be/src/core/string_ref.cpp
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/ClickHouse/ClickHouse/blob/master/base/base/StringRef. |
19 | | // And modified by Doris |
20 | | |
21 | | #include "core/string_ref.h" |
22 | | |
23 | | #include "common/compiler_util.h" // IWYU pragma: keep |
24 | | |
25 | | namespace doris { |
26 | 252 | StringRef StringRef::trim() const { |
27 | | // Remove leading and trailing spaces. |
28 | 252 | int64_t begin = 0; |
29 | | |
30 | 253 | while (begin < size && data[begin] == ' ') { |
31 | 1 | ++begin; |
32 | 1 | } |
33 | | |
34 | 252 | int64_t end = size - 1; |
35 | | |
36 | 252 | while (end > begin && data[end] == ' ') { |
37 | 0 | --end; |
38 | 0 | } |
39 | | |
40 | 252 | return StringRef(data + begin, end - begin + 1); |
41 | 252 | } |
42 | | |
43 | 0 | StringRef StringRef::trim_tail_padding_zero() const { |
44 | | // Remove trailing padding zero. |
45 | 0 | int64_t end = size - 1; |
46 | |
|
47 | 0 | while (end >= 0 && data[end] == '\0') { |
48 | 0 | --end; |
49 | 0 | } |
50 | |
|
51 | 0 | return StringRef(data, end + 1); |
52 | 0 | } |
53 | | |
54 | 126 | StringRef StringRef::trim_whitespace() const { |
55 | | // Remove leading and trailing whitespace. |
56 | 126 | int64_t begin = 0; |
57 | | |
58 | 188 | while (begin < size && std::isspace(data[begin])) { |
59 | 62 | ++begin; |
60 | 62 | } |
61 | | |
62 | 126 | int64_t end = size - 1; |
63 | | |
64 | 126 | while (end > begin && std::isspace(data[end])) { |
65 | 0 | --end; |
66 | 0 | } |
67 | | |
68 | 126 | return StringRef(data + begin, end - begin + 1); |
69 | 126 | } |
70 | | |
71 | 124 | StringRef StringRef::trim_quote() const { |
72 | 124 | if (size < 2) { |
73 | 35 | return *this; |
74 | 35 | } |
75 | 89 | if (data[0] == '\'' && data[size - 1] == '\'') { |
76 | 2 | return StringRef(data + 1, size - 2); |
77 | 87 | } else if (data[0] == '"' && data[size - 1] == '"') { |
78 | 26 | return StringRef(data + 1, size - 2); |
79 | 26 | } |
80 | 61 | return *this; |
81 | 89 | } |
82 | | |
83 | | // TODO: rewrite in AVX2 |
84 | 152 | size_t StringRef::find_first_of(char c) const { |
85 | 152 | const char* p = static_cast<const char*>(memchr(data, c, size)); |
86 | 152 | return p == nullptr ? -1 : p - data; |
87 | 152 | } |
88 | | |
89 | 0 | StringRef StringRef::min_string_val() { |
90 | 0 | return StringRef((char*)(&StringRef::MIN_CHAR), 0); |
91 | 0 | } |
92 | | |
93 | 0 | StringRef StringRef::max_string_val() { |
94 | 0 | return StringRef((char*)(&StringRef::MAX_CHAR), 1); |
95 | 0 | } |
96 | | |
97 | 0 | bool StringRef::start_with(char ch) const { |
98 | 0 | if (UNLIKELY(size == 0)) { |
99 | 0 | return false; |
100 | 0 | } |
101 | 0 | return data[0] == ch; |
102 | 0 | } |
103 | 0 | bool StringRef::end_with(char ch) const { |
104 | 0 | if (UNLIKELY(size == 0)) { |
105 | 0 | return false; |
106 | 0 | } |
107 | 0 | return data[size - 1] == ch; |
108 | 0 | } |
109 | | |
110 | 49 | bool StringRef::start_with(const StringRef& search_string) const { |
111 | 49 | if (search_string.size == 0) { |
112 | 0 | return true; |
113 | 0 | } |
114 | | |
115 | 49 | if (UNLIKELY(size < search_string.size)) { |
116 | 0 | return false; |
117 | 0 | } |
118 | | |
119 | 49 | #if defined(__SSE2__) || defined(__aarch64__) |
120 | 49 | return memequalSSE2Wide(data, search_string.data, search_string.size); |
121 | | #else |
122 | | return 0 == memcmp(data, search_string.data, search_string.size); |
123 | | #endif |
124 | 49 | } |
125 | 65 | bool StringRef::end_with(const StringRef& search_string) const { |
126 | 65 | DCHECK(size >= search_string.size); |
127 | 65 | if (search_string.size == 0) { |
128 | 0 | return true; |
129 | 0 | } |
130 | | |
131 | 65 | #if defined(__SSE2__) || defined(__aarch64__) |
132 | 65 | return memequalSSE2Wide(data + size - search_string.size, search_string.data, |
133 | 65 | search_string.size); |
134 | | #else |
135 | | return 0 == memcmp(data + size - search_string.size, search_string.data, search_string.size); |
136 | | #endif |
137 | 65 | } |
138 | | } // namespace doris |