[백업][가리사니] 솔라/루씬 아리랑 한글 형태소 6.2.x 버전 부터 오류 해결법
java, lucene, solr

이 문서는 가리사니 개발자 포럼에 올렸던 글의 백업 파일입니다. 오래된 문서가 많아 현재 상황과 맞지 않을 수 있습니다.

루씬 한글분석기 커뮤니티 (이하 주소) 에서 제공하는 한글형태소가 솔라/루씬 6.2.x 버전부터 오류가 발생하여 적습니다. http://cafe.naver.com/korlucene

문제

6.2.0 버전부터 java.lang.NoClassDefFoundError: org/apache/lucene/analysis/util/CharacterUtils 오류가 납니다. 원인 : CharacterUtils 가 이동됨과 동시에 기능도 조금 바뀌었습니다.

해결법

  1. http://cafe.naver.com/korlucene/1305 에서 프로젝트를 다운받습니다.
  2. http://cafe.naver.com/korlucene/1304 에서 arirang-morph-1.0.3.jar 를 다운받습니다.
  3. 위에서 받은 프로젝트의 lib 폴더를 생성 후 arirang-morph-1.0.3.jar 을 넣습니다. pom.xml 수정
  4. 아래 디펜던시를 추가해주세요. ``` java
com.argo arirang-morph 1.0.3 system ${project.basedir}/lib/arirang-morph-1.0.3.jar
5. 아래 디펜던시를 제거해주세요.
원래 써있는 버전이 더 높은 버전이지만.. https://github.com/soomyung 에도 해당 버전을 찾을 수 없습니다.
파일로 구할 수 있는 것은 현재 1.0.3이 최신인 것 같습니다.
``` java
<dependency>
	<groupId>com.argo</groupId>
	<artifactId>arirang-morph</artifactId>
	<version>1.1.0</version>
</dependency>
  1. 아래 디펜던시의 버전을 바꿔주세요. ``` java
org.apache.lucene lucene-core 6.2.1 org.apache.lucene lucene-analyzers-common 6.2.1
7. org.apache.lucene.analysis.ko.KoreanTokenizer 의 소스코드를 고쳐야합니다.
7-1. CharacterUtils 의 경로와 소스가 바뀌었습니다.
- 기존 : org.apache.lucene.analysis.util.CharacterUtils
- 현재 : org.apache.lucene.analysis.CharacterUtils (6.2 이후)
7-2. CharacterUtils 는 더 이상 인스턴스로 만들 수 없으며, 오직 static으로 써야합니다.
- ~~4-3. CharacterUtils 는 더 이상 codePointAt가 지원되지 않기 때문에 codePointAt를 구현 해야합니다.~~
- http://cafe.naver.com/korlucene/1320 에 댓글을 달아주신 netman2k님의 소스를 보고 Character.codePointAt 가 있다는 것을 알았습니다.
7-3. CharacterUtils.codePointAt (존재하지 않는 메서드)를 Character.codePointAt 로 교체합니다.
8. 이렇게 모두 수정해주셨다면 mvn clear install 로 리빌드하여 기존의 arirang.lucene-analyzer-5.0-1.1.0.jar 를 교체합니다.

http://cafe.naver.com/korlucene/1320 에 수정한 jar 파일 올려두었습니다.


# 추신 - CharacterUtils 변경점
기존 : org.apache.lucene.analysis.util.CharacterUtils
현재 : org.apache.lucene.analysis.CharacterUtils (6.2 이후)
기존코드
``` java
package org.apache.lucene.analysis.util;

import java.io.IOException;
import java.io.Reader;

public abstract class CharacterUtils
{
  private static final Java4CharacterUtils JAVA_4 = new Java4CharacterUtils();
  private static final Java5CharacterUtils JAVA_5 = new Java5CharacterUtils();

  public static CharacterUtils getInstance()
  {
    return JAVA_5;
  }

  @Deprecated
  public static CharacterUtils getJava4Instance()
  {
    return JAVA_4;
  }

  public abstract int codePointAt(CharSequence paramCharSequence, int paramInt);

  public abstract int codePointAt(char[] paramArrayOfChar, int paramInt1, int paramInt2);

  public abstract int codePointCount(CharSequence paramCharSequence);

  public static CharacterBuffer newCharacterBuffer(int bufferSize)
  {
    if (bufferSize < 2) {
      throw new IllegalArgumentException("buffersize must be >= 2");
    }
    return new CharacterBuffer(new char[bufferSize], 0, 0);
  }

  public final void toLowerCase(char[] buffer, int offset, int limit)
  {
    assert (buffer.length >= limit);
    assert ((offset <= 0) && (offset <= buffer.length));
    for (int i = offset; i < limit;) {
      i += Character.toChars(
        Character.toLowerCase(
        codePointAt(buffer, i, limit)), buffer, i);
    }
  }

  public final void toUpperCase(char[] buffer, int offset, int limit)
  {
    assert (buffer.length >= limit);
    assert ((offset <= 0) && (offset <= buffer.length));
    for (int i = offset; i < limit;) {
      i += Character.toChars(
        Character.toUpperCase(
        codePointAt(buffer, i, limit)), buffer, i);
    }
  }

  public final int toCodePoints(char[] src, int srcOff, int srcLen, int[] dest, int destOff)
  {
    if (srcLen < 0) {
      throw new IllegalArgumentException("srcLen must be >= 0");
    }
    int codePointCount = 0;
    for (int i = 0; i < srcLen;)
    {
      int cp = codePointAt(src, srcOff + i, srcOff + srcLen);
      int charCount = Character.charCount(cp);
      dest[(destOff + codePointCount++)] = cp;
      i += charCount;
    }
    return codePointCount;
  }

  public final int toChars(int[] src, int srcOff, int srcLen, char[] dest, int destOff)
  {
    if (srcLen < 0) {
      throw new IllegalArgumentException("srcLen must be >= 0");
    }
    int written = 0;
    for (int i = 0; i < srcLen; i++) {
      written += Character.toChars(src[(srcOff + i)], dest, destOff + written);
    }
    return written;
  }

  public abstract boolean fill(CharacterBuffer paramCharacterBuffer, Reader paramReader, int paramInt)
    throws IOException;

  public final boolean fill(CharacterBuffer buffer, Reader reader)
    throws IOException
  {
    return fill(buffer, reader, buffer.buffer.length);
  }

  public abstract int offsetByCodePoints(char[] paramArrayOfChar, int paramInt1, int paramInt2, int paramInt3, int paramInt4);

  static int readFully(Reader reader, char[] dest, int offset, int len)
    throws IOException
  {
    int read = 0;
    while (read < len)
    {
      int r = reader.read(dest, offset + read, len - read);
      if (r == -1) {
        break;
      }
      read += r;
    }
    return read;
  }

  private static final class Java5CharacterUtils
    extends CharacterUtils
  {
    public int codePointAt(CharSequence seq, int offset)
    {
      return Character.codePointAt(seq, offset);
    }

    public int codePointAt(char[] chars, int offset, int limit)
    {
      return Character.codePointAt(chars, offset, limit);
    }

    public boolean fill(CharacterUtils.CharacterBuffer buffer, Reader reader, int numChars)
      throws IOException
    {
      assert (buffer.buffer.length >= 2);
      if ((numChars < 2) || (numChars > buffer.buffer.length)) {
        throw new IllegalArgumentException("numChars must be >= 2 and <= the buffer size");
      }
      char[] charBuffer = buffer.buffer;
      buffer.offset = 0;
      int offset;
      int offset;
      if (buffer.lastTrailingHighSurrogate != 0)
      {
        charBuffer[0] = buffer.lastTrailingHighSurrogate;
        buffer.lastTrailingHighSurrogate = '\000';
        offset = 1;
      }
      else
      {
        offset = 0;
      }
      int read = readFully(reader, charBuffer, offset, numChars - offset);

      buffer.length = (offset + read);
      boolean result = buffer.length == numChars;
      if (buffer.length < numChars) {
        return result;
      }
      if (Character.isHighSurrogate(charBuffer[(buffer.length - 1)])) {
        buffer.lastTrailingHighSurrogate = charBuffer[CharacterUtils.CharacterBuffer.access$206(buffer)];
      }
      return result;
    }

    public int codePointCount(CharSequence seq)
    {
      return Character.codePointCount(seq, 0, seq.length());
    }

    public int offsetByCodePoints(char[] buf, int start, int count, int index, int offset)
    {
      return Character.offsetByCodePoints(buf, start, count, index, offset);
    }
  }

  private static final class Java4CharacterUtils
    extends CharacterUtils
  {
    public int codePointAt(CharSequence seq, int offset)
    {
      return seq.charAt(offset);
    }

    public int codePointAt(char[] chars, int offset, int limit)
    {
      if (offset >= limit) {
        throw new IndexOutOfBoundsException("offset must be less than limit");
      }
      return chars[offset];
    }

    public boolean fill(CharacterUtils.CharacterBuffer buffer, Reader reader, int numChars)
      throws IOException
    {
      assert (buffer.buffer.length >= 1);
      if ((numChars < 1) || (numChars > buffer.buffer.length)) {
        throw new IllegalArgumentException("numChars must be >= 1 and <= the buffer size");
      }
      buffer.offset = 0;
      int read = readFully(reader, buffer.buffer, 0, numChars);
      buffer.length = read;
      buffer.lastTrailingHighSurrogate = '\000';
      return read == numChars;
    }

    public int codePointCount(CharSequence seq)
    {
      return seq.length();
    }

    public int offsetByCodePoints(char[] buf, int start, int count, int index, int offset)
    {
      int result = index + offset;
      if ((result < 0) || (result > count)) {
        throw new IndexOutOfBoundsException();
      }
      return result;
    }
  }

  public static final class CharacterBuffer
  {
    private final char[] buffer;
    private int offset;
    private int length;
    char lastTrailingHighSurrogate;

    CharacterBuffer(char[] buffer, int offset, int length)
    {
      this.buffer = buffer;
      this.offset = offset;
      this.length = length;
    }

    public char[] getBuffer()
    {
      return this.buffer;
    }

    public int getOffset()
    {
      return this.offset;
    }

    public int getLength()
    {
      return this.length;
    }

    public void reset()
    {
      this.offset = 0;
      this.length = 0;
      this.lastTrailingHighSurrogate = '\000';
    }
  }
}

6.2 이후 코드

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.lucene.analysis;


import java.io.IOException;
import java.io.Reader;

/**
 * Utility class to write tokenizers or token filters.
 * @lucene.internal
 */
public final class CharacterUtils {

  private CharacterUtils() {} // no instantiation

  /**
   * Creates a new {@link CharacterBuffer} and allocates a <code>char[]</code>
   * of the given bufferSize.
   *
   * @param bufferSize
   *          the internal char buffer size, must be <code>&gt;= 2</code>
   * @return a new {@link CharacterBuffer} instance.
   */
  public static CharacterBuffer newCharacterBuffer(final int bufferSize) {
    if (bufferSize < 2) {
      throw new IllegalArgumentException("buffersize must be >= 2");
    }
    return new CharacterBuffer(new char[bufferSize], 0, 0);
  }


  /**
   * Converts each unicode codepoint to lowerCase via {@link Character#toLowerCase(int)} starting
   * at the given offset.
   * @param buffer the char buffer to lowercase
   * @param offset the offset to start at
   * @param limit the max char in the buffer to lower case
   */
  public static void toLowerCase(final char[] buffer, final int offset, final int limit) {
    assert buffer.length >= limit;
    assert offset <=0 && offset <= buffer.length;
    for (int i = offset; i < limit;) {
      i += Character.toChars(
              Character.toLowerCase(
                  Character.codePointAt(buffer, i, limit)), buffer, i);
     }
  }

  /**
   * Converts each unicode codepoint to UpperCase via {@link Character#toUpperCase(int)} starting
   * at the given offset.
   * @param buffer the char buffer to UPPERCASE
   * @param offset the offset to start at
   * @param limit the max char in the buffer to lower case
   */
  public static void toUpperCase(final char[] buffer, final int offset, final int limit) {
    assert buffer.length >= limit;
    assert offset <=0 && offset <= buffer.length;
    for (int i = offset; i < limit;) {
      i += Character.toChars(
              Character.toUpperCase(
                  Character.codePointAt(buffer, i, limit)), buffer, i);
     }
  }

  /** Converts a sequence of Java characters to a sequence of unicode code points.
   *  @return the number of code points written to the destination buffer */
  public static int toCodePoints(char[] src, int srcOff, int srcLen, int[] dest, int destOff) {
    if (srcLen < 0) {
      throw new IllegalArgumentException("srcLen must be >= 0");
    }
    int codePointCount = 0;
    for (int i = 0; i < srcLen; ) {
      final int cp = Character.codePointAt(src, srcOff + i, srcOff + srcLen);
      final int charCount = Character.charCount(cp);
      dest[destOff + codePointCount++] = cp;
      i += charCount;
    }
    return codePointCount;
  }

  /** Converts a sequence of unicode code points to a sequence of Java characters.
   *  @return the number of chars written to the destination buffer */
  public static int toChars(int[] src, int srcOff, int srcLen, char[] dest, int destOff) {
    if (srcLen < 0) {
      throw new IllegalArgumentException("srcLen must be >= 0");
    }
    int written = 0;
    for (int i = 0; i < srcLen; ++i) {
      written += Character.toChars(src[srcOff + i], dest, destOff + written);
    }
    return written;
  }

  /**
   * Fills the {@link CharacterBuffer} with characters read from the given
   * reader {@link Reader}. This method tries to read <code>numChars</code>
   * characters into the {@link CharacterBuffer}, each call to fill will start
   * filling the buffer from offset <code>0</code> up to <code>numChars</code>.
   * In case code points can span across 2 java characters, this method may
   * only fill <code>numChars - 1</code> characters in order not to split in
   * the middle of a surrogate pair, even if there are remaining characters in
   * the {@link Reader}.
   * <p>
   * This method guarantees
   * that the given {@link CharacterBuffer} will never contain a high surrogate
   * character as the last element in the buffer unless it is the last available
   * character in the reader. In other words, high and low surrogate pairs will
   * always be preserved across buffer boarders.
   * </p>
   * <p>
   * A return value of <code>false</code> means that this method call exhausted
   * the reader, but there may be some bytes which have been read, which can be
   * verified by checking whether <code>buffer.getLength() &gt; 0</code>.
   * </p>
   *
   * @param buffer
   *          the buffer to fill.
   * @param reader
   *          the reader to read characters from.
   * @param numChars
   *          the number of chars to read
   * @return <code>false</code> if and only if reader.read returned -1 while trying to fill the buffer
   * @throws IOException
   *           if the reader throws an {@link IOException}.
   */
  public static boolean fill(CharacterBuffer buffer, Reader reader, int numChars) throws IOException {
    assert buffer.buffer.length >= 2;
    if (numChars < 2 || numChars > buffer.buffer.length) {
      throw new IllegalArgumentException("numChars must be >= 2 and <= the buffer size");
    }
    final char[] charBuffer = buffer.buffer;
    buffer.offset = 0;
    final int offset;

    // Install the previously saved ending high surrogate:
    if (buffer.lastTrailingHighSurrogate != 0) {
      charBuffer[0] = buffer.lastTrailingHighSurrogate;
      buffer.lastTrailingHighSurrogate = 0;
      offset = 1;
    } else {
      offset = 0;
    }

    final int read = readFully(reader, charBuffer, offset, numChars - offset);

    buffer.length = offset + read;
    final boolean result = buffer.length == numChars;
    if (buffer.length < numChars) {
      // We failed to fill the buffer. Even if the last char is a high
      // surrogate, there is nothing we can do
      return result;
    }

    if (Character.isHighSurrogate(charBuffer[buffer.length - 1])) {
      buffer.lastTrailingHighSurrogate = charBuffer[--buffer.length];
    }
    return result;
  }

  /** Convenience method which calls <code>fill(buffer, reader, buffer.buffer.length)</code>. */
  public static boolean fill(CharacterBuffer buffer, Reader reader) throws IOException {
    return fill(buffer, reader, buffer.buffer.length);
  }

  static int readFully(Reader reader, char[] dest, int offset, int len) throws IOException {
    int read = 0;
    while (read < len) {
      final int r = reader.read(dest, offset + read, len - read);
      if (r == -1) {
        break;
      }
      read += r;
    }
    return read;
  }

  /**
   * A simple IO buffer to use with
   * {@link CharacterUtils#fill(CharacterBuffer, Reader)}.
   */
  public static final class CharacterBuffer {

    private final char[] buffer;
    private int offset;
    private int length;
    // NOTE: not private so outer class can access without
    // $access methods:
    char lastTrailingHighSurrogate;

    CharacterBuffer(char[] buffer, int offset, int length) {
      this.buffer = buffer;
      this.offset = offset;
      this.length = length;
    }

    /**
     * Returns the internal buffer
     *
     * @return the buffer
     */
    public char[] getBuffer() {
      return buffer;
    }

    /**
     * Returns the data offset in the internal buffer.
     *
     * @return the offset
     */
    public int getOffset() {
      return offset;
    }

    /**
     * Return the length of the data in the internal buffer starting at
     * {@link #getOffset()}
     *
     * @return the length
     */
    public int getLength() {
      return length;
    }

    /**
     * Resets the CharacterBuffer. All internals are reset to its default
     * values.
     */
    public void reset() {
      offset = 0;
      length = 0;
      lastTrailingHighSurrogate = 0;
    }
  }

}