organicmaps/icu
13
Fork 0
mirror of https://github.com/unicode-org/icu.git synced 2025-04-13 08:53:20 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 3

ICU-22789 Fix typos, add TODOs for future optimization design

Browse source
This commit is contained in:
Elango Cheran 2025-01-03 11:19:57 -08:00
parent 4163a6d898
commit d9017e0408
2 changed files with 17 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

25
icu4j/main/core/src/main/java/com/ibm/icu/text/segmenter/Segments.java
View file

@ -11,14 +11,6 @@ public interface Segments {
Stream<Segment> ranges();
/**
* Returns whether offset {@code i} is a segmentation boundary. Throws an exception when
* {@code i} is not a valid boundary position for the source sequence.
* @param i
* @return
*/
boolean isBoundary(int i);
Stream<Segment> rangesAfterIndex(int i);
Stream<Segment> rangesBeforeIndex(int i);
@ -29,6 +21,14 @@ public interface Segments {
Function<Segment, CharSequence> rangeToSequenceFn();
/**
* Returns whether offset {@code i} is a segmentation boundary. Throws an exception when
* {@code i} is not a valid boundary position for the source sequence.
* @param i
* @return
*/
boolean isBoundary(int i);
IntStream boundaries();
IntStream boundariesAfter(int i);
@ -48,16 +48,21 @@ public interface Segments {
// Inner classes for Segment, SegmentIterable, and SegmentIterator
//
// TODO: consider options in design for potential memory usage optimization:
// 1) keep simple class with public fields, but requires field per Segment to point to source
// 2) make Segment an interface (getSource, getStart, getLimit, getRuleStatus, newSegment), and
// maybe an abstract class that implements the interface, maybe with a default method impl
// for convenience for getting (allocating & returning) the subsequence
class Segment {
public final int start;
public final int limit;
public final int ruleStatus = 0;
public final CharSequence soruce;
public final CharSequence source;
public Segment(int start, int limit, CharSequence source) {
this.start = start;
this.limit = limit;
this.soruce = source;
this.source = source;
}
}

2
icu4j/main/core/src/main/java/com/ibm/icu/text/segmenter/SegmentsImplUtils.java
View file

@ -94,6 +94,7 @@ public class SegmentsImplUtils {
breakIter.setText(sourceSequence);
// create a Stream from a Spliterator of an Iterable so that the Stream can be lazy, not eager
// TODO: optimize IntStream creation to avoid autoboxing
BoundaryIterable iterable = new BoundaryIterable(breakIter, IterationDirection.FORWARDS, i);
Stream<Integer> boundariesAsIntegers = StreamSupport.stream(iterable.spliterator(), false);
return boundariesAsIntegers.mapToInt(Integer::intValue);
@ -112,6 +113,7 @@ public class SegmentsImplUtils {
int backFromIdx = isOnBoundary ? i + 1 : i;
// create a Stream from a Spliterator of an Iterable so that the Stream can be lazy, not eager
// TODO: optimize IntStream creation to avoid autoboxing
BoundaryIterable iterable = new BoundaryIterable(breakIter, IterationDirection.BACKWARDS, backFromIdx);
Stream<Integer> boundariesAsIntegers = StreamSupport.stream(iterable.spliterator(), false);
return boundariesAsIntegers.mapToInt(Integer::intValue);