/// <summary>
/// Returns a new (deterministic and minimal) automaton that accepts the union
/// of the given collection of <seealso cref="BytesRef"/>s representing UTF-8 encoded
/// strings.
/// </summary>
/// <param name="utf8Strings">
/// The input strings, UTF-8 encoded. The collection must be in sorted
/// order.
/// </param>
/// <returns> An <seealso cref="Automaton"/> accepting all input strings. The resulting
/// automaton is codepoint based (full unicode codepoints on
/// transitions). </returns>
public static Automaton MakeStringUnion(ICollection <BytesRef> utf8Strings)
{
if (utf8Strings.Count == 0)
{
return(MakeEmpty());
}
else
{
return(DaciukMihovAutomatonBuilder.Build(utf8Strings));
}
}
private CompiledAutomaton Build(params string[] strings)
{
List <BytesRef> terms = new List <BytesRef>();
foreach (string s in strings)
{
terms.Add(new BytesRef(s));
}
terms.Sort();
Automaton a = DaciukMihovAutomatonBuilder.Build(terms);
return(new CompiledAutomaton(a, true, false));
}
/// <summary>
/// Build a minimal, deterministic automaton from a sorted list of <see cref="BytesRef"/> representing
/// strings in UTF-8. These strings must be binary-sorted.
/// </summary>
public static Automaton Build(ICollection <BytesRef> input)
{
DaciukMihovAutomatonBuilder builder = new DaciukMihovAutomatonBuilder();
CharsRef scratch = new CharsRef();
foreach (BytesRef b in input)
{
UnicodeUtil.UTF8toUTF16(b, scratch);
builder.Add(scratch);
}
return(new Automaton
{
initial = Convert(builder.Complete(), new JCG.Dictionary <State, Lucene.Net.Util.Automaton.State>(IdentityEqualityComparer <State> .Default)),
deterministic = true
});
}
/// <summary>
/// Build a minimal, deterministic automaton from a sorted list of <seealso cref="BytesRef"/> representing
/// strings in UTF-8. These strings must be binary-sorted.
/// </summary>
public static Automaton Build(ICollection <BytesRef> input)
{
DaciukMihovAutomatonBuilder builder = new DaciukMihovAutomatonBuilder();
CharsRef scratch = new CharsRef();
foreach (BytesRef b in input)
{
UnicodeUtil.UTF8toUTF16(b, scratch);
builder.Add(scratch);
}
Automaton a = new Automaton();
a.initial = Convert(builder.Complete(), new IdentityHashMap <State, Lucene.Net.Util.Automaton.State>());
a.deterministic = true;
return(a);
}
/// <summary>
/// Build a minimal, deterministic automaton from a sorted list of <seealso cref="BytesRef"/> representing
/// strings in UTF-8. These strings must be binary-sorted.
/// </summary>
public static Automaton Build(ICollection<BytesRef> input)
{
DaciukMihovAutomatonBuilder builder = new DaciukMihovAutomatonBuilder();
CharsRef scratch = new CharsRef();
foreach (BytesRef b in input)
{
UnicodeUtil.UTF8toUTF16(b, scratch);
builder.Add(scratch);
}
Automaton a = new Automaton();
a.Initial = Convert(builder.Complete(), new IdentityHashMap<State, Lucene.Net.Util.Automaton.State>());
a.deterministic = true;
return a;
}