stanfordnlp/CoreNLP · error · NumberFormatException

Bad number put into wordToNumber. Word is: \"" + input +…

Error message

Bad number put into wordToNumber.  Word is: \"" + input + "\", originally part of \"" + originalString + "\", piece # " + curIndex

What it means

NumberNormalizer.parseNumberPart converts a numeric word piece via Long.parseLong/Double.parseDouble after normalizing commas and trailing 's'; if the piece matches no numeric pattern (numPattern / digitsPatternExtended) it throws NumberFormatException naming the input, the original string, and the piece index. It is a guard for non-numeric text routed into numeric parsing inside wordToNumber.

Solutions

Pre-clean the token so only numeric pieces (digits, commas, periods, optional trailing 's') reach wordToNumber.
Catch NumberFormatException around the normalizer call and treat the token as non-numeric.
Check the exact failing word/piece index reported in the message and strip unexpected characters before calling.

Example fix

// before
Number n = NumberNormalizer.wordToNumber("3rd-place");
// after
String token = "3rd-place".split("-")[0]; // pass only numeric piece
Number n = NumberNormalizer.wordToNumber(token);

Defensive patterns

Strategy: try-catch

Validate before calling

static boolean isParseableNumberPiece(String s) {
  return s != null && s.replaceAll("[,]", "").matches("\\d+(\\.\\d+)?s?");
}

Try / catch

try {
  Number n = NumberNormalizer.wordToNumber(input);
} catch (NumberFormatException e) {
  Number n = null; // treat token as non-numeric
}

Prevention

Pass only clean numeric tokens; strip letters, currency, and stray punctuation first.
Catch NumberFormatException at mention-normalization boundaries.
Log the failing word from the message to fix upstream tokenization.

When it happens

Trigger: wordToNumber splitting a string and delegating a piece to parseNumberPart that contains characters beyond the accepted numeric formats (e.g. '1st2', '1.2.3', currency symbols, 'x100').

Common situations: Normalizing extracted mention text containing mixed alphanumeric tokens (like '3x' or 'one-two'), noisy OCR/entity text, or regex changes in numPattern across CoreNLP versions altering which strings parse.

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/1f2ce48eefe3348e. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/ie/NumberNormalizer.java:287

            break;
          default:
            // unknown magnitude! Ignore it.
            break;
        }
        if (digitsPattern.matcher(numPart).matches()) {
          return Long.parseLong(numPart) * magnitude;
        } else {
          return Double.parseDouble(numPart) * magnitude;
        }
      } else {
        if (digitsPattern.matcher(numPart).matches()) {
          return Long.parseLong(numPart);
        } else {
          return Double.parseDouble(numPart);
        }
      }
    } else{
      throw new NumberFormatException("Bad number put into wordToNumber.  Word is: \"" + input + "\", originally part of \"" + originalString + "\", piece # " + curIndex);
    }

  }


  /**
   * Fairly generous utility function to convert a string representing
   * a number (hopefully) to a Number.
   * Assumes that something else has somehow determined that the string
   * makes ONE suitable number.
   * The value of the number is determined by:
   * 0. Breaking up the string into pieces using whitespace
   *    (stuff like "and", "-", "," is turned into whitespace);
   * 1. Determining the numeric value of the pieces;
   * 2. Finding the numeric value of each piece;
   * 3. Combining the pieces together to form the overall value:
   *    a. Find the largest component and its value (X),
   *    b. Let B = overall value of pieces to the left (recursive),

View on GitHub (pinned to 1b7edd19c4)