stanfordnlp/CoreNLP · error · RuntimeException

Expected a property " + name + ".model

Error message

Expected a property " + name + ".model

What it means

ArabicSegmenterAnnotator requires a segmentation model file. After scanning the properties for '<name>.model' (e.g. arabic.model), if no model path was found it throws this RuntimeException before loadModel().

Solutions

  1. Set the model property, e.g. props.setProperty("arabicsegmenter.model", "data/arabic-segmenter-atbtrain.ser.gz")
  2. Ensure the model file exists and is a valid serialized segmenter model
  3. Confirm the property prefix matches the name the annotator was created with

Example fix

// before
props.setProperty("annotators", "arabicsegmenter");
// after
props.setProperty("annotators", "arabicsegmenter");
props.setProperty("arabicsegmenter.model", "edu/stanford/nlp/models/arabic/segmentation/arabic-segmenter-atbtrain.ser.gz");
Defensive patterns

Strategy: validation

Validate before calling

if (props.getProperty("arabicsegmenter.model") == null && props.getProperty("model") == null) {
  throw new IllegalArgumentException("arabicsegmenter requires a model property");
}

Try / catch

try {
  new ArabicSegmenterAnnotator(name, props);
} catch (RuntimeException e) {
  if (e.getMessage().contains("Expected a property")) {
    // set the .model property and retry
  } else throw e;
}

Prevention

When it happens

Trigger: Adding the 'arabicsegmenter' annotator to a pipeline without setting the '<prefix>.model' property, or setting a misspelled key the property scan doesn't recognize.

Common situations: Forgetting to download the Arabic segmentation model; using the wrong properties prefix; model property set under a name that doesn't match the annotator's configured name.

Understand the failure class

Background: "is required", "must be set", "missing required field": configuration validation errors across open-source libraries — this error's family across 36 libraries.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/55dae625aa92a355. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/pipeline/ArabicSegmenterAnnotator.java:76

  public ArabicSegmenterAnnotator(String name, Properties props) {
    String model = null;
    // Keep only the properties that apply to this annotator
    Properties modelProps = new Properties();
    String desiredKey = name + '.';
    for (String key : props.stringPropertyNames()) {
      if (key.startsWith(desiredKey)) {
        // skip past name and the subsequent "."
        String modelKey = key.substring(desiredKey.length());
        if (modelKey.equals("model")) {
          model = props.getProperty(key);
        } else {
          modelProps.setProperty(modelKey, props.getProperty(key));
        }
      }
    }
    this.VERBOSE = PropertiesUtils.getBool(props, name + ".verbose", false);
    if (model == null) {
      throw new RuntimeException("Expected a property " + name + ".model");
    }
    loadModel(model, modelProps);

    // TODO: unify with ChineseSegmenterAnnotator somehow?
    // The issue here is the Chinese segmenter returns text chunks and
    // the Arabic segmenter has a method which returns CoreLabels, so
    // the project of unifying the two into one ur-SegmenterAnnotator
    // is larger than simply ripping some code into a superclass and
    // calling it a day

    // If newlines are treated as sentence split, we need to retain them in tokenization for ssplit to make use of them
    tokenizeNewline = (!props.getProperty(StanfordCoreNLP.NEWLINE_IS_SENTENCE_BREAK_PROPERTY, "never").equals("never"))
            || Boolean.valueOf(props.getProperty(StanfordCoreNLP.NEWLINE_SPLITTER_PROPERTY, "false"));

    // record whether or not sentence splitting on two newlines ; if so, need to remove single newlines
    sentenceSplitOnTwoNewlines =
        props.getProperty(StanfordCoreNLP.NEWLINE_IS_SENTENCE_BREAK_PROPERTY, "never").equals("two");
  }

View on GitHub (pinned to 1b7edd19c4)