A tokenizer/parser/executor for the PartiQL-language, in Python.
Much beta, such wow. Feel free to raise any issues you encounter.
import json
from py_partiql_parser import S3SelectParser
original_json = json.dumps({"a1": "b1", "a2": "b2"})
parser = S3SelectParser(source_data={"s3object": original_json})
result = parser.parse("SELECT * FROM s3object")
import json
from py_partiql_parser import DynamoDBStatementParser
parser = DynamoDBStatementParser(source_data={"table1": {"a1": {"S": "b1"}, "a2": {"S": "b2"}}})
result = parser.parse("SELECT * from table1 WHERE a1 = ?", parameters=[{"S": "b1"}])
The important logic of this library can be found here: https://github.com/bblommers/py-partiql-parser/blob/main/py_partiql_parser/_internal/parser.py
It is implemented as a naive, dependency-free, TDD-first tokenizer.
- Support for functions such as
count(*)
- Support for CSV conversion. A start has been made in
_internal/csv_converter.py
- .. and I'm sure many other things.
The first iteration of this library was based on the spec, found here: https://partiql.org/assets/PartiQL-Specification.pdf
AWS doesn't follow its own spec though, most notably:
- a file containing a list (with multiple JSON documents) cannot be queried normally (
select *
returns everything, but you cannotselect key
for each document in the list) select values
is not supported
To report a security vulnerability, please use the Tidelift security contact. Tidelift will coordinate the fix and disclosure.