Java StringHelper.startsWith方法代码示例

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
private TreeMap<String,Long> readFields(IndexInput in) throws IOException {
  ChecksumIndexInput input = new BufferedChecksumIndexInput(in);
  BytesRefBuilder scratch = new BytesRefBuilder();
  TreeMap<String,Long> fields = new TreeMap<>();
  
  while (true) {
    SimpleTextUtil.readLine(input, scratch);
    if (scratch.get().equals(END)) {
      SimpleTextUtil.checkFooter(input);
      return fields;
    } else if (StringHelper.startsWith(scratch.get(), FIELD)) {
      String fieldName = new String(scratch.bytes(), FIELD.length, scratch.length() - FIELD.length, StandardCharsets.UTF_8);
      fields.put(fieldName, input.getFilePointer());
    }
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
/**
 * Scans ({@code termsEnum.next()}) terms until a term is found that does
 * not start with curVNode's cell. If it finds a leaf cell or a cell at
 * level {@code scanDetailLevel} then it calls {@link
 * #visitScanned(org.apache.lucene.spatial.prefix.tree.Cell)}.
 */
protected void scan(int scanDetailLevel) throws IOException {
  for (;
       thisTerm != null && StringHelper.startsWith(thisTerm, curVNodeTerm);//TODO refactor to use method on curVNode.cell
       thisTerm = termsEnum.next()) {
    scanCell = grid.getCell(thisTerm.bytes, thisTerm.offset, thisTerm.length, scanCell);

    int termLevel = scanCell.getLevel();
    if (termLevel < scanDetailLevel) {
      if (scanCell.isLeaf())
        visitScanned(scanCell);
    } else if (termLevel == scanDetailLevel) {
      if (!scanCell.isLeaf())//LUCENE-5529
        visitScanned(scanCell);
    }
  }//term loop
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
private void getPrefixTerms(ObjectHashSet<Term> terms, final Term prefix, final IndexReader reader) throws IOException {
    // SlowCompositeReaderWrapper could be used... but this would merge all terms from each segment into one terms
    // instance, which is very expensive. Therefore I think it is better to iterate over each leaf individually.
    List<LeafReaderContext> leaves = reader.leaves();
    for (LeafReaderContext leaf : leaves) {
        Terms _terms = leaf.reader().terms(field);
        if (_terms == null) {
            continue;
        }

        TermsEnum termsEnum = _terms.iterator();
        TermsEnum.SeekStatus seekStatus = termsEnum.seekCeil(prefix.bytes());
        if (TermsEnum.SeekStatus.END == seekStatus) {
            continue;
        }

        for (BytesRef term = termsEnum.term(); term != null; term = termsEnum.next()) {
            if (!StringHelper.startsWith(term, prefix.bytes())) {
                break;
            }

            terms.add(new Term(field, BytesRef.deepCopyOf(term)));
            if (terms.size() >= maxExpansions) {
                return;
            }
        }
    }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
@Override
protected AcceptStatus accept(BytesRef term) {
  if (StringHelper.startsWith(term, prefixRef)) {
    return AcceptStatus.YES;
  } else {
    return AcceptStatus.END;
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
private BytesRef setTerm() throws IOException {
  term = termsEnum.term();
  //System.out.println("  setTerm() term=" + term.utf8ToString() + " vs prefix=" + (prefix == null ? "null" : prefix.utf8ToString()));
  if (prefix != null && !StringHelper.startsWith(term, prefix)) {
    term = null;
  }
  return term;
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
@Override
public void visitMatchingTerms(
  IndexReader reader,
  String fieldName,
  MatchingTermVisitor mtv) throws IOException
{
  int prefixLength = prefix.length();
  Terms terms = MultiFields.getTerms(reader, fieldName);
  if (terms != null) {
    Matcher matcher = pattern.matcher("");
    try {
      TermsEnum termsEnum = terms.iterator(null);

      TermsEnum.SeekStatus status = termsEnum.seekCeil(prefixRef);
      BytesRef text;
      if (status == TermsEnum.SeekStatus.FOUND) {
        text = prefixRef;
      } else if (status == TermsEnum.SeekStatus.NOT_FOUND) {
        text = termsEnum.term();
      } else {
        text = null;
      }

      while(text != null) {
        if (text != null && StringHelper.startsWith(text, prefixRef)) {
          String textString = text.utf8ToString();
          matcher.reset(textString.substring(prefixLength));
          if (matcher.matches()) {
            mtv.visitMatchingTerm(new Term(fieldName, textString));
          }
        } else {
          break;
        }
        text = termsEnum.next();
      }
    } finally {
      matcher.reset();
    }
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
@Override
public void visitMatchingTerms(
  IndexReader reader,
  String fieldName,
  MatchingTermVisitor mtv) throws IOException
{
  /* inspired by PrefixQuery.rewrite(): */
  Terms terms = MultiFields.getTerms(reader, fieldName);
  if (terms != null) {
    TermsEnum termsEnum = terms.iterator(null);

    boolean skip = false;
    TermsEnum.SeekStatus status = termsEnum.seekCeil(new BytesRef(getPrefix()));
    if (status == TermsEnum.SeekStatus.FOUND) {
      mtv.visitMatchingTerm(getLucenePrefixTerm(fieldName));
    } else if (status == TermsEnum.SeekStatus.NOT_FOUND) {
      if (StringHelper.startsWith(termsEnum.term(), prefixRef)) {
        mtv.visitMatchingTerm(new Term(fieldName, termsEnum.term().utf8ToString()));
      } else {
        skip = true;
      }
    } else {
      // EOF
      skip = true;
    }

    if (!skip) {
      while(true) {
        BytesRef text = termsEnum.next();
        if (text != null && StringHelper.startsWith(text, prefixRef)) {
          mtv.visitMatchingTerm(new Term(fieldName, text.utf8ToString()));
        } else {
          break;
        }
      }
    }
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
public static void checkFooter(ChecksumIndexInput input) throws IOException {
  BytesRefBuilder scratch = new BytesRefBuilder();
  String expectedChecksum = String.format(Locale.ROOT, "%020d", input.getChecksum());
  SimpleTextUtil.readLine(input, scratch);
  if (StringHelper.startsWith(scratch.get(), CHECKSUM) == false) {
    throw new CorruptIndexException("SimpleText failure: expected checksum line but got " + scratch.get().utf8ToString() + " (resource=" + input + ")");
  }
  String actualChecksum = new BytesRef(scratch.bytes(), CHECKSUM.length, scratch.length() - CHECKSUM.length).utf8ToString();
  if (!expectedChecksum.equals(actualChecksum)) {
    throw new CorruptIndexException("SimpleText checksum failure: " + actualChecksum + " != " + expectedChecksum + " (resource=" + input + ")");
  }
  if (input.length() != input.getFilePointer()) {
    throw new CorruptIndexException("Unexpected stuff at the end of file, please be careful with your text editor! (resource=" + input + ")");
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
@Override
protected AcceptStatus accept(BytesRef term) {
  if (StringHelper.startsWith(term, prefixRef)) {
    // TODO: set BoostAttr based on distance of
    // searchTerm.text() and term().text()
    return regexImpl.match(term) ? AcceptStatus.YES : AcceptStatus.NO;
  } else {
    return AcceptStatus.NO;
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
/**
 * <p>The termCompare method in FuzzyTermEnum uses Levenshtein distance to 
 * calculate the distance between the given term and the comparing term. 
 * </p>
 * <p>If the minSimilarity is >= 1.0, this uses the maxEdits as the comparison.
 * Otherwise, this method uses the following logic to calculate similarity.
 * <pre>
 *   similarity = 1 - ((float)distance / (float) (prefixLength + Math.min(textlen, targetlen)));
 *   </pre>
 * where distance is the Levenshtein distance for the two words.
 * </p>
 * 
 */
@Override
protected final AcceptStatus accept(BytesRef term) {
  if (StringHelper.startsWith(term, prefixBytesRef)) {
    utf32.copyUTF8Bytes(term);
    final int distance = calcDistance(utf32.ints(), realPrefixLength, utf32.length() - realPrefixLength);
   
    //Integer.MIN_VALUE is the sentinel that Levenshtein stopped early
    if (distance == Integer.MIN_VALUE){
       return AcceptStatus.NO;
    }
    //no need to calc similarity, if raw is true and distance > maxEdits
    if (raw == true && distance > maxEdits){
          return AcceptStatus.NO;
    } 
    final float similarity = calcSimilarity(distance, (utf32.length() - realPrefixLength), text.length);
    
    //if raw is true, then distance must also be <= maxEdits by now
    //given the previous if statement
    if (raw == true ||
          (raw == false && similarity > minSimilarity)) {
      boostAtt.setBoost((similarity - minSimilarity) * scale_factor);
      return AcceptStatus.YES;
    } else {
       return AcceptStatus.NO;
    }
  } else {
    return AcceptStatus.END;
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
private TreeMap<String,Long> readFields(IndexInput in) throws IOException {
  BytesRef scratch = new BytesRef(10);
  TreeMap<String,Long> fields = new TreeMap<String,Long>();
  
  while (true) {
    SimpleTextUtil.readLine(in, scratch);
    if (scratch.equals(END)) {
      return fields;
    } else if (StringHelper.startsWith(scratch, FIELD)) {
      String fieldName = new String(scratch.bytes, scratch.offset + FIELD.length, scratch.length - FIELD.length, "UTF-8");
      fields.put(fieldName, in.getFilePointer());
    }
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
/**
 * <p>The termCompare method in FuzzyTermEnum uses Levenshtein distance to 
 * calculate the distance between the given term and the comparing term. 
 * </p>
 * <p>If the minSimilarity is >= 1.0, this uses the maxEdits as the comparison.
 * Otherwise, this method uses the following logic to calculate similarity.
 * <pre>
 *   similarity = 1 - ((float)distance / (float) (prefixLength + Math.min(textlen, targetlen)));
 *   </pre>
 * where distance is the Levenshtein distance for the two words.
 * </p>
 * 
 */
@Override
protected final AcceptStatus accept(BytesRef term) {
  if (StringHelper.startsWith(term, prefixBytesRef)) {
    UnicodeUtil.UTF8toUTF32(term, utf32);
    final int distance = calcDistance(utf32.ints, realPrefixLength, utf32.length - realPrefixLength);
   
    //Integer.MIN_VALUE is the sentinel that Levenshtein stopped early
    if (distance == Integer.MIN_VALUE){
       return AcceptStatus.NO;
    }
    //no need to calc similarity, if raw is true and distance > maxEdits
    if (raw == true && distance > maxEdits){
          return AcceptStatus.NO;
    } 
    final float similarity = calcSimilarity(distance, (utf32.length - realPrefixLength), text.length);
    
    //if raw is true, then distance must also be <= maxEdits by now
    //given the previous if statement
    if (raw == true ||
          (raw == false && similarity > minSimilarity)) {
      boostAtt.setBoost((similarity - minSimilarity) * scale_factor);
      return AcceptStatus.YES;
    } else {
       return AcceptStatus.NO;
    }
  } else {
    return AcceptStatus.END;
  }
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
/**
 * Scans ({@code termsEnum.next()}) terms until a term is found that does
 * not start with curVNode's cell. If it finds a leaf cell or a cell at
 * level {@code scanDetailLevel} then it calls {@link
 * #visitScanned(org.apache.lucene.spatial.prefix.tree.Cell)}.
 */
protected void scan(int scanDetailLevel) throws IOException {
  for (;
       thisTerm != null && StringHelper.startsWith(thisTerm, curVNodeTerm);
       thisTerm = termsEnum.next()) {
    scanCell = grid.getCell(thisTerm.bytes, thisTerm.offset, thisTerm.length, scanCell);

    int termLevel = scanCell.getLevel();
    if (termLevel > scanDetailLevel)
      continue;
    if (termLevel == scanDetailLevel || scanCell.isLeaf()) {
      visitScanned(scanCell);
    }
  }//term loop
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
/** Used only in ctor: */
private boolean startsWith(BytesRef prefix) {
  return StringHelper.startsWith(scratch.get(), prefix);
}
 

import org.apache.lucene.util.StringHelper; //导入方法依赖的package包/类
private void loadTerms() throws IOException {
  PositiveIntOutputs posIntOutputs = PositiveIntOutputs.getSingleton();
  final Builder<PairOutputs.Pair<Long,PairOutputs.Pair<Long,Long>>> b;
  final PairOutputs<Long,Long> outputsInner = new PairOutputs<>(posIntOutputs, posIntOutputs);
  final PairOutputs<Long,PairOutputs.Pair<Long,Long>> outputs = new PairOutputs<>(posIntOutputs,
                                                                                                                  outputsInner);
  b = new Builder<>(FST.INPUT_TYPE.BYTE1, outputs);
  IndexInput in = SimpleTextFieldsReader.this.in.clone();
  in.seek(termsStart);
  final BytesRefBuilder lastTerm = new BytesRefBuilder();
  long lastDocsStart = -1;
  int docFreq = 0;
  long totalTermFreq = 0;
  FixedBitSet visitedDocs = new FixedBitSet(maxDoc);
  final IntsRefBuilder scratchIntsRef = new IntsRefBuilder();
  while(true) {
    SimpleTextUtil.readLine(in, scratch);
    if (scratch.get().equals(END) || StringHelper.startsWith(scratch.get(), FIELD)) {
      if (lastDocsStart != -1) {
        b.add(Util.toIntsRef(lastTerm.get(), scratchIntsRef),
              outputs.newPair(lastDocsStart,
                              outputsInner.newPair((long) docFreq, totalTermFreq)));
        sumTotalTermFreq += totalTermFreq;
      }
      break;
    } else if (StringHelper.startsWith(scratch.get(), DOC)) {
      docFreq++;
      sumDocFreq++;
      scratchUTF16.copyUTF8Bytes(scratch.bytes(), DOC.length, scratch.length()-DOC.length);
      int docID = ArrayUtil.parseInt(scratchUTF16.chars(), 0, scratchUTF16.length());
      visitedDocs.set(docID);
    } else if (StringHelper.startsWith(scratch.get(), FREQ)) {
      scratchUTF16.copyUTF8Bytes(scratch.bytes(), FREQ.length, scratch.length()-FREQ.length);
      totalTermFreq += ArrayUtil.parseInt(scratchUTF16.chars(), 0, scratchUTF16.length());
    } else if (StringHelper.startsWith(scratch.get(), TERM)) {
      if (lastDocsStart != -1) {
        b.add(Util.toIntsRef(lastTerm.get(), scratchIntsRef), outputs.newPair(lastDocsStart,
                                                                        outputsInner.newPair((long) docFreq, totalTermFreq)));
      }
      lastDocsStart = in.getFilePointer();
      final int len = scratch.length() - TERM.length;
      lastTerm.grow(len);
      System.arraycopy(scratch.bytes(), TERM.length, lastTerm.bytes(), 0, len);
      lastTerm.setLength(len);
      docFreq = 0;
      sumTotalTermFreq += totalTermFreq;
      totalTermFreq = 0;
      termCount++;
    }
  }
  docCount = visitedDocs.cardinality();
  fst = b.finish();
  /*
  PrintStream ps = new PrintStream("out.dot");
  fst.toDot(ps);
  ps.close();
  System.out.println("SAVED out.dot");
  */
  //System.out.println("FST " + fst.sizeInBytes());
}