Save This Page
Home » lucene-2.4.1-src » org.apache » lucene » index » [javadoc | source]
    1   package org.apache.lucene.index;
    2   
    3   /**
    4    * Licensed to the Apache Software Foundation (ASF) under one or more
    5    * contributor license agreements.  See the NOTICE file distributed with
    6    * this work for additional information regarding copyright ownership.
    7    * The ASF licenses this file to You under the Apache License, Version 2.0
    8    * (the "License"); you may not use this file except in compliance with
    9    * the License.  You may obtain a copy of the License at
   10    *
   11    *     http://www.apache.org/licenses/LICENSE-2.0
   12    *
   13    * Unless required by applicable law or agreed to in writing, software
   14    * distributed under the License is distributed on an "AS IS" BASIS,
   15    * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
   16    * See the License for the specific language governing permissions and
   17    * limitations under the License.
   18    */
   19   
   20   import java.io.IOException;
   21   import org.apache.lucene.store.IndexInput;
   22   
   23   final class SegmentTermEnum extends TermEnum implements Cloneable {
   24     private IndexInput input;
   25     FieldInfos fieldInfos;
   26     long size;
   27     long position = -1;
   28   
   29     private TermBuffer termBuffer = new TermBuffer();
   30     private TermBuffer prevBuffer = new TermBuffer();
   31     private TermBuffer scanBuffer = new TermBuffer(); // used for scanning
   32   
   33     private TermInfo termInfo = new TermInfo();
   34   
   35     private int format;
   36     private boolean isIndex = false;
   37     long indexPointer = 0;
   38     int indexInterval;
   39     int skipInterval;
   40     int maxSkipLevels;
   41     private int formatM1SkipInterval;
   42   
   43     SegmentTermEnum(IndexInput i, FieldInfos fis, boolean isi)
   44             throws CorruptIndexException, IOException {
   45       input = i;
   46       fieldInfos = fis;
   47       isIndex = isi;
   48       maxSkipLevels = 1; // use single-level skip lists for formats > -3 
   49       
   50       int firstInt = input.readInt();
   51       if (firstInt >= 0) {
   52         // original-format file, without explicit format version number
   53         format = 0;
   54         size = firstInt;
   55   
   56         // back-compatible settings
   57         indexInterval = 128;
   58         skipInterval = Integer.MAX_VALUE; // switch off skipTo optimization
   59       } else {
   60         // we have a format version number
   61         format = firstInt;
   62   
   63         // check that it is a format we can understand
   64         if (format < TermInfosWriter.FORMAT_CURRENT)
   65           throw new CorruptIndexException("Unknown format version:" + format + " expected " + TermInfosWriter.FORMAT_CURRENT + " or higher");
   66   
   67         size = input.readLong();                    // read the size
   68         
   69         if(format == -1){
   70           if (!isIndex) {
   71             indexInterval = input.readInt();
   72             formatM1SkipInterval = input.readInt();
   73           }
   74           // switch off skipTo optimization for file format prior to 1.4rc2 in order to avoid a bug in 
   75           // skipTo implementation of these versions
   76           skipInterval = Integer.MAX_VALUE;
   77         } else {
   78           indexInterval = input.readInt();
   79           skipInterval = input.readInt();
   80           if (format <= TermInfosWriter.FORMAT) {
   81             // this new format introduces multi-level skipping
   82             maxSkipLevels = input.readInt();
   83           }
   84         }
   85       }
   86       if (format > TermInfosWriter.FORMAT_VERSION_UTF8_LENGTH_IN_BYTES) {
   87         termBuffer.setPreUTF8Strings();
   88         scanBuffer.setPreUTF8Strings();
   89         prevBuffer.setPreUTF8Strings();
   90       }
   91     }
   92   
   93     protected Object clone() {
   94       SegmentTermEnum clone = null;
   95       try {
   96         clone = (SegmentTermEnum) super.clone();
   97       } catch (CloneNotSupportedException e) {}
   98   
   99       clone.input = (IndexInput) input.clone();
  100       clone.termInfo = new TermInfo(termInfo);
  101   
  102       clone.termBuffer = (TermBuffer)termBuffer.clone();
  103       clone.prevBuffer = (TermBuffer)prevBuffer.clone();
  104       clone.scanBuffer = new TermBuffer();
  105   
  106       return clone;
  107     }
  108   
  109     final void seek(long pointer, int p, Term t, TermInfo ti)
  110             throws IOException {
  111       input.seek(pointer);
  112       position = p;
  113       termBuffer.set(t);
  114       prevBuffer.reset();
  115       termInfo.set(ti);
  116     }
  117   
  118     /** Increments the enumeration to the next element.  True if one exists.*/
  119     public final boolean next() throws IOException {
  120       if (position++ >= size - 1) {
  121         prevBuffer.set(termBuffer);
  122         termBuffer.reset();
  123         return false;
  124       }
  125   
  126       prevBuffer.set(termBuffer);
  127       termBuffer.read(input, fieldInfos);
  128   
  129       termInfo.docFreq = input.readVInt();	  // read doc freq
  130       termInfo.freqPointer += input.readVLong();	  // read freq pointer
  131       termInfo.proxPointer += input.readVLong();	  // read prox pointer
  132       
  133       if(format == -1){
  134       //  just read skipOffset in order to increment  file pointer; 
  135       // value is never used since skipTo is switched off
  136         if (!isIndex) {
  137           if (termInfo.docFreq > formatM1SkipInterval) {
  138             termInfo.skipOffset = input.readVInt(); 
  139           }
  140         }
  141       }
  142       else{
  143         if (termInfo.docFreq >= skipInterval) 
  144           termInfo.skipOffset = input.readVInt();
  145       }
  146       
  147       if (isIndex)
  148         indexPointer += input.readVLong();	  // read index pointer
  149   
  150       return true;
  151     }
  152   
  153     /** Optimized scan, without allocating new terms. 
  154      *  Return number of invocations to next(). */
  155     final int scanTo(Term term) throws IOException {
  156       scanBuffer.set(term);
  157       int count = 0;
  158       while (scanBuffer.compareTo(termBuffer) > 0 && next()) {
  159         count++;
  160       }
  161       return count;
  162     }
  163   
  164     /** Returns the current Term in the enumeration.
  165      Initially invalid, valid after next() called for the first time.*/
  166     public final Term term() {
  167       return termBuffer.toTerm();
  168     }
  169   
  170     /** Returns the previous Term enumerated. Initially null.*/
  171     final Term prev() {
  172       return prevBuffer.toTerm();
  173     }
  174   
  175     /** Returns the current TermInfo in the enumeration.
  176      Initially invalid, valid after next() called for the first time.*/
  177     final TermInfo termInfo() {
  178       return new TermInfo(termInfo);
  179     }
  180   
  181     /** Sets the argument to the current TermInfo in the enumeration.
  182      Initially invalid, valid after next() called for the first time.*/
  183     final void termInfo(TermInfo ti) {
  184       ti.set(termInfo);
  185     }
  186   
  187     /** Returns the docFreq from the current TermInfo in the enumeration.
  188      Initially invalid, valid after next() called for the first time.*/
  189     public final int docFreq() {
  190       return termInfo.docFreq;
  191     }
  192   
  193     /* Returns the freqPointer from the current TermInfo in the enumeration.
  194       Initially invalid, valid after next() called for the first time.*/
  195     final long freqPointer() {
  196       return termInfo.freqPointer;
  197     }
  198   
  199     /* Returns the proxPointer from the current TermInfo in the enumeration.
  200       Initially invalid, valid after next() called for the first time.*/
  201     final long proxPointer() {
  202       return termInfo.proxPointer;
  203     }
  204   
  205     /** Closes the enumeration to further activity, freeing resources. */
  206     public final void close() throws IOException {
  207       input.close();
  208     }
  209   }

Save This Page
Home » lucene-2.4.1-src » org.apache » lucene » index » [javadoc | source]