stanfordnlp/CoreNLP · error · IllegalArgumentException

ChineseUtils: Unsupported parameter option: ascii=

Error message

ChineseUtils: Unsupported parameter option: ascii=

What it means

Inside ChineseUtils.normalizeBMP, the per-codepoint conversion switches on the ascii option; any value not matching a known case (LEAVE/ASCII/FULLWIDTH etc.) reaches the default branch and throws this IllegalArgumentException. It means the ascii normalization mode is a constant the BMP normalizer does not implement — a legal-but-unknown constant passes normalize's range check and only fails here.

Solutions

1. Pass one of the documented ascii constants from ChineseUtils (e.g. ASCII, FULLWIDTH, LEAVE) instead of a raw int
2. Log the ascii value at the failure point to identify the bad constant in your caller
3. Upgrade/synchronize your code and Stanford ChineseUtils versions so constants match

Example fix

// before
int asciiMode = 3; // unknown to the switch
String norm = ChineseUtils.normalize(in, asciiMode, spaceChar, midDot);
// after
String norm = ChineseUtils.normalize(in, ChineseUtils.FULLWIDTH, spaceChar, midDot);

Set<Integer> legalAscii = Set.of(ChineseUtils.LEAVE, ChineseUtils.ASCII, ChineseUtils.FULLWIDTH);
if (!legalAscii.contains(asciiMode)) throw new IllegalArgumentException("unknown ascii mode: " + asciiMode);

try {
  String norm = ChineseUtils.normalize(in, asciiMode, spaceChar, midDot);
} catch (IllegalArgumentException e) {
  if (!e.getMessage().contains("ascii=")) throw e;
  norm = ChineseUtils.normalize(in, ChineseUtils.LEAVE, spaceChar, midDot);
}

When it happens

Trigger: Calling ChineseUtils.normalize (or normalizeBMP directly) with an ascii value that is within [0, MAX_LEGAL] but not one of the recognized mode constants, so the switch falls through to default.

Common situations: Passing a raw integer that passes the range guard but maps to no case in the switch, or constants mixed from different library versions.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• ChineseUtils: Unknown parameter option
• ChineseUtils: Unsupported parameter option: midDot=
• We need at least 2 extractors for ExtractorMerger to make se
• Too many columns: <columnI>/<numColumns> (offset: <offset>)
• Too few columns: <columnI>/<numColumns> (offset: <offset>)

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/9aa13ede9dbf9ba0. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/trees/international/pennchinese/ChineseUtils.java:142

              cub == Character.UnicodeBlock.SUPPLEMENTARY_PRIVATE_USE_AREA_B) {
        EncodingPrintWriter.err.println("ChineseUtils.normalize warning: private use area codepoint U+" + Integer.toHexString(cp) + " in " + in);
      }
      boolean delete = false;
      switch (ascii) {
        case LEAVE:
          break;
        case ASCII:
          if (cp >= '\uFF01' && cp <= '\uFF5E') {
            cp -= (0xFF00 - 0x0020);
          }
          break;
        case FULLWIDTH:
          if (cp >= '\u0021' && cp <= '\u007E') {
            cp += (0xFF00 - 0x0020);
          }
          break;
        default:
          throw new IllegalArgumentException("ChineseUtils: Unsupported parameter option: ascii=" + ascii);
      }
      switch (spaceChar) {
        case LEAVE:
          break;
        case ASCII:
          if (Character.isSpaceChar(cp)) {
            cp = ' ';
          }
          break;
        case FULLWIDTH:
          if (Character.isSpaceChar(cp)) {
            cp = '\u3000';
          }
          break;
        case DELETE:
          if (Character.isSpaceChar(cp)) {
            delete = true;
          }

View on GitHub (pinned to 1b7edd19c4)