public void LCPInt()
{
var rnd = new Random();
for (int n = 1; n < 100; n++)
{
var str = rnd.NextIntArray(n, -5, 5);
var sa = SuffixArray.Create(str);
var saNative = GetNative((ReadOnlySpan <int>)str);
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
{
var str = Enumerable.Repeat(1, 100).ToArray();
var sa = SuffixArray.Create(str);
var saNative = GetNative((ReadOnlySpan <int>)str);
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
{
var str = Enumerable.Range(0, 100).ToArray();
var sa = SuffixArray.Create(str);
var saNative = GetNative((ReadOnlySpan <int>)str);
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
{
var str = new[] { -4210, 4219014, -5, -4210, -4210, 4219014, -5, -4210 };
var sa = SuffixArray.Create(str);
var saNative = GetNative((ReadOnlySpan <int>)str);
sa.LongestCommonPrefix(0, 3).Should().Be(1);
sa.LongestCommonPrefix(0, 4).Should().Be(4);
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
}
public void Occurences2()
{
var sa = SuffixArray.Create("papapa");
PrintSortedArray(sa);
Assert.AreEqual(new[] { 0, 2 }, sa.FindOccurences("papa".ToCharArray()).ToArray());
}
public void Sufa_Abracadabra()
{
var sa = SuffixArray.Create("abracadabra");
//DumpSortedArray(sa);
Debug.Print(sa.ToString());
Assert.AreEqual(new[] { 0, 3, 5, 7, 10 }, sa.FindOccurences("a".ToCharArray()).OrderBy(i => i).ToArray());
}
public void Sufa_ByteArray()
{
var sa = SuffixArray.Create(new byte[]
{
0x01, 0x02, 0x4E, 0x75,
0x01, 0x02, 0x4E, 0x75,
});
Assert.AreEqual(new[] { 2, 6 }, sa.FindOccurences(new byte[] { 0x4E, 0x75 }).OrderBy(i => i).ToArray());
}
public void LCPString()
{
var rnd = new Random();
for (int n = 1; n < 100; n++)
{
var str = rnd.NextString(n);
var sa = SuffixArray.Create(str);
var saNative = GetNative(str.AsSpan());
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
{
var str = "abcaabca";
var sa = SuffixArray.Create(str);
var saNative = GetNative(str.AsSpan());
sa.LongestCommonPrefix(0, 3).Should().Be(1);
sa.LongestCommonPrefix(0, 4).Should().Be(4);
sa.SA.Should().Equal(
7, // a
3, // aabca
4, // abca
0, // abcaabca
5, // bca
1, // bcaabca
6, // ca
2 // caabca
);
sa.LcpArray.Should().Equal(
1, // a - aabca
1, // aabca - abca
4, // abca - abcaabca
0, // abcaabca - bca
3, // bca - bcaabca
0, // bcaabca - ca
2 // ca - caabca
);
sa.Rank.Should().Equal(3, 5, 7, 1, 2, 4, 6, 0);
for (int i = 0; i < str.Length; i++)
{
for (int j = i; j < str.Length; j++)
{
sa.LongestCommonPrefix(i, j).Should().Be(sa.LongestCommonPrefix(j, i)).And.Be(saNative.GetLCP(i, j));
}
}
}
}
//[Test]
public void Timing()
{
for (int N = 2; N < 100000000; N = N * 3 / 2)
{
var rnd = new Random(0x4711);
var bytes = new byte[N];
rnd.NextBytes(bytes);
var str = new string(bytes.Select(b => (char)b).ToArray());
var sw1 = Time(str, s => SuffixArray.Create(s));
var sw2 = Time(str, s => { });
var beforeGc = GC.GetTotalMemory(false) / 1024.0 / 1024.0;
GC.Collect();
var afterGc = GC.GetTotalMemory(false) / 1024.0 / 1024.0;
Debug.Print("{0,15} {1,15} msec {2,15} msec {3} / {4} MiB", N, sw1.ElapsedMilliseconds, sw2.ElapsedMilliseconds,
beforeGc,
afterGc);
}
}
public void Test3()
{
string str = "abracadabra";
string[] expectedSubstrs = { "a", "abra", "abracadabra", "acadabra", "adabra", "bra", "bracadabra", "cadabra", "dabra", "ra", "racadabra" };
int[] expectedLcps = { 0, 1, 4, 1, 1, 0, 3, 0, 0, 0, 2 };
var sa = SuffixArray.Create(str);
//DumpSortedArray(sa);
Assert.AreEqual(sa.Length, str.Length, "Wrong SA length");
Assert.AreEqual(sa.Lcp.Length, str.Length + 1, "Wrong LCP length");
for (int i = 0; i < str.Length; ++i)
{
Assert.AreEqual(str.Substring(sa[i]), expectedSubstrs[i], String.Format("Wrong entry {0}", i));
Assert.AreEqual(sa.Lcp[i], expectedLcps[i], String.Format("Wrong LCP {0}", i));
}
}
/// <summary>
/// Creates the LCP array in O(n) time from the given text.
/// The last character in argument 'text' must be the unique $ char. Refer to SuffixArray class remarks..
/// </summary>
/// <param name="text">The text.</param>
/// <returns></returns>
public static int[] Build(string text)
{
var suffixArray = SuffixArray.Create(text);
if (suffixArray != null)
{
var suffixArrayReverse = new int[suffixArray.Length];
for (int k = 0; k < suffixArray.Length; ++k)
{
suffixArrayReverse[suffixArray[k]] = k;
}
var lcpArray = new int[suffixArray.Length];
lcpArray[0] = -1;
int m = 0;
for (int i = 0; i < suffixArray.Length - 1; i++)
{
int j = suffixArray[suffixArrayReverse[i] - 1];
while (text[m + i] == text[m + j])
{
++m;
}
lcpArray[suffixArrayReverse[i]] = m;
if (m > 0)
{
--m;
}
}
return(lcpArray);
}
else
{
return(null);
}
}
private object EnsureSuffixArray(string filename, byte[] image)
{
var fsSvc = Services.RequireService <IFileSystemService>();
var diagSvc = Services.RequireService <IDiagnosticsService>();
Stream?stm = null;
try
{
if (fsSvc.FileExists(filename))
{
stm = fsSvc.CreateFileStream(filename, FileMode.Open, FileAccess.Read);
try
{
var sSuffix = (int[])new UbjsonReader(stm).Read() !;
return(SuffixArray.Load(image, sSuffix));
}
catch (Exception ex)
{
diagSvc.Warn("Unable to load suffix array {0}. {1}", filename, ex.Message);
} finally
{
stm.Close();
}
}
var sa = SuffixArray.Create(image);
stm = fsSvc.CreateFileStream(filename, FileMode.Create, FileAccess.Write);
new UbjsonWriter(stm).Write(sa.Save());
return(sa);
}
finally
{
if (stm != null)
{
stm.Dispose();
}
}
}
public void TestSearch3()
{
var sa = SuffixArray.Create("yakawow");
Assert.AreEqual(sa.IndexOf(new char[] { 'z' }), -1, "Found wrong substring.");
}
public void TestSearch2()
{
var sa = SuffixArray.Create("yakawow");
Assert.AreEqual(sa.IndexOf("yakawow".ToCharArray()), 0, "Wrong index");
}
public void TestSearch1()
{
var sa = SuffixArray.Create("yakawow");
Assert.AreEqual(sa.IndexOf(new char[] { 'a' }), 1, "Substring not found/Wrong index");
}
public void Test2()
{
SuffixArray <char> sa = SuffixArray.Create((string)null);
Assert.AreEqual(sa.IndexOf(new[] { 'a' }), -1, "Found wrong substring");
}
public void Occurences()
{
var sa = SuffixArray.Create("zappa");
Assert.AreEqual(new[] { 1, 4 }, sa.FindOccurences("a".ToCharArray()).ToArray());
}
public void TestSearch5()
{
var sa = SuffixArray.Create("yakawow");
Assert.AreEqual(sa.IndexOf(new char[0]), -1, "Wrong index");
}