blob: fe324d40fd9529c4ea47733cffe567cb96c48c54 [file]
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
//! This module defines the interface for logical nodes
use crate::{Expr, LogicalPlan};
use datafusion_common::{DFSchema, DFSchemaRef, Result};
use std::cmp::Ordering;
use std::hash::{Hash, Hasher};
use std::{any::Any, collections::HashSet, fmt, sync::Arc};
use super::InvariantLevel;
/// This defines the interface for [`LogicalPlan`] nodes that can be
/// used to extend DataFusion with custom relational operators.
///
/// The [`UserDefinedLogicalNodeCore`] trait is *the recommended way to implement*
/// this trait and avoids having implementing some required boiler plate code.
pub trait UserDefinedLogicalNode: fmt::Debug + Send + Sync {
/// Return a reference to self as Any, to support dynamic downcasting
///
/// Typically this will look like:
///
/// ```
/// # use std::any::Any;
/// # struct Dummy { }
///
/// # impl Dummy {
/// // canonical boiler plate
/// fn as_any(&self) -> &dyn Any {
/// self
/// }
/// # }
/// ```
fn as_any(&self) -> &dyn Any;
/// Return the plan's name.
fn name(&self) -> &str;
/// Return the logical plan's inputs.
fn inputs(&self) -> Vec<&LogicalPlan>;
/// Return the output schema of this logical plan node.
fn schema(&self) -> &DFSchemaRef;
/// Perform check of invariants for the extension node.
fn check_invariants(&self, check: InvariantLevel) -> Result<()>;
/// Returns all expressions in the current logical plan node. This should
/// not include expressions of any inputs (aka non-recursively).
///
/// These expressions are used for optimizer
/// passes and rewrites. See [`LogicalPlan::expressions`] for more details.
fn expressions(&self) -> Vec<Expr>;
/// A list of output columns (e.g. the names of columns in
/// self.schema()) for which predicates can not be pushed below
/// this node without changing the output.
///
/// By default, this returns all columns and thus prevents any
/// predicates from being pushed below this node.
fn prevent_predicate_push_down_columns(&self) -> HashSet<String> {
// default (safe) is all columns in the schema.
get_all_columns_from_schema(self.schema())
}
/// Write a single line, human readable string to `f` for use in explain plan.
///
/// For example: `TopK: k=10`
fn fmt_for_explain(&self, f: &mut fmt::Formatter) -> fmt::Result;
/// Create a new `UserDefinedLogicalNode` with the specified children
/// and expressions. This function is used during optimization
/// when the plan is being rewritten and a new instance of the
/// `UserDefinedLogicalNode` must be created.
///
/// Note that exprs and inputs are in the same order as the result
/// of self.inputs and self.exprs.
///
/// So, `self.with_exprs_and_inputs(exprs, ..).expressions() == exprs
fn with_exprs_and_inputs(
&self,
exprs: Vec<Expr>,
inputs: Vec<LogicalPlan>,
) -> Result<Arc<dyn UserDefinedLogicalNode>>;
/// Returns the necessary input columns for this node required to compute
/// the columns in the output schema
///
/// This is used for projection push-down when DataFusion has determined that
/// only a subset of the output columns of this node are needed by its parents.
/// This API is used to tell DataFusion which, if any, of the input columns are no longer
/// needed.
///
/// Return `None`, the default, if this information can not be determined.
/// Returns `Some(_)` with the column indices for each child of this node that are
/// needed to compute `output_columns`
fn necessary_children_exprs(
&self,
_output_columns: &[usize],
) -> Option<Vec<Vec<usize>>> {
None
}
/// Update the hash `state` with this node requirements from
/// [`Hash`].
///
/// Note: consider using [`UserDefinedLogicalNodeCore`] instead of
/// [`UserDefinedLogicalNode`] directly.
///
/// This method is required to support hashing [`LogicalPlan`]s. To
/// implement it, typically the type implementing
/// [`UserDefinedLogicalNode`] typically implements [`Hash`] and
/// then the following boiler plate is used:
///
/// # Example:
/// ```
/// // User defined node that derives Hash
/// #[derive(Hash, Debug, PartialEq, Eq)]
/// struct MyNode {
/// val: u64,
/// }
///
/// // impl UserDefinedLogicalNode {
/// // ...
/// # impl MyNode {
/// // Boiler plate to call the derived Hash impl
/// fn dyn_hash(&self, state: &mut dyn std::hash::Hasher) {
/// use std::hash::Hash;
/// let mut s = state;
/// self.hash(&mut s);
/// }
/// // }
/// # }
/// ```
/// Note: [`UserDefinedLogicalNode`] is not constrained by [`Hash`]
/// directly because it must remain object safe.
fn dyn_hash(&self, state: &mut dyn Hasher);
/// Compare `other`, respecting requirements from [Eq].
///
/// Note: consider using [`UserDefinedLogicalNodeCore`] instead of
/// [`UserDefinedLogicalNode`] directly.
///
/// When `other` has an another type than `self`, then the values
/// are *not* equal.
///
/// This method is required to support Eq on [`LogicalPlan`]s. To
/// implement it, typically the type implementing
/// [`UserDefinedLogicalNode`] typically implements [`Eq`] and
/// then the following boiler plate is used:
///
/// # Example:
/// ```
/// # use datafusion_expr::UserDefinedLogicalNode;
/// // User defined node that derives Eq
/// #[derive(Hash, Debug, PartialEq, Eq)]
/// struct MyNode {
/// val: u64,
/// }
///
/// // impl UserDefinedLogicalNode {
/// // ...
/// # impl MyNode {
/// // Boiler plate to call the derived Eq impl
/// fn dyn_eq(&self, other: &dyn UserDefinedLogicalNode) -> bool {
/// match other.as_any().downcast_ref::<Self>() {
/// Some(o) => self == o,
/// None => false,
/// }
/// }
/// // }
/// # }
/// ```
/// Note: [`UserDefinedLogicalNode`] is not constrained by [`Eq`]
/// directly because it must remain object safe.
fn dyn_eq(&self, other: &dyn UserDefinedLogicalNode) -> bool;
/// Compare `other`, respecting requirements from [PartialOrd].
/// Must return `Some(Equal)` if and only if `self.dyn_eq(other)`.
fn dyn_ord(&self, other: &dyn UserDefinedLogicalNode) -> Option<Ordering>;
/// Returns `true` if a limit can be safely pushed down through this
/// `UserDefinedLogicalNode` node.
///
/// If this method returns `true`, and the query plan contains a limit at
/// the output of this node, DataFusion will push the limit to the input
/// of this node.
fn supports_limit_pushdown(&self) -> bool {
false
}
}
impl Hash for dyn UserDefinedLogicalNode {
fn hash<H: Hasher>(&self, state: &mut H) {
self.dyn_hash(state);
}
}
impl PartialEq for dyn UserDefinedLogicalNode {
fn eq(&self, other: &Self) -> bool {
self.dyn_eq(other)
}
}
impl PartialOrd for dyn UserDefinedLogicalNode {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
self.dyn_ord(other)
}
}
impl Eq for dyn UserDefinedLogicalNode {}
/// This trait facilitates implementation of the [`UserDefinedLogicalNode`].
///
/// See the example in
/// [user_defined_plan.rs](https://github.com/apache/datafusion/blob/main/datafusion/core/tests/user_defined/user_defined_plan.rs)
/// file for an example of how to use this extension API.
pub trait UserDefinedLogicalNodeCore:
fmt::Debug + Eq + PartialOrd + Hash + Sized + Send + Sync + 'static
{
/// Return the plan's name.
fn name(&self) -> &str;
/// Return the logical plan's inputs.
fn inputs(&self) -> Vec<&LogicalPlan>;
/// Return the output schema of this logical plan node.
fn schema(&self) -> &DFSchemaRef;
/// Perform check of invariants for the extension node.
///
/// This is the default implementation for extension nodes.
fn check_invariants(&self, _check: InvariantLevel) -> Result<()> {
Ok(())
}
/// Returns all expressions in the current logical plan node. This
/// should not include expressions of any inputs (aka
/// non-recursively). These expressions are used for optimizer
/// passes and rewrites.
fn expressions(&self) -> Vec<Expr>;
/// A list of output columns (e.g. the names of columns in
/// self.schema()) for which predicates can not be pushed below
/// this node without changing the output.
///
/// By default, this returns all columns and thus prevents any
/// predicates from being pushed below this node.
fn prevent_predicate_push_down_columns(&self) -> HashSet<String> {
// default (safe) is all columns in the schema.
get_all_columns_from_schema(self.schema())
}
/// Write a single line, human readable string to `f` for use in explain plan.
///
/// For example: `TopK: k=10`
fn fmt_for_explain(&self, f: &mut fmt::Formatter) -> fmt::Result;
/// Create a new `UserDefinedLogicalNode` with the specified children
/// and expressions. This function is used during optimization
/// when the plan is being rewritten and a new instance of the
/// `UserDefinedLogicalNode` must be created.
///
/// Note that exprs and inputs are in the same order as the result
/// of self.inputs and self.exprs.
///
/// So, `self.with_exprs_and_inputs(exprs, ..).expressions() == exprs
fn with_exprs_and_inputs(
&self,
exprs: Vec<Expr>,
inputs: Vec<LogicalPlan>,
) -> Result<Self>;
/// Returns the necessary input columns for this node required to compute
/// the columns in the output schema
///
/// This is used for projection push-down when DataFusion has determined that
/// only a subset of the output columns of this node are needed by its parents.
/// This API is used to tell DataFusion which, if any, of the input columns are no longer
/// needed.
///
/// Return `None`, the default, if this information can not be determined.
/// Returns `Some(_)` with the column indices for each child of this node that are
/// needed to compute `output_columns`
fn necessary_children_exprs(
&self,
_output_columns: &[usize],
) -> Option<Vec<Vec<usize>>> {
None
}
/// Returns `true` if a limit can be safely pushed down through this
/// `UserDefinedLogicalNode` node.
///
/// If this method returns `true`, and the query plan contains a limit at
/// the output of this node, DataFusion will push the limit to the input
/// of this node.
fn supports_limit_pushdown(&self) -> bool {
false // Disallow limit push-down by default
}
}
/// Automatically derive UserDefinedLogicalNode to `UserDefinedLogicalNode`
/// to avoid boiler plate for implementing `as_any`, `Hash`, `PartialEq` and `PartialOrd`.
impl<T: UserDefinedLogicalNodeCore> UserDefinedLogicalNode for T {
fn as_any(&self) -> &dyn Any {
self
}
fn name(&self) -> &str {
self.name()
}
fn inputs(&self) -> Vec<&LogicalPlan> {
self.inputs()
}
fn schema(&self) -> &DFSchemaRef {
self.schema()
}
fn check_invariants(&self, check: InvariantLevel) -> Result<()> {
self.check_invariants(check)
}
fn expressions(&self) -> Vec<Expr> {
self.expressions()
}
fn prevent_predicate_push_down_columns(&self) -> HashSet<String> {
self.prevent_predicate_push_down_columns()
}
fn fmt_for_explain(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.fmt_for_explain(f)
}
fn with_exprs_and_inputs(
&self,
exprs: Vec<Expr>,
inputs: Vec<LogicalPlan>,
) -> Result<Arc<dyn UserDefinedLogicalNode>> {
Ok(Arc::new(self.with_exprs_and_inputs(exprs, inputs)?))
}
fn necessary_children_exprs(
&self,
output_columns: &[usize],
) -> Option<Vec<Vec<usize>>> {
self.necessary_children_exprs(output_columns)
}
fn dyn_hash(&self, state: &mut dyn Hasher) {
let mut s = state;
self.hash(&mut s);
}
fn dyn_eq(&self, other: &dyn UserDefinedLogicalNode) -> bool {
match other.as_any().downcast_ref::<Self>() {
Some(o) => self == o,
None => false,
}
}
fn dyn_ord(&self, other: &dyn UserDefinedLogicalNode) -> Option<Ordering> {
other
.as_any()
.downcast_ref::<Self>()
.and_then(|other| self.partial_cmp(other))
}
fn supports_limit_pushdown(&self) -> bool {
self.supports_limit_pushdown()
}
}
fn get_all_columns_from_schema(schema: &DFSchema) -> HashSet<String> {
schema.fields().iter().map(|f| f.name().clone()).collect()
}