stanfordnlp/CoreNLP · error · IllegalArgumentException

TokensRegexNERAnnotator ERROR: Invalid group in line in…

Error message

TokensRegexNERAnnotator ERROR: Invalid group in line  in regexner file : ""!

What it means

When the header defines a 'group' column, each line's group value is parsed with Integer.parseInt; a non-integer throws IllegalArgumentException showing line number, mapping file, and line text. The group selects which capture group of the pattern gets annotated and must be a plain integer.

Solutions

  1. Replace the group value on the reported line with a plain integer (e.g. 0 or 1).
  2. Remove the group column from header and rows if you want the default group 0.
  3. Fix row alignment so each cell sits in its intended column.
  4. Ensure the integer stays within the pattern's group count (see the related 'Invalid match group' runtime error).

Example fix

// before
([ { word:/Dr\./ } ]) ([ { ner:PERSON } ])	TITLE	person	first
// after
([ { word:/Dr\./ } ]) ([ { ner:PERSON } ])	TITLE	person	2
Defensive patterns

Strategy: validation

Validate before calling

for (String[] row : rows) {
  String g = row[iGroup];
  if (g != null && !g.trim().isEmpty()) Integer.parseInt(g.trim()); // throws NumberFormatException early with context
}

Try / catch

try { annotator = new TokensRegexNERAnnotator(name, props); } catch (IllegalArgumentException e) { if (e.getMessage().contains("Invalid group")) { fixGroupOnLine(e.getMessage()); } else throw e; }

Prevention

When it happens

Trigger: A group column containing text like 'first', 'g1', '2.0', or an empty/whitespace value; a shifted row putting another cell (e.g. a description) into the group column.

Common situations: Hand-written mapping files with descriptive group names; copy-paste rows where the group cell was lost; rows misaligned after header changes so a string column lands under 'group'.

Understand the failure class

Background: "Invalid ... format", "must be in format X", "does not look like a ..." — invalid argument format errors across CLI tools and libraries — this error's family across 17 libraries.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/56a2d65ed92e4155. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/pipeline/TokensRegexNERAnnotator.java:800

      double weight = 0.0;
      if (iWeight >= 0 && split.length > iWeight) {
        try {
          weight = Double.parseDouble(split[iWeight].trim());
        } catch (NumberFormatException e) {
          throw new IllegalArgumentException("TokensRegexNERAnnotator " + annotatorName
              + " ERROR: Invalid weight in line " + lineCount
              + " in regexner file " + mappingFilename + ": \"" + line + "\"!", e);
        }
      }
      int annotateGroup = 0;
      // Get annotate group from input....
      if (iGroup>= 0 && split.length > iGroup) {
        // Which group to take (allow for context)
        String context = split[iGroup].trim();
        try {
          annotateGroup = Integer.parseInt(context);
        } catch (NumberFormatException e) {
          throw new IllegalArgumentException("TokensRegexNERAnnotator " + annotatorName
              + " ERROR: Invalid group in line " + lineCount
              + " in regexner file " + mappingFilename + ": \"" + line + "\"!", e);
        }
      }

      // Print some warnings about the type
      for (int i = 0; i < types.length; i++) {
        String type = types[i];
        // TODO: Have option to allow commas in types
        int commaPos = type.indexOf(',');
        if (commaPos > 0) {
          // Strip the "," and just take first type
          String newType = type.substring(0, commaPos).trim();
          logger.warn(annotatorName + ": Entry has multiple types for " +
                  annotationFieldnames[i] + ": " + line + ".  Taking type to be " + newType);
          types[i] = newType;
        }
      }

View on GitHub (pinned to 1b7edd19c4)