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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523
|
package minify
import (
"bytes"
"encoding/base64"
"github.com/tdewolff/parse/v2"
"github.com/tdewolff/parse/v2/strconv"
)
var (
textMimeBytes = []byte("text/plain")
charsetASCIIBytes = []byte("charset=us-ascii")
dataBytes = []byte("data:")
base64Bytes = []byte(";base64")
)
// Epsilon is the closest number to zero that is not considered to be zero.
var Epsilon = 0.00001
// Mediatype minifies a given mediatype by removing all whitespace and lowercasing all parts except strings (which may be case sensitive).
func Mediatype(b []byte) []byte {
j := 0
inString := false
start, lastString := 0, 0
for i, c := range b {
if !inString && parse.IsWhitespace(c) {
if start != 0 {
j += copy(b[j:], b[start:i])
} else {
j += i
}
start = i + 1
} else if c == '"' {
inString = !inString
if inString {
if i-lastString < 1024 { // ToLower may otherwise slow down minification greatly
parse.ToLower(b[lastString:i])
}
} else {
lastString = j + (i + 1 - start)
}
}
}
if start != 0 {
j += copy(b[j:], b[start:])
parse.ToLower(b[lastString:j])
return b[:j]
}
parse.ToLower(b[lastString:])
return b
}
// DataURI minifies a data URI and calls a minifier by the specified mediatype. Specifications: https://www.ietf.org/rfc/rfc2397.txt.
func DataURI(m *M, dataURI []byte) []byte {
origData := parse.Copy(dataURI)
mediatype, data, err := parse.DataURI(dataURI)
if err != nil {
return dataURI
}
data, _ = m.Bytes(string(mediatype), data)
base64Len := len(";base64") + base64.StdEncoding.EncodedLen(len(data))
asciiLen := len(data)
for _, c := range data {
if parse.DataURIEncodingTable[c] {
asciiLen += 2
}
if asciiLen > base64Len {
break
}
}
if len(origData) < base64Len && len(origData) < asciiLen {
return origData
}
if base64Len < asciiLen {
encoded := make([]byte, base64Len-len(";base64"))
base64.StdEncoding.Encode(encoded, data)
data = encoded
mediatype = append(mediatype, base64Bytes...)
} else {
data = parse.EncodeURL(data, parse.DataURIEncodingTable)
}
if len("text/plain") <= len(mediatype) && parse.EqualFold(mediatype[:len("text/plain")], textMimeBytes) {
mediatype = mediatype[len("text/plain"):]
}
for i := 0; i+len(";charset=us-ascii") <= len(mediatype); i++ {
// must start with semicolon and be followed by end of mediatype or semicolon
if mediatype[i] == ';' && parse.EqualFold(mediatype[i+1:i+len(";charset=us-ascii")], charsetASCIIBytes) && (i+len(";charset=us-ascii") >= len(mediatype) || mediatype[i+len(";charset=us-ascii")] == ';') {
mediatype = append(mediatype[:i], mediatype[i+len(";charset=us-ascii"):]...)
break
}
}
return append(append(append(dataBytes, mediatype...), ','), data...)
}
// MaxInt is the maximum value of int.
const MaxInt = int(^uint(0) >> 1)
// MinInt is the minimum value of int.
const MinInt = -MaxInt - 1
// Decimal minifies a given byte slice containing a decimal and removes superfluous characters. It differs from Number in that it does not parse exponents.
// It does not parse or output exponents. prec is the number of significant digits. When prec is zero it will keep all digits. Only digits after the dot can be removed to reach the number of significant digits. Very large number may thus have more significant digits.
func Decimal(num []byte, prec int) []byte {
if len(num) <= 1 {
return num
}
// omit first + and register mantissa start and end, whether it's negative and the exponent
neg := false
start := 0
dot := -1
end := len(num)
if 0 < end && (num[0] == '+' || num[0] == '-') {
if num[0] == '-' {
neg = true
}
start++
}
for i, c := range num[start:] {
if c == '.' {
dot = start + i
break
}
}
if dot == -1 {
dot = end
}
// trim leading zeros but leave at least one digit
for start < end-1 && num[start] == '0' {
start++
}
// trim trailing zeros
i := end - 1
for ; dot < i; i-- {
if num[i] != '0' {
end = i + 1
break
}
}
if i == dot {
end = dot
if start == end {
num[start] = '0'
return num[start : start+1]
}
} else if start == end-1 && num[start] == '0' {
return num[start:end]
}
// apply precision
if 0 < prec && dot <= start+prec {
precEnd := start + prec + 1 // include dot
if dot == start { // for numbers like .012
digit := start + 1
for digit < end && num[digit] == '0' {
digit++
}
precEnd = digit + prec
}
if precEnd < end {
end = precEnd
// process either an increase from a lesser significant decimal (>= 5)
// or remove trailing zeros after the dot, or both
i := end - 1
inc := '5' <= num[end]
for ; start < i; i-- {
if i == dot {
// no-op
} else if inc && num[i] != '9' {
num[i]++
inc = false
break
} else if inc && i < dot { // end inc for integer
num[i] = '0'
} else if !inc && (i < dot || num[i] != '0') {
break
}
}
if i < dot {
end = dot
} else {
end = i + 1
}
if inc {
if dot == start && end == start+1 {
num[start] = '1'
} else if num[start] == '9' {
num[start] = '1'
num[start+1] = '0'
end++
} else {
num[start]++
}
}
}
}
if neg {
start--
num[start] = '-'
}
return num[start:end]
}
// Number minifies a given byte slice containing a number and removes superfluous characters.
func Number(num []byte, prec int) []byte {
if len(num) <= 1 {
return num
}
// omit first + and register mantissa start and end, whether it's negative and the exponent
neg := false
start := 0
dot := -1
end := len(num)
origExp := 0
if num[0] == '+' || num[0] == '-' {
if num[0] == '-' {
neg = true
}
start++
}
for i, c := range num[start:] {
if c == '.' {
dot = start + i
} else if c == 'e' || c == 'E' {
end = start + i
i += start + 1
if i < len(num) && num[i] == '+' {
i++
}
if tmpOrigExp, n := strconv.ParseInt(num[i:]); 0 < n && int64(MinInt) <= tmpOrigExp && tmpOrigExp <= int64(MaxInt) {
// range checks for when int is 32 bit
origExp = int(tmpOrigExp)
} else {
return num
}
break
}
}
if dot == -1 {
dot = end
}
// trim leading zeros but leave at least one digit
for start < end-1 && num[start] == '0' {
start++
}
// trim trailing zeros
i := end - 1
for ; dot < i; i-- {
if num[i] != '0' {
end = i + 1
break
}
}
if i == dot {
end = dot
if start == end {
num[start] = '0'
return num[start : start+1]
}
} else if start == end-1 && num[start] == '0' {
return num[start:end]
}
// apply precision
if 0 < prec { //&& (dot <= start+prec || start+prec+1 < dot || 0 < origExp) { // don't minify 9 to 10, but do 999 to 1e3 and 99e1 to 1e3
precEnd := start + prec
if dot == start { // for numbers like .012
digit := start + 1
for digit < end && num[digit] == '0' {
digit++
}
precEnd = digit + prec
} else if dot < precEnd { // for numbers where precision will include the dot
precEnd++
}
if precEnd < end && (dot < end || 1 < dot-precEnd+origExp) { // do not minify 9=>10 or 99=>100 or 9e1=>1e2 (but 90), but 999=>1e3 and 99e1=>1e3
end = precEnd
inc := '5' <= num[end]
if dot == end {
inc = end+1 < len(num) && '5' <= num[end+1]
}
if precEnd < dot {
origExp += dot - precEnd
dot = precEnd
}
// process either an increase from a lesser significant decimal (>= 5)
// and remove trailing zeros
i := end - 1
for ; start < i; i-- {
if i == dot {
// no-op
} else if inc && num[i] != '9' {
num[i]++
inc = false
break
} else if !inc && num[i] != '0' {
break
}
}
end = i + 1
if end < dot {
origExp += dot - end
dot = end
}
if inc { // single digit left
if dot == start {
num[start] = '1'
dot = start + 1
} else if num[start] == '9' {
num[start] = '1'
origExp++
} else {
num[start]++
}
}
}
}
// n is the number of significant digits
// normExp would be the exponent if it were normalised (0.1 <= f < 1)
n := 0
normExp := 0
if start == end {
return num // no number before exponent
} else if dot == start {
for i = dot + 1; i < end; i++ {
if num[i] != '0' {
n = end - i
normExp = dot - i + 1
break
}
}
} else if dot == end {
normExp = end - start
for i = end - 1; start <= i; i-- {
if num[i] != '0' {
n = i + 1 - start
end = i + 1
break
}
}
} else {
n = end - start - 1
normExp = dot - start
}
if origExp < 0 && (normExp < MinInt-origExp || normExp-n < MinInt-origExp) || 0 < origExp && (MaxInt-origExp < normExp || MaxInt-origExp < normExp-n) {
return num // exponent overflow
}
normExp += origExp
// intExp would be the exponent if it were an integer
intExp := normExp - n
lenIntExp := strconv.LenInt(int64(intExp))
lenNormExp := strconv.LenInt(int64(normExp))
// there are three cases to consider when printing the number
// case 1: without decimals and with a positive exponent (large numbers: 5e4)
// case 2: with decimals and with a negative exponent (small numbers with many digits: .123456e-4)
// case 3: with decimals and without an exponent (around zero: 5.6)
// case 4: without decimals and with a negative exponent (small numbers: 123456e-9)
if n <= normExp {
// case 1: print number with positive exponent
if dot < end {
// remove dot, either from the front or copy the smallest part
if dot == start {
start = end - n
} else if dot-start < end-dot-1 {
copy(num[start+1:], num[start:dot])
start++
} else {
copy(num[dot:], num[dot+1:end])
end--
}
}
if n+3 <= normExp {
num[end] = 'e'
end++
for i := end + lenIntExp - 1; end <= i; i-- {
num[i] = byte(intExp%10) + '0'
intExp /= 10
}
end += lenIntExp
} else if n+2 == normExp {
num[end] = '0'
num[end+1] = '0'
end += 2
} else if n+1 == normExp {
num[end] = '0'
end++
}
} else if normExp < -3 && lenNormExp < lenIntExp && dot < end {
// case 2: print normalized number (0.1 <= f < 1)
zeroes := -normExp + origExp
if 0 < zeroes {
copy(num[start+1:], num[start+1+zeroes:end])
end -= zeroes
} else if zeroes < 0 {
copy(num[start+1:], num[start:dot])
num[start] = '.'
} else {
return num
}
num[end] = 'e'
num[end+1] = '-'
end += 2
for i := end + lenNormExp - 2; end <= i; i-- {
num[i] = -byte(normExp%10) + '0'
normExp /= 10
}
end += lenNormExp - 1
} else if -lenIntExp <= normExp {
// case 3: print number without exponent
zeroes := -normExp
if 0 < zeroes {
// place dot at the front, adding zeroes after the dot
if newDot := end - n - zeroes - 1; newDot != dot {
if d := start - newDot; 0 < d {
if dot < end {
// copy original digits after the dot towards the end
copy(num[dot+1+d:], num[dot+1:end])
if start < dot {
// copy original digits before the dot towards the end
copy(num[start+d+1:], num[start:dot])
}
} else if start < dot {
// copy original digits before the dot towards the end
copy(num[start+d:], num[start:dot])
}
newDot = start
end += d
} else {
start += -d
}
num[newDot] = '.'
for i := 0; i < zeroes; i++ {
num[newDot+1+i] = '0'
}
}
} else {
// place dot in the middle of the number
if end <= dot {
// when input has no dot in it
dot = end
end++
} else if dot == start {
// when there are zeroes after the dot
dot = end - n - 1
start = dot
}
// move digits between dot and newDot towards the end
newDot := start + normExp
if dot < newDot {
copy(num[dot:], num[dot+1:newDot+1])
} else if newDot < dot {
copy(num[newDot+1:], num[newDot:dot])
}
num[newDot] = '.'
}
} else {
// case 4: print number with negative exponent
// find new end, considering moving numbers to the front, removing the dot and increasing the length of the exponent
newEnd := end
if dot == start {
newEnd = dot + n
} else {
newEnd--
}
newEnd += 1 + lenIntExp
exp := intExp
lenExp := lenIntExp
if newEnd < len(num) {
// it saves space to convert the decimal to an integer and decrease the exponent
if dot < end {
if dot == start {
copy(num[start:], num[end-n:end])
end = start + n
} else {
copy(num[dot:], num[dot+1:end])
end--
}
}
} else {
// it does not save space and will panic, so we revert to the original representation
exp = origExp
lenExp = strconv.LenInt(int64(origExp))
}
num[end] = 'e'
num[end+1] = '-'
end += 2
for i := end + lenExp - 2; end <= i; i-- {
num[i] = -byte(exp%10) + '0'
exp /= 10
}
end += lenExp - 1
}
if neg {
start--
num[start] = '-'
}
return num[start:end]
}
func UpdateErrorPosition(err error, input *parse.Input, offset int) error {
if perr, ok := err.(*parse.Error); ok {
r := bytes.NewBuffer(input.Bytes())
line, column, _ := parse.Position(r, offset)
perr.Line += line - 1
perr.Column += column - 1
return perr
}
return err
}
|