public void Tokenize_Abbreviation_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer(new[] { "mr", "dr", "ms" });
Assert.That(tokenizer.TokenizeToStrings("Mr. Smith went to Washington."),
Is.EqualTo(new[] { "Mr.", "Smith", "went", "to", "Washington", "." }));
}
public void Tokenize_Quotes_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer();
Assert.That(tokenizer.TokenizeToStrings("\"This is a test.\""),
Is.EqualTo(new[] { "\"", "This", "is", "a", "test", ".", "\"" }));
}
public void Tokenize_PunctuationAtStartOfWord_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer();
Assert.That(tokenizer.TokenizeToStrings("Is this a \"test\"?"),
Is.EqualTo(new[] { "Is", "this", "a", "\"", "test", "\"", "?" }));
}
public void Tokenize_PunctuationInsideWord_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer();
Assert.That(tokenizer.TokenizeToStrings("This isn't a test."),
Is.EqualTo(new[] { "This", "isn't", "a", "test", "." }));
}
/// <summary>
/// This tests a workaround for a bug in Bridge.NET, see issue #2981.
/// </summary>
private static void Tokenize_NonAsciiCharacter_DoesNotThrow(Assert assert)
{
var tokenizer = new LatinWordTokenizer();
assert.DeepEqual(tokenizer.TokenizeToStrings("This is—a test.").ToArray(),
new[] { "This", "is", "—", "a", "test", "." });
}
private static void Tokenize_Abbreviation_ReturnsTokens(Assert assert)
{
var tokenizer = new LatinWordTokenizer(new[] { "mr", "dr", "ms" });
assert.DeepEqual(tokenizer.TokenizeToStrings("Mr. Smith went to Washington.").ToArray(),
new[] { "Mr.", "Smith", "went", "to", "Washington", "." });
}
private static void Tokenize_PunctuationInsideWord_ReturnsTokens(Assert assert)
{
var tokenizer = new LatinWordTokenizer();
assert.DeepEqual(tokenizer.TokenizeToStrings("This isn't a test.").ToArray(),
new[] { "This", "isn't", "a", "test", "." });
}
private static void Tokenize_PunctuationAtStartOfWord_ReturnsTokens(Assert assert)
{
var tokenizer = new LatinWordTokenizer();
assert.DeepEqual(tokenizer.TokenizeToStrings("Is this a \"test\"?").ToArray(),
new[] { "Is", "this", "a", "\"", "test", "\"", "?" });
}
public void Tokenize_ApostropheAsSingleQuote_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer {
TreatApostropheAsSingleQuote = true
};
Assert.That(tokenizer.TokenizeToStrings("'Moses's cat said 'Meow' to the dog.'"),
Is.EqualTo(new[] { "'", "Moses's", "cat", "said", "'", "Meow", "'", "to", "the", "dog", ".", "'" }));
}
public void Tokenize_ApostropheNotAsSingleQuote_ReturnsTokens()
{
var tokenizer = new LatinWordTokenizer();
Assert.That(tokenizer.TokenizeToStrings("“Moses' cat said ‘Meow’ to the dog.”"),
Is.EqualTo(new[] { "“", "Moses'", "cat", "said", "‘", "Meow", "’", "to", "the", "dog", ".", "”" }));
Assert.That(tokenizer.TokenizeToStrings("i ha''on ot ano'."),
Is.EqualTo(new[] { "i", "ha''on", "ot", "ano'", "." }));
}
public void Tokenize_Empty_ReturnsEmpty()
{
var tokenizer = new LatinWordTokenizer();
Assert.That(tokenizer.TokenizeToStrings(""), Is.Empty);
}
private static void Tokenize_Whitespace_ReturnsEmpty(Assert assert)
{
var tokenizer = new LatinWordTokenizer();
assert.DeepEqual(tokenizer.TokenizeToStrings(" ").ToArray(), new string[0]);
}
