1 // Copyright 2017 The Abseil Authors.
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 // https://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 #include "absl/strings/str_split.h"
16
17 #include <algorithm>
18 #include <cassert>
19 #include <cstdint>
20 #include <cstdlib>
21 #include <cstring>
22 #include <iterator>
23 #include <limits>
24 #include <memory>
25
26 #include "absl/base/internal/raw_logging.h"
27 #include "absl/strings/ascii.h"
28
29 namespace absl {
30 ABSL_NAMESPACE_BEGIN
31
32 namespace {
33
34 // This GenericFind() template function encapsulates the finding algorithm
35 // shared between the ByString and ByAnyChar delimiters. The FindPolicy
36 // template parameter allows each delimiter to customize the actual find
37 // function to use and the length of the found delimiter. For example, the
38 // Literal delimiter will ultimately use absl::string_view::find(), and the
39 // AnyOf delimiter will use absl::string_view::find_first_of().
40 template <typename FindPolicy>
GenericFind(absl::string_view text,absl::string_view delimiter,size_t pos,FindPolicy find_policy)41 absl::string_view GenericFind(absl::string_view text,
42 absl::string_view delimiter, size_t pos,
43 FindPolicy find_policy) {
44 if (delimiter.empty() && text.length() > 0) {
45 // Special case for empty string delimiters: always return a zero-length
46 // absl::string_view referring to the item at position 1 past pos.
47 return absl::string_view(text.data() + pos + 1, 0);
48 }
49 size_t found_pos = absl::string_view::npos;
50 absl::string_view found(text.data() + text.size(),
51 0); // By default, not found
52 found_pos = find_policy.Find(text, delimiter, pos);
53 if (found_pos != absl::string_view::npos) {
54 found = absl::string_view(text.data() + found_pos,
55 find_policy.Length(delimiter));
56 }
57 return found;
58 }
59
60 // Finds using absl::string_view::find(), therefore the length of the found
61 // delimiter is delimiter.length().
62 struct LiteralPolicy {
Findabsl::__anon7826610a0111::LiteralPolicy63 size_t Find(absl::string_view text, absl::string_view delimiter, size_t pos) {
64 return text.find(delimiter, pos);
65 }
Lengthabsl::__anon7826610a0111::LiteralPolicy66 size_t Length(absl::string_view delimiter) { return delimiter.length(); }
67 };
68
69 // Finds using absl::string_view::find_first_of(), therefore the length of the
70 // found delimiter is 1.
71 struct AnyOfPolicy {
Findabsl::__anon7826610a0111::AnyOfPolicy72 size_t Find(absl::string_view text, absl::string_view delimiter, size_t pos) {
73 return text.find_first_of(delimiter, pos);
74 }
Lengthabsl::__anon7826610a0111::AnyOfPolicy75 size_t Length(absl::string_view /* delimiter */) { return 1; }
76 };
77
78 } // namespace
79
80 //
81 // ByString
82 //
83
ByString(absl::string_view sp)84 ByString::ByString(absl::string_view sp) : delimiter_(sp) {}
85
Find(absl::string_view text,size_t pos) const86 absl::string_view ByString::Find(absl::string_view text, size_t pos) const {
87 if (delimiter_.length() == 1) {
88 // Much faster to call find on a single character than on an
89 // absl::string_view.
90 size_t found_pos = text.find(delimiter_[0], pos);
91 if (found_pos == absl::string_view::npos)
92 return absl::string_view(text.data() + text.size(), 0);
93 return text.substr(found_pos, 1);
94 }
95 return GenericFind(text, delimiter_, pos, LiteralPolicy());
96 }
97
98 //
99 // ByChar
100 //
101
Find(absl::string_view text,size_t pos) const102 absl::string_view ByChar::Find(absl::string_view text, size_t pos) const {
103 size_t found_pos = text.find(c_, pos);
104 if (found_pos == absl::string_view::npos)
105 return absl::string_view(text.data() + text.size(), 0);
106 return text.substr(found_pos, 1);
107 }
108
109 //
110 // ByAnyChar
111 //
112
ByAnyChar(absl::string_view sp)113 ByAnyChar::ByAnyChar(absl::string_view sp) : delimiters_(sp) {}
114
Find(absl::string_view text,size_t pos) const115 absl::string_view ByAnyChar::Find(absl::string_view text, size_t pos) const {
116 return GenericFind(text, delimiters_, pos, AnyOfPolicy());
117 }
118
119 //
120 // ByLength
121 //
ByLength(ptrdiff_t length)122 ByLength::ByLength(ptrdiff_t length) : length_(length) {
123 ABSL_RAW_CHECK(length > 0, "");
124 }
125
Find(absl::string_view text,size_t pos) const126 absl::string_view ByLength::Find(absl::string_view text,
127 size_t pos) const {
128 pos = std::min(pos, text.size()); // truncate `pos`
129 absl::string_view substr = text.substr(pos);
130 // If the string is shorter than the chunk size we say we
131 // "can't find the delimiter" so this will be the last chunk.
132 if (substr.length() <= static_cast<size_t>(length_))
133 return absl::string_view(text.data() + text.size(), 0);
134
135 return absl::string_view(substr.data() + length_, 0);
136 }
137
138 ABSL_NAMESPACE_END
139 } // namespace absl
140