public void NewCodePointSetHasNoElements()
{
// Act
var sut = new CodePointSet();
// Assert
Assert.That(sut.IsEmpty, Is.True);
Assert.That(sut.Count, Is.EqualTo(0));
}
public void AddManyUnrelatedElements()
{
// Arrange
var sut = new CodePointSet();
var arr = new [] { new CodePoint(1234), new CodePoint(123), new CodePoint(456), new CodePoint(5678), new CodePoint(789) };
// Act
foreach(var v in arr)
sut.Add(v);
// Assert
Assert.That(sut.IsEmpty, Is.False);
Assert.That(sut.Count, Is.EqualTo(arr.Length));
Assert.That(sut, Is.EquivalentTo(arr.OrderBy(i => i)));
foreach(var v in arr)
Assert.That(sut.Contains(v), Is.True);
Assert.That(sut.GetRanges(), Is.EquivalentTo(arr.Select(i => new CodePointRange(i, i))));
}
public void AddOneElement()
{
// Arrange
var sut = new CodePointSet();
var v = new CodePoint(123);
// Act
sut.Add(v);
// Assert
Assert.That(sut.IsEmpty, Is.False);
Assert.That(sut.Count, Is.EqualTo(1));
Assert.That(sut, Is.EquivalentTo(new [] { v }));
Assert.That(sut.Contains(v), Is.True);
Assert.That(sut.Contains(v - 1), Is.False);
Assert.That(sut.Contains(v + 1), Is.False);
Assert.That(sut.GetRanges(), Is.EquivalentTo(new []{ new CodePointRange(v, v) }));
}
public void AddAdjacentElementBeforeMerges()
{
// Arrange
var sut = new CodePointSet();
var v1 = new CodePoint(123);
var v2 = v1 + 1;
// Act
sut.Add(v2);
sut.Add(v1);
// Assert
Assert.That(sut.IsEmpty, Is.False);
Assert.That(sut.Count, Is.EqualTo(2));
Assert.That(sut, Is.EquivalentTo(new [] { v1, v2 }));
Assert.That(sut.Contains(v1), Is.True);
Assert.That(sut.Contains(v2), Is.True);
Assert.That(sut.GetRanges(), Is.EquivalentTo(new[] { new CodePointRange(v1, v2) }));
}
/// <summary>
/// Parses a character set from the parse table.
/// </summary>
/// <param name="listTerm">A list of characters and character ranges.</param>
/// <returns>The parsed character set.</returns>
/// <example>
/// The list term might look like this:
/// <code>
/// [range(9,10),13,32]
/// </code>
/// </example>
private IReadOnlySet<CodePoint> ParseCharacterRanges(IListTerm listTerm)
{
#region Contract
Contract.Requires<ArgumentNullException>(listTerm != null);
Contract.Ensures(Contract.Result<IReadOnlySet<CodePoint>>() != null);
#endregion
CodePointSet characters = new CodePointSet();
foreach (var term in listTerm.SubTerms)
{
int? charValue = term.AsInt32();
IConsTerm rangeTerm = term.AsCons("range", 2);
if (charValue != null)
{
if (charValue <= SpoofaxParseTableFormat.CharacterMax)
characters.Add(new CodePoint((int)charValue));
// Else: ignore. It's a label which we'll add later.
}
else if (rangeTerm != null)
{
int start = rangeTerm[0].ToInt32();
// Clip the end to the max character we read.
int end = Math.Min(rangeTerm[1].ToInt32(), SpoofaxParseTableFormat.CharacterMax);
if (start <= SpoofaxParseTableFormat.CharacterMax)
characters.AddRange(new CodePoint(start), new CodePoint(end));
// Else: ignore. It's a range of labels which we'll add later.
}
else
throw new InvalidParseTableException("Unrecognized term: " + term);
}
// If the range contains the 8-bit Eof, change it into the 16-bit Eof.
if (characters.Contains(new CodePoint(256)))
{
characters.Remove(new CodePoint(256));
characters.Add(CodePoint.Eof);
}
return characters;
}
public void RemoveElementInMiddleOfRange()
{
// Arrange
var sut = new CodePointSet();
var v1 = new CodePoint(123);
var v2 = v1 + 1;
var v3 = v2 + 1;
sut.Add(v1);
sut.Add(v2);
sut.Add(v3);
// Act
sut.Remove(v2);
// Assert
Assert.That(sut.IsEmpty, Is.False);
Assert.That(sut.Count, Is.EqualTo(2));
Assert.That(sut, Is.EquivalentTo(new [] { v1, v3 }));
Assert.That(sut.Contains(v1), Is.True);
Assert.That(sut.Contains(v2), Is.False);
Assert.That(sut.Contains(v3), Is.True);
Assert.That(sut.GetRanges(), Is.EquivalentTo(new []{ new CodePointRange(v1, v1), new CodePointRange(v3, v3) }));
}
public void RemoveStandAloneElement()
{
// Arrange
var sut = new CodePointSet();
var arr = new [] { new CodePoint(1234), new CodePoint(123), new CodePoint(456), new CodePoint(5678), new CodePoint(789) };
foreach(var v in arr) sut.Add(v);
// Act
sut.Remove(arr[2]);
// Assert
Assert.That(sut.IsEmpty, Is.False);
Assert.That(sut.Count, Is.EqualTo(arr.Length - 1));
Assert.That(sut, Is.EquivalentTo(arr.Where(i => i != arr[2]).OrderBy(i => i)));
foreach(var v in arr.Where(i => i != arr[2]))
Assert.That(sut.Contains(v), Is.True);
Assert.That(sut.GetRanges(), Is.EquivalentTo(arr.Where(i => i != arr[2]).Select(i => new CodePointRange(i, i))));
}
/// <summary>
/// Reads a character set from a list term.
/// </summary>
/// <param name="listTerm">A list of characters and character ranges.</param>
/// <returns>The read character set.</returns>
/// <example>
/// The list term might look like this:
/// <code>
/// [range(9,10),13,32]
/// </code>
/// </example>
private IReadOnlySet<CodePoint> ReadCharacterRanges(IListTerm listTerm)
{
#region Contract
Contract.Requires<ArgumentNullException>(listTerm != null);
Contract.Ensures(Contract.Result<IReadOnlySet<CodePoint>>() != null);
#endregion
CodePointSet characters = new CodePointSet();
foreach (var term in listTerm.SubTerms)
{
int? charValue = term.AsInt32();
IConsTerm rangeTerm = term.AsCons("range", 2);
if (charValue != null)
{
characters.Add(new CodePoint((int)charValue));
}
else if (rangeTerm != null)
{
int start = rangeTerm[0].ToInt32();
int end = rangeTerm[1].ToInt32();
characters.AddRange(new CodePoint(start), new CodePoint(end));
}
else
throw new InvalidOperationException("Unrecognized term: " + term);
}
// If the range contains the 8-bit Eof, change it into the 16-bit Eof.
if (characters.Contains(new CodePoint(256)))
{
characters.Remove(new CodePoint(256));
characters.Add(CodePoint.Eof);
}
return characters;
}
