Line | Count | Source |
1 | | // Licensed to the Apache Software Foundation (ASF) under one |
2 | | // or more contributor license agreements. See the NOTICE file |
3 | | // distributed with this work for additional information |
4 | | // regarding copyright ownership. The ASF licenses this file |
5 | | // to you under the Apache License, Version 2.0 (the |
6 | | // "License"); you may not use this file except in compliance |
7 | | // with the License. You may obtain a copy of the License at |
8 | | // |
9 | | // http://www.apache.org/licenses/LICENSE-2.0 |
10 | | // |
11 | | // Unless required by applicable law or agreed to in writing, |
12 | | // software distributed under the License is distributed on an |
13 | | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
14 | | // KIND, either express or implied. See the License for the |
15 | | // specific language governing permissions and limitations |
16 | | // under the License. |
17 | | // This file is copied from |
18 | | // https://github.com/apache/impala/blob/branch-2.9.0/be/src/runtime/raw-value.h |
19 | | // and modified by Doris |
20 | | |
21 | | #pragma once |
22 | | |
23 | | #include <string> |
24 | | |
25 | | #include "common/consts.h" |
26 | | #include "common/exception.h" |
27 | | #include "common/logging.h" |
28 | | #include "core/data_type/define_primitive_type.h" |
29 | | #include "core/packed_int128.h" |
30 | | #include "core/string_ref.h" |
31 | | #include "util/hash_util.hpp" |
32 | | |
33 | | namespace doris { |
34 | | class SlotDescriptor; |
35 | | |
36 | | // Useful utility functions for runtime values (which are passed around as void*). |
37 | | class RawValue { |
38 | | public: |
39 | | // Same as the up function, only use in vec exec engine. |
40 | | static uint32_t zlib_crc32(const void* value, size_t len, const PrimitiveType& type, |
41 | | uint32_t seed); |
42 | | }; |
43 | | |
44 | | // NOTE: this is just for split data, decimal use old doris hash function |
45 | | // Because crc32 hardware is not equal with zlib crc32 |
46 | | inline uint32_t RawValue::zlib_crc32(const void* v, size_t len, const PrimitiveType& type, |
47 | 39.4M | uint32_t seed) { |
48 | | // Reject binary even for NULL instead of reaching the default-type assertion or hash path. |
49 | 39.4M | if (type == TYPE_VARBINARY) { |
50 | 3 | throw Exception(ErrorCode::NOT_IMPLEMENTED_ERROR, |
51 | 3 | "VARBINARY tablet routing hash is not supported"); |
52 | 3 | } |
53 | | // Hash_combine with v = 0 |
54 | 39.4M | if (v == nullptr) { |
55 | 0 | uint32_t value = 0x9e3779b9; |
56 | 0 | return seed ^ (value + (seed << 6) + (seed >> 2)); |
57 | 0 | } |
58 | | |
59 | 39.4M | switch (type) { |
60 | 945k | case TYPE_VARCHAR: |
61 | 945k | case TYPE_HLL: |
62 | 945k | case TYPE_STRING: |
63 | 946k | case TYPE_CHAR: { |
64 | 946k | return HashUtil::zlib_crc_hash(v, (uint32_t)len, seed); |
65 | 945k | } |
66 | | |
67 | 135 | case TYPE_BOOLEAN: |
68 | 3.09k | case TYPE_TINYINT: |
69 | 3.09k | return HashUtil::zlib_crc_hash(v, 1, seed); |
70 | 1.31k | case TYPE_SMALLINT: |
71 | 1.31k | return HashUtil::zlib_crc_hash(v, 2, seed); |
72 | 21.7M | case TYPE_INT: |
73 | 21.7M | return HashUtil::zlib_crc_hash(v, 4, seed); |
74 | 16.7M | case TYPE_BIGINT: |
75 | 16.7M | return HashUtil::zlib_crc_hash(v, 8, seed); |
76 | 28.3k | case TYPE_LARGEINT: |
77 | 28.3k | return HashUtil::zlib_crc_hash(v, 16, seed); |
78 | 0 | case TYPE_FLOAT: |
79 | 0 | return HashUtil::zlib_crc_hash(v, 4, seed); |
80 | 0 | case TYPE_DOUBLE: |
81 | 0 | return HashUtil::zlib_crc_hash(v, 8, seed); |
82 | 23 | case TYPE_DATE: |
83 | 48 | case TYPE_DATETIME: { |
84 | 48 | auto* date_val = (const VecDateTimeValue*)v; |
85 | 48 | char buf[64]; |
86 | 48 | int date_len = date_val->to_buffer(buf); |
87 | 48 | return HashUtil::zlib_crc_hash(buf, date_len, seed); |
88 | 23 | } |
89 | | |
90 | 1.76k | case TYPE_DATEV2: { |
91 | 1.76k | return HashUtil::zlib_crc_hash(v, 4, seed); |
92 | 23 | } |
93 | | |
94 | 12.4k | case TYPE_DATETIMEV2: |
95 | 12.5k | case TYPE_TIMESTAMP_NS: { |
96 | 12.5k | return HashUtil::zlib_crc_hash(v, 8, seed); |
97 | 12.4k | } |
98 | | |
99 | 2.93k | case TYPE_TIMESTAMPTZ: { |
100 | 2.93k | return HashUtil::zlib_crc_hash(v, 8, seed); |
101 | 12.4k | } |
102 | | |
103 | 27 | case TYPE_DECIMALV2: { |
104 | 27 | const DecimalV2Value* dec_val = (const DecimalV2Value*)v; |
105 | 27 | int64_t int_val = dec_val->int_value(); |
106 | 27 | int32_t frac_val = dec_val->frac_value(); |
107 | 27 | seed = HashUtil::zlib_crc_hash(&int_val, sizeof(int_val), seed); |
108 | 27 | return HashUtil::zlib_crc_hash(&frac_val, sizeof(frac_val), seed); |
109 | 12.4k | } |
110 | 156 | case TYPE_DECIMAL32: |
111 | 156 | return HashUtil::zlib_crc_hash(v, 4, seed); |
112 | 224 | case TYPE_DECIMAL64: |
113 | 224 | return HashUtil::zlib_crc_hash(v, 8, seed); |
114 | 428 | case TYPE_DECIMAL128I: |
115 | 428 | return HashUtil::zlib_crc_hash(v, 16, seed); |
116 | 4.37k | case TYPE_DECIMAL256: |
117 | 4.37k | return HashUtil::zlib_crc_hash(v, 32, seed); |
118 | 3 | case TYPE_IPV4: |
119 | 3 | return HashUtil::zlib_crc_hash(v, 4, seed); |
120 | 3 | case TYPE_IPV6: |
121 | 3 | return HashUtil::zlib_crc_hash(v, 16, seed); |
122 | 0 | default: |
123 | 0 | DCHECK(false) << "invalid type: " << type; |
124 | 0 | return 0; |
125 | 39.4M | } |
126 | 39.4M | } |
127 | | } // namespace doris |