From cc60531ac22a7e9601055174a02a6e67c428f800 Mon Sep 17 00:00:00 2001 From: Jon Bratseth Date: Mon, 22 May 2023 23:08:48 +0200 Subject: Always treat each symbol as a separate token --- .../language/opennlp/OpenNlpTokenizationTestCase.java | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'opennlp-linguistics/src/test/java/com/yahoo/language/opennlp/OpenNlpTokenizationTestCase.java') diff --git a/opennlp-linguistics/src/test/java/com/yahoo/language/opennlp/OpenNlpTokenizationTestCase.java b/opennlp-linguistics/src/test/java/com/yahoo/language/opennlp/OpenNlpTokenizationTestCase.java index ef29ffd51cc..78412f94fd4 100644 --- a/opennlp-linguistics/src/test/java/com/yahoo/language/opennlp/OpenNlpTokenizationTestCase.java +++ b/opennlp-linguistics/src/test/java/com/yahoo/language/opennlp/OpenNlpTokenizationTestCase.java @@ -165,11 +165,18 @@ public class OpenNlpTokenizationTestCase { @Test public void testTokenizeEmojis() { - String emoji = "\uD83D\uDD2A"; // 🔪 - Iterator tokens = tokenizer.tokenize(emoji, Language.ENGLISH, StemMode.ALL, true).iterator(); - assertTrue(tokens.hasNext()); - assertEquals(emoji, tokens.next().getTokenString()); - assertFalse(tokens.hasNext()); + String emoji1 = "\uD83D\uDD2A"; // 🔪 + Iterator tokens1 = tokenizer.tokenize(emoji1, Language.ENGLISH, StemMode.ALL, true).iterator(); + assertTrue(tokens1.hasNext()); + assertEquals(emoji1, tokens1.next().getTokenString()); + assertFalse(tokens1.hasNext()); + + String emoji2 = "\uD83D\uDE00"; // 😀 + Iterator tokens2 = tokenizer.tokenize(emoji1 + emoji2, Language.ENGLISH, StemMode.ALL, true).iterator(); + assertTrue(tokens2.hasNext()); + assertEquals(emoji1, tokens2.next().getTokenString()); + assertEquals(emoji2, tokens2.next().getTokenString()); + assertFalse(tokens2.hasNext()); } @Test -- cgit v1.2.3