public static Tree processPatternsOnTree(List<Pair<TregexPattern, TsurgeonPattern>> ops, Tree t) {
matchedOnTree = false;
for (Pair<TregexPattern, TsurgeonPattern> op : ops) {
try {
if (DEBUG) {
System.err.println("Running pattern " + op.first());
}
TregexMatcher m = op.first().matcher(t);
while (m.find()) {
matchedOnTree = true;
t = op.second().evaluate(t, m);
if (t == null) {
return null;
}
m = op.first().matcher(t);
}
} catch (NullPointerException npe) {
throw new RuntimeException(
"Tsurgeon.processPatternsOnTree failed to match label for pattern: "
+ op.first()
+ ", "
+ op.second(),
npe);
}
}
return t;
}
/**
* Remove traces and non-terminal decorations (e.g., "-SUBJ" in "NP-SUBJ") from a Penn
* Treebank-style tree.
*
* @param inputTree
*/
public void normalizeTree(Tree inputTree) {
inputTree.label().setFromString("ROOT");
List<Pair<TregexPattern, TsurgeonPattern>> ops =
new ArrayList<Pair<TregexPattern, TsurgeonPattern>>();
List<TsurgeonPattern> ps = new ArrayList<TsurgeonPattern>();
String tregexOpStr;
TregexPattern matchPattern;
TsurgeonPattern p;
TregexMatcher matcher;
tregexOpStr = "/\\-NONE\\-/=emptynode";
matchPattern = TregexPatternFactory.getPattern(tregexOpStr);
matcher = matchPattern.matcher(inputTree);
ps.add(Tsurgeon.parseOperation("prune emptynode"));
matchPattern = TregexPatternFactory.getPattern(tregexOpStr);
p = Tsurgeon.collectOperations(ps);
ops.add(new Pair<TregexPattern, TsurgeonPattern>(matchPattern, p));
Tsurgeon.processPatternsOnTree(ops, inputTree);
Label nonterminalLabel;
tregexOpStr = "/.+\\-.+/=nonterminal < __";
matchPattern = TregexPatternFactory.getPattern(tregexOpStr);
matcher = matchPattern.matcher(inputTree);
while (matcher.find()) {
nonterminalLabel = matcher.getNode("nonterminal");
if (nonterminalLabel == null) continue;
nonterminalLabel.setFromString(tlp.basicCategory(nonterminalLabel.value()));
}
}
/**
* Tries to match a pattern against a tree. If it succeeds, apply the surgical operations
* contained in a {@link TsurgeonPattern}.
*
* @param matchPattern A {@link TregexPattern} to be matched against a {@link Tree}.
* @param p A {@link TsurgeonPattern} to apply.
* @param t the {@link Tree} to match against and perform surgery on.
* @return t, which has been surgically modified.
*/
public static Tree processPattern(TregexPattern matchPattern, TsurgeonPattern p, Tree t) {
TregexMatcher m = matchPattern.matcher(t);
while (m.find()) {
t = p.evaluate(t, m);
if (t == null) {
break;
}
m = matchPattern.matcher(t);
}
return t;
}
public static boolean filter(PhraseInfo phrase) {
if (phrase == null || phrase.getTree() == null) return false;
Tree phraseTree = phrase.getTree();
// __ < ( NP < (PRP !< it|them) ) | < ( NP < ( NP < (PRP !< it|them) ) )
String invalidPronounPattern =
" ( NP < (PRP=prp !< " + Utils.wordsConjuction(Rules.VALID_PRONOUNS) + ") ) ";
String filterPattern =
"__ < " + invalidPronounPattern + " | < ( NP <" + invalidPronounPattern + ")";
TregexPattern tregexPattern = TregexPattern.compile(filterPattern);
TregexMatcher matcher = tregexPattern.matcher(phraseTree);
if (matcher.matches()) {
Proof proof = new Proof(ProofType.FAIL_PERSONAL_PRONOUN);
Tree evdTree = matcher.getNode("prp");
proof.setEvidenceTree(evdTree);
phrase.addProof(proof);
return false;
}
return true;
}
public String apply(TregexMatcher m) {
String punc = m.getNode(key).value();
String punctClass = PunctEquivalenceClasser.getPunctClass(punc);
return punctClass.equals("") ? "" : annotationMark + punctClass;
}