Overview
Comment: | OFStdIOStream_Win32Console: Improve writing
When writing an incomplete surrogate, it now writes everything up to |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | trunk |
Files: | files | file ages | folders |
SHA3-256: |
3a0fdb670130629b9fcbdf5e878b9845 |
User & Date: | js on 2016-03-13 19:33:09 |
Other Links: | manifest | tags |
Context
2016-03-13
| ||
20:04 | OFStdIOStream_Win32Console: Use U+FFFD, not U+FFFE check-in: 9d70e660ea user: js tags: trunk | |
19:33 | OFStdIOStream_Win32Console: Improve writing check-in: 3a0fdb6701 user: js tags: trunk | |
13:16 | of_string_utf8_decode(): Check all surrogate bytes check-in: a83b261f72 user: js tags: trunk | |
Changes
Modified src/OFStdIOStream_Win32Console.h from [34f51c6008] to [b8fefbb451].
︙ | ︙ | |||
19 20 21 22 23 24 25 | #import "OFStdIOStream.h" OF_ASSUME_NONNULL_BEGIN @interface OFStdIOStream_Win32Console: OFStdIOStream { HANDLE _handle; | | > > | 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | #import "OFStdIOStream.h" OF_ASSUME_NONNULL_BEGIN @interface OFStdIOStream_Win32Console: OFStdIOStream { HANDLE _handle; of_char16_t _incompleteUTF16Surrogate; char _incompleteUTF8Surrogate[4]; size_t _incompleteUTF8SurrogateLen; } @end OF_ASSUME_NONNULL_END |
Modified src/OFStdIOStream_Win32Console.m from [f55d77f8b8] to [c806288ec6].
︙ | ︙ | |||
26 27 28 29 30 31 32 | * written and read() just returns 0 as soon as a Unicode character is being * read. * * Therefore, instead of just using the UTF-8 codepage, this captures all reads * and writes to of_std{in,out,err} on the lowlevel, interprets the buffer as * UTF-8 and converts to / from UTF-16 to use ReadConsoleW() / WriteConsoleW(). * Doing so is safe, as the console only supports text anyway and thus it does | | > > > | 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 | * written and read() just returns 0 as soon as a Unicode character is being * read. * * Therefore, instead of just using the UTF-8 codepage, this captures all reads * and writes to of_std{in,out,err} on the lowlevel, interprets the buffer as * UTF-8 and converts to / from UTF-16 to use ReadConsoleW() / WriteConsoleW(). * Doing so is safe, as the console only supports text anyway and thus it does * not matter if binary gets garbled by the conversion (e.g. because invalid * UTF-8 gets converted to U+FFFE). * * In order to not do this when redirecting input / output to a file (as the * file would then be read / written in the wrong encoding and break reading / * writing binary), it checks that the handle is indeed a console. */ #define OF_STDIO_STREAM_WIN32_CONSOLE_M #include "config.h" #include <assert.h> #import "OFStdIOStream_Win32Console.h" #import "OFStdIOStream+Private.h" #import "OFString.h" #import "OFDataArray.h" #import "OFInvalidArgumentException.h" |
︙ | ︙ | |||
110 111 112 113 114 115 116 | @try { DWORD UTF16Len; OFDataArray *rest = nil; if (!ReadConsoleW(_handle, UTF16, length, &UTF16Len, NULL)) @throw [OFReadFailedException exceptionWithObject: self | | | | | | | 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 | @try { DWORD UTF16Len; OFDataArray *rest = nil; if (!ReadConsoleW(_handle, UTF16, length, &UTF16Len, NULL)) @throw [OFReadFailedException exceptionWithObject: self requestedLength: length * 2]; if (UTF16Len > 0 && _incompleteUTF16Surrogate != 0) { of_unichar_t c = (((_incompleteUTF16Surrogate & 0x3FF) << 10) | (UTF16[0] & 0x3FF)) + 0x10000; char UTF8[4]; size_t UTF8Len; if ((UTF8Len = of_string_utf8_encode(c, UTF8)) == 0) @throw [OFInvalidEncodingException exception]; if (UTF8Len <= length) { memcpy(buffer, UTF8, UTF8Len); j += UTF8Len; } else { if (rest == nil) rest = [OFDataArray dataArray]; [rest addItems: UTF8 count: UTF8Len]; } _incompleteUTF16Surrogate = 0; } for (size_t i = 0; i < UTF16Len; i++) { of_unichar_t c = UTF16[i]; char UTF8[4]; size_t UTF8Len; /* Missing high surrogate */ if ((c & 0xFC00) == 0xDC00) @throw [OFInvalidEncodingException exception]; if ((c & 0xFC00) == 0xD800) { of_char16_t next; if (UTF16Len <= i + 1) { _incompleteUTF16Surrogate = c; if (rest != nil) { char *items = [rest items]; size_t count = [rest count]; [self unreadFromBuffer: items length: count]; |
︙ | ︙ | |||
208 209 210 211 212 213 214 215 216 217 218 219 220 221 | } - (void)lowlevelWriteBuffer: (const void*)buffer_ length: (size_t)length { const char *buffer = buffer_; of_char16_t *tmp; if (length > SIZE_MAX / 2) @throw [OFOutOfRangeException exception]; tmp = [self allocMemoryWithSize: sizeof(of_char16_t) count: length * 2]; @try { | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > < | > > > > > > > > > > | < > > > > | | 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 | } - (void)lowlevelWriteBuffer: (const void*)buffer_ length: (size_t)length { const char *buffer = buffer_; of_char16_t *tmp; size_t i = 0, j = 0; if (length > SIZE_MAX / 2) @throw [OFOutOfRangeException exception]; if (_incompleteUTF8SurrogateLen > 0) { of_unichar_t c; of_char16_t UTF16[2]; ssize_t UTF8Len; size_t toCopy, UTF16Len; DWORD written; UTF8Len = -of_string_utf8_decode( _incompleteUTF8Surrogate, _incompleteUTF8SurrogateLen, &c); OF_ENSURE(UTF8Len > 0); toCopy = UTF8Len - _incompleteUTF8SurrogateLen; if (toCopy > length) toCopy = length; memcpy(_incompleteUTF8Surrogate + _incompleteUTF8SurrogateLen, buffer, toCopy); _incompleteUTF8SurrogateLen += toCopy; if (_incompleteUTF8SurrogateLen < (size_t)UTF8Len) return; UTF8Len = of_string_utf8_decode( _incompleteUTF8Surrogate, _incompleteUTF8SurrogateLen, &c); if (UTF8Len <= 0 || c > 0x10FFFF) { assert(UTF8Len == 0 || UTF8Len < -4); UTF16[0] = 0xFFFE; UTF16Len = 1; } else { if (c > 0xFFFF) { c -= 0x10000; UTF16[0] = 0xD800 | (c >> 10); UTF16[1] = 0xDC00 | (c & 0x3FF); UTF16Len = 2; } else { UTF16[0] = c; UTF16Len = 1; } } if (!WriteConsoleW(_handle, UTF16, UTF16Len, &written, NULL) || written != UTF16Len) @throw [OFWriteFailedException exceptionWithObject: self requestedLength: UTF16Len * 2]; _incompleteUTF8SurrogateLen = 0; i += toCopy; } tmp = [self allocMemoryWithSize: sizeof(of_char16_t) count: length * 2]; @try { DWORD written; while (i < length) { of_unichar_t c; ssize_t UTF8Len; UTF8Len = of_string_utf8_decode(buffer + i, length - i, &c); if (UTF8Len < 0 && UTF8Len >= -4) { OF_ENSURE(length - i < 4); memcpy(_incompleteUTF8Surrogate, buffer + i, length - i); _incompleteUTF8SurrogateLen = length - i; break; } if (UTF8Len <= 0 || c > 0x10FFFF) { tmp[j++] = 0xFFFE; i++; continue; } if (c > 0xFFFF) { c -= 0x10000; tmp[j++] = 0xD800 | (c >> 10); tmp[j++] = 0xDC00 | (c & 0x3FF); } else tmp[j++] = c; i += UTF8Len; } if (!WriteConsoleW(_handle, tmp, j, &written, NULL) || written != j) @throw [OFWriteFailedException exceptionWithObject: self requestedLength: j * 2]; } @finally { [self freeMemory: tmp]; } } @end |