1// Copyright (C) 2022 The Qt Company Ltd.
2// SPDX-License-Identifier: LicenseRef-Qt-Commercial OR LGPL-3.0-only OR GPL-2.0-only OR GPL-3.0-only
3
4#ifndef QLATIN1STRINGMATCHER_H
5#define QLATIN1STRINGMATCHER_H
6
7#include <functional>
8#include <iterator>
9#include <limits>
10
11#include <QtCore/q20algorithm.h>
12#include <QtCore/qstring.h>
13
14QT_BEGIN_NAMESPACE
15
16namespace QtPrivate {
17template<class RandomIt1,
18 class Hash = std::hash<typename std::iterator_traits<RandomIt1>::value_type>,
19 class BinaryPredicate = std::equal_to<>>
20class q_boyer_moore_searcher_hashed_needle
21{
22public:
23 constexpr q_boyer_moore_searcher_hashed_needle(RandomIt1 pat_first, RandomIt1 pat_last)
24 : m_skiptable{}
25 {
26 const size_t n = std::distance(pat_first, pat_last);
27 constexpr auto uchar_max = (std::numeric_limits<uchar>::max)();
28 uchar max = n > uchar_max ? uchar_max : uchar(n);
29 q20::fill(first: std::begin(arr&: m_skiptable), last: std::end(arr&: m_skiptable), value: max);
30
31 RandomIt1 pattern = pat_first;
32 pattern += n - max;
33 while (max--)
34 m_skiptable[uchar(*pattern++)] = max;
35 }
36
37 template<class RandomIt2>
38 constexpr auto operator()(RandomIt2 first, RandomIt2 last, RandomIt1 pat_first,
39 RandomIt1 pat_last) const
40 {
41 struct R
42 {
43 RandomIt2 begin, end;
44 };
45 Hash hf;
46 BinaryPredicate pred;
47 auto pat_length = std::distance(pat_first, pat_last);
48 if (pat_length == 0)
49 return R{ first, first };
50
51 const qsizetype pl_minus_one = qsizetype(pat_length - 1);
52 RandomIt2 current = first + pl_minus_one;
53
54 while (current < last) {
55 qsizetype skip = m_skiptable[hf(*current)];
56 if (!skip) {
57 // possible match
58 while (skip < pat_length) {
59 if (!pred(hf(*(current - skip)), uchar(pat_first[pl_minus_one - skip])))
60 break;
61 skip++;
62 }
63 if (skip > pl_minus_one) { // we have a match
64 auto match = current - skip + 1;
65 return R{ match, match + pat_length };
66 }
67
68 // If we don't have a match we are a bit inefficient as we only skip by one
69 // when we have the non matching char in the string.
70 if (m_skiptable[hf(*(current - skip))] == pat_length)
71 skip = pat_length - skip;
72 else
73 skip = 1;
74 }
75 current += skip;
76 }
77
78 return R{ last, last };
79 }
80
81private:
82 alignas(16) uchar m_skiptable[256];
83};
84
85struct QCaseSensitiveLatin1Hash
86{
87 constexpr QCaseSensitiveLatin1Hash() = default;
88
89 constexpr std::size_t operator()(char c) const noexcept { return std::size_t(uchar(c)); }
90};
91
92struct QCaseInsensitiveLatin1Hash
93{
94 constexpr QCaseInsensitiveLatin1Hash() = default;
95
96 constexpr std::size_t operator()(char c) const noexcept
97 {
98 return std::size_t(latin1Lower[uchar(c)]);
99 }
100
101 static int difference(char lhs, char rhs)
102 {
103 return int(latin1Lower[uchar(lhs)]) - int(latin1Lower[uchar(rhs)]);
104 }
105
106 static auto matcher(char ch)
107 {
108 return [sought = latin1Lower[uchar(ch)]](char other) {
109 return latin1Lower[uchar(other)] == sought;
110 };
111 }
112
113private:
114 static constexpr uchar latin1Lower[256] = {
115 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,
116 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,
117 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,
118 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,
119 0x40,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,
120 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7a,0x5b,0x5c,0x5d,0x5e,0x5f,
121 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,
122 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,
123 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,
124 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,
125 0xa0,0xa1,0xa2,0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,
126 0xb0,0xb1,0xb2,0xb3,0xb4,0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,
127 // 0xd7 (multiplication sign) and 0xdf (sz ligature) complicate life
128 0xe0,0xe1,0xe2,0xe3,0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,
129 0xf0,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xd7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xdf,
130 0xe0,0xe1,0xe2,0xe3,0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,
131 0xf0,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
132 };
133};
134}
135
136class QLatin1StringMatcher
137{
138public:
139 Q_CORE_EXPORT QLatin1StringMatcher() noexcept;
140 Q_CORE_EXPORT explicit QLatin1StringMatcher(
141 QLatin1StringView pattern, Qt::CaseSensitivity cs = Qt::CaseSensitive) noexcept;
142 Q_CORE_EXPORT ~QLatin1StringMatcher() noexcept;
143
144 Q_CORE_EXPORT void setPattern(QLatin1StringView pattern) noexcept;
145 Q_CORE_EXPORT QLatin1StringView pattern() const noexcept;
146 Q_CORE_EXPORT void setCaseSensitivity(Qt::CaseSensitivity cs) noexcept;
147 Q_CORE_EXPORT Qt::CaseSensitivity caseSensitivity() const noexcept;
148
149 Q_CORE_EXPORT qsizetype indexIn(QLatin1StringView haystack, qsizetype from = 0) const noexcept;
150
151private:
152 void setSearcher() noexcept;
153 void freeSearcher() noexcept;
154
155 QLatin1StringView m_pattern;
156 Qt::CaseSensitivity m_cs;
157 typedef QtPrivate::q_boyer_moore_searcher_hashed_needle<const char *,
158 QtPrivate::QCaseSensitiveLatin1Hash>
159 CaseSensitiveSearcher;
160 typedef QtPrivate::q_boyer_moore_searcher_hashed_needle<const char *,
161 QtPrivate::QCaseInsensitiveLatin1Hash>
162 CaseInsensitiveSearcher;
163 union {
164 CaseSensitiveSearcher m_caseSensitiveSearcher;
165 CaseInsensitiveSearcher m_caseInsensitiveSearcher;
166 };
167 char m_foldBuffer[256];
168};
169
170QT_END_NAMESPACE
171
172#endif // QLATIN1MATCHER_H
173

source code of qtbase/src/corelib/text/qlatin1stringmatcher.h