stanfordnlp/CoreNLP · error · NumberFormatException

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

Error message

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

What it means

wordToNumber treats pieces ending in 'ths'/'rds' (mid-string) as fractions by stripping the final 's' and looking up ordWord2NumMap; if the lookup misses it throws NumberFormatException naming the piece, original string, and piece index. This is the fractional-word guard inside the multi-piece branch.

Solutions

  1. Rewrite the fraction into numeric form ('three fourths' -> '0.75') before normalization.
  2. Catch NumberFormatException and leave the token unnormalized.
  3. Verify the ordinal spelling is supported by ordWord2NumMap (fourths, thirds, halves, etc.).

Example fix

// before
Number n = NumberNormalizer.wordToNumber("sevens ths");
// after
Number n = NumberNormalizer.wordToNumber("0.75"); // pre-normalized fraction
Defensive patterns

Strategy: try-catch

Validate before calling

static boolean isKnownFractionWord(String part) {
  return part.matches("(half|halves|third|thirds|fourth|fourths|fifth|fifths|...)s?");
}

Try / catch

try {
  Number n = NumberNormalizer.wordToNumber(input);
} catch (NumberFormatException e) {
  Number n = null; // unsupported fraction word; skip
}

Prevention

When it happens

Trigger: A mid-string piece ending in 'ths' or 'rds' whose base form is not in the ordinal map, e.g. 'sixteenths' combined with other pieces, or misspelled ordinals like 'oneths'.

Common situations: Tokenizing unusual numeric phrases ('thirds half'), ordinal forms not present in the built-in word map (rare fractions like 'fifths' in some versions), or text with typos.

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/fa06d58b18b3b00a. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/ie/NumberNormalizer.java:374

      Matcher m = alphaPattern.matcher(curPart);
      if (m.find()) {
        // Some part of the word has alpha characters
        Number curNum;
        if (word2NumMap.containsKey(curPart)) {
          curNum = word2NumMap.get(curPart);
        } else if (ordWord2NumMap.containsKey(curPart)) {
          if (curIndex == numWords-1){
            curNum = ordWord2NumMap.get(curPart);
          } else {
            throw new NumberFormatException("Error in wordToNumber function.");
          }
        } else if (curIndex > 0 && (curPart.endsWith("ths") || curPart.endsWith("rds"))) {
          // Fractions?
          curNum = ordWord2NumMap.get(curPart.substring(0, curPart.length()-1));
          if (curNum != null) {
            curNum = 1/curNum.doubleValue();
          } else {
            throw new NumberFormatException("Bad number put into wordToNumber.  Word is: \"" + curPart + "\", originally part of \"" + originalString + "\", piece # " + curIndex);
          }
        } else if (Character.isDigit(curPart.charAt(0)) || curPart.charAt(0) == '.') {
          if (curPart.endsWith("th") || curPart.endsWith("rd") || curPart.endsWith("nd") || curPart.endsWith("st")) {
            curPart = curPart.substring(0, curPart.length()-2).trim();
          }
          curNum = parseNumberPart(curPart, originalString, curIndex);
        } else {
          throw new NumberFormatException("Bad number put into wordToNumber.  Word is: \"" + curPart + "\", originally part of \"" + originalString + "\", piece # " + curIndex);
        }
        numFields[curIndex] = curNum;
      } else {
        // Word is all numeric
        Matcher matcher = digitsPatternExtended.matcher(curPart);
        if (matcher.matches()) {
          numFields[curIndex] = parseNumberPart(curPart, originalString, curIndex);
        } else if (numPattern.matcher(curPart).matches()) {
          numFields[curIndex] = new BigDecimal(curPart);
        } else {

View on GitHub (pinned to 1b7edd19c4)