risingwave_frontend/optimizer/plan_node/generic/
expand.rs

1// Copyright 2025 RisingWave Labs
2//
3// Licensed under the Apache License, Version 2.0 (the "License");
4// you may not use this file except in compliance with the License.
5// You may obtain a copy of the License at
6//
7//     http://www.apache.org/licenses/LICENSE-2.0
8//
9// Unless required by applicable law or agreed to in writing, software
10// distributed under the License is distributed on an "AS IS" BASIS,
11// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12// See the License for the specific language governing permissions and
13// limitations under the License.
14
15use itertools::Itertools;
16use pretty_xmlish::{Pretty, Str, XmlNode};
17use risingwave_common::catalog::{Field, FieldDisplay, Schema};
18use risingwave_common::types::DataType;
19use risingwave_common::util::column_index_mapping::ColIndexMapping;
20
21use super::{DistillUnit, GenericPlanNode, GenericPlanRef};
22use crate::optimizer::optimizer_context::OptimizerContextRef;
23use crate::optimizer::plan_node::utils::childless_record;
24use crate::optimizer::property::FunctionalDependencySet;
25
26/// [`Expand`] expand one row multiple times according to `column_subsets` and also keep
27/// original columns of input. It can be used to implement distinct aggregation and group set.
28///
29/// This is the schema of `Expand`:
30/// | expanded columns(i.e. some columns are set to null) | original columns of input | flag |.
31///
32/// Aggregates use expanded columns as their arguments and original columns for their filter. `flag`
33/// is used to distinguish between different `subset`s in `column_subsets`.
34#[derive(Debug, Clone, PartialEq, Eq, Hash)]
35pub struct Expand<PlanRef> {
36    // `column_subsets` has many `subset`s which specifies the columns that need to be
37    // reserved and other columns will be filled with NULL.
38    pub column_subsets: Vec<Vec<usize>>,
39    pub input: PlanRef,
40}
41
42impl<PlanRef: GenericPlanRef> Expand<PlanRef> {
43    pub fn output_len(&self) -> usize {
44        self.input.schema().len() * 2 + 1
45    }
46
47    fn flag_index(&self) -> usize {
48        self.output_len() - 1
49    }
50}
51
52impl<PlanRef: GenericPlanRef> GenericPlanNode for Expand<PlanRef> {
53    fn schema(&self) -> Schema {
54        let mut fields = self
55            .input
56            .schema()
57            .fields()
58            .iter()
59            .map(|f| Field::with_name(f.data_type(), format!("{}_expanded", f.name)))
60            .collect::<Vec<_>>();
61        fields.extend(self.input.schema().fields().iter().cloned());
62        fields.push(Field::with_name(DataType::Int64, "flag"));
63        Schema::new(fields)
64    }
65
66    fn stream_key(&self) -> Option<Vec<usize>> {
67        let input_schema_len = self.input.schema().len();
68        let mut pk_indices = self
69            .input
70            .stream_key()?
71            .iter()
72            .map(|&pk| pk + input_schema_len)
73            .collect_vec();
74        // The last column should be the flag.
75        pk_indices.push(input_schema_len * 2);
76        Some(pk_indices)
77    }
78
79    fn ctx(&self) -> OptimizerContextRef {
80        self.input.ctx()
81    }
82
83    fn functional_dependency(&self) -> FunctionalDependencySet {
84        let input_fd = self
85            .input
86            .functional_dependency()
87            .clone()
88            .into_dependencies();
89        let output_len = self.output_len();
90        let flag_index = self.flag_index();
91
92        self.input
93            .functional_dependency()
94            .as_dependencies()
95            .iter()
96            .map(|_input_fd| {})
97            .collect_vec();
98
99        let mut current_fd = FunctionalDependencySet::new(output_len);
100        for mut fd in input_fd {
101            fd.grow(output_len);
102            fd.set_from(flag_index, true);
103            current_fd.add_functional_dependency(fd);
104        }
105        current_fd
106    }
107}
108
109impl<PlanRef: GenericPlanRef> DistillUnit for Expand<PlanRef> {
110    fn distill_with_name<'a>(&self, name: impl Into<Str<'a>>) -> XmlNode<'a> {
111        childless_record(name, vec![("column_subsets", self.column_subsets_pretty())])
112    }
113}
114
115impl<PlanRef: GenericPlanRef> Expand<PlanRef> {
116    fn column_subsets_pretty<'a>(&self) -> Pretty<'a> {
117        Pretty::Array(
118            self.column_subsets
119                .iter()
120                .map(|subset| {
121                    subset
122                        .iter()
123                        .map(|&i| FieldDisplay(self.input.schema().fields.get(i).unwrap()))
124                        .map(|d| Pretty::display(&d))
125                        .collect()
126                })
127                .map(Pretty::Array)
128                .collect(),
129        )
130    }
131
132    pub fn i2o_col_mapping(&self) -> ColIndexMapping {
133        let input_len = self.input.schema().len();
134        let map = (0..input_len)
135            .map(|source| Some(source + input_len))
136            .collect_vec();
137        ColIndexMapping::new(map, self.output_len())
138    }
139
140    pub fn o2i_col_mapping(&self) -> ColIndexMapping {
141        self.i2o_col_mapping()
142            .inverse()
143            .expect("must be invertible")
144    }
145
146    pub fn decompose(self) -> (PlanRef, Vec<Vec<usize>>) {
147        (self.input, self.column_subsets)
148    }
149}
risingwave_frontend/optimizer/plan_node/generic/expand.rs

risingwave_frontend/optimizer/plan_node/generic/
expand.rs
