datafusion_python/
lib.rs

1// Licensed to the Apache Software Foundation (ASF) under one
2// or more contributor license agreements.  See the NOTICE file
3// distributed with this work for additional information
4// regarding copyright ownership.  The ASF licenses this file
5// to you under the Apache License, Version 2.0 (the
6// "License"); you may not use this file except in compliance
7// with the License.  You may obtain a copy of the License at
8//
9//   http://www.apache.org/licenses/LICENSE-2.0
10//
11// Unless required by applicable law or agreed to in writing,
12// software distributed under the License is distributed on an
13// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
14// KIND, either express or implied.  See the License for the
15// specific language governing permissions and limitations
16// under the License.
17
18#[cfg(feature = "mimalloc")]
19use mimalloc::MiMalloc;
20use pyo3::prelude::*;
21
22// Re-export Apache Arrow DataFusion dependencies
23pub use datafusion;
24pub use datafusion::common as datafusion_common;
25pub use datafusion::logical_expr as datafusion_expr;
26pub use datafusion::optimizer;
27pub use datafusion::sql as datafusion_sql;
28
29#[cfg(feature = "substrait")]
30pub use datafusion_substrait;
31
32#[allow(clippy::borrow_deref_ref)]
33pub mod catalog;
34pub mod common;
35
36#[allow(clippy::borrow_deref_ref)]
37mod config;
38#[allow(clippy::borrow_deref_ref)]
39pub mod context;
40#[allow(clippy::borrow_deref_ref)]
41pub mod dataframe;
42mod dataset;
43mod dataset_exec;
44pub mod errors;
45#[allow(clippy::borrow_deref_ref)]
46pub mod expr;
47#[allow(clippy::borrow_deref_ref)]
48mod functions;
49pub mod physical_plan;
50mod pyarrow_filter_expression;
51pub mod pyarrow_util;
52mod record_batch;
53pub mod sql;
54pub mod store;
55
56#[cfg(feature = "substrait")]
57pub mod substrait;
58#[allow(clippy::borrow_deref_ref)]
59mod udaf;
60#[allow(clippy::borrow_deref_ref)]
61mod udf;
62mod udwf;
63pub mod utils;
64
65#[cfg(feature = "mimalloc")]
66#[global_allocator]
67static GLOBAL: MiMalloc = MiMalloc;
68
69// Used to define Tokio Runtime as a Python module attribute
70pub(crate) struct TokioRuntime(tokio::runtime::Runtime);
71
72/// Low-level DataFusion internal package.
73///
74/// The higher-level public API is defined in pure python files under the
75/// datafusion directory.
76#[pymodule]
77fn _internal(py: Python, m: Bound<'_, PyModule>) -> PyResult<()> {
78    // Register the python classes
79    m.add_class::<catalog::PyCatalog>()?;
80    m.add_class::<catalog::PyDatabase>()?;
81    m.add_class::<catalog::PyTable>()?;
82    m.add_class::<context::PyRuntimeEnvBuilder>()?;
83    m.add_class::<context::PySessionConfig>()?;
84    m.add_class::<context::PySessionContext>()?;
85    m.add_class::<context::PySQLOptions>()?;
86    m.add_class::<dataframe::PyDataFrame>()?;
87    m.add_class::<udf::PyScalarUDF>()?;
88    m.add_class::<udaf::PyAggregateUDF>()?;
89    m.add_class::<udwf::PyWindowUDF>()?;
90    m.add_class::<config::PyConfig>()?;
91    m.add_class::<sql::logical::PyLogicalPlan>()?;
92    m.add_class::<physical_plan::PyExecutionPlan>()?;
93    m.add_class::<record_batch::PyRecordBatch>()?;
94    m.add_class::<record_batch::PyRecordBatchStream>()?;
95
96    // Register `common` as a submodule. Matching `datafusion-common` https://docs.rs/datafusion-common/latest/datafusion_common/
97    let common = PyModule::new(py, "common")?;
98    common::init_module(&common)?;
99    m.add_submodule(&common)?;
100
101    // Register `expr` as a submodule. Matching `datafusion-expr` https://docs.rs/datafusion-expr/latest/datafusion_expr/
102    let expr = PyModule::new(py, "expr")?;
103    expr::init_module(&expr)?;
104    m.add_submodule(&expr)?;
105
106    // Register the functions as a submodule
107    let funcs = PyModule::new(py, "functions")?;
108    functions::init_module(&funcs)?;
109    m.add_submodule(&funcs)?;
110
111    let store = PyModule::new(py, "object_store")?;
112    store::init_module(&store)?;
113    m.add_submodule(&store)?;
114
115    // Register substrait as a submodule
116    #[cfg(feature = "substrait")]
117    setup_substrait_module(py, &m)?;
118
119    Ok(())
120}
121
122#[cfg(feature = "substrait")]
123fn setup_substrait_module(py: Python, m: &Bound<'_, PyModule>) -> PyResult<()> {
124    let substrait = PyModule::new(py, "substrait")?;
125    substrait::init_module(&substrait)?;
126    m.add_submodule(&substrait)?;
127    Ok(())
128}
datafusion_python/lib.rs

datafusion_python/
lib.rs
