include/boost/url/impl/encode.hpp
100.0% Lines (112/112)
85.0% Functions (17/20)
| Line | TLA | Hits | Source Code |
|---|---|---|---|
| 1 | // | ||
| 2 | // Copyright (c) 2019 Vinnie Falco (vinnie.falco@gmail.com) | ||
| 3 | // | ||
| 4 | // Distributed under the Boost Software License, Version 1.0. (See accompanying | ||
| 5 | // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) | ||
| 6 | // | ||
| 7 | // Official repository: https://github.com/boostorg/url | ||
| 8 | // | ||
| 9 | |||
| 10 | #ifndef BOOST_URL_IMPL_ENCODE_HPP | ||
| 11 | #define BOOST_URL_IMPL_ENCODE_HPP | ||
| 12 | |||
| 13 | #include <boost/url/grammar/token_rule.hpp> | ||
| 14 | #include <boost/assert.hpp> | ||
| 15 | #include <boost/core/detail/static_assert.hpp> | ||
| 16 | #include <boost/url/detail/encode.hpp> | ||
| 17 | #include <boost/url/detail/except.hpp> | ||
| 18 | #include <boost/url/encoding_opts.hpp> | ||
| 19 | #include <boost/url/grammar/charset.hpp> | ||
| 20 | #include <boost/url/grammar/hexdig_chars.hpp> | ||
| 21 | #include <boost/url/grammar/string_token.hpp> | ||
| 22 | #include <boost/url/grammar/type_traits.hpp> | ||
| 23 | |||
| 24 | namespace boost { | ||
| 25 | namespace urls { | ||
| 26 | |||
| 27 | //------------------------------------------------ | ||
| 28 | |||
| 29 | template<BOOST_URL_CONSTRAINT(grammar::CharSet) CS> | ||
| 30 | std::size_t | ||
| 31 | 5523 | encoded_size( | |
| 32 | core::string_view s, | ||
| 33 | CS const& allowed, | ||
| 34 | encoding_opts opt) noexcept | ||
| 35 | { | ||
| 36 | /* | ||
| 37 | If you get a compilation error here, it | ||
| 38 | means that the value you passed does | ||
| 39 | not meet the requirements stated in | ||
| 40 | the documentation. | ||
| 41 | */ | ||
| 42 | BOOST_CORE_STATIC_ASSERT( | ||
| 43 | grammar::is_charset<CS>::value); | ||
| 44 | |||
| 45 | 5523 | std::size_t n = 0; | |
| 46 | 5523 | auto it = s.data(); | |
| 47 | 5523 | auto const last = it + s.size(); | |
| 48 | |||
| 49 | 5523 | if (!opt.space_as_plus) | |
| 50 | { | ||
| 51 | 25592 | while (it != last) | |
| 52 | { | ||
| 53 | 22571 | char const c = *it; | |
| 54 | 22571 | if (allowed(c)) | |
| 55 | { | ||
| 56 | 19533 | ++n; | |
| 57 | } | ||
| 58 | else | ||
| 59 | { | ||
| 60 | 3038 | n += 3; | |
| 61 | } | ||
| 62 | 22571 | ++it; | |
| 63 | } | ||
| 64 | } | ||
| 65 | else | ||
| 66 | { | ||
| 67 | // '+' is always encoded (thus | ||
| 68 | // spending 3 chars) even if | ||
| 69 | // allowed because "%2B" and | ||
| 70 | // "+" have different meanings | ||
| 71 | // when space as plus is enabled | ||
| 72 | using FNT = bool (*)(CS const& allowed, char); | ||
| 73 | 2502 | FNT takes_one_char = | |
| 74 | 5004 | allowed('+') ? | |
| 75 | 2500 | (allowed(' ') ? | |
| 76 | 1 | FNT([](CS const& allowed, char c){ return allowed(c) && c != '+'; }) : | |
| 77 | 22580 | FNT([](CS const& allowed, char c){ return (allowed(c) || c == ' ') && c != '+'; })) : | |
| 78 | 2 | (allowed(' ') ? | |
| 79 | 1 | FNT([](CS const& allowed, char c){ return allowed(c); }) : | |
| 80 | 1 | FNT([](CS const& allowed, char c){ return allowed(c) || c == ' '; })); | |
| 81 | 22624 | while (it != last) | |
| 82 | { | ||
| 83 | 20122 | char const c = *it; | |
| 84 | 20122 | if (takes_one_char(allowed, c)) | |
| 85 | { | ||
| 86 | 16975 | ++n; | |
| 87 | } | ||
| 88 | else | ||
| 89 | { | ||
| 90 | 3147 | n += 3; | |
| 91 | } | ||
| 92 | 20122 | ++it; | |
| 93 | } | ||
| 94 | } | ||
| 95 | 5523 | return n; | |
| 96 | } | ||
| 97 | |||
| 98 | //------------------------------------------------ | ||
| 99 | |||
| 100 | template<BOOST_URL_CONSTRAINT(grammar::CharSet) CS> | ||
| 101 | std::size_t | ||
| 102 | 21818 | encode( | |
| 103 | char* dest, | ||
| 104 | std::size_t size, | ||
| 105 | core::string_view s, | ||
| 106 | CS const& allowed, | ||
| 107 | encoding_opts opt) | ||
| 108 | { | ||
| 109 | /* If you get a compilation error here, it | ||
| 110 | means that the value you passed does | ||
| 111 | not meet the requirements stated in | ||
| 112 | the documentation. | ||
| 113 | */ | ||
| 114 | BOOST_CORE_STATIC_ASSERT( | ||
| 115 | grammar::is_charset<CS>::value); | ||
| 116 | |||
| 117 | // '%' must be reserved | ||
| 118 | 21818 | BOOST_ASSERT(!allowed('%')); | |
| 119 | |||
| 120 | 21818 | char const* const hex = | |
| 121 | 21818 | detail::hexdigs[opt.lower_case]; | |
| 122 | 266448 | auto const encode = [hex]( | |
| 123 | char*& dest, | ||
| 124 | unsigned char c) noexcept | ||
| 125 | { | ||
| 126 | 244630 | *dest++ = '%'; | |
| 127 | 244630 | *dest++ = hex[c>>4]; | |
| 128 | 244630 | *dest++ = hex[c&0xf]; | |
| 129 | }; | ||
| 130 | |||
| 131 | 21818 | auto it = s.data(); | |
| 132 | 21818 | auto const end = dest + size; | |
| 133 | 21818 | auto const last = it + s.size(); | |
| 134 | 21818 | auto const dest0 = dest; | |
| 135 | |||
| 136 | 21818 | if (!opt.space_as_plus) | |
| 137 | { | ||
| 138 | 1303128 | while(it != last) | |
| 139 | { | ||
| 140 | 1294002 | char const c = *it; | |
| 141 | 1294002 | if (allowed(c)) | |
| 142 | { | ||
| 143 | 1048403 | if(dest == end) | |
| 144 | 6102 | return dest - dest0; | |
| 145 | 1042301 | *dest++ = c; | |
| 146 | 1042301 | ++it; | |
| 147 | 1042301 | continue; | |
| 148 | } | ||
| 149 | 245599 | if (end - dest < 3) | |
| 150 | 4088 | return dest - dest0; | |
| 151 | 241511 | encode(dest, c); | |
| 152 | 241511 | ++it; | |
| 153 | } | ||
| 154 | 9126 | return dest - dest0; | |
| 155 | } | ||
| 156 | else | ||
| 157 | { | ||
| 158 | 22594 | while (it != last) | |
| 159 | { | ||
| 160 | 20106 | char const c = *it; | |
| 161 | 20106 | if (c == ' ') | |
| 162 | { | ||
| 163 | 262 | if(dest == end) | |
| 164 | 2 | return dest - dest0; | |
| 165 | 260 | *dest++ = '+'; | |
| 166 | 260 | ++it; | |
| 167 | 260 | continue; | |
| 168 | } | ||
| 169 | 36537 | else if ( | |
| 170 | 19844 | allowed(c) && | |
| 171 | c != '+') | ||
| 172 | { | ||
| 173 | 16696 | if(dest == end) | |
| 174 | 3 | return dest - dest0; | |
| 175 | 16693 | *dest++ = c; | |
| 176 | 16693 | ++it; | |
| 177 | 16693 | continue; | |
| 178 | } | ||
| 179 | 3148 | if(end - dest < 3) | |
| 180 | 9 | return dest - dest0; | |
| 181 | 3139 | encode(dest, c); | |
| 182 | 3139 | ++it; | |
| 183 | } | ||
| 184 | } | ||
| 185 | 2488 | return dest - dest0; | |
| 186 | } | ||
| 187 | |||
| 188 | //------------------------------------------------ | ||
| 189 | |||
| 190 | // unsafe encode just | ||
| 191 | // asserts on the output buffer | ||
| 192 | // | ||
| 193 | template<BOOST_URL_CONSTRAINT(grammar::CharSet) CS> | ||
| 194 | std::size_t | ||
| 195 | 1569 | encode_unsafe( | |
| 196 | char* dest, | ||
| 197 | std::size_t size, | ||
| 198 | core::string_view s, | ||
| 199 | CS const& allowed, | ||
| 200 | encoding_opts opt) | ||
| 201 | { | ||
| 202 | BOOST_CORE_STATIC_ASSERT( | ||
| 203 | grammar::is_charset<CS>::value); | ||
| 204 | |||
| 205 | // '%' must be reserved | ||
| 206 | 1569 | BOOST_ASSERT(!allowed('%')); | |
| 207 | |||
| 208 | 1569 | auto it = s.data(); | |
| 209 | 1569 | auto const last = it + s.size(); | |
| 210 | 1569 | auto const end = dest + size; | |
| 211 | ignore_unused(end); | ||
| 212 | |||
| 213 | 1569 | char const* const hex = | |
| 214 | 1569 | detail::hexdigs[opt.lower_case]; | |
| 215 | 2923 | auto const encode = [end, hex]( | |
| 216 | char*& dest, | ||
| 217 | unsigned char c) noexcept | ||
| 218 | { | ||
| 219 | 677 | ignore_unused(end); | |
| 220 | 677 | *dest++ = '%'; | |
| 221 | 677 | BOOST_ASSERT(dest != end); | |
| 222 | 677 | *dest++ = hex[c>>4]; | |
| 223 | 677 | BOOST_ASSERT(dest != end); | |
| 224 | 677 | *dest++ = hex[c&0xf]; | |
| 225 | }; | ||
| 226 | |||
| 227 | 1569 | auto const dest0 = dest; | |
| 228 | 1569 | if (!opt.space_as_plus) | |
| 229 | { | ||
| 230 | 8140 | while(it != last) | |
| 231 | { | ||
| 232 | 6584 | BOOST_ASSERT(dest != end); | |
| 233 | 6584 | char const c = *it; | |
| 234 | 6584 | if(allowed(c)) | |
| 235 | { | ||
| 236 | 5899 | *dest++ = c; | |
| 237 | } | ||
| 238 | else | ||
| 239 | { | ||
| 240 | 685 | encode(dest, c); | |
| 241 | } | ||
| 242 | 6584 | ++it; | |
| 243 | } | ||
| 244 | } | ||
| 245 | else | ||
| 246 | { | ||
| 247 | 53 | while(it != last) | |
| 248 | { | ||
| 249 | 40 | BOOST_ASSERT(dest != end); | |
| 250 | 40 | char const c = *it; | |
| 251 | 40 | if (c == ' ') | |
| 252 | { | ||
| 253 | 9 | *dest++ = '+'; | |
| 254 | } | ||
| 255 | 31 | else if ( | |
| 256 | 31 | allowed(c) && | |
| 257 | c != '+') | ||
| 258 | { | ||
| 259 | 23 | *dest++ = c; | |
| 260 | } | ||
| 261 | else | ||
| 262 | { | ||
| 263 | 8 | encode(dest, c); | |
| 264 | } | ||
| 265 | 40 | ++it; | |
| 266 | } | ||
| 267 | } | ||
| 268 | 1569 | return dest - dest0; | |
| 269 | } | ||
| 270 | |||
| 271 | //------------------------------------------------ | ||
| 272 | |||
| 273 | template< | ||
| 274 | BOOST_URL_CONSTRAINT(string_token::StringToken) StringToken, | ||
| 275 | BOOST_URL_CONSTRAINT(grammar::CharSet) CS> | ||
| 276 | BOOST_URL_STRTOK_RETURN | ||
| 277 | 28 | encode( | |
| 278 | core::string_view s, | ||
| 279 | CS const& allowed, | ||
| 280 | encoding_opts opt, | ||
| 281 | StringToken&& token) | ||
| 282 | { | ||
| 283 | BOOST_CORE_STATIC_ASSERT( | ||
| 284 | grammar::is_charset<CS>::value); | ||
| 285 | |||
| 286 | 28 | auto const n = encoded_size( | |
| 287 | s, allowed, opt); | ||
| 288 | 28 | auto p = token.prepare(n); | |
| 289 | 28 | if(n > 0) | |
| 290 | 26 | encode_unsafe( | |
| 291 | p, n, s, allowed, opt); | ||
| 292 | 28 | return token.result(); | |
| 293 | } | ||
| 294 | |||
| 295 | } // urls | ||
| 296 | } // boost | ||
| 297 | |||
| 298 | #endif | ||
| 299 |