| 1 | // |
| 2 | // Copyright (c) 2009-2011 Artyom Beilis (Tonkikh) |
| 3 | // |
| 4 | // Distributed under the Boost Software License, Version 1.0. |
| 5 | // https://www.boost.org/LICENSE_1_0.txt |
| 6 | |
| 7 | #include <boost/locale/utf.hpp> |
| 8 | #include <boost/locale/util/string.hpp> |
| 9 | #include "boostLocale/test/tools.hpp" |
| 10 | #include "boostLocale/test/unit_test.hpp" |
| 11 | #include <boost/detail/workaround.hpp> |
| 12 | #include <cstring> |
| 13 | |
| 14 | using namespace boost::locale::utf; |
| 15 | |
| 16 | const std::uint32_t* u32_seq(std::uint32_t a) |
| 17 | { |
| 18 | static std::uint32_t buf[2]; |
| 19 | buf[0] = a; |
| 20 | buf[1] = 0; |
| 21 | return buf; |
| 22 | } |
| 23 | |
| 24 | const std::uint16_t* u16_seq(std::uint16_t a) |
| 25 | { |
| 26 | static std::uint16_t buf[2]; |
| 27 | buf[0] = a; |
| 28 | buf[1] = 0; |
| 29 | return buf; |
| 30 | } |
| 31 | |
| 32 | const std::uint16_t* u16_seq(std::uint16_t a, std::uint16_t b) |
| 33 | { |
| 34 | static std::uint16_t buf[3]; |
| 35 | buf[0] = a; |
| 36 | buf[1] = b; |
| 37 | buf[2] = 0; |
| 38 | return buf; |
| 39 | } |
| 40 | |
| 41 | const char16_t* c16_seq(std::uint16_t a) |
| 42 | { |
| 43 | static char16_t buf[2]; |
| 44 | buf[0] = static_cast<char16_t>(a); |
| 45 | buf[1] = 0; |
| 46 | return buf; |
| 47 | } |
| 48 | |
| 49 | const char32_t* c32_seq(std::uint32_t a) |
| 50 | { |
| 51 | static char32_t buf[2]; |
| 52 | buf[0] = static_cast<char32_t>(a); |
| 53 | buf[1] = 0; |
| 54 | return buf; |
| 55 | } |
| 56 | |
| 57 | template<typename CharType> |
| 58 | void test_from_utf(const CharType* const s, unsigned codepoint) |
| 59 | { |
| 60 | const CharType* cur = s; |
| 61 | const CharType* const end = boost::locale::util::str_end(s); |
| 62 | |
| 63 | typedef utf_traits<CharType> tr; |
| 64 | |
| 65 | static_assert(tr::max_width == 4 / sizeof(CharType), "Wrong max_width" ); |
| 66 | |
| 67 | TEST_EQ(tr::decode(cur, end), codepoint); |
| 68 | |
| 69 | if(codepoint != illegal) |
| 70 | TEST(cur == end); |
| 71 | |
| 72 | if(codepoint == incomplete) { |
| 73 | TEST(*s == 0 || tr::trail_length(*s) > 0); |
| 74 | TEST_GE(tr::trail_length(*s), end - s); |
| 75 | } |
| 76 | |
| 77 | if(codepoint != incomplete && codepoint != illegal) { |
| 78 | TEST(tr::is_lead(*s)); |
| 79 | TEST(!tr::is_trail(*s)); |
| 80 | cur = s; |
| 81 | while(++cur != end) { |
| 82 | TEST(tr::is_trail(*cur)); |
| 83 | TEST(!tr::is_lead(*cur)); |
| 84 | } |
| 85 | TEST_EQ(tr::width(codepoint), end - s); |
| 86 | TEST_EQ(tr::trail_length(*s), tr::width(codepoint) - 1); |
| 87 | cur = s; |
| 88 | TEST_EQ(tr::decode_valid(cur), codepoint); |
| 89 | TEST(cur == end); |
| 90 | } |
| 91 | } |
| 92 | |
| 93 | template<typename CharType> |
| 94 | void test_to_utf(const CharType* str, unsigned codepoint) |
| 95 | { |
| 96 | CharType buf[5] = {1, 1, 1, 1, 1}; |
| 97 | CharType* p = buf; |
| 98 | p = utf_traits<CharType>::encode(codepoint, p); |
| 99 | const CharType* const end = boost::locale::util::str_end(str); |
| 100 | TEST_EQ(end - str, p - buf); |
| 101 | TEST(*p); |
| 102 | *p = 0; |
| 103 | TEST_EQ(memcmp(str, buf, sizeof(CharType) * (end - str)), 0); |
| 104 | } |
| 105 | |
| 106 | template<typename CharType> |
| 107 | void test_valid_utf(const CharType* str, unsigned codepoint) |
| 108 | { |
| 109 | test_from_utf(str, codepoint); |
| 110 | test_to_utf(str, codepoint); |
| 111 | } |
| 112 | |
| 113 | void test_utf8() |
| 114 | { |
| 115 | std::cout << "- Test UTF-8\n" ; |
| 116 | |
| 117 | std::cout << "-- Correct" << std::endl; |
| 118 | test_valid_utf(str: "\x7f" , codepoint: 0x7f); |
| 119 | test_valid_utf(str: "\xc2\x80" , codepoint: 0x80); |
| 120 | test_valid_utf(str: "\xdf\xbf" , codepoint: 0x7ff); |
| 121 | test_valid_utf(str: "\xe0\xa0\x80" , codepoint: 0x800); |
| 122 | test_valid_utf(str: "\xef\xbf\xbf" , codepoint: 0xffff); |
| 123 | test_valid_utf(str: "\xf0\x90\x80\x80" , codepoint: 0x10000); |
| 124 | test_valid_utf(str: "\xf4\x8f\xbf\xbf" , codepoint: 0x10ffff); |
| 125 | |
| 126 | // test that this actually works |
| 127 | test_from_utf(s: make2(v: 0x80), codepoint: 0x80); |
| 128 | test_from_utf(s: make2(v: 0x7ff), codepoint: 0x7ff); |
| 129 | |
| 130 | test_from_utf(s: make3(v: 0x800), codepoint: 0x800); |
| 131 | test_from_utf(s: make3(v: 0xffff), codepoint: 0xffff); |
| 132 | |
| 133 | test_from_utf(s: make4(v: 0x10000), codepoint: 0x10000); |
| 134 | test_from_utf(s: make4(v: 0x10ffff), codepoint: 0x10ffff); |
| 135 | |
| 136 | std::cout << "-- Too big" << std::endl; |
| 137 | test_from_utf(s: "\xf4\x9f\x80\x80" , codepoint: illegal); // 11 0000 |
| 138 | test_from_utf(s: "\xfb\xbf\xbf\xbf" , codepoint: illegal); // 3ff ffff |
| 139 | test_from_utf(s: "\xf8\x90\x80\x80\x80" , codepoint: illegal); // 400 0000 |
| 140 | test_from_utf(s: "\xfd\xbf\xbf\xbf\xbf\xbf" , codepoint: illegal); // 7fff ffff |
| 141 | |
| 142 | std::cout << "-- Invalid length" << std::endl; |
| 143 | |
| 144 | test_from_utf(s: make2(v: 0), codepoint: illegal); |
| 145 | test_from_utf(s: make3(v: 0), codepoint: illegal); |
| 146 | test_from_utf(s: make4(v: 0), codepoint: illegal); |
| 147 | test_from_utf(s: make2(v: 0x7f), codepoint: illegal); |
| 148 | test_from_utf(s: make3(v: 0x7f), codepoint: illegal); |
| 149 | test_from_utf(s: make4(v: 0x7f), codepoint: illegal); |
| 150 | |
| 151 | test_from_utf(s: make3(v: 0x80), codepoint: illegal); |
| 152 | test_from_utf(s: make4(v: 0x80), codepoint: illegal); |
| 153 | test_from_utf(s: make3(v: 0x7ff), codepoint: illegal); |
| 154 | test_from_utf(s: make4(v: 0x7ff), codepoint: illegal); |
| 155 | |
| 156 | test_from_utf(s: make4(v: 0x8000), codepoint: illegal); |
| 157 | test_from_utf(s: make4(v: 0xffff), codepoint: illegal); |
| 158 | test_from_utf(s: make4(v: 0x110000), codepoint: illegal); |
| 159 | test_from_utf(s: make4(v: 0x1fffff), codepoint: illegal); |
| 160 | |
| 161 | std::cout << "-- Invalid surrogate" << std::endl; |
| 162 | |
| 163 | test_from_utf(s: make3(v: 0xd800), codepoint: illegal); |
| 164 | test_from_utf(s: make3(v: 0xdbff), codepoint: illegal); |
| 165 | test_from_utf(s: make3(v: 0xdc00), codepoint: illegal); |
| 166 | test_from_utf(s: make3(v: 0xdfff), codepoint: illegal); |
| 167 | |
| 168 | test_from_utf(s: make4(v: 0xd800), codepoint: illegal); |
| 169 | test_from_utf(s: make4(v: 0xdbff), codepoint: illegal); |
| 170 | test_from_utf(s: make4(v: 0xdc00), codepoint: illegal); |
| 171 | test_from_utf(s: make4(v: 0xdfff), codepoint: illegal); |
| 172 | |
| 173 | std::cout << "-- Incomplete" << std::endl; |
| 174 | |
| 175 | test_from_utf(s: "" , codepoint: incomplete); |
| 176 | |
| 177 | test_from_utf(s: "\x80" , codepoint: illegal); |
| 178 | test_from_utf(s: "\xc2" , codepoint: incomplete); |
| 179 | |
| 180 | test_from_utf(s: "\xdf" , codepoint: incomplete); |
| 181 | |
| 182 | test_from_utf(s: "\xe0" , codepoint: incomplete); |
| 183 | test_from_utf(s: "\xe0\xa0" , codepoint: incomplete); |
| 184 | |
| 185 | test_from_utf(s: "\xef\xbf" , codepoint: incomplete); |
| 186 | test_from_utf(s: "\xef" , codepoint: incomplete); |
| 187 | |
| 188 | test_from_utf(s: "\xf0\x90\x80" , codepoint: incomplete); |
| 189 | test_from_utf(s: "\xf0\x90" , codepoint: incomplete); |
| 190 | test_from_utf(s: "\xf0" , codepoint: incomplete); |
| 191 | |
| 192 | test_from_utf(s: "\xf4\x8f\xbf" , codepoint: incomplete); |
| 193 | test_from_utf(s: "\xf4\x8f" , codepoint: incomplete); |
| 194 | test_from_utf(s: "\xf4" , codepoint: incomplete); |
| 195 | } |
| 196 | |
| 197 | void test_utf16() |
| 198 | { |
| 199 | std::cout << "- Test UTF-16\n" ; |
| 200 | |
| 201 | std::cout << "-- Correct" << std::endl; |
| 202 | test_valid_utf(str: u16_seq(a: 0x10), codepoint: 0x10); |
| 203 | test_valid_utf(str: u16_seq(a: 0xffff), codepoint: 0xffff); |
| 204 | test_valid_utf(str: u16_seq(a: 0xD800, b: 0xDC00), codepoint: 0x10000); |
| 205 | test_valid_utf(str: u16_seq(a: 0xDBFF, b: 0xDFFF), codepoint: 0x10FFFF); |
| 206 | |
| 207 | std::cout << "-- Invalid surrogate" << std::endl; |
| 208 | test_from_utf(s: u16_seq(a: 0xDFFF), codepoint: illegal); |
| 209 | test_from_utf(s: u16_seq(a: 0xDC00), codepoint: illegal); |
| 210 | |
| 211 | std::cout << "-- Incomplete" << std::endl; |
| 212 | test_from_utf(s: u16_seq(a: 0), codepoint: incomplete); |
| 213 | test_from_utf(s: u16_seq(a: 0xD800), codepoint: incomplete); |
| 214 | test_from_utf(s: u16_seq(a: 0xDBFF), codepoint: incomplete); |
| 215 | |
| 216 | std::cout << "-- Test char16_t" << std::endl; |
| 217 | #if BOOST_WORKAROUND(BOOST_GCC_VERSION, < 50000) |
| 218 | test_valid_utf(u"\x0010" , 0x10); |
| 219 | test_valid_utf(u"\xffff" , 0xffff); |
| 220 | #else |
| 221 | test_valid_utf(str: u"\u0010" , codepoint: 0x10); |
| 222 | test_valid_utf(str: u"\uffff" , codepoint: 0xffff); |
| 223 | #endif |
| 224 | test_valid_utf(str: u"\U00010000" , codepoint: 0x10000); |
| 225 | test_valid_utf(str: u"\U0010FFFF" , codepoint: 0x10FFFF); |
| 226 | test_from_utf(s: c16_seq(a: 0xDFFF), codepoint: illegal); |
| 227 | test_from_utf(s: c16_seq(a: 0xDC00), codepoint: illegal); |
| 228 | } |
| 229 | |
| 230 | void test_utf32() |
| 231 | { |
| 232 | std::cout << "- Test UTF-32\n" ; |
| 233 | |
| 234 | std::cout << "-- Correct" << std::endl; |
| 235 | test_valid_utf(str: u32_seq(a: 0x10), codepoint: 0x10); |
| 236 | test_valid_utf(str: u32_seq(a: 0xffff), codepoint: 0xffff); |
| 237 | test_valid_utf(str: u32_seq(a: 0x10000), codepoint: 0x10000); |
| 238 | test_valid_utf(str: u32_seq(a: 0x10ffff), codepoint: 0x10ffff); |
| 239 | |
| 240 | std::cout << "-- Invalid surrogate" << std::endl; |
| 241 | test_from_utf(s: u32_seq(a: 0xD800), codepoint: illegal); |
| 242 | test_from_utf(s: u32_seq(a: 0xDBFF), codepoint: illegal); |
| 243 | test_from_utf(s: u32_seq(a: 0xDFFF), codepoint: illegal); |
| 244 | test_from_utf(s: u32_seq(a: 0xDC00), codepoint: illegal); |
| 245 | test_from_utf(s: u32_seq(a: 0x110000), codepoint: illegal); |
| 246 | |
| 247 | std::cout << "-- Incomplete" << std::endl; |
| 248 | test_from_utf(s: u32_seq(a: 0), codepoint: incomplete); |
| 249 | |
| 250 | std::cout << "-- Test char32_t" << std::endl; |
| 251 | #if BOOST_WORKAROUND(BOOST_GCC_VERSION, < 50000) |
| 252 | test_valid_utf(U"\x0010" , 0x10); |
| 253 | #else |
| 254 | test_valid_utf(str: U"\U00000010" , codepoint: 0x10); |
| 255 | #endif |
| 256 | test_valid_utf(str: U"\U0000ffff" , codepoint: 0xffff); |
| 257 | test_valid_utf(str: U"\U00010000" , codepoint: 0x10000); |
| 258 | test_valid_utf(str: U"\U0010ffff" , codepoint: 0x10ffff); |
| 259 | test_from_utf(s: c32_seq(a: 0xD800), codepoint: illegal); |
| 260 | test_from_utf(s: c32_seq(a: 0xDBFF), codepoint: illegal); |
| 261 | test_from_utf(s: c32_seq(a: 0xDFFF), codepoint: illegal); |
| 262 | test_from_utf(s: c32_seq(a: 0xDC00), codepoint: illegal); |
| 263 | test_from_utf(s: c32_seq(a: 0x110000), codepoint: illegal); |
| 264 | } |
| 265 | |
| 266 | void test_main(int /*argc*/, char** /*argv*/) |
| 267 | { |
| 268 | test_utf8(); |
| 269 | test_utf16(); |
| 270 | test_utf32(); |
| 271 | } |
| 272 | |
| 273 | // boostinspect:noascii |
| 274 | |