pyflink.table.expression.Expression.json_query#

Expression.json_query(path: str, wrapping_behavior=JsonQueryWrapper.WITHOUT_ARRAY, on_empty=JsonQueryOnEmptyOrError.NULL, on_error=JsonQueryOnEmptyOrError.NULL) → pyflink.table.expression.Expression[source]#

Extracts JSON values from a JSON string.

This follows the ISO/IEC TR 19075-6 specification for JSON support in SQL. The result is always returned as a STRING.

The wrapping_behavior determines whether the extracted value should be wrapped into an array, and whether to do so unconditionally or only if the value itself isn’t an array already.

on_empty and on_error determine the behavior in case the path expression is empty, or in case an error was raised, respectively. By default, in both cases null is returned. Other choices are to use an empty array, an empty object, or to raise an error.

seealso:: json_value()

Examples:

>>> lit('{"a":{"b":1}}').json_query('$.a') # '{"b":1}'
>>> lit('[1,2]').json_query('$') # '[1,2]'
>>> null_of(DataTypes.STRING()).json_query('$') # None

>>> lit('{}').json_query('$', JsonQueryWrapper.CONDITIONAL_ARRAY) # '[{}]'
>>> lit('[1,2]').json_query('$', JsonQueryWrapper.CONDITIONAL_ARRAY) # '[1,2]'
>>> lit('[1,2]').json_query('$', JsonQueryWrapper.UNCONDITIONAL_ARRAY) # '[[1,2]]'

>>> lit(1).json_query('$') # null
>>> lit(1).json_query('$', JsonQueryWrapper.CONDITIONAL_ARRAY) # '[1]'

>>> lit('{}').json_query('lax $.invalid', JsonQueryWrapper.WITHOUT_ARRAY,                                      JsonQueryOnEmptyOrError.EMPTY_OBJECT,                                     JsonQueryOnEmptyOrError.NULL) # '{}'
>>> lit('{}').json_query('strict $.invalid', JsonQueryWrapper.WITHOUT_ARRAY,                                      JsonQueryOnEmptyOrError.NULL,                                      JsonQueryOnEmptyOrError.EMPTY_ARRAY) # '[]'

New in version 1.12.0.