risingwave_stream/from_proto/
hash_agg.rs

1// Copyright 2022 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
15//! Streaming Hash Aggregator
16
17use std::sync::Arc;
18
19use risingwave_common::hash::{HashKey, HashKeyDispatcher};
20use risingwave_common::types::DataType;
21use risingwave_expr::aggregate::AggCall;
22use risingwave_pb::stream_plan::HashAggNode;
23
24use super::agg_common::{
25    build_agg_state_storages_from_proto, build_distinct_dedup_table_from_proto,
26};
27use super::*;
28use crate::common::table::state_table::StateTableBuilder;
29use crate::executor::aggregate::{AggExecutorArgs, HashAggExecutor, HashAggExecutorExtraArgs};
30
31pub struct HashAggExecutorDispatcherArgs<S: StateStore> {
32    args: AggExecutorArgs<S, HashAggExecutorExtraArgs>,
33    group_key_types: Vec<DataType>,
34}
35
36impl<S: StateStore> HashKeyDispatcher for HashAggExecutorDispatcherArgs<S> {
37    type Output = StreamResult<Box<dyn Execute>>;
38
39    fn dispatch_impl<K: HashKey>(self) -> Self::Output {
40        Ok(HashAggExecutor::<K, S>::new(self.args)?.boxed())
41    }
42
43    fn data_types(&self) -> &[DataType] {
44        &self.group_key_types
45    }
46}
47
48pub struct HashAggExecutorBuilder;
49
50impl_stream_node_body!(HashAgg(HashAggNode) => HashAggExecutorBuilder);
51
52impl ExecutorBuilder for HashAggExecutorBuilder {
53    type Node = HashAggNode;
54
55    async fn new_boxed_executor(
56        params: ExecutorParams,
57        node: &Self::Node,
58        store: impl StateStore,
59    ) -> StreamResult<Executor> {
60        let group_key_indices = node
61            .get_group_key()
62            .iter()
63            .map(|key| *key as usize)
64            .collect::<Vec<_>>();
65        let [input]: [_; 1] = params.input.try_into().unwrap();
66        let group_key_types = group_key_indices
67            .iter()
68            .map(|idx| input.schema().fields[*idx].data_type())
69            .collect_vec();
70
71        let agg_calls: Vec<AggCall> = node
72            .get_agg_calls()
73            .iter()
74            .map(AggCall::from_protobuf)
75            .try_collect()?;
76
77        let vnodes = Some(Arc::new(params.vnode_bitmap));
78        let storages = build_agg_state_storages_from_proto(
79            node.get_agg_call_states(),
80            store.clone(),
81            vnodes.clone(),
82            &params.config,
83        )
84        .await;
85        // disable sanity check so that old value is not required when updating states
86        let intermediate_state_table = StateTableBuilder::new(
87            node.get_intermediate_state_table().unwrap(),
88            store.clone(),
89            vnodes.clone(),
90        )
91        .enable_preload_all_rows_by_config(&params.config)
92        .build()
93        .await;
94        let distinct_dedup_tables = build_distinct_dedup_table_from_proto(
95            node.get_distinct_dedup_tables(),
96            store,
97            vnodes,
98            &params.config,
99        )
100        .await;
101
102        let exec = HashAggExecutorDispatcherArgs {
103            args: AggExecutorArgs {
104                version: node.version(),
105
106                input,
107                actor_ctx: params.actor_context,
108                info: params.info.clone(),
109
110                extreme_cache_size: params.config.developer.unsafe_extreme_cache_size,
111
112                agg_calls,
113                row_count_index: node.get_row_count_index() as usize,
114                storages,
115                intermediate_state_table,
116                distinct_dedup_tables,
117                watermark_epoch: params.watermark_epoch,
118                extra: HashAggExecutorExtraArgs {
119                    group_key_indices,
120                    chunk_size: params.config.developer.chunk_size,
121                    max_dirty_groups_heap_size: (params.config.developer)
122                        .hash_agg_max_dirty_groups_heap_size,
123                    emit_on_window_close: node.get_emit_on_window_close(),
124                },
125            },
126            group_key_types,
127        }
128        .dispatch()?;
129        Ok((params.info, exec).into())
130    }
131}
risingwave_stream/from_proto/hash_agg.rs

risingwave_stream/from_proto/
hash_agg.rs
