py-partiql-parser

A tokenizer/parser/executor for the PartiQL-language, in Python.

Much beta, such wow. Feel free to raise any issues you encounter.

S3 Usage

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")

DynamoDB Usage

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"}])

Meat

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.

Outstanding

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.

Notes

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 cannot select key for each document in the list)
select values is not supported