diff options
Diffstat (limited to 'linguistics/src/test/java/com/yahoo/language/simple/SimpleNormalizerTestCase.java')
-rw-r--r-- | linguistics/src/test/java/com/yahoo/language/simple/SimpleNormalizerTestCase.java | 34 |
1 files changed, 34 insertions, 0 deletions
diff --git a/linguistics/src/test/java/com/yahoo/language/simple/SimpleNormalizerTestCase.java b/linguistics/src/test/java/com/yahoo/language/simple/SimpleNormalizerTestCase.java new file mode 100644 index 00000000000..9c9c8b8fcc5 --- /dev/null +++ b/linguistics/src/test/java/com/yahoo/language/simple/SimpleNormalizerTestCase.java @@ -0,0 +1,34 @@ +// Copyright 2016 Yahoo Inc. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root. +package com.yahoo.language.simple; + +import com.yahoo.language.process.Normalizer; +import org.junit.Test; + +import static org.junit.Assert.assertEquals; + +/** + * @author <a href="mailto:simon@yahoo-inc.com">Simon Thoresen Hult</a> + */ +public class SimpleNormalizerTestCase { + + private static final Normalizer NORMALIZER = new SimpleNormalizer(); + + @Test + public void requireThatInputIsNfkcNormalized() { + assertNormalize("\u212B", "\u00C5"); + assertNormalize("\u2126", "\u03A9"); + assertNormalize("\u00C5", "\u00C5"); + assertNormalize("\u00F4", "\u00F4"); + assertNormalize("\u1E69", "\u1E69"); + assertNormalize("\u1E0B\u0323", "\u1E0D\u0307"); + assertNormalize("\u0071\u0307\u0323", "q\u0323\u0307"); + assertNormalize("\uFB01", "fi"); + assertNormalize("\u0032\u2075", "25"); + assertNormalize("\u1E9B\u0323", "\u1E69"); + } + + private static void assertNormalize(String input, String expectedNormalForm) { + assertEquals(expectedNormalForm, NORMALIZER.normalize(input)); + } + +}
\ No newline at end of file |