INCONSISTENT_BEHAVIOR_CROSS_VERSION error class

  • Article

SQLSTATE: 42K0B

You may get a different result due to the upgrading to:

DATETIME_PATTERN_RECOGNITION

Spark >= 3.0:

Fail to recognize <pattern> pattern in the DateTimeFormatter.

  1. You can set <config> to "LEGACY" to restore the behavior before Spark 3.0.
  2. You can form a valid datetime pattern with the guide from '<docroot>/sql-ref-datetime-pattern.html'.

DATETIME_WEEK_BASED_PATTERN

Spark >= 3.0:

All week-based patterns are unsupported since Spark 3.0, detected week-based character: <c>.

Please use the SQL function EXTRACT instead.

PARSE_DATETIME_BY_NEW_PARSER

Spark >= 3.0:

Fail to parse <datetime> in the new parser.

You can set <config> to "LEGACY" to restore the behavior before Spark 3.0, or set to "CORRECTED" and treat it as an invalid datetime string.

READ_ANCIENT_DATETIME

Spark >= 3.0:

reading dates before 1582-10-15 or timestamps before 1900-01-01T00:00:00Z

from <format> files can be ambiguous, as the files may be written by

Spark 2.x or legacy versions of Hive, which uses a legacy hybrid calendar

that is different from Spark 3.0+'s Proleptic Gregorian calendar.

See more details in SPARK-31404. You can set the SQL config <config> or

the datasource option <option> to "LEGACY" to rebase the datetime values

w.r.t. the calendar difference during reading. To read the datetime values

as it is, set the SQL config <config> or the datasource option <option>

to "CORRECTED".

TBD

Spark >= <sparkVersion>: <details>

WRITE_ANCIENT_DATETIME

Spark >= 3.0:

writing dates before 1582-10-15 or timestamps before 1900-01-01T00:00:00Z into <format> files can be dangerous, as the files may be read by Spark 2.x or legacy versions of Hive later, which uses a legacy hybrid calendar that is different from Spark 3.0+'s Proleptic Gregorian calendar.

See more details in SPARK-31404.

You can set <config> to "LEGACY" to rebase the datetime values w.r.t. the calendar difference during writing, to get maximum interoperability.

Or set the config to "CORRECTED" to write the datetime values as it is, if you are sure that the written files will only be read by Spark 3.0+ or other systems that use Proleptic Gregorian calendar.