Java Operations类代码示例

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
final Automaton toLookupAutomaton(final CharSequence key) throws IOException {
    // TODO: is there a Reader from a CharSequence?
    // Turn tokenstream into automaton:
    Automaton automaton = null;

    try (TokenStream ts = queryAnalyzer.tokenStream("", key.toString())) {
        automaton = getTokenStreamToAutomaton().toAutomaton(ts);
    }

    automaton = replaceSep(automaton);

    // TODO: we can optimize this somewhat by determinizing
    // while we convert

    // This automaton should not blow up during determinize:
    automaton = Operations.determinize(automaton, Integer.MAX_VALUE);
    return automaton;
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
private Automaton toAutomaton() {
    Automaton a = null;
    if (include != null) {
        a = include.toAutomaton();
    } else if (includeValues != null) {
        a = Automata.makeStringUnion(includeValues);
    } else {
        a = Automata.makeAnyString();
    }
    if (exclude != null) {
        a = Operations.minus(a, exclude.toAutomaton(), Operations.DEFAULT_MAX_DETERMINIZED_STATES);
    } else if (excludeValues != null) {
        a = Operations.minus(a, Automata.makeStringUnion(excludeValues), Operations.DEFAULT_MAX_DETERMINIZED_STATES);
    }
    return a;
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
public void testFuzzyQueryType() throws Exception {
    String mapping = jsonBuilder().startObject().startObject("type1")
            .startObject("properties").startObject("completion")
            .field("type", "completion")
            .endObject().endObject()
            .endObject().endObject().string();

    DocumentMapper defaultMapper = createIndex("test").mapperService().documentMapperParser().parse("type1", new CompressedXContent(mapping));
    FieldMapper fieldMapper = defaultMapper.mappers().getMapper("completion");
    CompletionFieldMapper completionFieldMapper = (CompletionFieldMapper) fieldMapper;
    Query prefixQuery = completionFieldMapper.fieldType().fuzzyQuery("co",
            Fuzziness.fromEdits(FuzzyCompletionQuery.DEFAULT_MAX_EDITS), FuzzyCompletionQuery.DEFAULT_NON_FUZZY_PREFIX,
            FuzzyCompletionQuery.DEFAULT_MIN_FUZZY_LENGTH, Operations.DEFAULT_MAX_DETERMINIZED_STATES,
            FuzzyCompletionQuery.DEFAULT_TRANSPOSITIONS, FuzzyCompletionQuery.DEFAULT_UNICODE_AWARE);
    assertThat(prefixQuery, instanceOf(FuzzyCompletionQuery.class));
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
/**
 * Create a automaton for a given context query this automaton will be used
 * to find the matching paths with the fst
 *
 * @param preserveSep set an additional char (<code>XAnalyzingSuggester.SEP_LABEL</code>) between each context query
 * @param queries list of {@link ContextQuery} defining the lookup context
 *
 * @return Automaton matching the given Query
 */
public static Automaton toAutomaton(boolean preserveSep, Iterable<ContextQuery> queries) {
    Automaton a = Automata.makeEmptyString();

    Automaton gap = Automata.makeChar(ContextMapping.SEPARATOR);
    if (preserveSep) {
        // if separators are preserved the fst contains a SEP_LABEL
        // behind each gap. To have a matching automaton, we need to
        // include the SEP_LABEL in the query as well
        gap = Operations.concatenate(gap, Automata.makeChar(XAnalyzingSuggester.SEP_LABEL));
    }

    for (ContextQuery query : queries) {
        a = Operations.concatenate(Arrays.asList(query.toAutomaton(), gap, a));
    }

    // TODO: should we limit this?  Do any of our ContextQuery impls really create exponential regexps?  GeoQuery looks safe (union
    // of strings).
    return Operations.determinize(a, Integer.MAX_VALUE);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
final Automaton toLookupAutomaton(final CharSequence key) throws IOException {
  // TODO: is there a Reader from a CharSequence?
  // Turn tokenstream into automaton:
  Automaton automaton = null;
  TokenStream ts = queryAnalyzer.tokenStream("", key.toString());
  try {
      automaton = getTokenStreamToAutomaton().toAutomaton(ts);
  } finally {
    IOUtils.closeWhileHandlingException(ts);
  }

  automaton = replaceSep(automaton);

  // TODO: we can optimize this somewhat by determinizing
  // while we convert
  automaton = Operations.determinize(automaton, DEFAULT_MAX_DETERMINIZED_STATES);
  return automaton;
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
/** Returns true iff <code>o</code> is equal to this. */
@Override
public boolean equals(Object o) {
  if (!(o instanceof TermAutomatonQuery)) {
    return false;
  }
  TermAutomatonQuery other = (TermAutomatonQuery) o;

  if (det == null) {
    throw new IllegalStateException("please call finish first");
  }
  if (other.det == null) {
    throw new IllegalStateException("please call other.finish first");
  }

  // NOTE: not quite correct, because if terms were added in different
  // order in each query but the language is the same, we return false:
  return (this.getBoost() == other.getBoost())
    && this.termToID.equals(other.termToID) &&
    Operations.sameLanguage(det, other.det);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
@Override
public void setUp() throws Exception {
  super.setUp();
  Automaton single = new Automaton();
  int initial = single.createState();
  int accept = single.createState();
  single.setAccept(accept, true);

  // build an automaton matching this jvm's letter definition
  for (int i = 0; i <= 0x10FFFF; i++) {
    if (Character.isLetter(i)) {
      single.addTransition(initial, accept, i);
    }
  }
  Automaton repeat = Operations.repeat(single);
  jvmLetter = new CharacterRunAutomaton(repeat);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
public void testCustomProvider() throws IOException {
  AutomatonProvider myProvider = new AutomatonProvider() {
    // automaton that matches quick or brown
    private Automaton quickBrownAutomaton = Operations.union(Arrays
        .asList(Automata.makeString("quick"),
        Automata.makeString("brown"),
        Automata.makeString("bob")));
    
    @Override
    public Automaton getAutomaton(String name) {
      if (name.equals("quickBrown")) return quickBrownAutomaton;
      else return null;
    }
  };
  RegexpQuery query = new RegexpQuery(newTerm("<quickBrown>"), RegExp.ALL,
    myProvider, DEFAULT_MAX_DETERMINIZED_STATES);
  assertEquals(1, searcher.search(query, 5).totalHits);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
/**
 * Test some very simple automata.
 */
public void testAutomata() throws IOException {
  assertAutomatonHits(0, Automata.makeEmpty());
  assertAutomatonHits(0, Automata.makeEmptyString());
  assertAutomatonHits(2, Automata.makeAnyChar());
  assertAutomatonHits(3, Automata.makeAnyString());
  assertAutomatonHits(2, Automata.makeString("doc"));
  assertAutomatonHits(1, Automata.makeChar('a'));
  assertAutomatonHits(2, Automata.makeCharRange('a', 'b'));
  assertAutomatonHits(2, Automata.makeInterval(1233, 2346, 0));
  assertAutomatonHits(1, Automata.makeInterval(0, 2000, 0));
  assertAutomatonHits(2, Operations.union(Automata.makeChar('a'),
      Automata.makeChar('b')));
  assertAutomatonHits(0, Operations.intersection(Automata
      .makeChar('a'), Automata.makeChar('b')));
  assertAutomatonHits(1, Operations.minus(Automata.makeCharRange('a', 'b'), 
      Automata.makeChar('a'), DEFAULT_MAX_DETERMINIZED_STATES));
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
/**
 * Build the {@link CharacterRunAutomaton} that represents the reindex-from-remote whitelist and make sure that it doesn't whitelist
 * the world.
 */
static CharacterRunAutomaton buildRemoteWhitelist(List<String> whitelist) {
    if (whitelist.isEmpty()) {
        return new CharacterRunAutomaton(Automata.makeEmpty());
    }
    Automaton automaton = Regex.simpleMatchToAutomaton(whitelist.toArray(Strings.EMPTY_ARRAY));
    automaton = MinimizationOperations.minimize(automaton, Operations.DEFAULT_MAX_DETERMINIZED_STATES);
    if (Operations.isTotal(automaton)) {
        throw new IllegalArgumentException("Refusing to start because whitelist " + whitelist + " accepts all addresses. "
                + "This would allow users to reindex-from-remote any URL they like effectively having Elasticsearch make HTTP GETs "
                + "for them.");
    }
    return new CharacterRunAutomaton(automaton);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
@Override
protected Automaton convertAutomaton(Automaton a) {
  if (unicodeAware) {
    // FLORIAN EDIT: get converted Automaton from superclass
    Automaton utf8automaton = new UTF32ToUTF8().convert(super.convertAutomaton(a));
    // This automaton should not blow up during determinize:
    utf8automaton = Operations.determinize(utf8automaton, Integer.MAX_VALUE);
    return utf8automaton;
  } else {
    return super.convertAutomaton(a);
  }
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
protected Automaton convertAutomaton(Automaton a) {
  if (queryPrefix != null) {
    a = Operations.concatenate(Arrays.asList(queryPrefix, a));
    // This automaton should not blow up during determinize:
    a = Operations.determinize(a, Integer.MAX_VALUE);
  }
  return a;
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
/** Return an {@link Automaton} that matches the given pattern. */
public static Automaton simpleMatchToAutomaton(String pattern) {
    List<Automaton> automata = new ArrayList<>();
    int previous = 0;
    for (int i = pattern.indexOf('*'); i != -1; i = pattern.indexOf('*', i + 1)) {
        automata.add(Automata.makeString(pattern.substring(previous, i)));
        automata.add(Automata.makeAnyString());
        previous = i + 1;
    }
    automata.add(Automata.makeString(pattern.substring(previous)));
    return Operations.concatenate(automata);
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
public void testRegexQueryType() throws Exception {
    String mapping = jsonBuilder().startObject().startObject("type1")
            .startObject("properties").startObject("completion")
            .field("type", "completion")
            .endObject().endObject()
            .endObject().endObject().string();

    DocumentMapper defaultMapper = createIndex("test").mapperService().documentMapperParser().parse("type1", new CompressedXContent(mapping));
    FieldMapper fieldMapper = defaultMapper.mappers().getMapper("completion");
    CompletionFieldMapper completionFieldMapper = (CompletionFieldMapper) fieldMapper;
    Query prefixQuery = completionFieldMapper.fieldType()
            .regexpQuery(new BytesRef("co"), RegExp.ALL, Operations.DEFAULT_MAX_DETERMINIZED_STATES);
    assertThat(prefixQuery, instanceOf(RegexCompletionQuery.class));
}
 

import org.apache.lucene.util.automaton.Operations; //导入依赖的package包/类
@Override
public Automaton toAutomaton() {
    List<Automaton> automatons = new ArrayList<>();
    for (CharSequence value : values) {
        automatons.add(Automata.makeString(value.toString()));
    }
    return Operations.union(automatons);
}