What causes "ExhaustivePCFGParser doesn't sample." in stanfordnlp/CoreNLP?

Sample manually from getKBestParses(k) output weighted by their scores.

stanfordnlp/CoreNLP · error · UnsupportedOperationException

ExhaustivePCFGParser doesn't sample.

Error message

ExhaustivePCFGParser doesn't sample.

What it means

Sentinel UnsupportedOperationException signalling a capability gap: ExhaustivePCFGParser only performs exact exhaustive parsing and can return k-best or good parses (getKGoodParses just delegates to getKBestParses), but it contains no sampling machinery to draw parses according to their relative probability. The error fires whenever getKSampledParses is called on this parser.

Solutions

Sample manually from getKBestParses(k) output weighted by their scores.
Use getKGoodParses(k) or getKBestParses(k) instead if an approximate list suffices.
Implement external sampling: draw parses from the returned scored list using exp(-score) weights.

Example fix

// before
List<ScoredObject<Tree>> samples = pq.getKSampledParses(10);
// after
List<ScoredObject<Tree>> best = pq.getKBestParses(100);
List<ScoredObject<Tree>> samples = weightedSampleWithoutReplacement(best, 10, new Random());

Defensive patterns

Strategy: try-catch

Validate before calling

boolean supportsSampling = query.getClass().getSimpleName().contains("Factored") && !query.getClass().getSimpleName().contains("ExhaustivePCFG");

Try / catch

try { samples = pq.getKSampledParses(k); } catch (UnsupportedOperationException e) { samples = weightedSample(pq.getKBestParses(100), k); }

Prevention

Assume ExhaustivePCFGParser provides only exact/approximate k-best, not sampling.
Derive samples from scored k-best lists with probability exp(score).
Keep sampling logic outside the parser abstraction layer.

When it happens

Trigger: Calling getKSampledParses(k) on a ParserQuery produced by ExhaustivePCFGParser after parsing.

Common situations: Code that uniformly requests sampled parses across parser implementations; PCFG parser has no sampling machinery despite having k-best support.

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/6559bf75b19e8479. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/parser/lexparser/ExhaustivePCFGParser.java:1807

   *  @param k The number of good parses to return
   *  @return A list of k good parses for the sentence, with
   *         each accompanied by its score
   */
  @Override
  public List<ScoredObject<Tree>> getKGoodParses(int k) {
    return getKBestParses(k);
  }

  /** Get k parse samples for the sentence.  It is expected that the
   *  parses are sampled based on their relative probability.
   *
   *  @param k The number of sampled parses to return
   *  @return A list of k parse samples for the sentence, with
   *         each accompanied by its score
   */
  @Override
  public List<ScoredObject<Tree>> getKSampledParses(int k) {
    throw new UnsupportedOperationException("ExhaustivePCFGParser doesn't sample.");
  }


  //
  // BEGIN K-BEST STUFF
  // taken straight out of "Better k-best Parsing" by Liang Huang and David
  // Chiang
  //

  /** Get the exact k best parses for the sentence.
   *
   *  @param k The number of best parses to return
   *  @return The exact k best parses for the sentence, with
   *         each accompanied by its score (typically a
   *         negative log probability).
   */
  @Override
  public List<ScoredObject<Tree>> getKBestParses(int k) {

View on GitHub (pinned to 1b7edd19c4)