Spaces:
Running
Running
File size: 14,117 Bytes
3c778ff | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 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 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 | import { toHexString } from 'src/utils/strings';
/**
* Encodes a string to UTF-8.
*
* @param input The string to be encoded.
* @param byteOrderMark Whether or not a byte order marker (BOM) should be added
* to the start of the encoding. (default `true`)
* @returns A Uint8Array containing the UTF-8 encoding of the input string.
*
* -----------------------------------------------------------------------------
*
* JavaScript strings are composed of Unicode code points. Code points are
* integers in the range 0 to 1,114,111 (0x10FFFF). When serializing a string,
* it must be encoded as a sequence of words. A word is typically 8, 16, or 32
* bytes in size. As such, Unicode defines three encoding forms: UTF-8, UTF-16,
* and UTF-32. These encoding forms are described in the Unicode standard [1].
* This function implements the UTF-8 encoding form.
*
* -----------------------------------------------------------------------------
*
* In UTF-8, each code point is mapped to a sequence of 1, 2, 3, or 4 bytes.
* Note that the logic which defines this mapping is slightly convoluted, and
* not as straightforward as the mapping logic for UTF-16 or UTF-32. The UTF-8
* mapping logic is as follows [2]:
*
* • If a code point is in the range U+0000..U+007F, then view it as a 7-bit
* integer: 0bxxxxxxx. Map the code point to 1 byte with the first high order
* bit set to 0:
*
* b1=0b0xxxxxxx
*
* • If a code point is in the range U+0080..U+07FF, then view it as an 11-bit
* integer: 0byyyyyxxxxxx. Map the code point to 2 bytes with the first 5 bits
* of the code point stored in the first byte, and the last 6 bits stored in
* the second byte:
*
* b1=0b110yyyyy b2=0b10xxxxxx
*
* • If a code point is in the range U+0800..U+FFFF, then view it as a 16-bit
* integer, 0bzzzzyyyyyyxxxxxx. Map the code point to 3 bytes with the first
* 4 bits stored in the first byte, the next 6 bits stored in the second byte,
* and the last 6 bits in the third byte:
*
* b1=0b1110zzzz b2=0b10yyyyyy b3=0b10xxxxxx
*
* • If a code point is in the range U+10000...U+10FFFF, then view it as a
* 21-bit integer, 0bvvvzzzzzzyyyyyyxxxxxx. Map the code point to 4 bytes with
* the first 3 bits stored in the first byte, the next 6 bits stored in the
* second byte, the next 6 bits stored in the third byte, and the last 6 bits
* stored in the fourth byte:
*
* b1=0b11110xxx b2=0b10zzzzzz b3=0b10yyyyyy b4=0b10xxxxxx
*
* -----------------------------------------------------------------------------
*
* It is important to note, when iterating through the code points of a string
* in JavaScript, that if a character is encoded as a surrogate pair it will
* increase the string's length by 2 instead of 1 [4]. For example:
*
* ```
* > 'a'.length
* 1
* > '💩'.length
* 2
* > '語'.length
* 1
* > 'a💩語'.length
* 4
* ```
*
* The results of the above example are explained by the fact that the
* characters 'a' and '語' are not represented by surrogate pairs, but '💩' is.
*
* Because of this idiosyncrasy in JavaScript's string implementation and APIs,
* we must "jump" an extra index after encoding a character as a surrogate
* pair. In practice, this means we must increment the index of our for loop by
* 2 if we encode a surrogate pair, and 1 in all other cases.
*
* -----------------------------------------------------------------------------
*
* References:
* - [1] https://www.unicode.org/versions/Unicode12.0.0/UnicodeStandard-12.0.pdf
* 3.9 Unicode Encoding Forms - UTF-8
* - [2] http://www.herongyang.com/Unicode/UTF-8-UTF-8-Encoding.html
* - [3] http://www.herongyang.com/Unicode/UTF-8-UTF-8-Encoding-Algorithm.html
* - [4] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/length#Description
*
*/
export const utf8Encode = (input: string, byteOrderMark = true): Uint8Array => {
const encoded = [];
if (byteOrderMark) encoded.push(0xef, 0xbb, 0xbf);
for (let idx = 0, len = input.length; idx < len; ) {
const codePoint = input.codePointAt(idx)!;
// One byte encoding
if (codePoint < 0x80) {
const byte1 = codePoint & 0x7f;
encoded.push(byte1);
idx += 1;
}
// Two byte encoding
else if (codePoint < 0x0800) {
const byte1 = ((codePoint >> 6) & 0x1f) | 0xc0;
const byte2 = (codePoint & 0x3f) | 0x80;
encoded.push(byte1, byte2);
idx += 1;
}
// Three byte encoding
else if (codePoint < 0x010000) {
const byte1 = ((codePoint >> 12) & 0x0f) | 0xe0;
const byte2 = ((codePoint >> 6) & 0x3f) | 0x80;
const byte3 = (codePoint & 0x3f) | 0x80;
encoded.push(byte1, byte2, byte3);
idx += 1;
}
// Four byte encoding (surrogate pair)
else if (codePoint < 0x110000) {
const byte1 = ((codePoint >> 18) & 0x07) | 0xf0;
const byte2 = ((codePoint >> 12) & 0x3f) | 0x80;
const byte3 = ((codePoint >> 6) & 0x3f) | 0x80;
const byte4 = ((codePoint >> 0) & 0x3f) | 0x80;
encoded.push(byte1, byte2, byte3, byte4);
idx += 2;
}
// Should never reach this case
else throw new Error(`Invalid code point: 0x${toHexString(codePoint)}`);
}
return new Uint8Array(encoded);
};
/**
* Encodes a string to UTF-16.
*
* @param input The string to be encoded.
* @param byteOrderMark Whether or not a byte order marker (BOM) should be added
* to the start of the encoding. (default `true`)
* @returns A Uint16Array containing the UTF-16 encoding of the input string.
*
* -----------------------------------------------------------------------------
*
* JavaScript strings are composed of Unicode code points. Code points are
* integers in the range 0 to 1,114,111 (0x10FFFF). When serializing a string,
* it must be encoded as a sequence of words. A word is typically 8, 16, or 32
* bytes in size. As such, Unicode defines three encoding forms: UTF-8, UTF-16,
* and UTF-32. These encoding forms are described in the Unicode standard [1].
* This function implements the UTF-16 encoding form.
*
* -----------------------------------------------------------------------------
*
* In UTF-16, each code point is mapped to one or two 16-bit integers. The
* UTF-16 mapping logic is as follows [2]:
*
* • If a code point is in the range U+0000..U+FFFF, then map the code point to
* a 16-bit integer with the most significant byte first.
*
* • If a code point is in the range U+10000..U+10000, then map the code point
* to two 16-bit integers. The first integer should contain the high surrogate
* and the second integer should contain the low surrogate. Both surrogates
* should be written with the most significant byte first.
*
* -----------------------------------------------------------------------------
*
* It is important to note, when iterating through the code points of a string
* in JavaScript, that if a character is encoded as a surrogate pair it will
* increase the string's length by 2 instead of 1 [4]. For example:
*
* ```
* > 'a'.length
* 1
* > '💩'.length
* 2
* > '語'.length
* 1
* > 'a💩語'.length
* 4
* ```
*
* The results of the above example are explained by the fact that the
* characters 'a' and '語' are not represented by surrogate pairs, but '💩' is.
*
* Because of this idiosyncrasy in JavaScript's string implementation and APIs,
* we must "jump" an extra index after encoding a character as a surrogate
* pair. In practice, this means we must increment the index of our for loop by
* 2 if we encode a surrogate pair, and 1 in all other cases.
*
* -----------------------------------------------------------------------------
*
* References:
* - [1] https://www.unicode.org/versions/Unicode12.0.0/UnicodeStandard-12.0.pdf
* 3.9 Unicode Encoding Forms - UTF-8
* - [2] http://www.herongyang.com/Unicode/UTF-16-UTF-16-Encoding.html
* - [3] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/length#Description
*
*/
export const utf16Encode = (
input: string,
byteOrderMark = true,
): Uint16Array => {
const encoded = [];
if (byteOrderMark) encoded.push(0xfeff);
for (let idx = 0, len = input.length; idx < len; ) {
const codePoint = input.codePointAt(idx)!;
// Two byte encoding
if (codePoint < 0x010000) {
encoded.push(codePoint);
idx += 1;
}
// Four byte encoding (surrogate pair)
else if (codePoint < 0x110000) {
encoded.push(highSurrogate(codePoint), lowSurrogate(codePoint));
idx += 2;
}
// Should never reach this case
else throw new Error(`Invalid code point: 0x${toHexString(codePoint)}`);
}
return new Uint16Array(encoded);
};
/**
* Returns `true` if the `codePoint` is within the
* Basic Multilingual Plane (BMP). Code points inside the BMP are not encoded
* with surrogate pairs.
* @param codePoint The code point to be evaluated.
*
* Reference: https://en.wikipedia.org/wiki/UTF-16#Description
*/
export const isWithinBMP = (codePoint: number) =>
codePoint >= 0 && codePoint <= 0xffff;
/**
* Returns `true` if the given `codePoint` is valid and must be represented
* with a surrogate pair when encoded.
* @param codePoint The code point to be evaluated.
*
* Reference: https://en.wikipedia.org/wiki/UTF-16#Description
*/
export const hasSurrogates = (codePoint: number) =>
codePoint >= 0x010000 && codePoint <= 0x10ffff;
// From Unicode 3.0 spec, section 3.7:
// http://unicode.org/versions/Unicode3.0.0/ch03.pdf
export const highSurrogate = (codePoint: number) =>
Math.floor((codePoint - 0x10000) / 0x400) + 0xd800;
// From Unicode 3.0 spec, section 3.7:
// http://unicode.org/versions/Unicode3.0.0/ch03.pdf
export const lowSurrogate = (codePoint: number) =>
((codePoint - 0x10000) % 0x400) + 0xdc00;
enum ByteOrder {
BigEndian = 'BigEndian',
LittleEndian = 'LittleEndian',
}
const REPLACEMENT = '�'.codePointAt(0)!;
/**
* Decodes a Uint8Array of data to a string using UTF-16.
*
* Note that this function attempts to recover from erronous input by
* inserting the replacement character (�) to mark invalid code points
* and surrogate pairs.
*
* @param input A Uint8Array containing UTF-16 encoded data
* @param byteOrderMark Whether or not a byte order marker (BOM) should be read
* at the start of the encoding. (default `true`)
* @returns The decoded string.
*/
export const utf16Decode = (
input: Uint8Array,
byteOrderMark = true,
): string => {
// Need at least 2 bytes of data in UTF-16 encodings
if (input.length <= 1) return String.fromCodePoint(REPLACEMENT);
const byteOrder = byteOrderMark ? readBOM(input) : ByteOrder.BigEndian;
// Skip byte order mark if needed
let idx = byteOrderMark ? 2 : 0;
const codePoints: number[] = [];
while (input.length - idx >= 2) {
const first = decodeValues(input[idx++], input[idx++], byteOrder);
if (isHighSurrogate(first)) {
if (input.length - idx < 2) {
// Need at least 2 bytes left for the low surrogate that is required
codePoints.push(REPLACEMENT);
} else {
const second = decodeValues(input[idx++], input[idx++], byteOrder);
if (isLowSurrogate(second)) {
codePoints.push(first, second);
} else {
// Low surrogates should always follow high surrogates
codePoints.push(REPLACEMENT);
}
}
} else if (isLowSurrogate(first)) {
// High surrogates should always come first since `decodeValues()`
// accounts for the byte ordering
idx += 2;
codePoints.push(REPLACEMENT);
} else {
codePoints.push(first);
}
}
// There shouldn't be extra byte(s) left over
if (idx < input.length) codePoints.push(REPLACEMENT);
return String.fromCodePoint(...codePoints);
};
/**
* Returns `true` if the given `codePoint` is a high surrogate.
* @param codePoint The code point to be evaluated.
*
* Reference: https://en.wikipedia.org/wiki/UTF-16#Description
*/
const isHighSurrogate = (codePoint: number) =>
codePoint >= 0xd800 && codePoint <= 0xdbff;
/**
* Returns `true` if the given `codePoint` is a low surrogate.
* @param codePoint The code point to be evaluated.
*
* Reference: https://en.wikipedia.org/wiki/UTF-16#Description
*/
const isLowSurrogate = (codePoint: number) =>
codePoint >= 0xdc00 && codePoint <= 0xdfff;
/**
* Decodes the given utf-16 values first and second using the specified
* byte order.
* @param first The first byte of the encoding.
* @param second The second byte of the encoding.
* @param byteOrder The byte order of the encoding.
* Reference: https://en.wikipedia.org/wiki/UTF-16#Examples
*/
const decodeValues = (first: number, second: number, byteOrder: ByteOrder) => {
// Append the binary representation of the preceding byte by shifting the
// first one 8 to the left and than applying a bitwise or-operator to append
// the second one.
if (byteOrder === ByteOrder.LittleEndian) return (second << 8) | first;
if (byteOrder === ByteOrder.BigEndian) return (first << 8) | second;
throw new Error(`Invalid byteOrder: ${byteOrder}`);
};
/**
* Returns whether the given array contains a byte order mark for the
* UTF-16BE or UTF-16LE encoding. If it has neither, BigEndian is assumed.
*
* Reference: https://en.wikipedia.org/wiki/Byte_order_mark#UTF-16
*
* @param bytes The byte array to be evaluated.
*/
// prettier-ignore
const readBOM = (bytes: Uint8Array): ByteOrder => (
hasUtf16BigEndianBOM(bytes) ? ByteOrder.BigEndian
: hasUtf16LittleEndianBOM(bytes) ? ByteOrder.LittleEndian
: ByteOrder.BigEndian
);
const hasUtf16BigEndianBOM = (bytes: Uint8Array) =>
bytes[0] === 0xfe && bytes[1] === 0xff;
const hasUtf16LittleEndianBOM = (bytes: Uint8Array) =>
bytes[0] === 0xff && bytes[1] === 0xfe;
export const hasUtf16BOM = (bytes: Uint8Array) =>
hasUtf16BigEndianBOM(bytes) || hasUtf16LittleEndianBOM(bytes);
|