sqloxide wraps rust bindings for sqlparser-rs into a python package using pyO3.
The original goal of this project was to have a very fast, efficient, and accurate SQL parser I could use for building data lineage graphs across large code bases (think hundreds of auto-generated .sql files). Most existing sql parsing approaches for python are either very slow or not accurate (especially in regards to deeply nested queries, sub-selects and/or table aliases). Looking to the rust community for support, I found the excellent sqlparser-rs crate which is quite easy to wrap in python code.
The project provides manylinux2014 wheels on pypi so it should be compatible with most linux distributions. Native wheels are also now available for OSX and Windows.
To install from pypi:
pip install sqloxideParsing a SQL query is relatively straight forward:
from sqloxide import parse_sql
sql = """
SELECT employee.first_name, employee.last_name,
call.start_time, call.end_time, call_outcome.outcome_text
FROM employee
INNER JOIN call ON call.employee_id = employee.id
INNER JOIN call_outcome ON call.call_outcome_id = call_outcome.id
ORDER BY call.start_time ASC;
"""
output = parse_sql(sql=sql, dialect='ansi')
print(output)
>>> [
{
"Query": {
"ctes": [],
"body": {
"Select": {
"distinct": false,
"top": null,
"projection": [
{
"UnnamedExpr": {
"CompoundIdentifier": [
{
"value": "employee",
"quote_style": null
},
{
"value": "first_name",
"quote_style": null
}
]
}
},
{
"UnnamedExpr": {
"CompoundIdentifier": [
{
"value": "employee",
"quote_style": null
},
{
"value": "last_name",
"quote_style": null
}
]
}
},
{
"UnnamedExpr": {
"CompoundIdentifier": [
{
"value": "call",
"quote_style": null
},
{
"value": "start_time",
"quote_style": null
}
]
}
},
{ # OUTPUT TRUNCATEDNote that you get back what looks like a JSON document but in actual python types, this is a typed AST that matches the sqlparser-rs AST schema.
We can convert this AST back into a SQL query by running:
from sqloxide import restore_ast
query = restore_ast(ast=output)
print(query)This reconstruction is helpful if you want to make manual edits to the AST in python.
If you want a more structured approach to AST edits, we also expose APIs that allow you to use the visitor pattern to make query modifications.
Here is an example for mutating a subset of the expressions in the query to be SHOUTING UPPERCASE:
from sqloxide import parse_sql, mutate_expressions
sql = "SELECT something from somewhere where something = 1 and something_else = 2"
def func(x):
if "CompoundIdentifier" in x.keys():
for y in x["CompoundIdentifier"]:
y["value"] = y["value"].upper()
return x
ast = parse_sql(sql=sql, dialect="ansi")
result = mutate_expressions(parsed_query=ast, func=func)
print(result)
---
>>> ['SELECT something FROM somewhere WHERE something = 1 AND something_else = 2']What if you needed to make a structured edit to the table name in the above query? There is also an API for that as well:
from sqloxide import parse_sql, mutate_relations
def func(x):
return x.replace("somewhere", "anywhere")
result = mutate_relations(parsed_query=ast, func=func)
print(result)
---
>>> ['SELECT something FROM anywhere WHERE something = 1 AND something_else = 2']These features combined allow for powerful semantic rewrites of queries, if you have any examples you'd like to share please contribue back to the examples/ folder!
We run 4 benchmarks, comparing to some python native sql parsing libraries:
test_sqloxide- parse query and get a python object back from rusttest_sqlparser- testing sqlparse, query -> ASTtest_mozsqlparser- testing moz-sql-parser, full roundtrip as in the docs, query -> JSONtest_sqlglot- testing sqlglot, query -> AST
To run them on your machine:
poetry run pytest tests/benchmark.py
------------------------------------------------------------------------------------------- benchmark: 4 tests -------------------------------------------------------------------------------------------
Name (time in us) Min Max Mean StdDev Median IQR Outliers OPS Rounds Iterations
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_sqloxide 29.6800 (1.0) 50.4300 (1.0) 30.6219 (1.0) 0.7367 (1.0) 30.4900 (1.0) 0.2390 (1.0) 527;716 32,656.3811 (1.0) 9099 1
test_sqlglot 365.8420 (12.33) 692.8950 (13.74) 377.2422 (12.32) 11.7692 (15.98) 375.7825 (12.32) 4.3145 (18.05) 62;97 2,650.8168 (0.08) 2260 1
test_sqlparser 1,577.7720 (53.16) 9,751.9699 (193.38) 1,651.5547 (53.93) 355.5511 (482.64) 1,620.7315 (53.16) 30.9200 (129.37) 3;60 605.4901 (0.02) 538 1
test_mozsqlparser 2,793.8400 (94.13) 12,358.7790 (245.07) 3,091.8519 (100.97) 960.4173 (>1000.0) 2,937.6310 (96.35) 243.3220 (>1000.0) 4;4 323.4308 (0.01) 316 1
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The depgraph example reads a bunch of .sql files from disk using glob, and builds a dependency graph of all of the objects using graphviz.
poetry run python ./examples/depgraph.py --path {path/to/folder/with/queries}
-
Install
rustup -
poetry installwill automatically create the venv, compile the package and install it into the venv via the build script.