1 | /* |
---|
2 | * |
---|
3 | * Copyright (c) 2004 |
---|
4 | * John Maddock |
---|
5 | * |
---|
6 | * Use, modification and distribution are subject to the |
---|
7 | * Boost Software License, Version 1.0. (See accompanying file |
---|
8 | * LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) |
---|
9 | * |
---|
10 | */ |
---|
11 | |
---|
12 | /* |
---|
13 | * LOCATION: see http://www.boost.org for most recent version. |
---|
14 | * FILE: c_regex_traits.cpp |
---|
15 | * VERSION: see <boost/version.hpp> |
---|
16 | * DESCRIPTION: Implements out of line c_regex_traits<char> members |
---|
17 | */ |
---|
18 | |
---|
19 | |
---|
20 | #define BOOST_REGEX_SOURCE |
---|
21 | |
---|
22 | #include <boost/config.hpp> |
---|
23 | #include <boost/detail/workaround.hpp> |
---|
24 | |
---|
25 | #if !BOOST_WORKAROUND(__BORLANDC__, < 0x560) |
---|
26 | |
---|
27 | #include <boost/regex/v4/c_regex_traits.hpp> |
---|
28 | #include <boost/regex/v4/primary_transform.hpp> |
---|
29 | #include <boost/regex/v4/regex_traits_defaults.hpp> |
---|
30 | |
---|
31 | #ifdef BOOST_NO_STDC_NAMESPACE |
---|
32 | namespace std{ |
---|
33 | using ::strxfrm; using ::isspace; |
---|
34 | using ::ispunct; using ::isalpha; |
---|
35 | using ::isalnum; using ::iscntrl; |
---|
36 | using ::isprint; using ::isupper; |
---|
37 | using ::islower; using ::isdigit; |
---|
38 | using ::isxdigit; using ::strtol; |
---|
39 | } |
---|
40 | #endif |
---|
41 | |
---|
42 | #ifdef BOOST_HAS_ABI_HEADERS |
---|
43 | # include BOOST_ABI_PREFIX |
---|
44 | #endif |
---|
45 | |
---|
46 | namespace boost{ |
---|
47 | |
---|
48 | c_regex_traits<char>::string_type BOOST_REGEX_CALL c_regex_traits<char>::transform(const char* p1, const char* p2) |
---|
49 | { |
---|
50 | std::string result(10, ' '); |
---|
51 | std::size_t s = result.size(); |
---|
52 | std::size_t r; |
---|
53 | std::string src(p1, p2); |
---|
54 | while(s < (r = std::strxfrm(&*result.begin(), src.c_str(), s))) |
---|
55 | { |
---|
56 | result.append(r - s + 3, ' '); |
---|
57 | s = result.size(); |
---|
58 | } |
---|
59 | result.erase(r); |
---|
60 | return result; |
---|
61 | } |
---|
62 | |
---|
63 | c_regex_traits<char>::string_type BOOST_REGEX_CALL c_regex_traits<char>::transform_primary(const char* p1, const char* p2) |
---|
64 | { |
---|
65 | static char s_delim; |
---|
66 | static const int s_collate_type = ::boost::re_detail::find_sort_syntax(static_cast<c_regex_traits<char>*>(0), &s_delim); |
---|
67 | std::string result; |
---|
68 | // |
---|
69 | // What we do here depends upon the format of the sort key returned by |
---|
70 | // sort key returned by this->transform: |
---|
71 | // |
---|
72 | switch(s_collate_type) |
---|
73 | { |
---|
74 | case ::boost::re_detail::sort_C: |
---|
75 | case ::boost::re_detail::sort_unknown: |
---|
76 | // the best we can do is translate to lower case, then get a regular sort key: |
---|
77 | { |
---|
78 | result.assign(p1, p2); |
---|
79 | for(std::string::size_type i = 0; i < result.size(); ++i) |
---|
80 | result[i] = static_cast<char>((std::tolower)(static_cast<unsigned char>(result[i]))); |
---|
81 | result = transform(&*result.begin(), &*result.begin() + result.size()); |
---|
82 | break; |
---|
83 | } |
---|
84 | case ::boost::re_detail::sort_fixed: |
---|
85 | { |
---|
86 | // get a regular sort key, and then truncate it: |
---|
87 | result = transform(p1, p2); |
---|
88 | result.erase(s_delim); |
---|
89 | break; |
---|
90 | } |
---|
91 | case ::boost::re_detail::sort_delim: |
---|
92 | // get a regular sort key, and then truncate everything after the delim: |
---|
93 | result = transform(p1, p2); |
---|
94 | if(result.size() && (result[0] == s_delim)) |
---|
95 | break; |
---|
96 | std::size_t i; |
---|
97 | for(i = 0; i < result.size(); ++i) |
---|
98 | { |
---|
99 | if(result[i] == s_delim) |
---|
100 | break; |
---|
101 | } |
---|
102 | result.erase(i); |
---|
103 | break; |
---|
104 | } |
---|
105 | if(result.empty()) |
---|
106 | result = std::string(1, char(0)); |
---|
107 | return result; |
---|
108 | } |
---|
109 | |
---|
110 | enum |
---|
111 | { |
---|
112 | char_class_space=1<<0, |
---|
113 | char_class_print=1<<1, |
---|
114 | char_class_cntrl=1<<2, |
---|
115 | char_class_upper=1<<3, |
---|
116 | char_class_lower=1<<4, |
---|
117 | char_class_alpha=1<<5, |
---|
118 | char_class_digit=1<<6, |
---|
119 | char_class_punct=1<<7, |
---|
120 | char_class_xdigit=1<<8, |
---|
121 | char_class_alnum=char_class_alpha|char_class_digit, |
---|
122 | char_class_graph=char_class_alnum|char_class_punct, |
---|
123 | char_class_blank=1<<9, |
---|
124 | char_class_word=1<<10, |
---|
125 | char_class_unicode=1<<11 |
---|
126 | }; |
---|
127 | |
---|
128 | c_regex_traits<char>::char_class_type BOOST_REGEX_CALL c_regex_traits<char>::lookup_classname(const char* p1, const char* p2) |
---|
129 | { |
---|
130 | static const char_class_type masks[] = |
---|
131 | { |
---|
132 | 0, |
---|
133 | char_class_alnum, |
---|
134 | char_class_alpha, |
---|
135 | char_class_blank, |
---|
136 | char_class_cntrl, |
---|
137 | char_class_digit, |
---|
138 | char_class_digit, |
---|
139 | char_class_graph, |
---|
140 | char_class_lower, |
---|
141 | char_class_lower, |
---|
142 | char_class_print, |
---|
143 | char_class_punct, |
---|
144 | char_class_space, |
---|
145 | char_class_space, |
---|
146 | char_class_upper, |
---|
147 | char_class_unicode, |
---|
148 | char_class_upper, |
---|
149 | char_class_alnum | char_class_word, |
---|
150 | char_class_alnum | char_class_word, |
---|
151 | char_class_xdigit, |
---|
152 | }; |
---|
153 | |
---|
154 | int id = ::boost::re_detail::get_default_class_id(p1, p2); |
---|
155 | if(id < 0) |
---|
156 | { |
---|
157 | std::string s(p1, p2); |
---|
158 | for(std::string::size_type i = 0; i < s.size(); ++i) |
---|
159 | s[i] = static_cast<char>((std::tolower)(static_cast<unsigned char>(s[i]))); |
---|
160 | id = ::boost::re_detail::get_default_class_id(&*s.begin(), &*s.begin() + s.size()); |
---|
161 | } |
---|
162 | BOOST_ASSERT(std::size_t(id+1) < sizeof(masks) / sizeof(masks[0])); |
---|
163 | return masks[id+1]; |
---|
164 | } |
---|
165 | |
---|
166 | bool BOOST_REGEX_CALL c_regex_traits<char>::isctype(char c, char_class_type mask) |
---|
167 | { |
---|
168 | return |
---|
169 | ((mask & char_class_space) && (std::isspace)(static_cast<unsigned char>(c))) |
---|
170 | || ((mask & char_class_print) && (std::isprint)(static_cast<unsigned char>(c))) |
---|
171 | || ((mask & char_class_cntrl) && (std::iscntrl)(static_cast<unsigned char>(c))) |
---|
172 | || ((mask & char_class_upper) && (std::isupper)(static_cast<unsigned char>(c))) |
---|
173 | || ((mask & char_class_lower) && (std::islower)(static_cast<unsigned char>(c))) |
---|
174 | || ((mask & char_class_alpha) && (std::isalpha)(static_cast<unsigned char>(c))) |
---|
175 | || ((mask & char_class_digit) && (std::isdigit)(static_cast<unsigned char>(c))) |
---|
176 | || ((mask & char_class_punct) && (std::ispunct)(static_cast<unsigned char>(c))) |
---|
177 | || ((mask & char_class_xdigit) && (std::isxdigit)(static_cast<unsigned char>(c))) |
---|
178 | || ((mask & char_class_blank) && (std::isspace)(static_cast<unsigned char>(c)) && !::boost::re_detail::is_separator(c)) |
---|
179 | || ((mask & char_class_word) && (c == '_')); |
---|
180 | } |
---|
181 | |
---|
182 | c_regex_traits<char>::string_type BOOST_REGEX_CALL c_regex_traits<char>::lookup_collatename(const char* p1, const char* p2) |
---|
183 | { |
---|
184 | std::string s(p1, p2); |
---|
185 | s = ::boost::re_detail::lookup_default_collate_name(s); |
---|
186 | if(s.empty() && (p2-p1 == 1)) |
---|
187 | s.append(1, *p1); |
---|
188 | return s; |
---|
189 | } |
---|
190 | |
---|
191 | int BOOST_REGEX_CALL c_regex_traits<char>::value(char c, int radix) |
---|
192 | { |
---|
193 | char b[2] = { c, '\0', }; |
---|
194 | char* ep; |
---|
195 | int result = std::strtol(b, &ep, radix); |
---|
196 | if(ep == b) |
---|
197 | return -1; |
---|
198 | return result; |
---|
199 | } |
---|
200 | |
---|
201 | } |
---|
202 | #ifdef BOOST_HAS_ABI_HEADERS |
---|
203 | # include BOOST_ABI_SUFFIX |
---|
204 | #endif |
---|
205 | |
---|
206 | #endif |
---|