alibaba/DataX · error · Exception

DATE类型数据 '%s' 格式不正确，必须为yyyy-mm-dd格式

Error message

DATE类型数据 '%s' 格式不正确，必须为yyyy-mm-dd格式

What it means

Thrown by CassandraWriterHelper's string-to-value converter for DATE columns. It expects a strict yyyy-mm-dd literal (exactly two '-' separators after split), and anything else — '2024/01/02', '2024-1-2', empty fragments, or a full timestamp — fails the a.length != 3 check. The message names the offending value and the required format.

Source

Thrown at cassandrawriter/src/main/java/com/alibaba/datax/plugin/writer/cassandrawriter/CassandraWriterHelper.java:102

    case BIGINT:
      return Long.valueOf(s);

    case VARINT:
      return new BigInteger(s, 10);

    case FLOAT:
      return Float.valueOf(s);

    case DOUBLE:
      return Double.valueOf(s);

    case DECIMAL:
      return new BigDecimal(s);

    case DATE: {
      String[] a = s.split("-");
      if (a.length != 3) {
        throw new Exception(String.format("DATE类型数据 '%s' 格式不正确，必须为yyyy-mm-dd格式", s));
      }
      return LocalDate.fromYearMonthDay(Integer.valueOf(a[0]), Integer.valueOf(a[1]),
          Integer.valueOf(a[2]));
    }

    case TIME:
      return Long.valueOf(s);

    case TIMESTAMP:
      return new Date(Long.valueOf(s));

    case UUID:
    case TIMEUUID:
      return UUID.fromString(s);

    case INET:
      String[] b = s.split("/");
      if (b.length < 2) {

View on GitHub (pinned to 80ec23d5c5)

Solutions

1. Normalize the source value to yyyy-mm-dd before it reaches cassandrawriter (convert in the reader or via a transform)
2. If the data is really a timestamp, change the Cassandra column/writer type mapping to TIMESTAMP instead of DATE
3. Fix malformed rows in the source data (wrong separators, missing parts)

Example fix

// before
{"name": "cassandrawriter", "parameter": {"column": [{"name": "d", "type": "DATE"}]}}
// source value: "2024/12/01" -> throws
// after: convert source to "2024-12-01" (yyyy-mm-dd) or declare the column as TIMESTAMP

// normalize before writing DATE columns
static String toIsoDate(String s) {
    String[] a = s.split("-");
    if (a.length != 3) throw new IllegalArgumentException("Not yyyy-mm-dd: " + s);
    return String.format("%04d-%02d-%02d", Integer.valueOf(a[0].trim()), Integer.valueOf(a[1].trim()), Integer.valueOf(a[2].trim()));
}

boolean isYyyyMmDd(String s) {
    return s != null && s.split("-").length == 3
        && s.matches("\\d{1,4}-\\d{1,2}-\\d{1,2}");
}

catch (Exception e) {
  if (e.getMessage() != null && e.getMessage().contains("yyyy-mm-dd")) {
    // route the row to a dirty-data sink or fix the source format
  }
  throw e;
}

When it happens

Trigger: A cassandrawriter column of CQL type DATE whose incoming string does not split on '-' into exactly 3 parts, e.g. '2024/12/01', '2024-12-01 00:00:00', or '2024-12'.

Common situations: Reader emitting timestamps or localized date strings into a DATE column; column configuration declaring a DATE type for data that is actually a timestamp; hand-written test data with mixed separators or missing zero-padding is fine ('2024-1-2' still splits to 3) but any format with != 3 dash-separated parts throws.

Related errors

• Date 类型转换错误：[%s]
• Date type conversion error: [%s]
• Unknown column type:
• Failed to flush data to Doris. %s

AI-assisted analysis of alibaba/DataX@80ec23d5c5 (2026-08-14). Data as JSON: /api/errors/bb03c4d17a35ebca. Report an issue: GitHub.