequence of Unicode containing 0-width
(invisible) characters. Specifically, there are characters defining the
text direction (LTR or RTL), which cause our algorithm to touch the
worst case quadratic runtime. For example, a person is able to send a
message containing a few visible characters, and also a lot of these
direction changing invisible ones. When the receiver touches the message
(causing the Layout#getOffsetForHorizontal method to be called), the
receiver's application would become not responsive.
This CL optimizes the method to run in O(n) worst case. This is achieved
by computing the measurements of all line prefixes at first, which can
be done in a single pass. Then, all the prefix measurement queries will
be answered in O(1), rather than O(n) as it was happening before.
Bug: 79215201
Test: manual testing
Change-Id: Ib66ef392c19c937718e7101f6d48fac3abe51ad0
Merged-In: Ib66ef392c19c937718e7101f6d48fac3abe51ad0
Add cached int buffers for sorting.
Test: All CtsTextTestCases
Merged-In: Ice0b3c3fffc541b26aca94c93fd01e30e13efe2e
Bug: 33609774
Change-Id: Ib728654a384cfc072d1c190611e4553aa8957fa8
The grow logic in SpannableStringInternal#setSpan assumes that the
size of mSpanData is consistent with that of mSpans, in particular
that if the latter doesn't need to grow, neither does the former.
The copySpans() method didn't enforce this, creating an mSpanData
array only big enough to hold the data.
This patch documents the invariant in a comment and enforces it.
Bug: 30359314
Change-Id: Ie25db70a76836e97af8476a7f5c10cb4b976c1cf
(cherry picked from commit 83549088c6)
The grow logic in SpannableStringInternal#setSpan assumes that the
size of mSpanData is consistent with that of mSpans, in particular
that if the latter doesn't need to grow, neither does the former.
The copySpans() method didn't enforce this, creating an mSpanData
array only big enough to hold the data.
This patch documents the invariant in a comment and enforces it.
Bug: 30359314
Change-Id: Ie25db70a76836e97af8476a7f5c10cb4b976c1cf
The gender-balanced emojis are made with ZWJ sequence and emoji
modifiers. For example, U+1F469 U+1F3FD U+200D U+1F4BC should be
deleted at the same time by single backsapce key event. Here, U+1F469
is WOMAN, U+1F3FD is EMOJI MODIFIER FITZPATRICK TYPE-4, U+200D is ZERO
WIDTH JOINER, U+1F4BC is BRIEFCASE.
This CL also renames the state name from STATE_BEFORE_ZWJ_EMOJI to
STATE_BEFORE_EMOJI since now all emoji can be a part of ZWJ sequence
after I572dad42ee108476962d4b3fe9f3a6019cb50098
BUG: 29728397
Change-Id: Ib114295db45c6592f1c65a0773ab236f8bf35209
(cherry picked from commit bba8d97c36)
During the offset calculation for selection, SpannableStringBuilder
had an overflow while multiplying two int values. This CL uses long to
calculate the multiplication, and also checks for overflow after
casting the final result into int again.
Bug: 29108549
Change-Id: I11eff4677916701074b38bc5214730fe704707c4
Add an extra parameter to LocaleHelper.getDisplayLocaleList()
specifying the maximum number of locales to output, as callers
probably won't need the whole list.
Bug: 28872122
Change-Id: Ief136bc1af2841e76ed4d8e65932a9a30821eae3
Preallocate storage lists and avoid TextUtils and its string
builder for a common code path.
Optimize list join helper to not have a check in the loop.
Bug: 28801010
Change-Id: Iafc582031f973d718252b34bcda6405a77425628
Since LocaleList needs to depend on android.os.Parcelable, we cannot let
that class belong to "android.util" package, which causes layering
violation.
Bug: 28819696
Change-Id: Ia8de2ee9df3dd0a42b1fe84574439519b680fe18
To properly apply ellipsis, we virtually concatenate the last line and
overflowed lines after line breaking for a paragraph.
Previously, width of the concatenated line was computed by summing up all
line's width. However, the width is wrong when there are any trailing
white spaces that can be normal white spaces by concatenating lines.
With this CL, we sum up widths of all characters in lines except the last
overflowed line.
Bug: 28599066
Change-Id: I41d828ee8eb8a702cd5096d626b307cbb3467047
This commit makes the LocaleList constructor require non-null
arguments in all cases, and fixes all uses of LocaleList that could
previously pass a null to use getEmptyLocaleList() instead (which is
preferred anyway becaues it avoids an allocation.
Bug: 28460668
Change-Id: I4b8b3cfa82914412731c2b79003951c46cb2afa1
Vendor may want to introduce their own ZWJ emoji sequence. To delete
them by one back space as the same behavior of Unicode emoji sequence,
use all emoji as the zwj emoji.
Bug: 28248662
Change-Id: I572dad42ee108476962d4b3fe9f3a6019cb50098
This undoes the automerger skip which occured in
commit e740c84dc3 and
replays it as a standard (NOT -s ours) merge.
Change-Id: If5a47be26f73d6a0735c425cd66310a3e2a89086
Because of a min/max issue, while processing each paragraph StaticLayout
would trim the left/right indent arrays incorrectly.
Bug: 28090810
Change-Id: Ib2b9b48963861e0952bd45a079179e3cca86ffcf
The returned descender value for BoringLayout and StaticLayout should
be equal to the font's "bottom" metric in the includePadding case.
Previously, the calculation incorrectly included an addition
mBottomPadding value in some cases (which was an attempt to work
around the case where maxLines was set on TextView but not the
corresponding StaticLayout, no longer the case).
Bug: 27901763
Change-Id: I008c5974b979087725a9bb9104e4771b0caac01c
Reset FontMetrics object used in BoringLayout.isBoring to get updated
and correct FontMetrics as a result of measurement.
Bug: 26704088
Change-Id: If77b0edba8dc4b5b1738a802c5f49e112e47b4f2
In addition, all English locales other than those explicitly mapped
to en-US are mapped to en-GB.
BUG: 26405413
Change-Id: Ie5b77af164c95a6ed5639da5752ddf21f92181bc
A new addLinks function is added that accepts a default scheme and a
set of known schemes. Default scheme is applied whenever the link found
does not start with one of the given known schemes.
Moreover, previously JavaDoc for addLinks functions with a single scheme
parameter described that the scheme attribute will be prepended to the
links that do not have 'a' scheme. The code actually checks if the link
starts with the given scheme and prepends if not. JavaDoc updated
accordingly.
Bug: 26985901
Change-Id: I94ea81dcf83ba7a6b6cd47c10fe8fb277964eb15
A point on a direction boundary can be mapped to two offset and
one offset on a direction boundary can be mapped to two points.
Previously, paragraph's primary direction is always used for deciding
offset and coordinates; thus, handle movement around a direction
boundary is often nonintuitive.
With this CL:
1. For selection end handle, direction of character at offset - 1 is
used for deciding handle shape and position.
2. For getting offset from coordinates, previous offset is used to
minimize the offset delta and primary .
3. For getting coordinates form offset, new logic chooses primary or
secondary horizontal coordinate depending on the current run
direction and paragraph direction.
4. When a handle passes another one because it passes a direction
boundary, new logic keeps the handle at the run boundary instead of
minimizing the selection.
Bug: 19199637
Bug: 21480356
Bug: 21649994
Change-Id: I2a7e87ad08416f4bd01a5f68e006240f77d9036b
Forward delete key now deletes characters until the next grapheme
cluster boundary.
Bug: 25737208
Bug: 27035430
Change-Id: Ie2fb510fefa115657cc48063be5319b1eecb30b9
