| 1 | // Copyright (c) 2019-2024 Alexander Medvednikov. All rights reserved. |
| 2 | // Use of this source code is governed by an MIT license |
| 3 | // that can be found in the LICENSE file. |
| 4 | module builtin |
| 5 | |
| 6 | import strconv |
| 7 | import strings |
| 8 | |
| 9 | /* |
| 10 | Note: A V string should be/is immutable from the point of view of |
| 11 | V user programs after it is first created. A V string is |
| 12 | also slightly larger than the equivalent C string because |
| 13 | the V string also has an integer length attached. |
| 14 | |
| 15 | This tradeoff is made, since V strings are created just *once*, |
| 16 | but potentially used *many times* over their lifetime. |
| 17 | |
| 18 | The V string implementation uses a struct, that has a .str field, |
| 19 | which points to a C style 0 terminated memory block. Although not |
| 20 | strictly necessary from the V point of view, that additional 0 |
| 21 | is *very useful for C interoperability*. |
| 22 | |
| 23 | The V string implementation also has an integer .len field, |
| 24 | containing the length of the .str field, excluding the |
| 25 | terminating 0 (just like the C's strlen(s) would do). |
| 26 | |
| 27 | The 0 ending of .str, and the .len field, mean that in practice: |
| 28 | a) a V string s can be used very easily, wherever a |
| 29 | C string is needed, just by passing s.str, |
| 30 | without a need for further conversion/copying. |
| 31 | |
| 32 | b) where strlen(s) is needed, you can just pass s.len, |
| 33 | without having to constantly recompute the length of s |
| 34 | *over and over again* like some C programs do. This is because |
| 35 | V strings are immutable and so their length does not change. |
| 36 | |
| 37 | Ordinary V code *does not need* to be concerned with the |
| 38 | additional 0 in the .str field. The 0 *must* be put there by the |
| 39 | low level string creating functions inside this module. |
| 40 | |
| 41 | Failing to do this will lead to programs that work most of the |
| 42 | time, when used with pure V functions, but fail in strange ways, |
| 43 | when used with modules using C functions (for example os and so on). |
| 44 | */ |
| 45 | pub struct string { |
| 46 | pub: |
| 47 | str &u8 = 0 // points to a C style 0 terminated string of bytes. |
| 48 | len int // the length of the .str field, excluding the ending 0 byte. It is always equal to strlen(.str). |
| 49 | mut: |
| 50 | is_lit int |
| 51 | // NB string.is_lit is an enumeration of the following: |
| 52 | // .is_lit == 0 => a fresh string, should be freed by autofree |
| 53 | // .is_lit == 1 => a literal string from .rodata, should NOT be freed |
| 54 | // .is_lit == -98761234 => already freed string, protects against double frees. |
| 55 | // ---------> ^^^^^^^^^ calling free on these is a bug. |
| 56 | // Any other value means that the string has been corrupted. |
| 57 | } |
| 58 | |
| 59 | // runes returns an array of all the utf runes in the string `s` |
| 60 | // which is useful if you want random access to them |
| 61 | @[direct_array_access] |
| 62 | pub fn (s string) runes() []rune { |
| 63 | mut runes := []rune{cap: s.len} |
| 64 | for i := 0; i < s.len; i++ { |
| 65 | char_len := utf8_char_len(unsafe { s.str[i] }) |
| 66 | if char_len > 1 { |
| 67 | end := if s.len - 1 >= i + char_len { i + char_len } else { s.len } |
| 68 | mut r := unsafe { s[i..end] } |
| 69 | runes << r.utf32_code() |
| 70 | i += char_len - 1 |
| 71 | } else { |
| 72 | runes << unsafe { s.str[i] } |
| 73 | } |
| 74 | } |
| 75 | return runes |
| 76 | } |
| 77 | |
| 78 | // graphemes returns the string split into Unicode grapheme clusters. |
| 79 | pub fn (s string) graphemes() []string { |
| 80 | return string_graphemes_impl(s) |
| 81 | } |
| 82 | |
| 83 | // cstring_to_vstring creates a new V string copy of the C style string, |
| 84 | // pointed by `s`. This function is most likely what you want to use when |
| 85 | // working with C style pointers to 0 terminated strings (i.e. `char*`). |
| 86 | // It is recommended to use it, unless you *do* understand the implications of |
| 87 | // tos/tos2/tos3/tos4/tos5 in terms of memory management and interactions with |
| 88 | // -autofree and `@[manualfree]`. |
| 89 | // It will panic, if the pointer `s` is 0. |
| 90 | @[unsafe] |
| 91 | pub fn cstring_to_vstring(const_s &char) string { |
| 92 | s := unsafe { tos2(byteptr(const_s)) } |
| 93 | return s.clone() |
| 94 | } |
| 95 | |
| 96 | // tos_clone creates a new V string copy of the C style string, pointed by `s`. |
| 97 | // See also cstring_to_vstring (it is the same as it, the only difference is, |
| 98 | // that tos_clone expects `&u8`, while cstring_to_vstring expects &char). |
| 99 | // It will panic, if the pointer `s` is 0. |
| 100 | @[unsafe] |
| 101 | pub fn tos_clone(const_s &u8) string { |
| 102 | s := unsafe { tos2(&u8(const_s)) } |
| 103 | return s.clone() |
| 104 | } |
| 105 | |
| 106 | // tos creates a V string, given a C style pointer to a 0 terminated block. |
| 107 | // Note: the memory block pointed by s is *reused, not copied*! |
| 108 | // It will panic, when the pointer `s` is 0. |
| 109 | // See also `tos_clone`. |
| 110 | @[unsafe] |
| 111 | pub fn tos(s &u8, len int) string { |
| 112 | if s == 0 { |
| 113 | panic('tos(): nil string') |
| 114 | } |
| 115 | return string{ |
| 116 | str: unsafe { s } |
| 117 | len: len |
| 118 | } |
| 119 | } |
| 120 | |
| 121 | // tos2 creates a V string, given a C style pointer to a 0 terminated block. |
| 122 | // Note: the memory block pointed by s is *reused, not copied*! |
| 123 | // It will calculate the length first, thus it is more costly than `tos`. |
| 124 | // It will panic, when the pointer `s` is 0. |
| 125 | // It is the same as `tos3`, but for &u8 pointers, avoiding callsite casts. |
| 126 | // See also `tos_clone`. |
| 127 | @[unsafe] |
| 128 | pub fn tos2(s &u8) string { |
| 129 | if s == 0 { |
| 130 | panic('tos2: nil string') |
| 131 | } |
| 132 | return string{ |
| 133 | str: unsafe { s } |
| 134 | len: unsafe { vstrlen(s) } |
| 135 | } |
| 136 | } |
| 137 | |
| 138 | // tos3 creates a V string, given a C style pointer to a 0 terminated block. |
| 139 | // Note: the memory block pointed by s is *reused, not copied*! |
| 140 | // It will calculate the length first, so it is more costly than tos. |
| 141 | // It will panic, when the pointer `s` is 0. |
| 142 | // It is the same as `tos2`, but for &char pointers, avoiding callsite casts. |
| 143 | // See also `tos_clone`. |
| 144 | @[unsafe] |
| 145 | pub fn tos3(s &char) string { |
| 146 | if s == 0 { |
| 147 | panic('tos3: nil string') |
| 148 | } |
| 149 | return string{ |
| 150 | str: unsafe { &u8(s) } |
| 151 | len: unsafe { vstrlen_char(s) } |
| 152 | } |
| 153 | } |
| 154 | |
| 155 | // tos4 creates a V string, given a C style pointer to a 0 terminated block. |
| 156 | // Note: the memory block pointed by s is *reused, not copied*! |
| 157 | // It will calculate the length first, so it is more costly than tos. |
| 158 | // It returns '', when given a 0 pointer `s`, it does NOT panic. |
| 159 | // It is the same as `tos5`, but for &u8 pointers, avoiding callsite casts. |
| 160 | // See also `tos_clone`. |
| 161 | @[unsafe] |
| 162 | pub fn tos4(s &u8) string { |
| 163 | if s == 0 { |
| 164 | return '' |
| 165 | } |
| 166 | return string{ |
| 167 | str: unsafe { s } |
| 168 | len: unsafe { vstrlen(s) } |
| 169 | } |
| 170 | } |
| 171 | |
| 172 | // tos5 creates a V string, given a C style pointer to a 0 terminated block. |
| 173 | // Note: the memory block pointed by s is *reused, not copied*! |
| 174 | // It will calculate the length first, so it is more costly than tos. |
| 175 | // It returns '', when given a 0 pointer `s`, it does NOT panic. |
| 176 | // It is the same as `tos4`, but for &char pointers, avoiding callsite casts. |
| 177 | // See also `tos_clone`. |
| 178 | @[unsafe] |
| 179 | pub fn tos5(s &char) string { |
| 180 | if s == 0 { |
| 181 | return '' |
| 182 | } |
| 183 | return string{ |
| 184 | str: unsafe { &u8(s) } |
| 185 | len: unsafe { vstrlen_char(s) } |
| 186 | } |
| 187 | } |
| 188 | |
| 189 | // vstring converts a C style string to a V string. |
| 190 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 191 | // Note: instead of `&u8(arr.data).vstring()`, do use `tos_clone(&u8(arr.data))`. |
| 192 | // Strings returned from this function will be normal V strings beside that, |
| 193 | // (i.e. they would be freed by V's -autofree mechanism, when they are no longer used). |
| 194 | // See also `tos_clone`. |
| 195 | @[unsafe] |
| 196 | pub fn (bp &u8) vstring() string { |
| 197 | return string{ |
| 198 | str: unsafe { bp } |
| 199 | len: unsafe { vstrlen(bp) } |
| 200 | } |
| 201 | } |
| 202 | |
| 203 | // vstring_with_len converts a C style 0 terminated string to a V string. |
| 204 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 205 | // This method has lower overhead compared to .vstring(), since it |
| 206 | // does not need to calculate the length of the 0 terminated string. |
| 207 | // See also `tos_clone`. |
| 208 | @[unsafe] |
| 209 | pub fn (bp &u8) vstring_with_len(len int) string { |
| 210 | return string{ |
| 211 | str: unsafe { bp } |
| 212 | len: len |
| 213 | is_lit: 0 |
| 214 | } |
| 215 | } |
| 216 | |
| 217 | // vstring converts a C style string to a V string. |
| 218 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 219 | // Strings returned from this function will be normal V strings beside that, |
| 220 | // (i.e. they would be freed by V's -autofree mechanism, when they are |
| 221 | // no longer used). |
| 222 | // Note: instead of `&u8(a.data).vstring()`, use `tos_clone(&u8(a.data))`. |
| 223 | // See also `tos_clone`. |
| 224 | @[unsafe] |
| 225 | pub fn (cp &char) vstring() string { |
| 226 | return string{ |
| 227 | str: &u8(cp) |
| 228 | len: unsafe { vstrlen_char(cp) } |
| 229 | is_lit: 0 |
| 230 | } |
| 231 | } |
| 232 | |
| 233 | // vstring_with_len converts a C style 0 terminated string to a V string. |
| 234 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 235 | // This method has lower overhead compared to .vstring(), since it |
| 236 | // does not calculate the length of the 0 terminated string. |
| 237 | // See also `tos_clone`. |
| 238 | @[unsafe] |
| 239 | pub fn (cp &char) vstring_with_len(len int) string { |
| 240 | return string{ |
| 241 | str: &u8(cp) |
| 242 | len: len |
| 243 | is_lit: 0 |
| 244 | } |
| 245 | } |
| 246 | |
| 247 | // vstring_literal converts a C style string to a V string. |
| 248 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 249 | // NB2: unlike vstring, vstring_literal will mark the string |
| 250 | // as a literal, so it will not be freed by -autofree. |
| 251 | // This is suitable for readonly strings, C string literals etc, |
| 252 | // that can be read by the V program, but that should not be |
| 253 | // managed/freed by it, for example `os.args` is implemented using it. |
| 254 | // See also `tos_clone`. |
| 255 | @[unsafe] |
| 256 | pub fn (bp &u8) vstring_literal() string { |
| 257 | return string{ |
| 258 | str: unsafe { bp } |
| 259 | len: unsafe { vstrlen(bp) } |
| 260 | is_lit: 1 |
| 261 | } |
| 262 | } |
| 263 | |
| 264 | // vstring_with_len converts a C style string to a V string. |
| 265 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 266 | // This method has lower overhead compared to .vstring_literal(), since it |
| 267 | // does not need to calculate the length of the 0 terminated string. |
| 268 | // See also `tos_clone`. |
| 269 | @[unsafe] |
| 270 | pub fn (bp &u8) vstring_literal_with_len(len int) string { |
| 271 | return string{ |
| 272 | str: unsafe { bp } |
| 273 | len: len |
| 274 | is_lit: 1 |
| 275 | } |
| 276 | } |
| 277 | |
| 278 | // vstring_literal converts a C style string char* pointer to a V string. |
| 279 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 280 | // See also `byteptr.vstring_literal` for more details. |
| 281 | // See also `tos_clone`. |
| 282 | @[unsafe] |
| 283 | pub fn (cp &char) vstring_literal() string { |
| 284 | return string{ |
| 285 | str: &u8(cp) |
| 286 | len: unsafe { vstrlen_char(cp) } |
| 287 | is_lit: 1 |
| 288 | } |
| 289 | } |
| 290 | |
| 291 | // vstring_literal_with_len converts a C style string char* pointer, |
| 292 | // to a V string. |
| 293 | // Note: the memory block pointed by `bp` is *reused, not copied*! |
| 294 | // This method has lower overhead compared to .vstring_literal(), since it |
| 295 | // does not need to calculate the length of the 0 terminated string. |
| 296 | // See also `tos_clone`. |
| 297 | @[unsafe] |
| 298 | pub fn (cp &char) vstring_literal_with_len(len int) string { |
| 299 | return string{ |
| 300 | str: &u8(cp) |
| 301 | len: len |
| 302 | is_lit: 1 |
| 303 | } |
| 304 | } |
| 305 | |
| 306 | // len_utf8 returns the number of runes contained in the string `s`. |
| 307 | pub fn (s string) len_utf8() int { |
| 308 | mut l := 0 |
| 309 | mut i := 0 |
| 310 | for i < s.len { |
| 311 | l++ |
| 312 | i += int(((u32(0xe5000000) >> ((unsafe { s.str[i] } >> 3) & 0x1e)) & 3) + 1) |
| 313 | } |
| 314 | return l |
| 315 | } |
| 316 | |
| 317 | // is_pure_ascii returns whether the string contains only ASCII characters. |
| 318 | // Note that UTF8 encodes such characters in just 1 byte: |
| 319 | // 1 byte: 0xxxxxxx |
| 320 | // 2 bytes: 110xxxxx 10xxxxxx |
| 321 | // 3 bytes: 1110xxxx 10xxxxxx 10xxxxxx |
| 322 | // 4 bytes: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx |
| 323 | @[direct_array_access] |
| 324 | pub fn (s string) is_pure_ascii() bool { |
| 325 | for i in 0 .. s.len { |
| 326 | if s[i] >= 0x80 { |
| 327 | return false |
| 328 | } |
| 329 | } |
| 330 | return true |
| 331 | } |
| 332 | |
| 333 | // clone_static returns an independent copy of a given array. |
| 334 | // It should be used only in -autofree generated code. |
| 335 | @[inline] |
| 336 | fn (a string) clone_static() string { |
| 337 | return a.clone() |
| 338 | } |
| 339 | |
| 340 | // option_clone_static returns an independent copy of a given array when lhs is an option type. |
| 341 | // It should be used only in -autofree generated code. |
| 342 | @[inline] |
| 343 | fn (a string) option_clone_static() ?string { |
| 344 | return ?string(a.clone()) |
| 345 | } |
| 346 | |
| 347 | // clone returns a copy of the V string `a`. |
| 348 | pub fn (a string) clone() string { |
| 349 | if a.len <= 0 { |
| 350 | return '' |
| 351 | } |
| 352 | mut b := string{ |
| 353 | str: unsafe { malloc_noscan(a.len + 1) } |
| 354 | len: a.len |
| 355 | } |
| 356 | unsafe { |
| 357 | vmemcpy(b.str, a.str, a.len) |
| 358 | b.str[a.len] = 0 |
| 359 | } |
| 360 | return b |
| 361 | } |
| 362 | |
| 363 | // replace_once replaces the first occurrence of `rep` with the string passed in `with`. |
| 364 | pub fn (s string) replace_once(rep string, with string) string { |
| 365 | idx := s.index_(rep) |
| 366 | if idx == -1 { |
| 367 | return s.clone() |
| 368 | } |
| 369 | // return s.substr(0, idx) + with + s.substr(idx + rep.len, s.len) |
| 370 | // |
| 371 | // Avoid an extra allocation here by using substr_unsafe |
| 372 | // string_plus copies from both strings via vmemcpy, so it's safe. |
| 373 | // |
| 374 | // return s.substr_unsafe(0, idx) + with + s.substr_unsafe(idx + rep.len, s.len) |
| 375 | return s.substr_unsafe(0, idx).plus_two(with, s.substr_unsafe(idx + rep.len, s.len)) |
| 376 | } |
| 377 | |
| 378 | const replace_stack_buffer_size = 10 |
| 379 | // replace replaces all occurrences of `rep` with the string passed in `with`. |
| 380 | @[direct_array_access; manualfree] |
| 381 | pub fn (s string) replace(rep string, with string) string { |
| 382 | if s.len == 0 || rep.len == 0 || rep.len > s.len { |
| 383 | return s.clone() |
| 384 | } |
| 385 | if !s.contains(rep) { |
| 386 | return s.clone() |
| 387 | } |
| 388 | mut pidxs_len := 0 |
| 389 | pidxs_cap := s.len / rep.len |
| 390 | mut stack_idxs := [replace_stack_buffer_size]int{} |
| 391 | mut pidxs := unsafe { &stack_idxs[0] } |
| 392 | if pidxs_cap > replace_stack_buffer_size { |
| 393 | pidxs = unsafe { &int(malloc(int(sizeof(int)) * pidxs_cap)) } |
| 394 | } |
| 395 | defer { |
| 396 | if pidxs_cap > replace_stack_buffer_size { |
| 397 | unsafe { free(pidxs) } |
| 398 | } |
| 399 | } |
| 400 | mut idx := 0 |
| 401 | for { |
| 402 | idx = s.index_after_(rep, idx) |
| 403 | if idx == -1 { |
| 404 | break |
| 405 | } |
| 406 | unsafe { |
| 407 | pidxs[pidxs_len] = idx |
| 408 | pidxs_len++ |
| 409 | } |
| 410 | idx += rep.len |
| 411 | } |
| 412 | // Dont change the string if there's nothing to replace |
| 413 | if pidxs_len == 0 { |
| 414 | return s.clone() |
| 415 | } |
| 416 | // Now we know the number of replacements we need to do and we can calc the len of the new string |
| 417 | new_len := s.len + pidxs_len * (with.len - rep.len) |
| 418 | mut b := unsafe { malloc_noscan(new_len + 1) } // add space for the null byte at the end |
| 419 | // Fill the new string |
| 420 | mut b_i := 0 |
| 421 | mut s_idx := 0 |
| 422 | for j in 0 .. pidxs_len { |
| 423 | rep_pos := unsafe { pidxs[j] } |
| 424 | // copy everything up to piece being replaced |
| 425 | before_len := rep_pos - s_idx |
| 426 | unsafe { vmemcpy(&b[b_i], &s.str[s_idx], before_len) } |
| 427 | b_i += before_len |
| 428 | s_idx = rep_pos + rep.len // move string index past replacement |
| 429 | // copy replacement piece |
| 430 | unsafe { vmemcpy(&b[b_i], &with.str[0], with.len) } |
| 431 | b_i += with.len |
| 432 | } |
| 433 | if s_idx < s.len { |
| 434 | // if any original after last replacement, copy it |
| 435 | unsafe { vmemcpy(&b[b_i], &s.str[s_idx], s.len - s_idx) } |
| 436 | } |
| 437 | unsafe { |
| 438 | b[new_len] = 0 |
| 439 | return tos(b, new_len) |
| 440 | } |
| 441 | } |
| 442 | |
| 443 | struct RepIndex { |
| 444 | idx int |
| 445 | val_idx int |
| 446 | } |
| 447 | |
| 448 | // replace_each replaces all occurrences of the string pairs given in `vals`. |
| 449 | // Example: assert 'ABCD'.replace_each(['B','C/','C','D','D','C']) == 'AC/DC' |
| 450 | @[direct_array_access] |
| 451 | pub fn (s string) replace_each(vals []string) string { |
| 452 | if s.len == 0 || vals.len == 0 { |
| 453 | return s.clone() |
| 454 | } |
| 455 | if vals.len % 2 != 0 { |
| 456 | eprintln('string.replace_each(): odd number of strings') |
| 457 | return s.clone() |
| 458 | } |
| 459 | // `rep` - string to replace |
| 460 | // `with` - string to replace with |
| 461 | // Remember positions of all rep strings, and calculate the length |
| 462 | // of the new string to do just one allocation. |
| 463 | mut new_len := s.len |
| 464 | mut idxs := []RepIndex{cap: 6} |
| 465 | defer { unsafe { idxs.free() } } |
| 466 | mut idx := 0 |
| 467 | s_ := s.clone() |
| 468 | for rep_i := 0; rep_i < vals.len; rep_i += 2 { |
| 469 | // vals: ['rep1, 'with1', 'rep2', 'with2'] |
| 470 | rep := vals[rep_i] |
| 471 | with := vals[rep_i + 1] |
| 472 | |
| 473 | for { |
| 474 | idx = s_.index_after_(rep, idx) |
| 475 | if idx == -1 { |
| 476 | break |
| 477 | } |
| 478 | // The string already found is set to `/del`, to avoid duplicate searches. |
| 479 | for i in 0 .. rep.len { |
| 480 | unsafe { |
| 481 | s_.str[idx + i] = 0 |
| 482 | } |
| 483 | } |
| 484 | // We need to remember both the position in the string, |
| 485 | // and which rep/with pair it refers to. |
| 486 | |
| 487 | idxs << RepIndex{ |
| 488 | idx: idx |
| 489 | val_idx: rep_i |
| 490 | } |
| 491 | |
| 492 | idx += rep.len |
| 493 | new_len += with.len - rep.len |
| 494 | } |
| 495 | } |
| 496 | |
| 497 | // Dont change the string if there's nothing to replace |
| 498 | if idxs.len == 0 { |
| 499 | return s.clone() |
| 500 | } |
| 501 | idxs.sort(a.idx < b.idx) |
| 502 | mut buf := unsafe { malloc_noscan(new_len + 1) } // add space for 0 terminator |
| 503 | // Fill the new string |
| 504 | mut idx_pos := 0 |
| 505 | mut cur_idx := idxs[idx_pos] |
| 506 | mut buf_i := 0 |
| 507 | for i := 0; i < s.len; i++ { |
| 508 | if i == cur_idx.idx { |
| 509 | // Reached the location of rep, replace it with "with" |
| 510 | rep := vals[cur_idx.val_idx] |
| 511 | with := vals[cur_idx.val_idx + 1] |
| 512 | for j in 0 .. with.len { |
| 513 | unsafe { |
| 514 | buf[buf_i] = with[j] |
| 515 | } |
| 516 | buf_i++ |
| 517 | } |
| 518 | // Skip the length of rep, since we just replaced it with "with" |
| 519 | i += rep.len - 1 |
| 520 | // Go to the next index |
| 521 | idx_pos++ |
| 522 | if idx_pos < idxs.len { |
| 523 | cur_idx = idxs[idx_pos] |
| 524 | } |
| 525 | } else { |
| 526 | // Rep doesnt start here, just copy |
| 527 | unsafe { |
| 528 | buf[buf_i] = s.str[i] |
| 529 | } |
| 530 | buf_i++ |
| 531 | } |
| 532 | } |
| 533 | unsafe { |
| 534 | buf[new_len] = 0 |
| 535 | return tos(buf, new_len) |
| 536 | } |
| 537 | } |
| 538 | |
| 539 | // format replaces positional placeholders like `{0}` and `{1}` in `s` |
| 540 | // with the corresponding values from `args`. |
| 541 | // Use `{{` and `}}` to output literal braces. |
| 542 | @[direct_array_access] |
| 543 | pub fn (s string) format(args ...string) string { |
| 544 | if s.len == 0 { |
| 545 | return '' |
| 546 | } |
| 547 | mut out := strings.new_builder(s.len) |
| 548 | mut i := 0 |
| 549 | for i < s.len { |
| 550 | ch := s[i] |
| 551 | if ch == `{` { |
| 552 | if i + 1 < s.len && s[i + 1] == `{` { |
| 553 | out.write_byte(`{`) |
| 554 | i += 2 |
| 555 | continue |
| 556 | } |
| 557 | mut j := i + 1 |
| 558 | if j >= s.len || !s[j].is_digit() { |
| 559 | out.write_byte(ch) |
| 560 | i++ |
| 561 | continue |
| 562 | } |
| 563 | mut idx := 0 |
| 564 | mut overflowed := false |
| 565 | for j < s.len && s[j].is_digit() { |
| 566 | digit := int(s[j] - `0`) |
| 567 | if idx > (max_int - digit) / 10 { |
| 568 | overflowed = true |
| 569 | break |
| 570 | } |
| 571 | idx = idx * 10 + digit |
| 572 | j++ |
| 573 | } |
| 574 | if !overflowed && j < s.len && s[j] == `}` { |
| 575 | if idx < args.len { |
| 576 | out.write_string(args[idx]) |
| 577 | } else { |
| 578 | out.write_string(s[i..j + 1]) |
| 579 | } |
| 580 | i = j + 1 |
| 581 | continue |
| 582 | } |
| 583 | out.write_byte(ch) |
| 584 | i++ |
| 585 | continue |
| 586 | } |
| 587 | if ch == `}` && i + 1 < s.len && s[i + 1] == `}` { |
| 588 | out.write_byte(`}`) |
| 589 | i += 2 |
| 590 | continue |
| 591 | } |
| 592 | out.write_byte(ch) |
| 593 | i++ |
| 594 | } |
| 595 | return out.str() |
| 596 | } |
| 597 | |
| 598 | // replace_char replaces all occurrences of the character `rep`, with `repeat` x the character passed in `with`. |
| 599 | // Example: assert '\tHello!'.replace_char(`\t`,` `,8) == ' Hello!' |
| 600 | @[direct_array_access] |
| 601 | pub fn (s string) replace_char(rep u8, with u8, repeat int) string { |
| 602 | $if !no_bounds_checking { |
| 603 | if repeat <= 0 { |
| 604 | panic('string.replace_char(): tab length too short') |
| 605 | } |
| 606 | } |
| 607 | if s.len == 0 { |
| 608 | return s.clone() |
| 609 | } |
| 610 | // TODO: Allocating ints is expensive. Should be a stack array |
| 611 | // - string.replace() |
| 612 | mut idxs := []int{cap: s.len >> 2} |
| 613 | defer { unsafe { idxs.free() } } |
| 614 | // No need to do a contains(), it already traverses the entire string |
| 615 | for i, ch in s { |
| 616 | if ch == rep { // Found char? Mark its location |
| 617 | idxs << i |
| 618 | } |
| 619 | } |
| 620 | if idxs.len == 0 { |
| 621 | return s.clone() |
| 622 | } |
| 623 | // Now we know the number of replacements we need to do and we can calc the len of the new string |
| 624 | new_len := s.len + idxs.len * (repeat - 1) |
| 625 | mut b := unsafe { malloc_noscan(new_len + 1) } // add space for the null byte at the end |
| 626 | // Fill the new string |
| 627 | mut b_i := 0 |
| 628 | mut s_idx := 0 |
| 629 | for rep_pos in idxs { |
| 630 | for i in s_idx .. rep_pos { // copy everything up to piece being replaced |
| 631 | unsafe { |
| 632 | b[b_i] = s[i] |
| 633 | } |
| 634 | b_i++ |
| 635 | } |
| 636 | s_idx = rep_pos + 1 // move string index past replacement |
| 637 | for _ in 0 .. repeat { // copy replacement piece |
| 638 | unsafe { |
| 639 | b[b_i] = with |
| 640 | } |
| 641 | b_i++ |
| 642 | } |
| 643 | } |
| 644 | if s_idx < s.len { // if any original after last replacement, copy it |
| 645 | for i in s_idx .. s.len { |
| 646 | unsafe { |
| 647 | b[b_i] = s[i] |
| 648 | } |
| 649 | b_i++ |
| 650 | } |
| 651 | } |
| 652 | unsafe { |
| 653 | b[new_len] = 0 |
| 654 | return tos(b, new_len) |
| 655 | } |
| 656 | } |
| 657 | |
| 658 | // normalize_tabs replaces all tab characters with `tab_len` amount of spaces. |
| 659 | // Example: assert '\t\tpop rax\t; pop rax'.normalize_tabs(2) == ' pop rax ; pop rax' |
| 660 | @[inline] |
| 661 | pub fn (s string) normalize_tabs(tab_len int) string { |
| 662 | return s.replace_char(`\t`, ` `, tab_len) |
| 663 | } |
| 664 | |
| 665 | // expand_tabs replaces tab characters (\t) in the input string with spaces to achieve proper column alignment . |
| 666 | // Example: assert 'AB\tHello!'.expand_tabs(4) == 'AB Hello!' |
| 667 | pub fn (s string) expand_tabs(tab_len int) string { |
| 668 | if tab_len <= 0 { |
| 669 | return s.clone() // Handle invalid tab length |
| 670 | } |
| 671 | mut output := strings.new_builder(s.len) |
| 672 | mut column := 0 |
| 673 | for r in s.runes_iterator() { |
| 674 | match r { |
| 675 | `\t` { |
| 676 | spaces := tab_len - (column % tab_len) |
| 677 | output.write_string(' '.repeat(spaces)) |
| 678 | column += spaces |
| 679 | } |
| 680 | `\n`, `\r` { |
| 681 | output.write_rune(r) |
| 682 | column = 0 // Reset on any line break |
| 683 | } |
| 684 | else { |
| 685 | output.write_rune(r) |
| 686 | column++ // Valid for most chars; consider Unicode wide chars |
| 687 | } |
| 688 | } |
| 689 | } |
| 690 | return output.str() |
| 691 | } |
| 692 | |
| 693 | // bool returns `true` if the string equals the word "true" it will return `false` otherwise. |
| 694 | @[inline] |
| 695 | pub fn (s string) bool() bool { |
| 696 | return s == 'true' || s == 't' // TODO: t for pg, remove |
| 697 | } |
| 698 | |
| 699 | // i8 returns the value of the string as i8 `'1'.i8() == i8(1)`. |
| 700 | @[inline] |
| 701 | pub fn (s string) i8() i8 { |
| 702 | return i8(strconv.common_parse_int(s, 0, 8, false, false) or { 0 }) |
| 703 | } |
| 704 | |
| 705 | // i16 returns the value of the string as i16 `'1'.i16() == i16(1)`. |
| 706 | @[inline] |
| 707 | pub fn (s string) i16() i16 { |
| 708 | return i16(strconv.common_parse_int(s, 0, 16, false, false) or { 0 }) |
| 709 | } |
| 710 | |
| 711 | // i32 returns the value of the string as i32 `'1'.i32() == i32(1)`. |
| 712 | @[inline] |
| 713 | pub fn (s string) i32() i32 { |
| 714 | return i32(strconv.common_parse_int(s, 0, 32, false, false) or { 0 }) |
| 715 | } |
| 716 | |
| 717 | // int returns the value of the string as an integer `'1'.int() == 1`. |
| 718 | @[inline] |
| 719 | pub fn (s string) int() int { |
| 720 | return int(strconv.common_parse_int(s, 0, 32, false, false) or { 0 }) |
| 721 | } |
| 722 | |
| 723 | // i64 returns the value of the string as i64 `'1'.i64() == i64(1)`. |
| 724 | @[inline] |
| 725 | pub fn (s string) i64() i64 { |
| 726 | return strconv.common_parse_int(s, 0, 64, false, false) or { 0 } |
| 727 | } |
| 728 | |
| 729 | // f32 returns the value of the string as f32 `'1.0'.f32() == f32(1)`. |
| 730 | @[inline] |
| 731 | pub fn (s string) f32() f32 { |
| 732 | return f32(strconv.atof64(s, allow_extra_chars: true) or { 0 }) |
| 733 | } |
| 734 | |
| 735 | // f64 returns the value of the string as f64 `'1.0'.f64() == f64(1)`. |
| 736 | @[inline] |
| 737 | pub fn (s string) f64() f64 { |
| 738 | return strconv.atof64(s, allow_extra_chars: true) or { 0 } |
| 739 | } |
| 740 | |
| 741 | // u8_array returns the value of the hex/bin string as u8 array. |
| 742 | // hex string example: `'0x11223344ee'.u8_array() == [u8(0x11),0x22,0x33,0x44,0xee]`. |
| 743 | // bin string example: `'0b1101_1101'.u8_array() == [u8(0xdd)]`. |
| 744 | // underscore in the string will be stripped. |
| 745 | pub fn (s string) u8_array() []u8 { |
| 746 | // strip underscore in the string |
| 747 | mut tmps := s.replace('_', '') |
| 748 | if tmps.len == 0 { |
| 749 | return []u8{} |
| 750 | } |
| 751 | tmps = tmps.to_lower_ascii() |
| 752 | if tmps.starts_with('0x') { |
| 753 | tmps = tmps[2..] |
| 754 | if tmps.len == 0 { |
| 755 | return []u8{} |
| 756 | } |
| 757 | // make sure every digit is valid hex digit |
| 758 | if !tmps.contains_only('0123456789abcdef') { |
| 759 | return []u8{} |
| 760 | } |
| 761 | // make sure tmps has even hex digits |
| 762 | if tmps.len % 2 == 1 { |
| 763 | tmps = '0' + tmps |
| 764 | } |
| 765 | |
| 766 | mut ret := []u8{len: tmps.len / 2} |
| 767 | for i in 0 .. ret.len { |
| 768 | ret[i] = u8(tmps[2 * i..2 * i + 2].parse_uint(16, 8) or { 0 }) |
| 769 | } |
| 770 | return ret |
| 771 | } else if tmps.starts_with('0b') { |
| 772 | tmps = tmps[2..] |
| 773 | if tmps.len == 0 { |
| 774 | return []u8{} |
| 775 | } |
| 776 | // make sure every digit is valid binary digit |
| 777 | if !tmps.contains_only('01') { |
| 778 | return []u8{} |
| 779 | } |
| 780 | // make sure tmps has multiple of 8 binary digits |
| 781 | if tmps.len % 8 != 0 { |
| 782 | tmps = '0'.repeat(8 - tmps.len % 8) + tmps |
| 783 | } |
| 784 | |
| 785 | mut ret := []u8{len: tmps.len / 8} |
| 786 | for i in 0 .. ret.len { |
| 787 | ret[i] = u8(tmps[8 * i..8 * i + 8].parse_uint(2, 8) or { 0 }) |
| 788 | } |
| 789 | return ret |
| 790 | } |
| 791 | return []u8{} |
| 792 | } |
| 793 | |
| 794 | // u8 returns the value of the string as u8 `'1'.u8() == u8(1)`. |
| 795 | @[inline] |
| 796 | pub fn (s string) u8() u8 { |
| 797 | return u8(strconv.common_parse_uint(s, 0, 8, false, false) or { 0 }) |
| 798 | } |
| 799 | |
| 800 | // u16 returns the value of the string as u16 `'1'.u16() == u16(1)`. |
| 801 | @[inline] |
| 802 | pub fn (s string) u16() u16 { |
| 803 | return u16(strconv.common_parse_uint(s, 0, 16, false, false) or { 0 }) |
| 804 | } |
| 805 | |
| 806 | // u32 returns the value of the string as u32 `'1'.u32() == u32(1)`. |
| 807 | @[inline] |
| 808 | pub fn (s string) u32() u32 { |
| 809 | return u32(strconv.common_parse_uint(s, 0, 32, false, false) or { 0 }) |
| 810 | } |
| 811 | |
| 812 | // u64 returns the value of the string as u64 `'1'.u64() == u64(1)`. |
| 813 | @[inline] |
| 814 | pub fn (s string) u64() u64 { |
| 815 | return strconv.common_parse_uint(s, 0, 64, false, false) or { 0 } |
| 816 | } |
| 817 | |
| 818 | // parse_uint is like `parse_int` but for unsigned numbers |
| 819 | // |
| 820 | // This method directly exposes the `parse_uint` function from `strconv` |
| 821 | // as a method on `string`. For more advanced features, |
| 822 | // consider calling `strconv.common_parse_uint` directly. |
| 823 | @[inline] |
| 824 | pub fn (s string) parse_uint(_base int, _bit_size int) !u64 { |
| 825 | return strconv.parse_uint(s, _base, _bit_size) |
| 826 | } |
| 827 | |
| 828 | // parse_int interprets a string s in the given base (0, 2 to 36) and |
| 829 | // bit size (0 to 64) and returns the corresponding value i. |
| 830 | // |
| 831 | // If the base argument is 0, the true base is implied by the string's |
| 832 | // prefix: 2 for "0b", 8 for "0" or "0o", 16 for "0x", and 10 otherwise. |
| 833 | // Also, for argument base 0 only, underscore characters are permitted |
| 834 | // as defined by the Go syntax for integer literals. |
| 835 | // |
| 836 | // The bitSize argument specifies the integer type |
| 837 | // that the result must fit into. Bit sizes 0, 8, 16, 32, and 64 |
| 838 | // correspond to int, int8, int16, int32, and int64. |
| 839 | // If bitSize is below 0 or above 64, an error is returned. |
| 840 | // |
| 841 | // This method directly exposes the `parse_int` function from `strconv` |
| 842 | // as a method on `string`. For more advanced features, |
| 843 | // consider calling `strconv.common_parse_int` directly. |
| 844 | @[inline] |
| 845 | pub fn (s string) parse_int(_base int, _bit_size int) !i64 { |
| 846 | return strconv.parse_int(s, _base, _bit_size) |
| 847 | } |
| 848 | |
| 849 | @[direct_array_access] |
| 850 | fn (s string) == (a string) bool { |
| 851 | if s.str == 0 { |
| 852 | // Nil string: equal only to another nil/empty string. |
| 853 | // This can happen with zero-initialized struct fields in the native backend. |
| 854 | return a.str == 0 || a.len == 0 |
| 855 | } |
| 856 | if s.len != a.len { |
| 857 | return false |
| 858 | } |
| 859 | unsafe { |
| 860 | return vmemcmp(s.str, a.str, a.len) == 0 |
| 861 | } |
| 862 | } |
| 863 | |
| 864 | // compare returns -1 if `s` < `a`, 0 if `s` == `a`, and 1 if `s` > `a` |
| 865 | @[direct_array_access] |
| 866 | pub fn (s string) compare(a string) int { |
| 867 | min_len := if s.len < a.len { s.len } else { a.len } |
| 868 | for i in 0 .. min_len { |
| 869 | if s[i] < a[i] { |
| 870 | return -1 |
| 871 | } |
| 872 | if s[i] > a[i] { |
| 873 | return 1 |
| 874 | } |
| 875 | } |
| 876 | if s.len < a.len { |
| 877 | return -1 |
| 878 | } |
| 879 | if s.len > a.len { |
| 880 | return 1 |
| 881 | } |
| 882 | return 0 |
| 883 | } |
| 884 | |
| 885 | @[direct_array_access] |
| 886 | fn (s string) < (a string) bool { |
| 887 | for i in 0 .. s.len { |
| 888 | if i >= a.len || s[i] > a[i] { |
| 889 | return false |
| 890 | } else if s[i] < a[i] { |
| 891 | return true |
| 892 | } |
| 893 | } |
| 894 | if s.len < a.len { |
| 895 | return true |
| 896 | } |
| 897 | return false |
| 898 | } |
| 899 | |
| 900 | @[direct_array_access] |
| 901 | fn (s string) + (a string) string { |
| 902 | slen := if s.len > 0 { s.len } else { 0 } |
| 903 | alen := if a.len > 0 { a.len } else { 0 } |
| 904 | new_len := alen + slen |
| 905 | mut res := string{ |
| 906 | str: unsafe { malloc_noscan(new_len + 1) } |
| 907 | len: new_len |
| 908 | } |
| 909 | unsafe { |
| 910 | if slen > 0 { |
| 911 | vmemcpy(res.str, s.str, slen) |
| 912 | } |
| 913 | if alen > 0 { |
| 914 | vmemcpy(res.str + slen, a.str, alen) |
| 915 | } |
| 916 | res.str[new_len] = 0 // V strings are not null terminated, but just in case |
| 917 | } |
| 918 | return res |
| 919 | } |
| 920 | |
| 921 | // string_plus_many concatenates several strings with a single allocation. |
| 922 | @[direct_array_access; markused] |
| 923 | fn string_plus_many(data_len int, input_base &string) string { |
| 924 | mut new_len := 0 |
| 925 | for i := 0; i < data_len; i++ { |
| 926 | part := unsafe { input_base[i] } |
| 927 | new_len += if part.len > 0 { part.len } else { 0 } |
| 928 | } |
| 929 | mut res := string{ |
| 930 | str: unsafe { malloc_noscan(new_len + 1) } |
| 931 | len: new_len |
| 932 | } |
| 933 | mut offset := 0 |
| 934 | unsafe { |
| 935 | for i := 0; i < data_len; i++ { |
| 936 | part := input_base[i] |
| 937 | part_len := if part.len > 0 { part.len } else { 0 } |
| 938 | if part_len > 0 { |
| 939 | vmemcpy(res.str + offset, part.str, part_len) |
| 940 | offset += part_len |
| 941 | } |
| 942 | } |
| 943 | res.str[new_len] = 0 // V strings are not null terminated, but just in case |
| 944 | } |
| 945 | return res |
| 946 | } |
| 947 | |
| 948 | // for `s + s2 + s3`, an optimization (faster than string_plus(string_plus(s1, s2), s3)) |
| 949 | @[direct_array_access] |
| 950 | fn (s string) plus_two(a string, b string) string { |
| 951 | slen := if s.len > 0 { s.len } else { 0 } |
| 952 | alen := if a.len > 0 { a.len } else { 0 } |
| 953 | blen := if b.len > 0 { b.len } else { 0 } |
| 954 | new_len := alen + blen + slen |
| 955 | mut res := string{ |
| 956 | str: unsafe { malloc_noscan(new_len + 1) } |
| 957 | len: new_len |
| 958 | } |
| 959 | unsafe { |
| 960 | if slen > 0 { |
| 961 | vmemcpy(res.str, s.str, slen) |
| 962 | } |
| 963 | if alen > 0 { |
| 964 | vmemcpy(res.str + slen, a.str, alen) |
| 965 | } |
| 966 | if blen > 0 { |
| 967 | vmemcpy(res.str + slen + alen, b.str, blen) |
| 968 | } |
| 969 | res.str[new_len] = 0 // V strings are not null terminated, but just in case |
| 970 | } |
| 971 | return res |
| 972 | } |
| 973 | |
| 974 | // split_any splits the string to an array by any of the `delim` chars. |
| 975 | // If the delimiter string is empty then `.split()` is used. |
| 976 | // Example: assert "first row\nsecond row".split_any(" \n") == ['first', 'row', 'second', 'row'] |
| 977 | @[direct_array_access] |
| 978 | pub fn (s string) split_any(delim string) []string { |
| 979 | mut res := []string{} |
| 980 | unsafe { res.flags.set(.noslices) } |
| 981 | defer { unsafe { res.flags.clear(.noslices) } } |
| 982 | mut i := 0 |
| 983 | // check empty source string |
| 984 | if s.len > 0 { |
| 985 | // if empty delimiter string using default split |
| 986 | if delim.len <= 0 { |
| 987 | return s.split('') |
| 988 | } |
| 989 | for index, ch in s { |
| 990 | for delim_ch in delim { |
| 991 | if ch == delim_ch { |
| 992 | res << s[i..index] |
| 993 | i = index + 1 |
| 994 | break |
| 995 | } |
| 996 | } |
| 997 | } |
| 998 | if i < s.len { |
| 999 | res << s[i..] |
| 1000 | } |
| 1001 | } |
| 1002 | return res |
| 1003 | } |
| 1004 | |
| 1005 | // rsplit_any splits the string to an array by any of the `delim` chars in reverse order. |
| 1006 | // If the delimiter string is empty then `.rsplit()` is used. |
| 1007 | // Example: assert "first row\nsecond row".rsplit_any(" \n") == ['row', 'second', 'row', 'first'] |
| 1008 | @[direct_array_access] |
| 1009 | pub fn (s string) rsplit_any(delim string) []string { |
| 1010 | mut res := []string{} |
| 1011 | unsafe { res.flags.set(.noslices) } |
| 1012 | defer { unsafe { res.flags.clear(.noslices) } } |
| 1013 | mut i := s.len - 1 |
| 1014 | if s.len > 0 { |
| 1015 | if delim.len <= 0 { |
| 1016 | return s.rsplit('') |
| 1017 | } |
| 1018 | mut rbound := s.len |
| 1019 | for i >= 0 { |
| 1020 | for delim_ch in delim { |
| 1021 | if s[i] == delim_ch { |
| 1022 | res << s[i + 1..rbound] |
| 1023 | rbound = i |
| 1024 | break |
| 1025 | } |
| 1026 | } |
| 1027 | i-- |
| 1028 | } |
| 1029 | if rbound > 0 { |
| 1030 | res << s[..rbound] |
| 1031 | } |
| 1032 | } |
| 1033 | return res |
| 1034 | } |
| 1035 | |
| 1036 | // split splits the string into an array of strings at the given delimiter. |
| 1037 | // If `delim` is empty the string is split by it's characters. |
| 1038 | // Example: assert 'DEF'.split('') == ['D','E','F'] |
| 1039 | // Example: assert 'A B C'.split(' ') == ['A','B','C'] |
| 1040 | @[inline] |
| 1041 | pub fn (s string) split(delim string) []string { |
| 1042 | return s.split_nth(delim, 0) |
| 1043 | } |
| 1044 | |
| 1045 | // rsplit splits the string into an array of strings at the given delimiter, starting from the right. |
| 1046 | // If `delim` is empty the string is split by it's characters. |
| 1047 | // Example: assert 'DEF'.rsplit('') == ['F','E','D'] |
| 1048 | // Example: assert 'A B C'.rsplit(' ') == ['C','B','A'] |
| 1049 | @[inline] |
| 1050 | pub fn (s string) rsplit(delim string) []string { |
| 1051 | return s.rsplit_nth(delim, 0) |
| 1052 | } |
| 1053 | |
| 1054 | // split_once splits the string into a pair of strings at the given delimiter. |
| 1055 | // Example: |
| 1056 | // ```v |
| 1057 | // path, ext := 'file.ts.dts'.split_once('.')? |
| 1058 | // assert path == 'file' |
| 1059 | // assert ext == 'ts.dts' |
| 1060 | pub fn (s string) split_once(delim string) ?(string, string) { |
| 1061 | result := s.split_nth(delim, 2) |
| 1062 | |
| 1063 | if result.len != 2 { |
| 1064 | return none |
| 1065 | } |
| 1066 | |
| 1067 | return result[0], result[1] |
| 1068 | } |
| 1069 | |
| 1070 | // rsplit_once splits the string into a pair of strings at the given delimiter, starting from the right. |
| 1071 | // NOTE: rsplit_once returns the string at the left side of the delimiter as first part of the pair. |
| 1072 | // Example: |
| 1073 | // ```v |
| 1074 | // path, ext := 'file.ts.dts'.rsplit_once('.')? |
| 1075 | // assert path == 'file.ts' |
| 1076 | // assert ext == 'dts' |
| 1077 | // ``` |
| 1078 | pub fn (s string) rsplit_once(delim string) ?(string, string) { |
| 1079 | result := s.rsplit_nth(delim, 2) |
| 1080 | |
| 1081 | if result.len != 2 { |
| 1082 | return none |
| 1083 | } |
| 1084 | |
| 1085 | return result[1], result[0] |
| 1086 | } |
| 1087 | |
| 1088 | // split_n splits the string based on the passed `delim` substring. |
| 1089 | // It returns the first Nth parts. When N=0, return all the splits. |
| 1090 | // The last returned element has the remainder of the string, even if |
| 1091 | // the remainder contains more `delim` substrings. |
| 1092 | pub fn (s string) split_n(delim string, n int) []string { |
| 1093 | return s.split_nth(delim, n) |
| 1094 | } |
| 1095 | |
| 1096 | // split_nth splits the string based on the passed `delim` substring. |
| 1097 | // It returns the first Nth parts. When N=0, return all the splits. |
| 1098 | // The last returned element has the remainder of the string, even if |
| 1099 | // the remainder contains more `delim` substrings. |
| 1100 | @[direct_array_access] |
| 1101 | pub fn (s string) split_nth(delim string, nth int) []string { |
| 1102 | mut res := []string{} |
| 1103 | unsafe { res.flags.set(.noslices) } // allow freeing of old data during << |
| 1104 | defer { unsafe { res.flags.clear(.noslices) } } |
| 1105 | match delim.len { |
| 1106 | 0 { |
| 1107 | for i, ch in s { |
| 1108 | if nth > 0 && res.len == nth - 1 { |
| 1109 | res << s[i..] |
| 1110 | break |
| 1111 | } |
| 1112 | res << ch.ascii_str() |
| 1113 | } |
| 1114 | } |
| 1115 | 1 { |
| 1116 | delim_byte := delim[0] |
| 1117 | mut start := 0 |
| 1118 | for i, ch in s { |
| 1119 | if ch == delim_byte { |
| 1120 | if nth > 0 && res.len == nth - 1 { |
| 1121 | break |
| 1122 | } |
| 1123 | res << s.substr(start, i) |
| 1124 | start = i + 1 |
| 1125 | } |
| 1126 | } |
| 1127 | if nth < 1 || res.len < nth { |
| 1128 | res << s[start..] |
| 1129 | } |
| 1130 | } |
| 1131 | else { |
| 1132 | mut start := 0 |
| 1133 | // Add up to `nth` segments left of every occurrence of the delimiter. |
| 1134 | for i := 0; i + delim.len <= s.len; { |
| 1135 | if unsafe { s.substr_unsafe(i, i + delim.len) } == delim { |
| 1136 | if nth > 0 && res.len == nth - 1 { |
| 1137 | break |
| 1138 | } |
| 1139 | res << s.substr(start, i) |
| 1140 | i += delim.len |
| 1141 | start = i |
| 1142 | } else { |
| 1143 | i++ |
| 1144 | } |
| 1145 | } |
| 1146 | // Then add the remaining part of the string as the last segment. |
| 1147 | if nth < 1 || res.len < nth { |
| 1148 | res << s[start..] |
| 1149 | } |
| 1150 | } |
| 1151 | } |
| 1152 | |
| 1153 | return res |
| 1154 | } |
| 1155 | |
| 1156 | // rsplit_nth splits the string based on the passed `delim` substring in revese order. |
| 1157 | // It returns the first Nth parts. When N=0, return all the splits. |
| 1158 | // The last returned element has the remainder of the string, even if |
| 1159 | // the remainder contains more `delim` substrings. |
| 1160 | @[direct_array_access] |
| 1161 | pub fn (s string) rsplit_nth(delim string, nth int) []string { |
| 1162 | mut res := []string{} |
| 1163 | unsafe { res.flags.set(.noslices) } // allow freeing of old data during << |
| 1164 | defer { unsafe { res.flags.clear(.noslices) } } |
| 1165 | match delim.len { |
| 1166 | 0 { |
| 1167 | for i := s.len - 1; i >= 0; i-- { |
| 1168 | if nth > 0 && res.len == nth - 1 { |
| 1169 | res << s[..i + 1] |
| 1170 | break |
| 1171 | } |
| 1172 | res << s[i].ascii_str() |
| 1173 | } |
| 1174 | } |
| 1175 | 1 { |
| 1176 | delim_byte := delim[0] |
| 1177 | mut rbound := s.len |
| 1178 | for i := s.len - 1; i >= 0; i-- { |
| 1179 | if s[i] == delim_byte { |
| 1180 | if nth > 0 && res.len == nth - 1 { |
| 1181 | break |
| 1182 | } |
| 1183 | res << s[i + 1..rbound] |
| 1184 | rbound = i |
| 1185 | } |
| 1186 | } |
| 1187 | if nth < 1 || res.len < nth { |
| 1188 | res << s[..rbound] |
| 1189 | } |
| 1190 | } |
| 1191 | else { |
| 1192 | mut rbound := s.len |
| 1193 | for i := s.len - 1; i >= 0; i-- { |
| 1194 | is_delim := i - delim.len >= 0 && s[i - delim.len..i] == delim |
| 1195 | if is_delim { |
| 1196 | if nth > 0 && res.len == nth - 1 { |
| 1197 | break |
| 1198 | } |
| 1199 | res << s[i..rbound] |
| 1200 | i -= delim.len |
| 1201 | rbound = i |
| 1202 | } |
| 1203 | } |
| 1204 | if nth < 1 || res.len < nth { |
| 1205 | res << s[..rbound] |
| 1206 | } |
| 1207 | } |
| 1208 | } |
| 1209 | |
| 1210 | return res |
| 1211 | } |
| 1212 | |
| 1213 | // split_into_lines splits the string by newline characters. |
| 1214 | // newlines are stripped. |
| 1215 | // `\r` (MacOS), `\n` (POSIX), and `\r\n` (WinOS) line endings are all supported (including mixed line endings). |
| 1216 | // NOTE: algorithm is "greedy", consuming '\r\n' as a single line ending with higher priority than '\r' and '\n' as multiple endings |
| 1217 | @[direct_array_access] |
| 1218 | pub fn (s string) split_into_lines() []string { |
| 1219 | mut res := []string{} |
| 1220 | if s.len == 0 { |
| 1221 | return res |
| 1222 | } |
| 1223 | unsafe { res.flags.set(.noslices) } // allow freeing of old data during << |
| 1224 | defer { unsafe { res.flags.clear(.noslices) } } |
| 1225 | cr := `\r` |
| 1226 | lf := `\n` |
| 1227 | mut line_start := 0 |
| 1228 | for i := 0; i < s.len; i++ { |
| 1229 | if line_start <= i { |
| 1230 | if s[i] == lf { |
| 1231 | res << if line_start == i { '' } else { s[line_start..i] } |
| 1232 | line_start = i + 1 |
| 1233 | } else if s[i] == cr { |
| 1234 | res << if line_start == i { '' } else { s[line_start..i] } |
| 1235 | if (i + 1) < s.len && s[i + 1] == lf { |
| 1236 | line_start = i + 2 |
| 1237 | } else { |
| 1238 | line_start = i + 1 |
| 1239 | } |
| 1240 | } |
| 1241 | } |
| 1242 | } |
| 1243 | if line_start < s.len { |
| 1244 | res << s[line_start..] |
| 1245 | } |
| 1246 | return res |
| 1247 | } |
| 1248 | |
| 1249 | // split_by_space splits the string by whitespace (any of ` `, `\n`, `\t`, `\v`, `\f`, `\r`). |
| 1250 | // Repeated, trailing or leading whitespaces will be omitted. |
| 1251 | pub fn (s string) split_by_space() []string { |
| 1252 | mut res := []string{} |
| 1253 | unsafe { res.flags.set(.noslices) } |
| 1254 | defer { unsafe { res.flags.clear(.noslices) } } |
| 1255 | for word in s.split_any(' \n\t\v\f\r') { |
| 1256 | if word != '' { |
| 1257 | res << word |
| 1258 | } |
| 1259 | } |
| 1260 | return res |
| 1261 | } |
| 1262 | |
| 1263 | // substr returns the string between index positions `start` and `end`. |
| 1264 | // Example: assert 'ABCD'.substr(1,3) == 'BC' |
| 1265 | @[direct_array_access] |
| 1266 | pub fn (s string) substr(start int, _end int) string { |
| 1267 | // WARNNING: The is a temp solution for bootstrap! |
| 1268 | end := if _end == max_i64 || _end == max_i32 { s.len } else { _end } // max_int |
| 1269 | $if !no_bounds_checking { |
| 1270 | if start > end || start > s.len || end > s.len || start < 0 || end < 0 { |
| 1271 | panic('substr(' + impl_i64_to_string(start) + ', ' + impl_i64_to_string(end) + |
| 1272 | ') out of bounds (len=' + impl_i64_to_string(s.len) + ') s=' + s) |
| 1273 | } |
| 1274 | } |
| 1275 | len := end - start |
| 1276 | if len == s.len { |
| 1277 | return s.clone() |
| 1278 | } |
| 1279 | mut res := string{ |
| 1280 | str: unsafe { malloc_noscan(len + 1) } |
| 1281 | len: len |
| 1282 | } |
| 1283 | unsafe { |
| 1284 | vmemcpy(res.str, s.str + start, len) |
| 1285 | res.str[len] = 0 |
| 1286 | } |
| 1287 | return res |
| 1288 | } |
| 1289 | |
| 1290 | // substr_unsafe works like substr(), but doesn't copy (allocate) the substring |
| 1291 | @[direct_array_access] |
| 1292 | pub fn (s string) substr_unsafe(start int, _end int) string { |
| 1293 | end := if _end == 2147483647 { s.len } else { _end } // max_int |
| 1294 | len := end - start |
| 1295 | if len == s.len { |
| 1296 | return s |
| 1297 | } |
| 1298 | return string{ |
| 1299 | str: unsafe { s.str + start } |
| 1300 | len: len |
| 1301 | } |
| 1302 | } |
| 1303 | |
| 1304 | // substr_or returns substr(start, end) if bounds are valid, otherwise returns fallback. |
| 1305 | // Used by the native backend for `s[start..end] or { fallback }` expressions. |
| 1306 | @[direct_array_access] |
| 1307 | pub fn (s string) substr_or(start int, _end int, fallback string) string { |
| 1308 | end := if _end == max_i64 || _end == max_i32 { s.len } else { _end } |
| 1309 | if start < 0 || start > end || end > s.len { |
| 1310 | return fallback |
| 1311 | } |
| 1312 | return s.substr(start, end) |
| 1313 | } |
| 1314 | |
| 1315 | // version of `substr()` that is used in `a[start..end] or {` |
| 1316 | // return an error when the index is out of range |
| 1317 | @[direct_array_access] |
| 1318 | pub fn (s string) substr_with_check(start int, _end int) !string { |
| 1319 | // WARNNING: The is a temp solution for bootstrap! |
| 1320 | end := if _end == max_i64 || _end == max_i32 { s.len } else { _end } // max_int |
| 1321 | if start > end || start > s.len || end > s.len || start < 0 || end < 0 { |
| 1322 | return error('substr(' + impl_i64_to_string(start) + ', ' + impl_i64_to_string(end) + |
| 1323 | ') out of bounds (len=' + impl_i64_to_string(s.len) + ')') |
| 1324 | } |
| 1325 | len := end - start |
| 1326 | if len == s.len { |
| 1327 | return s.clone() |
| 1328 | } |
| 1329 | mut res := string{ |
| 1330 | str: unsafe { malloc_noscan(len + 1) } |
| 1331 | len: len |
| 1332 | } |
| 1333 | unsafe { |
| 1334 | vmemcpy(res.str, s.str + start, len) |
| 1335 | res.str[len] = 0 |
| 1336 | } |
| 1337 | return res |
| 1338 | } |
| 1339 | |
| 1340 | // substr_ni returns the string between index positions `start` and `end` allowing negative indexes |
| 1341 | // This function always return a valid string. |
| 1342 | @[direct_array_access] |
| 1343 | pub fn (s string) substr_ni(_start int, _end int) string { |
| 1344 | mut start := _start |
| 1345 | // WARNNING: The is a temp solution for bootstrap! |
| 1346 | mut end := if _end == max_i64 || _end == max_i32 { s.len } else { _end } |
| 1347 | |
| 1348 | // borders math |
| 1349 | if start < 0 { |
| 1350 | start = s.len + start |
| 1351 | if start < 0 { |
| 1352 | start = 0 |
| 1353 | } |
| 1354 | } |
| 1355 | |
| 1356 | if end < 0 { |
| 1357 | end = s.len + end |
| 1358 | if end < 0 { |
| 1359 | end = 0 |
| 1360 | } |
| 1361 | } |
| 1362 | if end >= s.len { |
| 1363 | end = s.len |
| 1364 | } |
| 1365 | |
| 1366 | if start > s.len || end < start { |
| 1367 | return '' |
| 1368 | } |
| 1369 | |
| 1370 | len := end - start |
| 1371 | |
| 1372 | // string copy |
| 1373 | mut res := string{ |
| 1374 | str: unsafe { malloc_noscan(len + 1) } |
| 1375 | len: len |
| 1376 | } |
| 1377 | unsafe { |
| 1378 | vmemcpy(res.str, s.str + start, len) |
| 1379 | res.str[len] = 0 |
| 1380 | } |
| 1381 | return res |
| 1382 | } |
| 1383 | |
| 1384 | // index_ returns the position of the first character of the input string. |
| 1385 | // It will return `-1` if the input string can't be found. |
| 1386 | @[direct_array_access] |
| 1387 | pub fn (s string) index_(p string) int { |
| 1388 | if p.len > s.len || p.len == 0 || u64(s.str) <= 0xFFFF || u64(p.str) <= 0xFFFF { |
| 1389 | return -1 |
| 1390 | } |
| 1391 | if p.len > 2 { |
| 1392 | return s.index_kmp(p) |
| 1393 | } |
| 1394 | mut i := 0 |
| 1395 | for i < s.len { |
| 1396 | mut j := 0 |
| 1397 | for j < p.len && unsafe { s.str[i + j] == p.str[j] } { |
| 1398 | j++ |
| 1399 | } |
| 1400 | if j == p.len { |
| 1401 | return i |
| 1402 | } |
| 1403 | i++ |
| 1404 | } |
| 1405 | return -1 |
| 1406 | } |
| 1407 | |
| 1408 | // index returns the position of the first character of the first occurrence of the `needle` string in `s`. |
| 1409 | // It will return `none` if the `needle` string can't be found in `s`. |
| 1410 | pub fn (s string) index(p string) ?int { |
| 1411 | idx := s.index_(p) |
| 1412 | if idx == -1 { |
| 1413 | return none |
| 1414 | } |
| 1415 | return idx |
| 1416 | } |
| 1417 | |
| 1418 | // last_index returns the position of the first character of the *last* occurrence of the `needle` string in `s`. |
| 1419 | @[inline] |
| 1420 | pub fn (s string) last_index(needle string) ?int { |
| 1421 | idx := s.index_last_(needle) |
| 1422 | if idx == -1 { |
| 1423 | return none |
| 1424 | } |
| 1425 | return idx |
| 1426 | } |
| 1427 | |
| 1428 | const kmp_stack_buffer_size = 20 |
| 1429 | |
| 1430 | // index_kmp does KMP search inside the string `s` for the needle `p`. |
| 1431 | // It returns the first found index where the string `p` is found. |
| 1432 | // It returns -1, when the needle `p` is not present in `s`. |
| 1433 | @[direct_array_access; manualfree] |
| 1434 | fn (s string) index_kmp(p string) int { |
| 1435 | if p.len > s.len { |
| 1436 | return -1 |
| 1437 | } |
| 1438 | mut stack_prefixes := [kmp_stack_buffer_size]int{} |
| 1439 | mut p_prefixes := unsafe { &stack_prefixes[0] } |
| 1440 | if p.len > kmp_stack_buffer_size { |
| 1441 | p_prefixes = unsafe { &int(vcalloc(p.len * int(sizeof(int)))) } |
| 1442 | } |
| 1443 | defer { |
| 1444 | if p.len > kmp_stack_buffer_size { |
| 1445 | unsafe { free(p_prefixes) } |
| 1446 | } |
| 1447 | } |
| 1448 | mut j := 0 |
| 1449 | for i := 1; i < p.len; i++ { |
| 1450 | for unsafe { p.str[j] != p.str[i] } && j > 0 { |
| 1451 | j = unsafe { p_prefixes[j - 1] } |
| 1452 | } |
| 1453 | if unsafe { p.str[j] == p.str[i] } { |
| 1454 | j++ |
| 1455 | } |
| 1456 | unsafe { |
| 1457 | p_prefixes[i] = j |
| 1458 | } |
| 1459 | } |
| 1460 | j = 0 |
| 1461 | for i in 0 .. s.len { |
| 1462 | for unsafe { p.str[j] != s.str[i] } && j > 0 { |
| 1463 | j = unsafe { p_prefixes[j - 1] } |
| 1464 | } |
| 1465 | if unsafe { p.str[j] == s.str[i] } { |
| 1466 | j++ |
| 1467 | } |
| 1468 | if j == p.len { |
| 1469 | return i - p.len + 1 |
| 1470 | } |
| 1471 | } |
| 1472 | return -1 |
| 1473 | } |
| 1474 | |
| 1475 | // index_any returns the position of any of the characters in the input string - if found. |
| 1476 | pub fn (s string) index_any(chars string) int { |
| 1477 | for i, ss in s { |
| 1478 | for c in chars { |
| 1479 | if c == ss { |
| 1480 | return i |
| 1481 | } |
| 1482 | } |
| 1483 | } |
| 1484 | return -1 |
| 1485 | } |
| 1486 | |
| 1487 | // index_last_ returns the position of the last occurrence of the given string `p` in `s`. |
| 1488 | @[direct_array_access] |
| 1489 | fn (s string) index_last_(p string) int { |
| 1490 | if p.len > s.len || p.len == 0 { |
| 1491 | return -1 |
| 1492 | } |
| 1493 | mut i := s.len - p.len |
| 1494 | for i >= 0 { |
| 1495 | mut j := 0 |
| 1496 | for j < p.len && unsafe { s.str[i + j] == p.str[j] } { |
| 1497 | j++ |
| 1498 | } |
| 1499 | if j == p.len { |
| 1500 | return i |
| 1501 | } |
| 1502 | i-- |
| 1503 | } |
| 1504 | return -1 |
| 1505 | } |
| 1506 | |
| 1507 | // index_after returns the position of the input string, starting search from `start` position. |
| 1508 | @[direct_array_access] |
| 1509 | pub fn (s string) index_after(p string, start int) ?int { |
| 1510 | if p.len > s.len { |
| 1511 | return none |
| 1512 | } |
| 1513 | mut strt := start |
| 1514 | if start < 0 { |
| 1515 | strt = 0 |
| 1516 | } |
| 1517 | if start >= s.len { |
| 1518 | return none |
| 1519 | } |
| 1520 | mut i := strt |
| 1521 | for i < s.len { |
| 1522 | mut j := 0 |
| 1523 | mut ii := i |
| 1524 | for j < p.len && unsafe { s.str[ii] == p.str[j] } { |
| 1525 | j++ |
| 1526 | ii++ |
| 1527 | } |
| 1528 | if j == p.len { |
| 1529 | return i |
| 1530 | } |
| 1531 | i++ |
| 1532 | } |
| 1533 | return none |
| 1534 | } |
| 1535 | |
| 1536 | // index_after_ returns the position of the input string, starting search from `start` position. |
| 1537 | @[direct_array_access] |
| 1538 | pub fn (s string) index_after_(p string, start int) int { |
| 1539 | if p.len > s.len { |
| 1540 | return -1 |
| 1541 | } |
| 1542 | mut strt := start |
| 1543 | if start < 0 { |
| 1544 | strt = 0 |
| 1545 | } |
| 1546 | if start >= s.len { |
| 1547 | return -1 |
| 1548 | } |
| 1549 | mut i := strt |
| 1550 | for i < s.len { |
| 1551 | mut j := 0 |
| 1552 | mut ii := i |
| 1553 | for j < p.len && unsafe { s.str[ii] == p.str[j] } { |
| 1554 | j++ |
| 1555 | ii++ |
| 1556 | } |
| 1557 | if j == p.len { |
| 1558 | return i |
| 1559 | } |
| 1560 | i++ |
| 1561 | } |
| 1562 | return -1 |
| 1563 | } |
| 1564 | |
| 1565 | // index_u8 returns the index of byte `c` if found in the string. |
| 1566 | // index_u8 returns -1 if the byte can not be found. |
| 1567 | @[direct_array_access] |
| 1568 | pub fn (s string) index_u8(c u8) int { |
| 1569 | for i, b in s { |
| 1570 | if b == c { |
| 1571 | return i |
| 1572 | } |
| 1573 | } |
| 1574 | return -1 |
| 1575 | } |
| 1576 | |
| 1577 | // last_index_u8 returns the index of the last occurrence of byte `c` if it was found in the string. |
| 1578 | @[direct_array_access; inline] |
| 1579 | pub fn (s string) last_index_u8(c u8) int { |
| 1580 | for i := s.len - 1; i >= 0; i-- { |
| 1581 | if s[i] == c { |
| 1582 | return i |
| 1583 | } |
| 1584 | } |
| 1585 | return -1 |
| 1586 | } |
| 1587 | |
| 1588 | // count returns the number of occurrences of `substr` in the string. |
| 1589 | // count returns -1 if no `substr` could be found. |
| 1590 | @[direct_array_access] |
| 1591 | pub fn (s string) count(substr string) int { |
| 1592 | if s.len == 0 || substr.len == 0 { |
| 1593 | return 0 |
| 1594 | } |
| 1595 | if substr.len > s.len { |
| 1596 | return 0 |
| 1597 | } |
| 1598 | |
| 1599 | mut n := 0 |
| 1600 | |
| 1601 | if substr.len == 1 { |
| 1602 | target := substr[0] |
| 1603 | |
| 1604 | for letter in s { |
| 1605 | if letter == target { |
| 1606 | n++ |
| 1607 | } |
| 1608 | } |
| 1609 | |
| 1610 | return n |
| 1611 | } |
| 1612 | |
| 1613 | mut i := 0 |
| 1614 | for { |
| 1615 | i = s.index_after_(substr, i) |
| 1616 | if i == -1 { |
| 1617 | return n |
| 1618 | } |
| 1619 | i += substr.len |
| 1620 | n++ |
| 1621 | } |
| 1622 | return 0 // TODO: can never get here - v doesn't know that |
| 1623 | } |
| 1624 | |
| 1625 | // contains_u8 returns `true` if the string contains the byte value `x`. |
| 1626 | // See also: [`string.index_u8`](#string.index_u8) , to get the index of the byte as well. |
| 1627 | pub fn (s string) contains_u8(x u8) bool { |
| 1628 | for c in s { |
| 1629 | if x == c { |
| 1630 | return true |
| 1631 | } |
| 1632 | } |
| 1633 | return false |
| 1634 | } |
| 1635 | |
| 1636 | // contains returns `true` if the string contains `substr`. |
| 1637 | // See also: [`string.index`](#string.index) |
| 1638 | pub fn (s string) contains(substr string) bool { |
| 1639 | if substr.len == 0 { |
| 1640 | return true |
| 1641 | } |
| 1642 | if substr.len == 1 { |
| 1643 | return s.contains_u8(unsafe { substr.str[0] }) |
| 1644 | } |
| 1645 | return s.index_(substr) != -1 |
| 1646 | } |
| 1647 | |
| 1648 | // contains_any returns `true` if the string contains any chars in `chars`. |
| 1649 | pub fn (s string) contains_any(chars string) bool { |
| 1650 | for c in chars { |
| 1651 | if s.contains_u8(c) { |
| 1652 | return true |
| 1653 | } |
| 1654 | } |
| 1655 | return false |
| 1656 | } |
| 1657 | |
| 1658 | // contains_only returns `true`, if the string contains only the characters in `chars`. |
| 1659 | pub fn (s string) contains_only(chars string) bool { |
| 1660 | if chars.len == 0 { |
| 1661 | return false |
| 1662 | } |
| 1663 | for ch in s { |
| 1664 | mut res := 0 |
| 1665 | for i := 0; i < chars.len && res == 0; i++ { |
| 1666 | res += int(ch == unsafe { chars.str[i] }) |
| 1667 | } |
| 1668 | if res == 0 { |
| 1669 | return false |
| 1670 | } |
| 1671 | } |
| 1672 | return true |
| 1673 | } |
| 1674 | |
| 1675 | // contains_any_substr returns `true` if the string contains any of the strings in `substrs`. |
| 1676 | pub fn (s string) contains_any_substr(substrs []string) bool { |
| 1677 | if substrs.len == 0 { |
| 1678 | return true |
| 1679 | } |
| 1680 | for sub in substrs { |
| 1681 | if s.contains(sub) { |
| 1682 | return true |
| 1683 | } |
| 1684 | } |
| 1685 | return false |
| 1686 | } |
| 1687 | |
| 1688 | // starts_with returns `true` if the string starts with `p`. |
| 1689 | @[direct_array_access] |
| 1690 | pub fn (s string) starts_with(p string) bool { |
| 1691 | if p.len > s.len || u64(s.str) <= 0xFFFF || u64(p.str) <= 0xFFFF { |
| 1692 | return false |
| 1693 | } else if unsafe { vmemcmp(s.str, p.str, p.len) == 0 } { |
| 1694 | return true |
| 1695 | } |
| 1696 | return false |
| 1697 | } |
| 1698 | |
| 1699 | // ends_with returns `true` if the string ends with `p`. |
| 1700 | @[direct_array_access] |
| 1701 | pub fn (s string) ends_with(p string) bool { |
| 1702 | if p.len > s.len || u64(s.str) <= 0xFFFF || u64(p.str) <= 0xFFFF { |
| 1703 | return false |
| 1704 | } else if unsafe { vmemcmp(s.str + s.len - p.len, p.str, p.len) == 0 } { |
| 1705 | return true |
| 1706 | } |
| 1707 | return false |
| 1708 | } |
| 1709 | |
| 1710 | // to_lower_ascii returns the string in all lowercase characters. |
| 1711 | // It is faster than `s.to_lower()`, but works only when the input |
| 1712 | // string `s` is composed *entirely* from ASCII characters. |
| 1713 | // Use `s.to_lower()` instead, if you are not sure. |
| 1714 | @[direct_array_access] |
| 1715 | pub fn (s string) to_lower_ascii() string { |
| 1716 | unsafe { |
| 1717 | mut b := malloc_noscan(s.len + 1) |
| 1718 | for i in 0 .. s.len { |
| 1719 | if s.str[i] >= `A` && s.str[i] <= `Z` { |
| 1720 | b[i] = s.str[i] + 32 |
| 1721 | } else { |
| 1722 | b[i] = s.str[i] |
| 1723 | } |
| 1724 | } |
| 1725 | b[s.len] = 0 |
| 1726 | return tos(b, s.len) |
| 1727 | } |
| 1728 | } |
| 1729 | |
| 1730 | // to_lower returns the string in all lowercase characters. |
| 1731 | // Example: assert 'Hello V'.to_lower() == 'hello v' |
| 1732 | @[direct_array_access] |
| 1733 | pub fn (s string) to_lower() string { |
| 1734 | if s.is_pure_ascii() { |
| 1735 | return s.to_lower_ascii() |
| 1736 | } |
| 1737 | mut runes := s.runes() |
| 1738 | for i in 0 .. runes.len { |
| 1739 | runes[i] = runes[i].to_lower() |
| 1740 | } |
| 1741 | return runes.string() |
| 1742 | } |
| 1743 | |
| 1744 | // is_lower returns `true`, if all characters in the string are lowercase. |
| 1745 | // It only works when the input is composed entirely from ASCII characters. |
| 1746 | // Example: assert 'hello developer'.is_lower() == true |
| 1747 | @[direct_array_access] |
| 1748 | pub fn (s string) is_lower() bool { |
| 1749 | if s == '' || s[0].is_digit() { |
| 1750 | return false |
| 1751 | } |
| 1752 | for i in 0 .. s.len { |
| 1753 | if s[i] >= `A` && s[i] <= `Z` { |
| 1754 | return false |
| 1755 | } |
| 1756 | } |
| 1757 | return true |
| 1758 | } |
| 1759 | |
| 1760 | // to_upper_ascii returns the string in all UPPERCASE characters. |
| 1761 | // It is faster than `s.to_upper()`, but works only when the input |
| 1762 | // string `s` is composed *entirely* from ASCII characters. |
| 1763 | // Use `s.to_upper()` instead, if you are not sure. |
| 1764 | @[direct_array_access] |
| 1765 | pub fn (s string) to_upper_ascii() string { |
| 1766 | unsafe { |
| 1767 | mut b := malloc_noscan(s.len + 1) |
| 1768 | for i in 0 .. s.len { |
| 1769 | if s.str[i] >= `a` && s.str[i] <= `z` { |
| 1770 | b[i] = s.str[i] - 32 |
| 1771 | } else { |
| 1772 | b[i] = s.str[i] |
| 1773 | } |
| 1774 | } |
| 1775 | b[s.len] = 0 |
| 1776 | return tos(b, s.len) |
| 1777 | } |
| 1778 | } |
| 1779 | |
| 1780 | // to_upper returns the string in all uppercase characters. |
| 1781 | // Example: assert 'Hello V'.to_upper() == 'HELLO V' |
| 1782 | @[direct_array_access] |
| 1783 | pub fn (s string) to_upper() string { |
| 1784 | if s.is_pure_ascii() { |
| 1785 | return s.to_upper_ascii() |
| 1786 | } |
| 1787 | mut runes := s.runes() |
| 1788 | for i in 0 .. runes.len { |
| 1789 | runes[i] = runes[i].to_upper() |
| 1790 | } |
| 1791 | return runes.string() |
| 1792 | } |
| 1793 | |
| 1794 | // is_upper returns `true` if all ASCII letters in the string are uppercase, |
| 1795 | // and the string contains at least one uppercase ASCII letter. |
| 1796 | // It only works when the input is composed entirely from ASCII characters. |
| 1797 | // See also: [`byte.is_capital`](#byte.is_capital) |
| 1798 | // Example: assert 'HELLO V'.is_upper() == true |
| 1799 | @[direct_array_access] |
| 1800 | pub fn (s string) is_upper() bool { |
| 1801 | if s == '' { |
| 1802 | return false |
| 1803 | } |
| 1804 | mut has_upper := false |
| 1805 | for i in 0 .. s.len { |
| 1806 | if s[i] >= `a` && s[i] <= `z` { |
| 1807 | return false |
| 1808 | } |
| 1809 | if s[i] >= `A` && s[i] <= `Z` { |
| 1810 | has_upper = true |
| 1811 | } |
| 1812 | } |
| 1813 | return has_upper |
| 1814 | } |
| 1815 | |
| 1816 | // capitalize returns the string with the first character capitalized. |
| 1817 | // Example: assert 'hello'.capitalize() == 'Hello' |
| 1818 | @[direct_array_access] |
| 1819 | pub fn (s string) capitalize() string { |
| 1820 | if s.len == 0 { |
| 1821 | return '' |
| 1822 | } |
| 1823 | if s.len == 1 { |
| 1824 | return s[0].ascii_str().to_upper() |
| 1825 | } |
| 1826 | r := s.runes() |
| 1827 | letter := r[0].str() |
| 1828 | uletter := letter.to_upper() |
| 1829 | rrest := r[1..] |
| 1830 | srest := rrest.string() |
| 1831 | res := uletter + srest |
| 1832 | return res |
| 1833 | } |
| 1834 | |
| 1835 | // uncapitalize returns the string with the first character uncapitalized. |
| 1836 | // Example: assert 'Hello, Bob!'.uncapitalize() == 'hello, Bob!' |
| 1837 | @[direct_array_access] |
| 1838 | pub fn (s string) uncapitalize() string { |
| 1839 | if s.len == 0 { |
| 1840 | return '' |
| 1841 | } |
| 1842 | if s.len == 1 { |
| 1843 | return s[0].ascii_str().to_lower() |
| 1844 | } |
| 1845 | r := s.runes() |
| 1846 | letter := r[0].str() |
| 1847 | lletter := letter.to_lower() |
| 1848 | rrest := r[1..] |
| 1849 | srest := rrest.string() |
| 1850 | res := lletter + srest |
| 1851 | return res |
| 1852 | } |
| 1853 | |
| 1854 | // is_capital returns `true`, if the first character in the string `s`, |
| 1855 | // is a capital letter, and the rest are NOT. |
| 1856 | // Example: assert 'Hello'.is_capital() == true |
| 1857 | // Example: assert 'HelloWorld'.is_capital() == false |
| 1858 | @[direct_array_access] |
| 1859 | pub fn (s string) is_capital() bool { |
| 1860 | if s.len == 0 || !(s[0] >= `A` && s[0] <= `Z`) { |
| 1861 | return false |
| 1862 | } |
| 1863 | for i in 1 .. s.len { |
| 1864 | if s[i] >= `A` && s[i] <= `Z` { |
| 1865 | return false |
| 1866 | } |
| 1867 | } |
| 1868 | return true |
| 1869 | } |
| 1870 | |
| 1871 | // starts_with_capital returns `true`, if the first character in the string `s`, |
| 1872 | // is a capital letter, even if the rest are not. |
| 1873 | // Example: assert 'Hello'.starts_with_capital() == true |
| 1874 | // Example: assert 'Hello. World.'.starts_with_capital() == true |
| 1875 | @[direct_array_access] |
| 1876 | pub fn (s string) starts_with_capital() bool { |
| 1877 | if s.len == 0 || !s[0].is_capital() { |
| 1878 | return false |
| 1879 | } |
| 1880 | return true |
| 1881 | } |
| 1882 | |
| 1883 | // title returns the string with each word capitalized. |
| 1884 | // Example: assert 'hello v developer'.title() == 'Hello V Developer' |
| 1885 | pub fn (s string) title() string { |
| 1886 | words := s.split(' ') |
| 1887 | mut tit := []string{} |
| 1888 | for word in words { |
| 1889 | tit << word.capitalize() |
| 1890 | } |
| 1891 | title := tit.join(' ') |
| 1892 | return title |
| 1893 | } |
| 1894 | |
| 1895 | // is_title returns true if all words of the string are capitalized. |
| 1896 | // Example: assert 'Hello V Developer'.is_title() == true |
| 1897 | pub fn (s string) is_title() bool { |
| 1898 | words := s.split(' ') |
| 1899 | for word in words { |
| 1900 | if !word.is_capital() { |
| 1901 | return false |
| 1902 | } |
| 1903 | } |
| 1904 | return true |
| 1905 | } |
| 1906 | |
| 1907 | // find_between returns the string found between `start` string and `end` string. |
| 1908 | // Example: assert 'hey [man] how you doin'.find_between('[', ']') == 'man' |
| 1909 | pub fn (s string) find_between(start string, end string) string { |
| 1910 | start_pos := s.index_(start) |
| 1911 | if start_pos == -1 { |
| 1912 | return '' |
| 1913 | } |
| 1914 | // First get everything to the right of 'start' |
| 1915 | val := s[start_pos + start.len..] |
| 1916 | end_pos := val.index_(end) |
| 1917 | if end_pos == -1 { |
| 1918 | return '' |
| 1919 | } |
| 1920 | return val[..end_pos] |
| 1921 | } |
| 1922 | |
| 1923 | // trim_space strips any of ` `, `\n`, `\t`, `\v`, `\f`, `\r` from the start and end of the string. |
| 1924 | // Example: assert ' Hello V '.trim_space() == 'Hello V' |
| 1925 | @[inline] |
| 1926 | pub fn (s string) trim_space() string { |
| 1927 | return s.trim(' \n\t\v\f\r') |
| 1928 | } |
| 1929 | |
| 1930 | // trim_space_left strips any of ` `, `\n`, `\t`, `\v`, `\f`, `\r` from the start of the string. |
| 1931 | // Example: assert ' Hello V '.trim_space_left() == 'Hello V ' |
| 1932 | @[inline] |
| 1933 | pub fn (s string) trim_space_left() string { |
| 1934 | return s.trim_left(' \n\t\v\f\r') |
| 1935 | } |
| 1936 | |
| 1937 | // trim_space_right strips any of ` `, `\n`, `\t`, `\v`, `\f`, `\r` from the end of the string. |
| 1938 | // Example: assert ' Hello V '.trim_space_right() == ' Hello V' |
| 1939 | @[inline] |
| 1940 | pub fn (s string) trim_space_right() string { |
| 1941 | return s.trim_right(' \n\t\v\f\r') |
| 1942 | } |
| 1943 | |
| 1944 | // trim strips any of the characters given in `cutset` from the start and end of the string. |
| 1945 | // Example: assert ' ffHello V ffff'.trim(' f') == 'Hello V' |
| 1946 | pub fn (s string) trim(cutset string) string { |
| 1947 | if s == '' || cutset == '' { |
| 1948 | return s.clone() |
| 1949 | } |
| 1950 | if cutset.is_pure_ascii() { |
| 1951 | return s.trim_chars(cutset, .trim_both) |
| 1952 | } else { |
| 1953 | return s.trim_runes(cutset, .trim_both) |
| 1954 | } |
| 1955 | } |
| 1956 | |
| 1957 | // trim_indexes gets the new start and end indices of a string when any of the characters given in `cutset` were stripped from the start and end of the string. Should be used as an input to `substr()`. If the string contains only the characters in `cutset`, both values returned are zero. |
| 1958 | // Example: left, right := '-hi-'.trim_indexes('-'); assert left == 1; assert right == 3 |
| 1959 | @[direct_array_access] |
| 1960 | pub fn (s string) trim_indexes(cutset string) (int, int) { |
| 1961 | mut pos_left := 0 |
| 1962 | mut pos_right := s.len - 1 |
| 1963 | mut cs_match := true |
| 1964 | for pos_left <= s.len && pos_right >= -1 && cs_match { |
| 1965 | cs_match = false |
| 1966 | for cs in cutset { |
| 1967 | if s[pos_left] == cs { |
| 1968 | pos_left++ |
| 1969 | cs_match = true |
| 1970 | break |
| 1971 | } |
| 1972 | } |
| 1973 | for cs in cutset { |
| 1974 | if s[pos_right] == cs { |
| 1975 | pos_right-- |
| 1976 | cs_match = true |
| 1977 | break |
| 1978 | } |
| 1979 | } |
| 1980 | if pos_left > pos_right { |
| 1981 | return 0, 0 |
| 1982 | } |
| 1983 | } |
| 1984 | return pos_left, pos_right + 1 |
| 1985 | } |
| 1986 | |
| 1987 | enum TrimMode { |
| 1988 | trim_left |
| 1989 | trim_right |
| 1990 | trim_both |
| 1991 | } |
| 1992 | |
| 1993 | @[direct_array_access] |
| 1994 | fn (s string) trim_chars(cutset string, mode TrimMode) string { |
| 1995 | mut pos_left := 0 |
| 1996 | mut pos_right := s.len - 1 |
| 1997 | mut cs_match := true |
| 1998 | for pos_left <= s.len && pos_right >= -1 && cs_match { |
| 1999 | cs_match = false |
| 2000 | if mode in [.trim_left, .trim_both] { |
| 2001 | for cs in cutset { |
| 2002 | if s[pos_left] == cs { |
| 2003 | pos_left++ |
| 2004 | cs_match = true |
| 2005 | break |
| 2006 | } |
| 2007 | } |
| 2008 | } |
| 2009 | if mode in [.trim_right, .trim_both] { |
| 2010 | for cs in cutset { |
| 2011 | if s[pos_right] == cs { |
| 2012 | pos_right-- |
| 2013 | cs_match = true |
| 2014 | break |
| 2015 | } |
| 2016 | } |
| 2017 | } |
| 2018 | if pos_left > pos_right { |
| 2019 | return '' |
| 2020 | } |
| 2021 | } |
| 2022 | return s.substr(pos_left, pos_right + 1) |
| 2023 | } |
| 2024 | |
| 2025 | @[direct_array_access] |
| 2026 | fn (s string) trim_runes(cutset string, mode TrimMode) string { |
| 2027 | s_runes := s.runes() |
| 2028 | cs_runes := cutset.runes() |
| 2029 | mut pos_left := 0 |
| 2030 | mut pos_right := s_runes.len - 1 |
| 2031 | mut cs_match := true |
| 2032 | for pos_left <= s_runes.len && pos_right >= -1 && cs_match { |
| 2033 | cs_match = false |
| 2034 | if mode in [.trim_left, .trim_both] { |
| 2035 | for cs in cs_runes { |
| 2036 | if s_runes[pos_left] == cs { |
| 2037 | pos_left++ |
| 2038 | cs_match = true |
| 2039 | break |
| 2040 | } |
| 2041 | } |
| 2042 | } |
| 2043 | if mode in [.trim_right, .trim_both] { |
| 2044 | for cs in cs_runes { |
| 2045 | if s_runes[pos_right] == cs { |
| 2046 | pos_right-- |
| 2047 | cs_match = true |
| 2048 | break |
| 2049 | } |
| 2050 | } |
| 2051 | } |
| 2052 | if pos_left > pos_right { |
| 2053 | return '' |
| 2054 | } |
| 2055 | } |
| 2056 | return s_runes[pos_left..pos_right + 1].string() |
| 2057 | } |
| 2058 | |
| 2059 | // trim_left strips any of the characters given in `cutset` from the left of the string. |
| 2060 | // Example: assert 'd Hello V developer'.trim_left(' d') == 'Hello V developer' |
| 2061 | @[direct_array_access] |
| 2062 | pub fn (s string) trim_left(cutset string) string { |
| 2063 | if s == '' || cutset == '' { |
| 2064 | return s.clone() |
| 2065 | } |
| 2066 | if cutset.is_pure_ascii() { |
| 2067 | return s.trim_chars(cutset, .trim_left) |
| 2068 | } else { |
| 2069 | return s.trim_runes(cutset, .trim_left) |
| 2070 | } |
| 2071 | } |
| 2072 | |
| 2073 | // trim_right strips any of the characters given in `cutset` from the right of the string. |
| 2074 | // Example: assert ' Hello V d'.trim_right(' d') == ' Hello V' |
| 2075 | @[direct_array_access] |
| 2076 | pub fn (s string) trim_right(cutset string) string { |
| 2077 | if s.len < 1 || cutset.len < 1 { |
| 2078 | return s.clone() |
| 2079 | } |
| 2080 | if cutset.len == 1 { |
| 2081 | cut := cutset[0] |
| 2082 | mut pos_right := s.len - 1 |
| 2083 | for pos_right >= 0 && s[pos_right] == cut { |
| 2084 | pos_right-- |
| 2085 | } |
| 2086 | if pos_right < 0 { |
| 2087 | return '' |
| 2088 | } |
| 2089 | return s.substr(0, pos_right + 1) |
| 2090 | } |
| 2091 | if cutset.len == 2 && cutset.is_pure_ascii() { |
| 2092 | cut0 := cutset[0] |
| 2093 | cut1 := cutset[1] |
| 2094 | mut pos_right := s.len - 1 |
| 2095 | for pos_right >= 0 && (s[pos_right] == cut0 || s[pos_right] == cut1) { |
| 2096 | pos_right-- |
| 2097 | } |
| 2098 | if pos_right < 0 { |
| 2099 | return '' |
| 2100 | } |
| 2101 | return s.substr(0, pos_right + 1) |
| 2102 | } |
| 2103 | if cutset.is_pure_ascii() { |
| 2104 | return s.trim_chars(cutset, .trim_right) |
| 2105 | } else { |
| 2106 | return s.trim_runes(cutset, .trim_right) |
| 2107 | } |
| 2108 | } |
| 2109 | |
| 2110 | // trim_string_left strips `str` from the start of the string. |
| 2111 | // Example: assert 'WorldHello V'.trim_string_left('World') == 'Hello V' |
| 2112 | pub fn (s string) trim_string_left(str string) string { |
| 2113 | if s.starts_with(str) { |
| 2114 | return s[str.len..] |
| 2115 | } |
| 2116 | return s.clone() |
| 2117 | } |
| 2118 | |
| 2119 | // trim_string_right strips `str` from the end of the string. |
| 2120 | // Example: assert 'Hello VWorld'.trim_string_right('World') == 'Hello V' |
| 2121 | pub fn (s string) trim_string_right(str string) string { |
| 2122 | if s.ends_with(str) { |
| 2123 | return s[..s.len - str.len] |
| 2124 | } |
| 2125 | return s.clone() |
| 2126 | } |
| 2127 | |
| 2128 | // compare_strings returns `-1` if `a < b`, `1` if `a > b` else `0`. |
| 2129 | pub fn compare_strings(a &string, b &string) int { |
| 2130 | return match true { |
| 2131 | a < b { -1 } |
| 2132 | a > b { 1 } |
| 2133 | else { 0 } |
| 2134 | } |
| 2135 | } |
| 2136 | |
| 2137 | // compare_strings_by_len returns `-1` if `a.len < b.len`, `1` if `a.len > b.len` else `0`. |
| 2138 | fn compare_strings_by_len(a &string, b &string) int { |
| 2139 | return match true { |
| 2140 | a.len < b.len { -1 } |
| 2141 | a.len > b.len { 1 } |
| 2142 | else { 0 } |
| 2143 | } |
| 2144 | } |
| 2145 | |
| 2146 | // compare_lower_strings returns the same as compare_strings but converts `a` and `b` to lower case before comparing. |
| 2147 | fn compare_lower_strings(a &string, b &string) int { |
| 2148 | aa := a.to_lower() |
| 2149 | bb := b.to_lower() |
| 2150 | return compare_strings(&aa, &bb) |
| 2151 | } |
| 2152 | |
| 2153 | // sort_ignore_case sorts the string array using case insensitive comparing. |
| 2154 | @[inline] |
| 2155 | pub fn (mut s []string) sort_ignore_case() { |
| 2156 | s.sort_with_compare(compare_lower_strings) |
| 2157 | } |
| 2158 | |
| 2159 | // sort_by_len sorts the string array by each string's `.len` length. |
| 2160 | @[inline] |
| 2161 | pub fn (mut s []string) sort_by_len() { |
| 2162 | s.sort_with_compare(compare_strings_by_len) |
| 2163 | } |
| 2164 | |
| 2165 | // str returns a copy of the string |
| 2166 | @[inline] |
| 2167 | pub fn (s string) str() string { |
| 2168 | return s.clone() |
| 2169 | } |
| 2170 | |
| 2171 | // at returns the byte at index `idx`. |
| 2172 | // Example: assert 'ABC'.at(1) == u8(`B`) |
| 2173 | fn (s string) at(idx int) u8 { |
| 2174 | $if !no_bounds_checking { |
| 2175 | if idx < 0 || idx >= s.len { |
| 2176 | panic_n2('string index out of range(idx,s.len):', idx, s.len) |
| 2177 | } |
| 2178 | } |
| 2179 | return unsafe { s.str[idx] } |
| 2180 | } |
| 2181 | |
| 2182 | @[markused] |
| 2183 | fn (s string) at_i64(idx i64) u8 { |
| 2184 | $if !no_bounds_checking { |
| 2185 | if idx < 0 || idx >= i64(s.len) { |
| 2186 | panic_n2('string index out of range(idx,s.len):', idx, s.len) |
| 2187 | } |
| 2188 | } |
| 2189 | return unsafe { s.str[int(idx)] } |
| 2190 | } |
| 2191 | |
| 2192 | @[markused] |
| 2193 | fn (s string) at_u64(idx u64) u8 { |
| 2194 | $if !no_bounds_checking { |
| 2195 | if idx >= u64(s.len) { |
| 2196 | panic('string index out of range(idx,s.len): ' + idx.str() + ', ' + |
| 2197 | impl_i64_to_string(s.len)) |
| 2198 | } |
| 2199 | } |
| 2200 | return unsafe { s.str[int(idx)] } |
| 2201 | } |
| 2202 | |
| 2203 | @[markused] |
| 2204 | fn (s string) at_ni(idx int) u8 { |
| 2205 | return s.at(v_ni_index(idx, s.len)) |
| 2206 | } |
| 2207 | |
| 2208 | // version of `at()` that is used in `a[i] or {` |
| 2209 | // return an error when the index is out of range |
| 2210 | fn (s string) at_with_check(idx int) ?u8 { |
| 2211 | if idx < 0 || idx >= s.len { |
| 2212 | return none |
| 2213 | } |
| 2214 | unsafe { |
| 2215 | return s.str[idx] |
| 2216 | } |
| 2217 | } |
| 2218 | |
| 2219 | @[markused] |
| 2220 | fn (s string) at_with_check_i64(idx i64) ?u8 { |
| 2221 | if idx < 0 || idx >= i64(s.len) { |
| 2222 | return none |
| 2223 | } |
| 2224 | unsafe { |
| 2225 | return s.str[int(idx)] |
| 2226 | } |
| 2227 | } |
| 2228 | |
| 2229 | @[markused] |
| 2230 | fn (s string) at_with_check_u64(idx u64) ?u8 { |
| 2231 | if idx >= u64(s.len) { |
| 2232 | return none |
| 2233 | } |
| 2234 | unsafe { |
| 2235 | return s.str[int(idx)] |
| 2236 | } |
| 2237 | } |
| 2238 | |
| 2239 | @[markused] |
| 2240 | fn (s string) at_with_check_ni(idx int) ?u8 { |
| 2241 | return s.at_with_check(v_ni_index(idx, s.len)) |
| 2242 | } |
| 2243 | |
| 2244 | // Check if a string is an octal value. Returns 'true' if it is, or 'false' if it is not |
| 2245 | @[direct_array_access] |
| 2246 | pub fn (str string) is_oct() bool { |
| 2247 | mut i := 0 |
| 2248 | |
| 2249 | if str.len == 0 { |
| 2250 | return false |
| 2251 | } |
| 2252 | |
| 2253 | if str[i] == `0` { |
| 2254 | i++ |
| 2255 | } else if str[i] == `-` || str[i] == `+` { |
| 2256 | i++ |
| 2257 | |
| 2258 | if i < str.len && str[i] == `0` { |
| 2259 | i++ |
| 2260 | } else { |
| 2261 | return false |
| 2262 | } |
| 2263 | } else { |
| 2264 | return false |
| 2265 | } |
| 2266 | |
| 2267 | if i < str.len && str[i] == `o` { |
| 2268 | i++ |
| 2269 | } else { |
| 2270 | return false |
| 2271 | } |
| 2272 | |
| 2273 | if i == str.len { |
| 2274 | return false |
| 2275 | } |
| 2276 | |
| 2277 | for i < str.len { |
| 2278 | if str[i] < `0` || str[i] > `7` { |
| 2279 | return false |
| 2280 | } |
| 2281 | i++ |
| 2282 | } |
| 2283 | |
| 2284 | return true |
| 2285 | } |
| 2286 | |
| 2287 | // is_bin returns `true` if the string is a binary value. |
| 2288 | @[direct_array_access] |
| 2289 | pub fn (str string) is_bin() bool { |
| 2290 | mut i := 0 |
| 2291 | |
| 2292 | if str.len == 0 { |
| 2293 | return false |
| 2294 | } |
| 2295 | |
| 2296 | if str[i] == `0` { |
| 2297 | i++ |
| 2298 | } else if str[i] == `-` || str[i] == `+` { |
| 2299 | i++ |
| 2300 | |
| 2301 | if i < str.len && str[i] == `0` { |
| 2302 | i++ |
| 2303 | } else { |
| 2304 | return false |
| 2305 | } |
| 2306 | } else { |
| 2307 | return false |
| 2308 | } |
| 2309 | |
| 2310 | if i < str.len && str[i] == `b` { |
| 2311 | i++ |
| 2312 | } else { |
| 2313 | return false |
| 2314 | } |
| 2315 | |
| 2316 | if i == str.len { |
| 2317 | return false |
| 2318 | } |
| 2319 | |
| 2320 | for i < str.len { |
| 2321 | if str[i] < `0` || str[i] > `1` { |
| 2322 | return false |
| 2323 | } |
| 2324 | i++ |
| 2325 | } |
| 2326 | |
| 2327 | return true |
| 2328 | } |
| 2329 | |
| 2330 | // is_hex returns 'true' if the string is a hexadecimal value. |
| 2331 | @[direct_array_access] |
| 2332 | pub fn (str string) is_hex() bool { |
| 2333 | mut i := 0 |
| 2334 | |
| 2335 | if str.len == 0 { |
| 2336 | return false |
| 2337 | } |
| 2338 | |
| 2339 | if str[i] == `0` { |
| 2340 | i++ |
| 2341 | } else if str[i] == `-` || str[i] == `+` { |
| 2342 | i++ |
| 2343 | |
| 2344 | if i < str.len && str[i] == `0` { |
| 2345 | i++ |
| 2346 | } else { |
| 2347 | return false |
| 2348 | } |
| 2349 | } else { |
| 2350 | return false |
| 2351 | } |
| 2352 | |
| 2353 | if i < str.len && str[i] == `x` { |
| 2354 | i++ |
| 2355 | } else { |
| 2356 | return false |
| 2357 | } |
| 2358 | |
| 2359 | if i == str.len { |
| 2360 | return false |
| 2361 | } |
| 2362 | |
| 2363 | for i < str.len { |
| 2364 | // TODO: remove this workaround for v2's parser |
| 2365 | // vfmt off |
| 2366 | if (str[i] < `0` || str[i] > `9`) |
| 2367 | && ((str[i] < `a` || str[i] > `f`) && (str[i] < `A` || str[i] > `F`)) { |
| 2368 | return false |
| 2369 | } |
| 2370 | // vfmt on |
| 2371 | i++ |
| 2372 | } |
| 2373 | |
| 2374 | return true |
| 2375 | } |
| 2376 | |
| 2377 | // Check if a string is an integer value. Returns 'true' if it is, or 'false' if it is not |
| 2378 | @[direct_array_access] |
| 2379 | pub fn (str string) is_int() bool { |
| 2380 | mut i := 0 |
| 2381 | |
| 2382 | if str.len == 0 { |
| 2383 | return false |
| 2384 | } |
| 2385 | |
| 2386 | if (str[i] != `-` && str[i] != `+`) && (!str[i].is_digit()) { |
| 2387 | return false |
| 2388 | } else { |
| 2389 | i++ |
| 2390 | } |
| 2391 | |
| 2392 | if i == str.len && (!str[i - 1].is_digit()) { |
| 2393 | return false |
| 2394 | } |
| 2395 | |
| 2396 | for i < str.len { |
| 2397 | if str[i] < `0` || str[i] > `9` { |
| 2398 | return false |
| 2399 | } |
| 2400 | i++ |
| 2401 | } |
| 2402 | |
| 2403 | return true |
| 2404 | } |
| 2405 | |
| 2406 | // is_space returns `true` if the byte is a white space character. |
| 2407 | // The following list is considered white space characters: ` `, `\t`, `\n`, `\v`, `\f`, `\r`, 0x85, 0xa0 |
| 2408 | // Example: assert u8(` `).is_space() == true |
| 2409 | @[inline] |
| 2410 | pub fn (c u8) is_space() bool { |
| 2411 | // 0x85 is NEXT LINE (NEL) |
| 2412 | // 0xa0 is NO-BREAK SPACE |
| 2413 | return c == 32 || (c > 8 && c < 14) || c == 0x85 || c == 0xa0 |
| 2414 | } |
| 2415 | |
| 2416 | // is_digit returns `true` if the byte is in range 0-9 and `false` otherwise. |
| 2417 | // Example: assert u8(`9`).is_digit() == true |
| 2418 | @[inline] |
| 2419 | pub fn (c u8) is_digit() bool { |
| 2420 | return c >= `0` && c <= `9` |
| 2421 | } |
| 2422 | |
| 2423 | // is_hex_digit returns `true` if the byte is either in range 0-9, a-f or A-F and `false` otherwise. |
| 2424 | // Example: assert u8(`F`).is_hex_digit() == true |
| 2425 | @[inline] |
| 2426 | pub fn (c u8) is_hex_digit() bool { |
| 2427 | return c.is_digit() || (c >= `a` && c <= `f`) || (c >= `A` && c <= `F`) |
| 2428 | } |
| 2429 | |
| 2430 | // is_oct_digit returns `true` if the byte is in range 0-7 and `false` otherwise. |
| 2431 | // Example: assert u8(`7`).is_oct_digit() == true |
| 2432 | @[inline] |
| 2433 | pub fn (c u8) is_oct_digit() bool { |
| 2434 | return c >= `0` && c <= `7` |
| 2435 | } |
| 2436 | |
| 2437 | // is_bin_digit returns `true` if the byte is a binary digit (0 or 1) and `false` otherwise. |
| 2438 | // Example: assert u8(`0`).is_bin_digit() == true |
| 2439 | @[inline] |
| 2440 | pub fn (c u8) is_bin_digit() bool { |
| 2441 | return c == `0` || c == `1` |
| 2442 | } |
| 2443 | |
| 2444 | // is_letter returns `true` if the byte is in range a-z or A-Z and `false` otherwise. |
| 2445 | // Example: assert u8(`V`).is_letter() == true |
| 2446 | @[inline] |
| 2447 | pub fn (c u8) is_letter() bool { |
| 2448 | return (c >= `a` && c <= `z`) || (c >= `A` && c <= `Z`) |
| 2449 | } |
| 2450 | |
| 2451 | // is_alnum returns `true` if the byte is in range a-z, A-Z, 0-9 and `false` otherwise. |
| 2452 | // Example: assert u8(`V`).is_alnum() == true |
| 2453 | @[inline] |
| 2454 | pub fn (c u8) is_alnum() bool { |
| 2455 | return (c >= `a` && c <= `z`) || (c >= `A` && c <= `Z`) || (c >= `0` && c <= `9`) |
| 2456 | } |
| 2457 | |
| 2458 | // free allows for manually freeing the memory occupied by the string |
| 2459 | @[manualfree; unsafe] |
| 2460 | pub fn (s &string) free() { |
| 2461 | $if prealloc { |
| 2462 | return |
| 2463 | } |
| 2464 | if s.is_lit == -98761234 { |
| 2465 | double_free_msg := unsafe { &u8(c'double string.free() detected\n') } |
| 2466 | double_free_msg_len := unsafe { vstrlen(double_free_msg) } |
| 2467 | $if freestanding { |
| 2468 | bare_eprint(double_free_msg, u64(double_free_msg_len)) |
| 2469 | } $else { |
| 2470 | _write_buf_to_fd(1, double_free_msg, double_free_msg_len) |
| 2471 | } |
| 2472 | return |
| 2473 | } |
| 2474 | if s.is_lit == 1 || s.str == 0 { |
| 2475 | return |
| 2476 | } |
| 2477 | unsafe { |
| 2478 | // C.printf(c's: %x %s\n', s.str, s.str) |
| 2479 | free(s.str) |
| 2480 | s.str = nil |
| 2481 | } |
| 2482 | s.len = 0 |
| 2483 | s.is_lit = -98761234 |
| 2484 | } |
| 2485 | |
| 2486 | // before returns the contents before `sub` in the string. |
| 2487 | // If the substring is not found, it returns the full input string. |
| 2488 | // Example: assert '23:34:45.234'.before('.') == '23:34:45' |
| 2489 | // Example: assert 'abcd'.before('.') == 'abcd' |
| 2490 | // TODO: deprecate and remove either .before or .all_before |
| 2491 | pub fn (s string) before(sub string) string { |
| 2492 | pos := s.index_(sub) |
| 2493 | if pos == -1 { |
| 2494 | return s.clone() |
| 2495 | } |
| 2496 | return s[..pos] |
| 2497 | } |
| 2498 | |
| 2499 | // all_before returns the contents before `sub` in the string. |
| 2500 | // If the substring is not found, it returns the full input string. |
| 2501 | // Example: assert '23:34:45.234'.all_before('.') == '23:34:45' |
| 2502 | // Example: assert 'abcd'.all_before('.') == 'abcd' |
| 2503 | pub fn (s string) all_before(sub string) string { |
| 2504 | // TODO: remove dup method |
| 2505 | pos := s.index_(sub) |
| 2506 | if pos == -1 { |
| 2507 | return s.clone() |
| 2508 | } |
| 2509 | return s[..pos] |
| 2510 | } |
| 2511 | |
| 2512 | // all_before_last returns the contents before the last occurrence of `sub` in the string. |
| 2513 | // If the substring is not found, it returns the full input string. |
| 2514 | // Example: assert '23:34:45.234'.all_before_last(':') == '23:34' |
| 2515 | // Example: assert 'abcd'.all_before_last('.') == 'abcd' |
| 2516 | pub fn (s string) all_before_last(sub string) string { |
| 2517 | pos := s.index_last_(sub) |
| 2518 | if pos == -1 { |
| 2519 | return s.clone() |
| 2520 | } |
| 2521 | return s[..pos] |
| 2522 | } |
| 2523 | |
| 2524 | // all_after returns the contents after `sub` in the string. |
| 2525 | // If the substring is not found, it returns the full input string. |
| 2526 | // Example: assert '23:34:45.234'.all_after('.') == '234' |
| 2527 | // Example: assert 'abcd'.all_after('z') == 'abcd' |
| 2528 | pub fn (s string) all_after(sub string) string { |
| 2529 | pos := s.index_(sub) |
| 2530 | if pos == -1 { |
| 2531 | return s.clone() |
| 2532 | } |
| 2533 | return s[pos + sub.len..] |
| 2534 | } |
| 2535 | |
| 2536 | // all_after_last returns the contents after the last occurrence of `sub` in the string. |
| 2537 | // If the substring is not found, it returns the full input string. |
| 2538 | // Example: assert '23:34:45.234'.all_after_last(':') == '45.234' |
| 2539 | // Example: assert 'abcd'.all_after_last('z') == 'abcd' |
| 2540 | pub fn (s string) all_after_last(sub string) string { |
| 2541 | pos := s.index_last_(sub) |
| 2542 | if pos == -1 { |
| 2543 | return s.clone() |
| 2544 | } |
| 2545 | return s[pos + sub.len..] |
| 2546 | } |
| 2547 | |
| 2548 | // all_after_first returns the contents after the first occurrence of `sub` in the string. |
| 2549 | // If the substring is not found, it returns the full input string. |
| 2550 | // Example: assert '23:34:45.234'.all_after_first(':') == '34:45.234' |
| 2551 | // Example: assert 'abcd'.all_after_first('z') == 'abcd' |
| 2552 | pub fn (s string) all_after_first(sub string) string { |
| 2553 | pos := s.index_(sub) |
| 2554 | if pos == -1 { |
| 2555 | return s.clone() |
| 2556 | } |
| 2557 | return s[pos + sub.len..] |
| 2558 | } |
| 2559 | |
| 2560 | // after returns the contents after the last occurrence of `sub` in the string. |
| 2561 | // If the substring is not found, it returns the full input string. |
| 2562 | // Example: assert '23:34:45.234'.after(':') == '45.234' |
| 2563 | // Example: assert 'abcd'.after('z') == 'abcd' |
| 2564 | // TODO: deprecate either .all_after_last or .after |
| 2565 | @[inline] |
| 2566 | pub fn (s string) after(sub string) string { |
| 2567 | return s.all_after_last(sub) |
| 2568 | } |
| 2569 | |
| 2570 | // after_char returns the contents after the first occurrence of `sub` character in the string. |
| 2571 | // If the substring is not found, it returns the full input string. |
| 2572 | // Example: assert '23:34:45.234'.after_char(`:`) == '34:45.234' |
| 2573 | // Example: assert 'abcd'.after_char(`:`) == 'abcd' |
| 2574 | pub fn (s string) after_char(sub u8) string { |
| 2575 | mut pos := -1 |
| 2576 | for i, c in s { |
| 2577 | if c == sub { |
| 2578 | pos = i |
| 2579 | break |
| 2580 | } |
| 2581 | } |
| 2582 | if pos == -1 { |
| 2583 | return s.clone() |
| 2584 | } |
| 2585 | return s[pos + 1..] |
| 2586 | } |
| 2587 | |
| 2588 | // join joins a string array into a string using `sep` separator. |
| 2589 | // Example: assert ['Hello','V'].join(' ') == 'Hello V' |
| 2590 | pub fn (a []string) join(sep string) string { |
| 2591 | if a.len == 0 { |
| 2592 | return '' |
| 2593 | } |
| 2594 | mut len := 0 |
| 2595 | for val in a { |
| 2596 | len += val.len + sep.len |
| 2597 | } |
| 2598 | len -= sep.len |
| 2599 | // Allocate enough memory |
| 2600 | mut res := string{ |
| 2601 | str: unsafe { malloc_noscan(len + 1) } |
| 2602 | len: len |
| 2603 | } |
| 2604 | mut idx := 0 |
| 2605 | for i, val in a { |
| 2606 | unsafe { |
| 2607 | vmemcpy(voidptr(res.str + idx), val.str, val.len) |
| 2608 | idx += val.len |
| 2609 | } |
| 2610 | // Add sep if it's not last |
| 2611 | if i != a.len - 1 { |
| 2612 | unsafe { |
| 2613 | vmemcpy(voidptr(res.str + idx), sep.str, sep.len) |
| 2614 | idx += sep.len |
| 2615 | } |
| 2616 | } |
| 2617 | } |
| 2618 | unsafe { |
| 2619 | res.str[res.len] = 0 |
| 2620 | } |
| 2621 | return res |
| 2622 | } |
| 2623 | |
| 2624 | // join_lines joins a string array into a string using a `\n` newline delimiter. |
| 2625 | @[inline] |
| 2626 | pub fn (s []string) join_lines() string { |
| 2627 | return s.join('\n') |
| 2628 | } |
| 2629 | |
| 2630 | // reverse returns a reversed string. |
| 2631 | // Example: assert 'Hello V'.reverse() == 'V olleH' |
| 2632 | @[direct_array_access] |
| 2633 | pub fn (s string) reverse() string { |
| 2634 | if s.len == 0 || s.len == 1 { |
| 2635 | return s.clone() |
| 2636 | } |
| 2637 | mut res := string{ |
| 2638 | str: unsafe { malloc_noscan(s.len + 1) } |
| 2639 | len: s.len |
| 2640 | } |
| 2641 | for i := s.len - 1; i >= 0; i-- { |
| 2642 | unsafe { |
| 2643 | res.str[s.len - i - 1] = s[i] |
| 2644 | } |
| 2645 | } |
| 2646 | unsafe { |
| 2647 | res.str[res.len] = 0 |
| 2648 | } |
| 2649 | return res |
| 2650 | } |
| 2651 | |
| 2652 | // limit returns a portion of the string, starting at `0` and extending for a given number of characters afterward. |
| 2653 | // 'hello'.limit(2) => 'he' |
| 2654 | // 'hi'.limit(10) => 'hi' |
| 2655 | pub fn (s string) limit(max int) string { |
| 2656 | u := s.runes() |
| 2657 | if u.len <= max { |
| 2658 | return s.clone() |
| 2659 | } |
| 2660 | return u[0..max].string() |
| 2661 | } |
| 2662 | |
| 2663 | // hash returns an integer hash of the string. |
| 2664 | pub fn (s string) hash() int { |
| 2665 | mut h := u32(0) |
| 2666 | if h == 0 && s.len > 0 { |
| 2667 | for c in s { |
| 2668 | h = h * 31 + u32(c) |
| 2669 | } |
| 2670 | } |
| 2671 | return int(h) |
| 2672 | } |
| 2673 | |
| 2674 | // bytes returns the string converted to a byte array. |
| 2675 | pub fn (s string) bytes() []u8 { |
| 2676 | if s.len == 0 { |
| 2677 | return [] |
| 2678 | } |
| 2679 | mut buf := []u8{len: s.len} |
| 2680 | unsafe { vmemcpy(buf.data, s.str, s.len) } |
| 2681 | return buf |
| 2682 | } |
| 2683 | |
| 2684 | // repeat returns a new string with `count` number of copies of the string it was called on. |
| 2685 | @[direct_array_access] |
| 2686 | pub fn (s string) repeat(count int) string { |
| 2687 | if count <= 0 { |
| 2688 | return '' |
| 2689 | } else if count == 1 { |
| 2690 | return s.clone() |
| 2691 | } |
| 2692 | mut ret := unsafe { malloc_noscan(s.len * count + 1) } |
| 2693 | for i in 0 .. count { |
| 2694 | unsafe { |
| 2695 | vmemcpy(ret + i * s.len, s.str, s.len) |
| 2696 | } |
| 2697 | } |
| 2698 | new_len := s.len * count |
| 2699 | unsafe { |
| 2700 | ret[new_len] = 0 |
| 2701 | } |
| 2702 | return unsafe { ret.vstring_with_len(new_len) } |
| 2703 | } |
| 2704 | |
| 2705 | // fields returns a string array of the string split by `\t` and ` ` . |
| 2706 | // Example: assert '\t\tv = v'.fields() == ['v', '=', 'v'] |
| 2707 | // Example: assert ' sss ssss'.fields() == ['sss', 'ssss'] |
| 2708 | pub fn (s string) fields() []string { |
| 2709 | mut res := []string{} |
| 2710 | unsafe { res.flags.set(.noslices) } |
| 2711 | defer { unsafe { res.flags.clear(.noslices) } } |
| 2712 | mut word_start := 0 |
| 2713 | mut word_len := 0 |
| 2714 | mut is_in_word := false |
| 2715 | mut is_space := false |
| 2716 | for i, c in s { |
| 2717 | is_space = c in [32, 9, 10] |
| 2718 | if !is_space { |
| 2719 | word_len++ |
| 2720 | } |
| 2721 | if !is_in_word && !is_space { |
| 2722 | word_start = i |
| 2723 | is_in_word = true |
| 2724 | continue |
| 2725 | } |
| 2726 | if is_space && is_in_word { |
| 2727 | res << s[word_start..word_start + word_len] |
| 2728 | is_in_word = false |
| 2729 | word_len = 0 |
| 2730 | word_start = 0 |
| 2731 | continue |
| 2732 | } |
| 2733 | } |
| 2734 | if is_in_word && word_len > 0 { |
| 2735 | // collect the remainder word at the end |
| 2736 | res << s[word_start..s.len] |
| 2737 | } |
| 2738 | return res |
| 2739 | } |
| 2740 | |
| 2741 | // strip_margin allows multi-line strings to be formatted in a way that removes white-space |
| 2742 | // before a delimiter. By default `|` is used. |
| 2743 | // Note: the delimiter has to be a byte at this time. That means surrounding |
| 2744 | // the value in ``. |
| 2745 | // |
| 2746 | // See also: string.trim_indent() |
| 2747 | // |
| 2748 | // Example: |
| 2749 | // ```v |
| 2750 | // st := 'Hello there, |
| 2751 | // | this is a string, |
| 2752 | // | Everything before the first | is removed'.strip_margin() |
| 2753 | // |
| 2754 | // assert st == 'Hello there, |
| 2755 | // this is a string, |
| 2756 | // Everything before the first | is removed' |
| 2757 | // ``` |
| 2758 | @[inline] |
| 2759 | pub fn (s string) strip_margin() string { |
| 2760 | return s.strip_margin_custom(`|`) |
| 2761 | } |
| 2762 | |
| 2763 | // strip_margin_custom does the same as `strip_margin` but will use `del` as delimiter instead of `|` |
| 2764 | @[direct_array_access] |
| 2765 | pub fn (s string) strip_margin_custom(del u8) string { |
| 2766 | mut sep := del |
| 2767 | if sep.is_space() { |
| 2768 | println('Warning: `strip_margin` cannot use white-space as a delimiter') |
| 2769 | println(' Defaulting to `|`') |
| 2770 | sep = `|` |
| 2771 | } |
| 2772 | // don't know how much space the resulting string will be, but the max it |
| 2773 | // can be is this big |
| 2774 | mut ret := unsafe { malloc_noscan(s.len + 1) } |
| 2775 | mut count := 0 |
| 2776 | for i := 0; i < s.len; i++ { |
| 2777 | if s[i] in [10, 13] { |
| 2778 | unsafe { |
| 2779 | ret[count] = s[i] |
| 2780 | } |
| 2781 | count++ |
| 2782 | // CRLF |
| 2783 | if s[i] == 13 && i < s.len - 1 && s[i + 1] == 10 { |
| 2784 | unsafe { |
| 2785 | ret[count] = s[i + 1] |
| 2786 | } |
| 2787 | count++ |
| 2788 | i++ |
| 2789 | } |
| 2790 | for s[i] != sep { |
| 2791 | i++ |
| 2792 | if i >= s.len { |
| 2793 | break |
| 2794 | } |
| 2795 | } |
| 2796 | } else { |
| 2797 | unsafe { |
| 2798 | ret[count] = s[i] |
| 2799 | } |
| 2800 | count++ |
| 2801 | } |
| 2802 | } |
| 2803 | unsafe { |
| 2804 | ret[count] = 0 |
| 2805 | return ret.vstring_with_len(count) |
| 2806 | } |
| 2807 | } |
| 2808 | |
| 2809 | // trim_indent detects a common minimal indent of all the input lines, |
| 2810 | // removes it from every line and also removes the first and the last |
| 2811 | // lines if they are blank (notice difference blank vs empty). |
| 2812 | // |
| 2813 | // Note that blank lines do not affect the detected indent level. |
| 2814 | // |
| 2815 | // In case if there are non-blank lines with no leading whitespace characters |
| 2816 | // (no indent at all) then the common indent is 0, and therefore this function |
| 2817 | // doesn't change the indentation. |
| 2818 | // |
| 2819 | // Example: |
| 2820 | // ```v |
| 2821 | // st := ' |
| 2822 | // Hello there, |
| 2823 | // this is a string, |
| 2824 | // all the leading indents are removed |
| 2825 | // and also the first and the last lines if they are blank |
| 2826 | // '.trim_indent() |
| 2827 | // |
| 2828 | // assert st == 'Hello there, |
| 2829 | // this is a string, |
| 2830 | // all the leading indents are removed |
| 2831 | // and also the first and the last lines if they are blank' |
| 2832 | // ``` |
| 2833 | pub fn (s string) trim_indent() string { |
| 2834 | mut lines := s.split_into_lines() |
| 2835 | |
| 2836 | mut min_common_indent := int(max_int) // max int |
| 2837 | for line in lines { |
| 2838 | if line.is_blank() { |
| 2839 | continue |
| 2840 | } |
| 2841 | line_indent := line.indent_width() |
| 2842 | if line_indent < min_common_indent { |
| 2843 | min_common_indent = line_indent |
| 2844 | } |
| 2845 | } |
| 2846 | |
| 2847 | // trim first line if it's blank |
| 2848 | if lines.len > 0 && lines.first().is_blank() { |
| 2849 | lines = unsafe { lines[1..] } |
| 2850 | } |
| 2851 | |
| 2852 | // trim last line if it's blank |
| 2853 | if lines.len > 0 && lines.last().is_blank() { |
| 2854 | lines = unsafe { lines[..lines.len - 1] } |
| 2855 | } |
| 2856 | |
| 2857 | mut trimmed_lines := []string{cap: lines.len} |
| 2858 | |
| 2859 | for line in lines { |
| 2860 | if line.is_blank() { |
| 2861 | trimmed_lines << line |
| 2862 | continue |
| 2863 | } |
| 2864 | |
| 2865 | trimmed_lines << line[min_common_indent..] |
| 2866 | } |
| 2867 | |
| 2868 | return trimmed_lines.join('\n') |
| 2869 | } |
| 2870 | |
| 2871 | // indent_width returns the number of spaces or tabs at the beginning of the string. |
| 2872 | // Example: assert ' v'.indent_width() == 2 |
| 2873 | // Example: assert '\t\tv'.indent_width() == 2 |
| 2874 | pub fn (s string) indent_width() int { |
| 2875 | for i, c in s { |
| 2876 | if !c.is_space() { |
| 2877 | return i |
| 2878 | } |
| 2879 | } |
| 2880 | |
| 2881 | return 0 |
| 2882 | } |
| 2883 | |
| 2884 | // is_blank returns true if the string is empty or contains only white-space. |
| 2885 | // Example: assert ' '.is_blank() |
| 2886 | // Example: assert '\t'.is_blank() |
| 2887 | // Example: assert 'v'.is_blank() == false |
| 2888 | pub fn (s string) is_blank() bool { |
| 2889 | if s.len == 0 { |
| 2890 | return true |
| 2891 | } |
| 2892 | |
| 2893 | for c in s { |
| 2894 | if !c.is_space() { |
| 2895 | return false |
| 2896 | } |
| 2897 | } |
| 2898 | |
| 2899 | return true |
| 2900 | } |
| 2901 | |
| 2902 | // match_glob matches the string, with a Unix shell-style wildcard pattern. |
| 2903 | // Note: wildcard patterns are NOT the same as regular expressions. |
| 2904 | // They are much simpler, and do not allow backtracking, captures, etc. |
| 2905 | // The special characters used in shell-style wildcards are: |
| 2906 | // `*` - matches everything |
| 2907 | // `?` - matches any single character |
| 2908 | // `[seq]` - matches any of the characters in the sequence |
| 2909 | // `[^seq]` - matches any character that is NOT in the sequence |
| 2910 | // Any other character in `pattern`, is matched 1:1 to the corresponding |
| 2911 | // character in `name`, including / and \. |
| 2912 | // You can wrap the meta-characters in brackets too, i.e. `[?]` matches `?` |
| 2913 | // in the string, and `[*]` matches `*` in the string. |
| 2914 | // Example: assert 'ABCD'.match_glob('AB*') |
| 2915 | // Example: assert 'ABCD'.match_glob('*D') |
| 2916 | // Example: assert 'ABCD'.match_glob('*B*') |
| 2917 | // Example: assert !'ABCD'.match_glob('AB') |
| 2918 | @[direct_array_access] |
| 2919 | pub fn (name string) match_glob(pattern string) bool { |
| 2920 | // Initial port based on https://research.swtch.com/glob.go |
| 2921 | // See also https://research.swtch.com/glob |
| 2922 | mut px := 0 |
| 2923 | mut nx := 0 |
| 2924 | mut next_px := 0 |
| 2925 | mut next_nx := 0 |
| 2926 | plen := pattern.len |
| 2927 | nlen := name.len |
| 2928 | for px < plen || nx < nlen { |
| 2929 | if px < plen { |
| 2930 | c := pattern[px] |
| 2931 | match c { |
| 2932 | `?` { |
| 2933 | // single-character wildcard |
| 2934 | if nx < nlen { |
| 2935 | px++ |
| 2936 | nx++ |
| 2937 | continue |
| 2938 | } |
| 2939 | } |
| 2940 | `*` { |
| 2941 | // zero-or-more-character wildcard |
| 2942 | // Try to match at nx. |
| 2943 | // If that doesn't work out, restart at nx+1 next. |
| 2944 | next_px = px |
| 2945 | next_nx = nx + 1 |
| 2946 | px++ |
| 2947 | continue |
| 2948 | } |
| 2949 | `[` { |
| 2950 | if nx < nlen { |
| 2951 | wanted_c := name[nx] |
| 2952 | mut is_inverted := false |
| 2953 | mut inner_match := false |
| 2954 | mut inner_idx := px + 1 |
| 2955 | if inner_idx < plen && pattern[inner_idx] == `^` { |
| 2956 | is_inverted = true |
| 2957 | inner_idx++ |
| 2958 | } |
| 2959 | for ; inner_idx < plen && pattern[inner_idx] != `]`; inner_idx++ { |
| 2960 | if pattern[inner_idx] == wanted_c { |
| 2961 | inner_match = true |
| 2962 | } |
| 2963 | } |
| 2964 | if inner_idx < plen && ((inner_match && !is_inverted) |
| 2965 | || (!inner_match && is_inverted)) { |
| 2966 | px = inner_idx + 1 |
| 2967 | nx++ |
| 2968 | continue |
| 2969 | } |
| 2970 | } |
| 2971 | } |
| 2972 | else { |
| 2973 | // an ordinary character |
| 2974 | if nx < nlen && name[nx] == c { |
| 2975 | px++ |
| 2976 | nx++ |
| 2977 | continue |
| 2978 | } |
| 2979 | } |
| 2980 | } |
| 2981 | } |
| 2982 | if 0 < next_nx && next_nx <= nlen { |
| 2983 | // A mismatch, try restarting: |
| 2984 | px = next_px |
| 2985 | nx = next_nx |
| 2986 | continue |
| 2987 | } |
| 2988 | return false |
| 2989 | } |
| 2990 | // Matched all of `pattern` to all of `name` |
| 2991 | return true |
| 2992 | } |
| 2993 | |
| 2994 | // is_ascii returns true if all characters belong to the US-ASCII set ([` `..`~`]) |
| 2995 | @[direct_array_access; inline] |
| 2996 | pub fn (s string) is_ascii() bool { |
| 2997 | for i := 0; i < s.len; i++ { |
| 2998 | if s[i] < u8(` `) || s[i] > u8(`~`) { |
| 2999 | return false |
| 3000 | } |
| 3001 | } |
| 3002 | return true |
| 3003 | } |
| 3004 | |
| 3005 | // is_identifier checks if a string is a valid identifier (starts with letter/underscore, followed by letters, digits, or underscores) |
| 3006 | @[direct_array_access] |
| 3007 | pub fn (s string) is_identifier() bool { |
| 3008 | if s.len == 0 { |
| 3009 | return false |
| 3010 | } |
| 3011 | if !(s[0].is_letter() || s[0] == `_`) { |
| 3012 | return false |
| 3013 | } |
| 3014 | for i := 1; i < s.len; i++ { |
| 3015 | c := s[i] |
| 3016 | if !(c.is_letter() || c.is_digit() || c == `_`) { |
| 3017 | return false |
| 3018 | } |
| 3019 | } |
| 3020 | return true |
| 3021 | } |
| 3022 | |
| 3023 | // camel_to_snake convert string from camelCase to snake_case |
| 3024 | // Example: assert 'Abcd'.camel_to_snake() == 'abcd' |
| 3025 | // Example: assert 'aaBB'.camel_to_snake() == 'aa_bb' |
| 3026 | // Example: assert 'BBaa'.camel_to_snake() == 'bb_aa' |
| 3027 | // Example: assert 'HTTPServer'.camel_to_snake() == 'http_server' |
| 3028 | // Example: assert 'HTTP2Server'.camel_to_snake() == 'http2_server' |
| 3029 | // Example: assert 'XML2JSON'.camel_to_snake() == 'xml_2_json' |
| 3030 | @[direct_array_access] |
| 3031 | pub fn (s string) camel_to_snake() string { |
| 3032 | if s.len == 0 { |
| 3033 | return '' |
| 3034 | } |
| 3035 | if s.len == 1 { |
| 3036 | return s.to_lower_ascii() |
| 3037 | } |
| 3038 | mut b := unsafe { malloc_noscan(2 * s.len + 1) } |
| 3039 | // Rather than checking whether the iterator variable is > 1 inside the loop, |
| 3040 | // handle the first two chars separately to reduce load. |
| 3041 | mut pos := 2 |
| 3042 | mut prev_is_upper := false |
| 3043 | mut prev_inserted_boundary := false |
| 3044 | unsafe { |
| 3045 | if s[0].is_capital() { |
| 3046 | b[0] = s[0] + 32 |
| 3047 | b[1] = if s[1].is_capital() { |
| 3048 | prev_is_upper = true |
| 3049 | s[1] + 32 |
| 3050 | } else { |
| 3051 | s[1] |
| 3052 | } |
| 3053 | } else { |
| 3054 | b[0] = s[0] |
| 3055 | if s[1].is_capital() { |
| 3056 | prev_is_upper = true |
| 3057 | if s[0] != `_` && s.len > 2 && !s[2].is_capital() { |
| 3058 | b[1] = `_` |
| 3059 | b[2] = s[1] + 32 |
| 3060 | pos = 3 |
| 3061 | } else { |
| 3062 | b[1] = s[1] + 32 |
| 3063 | } |
| 3064 | } else { |
| 3065 | b[1] = s[1] |
| 3066 | } |
| 3067 | } |
| 3068 | } |
| 3069 | for i := 2; i < s.len; i++ { |
| 3070 | mut has_boundary_before_upper := false |
| 3071 | c := s[i] |
| 3072 | c_is_upper := c.is_capital() |
| 3073 | c_is_number := c.is_digit() |
| 3074 | next_is_lower := i + 1 < s.len && s[i + 1].is_letter() && !s[i + 1].is_capital() |
| 3075 | next2_is_lower := i + 2 < s.len && s[i + 2].is_letter() && !s[i + 2].is_capital() |
| 3076 | // Cases: `XML2JSON == xml_2_json` || `HTTP2Server == http2_server` |
| 3077 | skip_digit := c_is_number && prev_is_upper && !next_is_lower && next2_is_lower |
| 3078 | // Cases: `HTTPServer == http_server` || `getHTTPSUrl == get_https_url` |
| 3079 | if c_is_upper && prev_is_upper && i >= 2 && s[i - 2].is_capital() && next_is_lower |
| 3080 | && c != `_` { |
| 3081 | unsafe { |
| 3082 | if b[pos - 1] != `_` { |
| 3083 | b[pos] = `_` |
| 3084 | pos++ |
| 3085 | } |
| 3086 | } |
| 3087 | has_boundary_before_upper = true |
| 3088 | } |
| 3089 | // Cases: `aBcd == a_bcd` || `ABcd == ab_cd` |
| 3090 | // TODO: remove this workaround for v2's parser |
| 3091 | // vfmt off |
| 3092 | if ((c_is_upper && !prev_is_upper) || |
| 3093 | (!c_is_upper && prev_is_upper && s[i - 2].is_capital() && !prev_inserted_boundary && !skip_digit)) |
| 3094 | && c != `_` { |
| 3095 | unsafe { |
| 3096 | if b[pos - 1] != `_` { |
| 3097 | b[pos] = `_` |
| 3098 | pos++ |
| 3099 | } |
| 3100 | } |
| 3101 | } |
| 3102 | // vfmt on |
| 3103 | lower_c := if c_is_upper { c + 32 } else { c } |
| 3104 | unsafe { |
| 3105 | b[pos] = lower_c |
| 3106 | } |
| 3107 | prev_is_upper = c_is_upper |
| 3108 | prev_inserted_boundary = has_boundary_before_upper |
| 3109 | pos++ |
| 3110 | } |
| 3111 | unsafe { |
| 3112 | b[pos] = 0 |
| 3113 | } |
| 3114 | return unsafe { tos(b, pos) } |
| 3115 | } |
| 3116 | |
| 3117 | // snake_to_camel convert string from snake_case to camelCase |
| 3118 | // Example: assert 'abcd'.snake_to_camel() == 'Abcd' |
| 3119 | // Example: assert 'ab_cd'.snake_to_camel() == 'AbCd' |
| 3120 | // Example: assert '_abcd'.snake_to_camel() == 'Abcd' |
| 3121 | // Example: assert '_abcd_'.snake_to_camel() == 'Abcd' |
| 3122 | @[direct_array_access] |
| 3123 | pub fn (s string) snake_to_camel() string { |
| 3124 | if s.len == 0 { |
| 3125 | return '' |
| 3126 | } |
| 3127 | if s.len == 1 { |
| 3128 | return s |
| 3129 | } |
| 3130 | mut need_upper := true |
| 3131 | mut upper_c := `_` |
| 3132 | mut b := unsafe { malloc_noscan(s.len + 1) } |
| 3133 | mut i := 0 |
| 3134 | for c in s { |
| 3135 | upper_c = if c >= `a` && c <= `z` { c - 32 } else { c } |
| 3136 | if c == `_` { |
| 3137 | need_upper = true |
| 3138 | } else if need_upper { |
| 3139 | unsafe { |
| 3140 | b[i] = upper_c |
| 3141 | } |
| 3142 | i++ |
| 3143 | need_upper = false |
| 3144 | } else { |
| 3145 | unsafe { |
| 3146 | b[i] = c |
| 3147 | } |
| 3148 | i++ |
| 3149 | } |
| 3150 | } |
| 3151 | unsafe { |
| 3152 | b[i] = 0 |
| 3153 | } |
| 3154 | return unsafe { tos(b, i) } |
| 3155 | } |
| 3156 | |
| 3157 | @[params] |
| 3158 | pub struct WrapConfig { |
| 3159 | pub: |
| 3160 | width int = 80 |
| 3161 | end string = '\n' |
| 3162 | } |
| 3163 | |
| 3164 | // wrap wraps the string `s` when each line exceeds the width specified in `width` . |
| 3165 | // (default value is 80), and will use `end` (default value is '\n') as a line break. |
| 3166 | // Example: assert 'Hello, my name is Carl and I am a delivery'.wrap(width: 20) == 'Hello, my name is\nCarl and I am a\ndelivery' |
| 3167 | pub fn (s string) wrap(config WrapConfig) string { |
| 3168 | if config.width <= 0 { |
| 3169 | return '' |
| 3170 | } |
| 3171 | words := s.fields() |
| 3172 | if words.len == 0 { |
| 3173 | return '' |
| 3174 | } |
| 3175 | mut sb := strings.new_builder(s.len) |
| 3176 | sb.write_string(words[0]) |
| 3177 | mut space_left := config.width - words[0].len |
| 3178 | for i in 1 .. words.len { |
| 3179 | word := words[i] |
| 3180 | if word.len + 1 > space_left { |
| 3181 | sb.write_string(config.end) |
| 3182 | sb.write_string(word) |
| 3183 | space_left = config.width - word.len |
| 3184 | } else { |
| 3185 | sb.write_string(' ') |
| 3186 | sb.write_string(word) |
| 3187 | space_left -= 1 + word.len |
| 3188 | } |
| 3189 | } |
| 3190 | return sb.str() |
| 3191 | } |
| 3192 | |
| 3193 | // hex returns a string with the hexadecimal representation of the bytes of the string `s` . |
| 3194 | pub fn (s string) hex() string { |
| 3195 | if s == '' { |
| 3196 | return '' |
| 3197 | } |
| 3198 | return unsafe { data_to_hex_string(s.str, s.len) } |
| 3199 | } |
| 3200 | |
| 3201 | @[unsafe] |
| 3202 | fn data_to_hex_string(data &u8, len int) string { |
| 3203 | mut hex := malloc_noscan(u64(len) * 2 + 1) |
| 3204 | mut dst := 0 |
| 3205 | for c in 0 .. len { |
| 3206 | b := data[c] |
| 3207 | n0 := b >> 4 |
| 3208 | n1 := b & 0xF |
| 3209 | hex[dst] = if n0 < 10 { n0 + `0` } else { n0 + `W` } |
| 3210 | hex[dst + 1] = if n1 < 10 { n1 + `0` } else { n1 + `W` } |
| 3211 | dst += 2 |
| 3212 | } |
| 3213 | hex[dst] = 0 |
| 3214 | return tos(hex, dst) |
| 3215 | } |
| 3216 | |
| 3217 | pub struct RunesIterator { |
| 3218 | mut: |
| 3219 | s string |
| 3220 | i int |
| 3221 | } |
| 3222 | |
| 3223 | // runes_iterator creates an iterator over all the runes in the given string `s`. |
| 3224 | // It can be used in `for r in s.runes_iterator() {`, as a direct substitute to |
| 3225 | // calling .runes(): `for r in s.runes() {`, which needs an intermediate allocation |
| 3226 | // of an array. |
| 3227 | pub fn (s string) runes_iterator() RunesIterator { |
| 3228 | return RunesIterator{ |
| 3229 | s: s |
| 3230 | i: 0 |
| 3231 | } |
| 3232 | } |
| 3233 | |
| 3234 | // next is the method that will be called for each iteration in `for r in s.runes_iterator() {` . |
| 3235 | pub fn (mut ri RunesIterator) next() ?rune { |
| 3236 | if ri.i >= ri.s.len { |
| 3237 | return none |
| 3238 | } |
| 3239 | char_len := utf8_char_len(unsafe { ri.s.str[ri.i] }) |
| 3240 | if char_len == 1 { |
| 3241 | res := unsafe { ri.s.str[ri.i] } |
| 3242 | ri.i++ |
| 3243 | return res |
| 3244 | } |
| 3245 | start := &u8(unsafe { &ri.s.str[ri.i] }) |
| 3246 | len := if ri.s.len - 1 >= ri.i + char_len { char_len } else { ri.s.len - ri.i } |
| 3247 | ri.i += char_len |
| 3248 | if char_len > 4 { |
| 3249 | return 0 |
| 3250 | } |
| 3251 | return rune(impl_utf8_to_utf32(start, len)) |
| 3252 | } |
| 3253 | |