alamb commented on code in PR #9183: URL: https://github.com/apache/arrow-datafusion/pull/9183#discussion_r1484446902
########## datafusion-examples/examples/pruning.rs: ########## @@ -0,0 +1,186 @@ +// 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. + +use arrow::array::{ArrayRef, BooleanArray, Int32Array}; +use arrow::datatypes::{DataType, Field, Schema, SchemaRef}; +use datafusion::common::{DFSchema, ScalarValue}; +use datafusion::execution::context::ExecutionProps; +use datafusion::physical_expr::create_physical_expr; +use datafusion::physical_optimizer::pruning::{PruningPredicate, PruningStatistics}; +use datafusion::prelude::*; +use std::collections::HashSet; +use std::sync::Arc; + +/// This example shows how to use DataFusion's `PruningPredicate` to prove +/// filter expressions can never be true based on statistics such as min/max +/// values of columns. +/// +/// The process is called "pruning" and is commonly used in query engines to +/// quickly eliminate entire files / partitions / row groups of data from +/// consideration using statistical information from a catalog or other +/// metadata. +#[tokio::main] +async fn main() { + // In this example, we'll use the PruningPredicate to determine if + // the expression `x = 5 AND y = 10` can never be true based on statistics + + // Start with the expression `x = 5 AND y = 10` + let expr = col("x").eq(lit(5)).and(col("y").eq(lit(10))); + + // We can analyze this predicate using information provided by the + // `PruningStatistics` trait, in this case we'll use a simple catalog that + // models three files. For all rows in each file: + // + // File 1: x has values between `4` and `6` + // y has the value 10 + // + // File 1: x has values between `4` and `6` + // y has the value of `7` + // + // File 3: x has the value 1 + // nothing is known about the value of y + let my_catalog = MyCatalog::new(); + + // Create a `PruningPredicate`. + // + // Note the predicate does not automatically coerce types or simplify + // expressions. See expr_api.rs examples for how to do this if required + let predicate = create_pruning_predicate(expr, &my_catalog.schema); + + // Evaluate the predicate for the three files in the catalog + let prune_results = predicate.prune(&my_catalog).unwrap(); + println!("Pruning results: {prune_results:?}"); + + // The result is a `Vec` of bool values, one for each file in the catalog + assert_eq!( + prune_results, + vec![ + // File 1: `x = 5 AND y = 10` can evaluate to true if x has values + // between `4` and `6`, y has the value `10`, so the file can not be + // skipped + // + // NOTE this doesn't mean there actually are rows that evaluate to + // true, but the pruning predicate can't prove there aren't any. + true, + // File 2: `x = 5 AND y = 10` can never evaluate to true because y + // has only the value of 7. Thus this file can be skipped. + false, + // File 3: `x = 5 AND y = 10` can never evaluate to true because x Review Comment: FYI @appletreeisyellow here is an actual example showing that the pruning predicate does the right thing with unknown column values -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
