Rules and Restrictions - Advanced SQL Engine - Teradata Database

DATASET Data Type

Product
Advanced SQL Engine
Teradata Database
Release Number
17.05
17.00
Published
June 2020
Language
English (United States)
Last Update
2021-01-23
dita:mapPath
des1556232910526.ditamap
dita:ditavalPath
lze1555437562152.ditaval
dita:id
B035-1198
lifecycle
previous
Product Category
Teradata Vantageā„¢
The input data to the AvroContainerSplit table operator, as determined by the result of the SELECT statement in the ON clause, must be one or more rows of data, each with two columns defined as follows:
  • container_id includes the following types:
    • BYTEINT
    • SMALLINT
    • INTEGER
    • BIGINT
    • DOUBLE PRECISION
    • DECIMAL
    • NUMBER
    • CHAR
    • VARCHAR
  • container
An error is returned if the input data is not one of these data types, or if the input rows contain more or less than two columns. If the container value is a null value, a row with an avro_object_id and avro_value set to null is returned. The result of AvroContainerSplit is one or more output rows, where each row represents an individual Avro value from the Object Container File. Three output columns are returned as follows:
  • out_container_id: The same type and value as the container_id input column. out_container_id identifies the Object Container File that the Avro value came from.
  • avro_object_id: Each Avro value from a container is numbered from 0 to n-1, where n is the number of Avro values within the Object Container File. The output column is type INTEGER.
  • avro_val: The Avro Dataset value.

The Object Container File can be uncompressed (null codec) or compressed (deflate codec). The schema within the Object Container File is limited to 16 MB, and the uncompressed size of any file data block within the Object Container file is also limited to 16 MB.