X-Git-Url: https://git.mdrn.pl/pylucene.git/blobdiff_plain/a2e61f0c04805cfcb8706176758d1283c7e3a55c..aaeed5504b982cf3545252ab528713250aa33eed:/lucene-java-3.5.0/lucene/contrib/analyzers/common/src/java/org/apache/lucene/analysis/util/CharArrayIterator.java

diff --git a/lucene-java-3.5.0/lucene/contrib/analyzers/common/src/java/org/apache/lucene/analysis/util/CharArrayIterator.java b/lucene-java-3.5.0/lucene/contrib/analyzers/common/src/java/org/apache/lucene/analysis/util/CharArrayIterator.java
new file mode 100644
index 0000000..4d14479
--- /dev/null
+++ b/lucene-java-3.5.0/lucene/contrib/analyzers/common/src/java/org/apache/lucene/analysis/util/CharArrayIterator.java
@@ -0,0 +1,175 @@
+package org.apache.lucene.analysis.util;
+
+import java.text.BreakIterator; // javadoc
+import java.text.CharacterIterator;
+import java.util.Locale;
+
+/** 
+ * A CharacterIterator used internally for use with {@link BreakIterator}
+ * @lucene.internal
+ */
+public abstract class CharArrayIterator implements CharacterIterator {
+  private char array[];
+  private int start;
+  private int index;
+  private int length;
+  private int limit;
+
+  public char [] getText() {
+    return array;
+  }
+  
+  public int getStart() {
+    return start;
+  }
+  
+  public int getLength() {
+    return length;
+  }
+  
+  /**
+   * Set a new region of text to be examined by this iterator
+   * 
+   * @param array text buffer to examine
+   * @param start offset into buffer
+   * @param length maximum length to examine
+   */
+  public void setText(final char array[], int start, int length) {
+    this.array = array;
+    this.start = start;
+    this.index = start;
+    this.length = length;
+    this.limit = start + length;
+  }
+
+  public char current() {
+    return (index == limit) ? DONE : jreBugWorkaround(array[index]);
+  }
+  
+  protected abstract char jreBugWorkaround(char ch);
+
+  public char first() {
+    index = start;
+    return current();
+  }
+
+  public int getBeginIndex() {
+    return 0;
+  }
+
+  public int getEndIndex() {
+    return length;
+  }
+
+  public int getIndex() {
+    return index - start;
+  }
+
+  public char last() {
+    index = (limit == start) ? limit : limit - 1;
+    return current();
+  }
+
+  public char next() {
+    if (++index >= limit) {
+      index = limit;
+      return DONE;
+    } else {
+      return current();
+    }
+  }
+
+  public char previous() {
+    if (--index < start) {
+      index = start;
+      return DONE;
+    } else {
+      return current();
+    }
+  }
+
+  public char setIndex(int position) {
+    if (position < getBeginIndex() || position > getEndIndex())
+      throw new IllegalArgumentException("Illegal Position: " + position);
+    index = start + position;
+    return current();
+  }
+  
+  @Override
+  public Object clone() {
+    try {
+      return super.clone();
+    } catch (CloneNotSupportedException e) {
+      // CharacterIterator does not allow you to throw CloneNotSupported
+      throw new RuntimeException(e);
+    }
+  }
+  
+  /**
+   * Create a new CharArrayIterator that works around JRE bugs
+   * in a manner suitable for {@link BreakIterator#getSentenceInstance()}
+   */
+  public static CharArrayIterator newSentenceInstance() {
+    if (HAS_BUGGY_BREAKITERATORS) {
+      return new CharArrayIterator() {
+        // work around this for now by lying about all surrogates to 
+        // the sentence tokenizer, instead we treat them all as 
+        // SContinue so we won't break around them.
+        @Override
+        protected char jreBugWorkaround(char ch) {
+          return ch >= 0xD800 && ch <= 0xDFFF ? 0x002C : ch;
+        }
+      };
+    } else {
+      return new CharArrayIterator() {
+        // no bugs
+        @Override
+        protected char jreBugWorkaround(char ch) {
+          return ch;
+        }
+      };
+    }
+  }
+  
+  /**
+   * Create a new CharArrayIterator that works around JRE bugs
+   * in a manner suitable for {@link BreakIterator#getWordInstance()}
+   */
+  public static CharArrayIterator newWordInstance() {
+    if (HAS_BUGGY_BREAKITERATORS) {
+      return new CharArrayIterator() {
+        // work around this for now by lying about all surrogates to the word, 
+        // instead we treat them all as ALetter so we won't break around them.
+        @Override
+        protected char jreBugWorkaround(char ch) {
+          return ch >= 0xD800 && ch <= 0xDFFF ? 0x0041 : ch;
+        }
+      };
+    } else {
+      return new CharArrayIterator() {
+        // no bugs
+        @Override
+        protected char jreBugWorkaround(char ch) {
+          return ch;
+        }
+      };
+    }
+  }
+  
+  /**
+   * True if this JRE has a buggy BreakIterator implementation
+   */
+  public static final boolean HAS_BUGGY_BREAKITERATORS;
+  static {
+    boolean v;
+    try {
+      BreakIterator bi = BreakIterator.getSentenceInstance(Locale.US);
+      bi.setText("\udb40\udc53");
+      bi.next();
+      v = false;
+    } catch (Exception e) {
+      v = true;
+    }
+    HAS_BUGGY_BREAKITERATORS = v;
+  }
+}