stanfordnlp/CoreNLP · error · IllegalStateException

Someone didn't add a handler for a new docType.

Error message

Someone didn't add a handler for a new docType.

What it means

DocumentPreprocessor.iterator() dispatches on the DocType enum; only Plain and XML have implemented iterators. If docType holds any other value, the library throws this IllegalStateException because a new DocType was added without a corresponding iterator handler. In normal use this means the DocType passed to the constructor is not supported by this code path.

Solutions

  1. Use DocType.Plain for regular text or DocType.XML for XML documents
  2. Check for a version mismatch between library jars and switch to a consistent CoreNLP version
  3. If you added a custom DocType, implement and return a matching Iterator in iterator()'s dispatch chain
  4. Validate the DocType right after construction (fail early) rather than at iteration time

Example fix

// before
DocumentPreprocessor dp = new DocumentPreprocessor(reader, DocType.Media);
// after
DocumentPreprocessor dp = new DocumentPreprocessor(reader, DocType.Plain); // supported type
Defensive patterns

Strategy: validation

Validate before calling

if (docType != DocType.Plain && docType != DocType.XML) {
  throw new IllegalArgumentException("Unsupported DocType for DocumentPreprocessor: " + docType);
}

Try / catch

try (Iterable<List<HasWord>> sents = () -> dp.iterator()) {
  // consume sentences
} catch (IllegalStateException e) {
  if (e.getMessage() != null && e.getMessage().contains("didn't add a handler")) {
    throw new UnsupportedDocTypeException(docType, e);
  }
  throw e;
}

Prevention

When it happens

Trigger: Constructing a DocumentPreprocessor with a DocType other than Plain or XML (e.g. a custom/newer DocType constant) and then iterating via iterator(), tokens(), or a for-each over the preprocessor.

Common situations: Using a DocumentPreprocessor subclass or an upgraded enum from a newer CoreNLP version with an older iterator implementation; copy-pasted construction code setting an exotic DocType; reflection-based instantiation choosing the wrong enum constant.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/be701a40e44ed272. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/process/DocumentPreprocessor.java:231


  /**
   * Returns sentences until the document is exhausted. Calls close() if the end of the document
   * is reached. Otherwise, the user is required to close the stream.
   *
   * @return An Iterator over sentences (each a List of word tokens).
   * Although the type is given as {@code List<HasWord>}, in practice you get a List of CoreLabel,
   * and you can cast down to that. (Someday we might manage to fix the generic typing....)
   */
  @Override
  public Iterator<List<HasWord>> iterator() {
    // Add new document types here
    if (docType == DocType.Plain) {
      return new PlainTextIterator();
    } else if (docType == DocType.XML) {
      return new XMLIterator();
    } else {
      throw new IllegalStateException("Someone didn't add a handler for a new docType.");
    }
  }


  private class PlainTextIterator implements Iterator<List<HasWord>> {

    private final Tokenizer<? extends HasWord> tokenizer;
    private final Set<String> sentDelims;
    private final Set<String> delimFollowers;
    private final Function<String, String[]> splitTag;
    private List<HasWord> nextSent; // = null;
    private final List<HasWord> nextSentCarryover = Generics.newArrayList();

    public PlainTextIterator() {
      // Establish how to find sentence boundaries
      boolean eolIsSignificant = false;
      sentDelims = Generics.newHashSet();
      if (sentenceDelimiter == null) {

View on GitHub (pinned to 1b7edd19c4)