apache/druid · error · ParseException

Could not convert value [%s] to float for dimension [%s]. In

Error message

Could not convert value [%s] to float for dimension [%s]. Invalid type: [%s]

What it means

convertObjectToFloat throws this ParseException when the value's runtime type is not convertible to float (not Number, String, or List — e.g. a Map or boolean). The message reports the value and its class.

Source

Thrown at processing/src/main/java/org/apache/druid/segment/DimensionHandlerUtils.java:551

            message
        );
      } else {
        final String message;
        if (fieldName != null) {
          message = StringUtils.nonStrictFormat(
              "Could not convert value [%s] to float for dimension [%s]. Invalid type: [%s]",
              valObj,
              fieldName,
              valObj.getClass()
          );
        } else {
          message = StringUtils.nonStrictFormat(
              "Could not convert value [%s] to float. Invalid type: [%s]",
              valObj,
              valObj.getClass()
          );
        }
        throw new ParseException(
            valObj.getClass().toString(),
            message
        );
      }
    }
  }

  @Nullable
  public static Object convertObjectToType(
      @Nullable final Object obj,
      final TypeSignature<ValueType> type,
      final boolean reportParseExceptions,
      @Nullable final String fieldName
  )
  {
    Preconditions.checkNotNull(type, "type");

    switch (type.getType()) {

View on GitHub (pinned to 9b90983fd2)

Solutions

Extract the numeric sub-field with a transform/json_query before conversion.
Correct upstream serialization so floats are plain numbers.
Type the column as string (or nested) if data is structured.
Validate the input schema against samples before submitting the ingestion spec.

Example fix

// before: float column gets {"v": 1.2}
// after: transform extracts v
"transforms": [{"type": "expression", "name": "v", "expression": "CAST(json_value(payload, '$.v') AS DOUBLE)"}]

Defensive patterns

Strategy: type-guard

Validate before calling

boolean convertibleToFloat(Object v) {
  return v == null || v instanceof Number || v instanceof String;
}

Type guard

Float asFloat(Object v) {
  if (v instanceof Number) return ((Number) v).floatValue();
  if (v instanceof String) { try { return Float.parseFloat((String) v); } catch (NumberFormatException e) { return null; } }
  return null;
}

Try / catch

try {
  return DimensionHandlerUtils.convertObjectToFloat(value, fieldName);
} catch (ParseException e) {
  log.warn("Invalid type [%s] for float field [%s], dropping", value.getClass(), fieldName);
  return null;
}

Prevention

Verify value.getClass() is Number or String before conversion
Extract numeric sub-fields from nested objects via json_value transforms
JSON-schema validate producer payloads
Watch for auto-detected column types on heterogeneous data

When it happens

Trigger: A float column receives a nested JSON object, boolean, or other unhandled Java type; convertObjectToFloat falls to the final else; also reachable via convertObjectToType.

Common situations: Nested payloads mapped directly to a numeric metric column; heterogeneous upstream schemas; auto-detection mistyping a column as float.

Understand the failure class

Background: Type mismatch errors: IllegalArgumentException, TypeError and type guards across 150 open-source libraries — this error's family across 150 libraries.

Related errors

AI-assisted analysis of apache/druid@9b90983fd2 (2026-09-07). Data as JSON: /api/errors/4858164bd7457893. Report an issue: GitHub.