stanfordnlp/CoreNLP · error · UnsupportedOperationException

transform on actions not yet implemented

Error message

transform on actions not yet implemented

What it means

SequencePattern.transform() applies a NodePatternTransformer to the pattern's expression tree, but if the pattern carries an action (SequenceMatchAction) the transformation would lose or mis-handle it, so it throws UnsupportedOperationException ('transform on actions not yet implemented'). It is a declared feature gap, not a misuse error.

Solutions

1. Strip or split off the action before transforming; transform only the pattern part and re-attach the action afterward
2. Rebuild the pattern from a transformed pattern string that omits the action
3. Implement the missing transform support by extending the transformer to carry actions
4. Avoid transform() and recompile an equivalent pattern string instead

Example fix

// before
SequencePattern<CoreMap> t = patternWithAction.transform(transformer); // throws
// after
SequencePattern<CoreMap> t = new SequencePattern<>(patternStr,
    patternWithAction.getPatternExpr().transform(transformer),
    patternWithAction.getAction()); // keep action manually

if (pattern.getAction() != null) {
  throw new IllegalArgumentException("Cannot transform a pattern that has an action attached");
}

boolean isTransformable(SequencePattern<?> p) {
  return p.getAction() == null;
}

try {
  SequencePattern<T2> t = pattern.transform(transformer);
} catch (UnsupportedOperationException e) {
  // pattern has an action; handle pattern without transform
}

When it happens

Trigger: Calling pattern.transform(transformer) on a SequencePattern that was compiled with an action (e.g. a pattern string with ' :: ' action syntax, so pattern.getAction() != null).

Common situations: Programmatic pattern rewriting of TokensRegex patterns that include side-effect actions like tagging or annotation, common when normalizing or porting rule files.

Related errors

• BackRefPatternExpr.transform not implemented yet!!! Please…
• Unsupported findType
• Cannot evaluate type
• LogPrior.getSigmaSquaredM is undefined for any prior but…
• If you want to ask for the probability, you must train a…

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/b81621063d466d18. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/ling/tokensregex/SequencePattern.java:142

    nodeSequencePattern = new GroupPatternExpr(nodeSequencePattern, true);
    nodeSequencePattern = nodeSequencePattern.optimize();
    this.totalGroups = nodeSequencePattern.assignGroupIds(0);
    Frag f = nodeSequencePattern.build();
    f.connect(MATCH_STATE);
    this.root = f.start;
    varGroupBindings = new VarGroupBindings(totalGroups+1);
    nodeSequencePattern.updateBindings(varGroupBindings);
  }

  @Override
  public String toString() {
    return this.pattern();
  }

  public <T2> SequencePattern<T2> transform(NodePatternTransformer<T,T2> transformer) {
    if (action != null) {
      throw new UnsupportedOperationException("transform on actions not yet implemented");
    }
    SequencePattern.PatternExpr transformedPattern = this.patternExpr.transform(transformer);
    // TODO: Make string unique by indicating this pattern was transformed
    return new SequencePattern<>(this.patternStr, transformedPattern, null);
  }

  public String pattern() {
    return patternStr;
  }

  protected PatternExpr getPatternExpr() {
    return patternExpr;
  }

  public double getPriority() {
    return priority;
  }

View on GitHub (pinned to 1b7edd19c4)