/// <summary>
/// Find the given string in the pool. If not found, returns null.
/// </summary>
public NormStr Get(string str, bool add = false)
{
AssertValid();
if (str == null)
{
str = "";
}
var strSpan = str.AsSpan();
uint hash = Hashing.HashString(strSpan);
int ins = GetIns(hash);
while (ins >= 0)
{
ulong meta = _rgmeta[ins];
if ((int)Utils.GetLo(meta) == str.Length)
{
var ns = GetNs(ins);
if (strSpan.SequenceEqual(ns.Value.Span))
{
return(ns);
}
}
ins = (int)Utils.GetHi(meta);
}
Contracts.Assert(ins == -1);
return(add ? AddCore(str.AsMemory(), hash) : null);
}
public NormStr Get(ReadOnlyMemory <char> str, bool add = false, bool duplicateStr = true)
{
AssertValid();
var span = str.Span;
uint hash = Hashing.HashString(span);
int ins = GetIns(hash);
while (ins >= 0)
{
ulong meta = _rgmeta[ins];
if ((int)Utils.GetLo(meta) == str.Length)
{
var ns = GetNs(ins);
if (ns.Value.Span.SequenceEqual(span))
{
return(ns);
}
}
ins = (int)Utils.GetHi(meta);
}
Contracts.Assert(ins == -1);
if (duplicateStr)
{
// To avoid the case where 'str' actually stores a string with the
// content of a whole row in the dataset, a new 'str' is created
// See issue https://github.com/dotnet/machinelearning/issues/4571
// and PR https://github.com/dotnet/machinelearning/pull/4576
return(add ? AddCore(str.ToString().AsMemory(), hash) : null);
}
return(add ? AddCore(str, hash) : null);
}
/// <summary>
/// Make sure the given string has an equivalent NormStr in the pool and return it.
/// </summary>
public NormStr Get(StringBuilder sb, bool add = false)
{
AssertValid();
if (sb == null)
{
return(Get("", add));
}
int cch = sb.Length;
NormStr ns;
uint hash = Hashing.HashString(sb);
int ins = GetIns(hash);
while (ins >= 0)
{
ulong meta = _rgmeta[ins];
if ((int)Utils.GetLo(meta) == cch)
{
ns = GetNs(ins);
var value = ns.Value;
for (int ich = 0; ; ich++)
{
if (ich == cch)
{
return(ns);
}
if (value.Span[ich] != sb[ich])
{
break;
}
}
}
ins = (int)Utils.GetHi(meta);
}
Contracts.Assert(ins == -1);
return(add ? AddCore(sb.ToString().AsMemory(), hash) : null);
}
public static uint HashUint(uint u)
{
ulong uu = (ulong)u * 0x7ff19519UL; // this number is prime.
return(Utils.GetLo(uu) + Utils.GetHi(uu));
}
