Use structured, semi-structured, and unstructured data — SnowPro® Core Certification

Official references for this objective

• Snowflake — Querying Semi-structured Data | Snowflake Documentation

  the first two paths are equivalent, but the third is not

• Snowflake — Semi-structured data types | Snowflake Documentation

  A Snowflake OBJECT value is analogous to a JSON “object” .

• Snowflake — Structured data types | Snowflake Documentation

  The OBJECT value can’t contain any additional keys that aren’t specified in the OBJECT type. If there are additional keys, the cast fails.

• Snowflake — Introduction to loading semi-structured data | Snowflake Documentation

  If your data is a set of key-value pairs, you can load it into a column of type OBJECT.

• Snowflake — FLATTEN | Snowflake Documentation

  OUTER => TRUE )) f ; +-----+------+------+-------+-------+------+ | SEQ | KEY | PATH | INDEX | VALUE | THIS | |-----+------+------+-------+-------+------| | 1 | NULL

• Snowflake — Unstructured data types | Snowflake Documentation

  The FILE data type currently cannot be used in: CLUSTER BY, GROUP BY, and ORDER BY clauses Hybrid tables, Iceberg tables, and external tables

• Snowflake — Preparing to load data | Snowflake Documentation

  For JSON, Avro, ORC, and Parquet data, each top-level, complete object is loaded as a separate row in the table.

• Snowflake — Supported formats for semi-structured data | Snowflake Documentation

  The instance number is 0-based, not 1-based. So, the following queries are equivalent

• Snowflake — Transform data during a load | Snowflake Documentation

  Note that COPY transformations do not support the FLATTEN function, or JOIN or GROUP BY (aggregate) syntax.