Implement Trie Filter for special words

pom.xml

@@ -1,66 +1,30 @@
-<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0                       http://maven.apache.org/xsd/maven-4.0.0.xsd">
-  <modelVersion>4.0.0</modelVersion>
-
-  <properties>
-    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
-    <scm.uri>scm:git:[email protected]:rkapsi/patricia-trie.git</scm.uri>
-  </properties>
-
-  <groupId>org.ardverk</groupId>
-  <artifactId>patricia-trie</artifactId>
-  <version>0.7-SNAPSHOT</version>
-  <packaging>jar</packaging>
-  <url>https://github.com/rkapsi/patricia-trie</url>
-
-  <distributionManagement>
-    <repository>
-      <id>ardverk-release</id>
-      <url>scp://mvn.ardverk.org/repository/release</url>
-    </repository>
-    <snapshotRepository>
-      <id>ardverk-snapshot</id>
-      <url>scp://mvn.ardverk.org/repository/snapshot</url>
-    </snapshotRepository>
-  </distributionManagement>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+	<modelVersion>4.0.0</modelVersion>
 
-  <scm>
-    <connection>${scm.uri}</connection>
-    <developerConnection>${scm.uri}</developerConnection>
-    <url>${project.url}</url>
-  </scm>
-
-  <build>
-    <extensions>
-      <extension>
-        <groupId>org.apache.maven.wagon</groupId>
-        <artifactId>wagon-ssh</artifactId>
-        <version>1.0</version>
-      </extension>
-    </extensions>
-
-    <plugins>
-      <plugin>
-        <groupId>org.apache.maven.plugins</groupId>
-        <artifactId>maven-compiler-plugin</artifactId>
-        <configuration>
-            <source>1.6</source>
-            <target>1.6</target>
-        </configuration>
-      </plugin>
-
-      <plugin>
-        <groupId>org.apache.maven.plugins</groupId>
-        <artifactId>maven-release-plugin</artifactId>
-      </plugin>
-    </plugins>
-  </build>
-
-  <dependencies>
-    <dependency>
-      <groupId>junit</groupId>
-      <artifactId>junit</artifactId>
-      <version>[4.8.2,)</version>
-      <scope>test</scope>
-    </dependency>
-  </dependencies>
-</project>
+	<groupId>com.kim</groupId>
+	<artifactId>patricia-trie</artifactId>
+	<version>0.0.1-SNAPSHOT</version>
+	<packaging>jar</packaging>
+
+	<name>patricia-trie</name>
+	<url>http://maven.apache.org</url>
+
+	<properties>
+		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
+	</properties>
+
+	<dependencies>
+		<dependency>
+			<groupId>junit</groupId>
+			<artifactId>junit</artifactId>
+			<version>4.7</version>
+			<scope>test</scope>
+		</dependency>
+		<dependency>
+			<groupId>commons-io</groupId>
+			<artifactId>commons-io</artifactId>
+			<version>2.4</version>
+			<scope>test</scope>
+		</dependency>
+	</dependencies>
+</project>

src/main/java/org/ardverk/filter/TrieCounter.java

@@ -0,0 +1,11 @@
+package org.ardverk.filter;
+
+/**
+ * @author kim 2014年9月2日
+ */
+public interface TrieCounter {
+
+	public String source();
+
+	public int filtered();
+}

src/main/java/org/ardverk/filter/TrieFilter.java

@@ -0,0 +1,9 @@
+package org.ardverk.filter;
+
+/**
+ * @author kim 2014年9月2日
+ */
+public interface TrieFilter {
+
+	public TrieCounter filter(String source) throws Exception;
+}

src/main/java/org/ardverk/filter/impl/SimpleTrieFilter.java

@@ -0,0 +1,123 @@
+package org.ardverk.filter.impl;
+
+import java.util.Map;
+import java.util.concurrent.atomic.AtomicInteger;
+
+import org.ardverk.collection.PatriciaTrie;
+import org.ardverk.collection.StringKeyAnalyzer;
+import org.ardverk.filter.TrieCounter;
+import org.ardverk.filter.TrieFilter;
+
+/**
+ * @author kim 2014年9月2日
+ */
+public class SimpleTrieFilter implements TrieFilter {
+
+	private final static int ONE_STEP = 1;
+
+	private final PatriciaTrie<String, String> trie = new PatriciaTrie<String, String>(StringKeyAnalyzer.CHAR);
+
+	public SimpleTrieFilter(Map<String, String> properies) {
+		this.trie.putAll(properies);
+	}
+
+	public MergedTrieCounter filter(String source) throws Exception {
+		StringBuffer buffer = new StringBuffer(source);
+		Judgment judgment = new Judgment();
+		for (int index = 0; index < buffer.length(); index++) {
+			this.step(buffer, judgment, index);
+		}
+		return new MergedTrieCounter(buffer.toString(), judgment.count());
+	}
+
+	private void step(StringBuffer buffer, Judgment judgment, int index) {
+		String current = String.valueOf(buffer.charAt(index));
+		if (this.trie.select(current).getKey().startsWith(current)) {
+			for (int step = SimpleTrieFilter.ONE_STEP; step <= (buffer.length() - index); step++) {
+				String fragement = buffer.substring(index, index + step);
+				judgment.reset(this.trie.select(fragement), fragement);
+				if (judgment.same()) {
+					buffer.replace(index, index + step, judgment.replace());
+					return;
+				} else if (!judgment.challenge()) {
+					return;
+				}
+			}
+		}
+	}
+
+	protected class MergedTrieCounter implements TrieCounter {
+
+		private final StringBuffer buffer = new StringBuffer();
+
+		private final AtomicInteger counter = new AtomicInteger();
+
+		protected MergedTrieCounter(String source, int counter) {
+			this.buffer.append(source);
+			this.counter.addAndGet(counter);
+		}
+
+		public MergedTrieCounter merge(MergedTrieCounter counter) {
+			this.append(counter.source());
+			this.incr(counter.filtered());
+			return this;
+		}
+
+		public MergedTrieCounter incr() {
+			this.counter.incrementAndGet();
+			return this;
+		}
+
+		public MergedTrieCounter incr(int count) {
+			this.counter.addAndGet(count);
+			return this;
+		}
+
+		public MergedTrieCounter append(String buffer) {
+			this.buffer.append(buffer);
+			return this;
+		}
+
+		@Override
+		public String source() {
+			return this.buffer.toString();
+		}
+
+		@Override
+		public int filtered() {
+			return this.counter.get();
+		}
+	}
+
+	private class Judgment {
+
+		private final AtomicInteger counter = new AtomicInteger();
+
+		private Map.Entry<String, String> selected;
+
+		private String fragment;
+
+		private Judgment reset(Map.Entry<String, String> selected, String fragment) {
+			this.selected = selected;
+			this.fragment = fragment;
+			this.counter.addAndGet(this.same() ? 1 : 0);
+			return this;
+		}
+
+		public int count() {
+			return this.counter.get();
+		}
+
+		public String replace() {
+			return this.selected.getValue();
+		}
+
+		public boolean same() {
+			return this.selected.getKey().equals(this.fragment);
+		}
+
+		public boolean challenge() {
+			return this.selected.getKey().startsWith(this.fragment);
+		}
+	}
+}