// pre-order traversal
private void ReadRecursively(DataInput @in, TernaryTreeNode node)
{
node.splitchar = @in.ReadString().First();
sbyte mask = (sbyte)@in.ReadByte();
if ((mask & HAS_TOKEN) != 0)
{
node.token = @in.ReadString();
}
if ((mask & HAS_VALUE) != 0)
{
node.val = Convert.ToInt64(@in.ReadInt64());
}
if ((mask & LO_KID) != 0)
{
node.loKid = new TernaryTreeNode();
ReadRecursively(@in, node.loKid);
}
if ((mask & EQ_KID) != 0)
{
node.eqKid = new TernaryTreeNode();
ReadRecursively(@in, node.eqKid);
}
if ((mask & HI_KID) != 0)
{
node.hiKid = new TernaryTreeNode();
ReadRecursively(@in, node.hiKid);
}
}
public override void Build(IInputIterator tfit)
{
if (tfit.HasPayloads)
{
throw new System.ArgumentException("this suggester doesn't support payloads");
}
if (tfit.HasContexts)
{
throw new System.ArgumentException("this suggester doesn't support contexts");
}
root = new TernaryTreeNode();
// buffer first
#pragma warning disable 612, 618
if (tfit.Comparer != BytesRef.UTF8SortedAsUTF16Comparer)
{
// make sure it's sorted and the comparer uses UTF16 sort order
tfit = new SortedInputIterator(tfit, BytesRef.UTF8SortedAsUTF16Comparer);
}
#pragma warning restore 612, 618
List <string> tokens = new List <string>();
List <object> vals = new List <object>();
BytesRef spare;
CharsRef charsSpare = new CharsRef();
while ((spare = tfit.Next()) != null)
{
charsSpare.Grow(spare.Length);
UnicodeUtil.UTF8toUTF16(spare.Bytes, spare.Offset, spare.Length, charsSpare);
tokens.Add(charsSpare.ToString());
vals.Add(Convert.ToInt64(tfit.Weight));
}
autocomplete.BalancedTree(tokens.ToArray(), vals.ToArray(), 0, tokens.Count - 1, root);
}
public override bool Load(DataInput input)
{
lock (this)
{
count = input.ReadVInt64();
root = new TernaryTreeNode();
ReadRecursively(input, root);
return(true);
}
}
/// <summary>
/// Inserting keys in TST in the order middle,small,big (lexicographic measure)
/// recursively creates a balanced tree which reduces insertion and search
/// times significantly.
/// </summary>
/// <param name="tokens">
/// Sorted list of keys to be inserted in TST. </param>
/// <param name="lo">
/// stores the lower index of current list. </param>
/// <param name="hi">
/// stores the higher index of current list. </param>
/// <param name="root">
/// a reference object to root of TST. </param>
public virtual void BalancedTree(object[] tokens, object[] vals, int lo, int hi, TernaryTreeNode root)
{
if (lo > hi)
{
return;
}
int mid = (lo + hi) / 2;
root = Insert(root, (string)tokens[mid], vals[mid], 0);
BalancedTree(tokens, vals, lo, mid - 1, root);
BalancedTree(tokens, vals, mid + 1, hi, root);
}
// pre-order traversal
private void WriteRecursively(DataOutput @out, TernaryTreeNode node)
{
// write out the current node
@out.WriteString(new string(new char[] { node.splitchar }, 0, 1));
// prepare a mask of kids
sbyte mask = 0;
if (node.eqKid != null)
{
mask |= EQ_KID;
}
if (node.loKid != null)
{
mask |= LO_KID;
}
if (node.hiKid != null)
{
mask |= HI_KID;
}
if (node.token != null)
{
mask |= HAS_TOKEN;
}
if (node.val != null)
{
mask |= HAS_VALUE;
}
@out.WriteByte((byte)mask);
if (node.token != null)
{
@out.WriteString(node.token);
}
if (node.val != null)
{
@out.WriteInt64((long)node.val);
}
// recurse and write kids
if (node.loKid != null)
{
WriteRecursively(@out, node.loKid);
}
if (node.eqKid != null)
{
WriteRecursively(@out, node.eqKid);
}
if (node.hiKid != null)
{
WriteRecursively(@out, node.hiKid);
}
}
public override bool Load(DataInput input)
{
UninterruptableMonitor.Enter(this);
try
{
count = input.ReadVInt64();
root = new TernaryTreeNode();
ReadRecursively(input, root);
return(true);
}
finally
{
UninterruptableMonitor.Exit(this);
}
}
/// <summary>
/// Inserts a key in TST creating a series of Binary Search Trees at each node.
/// The key is actually stored across the eqKid of each node in a successive
/// manner.
/// </summary>
/// <param name="currentNode">
/// a reference node where the insertion will take currently. </param>
/// <param name="s">
/// key to be inserted in TST. </param>
/// <param name="x">
/// index of character in key to be inserted currently. </param>
/// <returns> currentNode The new reference to root node of TST </returns>
public virtual TernaryTreeNode Insert(TernaryTreeNode currentNode, string s, object val, int x)
{
if (s == null || s.Length <= x)
{
return currentNode;
}
if (currentNode == null)
{
TernaryTreeNode newNode = new TernaryTreeNode();
newNode.splitchar = s.charAt(x);
currentNode = newNode;
if (x < s.Length - 1)
{
currentNode.eqKid = Insert(currentNode.eqKid, s, val, x + 1);
}
else
{
currentNode.token = s.ToString();
currentNode.val = val;
return currentNode;
}
}
else if (currentNode.splitchar > s.charAt(x))
{
currentNode.loKid = Insert(currentNode.loKid, s, val, x);
}
else if (currentNode.splitchar == s.charAt(x))
{
if (x < s.Length - 1)
{
currentNode.eqKid = Insert(currentNode.eqKid, s, val, x + 1);
}
else
{
currentNode.token = s;
currentNode.val = val;
return currentNode;
}
}
else
{
currentNode.hiKid = Insert(currentNode.hiKid, s, val, x);
}
return currentNode;
}
/// <summary>
/// Inserts a key in TST creating a series of Binary Search Trees at each node.
/// The key is actually stored across the eqKid of each node in a successive
/// manner.
/// </summary>
/// <param name="currentNode">
/// a reference node where the insertion will take currently. </param>
/// <param name="s">
/// key to be inserted in TST. </param>
/// <param name="x">
/// index of character in key to be inserted currently. </param>
/// <returns> The new reference to root node of TST </returns>
public virtual TernaryTreeNode Insert(TernaryTreeNode currentNode, string s, object val, int x)
{
if (s is null || s.Length <= x)
{
return(currentNode);
}
if (currentNode is null)
{
TernaryTreeNode newNode = new TernaryTreeNode();
newNode.splitchar = s[x];
currentNode = newNode;
if (x < s.Length - 1)
{
currentNode.eqKid = Insert(currentNode.eqKid, s, val, x + 1);
}
else
{
currentNode.token = s.ToString();
currentNode.val = val;
return(currentNode);
}
}
else if (currentNode.splitchar > s[x])
{
currentNode.loKid = Insert(currentNode.loKid, s, val, x);
}
else if (currentNode.splitchar == s[x])
{
if (x < s.Length - 1)
{
currentNode.eqKid = Insert(currentNode.eqKid, s, val, x + 1);
}
else
{
currentNode.token = s;
currentNode.val = val;
return(currentNode);
}
}
else
{
currentNode.hiKid = Insert(currentNode.hiKid, s, val, x);
}
return(currentNode);
}
public override void Build(IInputEnumerator enumerator)
{
// LUCENENT: Added guard clause for null
if (enumerator is null)
{
throw new ArgumentNullException(nameof(enumerator));
}
if (enumerator.HasPayloads)
{
throw new ArgumentException("this suggester doesn't support payloads");
}
if (enumerator.HasContexts)
{
throw new ArgumentException("this suggester doesn't support contexts");
}
root = new TernaryTreeNode();
// buffer first
#pragma warning disable 612, 618
if (enumerator.Comparer != BytesRef.UTF8SortedAsUTF16Comparer)
{
// make sure it's sorted and the comparer uses UTF16 sort order
enumerator = new SortedInputEnumerator(enumerator, BytesRef.UTF8SortedAsUTF16Comparer);
}
#pragma warning restore 612, 618
JCG.List <string> tokens = new JCG.List <string>();
JCG.List <object> vals = new JCG.List <object>();
BytesRef spare;
CharsRef charsSpare = new CharsRef();
while (enumerator.MoveNext())
{
spare = enumerator.Current;
charsSpare.Grow(spare.Length);
UnicodeUtil.UTF8toUTF16(spare.Bytes, spare.Offset, spare.Length, charsSpare);
tokens.Add(charsSpare.ToString());
vals.Add(enumerator.Weight);
}
autocomplete.BalancedTree(tokens.ToArray(), vals.ToArray(), 0, tokens.Count - 1, root);
}
public override IList <LookupResult> DoLookup(string key, IEnumerable <BytesRef> contexts, bool onlyMorePopular, int num)
{
if (contexts != null)
{
throw new System.ArgumentException("this suggester doesn't support contexts");
}
IList <TernaryTreeNode> list = autocomplete.PrefixCompletion(root, key, 0);
List <LookupResult> res = new List <LookupResult>();
if (list == null || list.Count == 0)
{
return(res);
}
int maxCnt = Math.Min(num, list.Count);
if (onlyMorePopular)
{
LookupPriorityQueue queue = new LookupPriorityQueue(num);
foreach (TernaryTreeNode ttn in list)
{
queue.InsertWithOverflow(new LookupResult(ttn.token, (long)ttn.val));
}
foreach (LookupResult lr in queue.GetResults())
{
res.Add(lr);
}
}
else
{
for (int i = 0; i < maxCnt; i++)
{
TernaryTreeNode ttn = list[i];
res.Add(new LookupResult(ttn.token, (long)ttn.val));
}
}
return(res);
}
/// <summary>
/// Auto-completes a given prefix query using Depth-First Search with the end
/// of prefix as source node each time finding a new leaf to get a complete key
/// to be added in the suggest list.
/// </summary>
/// <param name="root">
/// a reference to root node of TST. </param>
/// <param name="s">
/// prefix query to be auto-completed. </param>
/// <param name="x">
/// index of current character to be searched while traversing through
/// the prefix in TST. </param>
/// <returns> suggest list of auto-completed keys for the given prefix query. </returns>
public virtual IList <TernaryTreeNode> PrefixCompletion(TernaryTreeNode root, string s, int x)
{
TernaryTreeNode p = root;
JCG.List <TernaryTreeNode> suggest = new JCG.List <TernaryTreeNode>();
while (p != null)
{
if (s[x] < p.splitchar)
{
p = p.loKid;
}
else if (s[x] == p.splitchar)
{
if (x == s.Length - 1)
{
break;
}
else
{
x++;
}
p = p.eqKid;
}
else
{
p = p.hiKid;
}
}
if (p is null)
{
return(suggest);
}
if (p.eqKid is null && p.token is null)
{
return(suggest);
}
if (p.eqKid is null && p.token != null)
{
suggest.Add(p);
return(suggest);
}
if (p.token != null)
{
suggest.Add(p);
}
p = p.eqKid;
var st = new Stack <TernaryTreeNode>();
st.Push(p);
while (st.Count > 0)
{
TernaryTreeNode top = st.Peek();
st.Pop();
if (top.token != null)
{
suggest.Add(top);
}
if (top.eqKid != null)
{
st.Push(top.eqKid);
}
if (top.loKid != null)
{
st.Push(top.loKid);
}
if (top.hiKid != null)
{
st.Push(top.hiKid);
}
}
return(suggest);
}
// pre-order traversal
private void WriteRecursively(DataOutput @out, TernaryTreeNode node)
{
// write out the current node
@out.WriteString(new string(new char[] { node.splitchar }, 0, 1));
// prepare a mask of kids
sbyte mask = 0;
if (node.eqKid != null)
{
mask |= EQ_KID;
}
if (node.loKid != null)
{
mask |= LO_KID;
}
if (node.hiKid != null)
{
mask |= HI_KID;
}
if (node.token != null)
{
mask |= HAS_TOKEN;
}
if (node.val != null)
{
mask |= HAS_VALUE;
}
@out.WriteByte((byte)mask);
if (node.token != null)
{
@out.WriteString(node.token);
}
if (node.val != null)
{
@out.WriteLong((long)node.val);
}
// recurse and write kids
if (node.loKid != null)
{
WriteRecursively(@out, node.loKid);
}
if (node.eqKid != null)
{
WriteRecursively(@out, node.eqKid);
}
if (node.hiKid != null)
{
WriteRecursively(@out, node.hiKid);
}
}
// pre-order traversal
private void ReadRecursively(DataInput @in, TernaryTreeNode node)
{
node.splitchar = @in.ReadString().First();
sbyte mask = (sbyte)@in.ReadByte();
if ((mask & HAS_TOKEN) != 0)
{
node.token = @in.ReadString();
}
if ((mask & HAS_VALUE) != 0)
{
node.val = Convert.ToInt64(@in.ReadLong());
}
if ((mask & LO_KID) != 0)
{
node.loKid = new TernaryTreeNode();
ReadRecursively(@in, node.loKid);
}
if ((mask & EQ_KID) != 0)
{
node.eqKid = new TernaryTreeNode();
ReadRecursively(@in, node.eqKid);
}
if ((mask & HI_KID) != 0)
{
node.hiKid = new TernaryTreeNode();
ReadRecursively(@in, node.hiKid);
}
}
public override bool Load(DataInput input)
{
lock (this)
{
count = input.ReadVLong();
root = new TernaryTreeNode();
ReadRecursively(input, root);
return true;
}
}
public override void Build(InputIterator tfit)
{
if (tfit.HasPayloads)
{
throw new System.ArgumentException("this suggester doesn't support payloads");
}
if (tfit.HasContexts)
{
throw new System.ArgumentException("this suggester doesn't support contexts");
}
root = new TernaryTreeNode();
// buffer first
if (tfit.Comparator != BytesRef.UTF8SortedAsUTF16Comparator)
{
// make sure it's sorted and the comparator uses UTF16 sort order
tfit = new SortedInputIterator(tfit, BytesRef.UTF8SortedAsUTF16Comparator);
}
List<string> tokens = new List<string>();
List<Number> vals = new List<Number>();
BytesRef spare;
CharsRef charsSpare = new CharsRef();
while ((spare = tfit.Next()) != null)
{
charsSpare.Grow(spare.Length);
UnicodeUtil.UTF8toUTF16(spare.Bytes, spare.Offset, spare.Length, charsSpare);
tokens.Add(charsSpare.ToString());
vals.Add(Convert.ToInt64(tfit.Weight));
}
autocomplete.BalancedTree(tokens.ToArray(), vals.ToArray(), 0, tokens.Count - 1, root);
}
/// <summary>
/// Inserting keys in TST in the order middle,small,big (lexicographic measure)
/// recursively creates a balanced tree which reduces insertion and search
/// times significantly.
/// </summary>
/// <param name="tokens">
/// Sorted list of keys to be inserted in TST. </param>
/// <param name="lo">
/// stores the lower index of current list. </param>
/// <param name="hi">
/// stores the higher index of current list. </param>
/// <param name="root">
/// a reference object to root of TST. </param>
public virtual void BalancedTree(object[] tokens, object[] vals, int lo, int hi, TernaryTreeNode root)
{
if (lo > hi)
{
return;
}
int mid = (lo + hi) / 2;
root = Insert(root, (string)tokens[mid], vals[mid], 0);
BalancedTree(tokens, vals, lo, mid - 1, root);
BalancedTree(tokens, vals, mid + 1, hi, root);
}
public override void Build(IInputIterator tfit)
{
if (tfit.HasPayloads)
{
throw new System.ArgumentException("this suggester doesn't support payloads");
}
if (tfit.HasContexts)
{
throw new System.ArgumentException("this suggester doesn't support contexts");
}
root = new TernaryTreeNode();
// buffer first
#pragma warning disable 612, 618
if (tfit.Comparator != BytesRef.UTF8SortedAsUTF16Comparer)
{
// make sure it's sorted and the comparator uses UTF16 sort order
tfit = new SortedInputIterator(tfit, BytesRef.UTF8SortedAsUTF16Comparer);
}
#pragma warning restore 612, 618
List<string> tokens = new List<string>();
List<object> vals = new List<object>(); // LUCENENET TODO: Should this be long? in Java it was Number, but we can probably do better than object
BytesRef spare;
CharsRef charsSpare = new CharsRef();
while ((spare = tfit.Next()) != null)
{
charsSpare.Grow(spare.Length);
UnicodeUtil.UTF8toUTF16(spare.Bytes, spare.Offset, spare.Length, charsSpare);
tokens.Add(charsSpare.ToString());
vals.Add(Convert.ToInt64(tfit.Weight));
}
autocomplete.BalancedTree(tokens.ToArray(), vals.ToArray(), 0, tokens.Count - 1, root);
}
/// <summary>
/// Auto-completes a given prefix query using Depth-First Search with the end
/// of prefix as source node each time finding a new leaf to get a complete key
/// to be added in the suggest list.
/// </summary>
/// <param name="root">
/// a reference to root node of TST. </param>
/// <param name="s">
/// prefix query to be auto-completed. </param>
/// <param name="x">
/// index of current character to be searched while traversing through
/// the prefix in TST. </param>
/// <returns> suggest list of auto-completed keys for the given prefix query. </returns>
public virtual List<TernaryTreeNode> PrefixCompletion(TernaryTreeNode root, string s, int x)
{
TernaryTreeNode p = root;
List<TernaryTreeNode> suggest = new List<TernaryTreeNode>();
while (p != null)
{
if (s.charAt(x) < p.splitchar)
{
p = p.loKid;
}
else if (s.charAt(x) == p.splitchar)
{
if (x == s.Length - 1)
{
break;
}
else
{
x++;
}
p = p.eqKid;
}
else
{
p = p.hiKid;
}
}
if (p == null)
{
return suggest;
}
if (p.eqKid == null && p.token == null)
{
return suggest;
}
if (p.eqKid == null && p.token != null)
{
suggest.Add(p);
return suggest;
}
if (p.token != null)
{
suggest.Add(p);
}
p = p.eqKid;
var st = new Stack<TernaryTreeNode>();
st.Push(p);
while (st.Count > 0)
{
TernaryTreeNode top = st.Peek();
st.Pop();
if (top.token != null)
{
suggest.Add(top);
}
if (top.eqKid != null)
{
st.Push(top.eqKid);
}
if (top.loKid != null)
{
st.Push(top.loKid);
}
if (top.hiKid != null)
{
st.Push(top.hiKid);
}
}
return suggest;
}