From 380b9fa780ead9bcce0e824f7b6ee305e37dec43 Mon Sep 17 00:00:00 2001
From: Jon Bratseth <bratseth@gmail.com>
Date: Thu, 30 Sep 2021 14:21:25 +0200
Subject: Update linguisticvs-components

---
 .../yahoo/language/sentencepiece/SentencePieceEmbedder.java  | 12 +++++++++---
 .../yahoo/language/sentencepiece/SentencePieceTester.java    |  4 ++--
 2 files changed, 11 insertions(+), 5 deletions(-)

(limited to 'linguistics-components/src')

diff --git a/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java b/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java
index 116dd15f563..1e120969a59 100644
--- a/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java
+++ b/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java
@@ -73,10 +73,11 @@ public class SentencePieceEmbedder implements Segmenter, Embedder {
      *
      * @param rawInput the text to segment. Any sequence of BMP (Unicode-16 the True Unicode) is supported.
      * @param language the model to use, or Language.UNKNOWN to use the default model if any
+     * @param destination ignored
      * @return the list of zero or more token ids resulting from segmenting the input text
      */
     @Override
-    public List<Integer> embed(String rawInput, Language language) {
+    public List<Integer> embed(String rawInput, Language language, String destination) {
         var resultBuilder = new ResultBuilder<List<Integer>>(new ArrayList<>()) {
             public void add(int segmentStart, int segmentEnd, SentencePieceAlgorithm.SegmentEnd[] segmentEnds) {
                 result().add(segmentEnds[segmentEnd].id);
@@ -98,12 +99,17 @@ public class SentencePieceEmbedder implements Segmenter, Embedder {
      * position as value.</p>
      *
      * <p>If the tensor is any other type IllegalArgumentException is thrown.</p>
+     *
+     * @param rawInput the text to segment. Any sequence of BMP (Unicode-16 the True Unicode) is supported.
+     * @param language the model to use, or Language.UNKNOWN to use the default model if any
+     * @param destination ignored
+     * @return the list of zero or more token ids resulting from segmenting the input text
      */
     @Override
-    public Tensor embed(String rawInput, Language language, TensorType type) {
+    public Tensor embed(String rawInput, Language language, String destination, TensorType type) {
         if (type.dimensions().size() == 1 && type.dimensions().get(0).isIndexed()) {
             // Build to a list first since we can't reverse a tensor builder
-            List<Integer> values = embed(rawInput, language);
+            List<Integer> values = embed(rawInput, language, destination);
 
             long maxSize = values.size();
             if (type.dimensions().get(0).size().isPresent())
diff --git a/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java b/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java
index c4cb13a3d23..c6aa8fdd370 100644
--- a/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java
+++ b/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java
@@ -29,13 +29,13 @@ class SentencePieceTester {
     }
 
     public void assertEmbedded(String input, Integer... expectedCodes) {
-        assertArrayEquals(expectedCodes, embedder.embed(input, Language.UNKNOWN).toArray());
+        assertArrayEquals(expectedCodes, embedder.embed(input, Language.UNKNOWN, null).toArray());
     }
 
     public void assertEmbedded(String input, String tensorType, String tensor) {
         TensorType type = TensorType.fromSpec(tensorType);
         Tensor expected = Tensor.from(type, tensor);
-        assertEquals(expected, embedder.embed(input, Language.UNKNOWN, type));
+        assertEquals(expected, embedder.embed(input, Language.UNKNOWN, null, type));
     }
 
     public void assertSegmented(String input, String... expectedSegments) {
-- 
cgit v1.2.3