public virtual void TestFilterPositions()
{
TokenStream ts = new MockTokenizer(new StringReader("abcde vwxyz"), MockTokenizer.WHITESPACE, false);
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, ts, EdgeNGramTokenFilter.Side.FRONT, 1, 3);
AssertTokenStreamContents(tokenizer, new string[] { "a", "ab", "abc", "v", "vw", "vwx" }, new int[] { 0, 0, 0, 6, 6, 6 }, new int[] { 5, 5, 5, 11, 11, 11 }, null, new int[] { 1, 0, 0, 1, 0, 0 }, null, null, false);
}
public virtual void TestFrontUnigram()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 1, 1);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "a" }, new int[] { 0 }, new int[] { 5 });
}
public virtual void TestBackRangeOfNgrams()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, input, EdgeNGramTokenFilter.Side.BACK, 1, 3);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "e", "de", "cde" }, new int[] { 4, 3, 2 }, new int[] { 5, 5, 5 }, null, null, null, null, false);
}
public virtual void TestBackUnigram()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, input, EdgeNGramTokenFilter.Side.BACK, 1, 1);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "e" }, new int[] { 4 }, new int[] { 5 });
}
public virtual void TestBackUnigram()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, input, EdgeNGramTokenFilter.Side.BACK, 1, 1);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "e" }, new int[] { 4 }, new int[] { 5 });
}
public virtual void TestOversizedNgrams()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 6, 6);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[0], new int[0], new int[0]);
}
public virtual void TestBackRangeOfNgrams()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, input, EdgeNGramTokenFilter.Side.BACK, 1, 3);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "e", "de", "cde" }, new int[] { 4, 3, 2 }, new int[] { 5, 5, 5 }, null, null, null, null, false);
}
public virtual void TestSmallTokenInStream()
{
input = new MockTokenizer(new StringReader("abc de fgh"), MockTokenizer.WHITESPACE, false);
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 3, 3);
AssertTokenStreamContents(tokenizer, new string[] { "abc", "fgh" }, new int[] { 0, 7 }, new int[] { 3, 10 });
}
public override TokenStreamComponents CreateComponents(string fieldName, TextReader reader)
{
Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
TokenFilter filters = new ASCIIFoldingFilter(tokenizer);
filters = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, filters, EdgeNGramTokenFilter.Side.FRONT, 2, 15);
return(new TokenStreamComponents(tokenizer, filters));
}
public virtual void TestFilterPositions()
{
TokenStream ts = new MockTokenizer(new StringReader("abcde vwxyz"), MockTokenizer.WHITESPACE, false);
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, ts, EdgeNGramTokenFilter.Side.FRONT, 1, 3);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "a", "ab", "abc", "v", "vw", "vwx" }, new int[] { 0, 0, 0, 6, 6, 6 }, new int[] { 5, 5, 5, 11, 11, 11 }, null, new int[] { 1, 0, 0, 1, 0, 0 }, null, null, false);
}
public virtual void TestGraphs()
{
TokenStream tk = new LetterTokenizer(TEST_VERSION_CURRENT, new StringReader("abc d efgh ij klmno p q"));
tk = new ShingleFilter(tk);
tk = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tk, 7, 10);
AssertTokenStreamContents(tk, new string[] { "efgh ij", "ij klmn", "ij klmno", "klmno p" }, new int[] { 6, 11, 11, 14 }, new int[] { 13, 19, 19, 21 }, new int[] { 3, 1, 0, 1 }, new int[] { 2, 2, 2, 2 }, 23);
}
public virtual void TestReset()
{
WhitespaceTokenizer tokenizer = new WhitespaceTokenizer(TEST_VERSION_CURRENT, new StringReader("abcde"));
EdgeNGramTokenFilter filter = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tokenizer, EdgeNGramTokenFilter.Side.FRONT, 1, 3);
AssertTokenStreamContents(filter, new string[] { "a", "ab", "abc" }, new int[] { 0, 0, 0 }, new int[] { 5, 5, 5 });
tokenizer.Reader = new StringReader("abcde");
AssertTokenStreamContents(filter, new string[] { "a", "ab", "abc" }, new int[] { 0, 0, 0 }, new int[] { 5, 5, 5 });
}
public virtual void TestFirstTokenPositionIncrement()
{
TokenStream ts = new MockTokenizer(new StringReader("a abc"), MockTokenizer.WHITESPACE, false);
ts = new PositionFilter(ts); // All but first token will get 0 position increment
EdgeNGramTokenFilter filter = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, ts, EdgeNGramTokenFilter.Side.FRONT, 2, 3);
// The first token "a" will not be output, since it's smaller than the mingram size of 2.
// The second token on input to EdgeNGramTokenFilter will have position increment of 0,
// which should be increased to 1, since this is the first output token in the stream.
AssertTokenStreamContents(filter, new string[] { "ab", "abc" }, new int[] { 2, 2 }, new int[] { 5, 5 }, new int[] { 1, 0 });
}
public virtual void TestSupplementaryCharacters()
{
string s = TestUtil.RandomUnicodeString(Random(), 10);
int codePointCount = Character.CodePointCount(s, 0, s.Length);
int minGram = TestUtil.NextInt(Random(), 1, 3);
int maxGram = TestUtil.NextInt(Random(), minGram, 10);
TokenStream tk = new KeywordTokenizer(new StringReader(s));
tk = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tk, minGram, maxGram);
ICharTermAttribute termAtt = tk.AddAttribute <ICharTermAttribute>();
IOffsetAttribute offsetAtt = tk.AddAttribute <IOffsetAttribute>();
tk.Reset();
for (int i = minGram; i <= Math.Min(codePointCount, maxGram); ++i)
{
assertTrue(tk.IncrementToken());
assertEquals(0, offsetAtt.StartOffset());
assertEquals(s.Length, offsetAtt.EndOffset());
int end = Character.OffsetByCodePoints(s, 0, i);
assertEquals(s.Substring(0, end), termAtt.ToString());
}
assertFalse(tk.IncrementToken());
}
public virtual void TestGraphs()
{
TokenStream tk = new LetterTokenizer(TEST_VERSION_CURRENT, new StringReader("abc d efgh ij klmno p q"));
tk = new ShingleFilter(tk);
tk = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tk, 7, 10);
AssertTokenStreamContents(tk, new string[] { "efgh ij", "ij klmn", "ij klmno", "klmno p" }, new int[] { 6, 11, 11, 14 }, new int[] { 13, 19, 19, 21 }, new int[] { 3, 1, 0, 1 }, new int[] { 2, 2, 2, 2 }, 23);
}
public virtual void TestSmallTokenInStream()
{
input = new MockTokenizer(new StringReader("abc de fgh"), MockTokenizer.WHITESPACE, false);
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 3, 3);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "abc", "fgh" }, new int[] { 0, 7 }, new int[] { 3, 10 });
}
public virtual void TestSupplementaryCharacters()
{
string s = TestUtil.RandomUnicodeString(Random(), 10);
int codePointCount = Character.CodePointCount(s, 0, s.Length);
int minGram = TestUtil.NextInt(Random(), 1, 3);
int maxGram = TestUtil.NextInt(Random(), minGram, 10);
TokenStream tk = new KeywordTokenizer(new StringReader(s));
tk = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tk, minGram, maxGram);
ICharTermAttribute termAtt = tk.AddAttribute<ICharTermAttribute>();
IOffsetAttribute offsetAtt = tk.AddAttribute<IOffsetAttribute>();
tk.Reset();
for (int i = minGram; i <= Math.Min(codePointCount, maxGram); ++i)
{
assertTrue(tk.IncrementToken());
assertEquals(0, offsetAtt.StartOffset());
assertEquals(s.Length, offsetAtt.EndOffset());
int end = Character.OffsetByCodePoints(s, 0, i);
assertEquals(s.Substring(0, end), termAtt.ToString());
}
assertFalse(tk.IncrementToken());
}
public override TokenStreamComponents CreateComponents(string fieldName, TextReader reader)
{
Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
TokenFilter filters = new ASCIIFoldingFilter(tokenizer);
#pragma warning disable 612, 618
filters = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, filters, EdgeNGramTokenFilter.Side.FRONT, 2, 15);
#pragma warning restore 612, 618
return new TokenStreamComponents(tokenizer, filters);
}
public virtual void TestBackUnigram()
{
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(LuceneVersion.LUCENE_43, input, EdgeNGramTokenFilter.Side.BACK, 1, 1);
AssertTokenStreamContents(tokenizer, new string[] { "e" }, new int[] { 4 }, new int[] { 5 });
}
public virtual void TestFrontUnigram()
{
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 1, 1);
AssertTokenStreamContents(tokenizer, new string[] { "a" }, new int[] { 0 }, new int[] { 5 });
}
public virtual void TestFirstTokenPositionIncrement()
{
TokenStream ts = new MockTokenizer(new StringReader("a abc"), MockTokenizer.WHITESPACE, false);
ts = new PositionFilter(ts); // All but first token will get 0 position increment
#pragma warning disable 612, 618
EdgeNGramTokenFilter filter = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, ts, EdgeNGramTokenFilter.Side.FRONT, 2, 3);
#pragma warning restore 612, 618
// The first token "a" will not be output, since it's smaller than the mingram size of 2.
// The second token on input to EdgeNGramTokenFilter will have position increment of 0,
// which should be increased to 1, since this is the first output token in the stream.
AssertTokenStreamContents(filter, new string[] { "ab", "abc" }, new int[] { 2, 2 }, new int[] { 5, 5 }, new int[] { 1, 0 });
}
public virtual void TestOversizedNgrams()
{
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 6, 6);
AssertTokenStreamContents(tokenizer, new string[0], new int[0], new int[0]);
}
public virtual void TestFrontRangeOfNgrams()
{
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 1, 3);
AssertTokenStreamContents(tokenizer, new string[] { "a", "ab", "abc" }, new int[] { 0, 0, 0 }, new int[] { 5, 5, 5 });
}
public virtual void TestReset()
{
WhitespaceTokenizer tokenizer = new WhitespaceTokenizer(TEST_VERSION_CURRENT, new StringReader("abcde"));
#pragma warning disable 612, 618
EdgeNGramTokenFilter filter = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, tokenizer, EdgeNGramTokenFilter.Side.FRONT, 1, 3);
#pragma warning restore 612, 618
AssertTokenStreamContents(filter, new string[] { "a", "ab", "abc" }, new int[] { 0, 0, 0 }, new int[] { 5, 5, 5 });
tokenizer.Reader = new StringReader("abcde");
AssertTokenStreamContents(filter, new string[] { "a", "ab", "abc" }, new int[] { 0, 0, 0 }, new int[] { 5, 5, 5 });
}
public virtual void TestFrontUnigram()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 1, 1);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[] { "a" }, new int[] { 0 }, new int[] { 5 });
}
public virtual void TestOversizedNgrams()
{
#pragma warning disable 612, 618
EdgeNGramTokenFilter tokenizer = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, input, EdgeNGramTokenFilter.Side.FRONT, 6, 6);
#pragma warning restore 612, 618
AssertTokenStreamContents(tokenizer, new string[0], new int[0], new int[0]);
}