/// <summary>
/// Optimize (remove empty rows) from the given Trie and return the resulting
/// Trie.
/// </summary>
/// <param name="orig">the <see cref="Trie"/> to consolidate</param>
/// <returns>the newly consolidated Trie</returns>
public override Trie Optimize(Trie orig)
{
IList<string> cmds = orig.cmds;
IList<Row> rows = new List<Row>();
IList<Row> orows = orig.rows;
int[] remap = new int[orows.Count];
for (int j = orows.Count - 1; j >= 0; j--)
{
Row now = new Remap(orows[j], remap);
bool merged = false;
for (int i = 0; i < rows.Count; i++)
{
Row q = Merge(now, rows[i]);
if (q != null)
{
rows[i] = q;
merged = true;
remap[j] = i;
break;
}
}
if (merged == false)
{
remap[j] = rows.Count;
rows.Add(now);
}
}
int root = remap[orig.root];
Arrays.Fill(remap, -1);
rows = RemoveGaps(root, rows, new List<Row>(), remap);
return new Trie(orig.forward, remap[root], cmds, rows);
}
/// <summary>
/// Create a Stemmer using pre-loaded stemmer table
/// </summary>
/// <param name="stemmer">pre-loaded stemmer table</param>
public StempelStemmer(Trie stemmer)
{
this.stemmer = stemmer;
}
/// <summary>
/// Entry point to the Compile application.
/// <para/>
/// This program takes any number of arguments: the first is the name of the
/// desired stemming algorithm to use (a list is available in the package
/// description) , all of the rest should be the path or paths to a file or
/// files containing a stemmer table to compile.
/// </summary>
/// <param name="args">the command line arguments</param>
public static void Main(string[] args)
{
if (args.Length < 1)
{
return;
}
// LUCENENET NOTE: This line does nothing in .NET
// and also does nothing in Java...what?
//args[0].ToUpperInvariant();
// Reads the first char of the first arg
backward = args[0][0] == '-';
int qq = (backward) ? 1 : 0;
bool storeorig = false;
if (args[0][qq] == '0')
{
storeorig = true;
qq++;
}
multi = args[0][qq] == 'M';
if (multi)
{
qq++;
}
string charset = SystemProperties.GetProperty("egothor.stemmer.charset", "UTF-8");
var stemmerTables = new List <string>();
// LUCENENET specific
// command line argument overrides environment variable or default, if supplied
for (int i = 1; i < args.Length; i++)
{
if ("-e".Equals(args[i]) || "--encoding".Equals(args[i]))
{
charset = args[i];
}
else
{
stemmerTables.Add(args[i]);
}
}
char[] optimizer = new char[args[0].Length - qq];
for (int i = 0; i < optimizer.Length; i++)
{
optimizer[i] = args[0][qq + i];
}
foreach (var stemmerTable in stemmerTables)
{
// System.out.println("[" + args[i] + "]");
Diff diff = new Diff();
//int stems = 0; // not used
int words = 0;
AllocTrie();
Console.WriteLine(stemmerTable);
using (TextReader input = new StreamReader(
new FileStream(stemmerTable, FileMode.Open, FileAccess.Read), Encoding.GetEncoding(charset)))
{
string line;
while ((line = input.ReadLine()) != null)
{
try
{
line = line.ToLowerInvariant();
StringTokenizer st = new StringTokenizer(line);
string stem = st.NextToken();
if (storeorig)
{
trie.Add(stem, "-a");
words++;
}
while (st.HasMoreTokens())
{
string token = st.NextToken();
if (token.Equals(stem) == false)
{
trie.Add(token, diff.Exec(token, stem));
words++;
}
}
}
catch (InvalidOperationException /*x*/)
{
// no base token (stem) on a line
}
}
}
Optimizer o = new Optimizer();
Optimizer2 o2 = new Optimizer2();
Lift l = new Lift(true);
Lift e = new Lift(false);
Gener g = new Gener();
for (int j = 0; j < optimizer.Length; j++)
{
string prefix;
switch (optimizer[j])
{
case 'G':
trie = trie.Reduce(g);
prefix = "G: ";
break;
case 'L':
trie = trie.Reduce(l);
prefix = "L: ";
break;
case 'E':
trie = trie.Reduce(e);
prefix = "E: ";
break;
case '2':
trie = trie.Reduce(o2);
prefix = "2: ";
break;
case '1':
trie = trie.Reduce(o);
prefix = "1: ";
break;
default:
continue;
}
trie.PrintInfo(Console.Out, prefix + " ");
}
using (DataOutputStream os = new DataOutputStream(
new FileStream(stemmerTable + ".out", FileMode.OpenOrCreate, FileAccess.Write)))
{
os.WriteUTF(args[0]);
trie.Store(os);
}
}
}
static DefaultsHolder()
{
try
{
DEFAULT_STOP_SET = WordlistLoader.GetWordSet(IOUtils.GetDecodingReader(typeof(PolishAnalyzer),
typeof(PolishAnalyzer).Namespace + "." + DEFAULT_STOPWORD_FILE, Encoding.UTF8), "#",
#pragma warning disable 612, 618
LuceneVersion.LUCENE_CURRENT);
#pragma warning restore 612, 618
}
catch (IOException ex)
{
// default set should always be present as it is part of the
// distribution (embedded resource)
throw new SystemException("Unable to load default stopword set", ex);
}
try
{
DEFAULT_TABLE = StempelStemmer.Load(typeof(PolishAnalyzer).Assembly.GetManifestResourceStream(
typeof(PolishAnalyzer).Namespace + "." + DEFAULT_STEMMER_FILE));
}
catch (IOException ex)
{
// default set should always be present as it is part of the
// distribution (embedded resource)
throw new SystemException("Unable to load default stemming tables", ex);
}
}
/// <summary>
/// Builds an analyzer with the given stop words. If a non-empty stem exclusion set is
/// provided this analyzer will add a <see cref="SetKeywordMarkerFilter"/> before
/// stemming.
/// </summary>
/// <param name="matchVersion">lucene compatibility version</param>
/// <param name="stopwords">a stopword set</param>
/// <param name="stemExclusionSet">a set of terms not to be stemmed</param>
public PolishAnalyzer(LuceneVersion matchVersion, CharArraySet stopwords, CharArraySet stemExclusionSet)
: base(matchVersion, stopwords)
{
this.stemTable = DefaultsHolder.DEFAULT_TABLE;
this.stemExclusionSet = CharArraySet.UnmodifiableSet(CharArraySet.Copy(
matchVersion, stemExclusionSet));
}
internal static Trie LoadTrie(string path)
{
Trie trie;
using (DataInputStream @is = new DataInputStream(
new FileStream(path, FileMode.Open, FileAccess.Read)))
{
string method = @is.ReadUTF().ToUpperInvariant();
if (method.IndexOf('M') < 0)
{
trie = new Trie(@is);
}
else
{
trie = new MultiTrie(@is);
}
}
return trie;
}
private static void AssertTrie(Trie trie, string file, bool usefull,
bool storeorig)
{
using (TextReader @in =
new StreamReader(new FileStream(file, FileMode.Open), Encoding.UTF8))
{
for (string line = @in.ReadLine(); line != null; line = @in.ReadLine())
{
try
{
line = line.ToLowerInvariant();
StringTokenizer st = new StringTokenizer(line);
string stem = st.NextToken();
if (storeorig)
{
string cmd = (usefull) ? trie.GetFully(stem) : trie
.GetLastOnPath(stem);
StringBuilder stm = new StringBuilder(stem);
Diff.Apply(stm, cmd);
assertEquals(stem.ToLowerInvariant(), stm.ToString().ToLowerInvariant());
}
while (st.HasMoreTokens())
{
string token = st.NextToken();
if (token.Equals(stem))
{
continue;
}
string cmd = (usefull) ? trie.GetFully(token) : trie
.GetLastOnPath(token);
StringBuilder stm = new StringBuilder(token);
Diff.Apply(stm, cmd);
assertEquals(stem.ToLowerInvariant(), stm.ToString().ToLowerInvariant());
}
}
catch (InvalidOperationException /*x*/)
{
// no base token (stem) on a line
}
}
}
}
internal static void AllocTrie()
{
if (multi)
{
trie = new MultiTrie2(!backward);
}
else
{
trie = new Trie(!backward);
}
}
/**
* Entry point to the Compile application.
* <p>
* This program takes any number of arguments: the first is the name of the
* desired stemming algorithm to use (a list is available in the package
* description) , all of the rest should be the path or paths to a file or
* files containing a stemmer table to compile.
*
* @param args the command line arguments
*/
public static void Main(string[] args)
{
if (args.Length < 1)
{
return;
}
args[0].ToUpperInvariant();
backward = args[0][0] == '-';
int qq = (backward) ? 1 : 0;
bool storeorig = false;
if (args[0][qq] == '0')
{
storeorig = true;
qq++;
}
multi = args[0][qq] == 'M';
if (multi)
{
qq++;
}
// LUCENENET TODO: Is this any different than Encoding.UTF8?
//String charset = System.getProperty("egothor.stemmer.charset", "UTF-8");
char[] optimizer = new char[args[0].Length - qq];
for (int i = 0; i < optimizer.Length; i++)
{
optimizer[i] = args[0][qq + i];
}
for (int i = 1; i < args.Length; i++)
{
TextReader @in;
// System.out.println("[" + args[i] + "]");
Diff diff = new Diff();
//int stems = 0; // not used
int words = 0;
AllocTrie();
Console.WriteLine(args[i]);
using (@in = new StreamReader(
new FileStream(args[i], FileMode.Open, FileAccess.Read), Encoding.UTF8))
{
for (string line = @in.ReadLine(); line != null; line = @in.ReadLine())
{
try
{
line = line.ToLowerInvariant();
StringTokenizer st = new StringTokenizer(line);
string stem = st.NextToken();
if (storeorig)
{
trie.Add(stem, "-a");
words++;
}
while (st.HasMoreTokens())
{
string token = st.NextToken();
if (token.Equals(stem) == false)
{
trie.Add(token, diff.Exec(token, stem));
words++;
}
}
}
catch (InvalidOperationException /*x*/)
{
// no base token (stem) on a line
}
}
}
Optimizer o = new Optimizer();
Optimizer2 o2 = new Optimizer2();
Lift l = new Lift(true);
Lift e = new Lift(false);
Gener g = new Gener();
for (int j = 0; j < optimizer.Length; j++)
{
string prefix;
switch (optimizer[j])
{
case 'G':
trie = trie.Reduce(g);
prefix = "G: ";
break;
case 'L':
trie = trie.Reduce(l);
prefix = "L: ";
break;
case 'E':
trie = trie.Reduce(e);
prefix = "E: ";
break;
case '2':
trie = trie.Reduce(o2);
prefix = "2: ";
break;
case '1':
trie = trie.Reduce(o);
prefix = "1: ";
break;
default:
continue;
}
trie.PrintInfo(System.Console.Out, prefix + " ");
}
using (DataOutputStream os = new DataOutputStream(
new FileStream(args[i] + ".out", FileMode.OpenOrCreate, FileAccess.Write)))
{
os.WriteUTF(args[0]);
trie.Store(os);
}
}
}
private static void AssertTrieContents(Trie trie, string[] keys, string[] vals)
{
Trie[] tries = new Trie[] {
trie,
trie.Reduce(new Optimizer()),
trie.Reduce(new Optimizer2()),
trie.Reduce(new Gener()),
trie.Reduce(new Lift(true)),
trie.Reduce(new Lift(false))
};
foreach (Trie t in tries)
{
for (int i = 0; i < keys.Length; i++)
{
assertEquals(vals[i], t.GetFully(keys[i]).ToString());
assertEquals(vals[i], t.GetLastOnPath(keys[i]).ToString());
}
}
}
public void TestTrieBackwards()
{
Trie t = new Trie(false);
string[] keys = { "a", "ba", "bb", "c" };
string[] vals = { "1", "2", "2", "4" };
for (int i = 0; i < keys.Length; i++)
{
t.Add(keys[i], vals[i]);
}
AssertTrieContents(t, keys, vals);
}
public void TestTrie()
{
Trie t = new Trie(true);
string[] keys = { "a", "ba", "bb", "c" };
string[] vals = { "1", "2", "2", "4" };
for (int i = 0; i < keys.Length; i++)
{
t.Add(keys[i], vals[i]);
}
assertEquals(0, t.root);
assertEquals(2, t.rows.Count);
assertEquals(3, t.cmds.Count);
AssertTrieContents(t, keys, vals);
}
/// <summary>
/// Return a Trie with infrequent values occurring in the given Trie removed.
/// </summary>
/// <param name="orig">the Trie to optimize</param>
/// <returns>a new optimized Trie</returns>
public override Trie Optimize(Trie orig)
{
IList<string> cmds = orig.cmds;
IList<Row> rows = new List<Row>();
IList<Row> orows = orig.rows;
int[] remap = new int[orows.Count];
Arrays.Fill(remap, 1);
for (int j = orows.Count - 1; j >= 0; j--)
{
if (Eat(orows[j], remap))
{
remap[j] = 0;
}
}
Arrays.Fill(remap, -1);
rows = RemoveGaps(orig.root, orows, new List<Row>(), remap);
return new Trie(orig.forward, remap[orig.root], cmds, rows);
}
/**
* Entry point to the Compile application.
* <p>
* This program takes any number of arguments: the first is the name of the
* desired stemming algorithm to use (a list is available in the package
* description) , all of the rest should be the path or paths to a file or
* files containing a stemmer table to compile.
*
* @param args the command line arguments
*/
public static void Main(string[] args)
{
if (args.Length < 1)
{
return;
}
args[0].ToUpperInvariant();
backward = args[0][0] == '-';
int qq = (backward) ? 1 : 0;
bool storeorig = false;
if (args[0][qq] == '0')
{
storeorig = true;
qq++;
}
multi = args[0][qq] == 'M';
if (multi)
{
qq++;
}
// LUCENENET TODO: Is this any different than Encoding.UTF8?
//String charset = System.getProperty("egothor.stemmer.charset", "UTF-8");
char[] optimizer = new char[args[0].Length - qq];
for (int i = 0; i < optimizer.Length; i++)
{
optimizer[i] = args[0][qq + i];
}
for (int i = 1; i < args.Length; i++)
{
TextReader @in;
// System.out.println("[" + args[i] + "]");
Diff diff = new Diff();
//int stems = 0; // not used
int words = 0;
AllocTrie();
Console.WriteLine(args[i]);
using (@in = new StreamReader(
new FileStream(args[i], FileMode.Open, FileAccess.Read), Encoding.UTF8))
{
for (string line = @in.ReadLine(); line != null; line = @in.ReadLine())
{
try
{
line = line.ToLowerInvariant();
StringTokenizer st = new StringTokenizer(line);
string stem = st.NextToken();
if (storeorig)
{
trie.Add(stem, "-a");
words++;
}
while (st.HasMoreTokens())
{
string token = st.NextToken();
if (token.Equals(stem) == false)
{
trie.Add(token, diff.Exec(token, stem));
words++;
}
}
}
catch (InvalidOperationException /*x*/)
{
// no base token (stem) on a line
}
}
}
Optimizer o = new Optimizer();
Optimizer2 o2 = new Optimizer2();
Lift l = new Lift(true);
Lift e = new Lift(false);
Gener g = new Gener();
for (int j = 0; j < optimizer.Length; j++)
{
string prefix;
switch (optimizer[j])
{
case 'G':
trie = trie.Reduce(g);
prefix = "G: ";
break;
case 'L':
trie = trie.Reduce(l);
prefix = "L: ";
break;
case 'E':
trie = trie.Reduce(e);
prefix = "E: ";
break;
case '2':
trie = trie.Reduce(o2);
prefix = "2: ";
break;
case '1':
trie = trie.Reduce(o);
prefix = "1: ";
break;
default:
continue;
}
trie.PrintInfo(System.Console.Out, prefix + " ");
}
using (DataOutputStream os = new DataOutputStream(
new FileStream(args[i] + ".out", FileMode.OpenOrCreate, FileAccess.Write)))
{
os.WriteUTF(args[0]);
trie.Store(os);
}
}
}
