|
| 1 | +// Copyright 2021 Datafuse 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 | +use databend_common_exception::Result; |
| 16 | +use databend_common_hashtable::BloomHash; |
| 17 | +use databend_common_hashtable::FastHash; |
| 18 | + |
| 19 | +use crate::HashMethod; |
| 20 | +use crate::HashMethodKind; |
| 21 | +use crate::ProjectedBlock; |
| 22 | + |
| 23 | +/// Get row hash by [`HashMethod`] (uses [`FastHash`], i.e. the hashtable hash). |
| 24 | +pub fn hash_by_method<T>( |
| 25 | + method: &HashMethodKind, |
| 26 | + columns: ProjectedBlock, |
| 27 | + num_rows: usize, |
| 28 | + hashes: &mut T, |
| 29 | +) -> Result<()> |
| 30 | +where |
| 31 | + T: Extend<u64>, |
| 32 | +{ |
| 33 | + match method { |
| 34 | + HashMethodKind::Serializer(method) => { |
| 35 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 36 | + hashes.extend( |
| 37 | + method |
| 38 | + .build_keys_iter(&keys_state)? |
| 39 | + .map(|key| key.fast_hash()), |
| 40 | + ); |
| 41 | + } |
| 42 | + HashMethodKind::SingleBinary(method) => { |
| 43 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 44 | + hashes.extend( |
| 45 | + method |
| 46 | + .build_keys_iter(&keys_state)? |
| 47 | + .map(|key| key.fast_hash()), |
| 48 | + ); |
| 49 | + } |
| 50 | + HashMethodKind::KeysU8(method) => { |
| 51 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 52 | + hashes.extend( |
| 53 | + method |
| 54 | + .build_keys_iter(&keys_state)? |
| 55 | + .map(|key| key.fast_hash()), |
| 56 | + ); |
| 57 | + } |
| 58 | + HashMethodKind::KeysU16(method) => { |
| 59 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 60 | + hashes.extend( |
| 61 | + method |
| 62 | + .build_keys_iter(&keys_state)? |
| 63 | + .map(|key| key.fast_hash()), |
| 64 | + ); |
| 65 | + } |
| 66 | + HashMethodKind::KeysU32(method) => { |
| 67 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 68 | + hashes.extend( |
| 69 | + method |
| 70 | + .build_keys_iter(&keys_state)? |
| 71 | + .map(|key| key.fast_hash()), |
| 72 | + ); |
| 73 | + } |
| 74 | + HashMethodKind::KeysU64(method) => { |
| 75 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 76 | + hashes.extend( |
| 77 | + method |
| 78 | + .build_keys_iter(&keys_state)? |
| 79 | + .map(|key| key.fast_hash()), |
| 80 | + ); |
| 81 | + } |
| 82 | + HashMethodKind::KeysU128(method) => { |
| 83 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 84 | + hashes.extend( |
| 85 | + method |
| 86 | + .build_keys_iter(&keys_state)? |
| 87 | + .map(|key| key.fast_hash()), |
| 88 | + ); |
| 89 | + } |
| 90 | + HashMethodKind::KeysU256(method) => { |
| 91 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 92 | + hashes.extend( |
| 93 | + method |
| 94 | + .build_keys_iter(&keys_state)? |
| 95 | + .map(|key| key.fast_hash()), |
| 96 | + ); |
| 97 | + } |
| 98 | + } |
| 99 | + Ok(()) |
| 100 | +} |
| 101 | + |
| 102 | +/// Get row hash for Bloom filter by [`HashMethod`]. This always uses [`BloomHash`], |
| 103 | +/// which is independent of SSE4.2 and provides a well-distributed 64-bit hash. |
| 104 | +pub fn hash_by_method_for_bloom<T>( |
| 105 | + method: &HashMethodKind, |
| 106 | + columns: ProjectedBlock, |
| 107 | + num_rows: usize, |
| 108 | + hashes: &mut T, |
| 109 | +) -> Result<()> |
| 110 | +where |
| 111 | + T: Extend<u64>, |
| 112 | +{ |
| 113 | + match method { |
| 114 | + HashMethodKind::Serializer(method) => { |
| 115 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 116 | + hashes.extend( |
| 117 | + method |
| 118 | + .build_keys_iter(&keys_state)? |
| 119 | + .map(|key| key.bloom_hash()), |
| 120 | + ); |
| 121 | + } |
| 122 | + HashMethodKind::SingleBinary(method) => { |
| 123 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 124 | + hashes.extend( |
| 125 | + method |
| 126 | + .build_keys_iter(&keys_state)? |
| 127 | + .map(|key| key.bloom_hash()), |
| 128 | + ); |
| 129 | + } |
| 130 | + HashMethodKind::KeysU8(method) => { |
| 131 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 132 | + hashes.extend( |
| 133 | + method |
| 134 | + .build_keys_iter(&keys_state)? |
| 135 | + .map(|key| key.bloom_hash()), |
| 136 | + ); |
| 137 | + } |
| 138 | + HashMethodKind::KeysU16(method) => { |
| 139 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 140 | + hashes.extend( |
| 141 | + method |
| 142 | + .build_keys_iter(&keys_state)? |
| 143 | + .map(|key| key.bloom_hash()), |
| 144 | + ); |
| 145 | + } |
| 146 | + HashMethodKind::KeysU32(method) => { |
| 147 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 148 | + hashes.extend( |
| 149 | + method |
| 150 | + .build_keys_iter(&keys_state)? |
| 151 | + .map(|key| key.bloom_hash()), |
| 152 | + ); |
| 153 | + } |
| 154 | + HashMethodKind::KeysU64(method) => { |
| 155 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 156 | + hashes.extend( |
| 157 | + method |
| 158 | + .build_keys_iter(&keys_state)? |
| 159 | + .map(|key| key.bloom_hash()), |
| 160 | + ); |
| 161 | + } |
| 162 | + HashMethodKind::KeysU128(method) => { |
| 163 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 164 | + hashes.extend( |
| 165 | + method |
| 166 | + .build_keys_iter(&keys_state)? |
| 167 | + .map(|key| key.bloom_hash()), |
| 168 | + ); |
| 169 | + } |
| 170 | + HashMethodKind::KeysU256(method) => { |
| 171 | + let keys_state = method.build_keys_state(columns, num_rows)?; |
| 172 | + hashes.extend( |
| 173 | + method |
| 174 | + .build_keys_iter(&keys_state)? |
| 175 | + .map(|key| key.bloom_hash()), |
| 176 | + ); |
| 177 | + } |
| 178 | + } |
| 179 | + Ok(()) |
| 180 | +} |
0 commit comments