Examples with IntsRefBuilder org.apache.lucene.util.IntsRefBuilder used on opensource projects

Example 16 with IntsRefBuilder

use of org.apache.lucene.util.IntsRefBuilder in project lucene-solr by apache.

the class FiniteStringsIteratorTest method testSingletonNoLimit.

public void testSingletonNoLimit() {
    Automaton a = Automata.makeString("foobar");
    FiniteStringsIterator iterator = new FiniteStringsIterator(a);
    List<IntsRef> actual = getFiniteStrings(iterator);
    assertEquals(1, actual.size());
    IntsRefBuilder scratch = new IntsRefBuilder();
    Util.toUTF32("foobar".toCharArray(), 0, 6, scratch);
    assertTrue(actual.contains(scratch.get()));
}

Example 17 with IntsRefBuilder

use of org.apache.lucene.util.IntsRefBuilder in project lucene-solr by apache.

the class FiniteStringsIteratorTest method testRandomFiniteStrings1.

public void testRandomFiniteStrings1() {
    int numStrings = atLeast(100);
    if (VERBOSE) {
        System.out.println("TEST: numStrings=" + numStrings);
    }
    Set<IntsRef> strings = new HashSet<>();
    List<Automaton> automata = new ArrayList<>();
    IntsRefBuilder scratch = new IntsRefBuilder();
    for (int i = 0; i < numStrings; i++) {
        String s = TestUtil.randomSimpleString(random(), 1, 200);
        Util.toUTF32(s.toCharArray(), 0, s.length(), scratch);
        if (strings.add(scratch.toIntsRef())) {
            automata.add(Automata.makeString(s));
            if (VERBOSE) {
                System.out.println("  add string=" + s);
            }
        }
    }
    // TODO: we could sometimes use
    // DaciukMihovAutomatonBuilder here
    // TODO: what other random things can we do here...
    Automaton a = Operations.union(automata);
    if (random().nextBoolean()) {
        a = MinimizationOperations.minimize(a, 1000000);
        if (VERBOSE) {
            System.out.println("TEST: a.minimize numStates=" + a.getNumStates());
        }
    } else if (random().nextBoolean()) {
        if (VERBOSE) {
            System.out.println("TEST: a.determinize");
        }
        a = Operations.determinize(a, 1000000);
    } else if (random().nextBoolean()) {
        if (VERBOSE) {
            System.out.println("TEST: a.removeDeadStates");
        }
        a = Operations.removeDeadStates(a);
    }
    FiniteStringsIterator iterator = new FiniteStringsIterator(a);
    List<IntsRef> actual = getFiniteStrings(iterator);
    assertFiniteStringsRecursive(a, actual);
    if (!strings.equals(new HashSet<>(actual))) {
        System.out.println("strings.size()=" + strings.size() + " actual.size=" + actual.size());
        List<IntsRef> x = new ArrayList<>(strings);
        Collections.sort(x);
        List<IntsRef> y = new ArrayList<>(actual);
        Collections.sort(y);
        int end = Math.min(x.size(), y.size());
        for (int i = 0; i < end; i++) {
            System.out.println("  i=" + i + " string=" + toString(x.get(i)) + " actual=" + toString(y.get(i)));
        }
        fail("wrong strings found");
    }
}

Example 18 with IntsRefBuilder

use of org.apache.lucene.util.IntsRefBuilder in project lucene-solr by apache.

the class TestAutomaton method testMakeBinaryIntervalRandom.

public void testMakeBinaryIntervalRandom() throws Exception {
    int iters = atLeast(100);
    for (int iter = 0; iter < iters; iter++) {
        BytesRef minTerm = TestUtil.randomBinaryTerm(random());
        boolean minInclusive = random().nextBoolean();
        BytesRef maxTerm = TestUtil.randomBinaryTerm(random());
        boolean maxInclusive = random().nextBoolean();
        Automaton a = makeBinaryInterval(minTerm, minInclusive, maxTerm, maxInclusive);
        for (int iter2 = 0; iter2 < 500; iter2++) {
            BytesRef term = TestUtil.randomBinaryTerm(random());
            int minCmp = minTerm.compareTo(term);
            int maxCmp = maxTerm.compareTo(term);
            boolean expected;
            if (minCmp > 0 || maxCmp < 0) {
                expected = false;
            } else if (minCmp == 0 && maxCmp == 0) {
                expected = minInclusive && maxInclusive;
            } else if (minCmp == 0) {
                expected = minInclusive;
            } else if (maxCmp == 0) {
                expected = maxInclusive;
            } else {
                expected = true;
            }
            if (VERBOSE) {
                System.out.println("  check term=" + term + " expected=" + expected);
            }
            IntsRefBuilder intsBuilder = new IntsRefBuilder();
            Util.toIntsRef(term, intsBuilder);
            assertEquals(expected, Operations.run(a, intsBuilder.toIntsRef()));
        }
    }
}

Example 19 with IntsRefBuilder

use of org.apache.lucene.util.IntsRefBuilder in project lucene-solr by apache.

the class Operations method getSingleton.

/** If this automaton accepts a single input, return it.  Else, return null.
   *  The automaton must be deterministic. */
public static IntsRef getSingleton(Automaton a) {
    if (a.isDeterministic() == false) {
        throw new IllegalArgumentException("input automaton must be deterministic");
    }
    IntsRefBuilder builder = new IntsRefBuilder();
    HashSet<Integer> visited = new HashSet<>();
    int s = 0;
    Transition t = new Transition();
    while (true) {
        visited.add(s);
        if (a.isAccept(s) == false) {
            if (a.getNumTransitions(s) == 1) {
                a.getTransition(s, 0, t);
                if (t.min == t.max && !visited.contains(t.dest)) {
                    builder.append(t.min);
                    s = t.dest;
                    continue;
                }
            }
        } else if (a.getNumTransitions(s) == 0) {
            return builder.get();
        }
        // Automaton accepts more than one string:
        return null;
    }
}

Example 20 with IntsRefBuilder

use of org.apache.lucene.util.IntsRefBuilder in project lucene-solr by apache.

the class BaseSynonymParserTestCase method assertEntryAbsent.

/**
   * Validates that there are no synonyms for the given word.
   * @param synonynMap  the generated synonym map after parsing
   * @param word        word (phrase) we are validating the synonyms for. Should be the value that comes out of the analyzer.
   *                    All spaces will be replaced by word separators.
   */
public static void assertEntryAbsent(SynonymMap synonynMap, String word) throws IOException {
    word = word.replace(' ', SynonymMap.WORD_SEPARATOR);
    BytesRef value = Util.get(synonynMap.fst, Util.toUTF32(new CharsRef(word), new IntsRefBuilder()));
    assertNull("There should be no synonyms for: " + word, value);
}