public void testMultipleHoles() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("a", 1, 1), token("b", 3, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected = join(s2a("a"), SEP_A, HOLE_A, SEP_A, HOLE_A, SEP_A, s2a("b"));
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testSynHangingOverEnd() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("a", 1, 1), token("X", 0, 10),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected =
BasicOperations.union(BasicAutomata.makeString("a"), BasicAutomata.makeString("X"));
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testStartsWithHole() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("abc", 2, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected = join(HOLE_A, SEP_A, s2a("abc"));
// toDot(actual);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testOverlappedTokensSausage() throws Exception {
// Two tokens on top of each other (sausage):
final TokenStream ts =
new CannedTokenStream(new Token[] {token("abc", 1, 1), token("xyz", 0, 1)});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton a1 = BasicAutomata.makeString("abc");
final Automaton a2 = BasicAutomata.makeString("xyz");
final Automaton expected = BasicOperations.union(a1, a2);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testSingleToken() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("abc", 1, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected = BasicAutomata.makeString("abc");
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testSynOverHole2() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("xyz", 1, 1), token("abc", 0, 3), token("def", 2, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected =
BasicOperations.union(
join(s2a("xyz"), SEP_A, HOLE_A, SEP_A, s2a("def")), BasicAutomata.makeString("abc"));
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testTwoTokens() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("abc", 1, 1), token("def", 1, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton expected = join("abc", "def");
// toDot(actual);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testOverlappedTokensLattice2() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("abc", 1, 1), token("xyz", 0, 3), token("def", 1, 1), token("ghi", 1, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton a1 = BasicAutomata.makeString("xyz");
final Automaton a2 = join("abc", "def", "ghi");
final Automaton expected = BasicOperations.union(a1, a2);
// toDot(actual);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
public void testSynOverHole() throws Exception {
final TokenStream ts =
new CannedTokenStream(
new Token[] {
token("a", 1, 1), token("X", 0, 2), token("b", 2, 1),
});
final Automaton actual = (new TokenStreamToAutomaton()).toAutomaton(ts);
final Automaton a1 =
BasicOperations.union(join(s2a("a"), SEP_A, HOLE_A), BasicAutomata.makeString("X"));
final Automaton expected = BasicOperations.concatenate(a1, join(SEP_A, s2a("b")));
// toDot(actual);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
/** tests intersect: TODO start at a random term! */
public void testIntersect() throws Exception {
for (int i = 0; i < numIterations; i++) {
String reg = AutomatonTestUtil.randomRegexp(random());
Automaton automaton = new RegExp(reg, RegExp.NONE).toAutomaton();
CompiledAutomaton ca =
new CompiledAutomaton(automaton, SpecialOperations.isFinite(automaton), false);
TermsEnum te = MultiFields.getTerms(reader, "field").intersect(ca, null);
Automaton expected = BasicOperations.intersection(termsAutomaton, automaton);
TreeSet<BytesRef> found = new TreeSet<BytesRef>();
while (te.next() != null) {
found.add(BytesRef.deepCopyOf(te.term()));
}
Automaton actual = BasicAutomata.makeStringUnion(found);
assertTrue(BasicOperations.sameLanguage(expected, actual));
}
}
