1 | /**************************************************************************/ |
2 | /* ustring.cpp */ |
3 | /**************************************************************************/ |
4 | /* This file is part of: */ |
5 | /* GODOT ENGINE */ |
6 | /* https://godotengine.org */ |
7 | /**************************************************************************/ |
8 | /* Copyright (c) 2014-present Godot Engine contributors (see AUTHORS.md). */ |
9 | /* Copyright (c) 2007-2014 Juan Linietsky, Ariel Manzur. */ |
10 | /* */ |
11 | /* Permission is hereby granted, free of charge, to any person obtaining */ |
12 | /* a copy of this software and associated documentation files (the */ |
13 | /* "Software"), to deal in the Software without restriction, including */ |
14 | /* without limitation the rights to use, copy, modify, merge, publish, */ |
15 | /* distribute, sublicense, and/or sell copies of the Software, and to */ |
16 | /* permit persons to whom the Software is furnished to do so, subject to */ |
17 | /* the following conditions: */ |
18 | /* */ |
19 | /* The above copyright notice and this permission notice shall be */ |
20 | /* included in all copies or substantial portions of the Software. */ |
21 | /* */ |
22 | /* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */ |
23 | /* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */ |
24 | /* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. */ |
25 | /* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */ |
26 | /* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */ |
27 | /* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */ |
28 | /* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ |
29 | /**************************************************************************/ |
30 | |
31 | #include "ustring.h" |
32 | |
33 | #include "core/crypto/crypto_core.h" |
34 | #include "core/math/color.h" |
35 | #include "core/math/math_funcs.h" |
36 | #include "core/os/memory.h" |
37 | #include "core/string/print_string.h" |
38 | #include "core/string/string_name.h" |
39 | #include "core/string/translation.h" |
40 | #include "core/string/ucaps.h" |
41 | #include "core/variant/variant.h" |
42 | #include "core/version_generated.gen.h" |
43 | |
44 | #include <stdio.h> |
45 | #include <stdlib.h> |
46 | #include <cstdint> |
47 | |
48 | #ifdef _MSC_VER |
49 | #define _CRT_SECURE_NO_WARNINGS // to disable build-time warning which suggested to use strcpy_s instead strcpy |
50 | #endif |
51 | |
52 | #if defined(MINGW_ENABLED) || defined(_MSC_VER) |
53 | #define snprintf _snprintf_s |
54 | #endif |
55 | |
56 | static const int MAX_DECIMALS = 32; |
57 | |
58 | static _FORCE_INLINE_ char32_t lower_case(char32_t c) { |
59 | return (is_ascii_upper_case(c) ? (c + ('a' - 'A')) : c); |
60 | } |
61 | |
62 | const char CharString::_null = 0; |
63 | const char16_t Char16String::_null = 0; |
64 | const char32_t String::_null = 0; |
65 | const char32_t String::_replacement_char = 0xfffd; |
66 | |
67 | bool select_word(const String &p_s, int p_col, int &r_beg, int &r_end) { |
68 | const String &s = p_s; |
69 | int beg = CLAMP(p_col, 0, s.length()); |
70 | int end = beg; |
71 | |
72 | if (s[beg] > 32 || beg == s.length()) { |
73 | bool symbol = beg < s.length() && is_symbol(s[beg]); |
74 | |
75 | while (beg > 0 && s[beg - 1] > 32 && (symbol == is_symbol(s[beg - 1]))) { |
76 | beg--; |
77 | } |
78 | while (end < s.length() && s[end + 1] > 32 && (symbol == is_symbol(s[end + 1]))) { |
79 | end++; |
80 | } |
81 | |
82 | if (end < s.length()) { |
83 | end += 1; |
84 | } |
85 | |
86 | r_beg = beg; |
87 | r_end = end; |
88 | |
89 | return true; |
90 | } else { |
91 | return false; |
92 | } |
93 | } |
94 | |
95 | /*************************************************************************/ |
96 | /* Char16String */ |
97 | /*************************************************************************/ |
98 | |
99 | bool Char16String::operator<(const Char16String &p_right) const { |
100 | if (length() == 0) { |
101 | return p_right.length() != 0; |
102 | } |
103 | |
104 | return is_str_less(get_data(), p_right.get_data()); |
105 | } |
106 | |
107 | Char16String &Char16String::operator+=(char16_t p_char) { |
108 | const int lhs_len = length(); |
109 | resize(lhs_len + 2); |
110 | |
111 | char16_t *dst = ptrw(); |
112 | dst[lhs_len] = p_char; |
113 | dst[lhs_len + 1] = 0; |
114 | |
115 | return *this; |
116 | } |
117 | |
118 | void Char16String::operator=(const char16_t *p_cstr) { |
119 | copy_from(p_cstr); |
120 | } |
121 | |
122 | const char16_t *Char16String::get_data() const { |
123 | if (size()) { |
124 | return &operator[](0); |
125 | } else { |
126 | return u"" ; |
127 | } |
128 | } |
129 | |
130 | void Char16String::copy_from(const char16_t *p_cstr) { |
131 | if (!p_cstr) { |
132 | resize(0); |
133 | return; |
134 | } |
135 | |
136 | const char16_t *s = p_cstr; |
137 | for (; *s; s++) { |
138 | } |
139 | size_t len = s - p_cstr; |
140 | |
141 | if (len == 0) { |
142 | resize(0); |
143 | return; |
144 | } |
145 | |
146 | Error err = resize(++len); // include terminating null char |
147 | |
148 | ERR_FAIL_COND_MSG(err != OK, "Failed to copy char16_t string." ); |
149 | |
150 | memcpy(ptrw(), p_cstr, len * sizeof(char16_t)); |
151 | } |
152 | |
153 | /*************************************************************************/ |
154 | /* CharString */ |
155 | /*************************************************************************/ |
156 | |
157 | bool CharString::operator<(const CharString &p_right) const { |
158 | if (length() == 0) { |
159 | return p_right.length() != 0; |
160 | } |
161 | |
162 | return is_str_less(get_data(), p_right.get_data()); |
163 | } |
164 | |
165 | bool CharString::operator==(const CharString &p_right) const { |
166 | if (length() == 0) { |
167 | // True if both have length 0, false if only p_right has a length |
168 | return p_right.length() == 0; |
169 | } else if (p_right.length() == 0) { |
170 | // False due to unequal length |
171 | return false; |
172 | } |
173 | |
174 | return strcmp(ptr(), p_right.ptr()) == 0; |
175 | } |
176 | |
177 | CharString &CharString::operator+=(char p_char) { |
178 | const int lhs_len = length(); |
179 | resize(lhs_len + 2); |
180 | |
181 | char *dst = ptrw(); |
182 | dst[lhs_len] = p_char; |
183 | dst[lhs_len + 1] = 0; |
184 | |
185 | return *this; |
186 | } |
187 | |
188 | void CharString::operator=(const char *p_cstr) { |
189 | copy_from(p_cstr); |
190 | } |
191 | |
192 | const char *CharString::get_data() const { |
193 | if (size()) { |
194 | return &operator[](0); |
195 | } else { |
196 | return "" ; |
197 | } |
198 | } |
199 | |
200 | void CharString::copy_from(const char *p_cstr) { |
201 | if (!p_cstr) { |
202 | resize(0); |
203 | return; |
204 | } |
205 | |
206 | size_t len = strlen(p_cstr); |
207 | |
208 | if (len == 0) { |
209 | resize(0); |
210 | return; |
211 | } |
212 | |
213 | Error err = resize(++len); // include terminating null char |
214 | |
215 | ERR_FAIL_COND_MSG(err != OK, "Failed to copy C-string." ); |
216 | |
217 | memcpy(ptrw(), p_cstr, len); |
218 | } |
219 | |
220 | /*************************************************************************/ |
221 | /* String */ |
222 | /*************************************************************************/ |
223 | |
224 | Error String::parse_url(String &r_scheme, String &r_host, int &r_port, String &r_path) const { |
225 | // Splits the URL into scheme, host, port, path. Strip credentials when present. |
226 | String base = *this; |
227 | r_scheme = "" ; |
228 | r_host = "" ; |
229 | r_port = 0; |
230 | r_path = "" ; |
231 | int pos = base.find("://" ); |
232 | // Scheme |
233 | if (pos != -1) { |
234 | r_scheme = base.substr(0, pos + 3).to_lower(); |
235 | base = base.substr(pos + 3, base.length() - pos - 3); |
236 | } |
237 | pos = base.find("/" ); |
238 | // Path |
239 | if (pos != -1) { |
240 | r_path = base.substr(pos, base.length() - pos); |
241 | base = base.substr(0, pos); |
242 | } |
243 | // Host |
244 | pos = base.find("@" ); |
245 | if (pos != -1) { |
246 | // Strip credentials |
247 | base = base.substr(pos + 1, base.length() - pos - 1); |
248 | } |
249 | if (base.begins_with("[" )) { |
250 | // Literal IPv6 |
251 | pos = base.rfind("]" ); |
252 | if (pos == -1) { |
253 | return ERR_INVALID_PARAMETER; |
254 | } |
255 | r_host = base.substr(1, pos - 1); |
256 | base = base.substr(pos + 1, base.length() - pos - 1); |
257 | } else { |
258 | // Anything else |
259 | if (base.get_slice_count(":" ) > 2) { |
260 | return ERR_INVALID_PARAMETER; |
261 | } |
262 | pos = base.rfind(":" ); |
263 | if (pos == -1) { |
264 | r_host = base; |
265 | base = "" ; |
266 | } else { |
267 | r_host = base.substr(0, pos); |
268 | base = base.substr(pos, base.length() - pos); |
269 | } |
270 | } |
271 | if (r_host.is_empty()) { |
272 | return ERR_INVALID_PARAMETER; |
273 | } |
274 | r_host = r_host.to_lower(); |
275 | // Port |
276 | if (base.begins_with(":" )) { |
277 | base = base.substr(1, base.length() - 1); |
278 | if (!base.is_valid_int()) { |
279 | return ERR_INVALID_PARAMETER; |
280 | } |
281 | r_port = base.to_int(); |
282 | if (r_port < 1 || r_port > 65535) { |
283 | return ERR_INVALID_PARAMETER; |
284 | } |
285 | } |
286 | return OK; |
287 | } |
288 | |
289 | void String::copy_from(const char *p_cstr) { |
290 | // copy Latin-1 encoded c-string directly |
291 | if (!p_cstr) { |
292 | resize(0); |
293 | return; |
294 | } |
295 | |
296 | const size_t len = strlen(p_cstr); |
297 | |
298 | if (len == 0) { |
299 | resize(0); |
300 | return; |
301 | } |
302 | |
303 | resize(len + 1); // include 0 |
304 | |
305 | char32_t *dst = this->ptrw(); |
306 | |
307 | for (size_t i = 0; i <= len; i++) { |
308 | uint8_t c = p_cstr[i] >= 0 ? p_cstr[i] : uint8_t(256 + p_cstr[i]); |
309 | if (c == 0 && i < len) { |
310 | print_unicode_error("NUL character" , true); |
311 | dst[i] = _replacement_char; |
312 | } else { |
313 | dst[i] = c; |
314 | } |
315 | } |
316 | } |
317 | |
318 | void String::copy_from(const char *p_cstr, const int p_clip_to) { |
319 | // copy Latin-1 encoded c-string directly |
320 | if (!p_cstr) { |
321 | resize(0); |
322 | return; |
323 | } |
324 | |
325 | int len = 0; |
326 | const char *ptr = p_cstr; |
327 | while ((p_clip_to < 0 || len < p_clip_to) && *(ptr++) != 0) { |
328 | len++; |
329 | } |
330 | |
331 | if (len == 0) { |
332 | resize(0); |
333 | return; |
334 | } |
335 | |
336 | resize(len + 1); // include 0 |
337 | |
338 | char32_t *dst = this->ptrw(); |
339 | |
340 | for (int i = 0; i < len; i++) { |
341 | uint8_t c = p_cstr[i] >= 0 ? p_cstr[i] : uint8_t(256 + p_cstr[i]); |
342 | if (c == 0) { |
343 | print_unicode_error("NUL character" , true); |
344 | dst[i] = _replacement_char; |
345 | } else { |
346 | dst[i] = c; |
347 | } |
348 | } |
349 | dst[len] = 0; |
350 | } |
351 | |
352 | void String::copy_from(const wchar_t *p_cstr) { |
353 | #ifdef WINDOWS_ENABLED |
354 | // wchar_t is 16-bit, parse as UTF-16 |
355 | parse_utf16((const char16_t *)p_cstr); |
356 | #else |
357 | // wchar_t is 32-bit, copy directly |
358 | copy_from((const char32_t *)p_cstr); |
359 | #endif |
360 | } |
361 | |
362 | void String::copy_from(const wchar_t *p_cstr, const int p_clip_to) { |
363 | #ifdef WINDOWS_ENABLED |
364 | // wchar_t is 16-bit, parse as UTF-16 |
365 | parse_utf16((const char16_t *)p_cstr, p_clip_to); |
366 | #else |
367 | // wchar_t is 32-bit, copy directly |
368 | copy_from((const char32_t *)p_cstr, p_clip_to); |
369 | #endif |
370 | } |
371 | |
372 | void String::copy_from(const char32_t &p_char) { |
373 | if (p_char == 0) { |
374 | print_unicode_error("NUL character" , true); |
375 | return; |
376 | } |
377 | |
378 | resize(2); |
379 | |
380 | char32_t *dst = ptrw(); |
381 | |
382 | if ((p_char & 0xfffff800) == 0xd800) { |
383 | print_unicode_error(vformat("Unpaired surrogate (%x)" , (uint32_t)p_char)); |
384 | dst[0] = _replacement_char; |
385 | } else if (p_char > 0x10ffff) { |
386 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , (uint32_t)p_char)); |
387 | dst[0] = _replacement_char; |
388 | } else { |
389 | dst[0] = p_char; |
390 | } |
391 | |
392 | dst[1] = 0; |
393 | } |
394 | |
395 | void String::copy_from(const char32_t *p_cstr) { |
396 | if (!p_cstr) { |
397 | resize(0); |
398 | return; |
399 | } |
400 | |
401 | int len = 0; |
402 | const char32_t *ptr = p_cstr; |
403 | while (*(ptr++) != 0) { |
404 | len++; |
405 | } |
406 | |
407 | if (len == 0) { |
408 | resize(0); |
409 | return; |
410 | } |
411 | |
412 | copy_from_unchecked(p_cstr, len); |
413 | } |
414 | |
415 | void String::copy_from(const char32_t *p_cstr, const int p_clip_to) { |
416 | if (!p_cstr) { |
417 | resize(0); |
418 | return; |
419 | } |
420 | |
421 | int len = 0; |
422 | const char32_t *ptr = p_cstr; |
423 | while ((p_clip_to < 0 || len < p_clip_to) && *(ptr++) != 0) { |
424 | len++; |
425 | } |
426 | |
427 | if (len == 0) { |
428 | resize(0); |
429 | return; |
430 | } |
431 | |
432 | copy_from_unchecked(p_cstr, len); |
433 | } |
434 | |
435 | // assumes the following have already been validated: |
436 | // p_char != nullptr |
437 | // p_length > 0 |
438 | // p_length <= p_char strlen |
439 | void String::copy_from_unchecked(const char32_t *p_char, const int p_length) { |
440 | resize(p_length + 1); |
441 | char32_t *dst = ptrw(); |
442 | dst[p_length] = 0; |
443 | |
444 | for (int i = 0; i < p_length; i++) { |
445 | if (p_char[i] == 0) { |
446 | print_unicode_error("NUL character" , true); |
447 | dst[i] = _replacement_char; |
448 | continue; |
449 | } |
450 | if ((p_char[i] & 0xfffff800) == 0xd800) { |
451 | print_unicode_error(vformat("Unpaired surrogate (%x)" , (uint32_t)p_char[i])); |
452 | dst[i] = _replacement_char; |
453 | continue; |
454 | } |
455 | if (p_char[i] > 0x10ffff) { |
456 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , (uint32_t)p_char[i])); |
457 | dst[i] = _replacement_char; |
458 | continue; |
459 | } |
460 | dst[i] = p_char[i]; |
461 | } |
462 | } |
463 | |
464 | void String::operator=(const char *p_str) { |
465 | copy_from(p_str); |
466 | } |
467 | |
468 | void String::operator=(const char32_t *p_str) { |
469 | copy_from(p_str); |
470 | } |
471 | |
472 | void String::operator=(const wchar_t *p_str) { |
473 | copy_from(p_str); |
474 | } |
475 | |
476 | String String::operator+(const String &p_str) const { |
477 | String res = *this; |
478 | res += p_str; |
479 | return res; |
480 | } |
481 | |
482 | String String::operator+(char32_t p_char) const { |
483 | String res = *this; |
484 | res += p_char; |
485 | return res; |
486 | } |
487 | |
488 | String operator+(const char *p_chr, const String &p_str) { |
489 | String tmp = p_chr; |
490 | tmp += p_str; |
491 | return tmp; |
492 | } |
493 | |
494 | String operator+(const wchar_t *p_chr, const String &p_str) { |
495 | #ifdef WINDOWS_ENABLED |
496 | // wchar_t is 16-bit |
497 | String tmp = String::utf16((const char16_t *)p_chr); |
498 | #else |
499 | // wchar_t is 32-bit |
500 | String tmp = (const char32_t *)p_chr; |
501 | #endif |
502 | tmp += p_str; |
503 | return tmp; |
504 | } |
505 | |
506 | String operator+(char32_t p_chr, const String &p_str) { |
507 | return (String::chr(p_chr) + p_str); |
508 | } |
509 | |
510 | String &String::operator+=(const String &p_str) { |
511 | const int lhs_len = length(); |
512 | if (lhs_len == 0) { |
513 | *this = p_str; |
514 | return *this; |
515 | } |
516 | |
517 | const int rhs_len = p_str.length(); |
518 | if (rhs_len == 0) { |
519 | return *this; |
520 | } |
521 | |
522 | resize(lhs_len + rhs_len + 1); |
523 | |
524 | const char32_t *src = p_str.ptr(); |
525 | char32_t *dst = ptrw() + lhs_len; |
526 | |
527 | // Don't copy the terminating null with `memcpy` to avoid undefined behavior when string is being added to itself (it would overlap the destination). |
528 | memcpy(dst, src, rhs_len * sizeof(char32_t)); |
529 | *(dst + rhs_len) = _null; |
530 | |
531 | return *this; |
532 | } |
533 | |
534 | String &String::operator+=(const char *p_str) { |
535 | if (!p_str || p_str[0] == 0) { |
536 | return *this; |
537 | } |
538 | |
539 | const int lhs_len = length(); |
540 | const size_t rhs_len = strlen(p_str); |
541 | |
542 | resize(lhs_len + rhs_len + 1); |
543 | |
544 | char32_t *dst = ptrw() + lhs_len; |
545 | |
546 | for (size_t i = 0; i <= rhs_len; i++) { |
547 | uint8_t c = p_str[i] >= 0 ? p_str[i] : uint8_t(256 + p_str[i]); |
548 | if (c == 0 && i < rhs_len) { |
549 | print_unicode_error("NUL character" , true); |
550 | dst[i] = _replacement_char; |
551 | } else { |
552 | dst[i] = c; |
553 | } |
554 | } |
555 | |
556 | return *this; |
557 | } |
558 | |
559 | String &String::operator+=(const wchar_t *p_str) { |
560 | #ifdef WINDOWS_ENABLED |
561 | // wchar_t is 16-bit |
562 | *this += String::utf16((const char16_t *)p_str); |
563 | #else |
564 | // wchar_t is 32-bit |
565 | *this += String((const char32_t *)p_str); |
566 | #endif |
567 | return *this; |
568 | } |
569 | |
570 | String &String::operator+=(const char32_t *p_str) { |
571 | *this += String(p_str); |
572 | return *this; |
573 | } |
574 | |
575 | String &String::operator+=(char32_t p_char) { |
576 | if (p_char == 0) { |
577 | print_unicode_error("NUL character" , true); |
578 | return *this; |
579 | } |
580 | |
581 | const int lhs_len = length(); |
582 | resize(lhs_len + 2); |
583 | char32_t *dst = ptrw(); |
584 | |
585 | if ((p_char & 0xfffff800) == 0xd800) { |
586 | print_unicode_error(vformat("Unpaired surrogate (%x)" , (uint32_t)p_char)); |
587 | dst[lhs_len] = _replacement_char; |
588 | } else if (p_char > 0x10ffff) { |
589 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , (uint32_t)p_char)); |
590 | dst[lhs_len] = _replacement_char; |
591 | } else { |
592 | dst[lhs_len] = p_char; |
593 | } |
594 | |
595 | dst[lhs_len + 1] = 0; |
596 | |
597 | return *this; |
598 | } |
599 | |
600 | bool String::operator==(const char *p_str) const { |
601 | // compare Latin-1 encoded c-string |
602 | int len = 0; |
603 | const char *aux = p_str; |
604 | |
605 | while (*(aux++) != 0) { |
606 | len++; |
607 | } |
608 | |
609 | if (length() != len) { |
610 | return false; |
611 | } |
612 | if (is_empty()) { |
613 | return true; |
614 | } |
615 | |
616 | int l = length(); |
617 | |
618 | const char32_t *dst = get_data(); |
619 | |
620 | // Compare char by char |
621 | for (int i = 0; i < l; i++) { |
622 | if ((char32_t)p_str[i] != dst[i]) { |
623 | return false; |
624 | } |
625 | } |
626 | |
627 | return true; |
628 | } |
629 | |
630 | bool String::operator==(const wchar_t *p_str) const { |
631 | #ifdef WINDOWS_ENABLED |
632 | // wchar_t is 16-bit, parse as UTF-16 |
633 | return *this == String::utf16((const char16_t *)p_str); |
634 | #else |
635 | // wchar_t is 32-bit, compare char by char |
636 | return *this == (const char32_t *)p_str; |
637 | #endif |
638 | } |
639 | |
640 | bool String::operator==(const char32_t *p_str) const { |
641 | int len = 0; |
642 | const char32_t *aux = p_str; |
643 | |
644 | while (*(aux++) != 0) { |
645 | len++; |
646 | } |
647 | |
648 | if (length() != len) { |
649 | return false; |
650 | } |
651 | if (is_empty()) { |
652 | return true; |
653 | } |
654 | |
655 | int l = length(); |
656 | |
657 | const char32_t *dst = get_data(); |
658 | |
659 | /* Compare char by char */ |
660 | for (int i = 0; i < l; i++) { |
661 | if (p_str[i] != dst[i]) { |
662 | return false; |
663 | } |
664 | } |
665 | |
666 | return true; |
667 | } |
668 | |
669 | bool String::operator==(const String &p_str) const { |
670 | if (length() != p_str.length()) { |
671 | return false; |
672 | } |
673 | if (is_empty()) { |
674 | return true; |
675 | } |
676 | |
677 | int l = length(); |
678 | |
679 | const char32_t *src = get_data(); |
680 | const char32_t *dst = p_str.get_data(); |
681 | |
682 | /* Compare char by char */ |
683 | for (int i = 0; i < l; i++) { |
684 | if (src[i] != dst[i]) { |
685 | return false; |
686 | } |
687 | } |
688 | |
689 | return true; |
690 | } |
691 | |
692 | bool String::operator==(const StrRange &p_str_range) const { |
693 | int len = p_str_range.len; |
694 | |
695 | if (length() != len) { |
696 | return false; |
697 | } |
698 | if (is_empty()) { |
699 | return true; |
700 | } |
701 | |
702 | const char32_t *c_str = p_str_range.c_str; |
703 | const char32_t *dst = &operator[](0); |
704 | |
705 | /* Compare char by char */ |
706 | for (int i = 0; i < len; i++) { |
707 | if (c_str[i] != dst[i]) { |
708 | return false; |
709 | } |
710 | } |
711 | |
712 | return true; |
713 | } |
714 | |
715 | bool operator==(const char *p_chr, const String &p_str) { |
716 | return p_str == p_chr; |
717 | } |
718 | |
719 | bool operator==(const wchar_t *p_chr, const String &p_str) { |
720 | #ifdef WINDOWS_ENABLED |
721 | // wchar_t is 16-bit |
722 | return p_str == String::utf16((const char16_t *)p_chr); |
723 | #else |
724 | // wchar_t is 32-bi |
725 | return p_str == String((const char32_t *)p_chr); |
726 | #endif |
727 | } |
728 | |
729 | bool operator!=(const char *p_chr, const String &p_str) { |
730 | return !(p_str == p_chr); |
731 | } |
732 | |
733 | bool operator!=(const wchar_t *p_chr, const String &p_str) { |
734 | #ifdef WINDOWS_ENABLED |
735 | // wchar_t is 16-bit |
736 | return !(p_str == String::utf16((const char16_t *)p_chr)); |
737 | #else |
738 | // wchar_t is 32-bi |
739 | return !(p_str == String((const char32_t *)p_chr)); |
740 | #endif |
741 | } |
742 | |
743 | bool String::operator!=(const char *p_str) const { |
744 | return (!(*this == p_str)); |
745 | } |
746 | |
747 | bool String::operator!=(const wchar_t *p_str) const { |
748 | return (!(*this == p_str)); |
749 | } |
750 | |
751 | bool String::operator!=(const char32_t *p_str) const { |
752 | return (!(*this == p_str)); |
753 | } |
754 | |
755 | bool String::operator!=(const String &p_str) const { |
756 | return !((*this == p_str)); |
757 | } |
758 | |
759 | bool String::operator<=(const String &p_str) const { |
760 | return !(p_str < *this); |
761 | } |
762 | |
763 | bool String::operator>(const String &p_str) const { |
764 | return p_str < *this; |
765 | } |
766 | |
767 | bool String::operator>=(const String &p_str) const { |
768 | return !(*this < p_str); |
769 | } |
770 | |
771 | bool String::operator<(const char *p_str) const { |
772 | if (is_empty() && p_str[0] == 0) { |
773 | return false; |
774 | } |
775 | if (is_empty()) { |
776 | return true; |
777 | } |
778 | return is_str_less(get_data(), p_str); |
779 | } |
780 | |
781 | bool String::operator<(const wchar_t *p_str) const { |
782 | if (is_empty() && p_str[0] == 0) { |
783 | return false; |
784 | } |
785 | if (is_empty()) { |
786 | return true; |
787 | } |
788 | |
789 | #ifdef WINDOWS_ENABLED |
790 | // wchar_t is 16-bit |
791 | return is_str_less(get_data(), String::utf16((const char16_t *)p_str).get_data()); |
792 | #else |
793 | // wchar_t is 32-bit |
794 | return is_str_less(get_data(), (const char32_t *)p_str); |
795 | #endif |
796 | } |
797 | |
798 | bool String::operator<(const char32_t *p_str) const { |
799 | if (is_empty() && p_str[0] == 0) { |
800 | return false; |
801 | } |
802 | if (is_empty()) { |
803 | return true; |
804 | } |
805 | |
806 | return is_str_less(get_data(), p_str); |
807 | } |
808 | |
809 | bool String::operator<(const String &p_str) const { |
810 | return operator<(p_str.get_data()); |
811 | } |
812 | |
813 | signed char String::nocasecmp_to(const String &p_str) const { |
814 | if (is_empty() && p_str.is_empty()) { |
815 | return 0; |
816 | } |
817 | if (is_empty()) { |
818 | return -1; |
819 | } |
820 | if (p_str.is_empty()) { |
821 | return 1; |
822 | } |
823 | |
824 | const char32_t *that_str = p_str.get_data(); |
825 | const char32_t *this_str = get_data(); |
826 | |
827 | while (true) { |
828 | if (*that_str == 0 && *this_str == 0) { // If both strings are at the end, they are equal. |
829 | return 0; |
830 | } else if (*this_str == 0) { // If at the end of this, and not of other, we are less. |
831 | return -1; |
832 | } else if (*that_str == 0) { // If at end of other, and not of this, we are greater. |
833 | return 1; |
834 | } else if (_find_upper(*this_str) < _find_upper(*that_str)) { // If current character in this is less, we are less. |
835 | return -1; |
836 | } else if (_find_upper(*this_str) > _find_upper(*that_str)) { // If current character in this is greater, we are greater. |
837 | return 1; |
838 | } |
839 | |
840 | this_str++; |
841 | that_str++; |
842 | } |
843 | } |
844 | |
845 | signed char String::casecmp_to(const String &p_str) const { |
846 | if (is_empty() && p_str.is_empty()) { |
847 | return 0; |
848 | } |
849 | if (is_empty()) { |
850 | return -1; |
851 | } |
852 | if (p_str.is_empty()) { |
853 | return 1; |
854 | } |
855 | |
856 | const char32_t *that_str = p_str.get_data(); |
857 | const char32_t *this_str = get_data(); |
858 | |
859 | while (true) { |
860 | if (*that_str == 0 && *this_str == 0) { // If both strings are at the end, they are equal. |
861 | return 0; |
862 | } else if (*this_str == 0) { // If at the end of this, and not of other, we are less. |
863 | return -1; |
864 | } else if (*that_str == 0) { // If at end of other, and not of this, we are greater. |
865 | return 1; |
866 | } else if (*this_str < *that_str) { // If current character in this is less, we are less. |
867 | return -1; |
868 | } else if (*this_str > *that_str) { // If current character in this is greater, we are greater. |
869 | return 1; |
870 | } |
871 | |
872 | this_str++; |
873 | that_str++; |
874 | } |
875 | } |
876 | |
877 | static _FORCE_INLINE_ signed char natural_cmp_common(const char32_t *&r_this_str, const char32_t *&r_that_str) { |
878 | // Keep ptrs to start of numerical sequences. |
879 | const char32_t *this_substr = r_this_str; |
880 | const char32_t *that_substr = r_that_str; |
881 | |
882 | // Compare lengths of both numerical sequences, ignoring leading zeros. |
883 | while (is_digit(*r_this_str)) { |
884 | r_this_str++; |
885 | } |
886 | while (is_digit(*r_that_str)) { |
887 | r_that_str++; |
888 | } |
889 | while (*this_substr == '0') { |
890 | this_substr++; |
891 | } |
892 | while (*that_substr == '0') { |
893 | that_substr++; |
894 | } |
895 | int this_len = r_this_str - this_substr; |
896 | int that_len = r_that_str - that_substr; |
897 | |
898 | if (this_len < that_len) { |
899 | return -1; |
900 | } else if (this_len > that_len) { |
901 | return 1; |
902 | } |
903 | |
904 | // If lengths equal, compare lexicographically. |
905 | while (this_substr != r_this_str && that_substr != r_that_str) { |
906 | if (*this_substr < *that_substr) { |
907 | return -1; |
908 | } else if (*this_substr > *that_substr) { |
909 | return 1; |
910 | } |
911 | this_substr++; |
912 | that_substr++; |
913 | } |
914 | |
915 | return 0; |
916 | } |
917 | |
918 | signed char String::naturalcasecmp_to(const String &p_str) const { |
919 | const char32_t *this_str = get_data(); |
920 | const char32_t *that_str = p_str.get_data(); |
921 | |
922 | if (this_str && that_str) { |
923 | while (*this_str == '.' || *that_str == '.') { |
924 | if (*this_str++ != '.') { |
925 | return 1; |
926 | } |
927 | if (*that_str++ != '.') { |
928 | return -1; |
929 | } |
930 | if (!*that_str) { |
931 | return 1; |
932 | } |
933 | if (!*this_str) { |
934 | return -1; |
935 | } |
936 | } |
937 | |
938 | while (*this_str) { |
939 | if (!*that_str) { |
940 | return 1; |
941 | } else if (is_digit(*this_str)) { |
942 | if (!is_digit(*that_str)) { |
943 | return -1; |
944 | } |
945 | |
946 | signed char ret = natural_cmp_common(this_str, that_str); |
947 | if (ret) { |
948 | return ret; |
949 | } |
950 | } else if (is_digit(*that_str)) { |
951 | return 1; |
952 | } else { |
953 | if (*this_str < *that_str) { // If current character in this is less, we are less. |
954 | return -1; |
955 | } else if (*this_str > *that_str) { // If current character in this is greater, we are greater. |
956 | return 1; |
957 | } |
958 | |
959 | this_str++; |
960 | that_str++; |
961 | } |
962 | } |
963 | if (*that_str) { |
964 | return -1; |
965 | } |
966 | } |
967 | |
968 | return 0; |
969 | } |
970 | |
971 | signed char String::naturalnocasecmp_to(const String &p_str) const { |
972 | const char32_t *this_str = get_data(); |
973 | const char32_t *that_str = p_str.get_data(); |
974 | |
975 | if (this_str && that_str) { |
976 | while (*this_str == '.' || *that_str == '.') { |
977 | if (*this_str++ != '.') { |
978 | return 1; |
979 | } |
980 | if (*that_str++ != '.') { |
981 | return -1; |
982 | } |
983 | if (!*that_str) { |
984 | return 1; |
985 | } |
986 | if (!*this_str) { |
987 | return -1; |
988 | } |
989 | } |
990 | |
991 | while (*this_str) { |
992 | if (!*that_str) { |
993 | return 1; |
994 | } else if (is_digit(*this_str)) { |
995 | if (!is_digit(*that_str)) { |
996 | return -1; |
997 | } |
998 | |
999 | signed char ret = natural_cmp_common(this_str, that_str); |
1000 | if (ret) { |
1001 | return ret; |
1002 | } |
1003 | } else if (is_digit(*that_str)) { |
1004 | return 1; |
1005 | } else { |
1006 | if (_find_upper(*this_str) < _find_upper(*that_str)) { // If current character in this is less, we are less. |
1007 | return -1; |
1008 | } else if (_find_upper(*this_str) > _find_upper(*that_str)) { // If current character in this is greater, we are greater. |
1009 | return 1; |
1010 | } |
1011 | |
1012 | this_str++; |
1013 | that_str++; |
1014 | } |
1015 | } |
1016 | if (*that_str) { |
1017 | return -1; |
1018 | } |
1019 | } |
1020 | |
1021 | return 0; |
1022 | } |
1023 | |
1024 | const char32_t *String::get_data() const { |
1025 | static const char32_t zero = 0; |
1026 | return size() ? &operator[](0) : &zero; |
1027 | } |
1028 | |
1029 | String String::_camelcase_to_underscore() const { |
1030 | const char32_t *cstr = get_data(); |
1031 | String new_string; |
1032 | int start_index = 0; |
1033 | |
1034 | for (int i = 1; i < this->size(); i++) { |
1035 | bool is_prev_upper = is_ascii_upper_case(cstr[i - 1]); |
1036 | bool is_prev_lower = is_ascii_lower_case(cstr[i - 1]); |
1037 | bool is_prev_digit = is_digit(cstr[i - 1]); |
1038 | |
1039 | bool is_curr_upper = is_ascii_upper_case(cstr[i]); |
1040 | bool is_curr_lower = is_ascii_lower_case(cstr[i]); |
1041 | bool is_curr_digit = is_digit(cstr[i]); |
1042 | |
1043 | bool is_next_lower = false; |
1044 | if (i + 1 < this->size()) { |
1045 | is_next_lower = is_ascii_lower_case(cstr[i + 1]); |
1046 | } |
1047 | |
1048 | const bool cond_a = is_prev_lower && is_curr_upper; // aA |
1049 | const bool cond_b = (is_prev_upper || is_prev_digit) && is_curr_upper && is_next_lower; // AAa, 2Aa |
1050 | const bool cond_c = is_prev_digit && is_curr_lower && is_next_lower; // 2aa |
1051 | const bool cond_d = (is_prev_upper || is_prev_lower) && is_curr_digit; // A2, a2 |
1052 | |
1053 | if (cond_a || cond_b || cond_c || cond_d) { |
1054 | new_string += this->substr(start_index, i - start_index) + "_" ; |
1055 | start_index = i; |
1056 | } |
1057 | } |
1058 | |
1059 | new_string += this->substr(start_index, this->size() - start_index); |
1060 | return new_string.to_lower(); |
1061 | } |
1062 | |
1063 | String String::capitalize() const { |
1064 | String aux = this->_camelcase_to_underscore().replace("_" , " " ).strip_edges(); |
1065 | String cap; |
1066 | for (int i = 0; i < aux.get_slice_count(" " ); i++) { |
1067 | String slice = aux.get_slicec(' ', i); |
1068 | if (slice.length() > 0) { |
1069 | slice[0] = _find_upper(slice[0]); |
1070 | if (i > 0) { |
1071 | cap += " " ; |
1072 | } |
1073 | cap += slice; |
1074 | } |
1075 | } |
1076 | |
1077 | return cap; |
1078 | } |
1079 | |
1080 | String String::to_camel_case() const { |
1081 | String s = this->to_pascal_case(); |
1082 | if (!s.is_empty()) { |
1083 | s[0] = _find_lower(s[0]); |
1084 | } |
1085 | return s; |
1086 | } |
1087 | |
1088 | String String::to_pascal_case() const { |
1089 | return this->capitalize().replace(" " , "" ); |
1090 | } |
1091 | |
1092 | String String::to_snake_case() const { |
1093 | return this->_camelcase_to_underscore().replace(" " , "_" ).strip_edges(); |
1094 | } |
1095 | |
1096 | String String::get_with_code_lines() const { |
1097 | const Vector<String> lines = split("\n" ); |
1098 | String ret; |
1099 | for (int i = 0; i < lines.size(); i++) { |
1100 | if (i > 0) { |
1101 | ret += "\n" ; |
1102 | } |
1103 | ret += vformat("%4d | %s" , i + 1, lines[i]); |
1104 | } |
1105 | return ret; |
1106 | } |
1107 | |
1108 | int String::get_slice_count(String p_splitter) const { |
1109 | if (is_empty()) { |
1110 | return 0; |
1111 | } |
1112 | if (p_splitter.is_empty()) { |
1113 | return 0; |
1114 | } |
1115 | |
1116 | int pos = 0; |
1117 | int slices = 1; |
1118 | |
1119 | while ((pos = find(p_splitter, pos)) >= 0) { |
1120 | slices++; |
1121 | pos += p_splitter.length(); |
1122 | } |
1123 | |
1124 | return slices; |
1125 | } |
1126 | |
1127 | String String::get_slice(String p_splitter, int p_slice) const { |
1128 | if (is_empty() || p_splitter.is_empty()) { |
1129 | return "" ; |
1130 | } |
1131 | |
1132 | int pos = 0; |
1133 | int prev_pos = 0; |
1134 | //int slices=1; |
1135 | if (p_slice < 0) { |
1136 | return "" ; |
1137 | } |
1138 | if (find(p_splitter) == -1) { |
1139 | return *this; |
1140 | } |
1141 | |
1142 | int i = 0; |
1143 | while (true) { |
1144 | pos = find(p_splitter, pos); |
1145 | if (pos == -1) { |
1146 | pos = length(); //reached end |
1147 | } |
1148 | |
1149 | int from = prev_pos; |
1150 | //int to=pos; |
1151 | |
1152 | if (p_slice == i) { |
1153 | return substr(from, pos - from); |
1154 | } |
1155 | |
1156 | if (pos == length()) { //reached end and no find |
1157 | break; |
1158 | } |
1159 | pos += p_splitter.length(); |
1160 | prev_pos = pos; |
1161 | i++; |
1162 | } |
1163 | |
1164 | return "" ; //no find! |
1165 | } |
1166 | |
1167 | String String::get_slicec(char32_t p_splitter, int p_slice) const { |
1168 | if (is_empty()) { |
1169 | return String(); |
1170 | } |
1171 | |
1172 | if (p_slice < 0) { |
1173 | return String(); |
1174 | } |
1175 | |
1176 | const char32_t *c = this->ptr(); |
1177 | int i = 0; |
1178 | int prev = 0; |
1179 | int count = 0; |
1180 | while (true) { |
1181 | if (c[i] == 0 || c[i] == p_splitter) { |
1182 | if (p_slice == count) { |
1183 | return substr(prev, i - prev); |
1184 | } else if (c[i] == 0) { |
1185 | return String(); |
1186 | } else { |
1187 | count++; |
1188 | prev = i + 1; |
1189 | } |
1190 | } |
1191 | |
1192 | i++; |
1193 | } |
1194 | } |
1195 | |
1196 | Vector<String> String::split_spaces() const { |
1197 | Vector<String> ret; |
1198 | int from = 0; |
1199 | int i = 0; |
1200 | int len = length(); |
1201 | if (len == 0) { |
1202 | return ret; |
1203 | } |
1204 | |
1205 | bool inside = false; |
1206 | |
1207 | while (true) { |
1208 | bool empty = operator[](i) < 33; |
1209 | |
1210 | if (i == 0) { |
1211 | inside = !empty; |
1212 | } |
1213 | |
1214 | if (!empty && !inside) { |
1215 | inside = true; |
1216 | from = i; |
1217 | } |
1218 | |
1219 | if (empty && inside) { |
1220 | ret.push_back(substr(from, i - from)); |
1221 | inside = false; |
1222 | } |
1223 | |
1224 | if (i == len) { |
1225 | break; |
1226 | } |
1227 | i++; |
1228 | } |
1229 | |
1230 | return ret; |
1231 | } |
1232 | |
1233 | Vector<String> String::split(const String &p_splitter, bool p_allow_empty, int p_maxsplit) const { |
1234 | Vector<String> ret; |
1235 | |
1236 | if (is_empty()) { |
1237 | if (p_allow_empty) { |
1238 | ret.push_back("" ); |
1239 | } |
1240 | return ret; |
1241 | } |
1242 | |
1243 | int from = 0; |
1244 | int len = length(); |
1245 | |
1246 | while (true) { |
1247 | int end; |
1248 | if (p_splitter.is_empty()) { |
1249 | end = from + 1; |
1250 | } else { |
1251 | end = find(p_splitter, from); |
1252 | if (end < 0) { |
1253 | end = len; |
1254 | } |
1255 | } |
1256 | if (p_allow_empty || (end > from)) { |
1257 | if (p_maxsplit <= 0) { |
1258 | ret.push_back(substr(from, end - from)); |
1259 | } else { |
1260 | // Put rest of the string and leave cycle. |
1261 | if (p_maxsplit == ret.size()) { |
1262 | ret.push_back(substr(from, len)); |
1263 | break; |
1264 | } |
1265 | |
1266 | // Otherwise, push items until positive limit is reached. |
1267 | ret.push_back(substr(from, end - from)); |
1268 | } |
1269 | } |
1270 | |
1271 | if (end == len) { |
1272 | break; |
1273 | } |
1274 | |
1275 | from = end + p_splitter.length(); |
1276 | } |
1277 | |
1278 | return ret; |
1279 | } |
1280 | |
1281 | Vector<String> String::rsplit(const String &p_splitter, bool p_allow_empty, int p_maxsplit) const { |
1282 | Vector<String> ret; |
1283 | const int len = length(); |
1284 | int remaining_len = len; |
1285 | |
1286 | while (true) { |
1287 | if (remaining_len < p_splitter.length() || (p_maxsplit > 0 && p_maxsplit == ret.size())) { |
1288 | // no room for another splitter or hit max splits, push what's left and we're done |
1289 | if (p_allow_empty || remaining_len > 0) { |
1290 | ret.push_back(substr(0, remaining_len)); |
1291 | } |
1292 | break; |
1293 | } |
1294 | |
1295 | int left_edge; |
1296 | if (p_splitter.is_empty()) { |
1297 | left_edge = remaining_len - 1; |
1298 | if (left_edge == 0) { |
1299 | left_edge--; // Skip to the < 0 condition. |
1300 | } |
1301 | } else { |
1302 | left_edge = rfind(p_splitter, remaining_len - p_splitter.length()); |
1303 | } |
1304 | |
1305 | if (left_edge < 0) { |
1306 | // no more splitters, we're done |
1307 | ret.push_back(substr(0, remaining_len)); |
1308 | break; |
1309 | } |
1310 | |
1311 | int substr_start = left_edge + p_splitter.length(); |
1312 | if (p_allow_empty || substr_start < remaining_len) { |
1313 | ret.push_back(substr(substr_start, remaining_len - substr_start)); |
1314 | } |
1315 | |
1316 | remaining_len = left_edge; |
1317 | } |
1318 | |
1319 | ret.reverse(); |
1320 | return ret; |
1321 | } |
1322 | |
1323 | Vector<double> String::split_floats(const String &p_splitter, bool p_allow_empty) const { |
1324 | Vector<double> ret; |
1325 | int from = 0; |
1326 | int len = length(); |
1327 | |
1328 | while (true) { |
1329 | int end = find(p_splitter, from); |
1330 | if (end < 0) { |
1331 | end = len; |
1332 | } |
1333 | if (p_allow_empty || (end > from)) { |
1334 | ret.push_back(String::to_float(&get_data()[from])); |
1335 | } |
1336 | |
1337 | if (end == len) { |
1338 | break; |
1339 | } |
1340 | |
1341 | from = end + p_splitter.length(); |
1342 | } |
1343 | |
1344 | return ret; |
1345 | } |
1346 | |
1347 | Vector<float> String::split_floats_mk(const Vector<String> &p_splitters, bool p_allow_empty) const { |
1348 | Vector<float> ret; |
1349 | int from = 0; |
1350 | int len = length(); |
1351 | |
1352 | while (true) { |
1353 | int idx; |
1354 | int end = findmk(p_splitters, from, &idx); |
1355 | int spl_len = 1; |
1356 | if (end < 0) { |
1357 | end = len; |
1358 | } else { |
1359 | spl_len = p_splitters[idx].length(); |
1360 | } |
1361 | |
1362 | if (p_allow_empty || (end > from)) { |
1363 | ret.push_back(String::to_float(&get_data()[from])); |
1364 | } |
1365 | |
1366 | if (end == len) { |
1367 | break; |
1368 | } |
1369 | |
1370 | from = end + spl_len; |
1371 | } |
1372 | |
1373 | return ret; |
1374 | } |
1375 | |
1376 | Vector<int> String::split_ints(const String &p_splitter, bool p_allow_empty) const { |
1377 | Vector<int> ret; |
1378 | int from = 0; |
1379 | int len = length(); |
1380 | |
1381 | while (true) { |
1382 | int end = find(p_splitter, from); |
1383 | if (end < 0) { |
1384 | end = len; |
1385 | } |
1386 | if (p_allow_empty || (end > from)) { |
1387 | ret.push_back(String::to_int(&get_data()[from], end - from)); |
1388 | } |
1389 | |
1390 | if (end == len) { |
1391 | break; |
1392 | } |
1393 | |
1394 | from = end + p_splitter.length(); |
1395 | } |
1396 | |
1397 | return ret; |
1398 | } |
1399 | |
1400 | Vector<int> String::split_ints_mk(const Vector<String> &p_splitters, bool p_allow_empty) const { |
1401 | Vector<int> ret; |
1402 | int from = 0; |
1403 | int len = length(); |
1404 | |
1405 | while (true) { |
1406 | int idx; |
1407 | int end = findmk(p_splitters, from, &idx); |
1408 | int spl_len = 1; |
1409 | if (end < 0) { |
1410 | end = len; |
1411 | } else { |
1412 | spl_len = p_splitters[idx].length(); |
1413 | } |
1414 | |
1415 | if (p_allow_empty || (end > from)) { |
1416 | ret.push_back(String::to_int(&get_data()[from], end - from)); |
1417 | } |
1418 | |
1419 | if (end == len) { |
1420 | break; |
1421 | } |
1422 | |
1423 | from = end + spl_len; |
1424 | } |
1425 | |
1426 | return ret; |
1427 | } |
1428 | |
1429 | String String::join(Vector<String> parts) const { |
1430 | String ret; |
1431 | for (int i = 0; i < parts.size(); ++i) { |
1432 | if (i > 0) { |
1433 | ret += *this; |
1434 | } |
1435 | ret += parts[i]; |
1436 | } |
1437 | return ret; |
1438 | } |
1439 | |
1440 | char32_t String::char_uppercase(char32_t p_char) { |
1441 | return _find_upper(p_char); |
1442 | } |
1443 | |
1444 | char32_t String::char_lowercase(char32_t p_char) { |
1445 | return _find_lower(p_char); |
1446 | } |
1447 | |
1448 | String String::to_upper() const { |
1449 | String upper = *this; |
1450 | |
1451 | for (int i = 0; i < upper.size(); i++) { |
1452 | const char32_t s = upper[i]; |
1453 | const char32_t t = _find_upper(s); |
1454 | if (s != t) { // avoid copy on write |
1455 | upper[i] = t; |
1456 | } |
1457 | } |
1458 | |
1459 | return upper; |
1460 | } |
1461 | |
1462 | String String::to_lower() const { |
1463 | String lower = *this; |
1464 | |
1465 | for (int i = 0; i < lower.size(); i++) { |
1466 | const char32_t s = lower[i]; |
1467 | const char32_t t = _find_lower(s); |
1468 | if (s != t) { // avoid copy on write |
1469 | lower[i] = t; |
1470 | } |
1471 | } |
1472 | |
1473 | return lower; |
1474 | } |
1475 | |
1476 | String String::chr(char32_t p_char) { |
1477 | char32_t c[2] = { p_char, 0 }; |
1478 | return String(c); |
1479 | } |
1480 | |
1481 | String String::num(double p_num, int p_decimals) { |
1482 | if (Math::is_nan(p_num)) { |
1483 | return "nan" ; |
1484 | } |
1485 | |
1486 | if (Math::is_inf(p_num)) { |
1487 | if (signbit(p_num)) { |
1488 | return "-inf" ; |
1489 | } else { |
1490 | return "inf" ; |
1491 | } |
1492 | } |
1493 | |
1494 | if (p_decimals < 0) { |
1495 | p_decimals = 14; |
1496 | const double abs_num = Math::abs(p_num); |
1497 | if (abs_num > 10) { |
1498 | // We want to align the digits to the above reasonable default, so we only |
1499 | // need to subtract log10 for numbers with a positive power of ten. |
1500 | p_decimals -= (int)floor(log10(abs_num)); |
1501 | } |
1502 | } |
1503 | if (p_decimals > MAX_DECIMALS) { |
1504 | p_decimals = MAX_DECIMALS; |
1505 | } |
1506 | |
1507 | char fmt[7]; |
1508 | fmt[0] = '%'; |
1509 | fmt[1] = '.'; |
1510 | |
1511 | if (p_decimals < 0) { |
1512 | fmt[1] = 'l'; |
1513 | fmt[2] = 'f'; |
1514 | fmt[3] = 0; |
1515 | } else if (p_decimals < 10) { |
1516 | fmt[2] = '0' + p_decimals; |
1517 | fmt[3] = 'l'; |
1518 | fmt[4] = 'f'; |
1519 | fmt[5] = 0; |
1520 | } else { |
1521 | fmt[2] = '0' + (p_decimals / 10); |
1522 | fmt[3] = '0' + (p_decimals % 10); |
1523 | fmt[4] = 'l'; |
1524 | fmt[5] = 'f'; |
1525 | fmt[6] = 0; |
1526 | } |
1527 | // if we want to convert a double with as much decimal places as as |
1528 | // DBL_MAX or DBL_MIN then we would theoretically need a buffer of at least |
1529 | // DBL_MAX_10_EXP + 2 for DBL_MAX and DBL_MAX_10_EXP + 4 for DBL_MIN. |
1530 | // BUT those values where still giving me exceptions, so I tested from |
1531 | // DBL_MAX_10_EXP + 10 incrementing one by one and DBL_MAX_10_EXP + 17 (325) |
1532 | // was the first buffer size not to throw an exception |
1533 | char buf[325]; |
1534 | |
1535 | #if defined(__GNUC__) || defined(_MSC_VER) |
1536 | // PLEASE NOTE that, albeit vcrt online reference states that snprintf |
1537 | // should safely truncate the output to the given buffer size, we have |
1538 | // found a case where this is not true, so we should create a buffer |
1539 | // as big as needed |
1540 | snprintf(buf, 325, fmt, p_num); |
1541 | #else |
1542 | sprintf(buf, fmt, p_num); |
1543 | #endif |
1544 | |
1545 | buf[324] = 0; |
1546 | //destroy trailing zeroes |
1547 | { |
1548 | bool period = false; |
1549 | int z = 0; |
1550 | while (buf[z]) { |
1551 | if (buf[z] == '.') { |
1552 | period = true; |
1553 | } |
1554 | z++; |
1555 | } |
1556 | |
1557 | if (period) { |
1558 | z--; |
1559 | while (z > 0) { |
1560 | if (buf[z] == '0') { |
1561 | buf[z] = 0; |
1562 | } else if (buf[z] == '.') { |
1563 | buf[z] = 0; |
1564 | break; |
1565 | } else { |
1566 | break; |
1567 | } |
1568 | |
1569 | z--; |
1570 | } |
1571 | } |
1572 | } |
1573 | |
1574 | return buf; |
1575 | } |
1576 | |
1577 | String String::num_int64(int64_t p_num, int base, bool capitalize_hex) { |
1578 | bool sign = p_num < 0; |
1579 | |
1580 | int64_t n = p_num; |
1581 | |
1582 | int chars = 0; |
1583 | do { |
1584 | n /= base; |
1585 | chars++; |
1586 | } while (n); |
1587 | |
1588 | if (sign) { |
1589 | chars++; |
1590 | } |
1591 | String s; |
1592 | s.resize(chars + 1); |
1593 | char32_t *c = s.ptrw(); |
1594 | c[chars] = 0; |
1595 | n = p_num; |
1596 | do { |
1597 | int mod = ABS(n % base); |
1598 | if (mod >= 10) { |
1599 | char a = (capitalize_hex ? 'A' : 'a'); |
1600 | c[--chars] = a + (mod - 10); |
1601 | } else { |
1602 | c[--chars] = '0' + mod; |
1603 | } |
1604 | |
1605 | n /= base; |
1606 | } while (n); |
1607 | |
1608 | if (sign) { |
1609 | c[0] = '-'; |
1610 | } |
1611 | |
1612 | return s; |
1613 | } |
1614 | |
1615 | String String::num_uint64(uint64_t p_num, int base, bool capitalize_hex) { |
1616 | uint64_t n = p_num; |
1617 | |
1618 | int chars = 0; |
1619 | do { |
1620 | n /= base; |
1621 | chars++; |
1622 | } while (n); |
1623 | |
1624 | String s; |
1625 | s.resize(chars + 1); |
1626 | char32_t *c = s.ptrw(); |
1627 | c[chars] = 0; |
1628 | n = p_num; |
1629 | do { |
1630 | int mod = n % base; |
1631 | if (mod >= 10) { |
1632 | char a = (capitalize_hex ? 'A' : 'a'); |
1633 | c[--chars] = a + (mod - 10); |
1634 | } else { |
1635 | c[--chars] = '0' + mod; |
1636 | } |
1637 | |
1638 | n /= base; |
1639 | } while (n); |
1640 | |
1641 | return s; |
1642 | } |
1643 | |
1644 | String String::num_real(double p_num, bool p_trailing) { |
1645 | if (p_num == (double)(int64_t)p_num) { |
1646 | if (p_trailing) { |
1647 | return num_int64((int64_t)p_num) + ".0" ; |
1648 | } else { |
1649 | return num_int64((int64_t)p_num); |
1650 | } |
1651 | } |
1652 | #ifdef REAL_T_IS_DOUBLE |
1653 | int decimals = 14; |
1654 | #else |
1655 | int decimals = 6; |
1656 | #endif |
1657 | // We want to align the digits to the above sane default, so we only need |
1658 | // to subtract log10 for numbers with a positive power of ten magnitude. |
1659 | double abs_num = Math::abs(p_num); |
1660 | if (abs_num > 10) { |
1661 | decimals -= (int)floor(log10(abs_num)); |
1662 | } |
1663 | return num(p_num, decimals); |
1664 | } |
1665 | |
1666 | String String::num_scientific(double p_num) { |
1667 | if (Math::is_nan(p_num)) { |
1668 | return "nan" ; |
1669 | } |
1670 | |
1671 | if (Math::is_inf(p_num)) { |
1672 | if (signbit(p_num)) { |
1673 | return "-inf" ; |
1674 | } else { |
1675 | return "inf" ; |
1676 | } |
1677 | } |
1678 | |
1679 | char buf[256]; |
1680 | |
1681 | #if defined(__GNUC__) || defined(_MSC_VER) |
1682 | |
1683 | #if defined(__MINGW32__) && defined(_TWO_DIGIT_EXPONENT) && !defined(_UCRT) |
1684 | // MinGW requires _set_output_format() to conform to C99 output for printf |
1685 | unsigned int old_exponent_format = _set_output_format(_TWO_DIGIT_EXPONENT); |
1686 | #endif |
1687 | snprintf(buf, 256, "%lg" , p_num); |
1688 | |
1689 | #if defined(__MINGW32__) && defined(_TWO_DIGIT_EXPONENT) && !defined(_UCRT) |
1690 | _set_output_format(old_exponent_format); |
1691 | #endif |
1692 | |
1693 | #else |
1694 | sprintf(buf, "%.16lg" , p_num); |
1695 | #endif |
1696 | |
1697 | buf[255] = 0; |
1698 | |
1699 | return buf; |
1700 | } |
1701 | |
1702 | String String::md5(const uint8_t *p_md5) { |
1703 | return String::hex_encode_buffer(p_md5, 16); |
1704 | } |
1705 | |
1706 | String String::hex_encode_buffer(const uint8_t *p_buffer, int p_len) { |
1707 | static const char hex[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'a', 'b', 'c', 'd', 'e', 'f' }; |
1708 | |
1709 | String ret; |
1710 | char v[2] = { 0, 0 }; |
1711 | |
1712 | for (int i = 0; i < p_len; i++) { |
1713 | v[0] = hex[p_buffer[i] >> 4]; |
1714 | ret += v; |
1715 | v[0] = hex[p_buffer[i] & 0xF]; |
1716 | ret += v; |
1717 | } |
1718 | |
1719 | return ret; |
1720 | } |
1721 | |
1722 | Vector<uint8_t> String::hex_decode() const { |
1723 | ERR_FAIL_COND_V_MSG(length() % 2 != 0, Vector<uint8_t>(), "Hexadecimal string of uneven length." ); |
1724 | |
1725 | #define HEX_TO_BYTE(m_output, m_index) \ |
1726 | uint8_t m_output; \ |
1727 | c = operator[](m_index); \ |
1728 | if (is_digit(c)) { \ |
1729 | m_output = c - '0'; \ |
1730 | } else if (c >= 'a' && c <= 'f') { \ |
1731 | m_output = c - 'a' + 10; \ |
1732 | } else if (c >= 'A' && c <= 'F') { \ |
1733 | m_output = c - 'A' + 10; \ |
1734 | } else { \ |
1735 | ERR_FAIL_V_MSG(Vector<uint8_t>(), "Invalid hexadecimal character \"" + chr(c) + "\" at index " + m_index + "."); \ |
1736 | } |
1737 | |
1738 | Vector<uint8_t> out; |
1739 | int len = length() / 2; |
1740 | out.resize(len); |
1741 | for (int i = 0; i < len; i++) { |
1742 | char32_t c; |
1743 | HEX_TO_BYTE(first, i * 2); |
1744 | HEX_TO_BYTE(second, i * 2 + 1); |
1745 | out.write[i] = first * 16 + second; |
1746 | } |
1747 | return out; |
1748 | #undef HEX_TO_BYTE |
1749 | } |
1750 | |
1751 | void String::print_unicode_error(const String &p_message, bool p_critical) const { |
1752 | if (p_critical) { |
1753 | print_error(vformat(U"Unicode parsing error, some characters were replaced with � (U+FFFD): %s" , p_message)); |
1754 | } else { |
1755 | print_error(vformat("Unicode parsing error: %s" , p_message)); |
1756 | } |
1757 | } |
1758 | |
1759 | CharString String::ascii(bool p_allow_extended) const { |
1760 | if (!length()) { |
1761 | return CharString(); |
1762 | } |
1763 | |
1764 | CharString cs; |
1765 | cs.resize(size()); |
1766 | |
1767 | for (int i = 0; i < size(); i++) { |
1768 | char32_t c = operator[](i); |
1769 | if ((c <= 0x7f) || (c <= 0xff && p_allow_extended)) { |
1770 | cs[i] = c; |
1771 | } else { |
1772 | print_unicode_error(vformat("Invalid unicode codepoint (%x), cannot represent as ASCII/Latin-1" , (uint32_t)c)); |
1773 | cs[i] = 0x20; // ascii doesn't have a replacement character like unicode, 0x1a is sometimes used but is kinda arcane |
1774 | } |
1775 | } |
1776 | |
1777 | return cs; |
1778 | } |
1779 | |
1780 | String String::utf8(const char *p_utf8, int p_len) { |
1781 | String ret; |
1782 | ret.parse_utf8(p_utf8, p_len); |
1783 | |
1784 | return ret; |
1785 | } |
1786 | |
1787 | Error String::parse_utf8(const char *p_utf8, int p_len, bool p_skip_cr) { |
1788 | if (!p_utf8) { |
1789 | return ERR_INVALID_DATA; |
1790 | } |
1791 | |
1792 | String aux; |
1793 | |
1794 | int cstr_size = 0; |
1795 | int str_size = 0; |
1796 | |
1797 | /* HANDLE BOM (Byte Order Mark) */ |
1798 | if (p_len < 0 || p_len >= 3) { |
1799 | bool has_bom = uint8_t(p_utf8[0]) == 0xef && uint8_t(p_utf8[1]) == 0xbb && uint8_t(p_utf8[2]) == 0xbf; |
1800 | if (has_bom) { |
1801 | //8-bit encoding, byte order has no meaning in UTF-8, just skip it |
1802 | if (p_len >= 0) { |
1803 | p_len -= 3; |
1804 | } |
1805 | p_utf8 += 3; |
1806 | } |
1807 | } |
1808 | |
1809 | bool decode_error = false; |
1810 | bool decode_failed = false; |
1811 | { |
1812 | const char *ptrtmp = p_utf8; |
1813 | const char *ptrtmp_limit = &p_utf8[p_len]; |
1814 | int skip = 0; |
1815 | uint8_t c_start = 0; |
1816 | while (ptrtmp != ptrtmp_limit && *ptrtmp) { |
1817 | uint8_t c = *ptrtmp >= 0 ? *ptrtmp : uint8_t(256 + *ptrtmp); |
1818 | |
1819 | if (skip == 0) { |
1820 | if (p_skip_cr && c == '\r') { |
1821 | ptrtmp++; |
1822 | continue; |
1823 | } |
1824 | /* Determine the number of characters in sequence */ |
1825 | if ((c & 0x80) == 0) { |
1826 | skip = 0; |
1827 | } else if ((c & 0xe0) == 0xc0) { |
1828 | skip = 1; |
1829 | } else if ((c & 0xf0) == 0xe0) { |
1830 | skip = 2; |
1831 | } else if ((c & 0xf8) == 0xf0) { |
1832 | skip = 3; |
1833 | } else if ((c & 0xfc) == 0xf8) { |
1834 | skip = 4; |
1835 | } else if ((c & 0xfe) == 0xfc) { |
1836 | skip = 5; |
1837 | } else { |
1838 | skip = 0; |
1839 | print_unicode_error(vformat("Invalid UTF-8 leading byte (%x)" , c), true); |
1840 | decode_failed = true; |
1841 | } |
1842 | c_start = c; |
1843 | |
1844 | if (skip == 1 && (c & 0x1e) == 0) { |
1845 | print_unicode_error(vformat("Overlong encoding (%x ...)" , c)); |
1846 | decode_error = true; |
1847 | } |
1848 | str_size++; |
1849 | } else { |
1850 | if ((c_start == 0xe0 && skip == 2 && c < 0xa0) || (c_start == 0xf0 && skip == 3 && c < 0x90) || (c_start == 0xf8 && skip == 4 && c < 0x88) || (c_start == 0xfc && skip == 5 && c < 0x84)) { |
1851 | print_unicode_error(vformat("Overlong encoding (%x %x ...)" , c_start, c)); |
1852 | decode_error = true; |
1853 | } |
1854 | if (c < 0x80 || c > 0xbf) { |
1855 | print_unicode_error(vformat("Invalid UTF-8 continuation byte (%x ... %x ...)" , c_start, c), true); |
1856 | decode_failed = true; |
1857 | skip = 0; |
1858 | } else { |
1859 | --skip; |
1860 | } |
1861 | } |
1862 | |
1863 | cstr_size++; |
1864 | ptrtmp++; |
1865 | } |
1866 | |
1867 | if (skip) { |
1868 | print_unicode_error(vformat("Missing %d UTF-8 continuation byte(s)" , skip), true); |
1869 | decode_failed = true; |
1870 | } |
1871 | } |
1872 | |
1873 | if (str_size == 0) { |
1874 | clear(); |
1875 | return OK; // empty string |
1876 | } |
1877 | |
1878 | resize(str_size + 1); |
1879 | char32_t *dst = ptrw(); |
1880 | dst[str_size] = 0; |
1881 | |
1882 | int skip = 0; |
1883 | uint32_t unichar = 0; |
1884 | while (cstr_size) { |
1885 | uint8_t c = *p_utf8 >= 0 ? *p_utf8 : uint8_t(256 + *p_utf8); |
1886 | |
1887 | if (skip == 0) { |
1888 | if (p_skip_cr && c == '\r') { |
1889 | p_utf8++; |
1890 | continue; |
1891 | } |
1892 | /* Determine the number of characters in sequence */ |
1893 | if ((c & 0x80) == 0) { |
1894 | *(dst++) = c; |
1895 | unichar = 0; |
1896 | skip = 0; |
1897 | } else if ((c & 0xe0) == 0xc0) { |
1898 | unichar = (0xff >> 3) & c; |
1899 | skip = 1; |
1900 | } else if ((c & 0xf0) == 0xe0) { |
1901 | unichar = (0xff >> 4) & c; |
1902 | skip = 2; |
1903 | } else if ((c & 0xf8) == 0xf0) { |
1904 | unichar = (0xff >> 5) & c; |
1905 | skip = 3; |
1906 | } else if ((c & 0xfc) == 0xf8) { |
1907 | unichar = (0xff >> 6) & c; |
1908 | skip = 4; |
1909 | } else if ((c & 0xfe) == 0xfc) { |
1910 | unichar = (0xff >> 7) & c; |
1911 | skip = 5; |
1912 | } else { |
1913 | *(dst++) = _replacement_char; |
1914 | unichar = 0; |
1915 | skip = 0; |
1916 | } |
1917 | } else { |
1918 | if (c < 0x80 || c > 0xbf) { |
1919 | *(dst++) = _replacement_char; |
1920 | skip = 0; |
1921 | } else { |
1922 | unichar = (unichar << 6) | (c & 0x3f); |
1923 | --skip; |
1924 | if (skip == 0) { |
1925 | if (unichar == 0) { |
1926 | print_unicode_error("NUL character" , true); |
1927 | decode_failed = true; |
1928 | unichar = _replacement_char; |
1929 | } else if ((unichar & 0xfffff800) == 0xd800) { |
1930 | print_unicode_error(vformat("Unpaired surrogate (%x)" , unichar), true); |
1931 | decode_failed = true; |
1932 | unichar = _replacement_char; |
1933 | } else if (unichar > 0x10ffff) { |
1934 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , unichar), true); |
1935 | decode_failed = true; |
1936 | unichar = _replacement_char; |
1937 | } |
1938 | *(dst++) = unichar; |
1939 | } |
1940 | } |
1941 | } |
1942 | |
1943 | cstr_size--; |
1944 | p_utf8++; |
1945 | } |
1946 | if (skip) { |
1947 | *(dst++) = 0x20; |
1948 | } |
1949 | |
1950 | if (decode_failed) { |
1951 | return ERR_INVALID_DATA; |
1952 | } else if (decode_error) { |
1953 | return ERR_PARSE_ERROR; |
1954 | } else { |
1955 | return OK; |
1956 | } |
1957 | } |
1958 | |
1959 | CharString String::utf8() const { |
1960 | int l = length(); |
1961 | if (!l) { |
1962 | return CharString(); |
1963 | } |
1964 | |
1965 | const char32_t *d = &operator[](0); |
1966 | int fl = 0; |
1967 | for (int i = 0; i < l; i++) { |
1968 | uint32_t c = d[i]; |
1969 | if (c <= 0x7f) { // 7 bits. |
1970 | fl += 1; |
1971 | } else if (c <= 0x7ff) { // 11 bits |
1972 | fl += 2; |
1973 | } else if (c <= 0xffff) { // 16 bits |
1974 | fl += 3; |
1975 | } else if (c <= 0x001fffff) { // 21 bits |
1976 | fl += 4; |
1977 | } else if (c <= 0x03ffffff) { // 26 bits |
1978 | fl += 5; |
1979 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , c)); |
1980 | } else if (c <= 0x7fffffff) { // 31 bits |
1981 | fl += 6; |
1982 | print_unicode_error(vformat("Invalid unicode codepoint (%x)" , c)); |
1983 | } else { |
1984 | fl += 1; |
1985 | print_unicode_error(vformat("Invalid unicode codepoint (%x), cannot represent as UTF-8" , c), true); |
1986 | } |
1987 | } |
1988 | |
1989 | CharString utf8s; |
1990 | if (fl == 0) { |
1991 | return utf8s; |
1992 | } |
1993 | |
1994 | utf8s.resize(fl + 1); |
1995 | uint8_t *cdst = (uint8_t *)utf8s.get_data(); |
1996 | |
1997 | #define APPEND_CHAR(m_c) *(cdst++) = m_c |
1998 | |
1999 | for (int i = 0; i < l; i++) { |
2000 | uint32_t c = d[i]; |
2001 | |
2002 | if (c <= 0x7f) { // 7 bits. |
2003 | APPEND_CHAR(c); |
2004 | } else if (c <= 0x7ff) { // 11 bits |
2005 | APPEND_CHAR(uint32_t(0xc0 | ((c >> 6) & 0x1f))); // Top 5 bits. |
2006 | APPEND_CHAR(uint32_t(0x80 | (c & 0x3f))); // Bottom 6 bits. |
2007 | } else if (c <= 0xffff) { // 16 bits |
2008 | APPEND_CHAR(uint32_t(0xe0 | ((c >> 12) & 0x0f))); // Top 4 bits. |
2009 | APPEND_CHAR(uint32_t(0x80 | ((c >> 6) & 0x3f))); // Middle 6 bits. |
2010 | APPEND_CHAR(uint32_t(0x80 | (c & 0x3f))); // Bottom 6 bits. |
2011 | } else if (c <= 0x001fffff) { // 21 bits |
2012 | APPEND_CHAR(uint32_t(0xf0 | ((c >> 18) & 0x07))); // Top 3 bits. |
2013 | APPEND_CHAR(uint32_t(0x80 | ((c >> 12) & 0x3f))); // Upper middle 6 bits. |
2014 | APPEND_CHAR(uint32_t(0x80 | ((c >> 6) & 0x3f))); // Lower middle 6 bits. |
2015 | APPEND_CHAR(uint32_t(0x80 | (c & 0x3f))); // Bottom 6 bits. |
2016 | } else if (c <= 0x03ffffff) { // 26 bits |
2017 | APPEND_CHAR(uint32_t(0xf8 | ((c >> 24) & 0x03))); // Top 2 bits. |
2018 | APPEND_CHAR(uint32_t(0x80 | ((c >> 18) & 0x3f))); // Upper middle 6 bits. |
2019 | APPEND_CHAR(uint32_t(0x80 | ((c >> 12) & 0x3f))); // middle 6 bits. |
2020 | APPEND_CHAR(uint32_t(0x80 | ((c >> 6) & 0x3f))); // Lower middle 6 bits. |
2021 | APPEND_CHAR(uint32_t(0x80 | (c & 0x3f))); // Bottom 6 bits. |
2022 | } else if (c <= 0x7fffffff) { // 31 bits |
2023 | APPEND_CHAR(uint32_t(0xfc | ((c >> 30) & 0x01))); // Top 1 bit. |
2024 | APPEND_CHAR(uint32_t(0x80 | ((c >> 24) & 0x3f))); // Upper upper middle 6 bits. |
2025 | APPEND_CHAR(uint32_t(0x80 | ((c >> 18) & 0x3f))); // Lower upper middle 6 bits. |
2026 | APPEND_CHAR(uint32_t(0x80 | ((c >> 12) & 0x3f))); // Upper lower middle 6 bits. |
2027 | APPEND_CHAR(uint32_t(0x80 | ((c >> 6) & 0x3f))); // Lower lower middle 6 bits. |
2028 | APPEND_CHAR(uint32_t(0x80 | (c & 0x3f))); // Bottom 6 bits. |
2029 | } else { |
2030 | // the string is a valid UTF32, so it should never happen ... |
2031 | print_unicode_error(vformat("Non scalar value (%x)" , c), true); |
2032 | APPEND_CHAR(uint32_t(0xe0 | ((_replacement_char >> 12) & 0x0f))); // Top 4 bits. |
2033 | APPEND_CHAR(uint32_t(0x80 | ((_replacement_char >> 6) & 0x3f))); // Middle 6 bits. |
2034 | APPEND_CHAR(uint32_t(0x80 | (_replacement_char & 0x3f))); // Bottom 6 bits. |
2035 | } |
2036 | } |
2037 | #undef APPEND_CHAR |
2038 | *cdst = 0; //trailing zero |
2039 | |
2040 | return utf8s; |
2041 | } |
2042 | |
2043 | String String::utf16(const char16_t *p_utf16, int p_len) { |
2044 | String ret; |
2045 | ret.parse_utf16(p_utf16, p_len); |
2046 | |
2047 | return ret; |
2048 | } |
2049 | |
2050 | Error String::parse_utf16(const char16_t *p_utf16, int p_len) { |
2051 | if (!p_utf16) { |
2052 | return ERR_INVALID_DATA; |
2053 | } |
2054 | |
2055 | String aux; |
2056 | |
2057 | int cstr_size = 0; |
2058 | int str_size = 0; |
2059 | |
2060 | /* HANDLE BOM (Byte Order Mark) */ |
2061 | bool byteswap = false; // assume correct endianness if no BOM found |
2062 | if (p_len < 0 || p_len >= 1) { |
2063 | bool has_bom = false; |
2064 | if (uint16_t(p_utf16[0]) == 0xfeff) { // correct BOM, read as is |
2065 | has_bom = true; |
2066 | byteswap = false; |
2067 | } else if (uint16_t(p_utf16[0]) == 0xfffe) { // backwards BOM, swap bytes |
2068 | has_bom = true; |
2069 | byteswap = true; |
2070 | } |
2071 | if (has_bom) { |
2072 | if (p_len >= 0) { |
2073 | p_len -= 1; |
2074 | } |
2075 | p_utf16 += 1; |
2076 | } |
2077 | } |
2078 | |
2079 | bool decode_error = false; |
2080 | { |
2081 | const char16_t *ptrtmp = p_utf16; |
2082 | const char16_t *ptrtmp_limit = &p_utf16[p_len]; |
2083 | uint32_t c_prev = 0; |
2084 | bool skip = false; |
2085 | while (ptrtmp != ptrtmp_limit && *ptrtmp) { |
2086 | uint32_t c = (byteswap) ? BSWAP16(*ptrtmp) : *ptrtmp; |
2087 | |
2088 | if ((c & 0xfffffc00) == 0xd800) { // lead surrogate |
2089 | if (skip) { |
2090 | print_unicode_error(vformat("Unpaired lead surrogate (%x [trail?] %x)" , c_prev, c)); |
2091 | decode_error = true; |
2092 | } |
2093 | skip = true; |
2094 | } else if ((c & 0xfffffc00) == 0xdc00) { // trail surrogate |
2095 | if (skip) { |
2096 | str_size--; |
2097 | } else { |
2098 | print_unicode_error(vformat("Unpaired trail surrogate (%x [lead?] %x)" , c_prev, c)); |
2099 | decode_error = true; |
2100 | } |
2101 | skip = false; |
2102 | } else { |
2103 | skip = false; |
2104 | } |
2105 | |
2106 | c_prev = c; |
2107 | str_size++; |
2108 | cstr_size++; |
2109 | ptrtmp++; |
2110 | } |
2111 | |
2112 | if (skip) { |
2113 | print_unicode_error(vformat("Unpaired lead surrogate (%x [eol])" , c_prev)); |
2114 | decode_error = true; |
2115 | } |
2116 | } |
2117 | |
2118 | if (str_size == 0) { |
2119 | clear(); |
2120 | return OK; // empty string |
2121 | } |
2122 | |
2123 | resize(str_size + 1); |
2124 | char32_t *dst = ptrw(); |
2125 | dst[str_size] = 0; |
2126 | |
2127 | bool skip = false; |
2128 | uint32_t c_prev = 0; |
2129 | while (cstr_size) { |
2130 | uint32_t c = (byteswap) ? BSWAP16(*p_utf16) : *p_utf16; |
2131 | |
2132 | if ((c & 0xfffffc00) == 0xd800) { // lead surrogate |
2133 | if (skip) { |
2134 | *(dst++) = c_prev; // unpaired, store as is |
2135 | } |
2136 | skip = true; |
2137 | } else if ((c & 0xfffffc00) == 0xdc00) { // trail surrogate |
2138 | if (skip) { |
2139 | *(dst++) = (c_prev << 10UL) + c - ((0xd800 << 10UL) + 0xdc00 - 0x10000); // decode pair |
2140 | } else { |
2141 | *(dst++) = c; // unpaired, store as is |
2142 | } |
2143 | skip = false; |
2144 | } else { |
2145 | *(dst++) = c; |
2146 | skip = false; |
2147 | } |
2148 | |
2149 | cstr_size--; |
2150 | p_utf16++; |
2151 | c_prev = c; |
2152 | } |
2153 | |
2154 | if (skip) { |
2155 | *(dst++) = c_prev; |
2156 | } |
2157 | |
2158 | if (decode_error) { |
2159 | return ERR_PARSE_ERROR; |
2160 | } else { |
2161 | return OK; |
2162 | } |
2163 | } |
2164 | |
2165 | Char16String String::utf16() const { |
2166 | int l = length(); |
2167 | if (!l) { |
2168 | return Char16String(); |
2169 | } |
2170 | |
2171 | const char32_t *d = &operator[](0); |
2172 | int fl = 0; |
2173 | for (int i = 0; i < l; i++) { |
2174 | uint32_t c = d[i]; |
2175 | if (c <= 0xffff) { // 16 bits. |
2176 | fl += 1; |
2177 | if ((c & 0xfffff800) == 0xd800) { |
2178 | print_unicode_error(vformat("Unpaired surrogate (%x)" , c)); |
2179 | } |
2180 | } else if (c <= 0x10ffff) { // 32 bits. |
2181 | fl += 2; |
2182 | } else { |
2183 | print_unicode_error(vformat("Invalid unicode codepoint (%x), cannot represent as UTF-16" , c), true); |
2184 | fl += 1; |
2185 | } |
2186 | } |
2187 | |
2188 | Char16String utf16s; |
2189 | if (fl == 0) { |
2190 | return utf16s; |
2191 | } |
2192 | |
2193 | utf16s.resize(fl + 1); |
2194 | uint16_t *cdst = (uint16_t *)utf16s.get_data(); |
2195 | |
2196 | #define APPEND_CHAR(m_c) *(cdst++) = m_c |
2197 | |
2198 | for (int i = 0; i < l; i++) { |
2199 | uint32_t c = d[i]; |
2200 | |
2201 | if (c <= 0xffff) { // 16 bits. |
2202 | APPEND_CHAR(c); |
2203 | } else if (c <= 0x10ffff) { // 32 bits. |
2204 | APPEND_CHAR(uint32_t((c >> 10) + 0xd7c0)); // lead surrogate. |
2205 | APPEND_CHAR(uint32_t((c & 0x3ff) | 0xdc00)); // trail surrogate. |
2206 | } else { |
2207 | // the string is a valid UTF32, so it should never happen ... |
2208 | APPEND_CHAR(uint32_t((_replacement_char >> 10) + 0xd7c0)); |
2209 | APPEND_CHAR(uint32_t((_replacement_char & 0x3ff) | 0xdc00)); |
2210 | } |
2211 | } |
2212 | #undef APPEND_CHAR |
2213 | *cdst = 0; //trailing zero |
2214 | |
2215 | return utf16s; |
2216 | } |
2217 | |
2218 | String::String(const char *p_str) { |
2219 | copy_from(p_str); |
2220 | } |
2221 | |
2222 | String::String(const wchar_t *p_str) { |
2223 | copy_from(p_str); |
2224 | } |
2225 | |
2226 | String::String(const char32_t *p_str) { |
2227 | copy_from(p_str); |
2228 | } |
2229 | |
2230 | String::String(const char *p_str, int p_clip_to_len) { |
2231 | copy_from(p_str, p_clip_to_len); |
2232 | } |
2233 | |
2234 | String::String(const wchar_t *p_str, int p_clip_to_len) { |
2235 | copy_from(p_str, p_clip_to_len); |
2236 | } |
2237 | |
2238 | String::String(const char32_t *p_str, int p_clip_to_len) { |
2239 | copy_from(p_str, p_clip_to_len); |
2240 | } |
2241 | |
2242 | String::String(const StrRange &p_range) { |
2243 | if (!p_range.c_str) { |
2244 | return; |
2245 | } |
2246 | copy_from(p_range.c_str, p_range.len); |
2247 | } |
2248 | |
2249 | int64_t String::hex_to_int() const { |
2250 | int len = length(); |
2251 | if (len == 0) { |
2252 | return 0; |
2253 | } |
2254 | |
2255 | const char32_t *s = ptr(); |
2256 | |
2257 | int64_t sign = s[0] == '-' ? -1 : 1; |
2258 | |
2259 | if (sign < 0) { |
2260 | s++; |
2261 | } |
2262 | |
2263 | if (len > 2 && s[0] == '0' && lower_case(s[1]) == 'x') { |
2264 | s += 2; |
2265 | } |
2266 | |
2267 | int64_t hex = 0; |
2268 | |
2269 | while (*s) { |
2270 | char32_t c = lower_case(*s); |
2271 | int64_t n; |
2272 | if (is_digit(c)) { |
2273 | n = c - '0'; |
2274 | } else if (c >= 'a' && c <= 'f') { |
2275 | n = (c - 'a') + 10; |
2276 | } else { |
2277 | ERR_FAIL_V_MSG(0, vformat(R"(Invalid hexadecimal notation character "%c" (U+%04X) in string "%s".)" , *s, static_cast<int32_t>(*s), *this)); |
2278 | } |
2279 | // Check for overflow/underflow, with special case to ensure INT64_MIN does not result in error |
2280 | bool overflow = ((hex > INT64_MAX / 16) && (sign == 1 || (sign == -1 && hex != (INT64_MAX >> 4) + 1))) || (sign == -1 && hex == (INT64_MAX >> 4) + 1 && c > '0'); |
2281 | ERR_FAIL_COND_V_MSG(overflow, sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + *this + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2282 | hex *= 16; |
2283 | hex += n; |
2284 | s++; |
2285 | } |
2286 | |
2287 | return hex * sign; |
2288 | } |
2289 | |
2290 | int64_t String::bin_to_int() const { |
2291 | int len = length(); |
2292 | if (len == 0) { |
2293 | return 0; |
2294 | } |
2295 | |
2296 | const char32_t *s = ptr(); |
2297 | |
2298 | int64_t sign = s[0] == '-' ? -1 : 1; |
2299 | |
2300 | if (sign < 0) { |
2301 | s++; |
2302 | } |
2303 | |
2304 | if (len > 2 && s[0] == '0' && lower_case(s[1]) == 'b') { |
2305 | s += 2; |
2306 | } |
2307 | |
2308 | int64_t binary = 0; |
2309 | |
2310 | while (*s) { |
2311 | char32_t c = lower_case(*s); |
2312 | int64_t n; |
2313 | if (c == '0' || c == '1') { |
2314 | n = c - '0'; |
2315 | } else { |
2316 | return 0; |
2317 | } |
2318 | // Check for overflow/underflow, with special case to ensure INT64_MIN does not result in error |
2319 | bool overflow = ((binary > INT64_MAX / 2) && (sign == 1 || (sign == -1 && binary != (INT64_MAX >> 1) + 1))) || (sign == -1 && binary == (INT64_MAX >> 1) + 1 && c > '0'); |
2320 | ERR_FAIL_COND_V_MSG(overflow, sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + *this + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2321 | binary *= 2; |
2322 | binary += n; |
2323 | s++; |
2324 | } |
2325 | |
2326 | return binary * sign; |
2327 | } |
2328 | |
2329 | int64_t String::to_int() const { |
2330 | if (length() == 0) { |
2331 | return 0; |
2332 | } |
2333 | |
2334 | int to = (find("." ) >= 0) ? find("." ) : length(); |
2335 | |
2336 | int64_t integer = 0; |
2337 | int64_t sign = 1; |
2338 | |
2339 | for (int i = 0; i < to; i++) { |
2340 | char32_t c = operator[](i); |
2341 | if (is_digit(c)) { |
2342 | bool overflow = (integer > INT64_MAX / 10) || (integer == INT64_MAX / 10 && ((sign == 1 && c > '7') || (sign == -1 && c > '8'))); |
2343 | ERR_FAIL_COND_V_MSG(overflow, sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + *this + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2344 | integer *= 10; |
2345 | integer += c - '0'; |
2346 | |
2347 | } else if (integer == 0 && c == '-') { |
2348 | sign = -sign; |
2349 | } |
2350 | } |
2351 | |
2352 | return integer * sign; |
2353 | } |
2354 | |
2355 | int64_t String::to_int(const char *p_str, int p_len) { |
2356 | int to = 0; |
2357 | if (p_len >= 0) { |
2358 | to = p_len; |
2359 | } else { |
2360 | while (p_str[to] != 0 && p_str[to] != '.') { |
2361 | to++; |
2362 | } |
2363 | } |
2364 | |
2365 | int64_t integer = 0; |
2366 | int64_t sign = 1; |
2367 | |
2368 | for (int i = 0; i < to; i++) { |
2369 | char c = p_str[i]; |
2370 | if (is_digit(c)) { |
2371 | bool overflow = (integer > INT64_MAX / 10) || (integer == INT64_MAX / 10 && ((sign == 1 && c > '7') || (sign == -1 && c > '8'))); |
2372 | ERR_FAIL_COND_V_MSG(overflow, sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + String(p_str).substr(0, to) + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2373 | integer *= 10; |
2374 | integer += c - '0'; |
2375 | |
2376 | } else if (c == '-' && integer == 0) { |
2377 | sign = -sign; |
2378 | } else if (c != ' ') { |
2379 | break; |
2380 | } |
2381 | } |
2382 | |
2383 | return integer * sign; |
2384 | } |
2385 | |
2386 | int64_t String::to_int(const wchar_t *p_str, int p_len) { |
2387 | int to = 0; |
2388 | if (p_len >= 0) { |
2389 | to = p_len; |
2390 | } else { |
2391 | while (p_str[to] != 0 && p_str[to] != '.') { |
2392 | to++; |
2393 | } |
2394 | } |
2395 | |
2396 | int64_t integer = 0; |
2397 | int64_t sign = 1; |
2398 | |
2399 | for (int i = 0; i < to; i++) { |
2400 | wchar_t c = p_str[i]; |
2401 | if (is_digit(c)) { |
2402 | bool overflow = (integer > INT64_MAX / 10) || (integer == INT64_MAX / 10 && ((sign == 1 && c > '7') || (sign == -1 && c > '8'))); |
2403 | ERR_FAIL_COND_V_MSG(overflow, sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + String(p_str).substr(0, to) + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2404 | integer *= 10; |
2405 | integer += c - '0'; |
2406 | |
2407 | } else if (c == '-' && integer == 0) { |
2408 | sign = -sign; |
2409 | } else if (c != ' ') { |
2410 | break; |
2411 | } |
2412 | } |
2413 | |
2414 | return integer * sign; |
2415 | } |
2416 | |
2417 | bool String::is_numeric() const { |
2418 | if (length() == 0) { |
2419 | return false; |
2420 | } |
2421 | |
2422 | int s = 0; |
2423 | if (operator[](0) == '-') { |
2424 | ++s; |
2425 | } |
2426 | bool dot = false; |
2427 | for (int i = s; i < length(); i++) { |
2428 | char32_t c = operator[](i); |
2429 | if (c == '.') { |
2430 | if (dot) { |
2431 | return false; |
2432 | } |
2433 | dot = true; |
2434 | } else if (!is_digit(c)) { |
2435 | return false; |
2436 | } |
2437 | } |
2438 | |
2439 | return true; // TODO: Use the parser below for this instead |
2440 | } |
2441 | |
2442 | template <class C> |
2443 | static double built_in_strtod( |
2444 | /* A decimal ASCII floating-point number, |
2445 | * optionally preceded by white space. Must |
2446 | * have form "-I.FE-X", where I is the integer |
2447 | * part of the mantissa, F is the fractional |
2448 | * part of the mantissa, and X is the |
2449 | * exponent. Either of the signs may be "+", |
2450 | * "-", or omitted. Either I or F may be |
2451 | * omitted, or both. The decimal point isn't |
2452 | * necessary unless F is present. The "E" may |
2453 | * actually be an "e". E and X may both be |
2454 | * omitted (but not just one). */ |
2455 | const C *string, |
2456 | /* If non-nullptr, store terminating Cacter's |
2457 | * address here. */ |
2458 | C **endPtr = nullptr) { |
2459 | /* Largest possible base 10 exponent. Any |
2460 | * exponent larger than this will already |
2461 | * produce underflow or overflow, so there's |
2462 | * no need to worry about additional digits. */ |
2463 | static const int maxExponent = 511; |
2464 | /* Table giving binary powers of 10. Entry |
2465 | * is 10^2^i. Used to convert decimal |
2466 | * exponents into floating-point numbers. */ |
2467 | static const double powersOf10[] = { |
2468 | 10., |
2469 | 100., |
2470 | 1.0e4, |
2471 | 1.0e8, |
2472 | 1.0e16, |
2473 | 1.0e32, |
2474 | 1.0e64, |
2475 | 1.0e128, |
2476 | 1.0e256 |
2477 | }; |
2478 | |
2479 | bool sign, expSign = false; |
2480 | double fraction, dblExp; |
2481 | const double *d; |
2482 | const C *p; |
2483 | int c; |
2484 | /* Exponent read from "EX" field. */ |
2485 | int exp = 0; |
2486 | /* Exponent that derives from the fractional |
2487 | * part. Under normal circumstances, it is |
2488 | * the negative of the number of digits in F. |
2489 | * However, if I is very long, the last digits |
2490 | * of I get dropped (otherwise a long I with a |
2491 | * large negative exponent could cause an |
2492 | * unnecessary overflow on I alone). In this |
2493 | * case, fracExp is incremented one for each |
2494 | * dropped digit. */ |
2495 | int fracExp = 0; |
2496 | /* Number of digits in mantissa. */ |
2497 | int mantSize; |
2498 | /* Number of mantissa digits BEFORE decimal point. */ |
2499 | int decPt; |
2500 | /* Temporarily holds location of exponent in string. */ |
2501 | const C *pExp; |
2502 | |
2503 | /* |
2504 | * Strip off leading blanks and check for a sign. |
2505 | */ |
2506 | |
2507 | p = string; |
2508 | while (*p == ' ' || *p == '\t' || *p == '\n') { |
2509 | p += 1; |
2510 | } |
2511 | if (*p == '-') { |
2512 | sign = true; |
2513 | p += 1; |
2514 | } else { |
2515 | if (*p == '+') { |
2516 | p += 1; |
2517 | } |
2518 | sign = false; |
2519 | } |
2520 | |
2521 | /* |
2522 | * Count the number of digits in the mantissa (including the decimal |
2523 | * point), and also locate the decimal point. |
2524 | */ |
2525 | |
2526 | decPt = -1; |
2527 | for (mantSize = 0;; mantSize += 1) { |
2528 | c = *p; |
2529 | if (!is_digit(c)) { |
2530 | if ((c != '.') || (decPt >= 0)) { |
2531 | break; |
2532 | } |
2533 | decPt = mantSize; |
2534 | } |
2535 | p += 1; |
2536 | } |
2537 | |
2538 | /* |
2539 | * Now suck up the digits in the mantissa. Use two integers to collect 9 |
2540 | * digits each (this is faster than using floating-point). If the mantissa |
2541 | * has more than 18 digits, ignore the extras, since they can't affect the |
2542 | * value anyway. |
2543 | */ |
2544 | |
2545 | pExp = p; |
2546 | p -= mantSize; |
2547 | if (decPt < 0) { |
2548 | decPt = mantSize; |
2549 | } else { |
2550 | mantSize -= 1; /* One of the digits was the point. */ |
2551 | } |
2552 | if (mantSize > 18) { |
2553 | fracExp = decPt - 18; |
2554 | mantSize = 18; |
2555 | } else { |
2556 | fracExp = decPt - mantSize; |
2557 | } |
2558 | if (mantSize == 0) { |
2559 | fraction = 0.0; |
2560 | p = string; |
2561 | goto done; |
2562 | } else { |
2563 | int frac1, frac2; |
2564 | |
2565 | frac1 = 0; |
2566 | for (; mantSize > 9; mantSize -= 1) { |
2567 | c = *p; |
2568 | p += 1; |
2569 | if (c == '.') { |
2570 | c = *p; |
2571 | p += 1; |
2572 | } |
2573 | frac1 = 10 * frac1 + (c - '0'); |
2574 | } |
2575 | frac2 = 0; |
2576 | for (; mantSize > 0; mantSize -= 1) { |
2577 | c = *p; |
2578 | p += 1; |
2579 | if (c == '.') { |
2580 | c = *p; |
2581 | p += 1; |
2582 | } |
2583 | frac2 = 10 * frac2 + (c - '0'); |
2584 | } |
2585 | fraction = (1.0e9 * frac1) + frac2; |
2586 | } |
2587 | |
2588 | /* |
2589 | * Skim off the exponent. |
2590 | */ |
2591 | |
2592 | p = pExp; |
2593 | if ((*p == 'E') || (*p == 'e')) { |
2594 | p += 1; |
2595 | if (*p == '-') { |
2596 | expSign = true; |
2597 | p += 1; |
2598 | } else { |
2599 | if (*p == '+') { |
2600 | p += 1; |
2601 | } |
2602 | expSign = false; |
2603 | } |
2604 | if (!is_digit(char32_t(*p))) { |
2605 | p = pExp; |
2606 | goto done; |
2607 | } |
2608 | while (is_digit(char32_t(*p))) { |
2609 | exp = exp * 10 + (*p - '0'); |
2610 | p += 1; |
2611 | } |
2612 | } |
2613 | if (expSign) { |
2614 | exp = fracExp - exp; |
2615 | } else { |
2616 | exp = fracExp + exp; |
2617 | } |
2618 | |
2619 | /* |
2620 | * Generate a floating-point number that represents the exponent. Do this |
2621 | * by processing the exponent one bit at a time to combine many powers of |
2622 | * 2 of 10. Then combine the exponent with the fraction. |
2623 | */ |
2624 | |
2625 | if (exp < 0) { |
2626 | expSign = true; |
2627 | exp = -exp; |
2628 | } else { |
2629 | expSign = false; |
2630 | } |
2631 | |
2632 | if (exp > maxExponent) { |
2633 | exp = maxExponent; |
2634 | WARN_PRINT("Exponent too high" ); |
2635 | } |
2636 | dblExp = 1.0; |
2637 | for (d = powersOf10; exp != 0; exp >>= 1, ++d) { |
2638 | if (exp & 01) { |
2639 | dblExp *= *d; |
2640 | } |
2641 | } |
2642 | if (expSign) { |
2643 | fraction /= dblExp; |
2644 | } else { |
2645 | fraction *= dblExp; |
2646 | } |
2647 | |
2648 | done: |
2649 | if (endPtr != nullptr) { |
2650 | *endPtr = (C *)p; |
2651 | } |
2652 | |
2653 | if (sign) { |
2654 | return -fraction; |
2655 | } |
2656 | return fraction; |
2657 | } |
2658 | |
2659 | #define READING_SIGN 0 |
2660 | #define READING_INT 1 |
2661 | #define READING_DEC 2 |
2662 | #define READING_EXP 3 |
2663 | #define READING_DONE 4 |
2664 | |
2665 | double String::to_float(const char *p_str) { |
2666 | return built_in_strtod<char>(p_str); |
2667 | } |
2668 | |
2669 | double String::to_float(const char32_t *p_str, const char32_t **r_end) { |
2670 | return built_in_strtod<char32_t>(p_str, (char32_t **)r_end); |
2671 | } |
2672 | |
2673 | double String::to_float(const wchar_t *p_str, const wchar_t **r_end) { |
2674 | return built_in_strtod<wchar_t>(p_str, (wchar_t **)r_end); |
2675 | } |
2676 | |
2677 | uint32_t String::num_characters(int64_t p_int) { |
2678 | int r = 1; |
2679 | if (p_int < 0) { |
2680 | r += 1; |
2681 | if (p_int == INT64_MIN) { |
2682 | p_int = INT64_MAX; |
2683 | } else { |
2684 | p_int = -p_int; |
2685 | } |
2686 | } |
2687 | while (p_int >= 10) { |
2688 | p_int /= 10; |
2689 | r++; |
2690 | } |
2691 | return r; |
2692 | } |
2693 | |
2694 | int64_t String::to_int(const char32_t *p_str, int p_len, bool p_clamp) { |
2695 | if (p_len == 0 || !p_str[0]) { |
2696 | return 0; |
2697 | } |
2698 | ///@todo make more exact so saving and loading does not lose precision |
2699 | |
2700 | int64_t integer = 0; |
2701 | int64_t sign = 1; |
2702 | int reading = READING_SIGN; |
2703 | |
2704 | const char32_t *str = p_str; |
2705 | const char32_t *limit = &p_str[p_len]; |
2706 | |
2707 | while (*str && reading != READING_DONE && str != limit) { |
2708 | char32_t c = *(str++); |
2709 | switch (reading) { |
2710 | case READING_SIGN: { |
2711 | if (is_digit(c)) { |
2712 | reading = READING_INT; |
2713 | // let it fallthrough |
2714 | } else if (c == '-') { |
2715 | sign = -1; |
2716 | reading = READING_INT; |
2717 | break; |
2718 | } else if (c == '+') { |
2719 | sign = 1; |
2720 | reading = READING_INT; |
2721 | break; |
2722 | } else { |
2723 | break; |
2724 | } |
2725 | [[fallthrough]]; |
2726 | } |
2727 | case READING_INT: { |
2728 | if (is_digit(c)) { |
2729 | if (integer > INT64_MAX / 10) { |
2730 | String number("" ); |
2731 | str = p_str; |
2732 | while (*str && str != limit) { |
2733 | number += *(str++); |
2734 | } |
2735 | if (p_clamp) { |
2736 | if (sign == 1) { |
2737 | return INT64_MAX; |
2738 | } else { |
2739 | return INT64_MIN; |
2740 | } |
2741 | } else { |
2742 | ERR_FAIL_V_MSG(sign == 1 ? INT64_MAX : INT64_MIN, "Cannot represent " + number + " as a 64-bit signed integer, since the value is " + (sign == 1 ? "too large." : "too small." )); |
2743 | } |
2744 | } |
2745 | integer *= 10; |
2746 | integer += c - '0'; |
2747 | } else { |
2748 | reading = READING_DONE; |
2749 | } |
2750 | |
2751 | } break; |
2752 | } |
2753 | } |
2754 | |
2755 | return sign * integer; |
2756 | } |
2757 | |
2758 | double String::to_float() const { |
2759 | if (is_empty()) { |
2760 | return 0; |
2761 | } |
2762 | return built_in_strtod<char32_t>(get_data()); |
2763 | } |
2764 | |
2765 | uint32_t String::hash(const char *p_cstr) { |
2766 | uint32_t hashv = 5381; |
2767 | uint32_t c = *p_cstr++; |
2768 | |
2769 | while (c) { |
2770 | hashv = ((hashv << 5) + hashv) + c; /* hash * 33 + c */ |
2771 | c = *p_cstr++; |
2772 | } |
2773 | |
2774 | return hashv; |
2775 | } |
2776 | |
2777 | uint32_t String::hash(const char *p_cstr, int p_len) { |
2778 | uint32_t hashv = 5381; |
2779 | for (int i = 0; i < p_len; i++) { |
2780 | hashv = ((hashv << 5) + hashv) + p_cstr[i]; /* hash * 33 + c */ |
2781 | } |
2782 | |
2783 | return hashv; |
2784 | } |
2785 | |
2786 | uint32_t String::hash(const wchar_t *p_cstr, int p_len) { |
2787 | uint32_t hashv = 5381; |
2788 | for (int i = 0; i < p_len; i++) { |
2789 | hashv = ((hashv << 5) + hashv) + p_cstr[i]; /* hash * 33 + c */ |
2790 | } |
2791 | |
2792 | return hashv; |
2793 | } |
2794 | |
2795 | uint32_t String::hash(const wchar_t *p_cstr) { |
2796 | uint32_t hashv = 5381; |
2797 | uint32_t c = *p_cstr++; |
2798 | |
2799 | while (c) { |
2800 | hashv = ((hashv << 5) + hashv) + c; /* hash * 33 + c */ |
2801 | c = *p_cstr++; |
2802 | } |
2803 | |
2804 | return hashv; |
2805 | } |
2806 | |
2807 | uint32_t String::hash(const char32_t *p_cstr, int p_len) { |
2808 | uint32_t hashv = 5381; |
2809 | for (int i = 0; i < p_len; i++) { |
2810 | hashv = ((hashv << 5) + hashv) + p_cstr[i]; /* hash * 33 + c */ |
2811 | } |
2812 | |
2813 | return hashv; |
2814 | } |
2815 | |
2816 | uint32_t String::hash(const char32_t *p_cstr) { |
2817 | uint32_t hashv = 5381; |
2818 | uint32_t c = *p_cstr++; |
2819 | |
2820 | while (c) { |
2821 | hashv = ((hashv << 5) + hashv) + c; /* hash * 33 + c */ |
2822 | c = *p_cstr++; |
2823 | } |
2824 | |
2825 | return hashv; |
2826 | } |
2827 | |
2828 | uint32_t String::hash() const { |
2829 | /* simple djb2 hashing */ |
2830 | |
2831 | const char32_t *chr = get_data(); |
2832 | uint32_t hashv = 5381; |
2833 | uint32_t c = *chr++; |
2834 | |
2835 | while (c) { |
2836 | hashv = ((hashv << 5) + hashv) + c; /* hash * 33 + c */ |
2837 | c = *chr++; |
2838 | } |
2839 | |
2840 | return hashv; |
2841 | } |
2842 | |
2843 | uint64_t String::hash64() const { |
2844 | /* simple djb2 hashing */ |
2845 | |
2846 | const char32_t *chr = get_data(); |
2847 | uint64_t hashv = 5381; |
2848 | uint64_t c = *chr++; |
2849 | |
2850 | while (c) { |
2851 | hashv = ((hashv << 5) + hashv) + c; /* hash * 33 + c */ |
2852 | c = *chr++; |
2853 | } |
2854 | |
2855 | return hashv; |
2856 | } |
2857 | |
2858 | String String::md5_text() const { |
2859 | CharString cs = utf8(); |
2860 | unsigned char hash[16]; |
2861 | CryptoCore::md5((unsigned char *)cs.ptr(), cs.length(), hash); |
2862 | return String::hex_encode_buffer(hash, 16); |
2863 | } |
2864 | |
2865 | String String::sha1_text() const { |
2866 | CharString cs = utf8(); |
2867 | unsigned char hash[20]; |
2868 | CryptoCore::sha1((unsigned char *)cs.ptr(), cs.length(), hash); |
2869 | return String::hex_encode_buffer(hash, 20); |
2870 | } |
2871 | |
2872 | String String::sha256_text() const { |
2873 | CharString cs = utf8(); |
2874 | unsigned char hash[32]; |
2875 | CryptoCore::sha256((unsigned char *)cs.ptr(), cs.length(), hash); |
2876 | return String::hex_encode_buffer(hash, 32); |
2877 | } |
2878 | |
2879 | Vector<uint8_t> String::md5_buffer() const { |
2880 | CharString cs = utf8(); |
2881 | unsigned char hash[16]; |
2882 | CryptoCore::md5((unsigned char *)cs.ptr(), cs.length(), hash); |
2883 | |
2884 | Vector<uint8_t> ret; |
2885 | ret.resize(16); |
2886 | for (int i = 0; i < 16; i++) { |
2887 | ret.write[i] = hash[i]; |
2888 | } |
2889 | return ret; |
2890 | } |
2891 | |
2892 | Vector<uint8_t> String::sha1_buffer() const { |
2893 | CharString cs = utf8(); |
2894 | unsigned char hash[20]; |
2895 | CryptoCore::sha1((unsigned char *)cs.ptr(), cs.length(), hash); |
2896 | |
2897 | Vector<uint8_t> ret; |
2898 | ret.resize(20); |
2899 | for (int i = 0; i < 20; i++) { |
2900 | ret.write[i] = hash[i]; |
2901 | } |
2902 | |
2903 | return ret; |
2904 | } |
2905 | |
2906 | Vector<uint8_t> String::sha256_buffer() const { |
2907 | CharString cs = utf8(); |
2908 | unsigned char hash[32]; |
2909 | CryptoCore::sha256((unsigned char *)cs.ptr(), cs.length(), hash); |
2910 | |
2911 | Vector<uint8_t> ret; |
2912 | ret.resize(32); |
2913 | for (int i = 0; i < 32; i++) { |
2914 | ret.write[i] = hash[i]; |
2915 | } |
2916 | return ret; |
2917 | } |
2918 | |
2919 | String String::insert(int p_at_pos, const String &p_string) const { |
2920 | if (p_at_pos < 0) { |
2921 | return *this; |
2922 | } |
2923 | |
2924 | if (p_at_pos > length()) { |
2925 | p_at_pos = length(); |
2926 | } |
2927 | |
2928 | String pre; |
2929 | if (p_at_pos > 0) { |
2930 | pre = substr(0, p_at_pos); |
2931 | } |
2932 | |
2933 | String post; |
2934 | if (p_at_pos < length()) { |
2935 | post = substr(p_at_pos, length() - p_at_pos); |
2936 | } |
2937 | |
2938 | return pre + p_string + post; |
2939 | } |
2940 | |
2941 | String String::erase(int p_pos, int p_chars) const { |
2942 | ERR_FAIL_COND_V_MSG(p_pos < 0, "" , vformat("Invalid starting position for `String.erase()`: %d. Starting position must be positive or zero." , p_pos)); |
2943 | ERR_FAIL_COND_V_MSG(p_chars < 0, "" , vformat("Invalid character count for `String.erase()`: %d. Character count must be positive or zero." , p_chars)); |
2944 | return left(p_pos) + substr(p_pos + p_chars); |
2945 | } |
2946 | |
2947 | String String::substr(int p_from, int p_chars) const { |
2948 | if (p_chars == -1) { |
2949 | p_chars = length() - p_from; |
2950 | } |
2951 | |
2952 | if (is_empty() || p_from < 0 || p_from >= length() || p_chars <= 0) { |
2953 | return "" ; |
2954 | } |
2955 | |
2956 | if ((p_from + p_chars) > length()) { |
2957 | p_chars = length() - p_from; |
2958 | } |
2959 | |
2960 | if (p_from == 0 && p_chars >= length()) { |
2961 | return String(*this); |
2962 | } |
2963 | |
2964 | String s; |
2965 | s.copy_from_unchecked(&get_data()[p_from], p_chars); |
2966 | return s; |
2967 | } |
2968 | |
2969 | int String::find(const String &p_str, int p_from) const { |
2970 | if (p_from < 0) { |
2971 | return -1; |
2972 | } |
2973 | |
2974 | const int src_len = p_str.length(); |
2975 | |
2976 | const int len = length(); |
2977 | |
2978 | if (src_len == 0 || len == 0) { |
2979 | return -1; // won't find anything! |
2980 | } |
2981 | |
2982 | const char32_t *src = get_data(); |
2983 | const char32_t *str = p_str.get_data(); |
2984 | |
2985 | for (int i = p_from; i <= (len - src_len); i++) { |
2986 | bool found = true; |
2987 | for (int j = 0; j < src_len; j++) { |
2988 | int read_pos = i + j; |
2989 | |
2990 | if (read_pos >= len) { |
2991 | ERR_PRINT("read_pos>=len" ); |
2992 | return -1; |
2993 | } |
2994 | |
2995 | if (src[read_pos] != str[j]) { |
2996 | found = false; |
2997 | break; |
2998 | } |
2999 | } |
3000 | |
3001 | if (found) { |
3002 | return i; |
3003 | } |
3004 | } |
3005 | |
3006 | return -1; |
3007 | } |
3008 | |
3009 | int String::find(const char *p_str, int p_from) const { |
3010 | if (p_from < 0) { |
3011 | return -1; |
3012 | } |
3013 | |
3014 | const int len = length(); |
3015 | |
3016 | if (len == 0) { |
3017 | return -1; // won't find anything! |
3018 | } |
3019 | |
3020 | const char32_t *src = get_data(); |
3021 | |
3022 | int src_len = 0; |
3023 | while (p_str[src_len] != '\0') { |
3024 | src_len++; |
3025 | } |
3026 | |
3027 | if (src_len == 1) { |
3028 | const char32_t needle = p_str[0]; |
3029 | |
3030 | for (int i = p_from; i < len; i++) { |
3031 | if (src[i] == needle) { |
3032 | return i; |
3033 | } |
3034 | } |
3035 | |
3036 | } else { |
3037 | for (int i = p_from; i <= (len - src_len); i++) { |
3038 | bool found = true; |
3039 | for (int j = 0; j < src_len; j++) { |
3040 | int read_pos = i + j; |
3041 | |
3042 | if (read_pos >= len) { |
3043 | ERR_PRINT("read_pos>=len" ); |
3044 | return -1; |
3045 | } |
3046 | |
3047 | if (src[read_pos] != (char32_t)p_str[j]) { |
3048 | found = false; |
3049 | break; |
3050 | } |
3051 | } |
3052 | |
3053 | if (found) { |
3054 | return i; |
3055 | } |
3056 | } |
3057 | } |
3058 | |
3059 | return -1; |
3060 | } |
3061 | |
3062 | int String::find_char(const char32_t &p_char, int p_from) const { |
3063 | return _cowdata.find(p_char, p_from); |
3064 | } |
3065 | |
3066 | int String::findmk(const Vector<String> &p_keys, int p_from, int *r_key) const { |
3067 | if (p_from < 0) { |
3068 | return -1; |
3069 | } |
3070 | if (p_keys.size() == 0) { |
3071 | return -1; |
3072 | } |
3073 | |
3074 | //int src_len=p_str.length(); |
3075 | const String *keys = &p_keys[0]; |
3076 | int key_count = p_keys.size(); |
3077 | int len = length(); |
3078 | |
3079 | if (len == 0) { |
3080 | return -1; // won't find anything! |
3081 | } |
3082 | |
3083 | const char32_t *src = get_data(); |
3084 | |
3085 | for (int i = p_from; i < len; i++) { |
3086 | bool found = true; |
3087 | for (int k = 0; k < key_count; k++) { |
3088 | found = true; |
3089 | if (r_key) { |
3090 | *r_key = k; |
3091 | } |
3092 | const char32_t *cmp = keys[k].get_data(); |
3093 | int l = keys[k].length(); |
3094 | |
3095 | for (int j = 0; j < l; j++) { |
3096 | int read_pos = i + j; |
3097 | |
3098 | if (read_pos >= len) { |
3099 | found = false; |
3100 | break; |
3101 | } |
3102 | |
3103 | if (src[read_pos] != cmp[j]) { |
3104 | found = false; |
3105 | break; |
3106 | } |
3107 | } |
3108 | if (found) { |
3109 | break; |
3110 | } |
3111 | } |
3112 | |
3113 | if (found) { |
3114 | return i; |
3115 | } |
3116 | } |
3117 | |
3118 | return -1; |
3119 | } |
3120 | |
3121 | int String::findn(const String &p_str, int p_from) const { |
3122 | if (p_from < 0) { |
3123 | return -1; |
3124 | } |
3125 | |
3126 | int src_len = p_str.length(); |
3127 | |
3128 | if (src_len == 0 || length() == 0) { |
3129 | return -1; // won't find anything! |
3130 | } |
3131 | |
3132 | const char32_t *srcd = get_data(); |
3133 | |
3134 | for (int i = p_from; i <= (length() - src_len); i++) { |
3135 | bool found = true; |
3136 | for (int j = 0; j < src_len; j++) { |
3137 | int read_pos = i + j; |
3138 | |
3139 | if (read_pos >= length()) { |
3140 | ERR_PRINT("read_pos>=length()" ); |
3141 | return -1; |
3142 | } |
3143 | |
3144 | char32_t src = _find_lower(srcd[read_pos]); |
3145 | char32_t dst = _find_lower(p_str[j]); |
3146 | |
3147 | if (src != dst) { |
3148 | found = false; |
3149 | break; |
3150 | } |
3151 | } |
3152 | |
3153 | if (found) { |
3154 | return i; |
3155 | } |
3156 | } |
3157 | |
3158 | return -1; |
3159 | } |
3160 | |
3161 | int String::rfind(const String &p_str, int p_from) const { |
3162 | // establish a limit |
3163 | int limit = length() - p_str.length(); |
3164 | if (limit < 0) { |
3165 | return -1; |
3166 | } |
3167 | |
3168 | // establish a starting point |
3169 | if (p_from < 0) { |
3170 | p_from = limit; |
3171 | } else if (p_from > limit) { |
3172 | p_from = limit; |
3173 | } |
3174 | |
3175 | int src_len = p_str.length(); |
3176 | int len = length(); |
3177 | |
3178 | if (src_len == 0 || len == 0) { |
3179 | return -1; // won't find anything! |
3180 | } |
3181 | |
3182 | const char32_t *src = get_data(); |
3183 | |
3184 | for (int i = p_from; i >= 0; i--) { |
3185 | bool found = true; |
3186 | for (int j = 0; j < src_len; j++) { |
3187 | int read_pos = i + j; |
3188 | |
3189 | if (read_pos >= len) { |
3190 | ERR_PRINT("read_pos>=len" ); |
3191 | return -1; |
3192 | } |
3193 | |
3194 | if (src[read_pos] != p_str[j]) { |
3195 | found = false; |
3196 | break; |
3197 | } |
3198 | } |
3199 | |
3200 | if (found) { |
3201 | return i; |
3202 | } |
3203 | } |
3204 | |
3205 | return -1; |
3206 | } |
3207 | |
3208 | int String::rfindn(const String &p_str, int p_from) const { |
3209 | // establish a limit |
3210 | int limit = length() - p_str.length(); |
3211 | if (limit < 0) { |
3212 | return -1; |
3213 | } |
3214 | |
3215 | // establish a starting point |
3216 | if (p_from < 0) { |
3217 | p_from = limit; |
3218 | } else if (p_from > limit) { |
3219 | p_from = limit; |
3220 | } |
3221 | |
3222 | int src_len = p_str.length(); |
3223 | int len = length(); |
3224 | |
3225 | if (src_len == 0 || len == 0) { |
3226 | return -1; // won't find anything! |
3227 | } |
3228 | |
3229 | const char32_t *src = get_data(); |
3230 | |
3231 | for (int i = p_from; i >= 0; i--) { |
3232 | bool found = true; |
3233 | for (int j = 0; j < src_len; j++) { |
3234 | int read_pos = i + j; |
3235 | |
3236 | if (read_pos >= len) { |
3237 | ERR_PRINT("read_pos>=len" ); |
3238 | return -1; |
3239 | } |
3240 | |
3241 | char32_t srcc = _find_lower(src[read_pos]); |
3242 | char32_t dstc = _find_lower(p_str[j]); |
3243 | |
3244 | if (srcc != dstc) { |
3245 | found = false; |
3246 | break; |
3247 | } |
3248 | } |
3249 | |
3250 | if (found) { |
3251 | return i; |
3252 | } |
3253 | } |
3254 | |
3255 | return -1; |
3256 | } |
3257 | |
3258 | bool String::ends_with(const String &p_string) const { |
3259 | int l = p_string.length(); |
3260 | if (l > length()) { |
3261 | return false; |
3262 | } |
3263 | |
3264 | if (l == 0) { |
3265 | return true; |
3266 | } |
3267 | |
3268 | const char32_t *p = &p_string[0]; |
3269 | const char32_t *s = &operator[](length() - l); |
3270 | |
3271 | for (int i = 0; i < l; i++) { |
3272 | if (p[i] != s[i]) { |
3273 | return false; |
3274 | } |
3275 | } |
3276 | |
3277 | return true; |
3278 | } |
3279 | |
3280 | bool String::begins_with(const String &p_string) const { |
3281 | int l = p_string.length(); |
3282 | if (l > length()) { |
3283 | return false; |
3284 | } |
3285 | |
3286 | if (l == 0) { |
3287 | return true; |
3288 | } |
3289 | |
3290 | const char32_t *p = &p_string[0]; |
3291 | const char32_t *s = &operator[](0); |
3292 | |
3293 | for (int i = 0; i < l; i++) { |
3294 | if (p[i] != s[i]) { |
3295 | return false; |
3296 | } |
3297 | } |
3298 | |
3299 | return true; |
3300 | } |
3301 | |
3302 | bool String::begins_with(const char *p_string) const { |
3303 | int l = length(); |
3304 | if (l == 0 || !p_string) { |
3305 | return false; |
3306 | } |
3307 | |
3308 | const char32_t *str = &operator[](0); |
3309 | int i = 0; |
3310 | |
3311 | while (*p_string && i < l) { |
3312 | if ((char32_t)*p_string != str[i]) { |
3313 | return false; |
3314 | } |
3315 | i++; |
3316 | p_string++; |
3317 | } |
3318 | |
3319 | return *p_string == 0; |
3320 | } |
3321 | |
3322 | bool String::is_enclosed_in(const String &p_string) const { |
3323 | return begins_with(p_string) && ends_with(p_string); |
3324 | } |
3325 | |
3326 | bool String::is_subsequence_of(const String &p_string) const { |
3327 | return _base_is_subsequence_of(p_string, false); |
3328 | } |
3329 | |
3330 | bool String::is_subsequence_ofn(const String &p_string) const { |
3331 | return _base_is_subsequence_of(p_string, true); |
3332 | } |
3333 | |
3334 | bool String::is_quoted() const { |
3335 | return is_enclosed_in("\"" ) || is_enclosed_in("'" ); |
3336 | } |
3337 | |
3338 | int String::_count(const String &p_string, int p_from, int p_to, bool p_case_insensitive) const { |
3339 | if (p_string.is_empty()) { |
3340 | return 0; |
3341 | } |
3342 | int len = length(); |
3343 | int slen = p_string.length(); |
3344 | if (len < slen) { |
3345 | return 0; |
3346 | } |
3347 | String str; |
3348 | if (p_from >= 0 && p_to >= 0) { |
3349 | if (p_to == 0) { |
3350 | p_to = len; |
3351 | } else if (p_from >= p_to) { |
3352 | return 0; |
3353 | } |
3354 | if (p_from == 0 && p_to == len) { |
3355 | str = String(); |
3356 | str.copy_from_unchecked(&get_data()[0], len); |
3357 | } else { |
3358 | str = substr(p_from, p_to - p_from); |
3359 | } |
3360 | } else { |
3361 | return 0; |
3362 | } |
3363 | int c = 0; |
3364 | int idx = -1; |
3365 | do { |
3366 | idx = p_case_insensitive ? str.findn(p_string) : str.find(p_string); |
3367 | if (idx != -1) { |
3368 | str = str.substr(idx + slen, str.length() - slen); |
3369 | ++c; |
3370 | } |
3371 | } while (idx != -1); |
3372 | return c; |
3373 | } |
3374 | |
3375 | int String::count(const String &p_string, int p_from, int p_to) const { |
3376 | return _count(p_string, p_from, p_to, false); |
3377 | } |
3378 | |
3379 | int String::countn(const String &p_string, int p_from, int p_to) const { |
3380 | return _count(p_string, p_from, p_to, true); |
3381 | } |
3382 | |
3383 | bool String::_base_is_subsequence_of(const String &p_string, bool case_insensitive) const { |
3384 | int len = length(); |
3385 | if (len == 0) { |
3386 | // Technically an empty string is subsequence of any string |
3387 | return true; |
3388 | } |
3389 | |
3390 | if (len > p_string.length()) { |
3391 | return false; |
3392 | } |
3393 | |
3394 | const char32_t *src = &operator[](0); |
3395 | const char32_t *tgt = &p_string[0]; |
3396 | |
3397 | for (; *src && *tgt; tgt++) { |
3398 | bool match = false; |
3399 | if (case_insensitive) { |
3400 | char32_t srcc = _find_lower(*src); |
3401 | char32_t tgtc = _find_lower(*tgt); |
3402 | match = srcc == tgtc; |
3403 | } else { |
3404 | match = *src == *tgt; |
3405 | } |
3406 | if (match) { |
3407 | src++; |
3408 | if (!*src) { |
3409 | return true; |
3410 | } |
3411 | } |
3412 | } |
3413 | |
3414 | return false; |
3415 | } |
3416 | |
3417 | Vector<String> String::bigrams() const { |
3418 | int n_pairs = length() - 1; |
3419 | Vector<String> b; |
3420 | if (n_pairs <= 0) { |
3421 | return b; |
3422 | } |
3423 | b.resize(n_pairs); |
3424 | for (int i = 0; i < n_pairs; i++) { |
3425 | b.write[i] = substr(i, 2); |
3426 | } |
3427 | return b; |
3428 | } |
3429 | |
3430 | // Similarity according to Sorensen-Dice coefficient |
3431 | float String::similarity(const String &p_string) const { |
3432 | if (operator==(p_string)) { |
3433 | // Equal strings are totally similar |
3434 | return 1.0f; |
3435 | } |
3436 | if (length() < 2 || p_string.length() < 2) { |
3437 | // No way to calculate similarity without a single bigram |
3438 | return 0.0f; |
3439 | } |
3440 | |
3441 | Vector<String> src_bigrams = bigrams(); |
3442 | Vector<String> tgt_bigrams = p_string.bigrams(); |
3443 | |
3444 | int src_size = src_bigrams.size(); |
3445 | int tgt_size = tgt_bigrams.size(); |
3446 | |
3447 | int sum = src_size + tgt_size; |
3448 | int inter = 0; |
3449 | for (int i = 0; i < src_size; i++) { |
3450 | for (int j = 0; j < tgt_size; j++) { |
3451 | if (src_bigrams[i] == tgt_bigrams[j]) { |
3452 | inter++; |
3453 | break; |
3454 | } |
3455 | } |
3456 | } |
3457 | |
3458 | return (2.0f * inter) / sum; |
3459 | } |
3460 | |
3461 | static bool _wildcard_match(const char32_t *p_pattern, const char32_t *p_string, bool p_case_sensitive) { |
3462 | switch (*p_pattern) { |
3463 | case '\0': |
3464 | return !*p_string; |
3465 | case '*': |
3466 | return _wildcard_match(p_pattern + 1, p_string, p_case_sensitive) || (*p_string && _wildcard_match(p_pattern, p_string + 1, p_case_sensitive)); |
3467 | case '?': |
3468 | return *p_string && (*p_string != '.') && _wildcard_match(p_pattern + 1, p_string + 1, p_case_sensitive); |
3469 | default: |
3470 | |
3471 | return (p_case_sensitive ? (*p_string == *p_pattern) : (_find_upper(*p_string) == _find_upper(*p_pattern))) && _wildcard_match(p_pattern + 1, p_string + 1, p_case_sensitive); |
3472 | } |
3473 | } |
3474 | |
3475 | bool String::match(const String &p_wildcard) const { |
3476 | if (!p_wildcard.length() || !length()) { |
3477 | return false; |
3478 | } |
3479 | |
3480 | return _wildcard_match(p_wildcard.get_data(), get_data(), true); |
3481 | } |
3482 | |
3483 | bool String::matchn(const String &p_wildcard) const { |
3484 | if (!p_wildcard.length() || !length()) { |
3485 | return false; |
3486 | } |
3487 | return _wildcard_match(p_wildcard.get_data(), get_data(), false); |
3488 | } |
3489 | |
3490 | String String::format(const Variant &values, String placeholder) const { |
3491 | String new_string = String(this->ptr()); |
3492 | |
3493 | if (values.get_type() == Variant::ARRAY) { |
3494 | Array values_arr = values; |
3495 | |
3496 | for (int i = 0; i < values_arr.size(); i++) { |
3497 | String i_as_str = String::num_int64(i); |
3498 | |
3499 | if (values_arr[i].get_type() == Variant::ARRAY) { //Array in Array structure [["name","RobotGuy"],[0,"godot"],["strength",9000.91]] |
3500 | Array value_arr = values_arr[i]; |
3501 | |
3502 | if (value_arr.size() == 2) { |
3503 | Variant v_key = value_arr[0]; |
3504 | String key = v_key; |
3505 | |
3506 | Variant v_val = value_arr[1]; |
3507 | String val = v_val; |
3508 | |
3509 | new_string = new_string.replace(placeholder.replace("_" , key), val); |
3510 | } else { |
3511 | ERR_PRINT(String("STRING.format Inner Array size != 2 " ).ascii().get_data()); |
3512 | } |
3513 | } else { //Array structure ["RobotGuy","Logis","rookie"] |
3514 | Variant v_val = values_arr[i]; |
3515 | String val = v_val; |
3516 | |
3517 | if (placeholder.find("_" ) > -1) { |
3518 | new_string = new_string.replace(placeholder.replace("_" , i_as_str), val); |
3519 | } else { |
3520 | new_string = new_string.replace_first(placeholder, val); |
3521 | } |
3522 | } |
3523 | } |
3524 | } else if (values.get_type() == Variant::DICTIONARY) { |
3525 | Dictionary d = values; |
3526 | List<Variant> keys; |
3527 | d.get_key_list(&keys); |
3528 | |
3529 | for (const Variant &key : keys) { |
3530 | new_string = new_string.replace(placeholder.replace("_" , key), d[key]); |
3531 | } |
3532 | } else { |
3533 | ERR_PRINT(String("Invalid type: use Array or Dictionary." ).ascii().get_data()); |
3534 | } |
3535 | |
3536 | return new_string; |
3537 | } |
3538 | |
3539 | String String::replace(const String &p_key, const String &p_with) const { |
3540 | String new_string; |
3541 | int search_from = 0; |
3542 | int result = 0; |
3543 | |
3544 | while ((result = find(p_key, search_from)) >= 0) { |
3545 | new_string += substr(search_from, result - search_from); |
3546 | new_string += p_with; |
3547 | search_from = result + p_key.length(); |
3548 | } |
3549 | |
3550 | if (search_from == 0) { |
3551 | return *this; |
3552 | } |
3553 | |
3554 | new_string += substr(search_from, length() - search_from); |
3555 | |
3556 | return new_string; |
3557 | } |
3558 | |
3559 | String String::replace(const char *p_key, const char *p_with) const { |
3560 | String new_string; |
3561 | int search_from = 0; |
3562 | int result = 0; |
3563 | |
3564 | while ((result = find(p_key, search_from)) >= 0) { |
3565 | new_string += substr(search_from, result - search_from); |
3566 | new_string += p_with; |
3567 | int k = 0; |
3568 | while (p_key[k] != '\0') { |
3569 | k++; |
3570 | } |
3571 | search_from = result + k; |
3572 | } |
3573 | |
3574 | if (search_from == 0) { |
3575 | return *this; |
3576 | } |
3577 | |
3578 | new_string += substr(search_from, length() - search_from); |
3579 | |
3580 | return new_string; |
3581 | } |
3582 | |
3583 | String String::replace_first(const String &p_key, const String &p_with) const { |
3584 | int pos = find(p_key); |
3585 | if (pos >= 0) { |
3586 | return substr(0, pos) + p_with + substr(pos + p_key.length(), length()); |
3587 | } |
3588 | |
3589 | return *this; |
3590 | } |
3591 | |
3592 | String String::replacen(const String &p_key, const String &p_with) const { |
3593 | String new_string; |
3594 | int search_from = 0; |
3595 | int result = 0; |
3596 | |
3597 | while ((result = findn(p_key, search_from)) >= 0) { |
3598 | new_string += substr(search_from, result - search_from); |
3599 | new_string += p_with; |
3600 | search_from = result + p_key.length(); |
3601 | } |
3602 | |
3603 | if (search_from == 0) { |
3604 | return *this; |
3605 | } |
3606 | |
3607 | new_string += substr(search_from, length() - search_from); |
3608 | return new_string; |
3609 | } |
3610 | |
3611 | String String::repeat(int p_count) const { |
3612 | ERR_FAIL_COND_V_MSG(p_count < 0, "" , "Parameter count should be a positive number." ); |
3613 | |
3614 | if (p_count == 0) { |
3615 | return "" ; |
3616 | } |
3617 | |
3618 | if (p_count == 1) { |
3619 | return *this; |
3620 | } |
3621 | |
3622 | int len = length(); |
3623 | String new_string = *this; |
3624 | new_string.resize(p_count * len + 1); |
3625 | |
3626 | char32_t *dst = new_string.ptrw(); |
3627 | int offset = 1; |
3628 | int stride = 1; |
3629 | while (offset < p_count) { |
3630 | memcpy(dst + offset * len, dst, stride * len * sizeof(char32_t)); |
3631 | offset += stride; |
3632 | stride = MIN(stride * 2, p_count - offset); |
3633 | } |
3634 | dst[p_count * len] = _null; |
3635 | return new_string; |
3636 | } |
3637 | |
3638 | String String::reverse() const { |
3639 | int len = length(); |
3640 | if (len <= 1) { |
3641 | return *this; |
3642 | } |
3643 | String new_string; |
3644 | new_string.resize(len + 1); |
3645 | |
3646 | const char32_t *src = ptr(); |
3647 | char32_t *dst = new_string.ptrw(); |
3648 | for (int i = 0; i < len; i++) { |
3649 | dst[i] = src[len - i - 1]; |
3650 | } |
3651 | dst[len] = _null; |
3652 | return new_string; |
3653 | } |
3654 | |
3655 | String String::left(int p_len) const { |
3656 | if (p_len < 0) { |
3657 | p_len = length() + p_len; |
3658 | } |
3659 | |
3660 | if (p_len <= 0) { |
3661 | return "" ; |
3662 | } |
3663 | |
3664 | if (p_len >= length()) { |
3665 | return *this; |
3666 | } |
3667 | |
3668 | String s; |
3669 | s.copy_from_unchecked(&get_data()[0], p_len); |
3670 | return s; |
3671 | } |
3672 | |
3673 | String String::right(int p_len) const { |
3674 | if (p_len < 0) { |
3675 | p_len = length() + p_len; |
3676 | } |
3677 | |
3678 | if (p_len <= 0) { |
3679 | return "" ; |
3680 | } |
3681 | |
3682 | if (p_len >= length()) { |
3683 | return *this; |
3684 | } |
3685 | |
3686 | String s; |
3687 | s.copy_from_unchecked(&get_data()[length() - p_len], p_len); |
3688 | return s; |
3689 | } |
3690 | |
3691 | char32_t String::unicode_at(int p_idx) const { |
3692 | ERR_FAIL_INDEX_V(p_idx, length(), 0); |
3693 | return operator[](p_idx); |
3694 | } |
3695 | |
3696 | String String::indent(const String &p_prefix) const { |
3697 | String new_string; |
3698 | int line_start = 0; |
3699 | |
3700 | for (int i = 0; i < length(); i++) { |
3701 | const char32_t c = operator[](i); |
3702 | if (c == '\n') { |
3703 | if (i == line_start) { |
3704 | new_string += c; // Leave empty lines empty. |
3705 | } else { |
3706 | new_string += p_prefix + substr(line_start, i - line_start + 1); |
3707 | } |
3708 | line_start = i + 1; |
3709 | } |
3710 | } |
3711 | if (line_start != length()) { |
3712 | new_string += p_prefix + substr(line_start); |
3713 | } |
3714 | return new_string; |
3715 | } |
3716 | |
3717 | String String::dedent() const { |
3718 | String new_string; |
3719 | String indent; |
3720 | bool has_indent = false; |
3721 | bool has_text = false; |
3722 | int line_start = 0; |
3723 | int indent_stop = -1; |
3724 | |
3725 | for (int i = 0; i < length(); i++) { |
3726 | char32_t c = operator[](i); |
3727 | if (c == '\n') { |
3728 | if (has_text) { |
3729 | new_string += substr(indent_stop, i - indent_stop); |
3730 | } |
3731 | new_string += "\n" ; |
3732 | has_text = false; |
3733 | line_start = i + 1; |
3734 | indent_stop = -1; |
3735 | } else if (!has_text) { |
3736 | if (c > 32) { |
3737 | has_text = true; |
3738 | if (!has_indent) { |
3739 | has_indent = true; |
3740 | indent = substr(line_start, i - line_start); |
3741 | indent_stop = i; |
3742 | } |
3743 | } |
3744 | if (has_indent && indent_stop < 0) { |
3745 | int j = i - line_start; |
3746 | if (j >= indent.length() || c != indent[j]) { |
3747 | indent_stop = i; |
3748 | } |
3749 | } |
3750 | } |
3751 | } |
3752 | |
3753 | if (has_text) { |
3754 | new_string += substr(indent_stop, length() - indent_stop); |
3755 | } |
3756 | |
3757 | return new_string; |
3758 | } |
3759 | |
3760 | String String::strip_edges(bool left, bool right) const { |
3761 | int len = length(); |
3762 | int beg = 0, end = len; |
3763 | |
3764 | if (left) { |
3765 | for (int i = 0; i < len; i++) { |
3766 | if (operator[](i) <= 32) { |
3767 | beg++; |
3768 | } else { |
3769 | break; |
3770 | } |
3771 | } |
3772 | } |
3773 | |
3774 | if (right) { |
3775 | for (int i = len - 1; i >= 0; i--) { |
3776 | if (operator[](i) <= 32) { |
3777 | end--; |
3778 | } else { |
3779 | break; |
3780 | } |
3781 | } |
3782 | } |
3783 | |
3784 | if (beg == 0 && end == len) { |
3785 | return *this; |
3786 | } |
3787 | |
3788 | return substr(beg, end - beg); |
3789 | } |
3790 | |
3791 | String String::strip_escapes() const { |
3792 | String new_string; |
3793 | for (int i = 0; i < length(); i++) { |
3794 | // Escape characters on first page of the ASCII table, before 32 (Space). |
3795 | if (operator[](i) < 32) { |
3796 | continue; |
3797 | } |
3798 | new_string += operator[](i); |
3799 | } |
3800 | |
3801 | return new_string; |
3802 | } |
3803 | |
3804 | String String::lstrip(const String &p_chars) const { |
3805 | int len = length(); |
3806 | int beg; |
3807 | |
3808 | for (beg = 0; beg < len; beg++) { |
3809 | if (p_chars.find_char(get(beg)) == -1) { |
3810 | break; |
3811 | } |
3812 | } |
3813 | |
3814 | if (beg == 0) { |
3815 | return *this; |
3816 | } |
3817 | |
3818 | return substr(beg, len - beg); |
3819 | } |
3820 | |
3821 | String String::rstrip(const String &p_chars) const { |
3822 | int len = length(); |
3823 | int end; |
3824 | |
3825 | for (end = len - 1; end >= 0; end--) { |
3826 | if (p_chars.find_char(get(end)) == -1) { |
3827 | break; |
3828 | } |
3829 | } |
3830 | |
3831 | if (end == len - 1) { |
3832 | return *this; |
3833 | } |
3834 | |
3835 | return substr(0, end + 1); |
3836 | } |
3837 | |
3838 | bool String::is_network_share_path() const { |
3839 | return begins_with("//" ) || begins_with("\\\\" ); |
3840 | } |
3841 | |
3842 | String String::simplify_path() const { |
3843 | String s = *this; |
3844 | String drive; |
3845 | |
3846 | // Check if we have a special path (like res://) or a protocol identifier. |
3847 | int p = s.find("://" ); |
3848 | bool found = false; |
3849 | if (p > 0) { |
3850 | bool only_chars = true; |
3851 | for (int i = 0; i < p; i++) { |
3852 | if (!is_ascii_alphanumeric_char(s[i])) { |
3853 | only_chars = false; |
3854 | break; |
3855 | } |
3856 | } |
3857 | if (only_chars) { |
3858 | found = true; |
3859 | drive = s.substr(0, p + 3); |
3860 | s = s.substr(p + 3); |
3861 | } |
3862 | } |
3863 | if (!found) { |
3864 | if (is_network_share_path()) { |
3865 | // Network path, beginning with // or \\. |
3866 | drive = s.substr(0, 2); |
3867 | s = s.substr(2); |
3868 | } else if (s.begins_with("/" ) || s.begins_with("\\" )) { |
3869 | // Absolute path. |
3870 | drive = s.substr(0, 1); |
3871 | s = s.substr(1); |
3872 | } else { |
3873 | // Windows-style drive path, like C:/ or C:\. |
3874 | p = s.find(":/" ); |
3875 | if (p == -1) { |
3876 | p = s.find(":\\" ); |
3877 | } |
3878 | if (p != -1 && p < s.find("/" )) { |
3879 | drive = s.substr(0, p + 2); |
3880 | s = s.substr(p + 2); |
3881 | } |
3882 | } |
3883 | } |
3884 | |
3885 | s = s.replace("\\" , "/" ); |
3886 | while (true) { // in case of using 2 or more slash |
3887 | String compare = s.replace("//" , "/" ); |
3888 | if (s == compare) { |
3889 | break; |
3890 | } else { |
3891 | s = compare; |
3892 | } |
3893 | } |
3894 | Vector<String> dirs = s.split("/" , false); |
3895 | |
3896 | for (int i = 0; i < dirs.size(); i++) { |
3897 | String d = dirs[i]; |
3898 | if (d == "." ) { |
3899 | dirs.remove_at(i); |
3900 | i--; |
3901 | } else if (d == ".." ) { |
3902 | if (i == 0) { |
3903 | dirs.remove_at(i); |
3904 | i--; |
3905 | } else { |
3906 | dirs.remove_at(i); |
3907 | dirs.remove_at(i - 1); |
3908 | i -= 2; |
3909 | } |
3910 | } |
3911 | } |
3912 | |
3913 | s = "" ; |
3914 | |
3915 | for (int i = 0; i < dirs.size(); i++) { |
3916 | if (i > 0) { |
3917 | s += "/" ; |
3918 | } |
3919 | s += dirs[i]; |
3920 | } |
3921 | |
3922 | return drive + s; |
3923 | } |
3924 | |
3925 | static int _humanize_digits(int p_num) { |
3926 | if (p_num < 100) { |
3927 | return 2; |
3928 | } else if (p_num < 1024) { |
3929 | return 1; |
3930 | } else { |
3931 | return 0; |
3932 | } |
3933 | } |
3934 | |
3935 | String String::humanize_size(uint64_t p_size) { |
3936 | uint64_t _div = 1; |
3937 | Vector<String> prefixes; |
3938 | prefixes.push_back(RTR("B" )); |
3939 | prefixes.push_back(RTR("KiB" )); |
3940 | prefixes.push_back(RTR("MiB" )); |
3941 | prefixes.push_back(RTR("GiB" )); |
3942 | prefixes.push_back(RTR("TiB" )); |
3943 | prefixes.push_back(RTR("PiB" )); |
3944 | prefixes.push_back(RTR("EiB" )); |
3945 | |
3946 | int prefix_idx = 0; |
3947 | |
3948 | while (prefix_idx < prefixes.size() - 1 && p_size > (_div * 1024)) { |
3949 | _div *= 1024; |
3950 | prefix_idx++; |
3951 | } |
3952 | |
3953 | const int digits = prefix_idx > 0 ? _humanize_digits(p_size / _div) : 0; |
3954 | const double divisor = prefix_idx > 0 ? _div : 1; |
3955 | |
3956 | return String::num(p_size / divisor).pad_decimals(digits) + " " + prefixes[prefix_idx]; |
3957 | } |
3958 | |
3959 | bool String::is_absolute_path() const { |
3960 | if (length() > 1) { |
3961 | return (operator[](0) == '/' || operator[](0) == '\\' || find(":/" ) != -1 || find(":\\" ) != -1); |
3962 | } else if ((length()) == 1) { |
3963 | return (operator[](0) == '/' || operator[](0) == '\\'); |
3964 | } else { |
3965 | return false; |
3966 | } |
3967 | } |
3968 | |
3969 | static _FORCE_INLINE_ bool _is_valid_identifier_bit(int p_index, char32_t p_char) { |
3970 | if (p_index == 0 && is_digit(p_char)) { |
3971 | return false; // No start with number plz. |
3972 | } |
3973 | return is_ascii_identifier_char(p_char); |
3974 | } |
3975 | |
3976 | String String::validate_identifier() const { |
3977 | if (is_empty()) { |
3978 | return "_" ; // Empty string is not a valid identifier; |
3979 | } |
3980 | |
3981 | String result = *this; |
3982 | int len = result.length(); |
3983 | char32_t *buffer = result.ptrw(); |
3984 | |
3985 | for (int i = 0; i < len; i++) { |
3986 | if (!_is_valid_identifier_bit(i, buffer[i])) { |
3987 | buffer[i] = '_'; |
3988 | } |
3989 | } |
3990 | |
3991 | return result; |
3992 | } |
3993 | |
3994 | bool String::is_valid_identifier() const { |
3995 | int len = length(); |
3996 | |
3997 | if (len == 0) { |
3998 | return false; |
3999 | } |
4000 | |
4001 | const char32_t *str = &operator[](0); |
4002 | |
4003 | for (int i = 0; i < len; i++) { |
4004 | if (!_is_valid_identifier_bit(i, str[i])) { |
4005 | return false; |
4006 | } |
4007 | } |
4008 | |
4009 | return true; |
4010 | } |
4011 | |
4012 | bool String::is_valid_string() const { |
4013 | int l = length(); |
4014 | const char32_t *src = get_data(); |
4015 | bool valid = true; |
4016 | for (int i = 0; i < l; i++) { |
4017 | valid = valid && (src[i] < 0xd800 || (src[i] > 0xdfff && src[i] <= 0x10ffff)); |
4018 | } |
4019 | return valid; |
4020 | } |
4021 | |
4022 | String String::uri_encode() const { |
4023 | const CharString temp = utf8(); |
4024 | String res; |
4025 | for (int i = 0; i < temp.length(); ++i) { |
4026 | uint8_t ord = temp[i]; |
4027 | if (ord == '.' || ord == '-' || ord == '~' || is_ascii_identifier_char(ord)) { |
4028 | res += ord; |
4029 | } else { |
4030 | char p[4] = { '%', 0, 0, 0 }; |
4031 | static const char hex[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' }; |
4032 | p[1] = hex[ord >> 4]; |
4033 | p[2] = hex[ord & 0xF]; |
4034 | res += p; |
4035 | } |
4036 | } |
4037 | return res; |
4038 | } |
4039 | |
4040 | String String::uri_decode() const { |
4041 | CharString src = utf8(); |
4042 | CharString res; |
4043 | for (int i = 0; i < src.length(); ++i) { |
4044 | if (src[i] == '%' && i + 2 < src.length()) { |
4045 | char ord1 = src[i + 1]; |
4046 | if (is_digit(ord1) || is_ascii_upper_case(ord1)) { |
4047 | char ord2 = src[i + 2]; |
4048 | if (is_digit(ord2) || is_ascii_upper_case(ord2)) { |
4049 | char bytes[3] = { (char)ord1, (char)ord2, 0 }; |
4050 | res += (char)strtol(bytes, nullptr, 16); |
4051 | i += 2; |
4052 | } |
4053 | } else { |
4054 | res += src[i]; |
4055 | } |
4056 | } else if (src[i] == '+') { |
4057 | res += ' '; |
4058 | } else { |
4059 | res += src[i]; |
4060 | } |
4061 | } |
4062 | return String::utf8(res); |
4063 | } |
4064 | |
4065 | String String::c_unescape() const { |
4066 | String escaped = *this; |
4067 | escaped = escaped.replace("\\a" , "\a" ); |
4068 | escaped = escaped.replace("\\b" , "\b" ); |
4069 | escaped = escaped.replace("\\f" , "\f" ); |
4070 | escaped = escaped.replace("\\n" , "\n" ); |
4071 | escaped = escaped.replace("\\r" , "\r" ); |
4072 | escaped = escaped.replace("\\t" , "\t" ); |
4073 | escaped = escaped.replace("\\v" , "\v" ); |
4074 | escaped = escaped.replace("\\'" , "\'" ); |
4075 | escaped = escaped.replace("\\\"" , "\"" ); |
4076 | escaped = escaped.replace("\\\\" , "\\" ); |
4077 | |
4078 | return escaped; |
4079 | } |
4080 | |
4081 | String String::c_escape() const { |
4082 | String escaped = *this; |
4083 | escaped = escaped.replace("\\" , "\\\\" ); |
4084 | escaped = escaped.replace("\a" , "\\a" ); |
4085 | escaped = escaped.replace("\b" , "\\b" ); |
4086 | escaped = escaped.replace("\f" , "\\f" ); |
4087 | escaped = escaped.replace("\n" , "\\n" ); |
4088 | escaped = escaped.replace("\r" , "\\r" ); |
4089 | escaped = escaped.replace("\t" , "\\t" ); |
4090 | escaped = escaped.replace("\v" , "\\v" ); |
4091 | escaped = escaped.replace("\'" , "\\'" ); |
4092 | escaped = escaped.replace("\"" , "\\\"" ); |
4093 | |
4094 | return escaped; |
4095 | } |
4096 | |
4097 | String String::c_escape_multiline() const { |
4098 | String escaped = *this; |
4099 | escaped = escaped.replace("\\" , "\\\\" ); |
4100 | escaped = escaped.replace("\"" , "\\\"" ); |
4101 | |
4102 | return escaped; |
4103 | } |
4104 | |
4105 | String String::json_escape() const { |
4106 | String escaped = *this; |
4107 | escaped = escaped.replace("\\" , "\\\\" ); |
4108 | escaped = escaped.replace("\b" , "\\b" ); |
4109 | escaped = escaped.replace("\f" , "\\f" ); |
4110 | escaped = escaped.replace("\n" , "\\n" ); |
4111 | escaped = escaped.replace("\r" , "\\r" ); |
4112 | escaped = escaped.replace("\t" , "\\t" ); |
4113 | escaped = escaped.replace("\v" , "\\v" ); |
4114 | escaped = escaped.replace("\"" , "\\\"" ); |
4115 | |
4116 | return escaped; |
4117 | } |
4118 | |
4119 | String String::xml_escape(bool p_escape_quotes) const { |
4120 | String str = *this; |
4121 | str = str.replace("&" , "&" ); |
4122 | str = str.replace("<" , "<" ); |
4123 | str = str.replace(">" , ">" ); |
4124 | if (p_escape_quotes) { |
4125 | str = str.replace("'" , "'" ); |
4126 | str = str.replace("\"" , """ ); |
4127 | } |
4128 | /* |
4129 | for (int i=1;i<32;i++) { |
4130 | char chr[2]={i,0}; |
4131 | str=str.replace(chr,"&#"+String::num(i)+";"); |
4132 | }*/ |
4133 | return str; |
4134 | } |
4135 | |
4136 | static _FORCE_INLINE_ int _xml_unescape(const char32_t *p_src, int p_src_len, char32_t *p_dst) { |
4137 | int len = 0; |
4138 | while (p_src_len) { |
4139 | if (*p_src == '&') { |
4140 | int eat = 0; |
4141 | |
4142 | if (p_src_len >= 4 && p_src[1] == '#') { |
4143 | char32_t c = 0; |
4144 | bool overflow = false; |
4145 | if (p_src[2] == 'x') { |
4146 | // Hex entity &#x<num>; |
4147 | for (int i = 3; i < p_src_len; i++) { |
4148 | eat = i + 1; |
4149 | char32_t ct = p_src[i]; |
4150 | if (ct == ';') { |
4151 | break; |
4152 | } else if (is_digit(ct)) { |
4153 | ct = ct - '0'; |
4154 | } else if (ct >= 'a' && ct <= 'f') { |
4155 | ct = (ct - 'a') + 10; |
4156 | } else if (ct >= 'A' && ct <= 'F') { |
4157 | ct = (ct - 'A') + 10; |
4158 | } else { |
4159 | break; |
4160 | } |
4161 | if (c > (UINT32_MAX >> 4)) { |
4162 | overflow = true; |
4163 | break; |
4164 | } |
4165 | c <<= 4; |
4166 | c |= ct; |
4167 | } |
4168 | } else { |
4169 | // Decimal entity &#<num>; |
4170 | for (int i = 2; i < p_src_len; i++) { |
4171 | eat = i + 1; |
4172 | char32_t ct = p_src[i]; |
4173 | if (ct == ';' || !is_digit(ct)) { |
4174 | break; |
4175 | } |
4176 | } |
4177 | if (p_src[eat - 1] == ';') { |
4178 | int64_t val = String::to_int(p_src + 2, eat - 3); |
4179 | if (val > 0 && val <= UINT32_MAX) { |
4180 | c = (char32_t)val; |
4181 | } else { |
4182 | overflow = true; |
4183 | } |
4184 | } |
4185 | } |
4186 | |
4187 | // Value must be non-zero, in the range of char32_t, |
4188 | // actually end with ';'. If invalid, leave the entity as-is |
4189 | if (c == '\0' || overflow || p_src[eat - 1] != ';') { |
4190 | eat = 1; |
4191 | c = *p_src; |
4192 | } |
4193 | if (p_dst) { |
4194 | *p_dst = c; |
4195 | } |
4196 | |
4197 | } else if (p_src_len >= 4 && p_src[1] == 'g' && p_src[2] == 't' && p_src[3] == ';') { |
4198 | if (p_dst) { |
4199 | *p_dst = '>'; |
4200 | } |
4201 | eat = 4; |
4202 | } else if (p_src_len >= 4 && p_src[1] == 'l' && p_src[2] == 't' && p_src[3] == ';') { |
4203 | if (p_dst) { |
4204 | *p_dst = '<'; |
4205 | } |
4206 | eat = 4; |
4207 | } else if (p_src_len >= 5 && p_src[1] == 'a' && p_src[2] == 'm' && p_src[3] == 'p' && p_src[4] == ';') { |
4208 | if (p_dst) { |
4209 | *p_dst = '&'; |
4210 | } |
4211 | eat = 5; |
4212 | } else if (p_src_len >= 6 && p_src[1] == 'q' && p_src[2] == 'u' && p_src[3] == 'o' && p_src[4] == 't' && p_src[5] == ';') { |
4213 | if (p_dst) { |
4214 | *p_dst = '"'; |
4215 | } |
4216 | eat = 6; |
4217 | } else if (p_src_len >= 6 && p_src[1] == 'a' && p_src[2] == 'p' && p_src[3] == 'o' && p_src[4] == 's' && p_src[5] == ';') { |
4218 | if (p_dst) { |
4219 | *p_dst = '\''; |
4220 | } |
4221 | eat = 6; |
4222 | } else { |
4223 | if (p_dst) { |
4224 | *p_dst = *p_src; |
4225 | } |
4226 | eat = 1; |
4227 | } |
4228 | |
4229 | if (p_dst) { |
4230 | p_dst++; |
4231 | } |
4232 | |
4233 | len++; |
4234 | p_src += eat; |
4235 | p_src_len -= eat; |
4236 | } else { |
4237 | if (p_dst) { |
4238 | *p_dst = *p_src; |
4239 | p_dst++; |
4240 | } |
4241 | len++; |
4242 | p_src++; |
4243 | p_src_len--; |
4244 | } |
4245 | } |
4246 | |
4247 | return len; |
4248 | } |
4249 | |
4250 | String String::xml_unescape() const { |
4251 | String str; |
4252 | int l = length(); |
4253 | int len = _xml_unescape(get_data(), l, nullptr); |
4254 | if (len == 0) { |
4255 | return String(); |
4256 | } |
4257 | str.resize(len + 1); |
4258 | _xml_unescape(get_data(), l, str.ptrw()); |
4259 | str[len] = 0; |
4260 | return str; |
4261 | } |
4262 | |
4263 | String String::pad_decimals(int p_digits) const { |
4264 | String s = *this; |
4265 | int c = s.find("." ); |
4266 | |
4267 | if (c == -1) { |
4268 | if (p_digits <= 0) { |
4269 | return s; |
4270 | } |
4271 | s += "." ; |
4272 | c = s.length() - 1; |
4273 | } else { |
4274 | if (p_digits <= 0) { |
4275 | return s.substr(0, c); |
4276 | } |
4277 | } |
4278 | |
4279 | if (s.length() - (c + 1) > p_digits) { |
4280 | return s.substr(0, c + p_digits + 1); |
4281 | } else { |
4282 | int zeros_to_add = p_digits - s.length() + (c + 1); |
4283 | return s + String("0" ).repeat(zeros_to_add); |
4284 | } |
4285 | } |
4286 | |
4287 | String String::pad_zeros(int p_digits) const { |
4288 | String s = *this; |
4289 | int end = s.find("." ); |
4290 | |
4291 | if (end == -1) { |
4292 | end = s.length(); |
4293 | } |
4294 | |
4295 | if (end == 0) { |
4296 | return s; |
4297 | } |
4298 | |
4299 | int begin = 0; |
4300 | |
4301 | while (begin < end && !is_digit(s[begin])) { |
4302 | begin++; |
4303 | } |
4304 | |
4305 | int zeros_to_add = p_digits - (end - begin); |
4306 | |
4307 | if (zeros_to_add <= 0) { |
4308 | return s; |
4309 | } else { |
4310 | return s.insert(begin, String("0" ).repeat(zeros_to_add)); |
4311 | } |
4312 | } |
4313 | |
4314 | String String::trim_prefix(const String &p_prefix) const { |
4315 | String s = *this; |
4316 | if (s.begins_with(p_prefix)) { |
4317 | return s.substr(p_prefix.length(), s.length() - p_prefix.length()); |
4318 | } |
4319 | return s; |
4320 | } |
4321 | |
4322 | String String::trim_suffix(const String &p_suffix) const { |
4323 | String s = *this; |
4324 | if (s.ends_with(p_suffix)) { |
4325 | return s.substr(0, s.length() - p_suffix.length()); |
4326 | } |
4327 | return s; |
4328 | } |
4329 | |
4330 | bool String::is_valid_int() const { |
4331 | int len = length(); |
4332 | |
4333 | if (len == 0) { |
4334 | return false; |
4335 | } |
4336 | |
4337 | int from = 0; |
4338 | if (len != 1 && (operator[](0) == '+' || operator[](0) == '-')) { |
4339 | from++; |
4340 | } |
4341 | |
4342 | for (int i = from; i < len; i++) { |
4343 | if (!is_digit(operator[](i))) { |
4344 | return false; // no start with number plz |
4345 | } |
4346 | } |
4347 | |
4348 | return true; |
4349 | } |
4350 | |
4351 | bool String::is_valid_hex_number(bool p_with_prefix) const { |
4352 | int len = length(); |
4353 | |
4354 | if (len == 0) { |
4355 | return false; |
4356 | } |
4357 | |
4358 | int from = 0; |
4359 | if (len != 1 && (operator[](0) == '+' || operator[](0) == '-')) { |
4360 | from++; |
4361 | } |
4362 | |
4363 | if (p_with_prefix) { |
4364 | if (len < 3) { |
4365 | return false; |
4366 | } |
4367 | if (operator[](from) != '0' || operator[](from + 1) != 'x') { |
4368 | return false; |
4369 | } |
4370 | from += 2; |
4371 | } |
4372 | |
4373 | for (int i = from; i < len; i++) { |
4374 | char32_t c = operator[](i); |
4375 | if (is_hex_digit(c)) { |
4376 | continue; |
4377 | } |
4378 | return false; |
4379 | } |
4380 | |
4381 | return true; |
4382 | } |
4383 | |
4384 | bool String::is_valid_float() const { |
4385 | int len = length(); |
4386 | |
4387 | if (len == 0) { |
4388 | return false; |
4389 | } |
4390 | |
4391 | int from = 0; |
4392 | if (operator[](0) == '+' || operator[](0) == '-') { |
4393 | from++; |
4394 | } |
4395 | |
4396 | bool exponent_found = false; |
4397 | bool period_found = false; |
4398 | bool sign_found = false; |
4399 | bool exponent_values_found = false; |
4400 | bool numbers_found = false; |
4401 | |
4402 | for (int i = from; i < len; i++) { |
4403 | if (is_digit(operator[](i))) { |
4404 | if (exponent_found) { |
4405 | exponent_values_found = true; |
4406 | } else { |
4407 | numbers_found = true; |
4408 | } |
4409 | } else if (numbers_found && !exponent_found && operator[](i) == 'e') { |
4410 | exponent_found = true; |
4411 | } else if (!period_found && !exponent_found && operator[](i) == '.') { |
4412 | period_found = true; |
4413 | } else if ((operator[](i) == '-' || operator[](i) == '+') && exponent_found && !exponent_values_found && !sign_found) { |
4414 | sign_found = true; |
4415 | } else { |
4416 | return false; // no start with number plz |
4417 | } |
4418 | } |
4419 | |
4420 | return numbers_found; |
4421 | } |
4422 | |
4423 | String String::path_to_file(const String &p_path) const { |
4424 | // Don't get base dir for src, this is expected to be a dir already. |
4425 | String src = this->replace("\\" , "/" ); |
4426 | String dst = p_path.replace("\\" , "/" ).get_base_dir(); |
4427 | String rel = src.path_to(dst); |
4428 | if (rel == dst) { // failed |
4429 | return p_path; |
4430 | } else { |
4431 | return rel + p_path.get_file(); |
4432 | } |
4433 | } |
4434 | |
4435 | String String::path_to(const String &p_path) const { |
4436 | String src = this->replace("\\" , "/" ); |
4437 | String dst = p_path.replace("\\" , "/" ); |
4438 | if (!src.ends_with("/" )) { |
4439 | src += "/" ; |
4440 | } |
4441 | if (!dst.ends_with("/" )) { |
4442 | dst += "/" ; |
4443 | } |
4444 | |
4445 | if (src.begins_with("res://" ) && dst.begins_with("res://" )) { |
4446 | src = src.replace("res://" , "/" ); |
4447 | dst = dst.replace("res://" , "/" ); |
4448 | |
4449 | } else if (src.begins_with("user://" ) && dst.begins_with("user://" )) { |
4450 | src = src.replace("user://" , "/" ); |
4451 | dst = dst.replace("user://" , "/" ); |
4452 | |
4453 | } else if (src.begins_with("/" ) && dst.begins_with("/" )) { |
4454 | //nothing |
4455 | } else { |
4456 | //dos style |
4457 | String src_begin = src.get_slicec('/', 0); |
4458 | String dst_begin = dst.get_slicec('/', 0); |
4459 | |
4460 | if (src_begin != dst_begin) { |
4461 | return p_path; //impossible to do this |
4462 | } |
4463 | |
4464 | src = src.substr(src_begin.length(), src.length()); |
4465 | dst = dst.substr(dst_begin.length(), dst.length()); |
4466 | } |
4467 | |
4468 | //remove leading and trailing slash and split |
4469 | Vector<String> src_dirs = src.substr(1, src.length() - 2).split("/" ); |
4470 | Vector<String> dst_dirs = dst.substr(1, dst.length() - 2).split("/" ); |
4471 | |
4472 | //find common parent |
4473 | int common_parent = 0; |
4474 | |
4475 | while (true) { |
4476 | if (src_dirs.size() == common_parent) { |
4477 | break; |
4478 | } |
4479 | if (dst_dirs.size() == common_parent) { |
4480 | break; |
4481 | } |
4482 | if (src_dirs[common_parent] != dst_dirs[common_parent]) { |
4483 | break; |
4484 | } |
4485 | common_parent++; |
4486 | } |
4487 | |
4488 | common_parent--; |
4489 | |
4490 | int dirs_to_backtrack = (src_dirs.size() - 1) - common_parent; |
4491 | String dir = String("../" ).repeat(dirs_to_backtrack); |
4492 | |
4493 | for (int i = common_parent + 1; i < dst_dirs.size(); i++) { |
4494 | dir += dst_dirs[i] + "/" ; |
4495 | } |
4496 | |
4497 | if (dir.length() == 0) { |
4498 | dir = "./" ; |
4499 | } |
4500 | return dir; |
4501 | } |
4502 | |
4503 | bool String::is_valid_html_color() const { |
4504 | return Color::html_is_valid(*this); |
4505 | } |
4506 | |
4507 | // Changes made to the set of invalid filename characters must also be reflected in the String documentation for is_valid_filename. |
4508 | static const char *invalid_filename_characters = ": / \\ ? * \" | % < >" ; |
4509 | |
4510 | bool String::is_valid_filename() const { |
4511 | String stripped = strip_edges(); |
4512 | if (*this != stripped) { |
4513 | return false; |
4514 | } |
4515 | |
4516 | if (stripped.is_empty()) { |
4517 | return false; |
4518 | } |
4519 | |
4520 | Vector<String> chars = String(invalid_filename_characters).split(" " ); |
4521 | for (const String &ch : chars) { |
4522 | if (contains(ch)) { |
4523 | return false; |
4524 | } |
4525 | } |
4526 | return true; |
4527 | } |
4528 | |
4529 | String String::validate_filename() const { |
4530 | Vector<String> chars = String(invalid_filename_characters).split(" " ); |
4531 | String name = strip_edges(); |
4532 | for (int i = 0; i < chars.size(); i++) { |
4533 | name = name.replace(chars[i], "_" ); |
4534 | } |
4535 | return name; |
4536 | } |
4537 | |
4538 | bool String::is_valid_ip_address() const { |
4539 | if (find(":" ) >= 0) { |
4540 | Vector<String> ip = split(":" ); |
4541 | for (int i = 0; i < ip.size(); i++) { |
4542 | String n = ip[i]; |
4543 | if (n.is_empty()) { |
4544 | continue; |
4545 | } |
4546 | if (n.is_valid_hex_number(false)) { |
4547 | int64_t nint = n.hex_to_int(); |
4548 | if (nint < 0 || nint > 0xffff) { |
4549 | return false; |
4550 | } |
4551 | continue; |
4552 | } |
4553 | if (!n.is_valid_ip_address()) { |
4554 | return false; |
4555 | } |
4556 | } |
4557 | |
4558 | } else { |
4559 | Vector<String> ip = split("." ); |
4560 | if (ip.size() != 4) { |
4561 | return false; |
4562 | } |
4563 | for (int i = 0; i < ip.size(); i++) { |
4564 | String n = ip[i]; |
4565 | if (!n.is_valid_int()) { |
4566 | return false; |
4567 | } |
4568 | int val = n.to_int(); |
4569 | if (val < 0 || val > 255) { |
4570 | return false; |
4571 | } |
4572 | } |
4573 | } |
4574 | |
4575 | return true; |
4576 | } |
4577 | |
4578 | bool String::is_resource_file() const { |
4579 | return begins_with("res://" ) && find("::" ) == -1; |
4580 | } |
4581 | |
4582 | bool String::is_relative_path() const { |
4583 | return !is_absolute_path(); |
4584 | } |
4585 | |
4586 | String String::get_base_dir() const { |
4587 | int end = 0; |
4588 | |
4589 | // URL scheme style base. |
4590 | int basepos = find("://" ); |
4591 | if (basepos != -1) { |
4592 | end = basepos + 3; |
4593 | } |
4594 | |
4595 | // Windows top level directory base. |
4596 | if (end == 0) { |
4597 | basepos = find(":/" ); |
4598 | if (basepos == -1) { |
4599 | basepos = find(":\\" ); |
4600 | } |
4601 | if (basepos != -1) { |
4602 | end = basepos + 2; |
4603 | } |
4604 | } |
4605 | |
4606 | // Windows UNC network share path. |
4607 | if (end == 0) { |
4608 | if (is_network_share_path()) { |
4609 | basepos = find("/" , 2); |
4610 | if (basepos == -1) { |
4611 | basepos = find("\\" , 2); |
4612 | } |
4613 | int servpos = find("/" , basepos + 1); |
4614 | if (servpos == -1) { |
4615 | servpos = find("\\" , basepos + 1); |
4616 | } |
4617 | if (servpos != -1) { |
4618 | end = servpos + 1; |
4619 | } |
4620 | } |
4621 | } |
4622 | |
4623 | // Unix root directory base. |
4624 | if (end == 0) { |
4625 | if (begins_with("/" )) { |
4626 | end = 1; |
4627 | } |
4628 | } |
4629 | |
4630 | String rs; |
4631 | String base; |
4632 | if (end != 0) { |
4633 | rs = substr(end, length()); |
4634 | base = substr(0, end); |
4635 | } else { |
4636 | rs = *this; |
4637 | } |
4638 | |
4639 | int sep = MAX(rs.rfind("/" ), rs.rfind("\\" )); |
4640 | if (sep == -1) { |
4641 | return base; |
4642 | } |
4643 | |
4644 | return base + rs.substr(0, sep); |
4645 | } |
4646 | |
4647 | String String::get_file() const { |
4648 | int sep = MAX(rfind("/" ), rfind("\\" )); |
4649 | if (sep == -1) { |
4650 | return *this; |
4651 | } |
4652 | |
4653 | return substr(sep + 1, length()); |
4654 | } |
4655 | |
4656 | String String::get_extension() const { |
4657 | int pos = rfind("." ); |
4658 | if (pos < 0 || pos < MAX(rfind("/" ), rfind("\\" ))) { |
4659 | return "" ; |
4660 | } |
4661 | |
4662 | return substr(pos + 1, length()); |
4663 | } |
4664 | |
4665 | String String::path_join(const String &p_file) const { |
4666 | if (is_empty()) { |
4667 | return p_file; |
4668 | } |
4669 | if (operator[](length() - 1) == '/' || (p_file.size() > 0 && p_file.operator[](0) == '/')) { |
4670 | return *this + p_file; |
4671 | } |
4672 | return *this + "/" + p_file; |
4673 | } |
4674 | |
4675 | String String::property_name_encode() const { |
4676 | // Escape and quote strings with extended ASCII or further Unicode characters |
4677 | // as well as '"', '=' or ' ' (32) |
4678 | const char32_t *cstr = get_data(); |
4679 | for (int i = 0; cstr[i]; i++) { |
4680 | if (cstr[i] == '=' || cstr[i] == '"' || cstr[i] == ';' || cstr[i] == '[' || cstr[i] == ']' || cstr[i] < 33 || cstr[i] > 126) { |
4681 | return "\"" + c_escape_multiline() + "\"" ; |
4682 | } |
4683 | } |
4684 | // Keep as is |
4685 | return *this; |
4686 | } |
4687 | |
4688 | // Changes made to the set of invalid characters must also be reflected in the String documentation. |
4689 | |
4690 | static const char32_t invalid_node_name_characters[] = { '.', ':', '@', '/', '\"', UNIQUE_NODE_PREFIX[0], 0 }; |
4691 | |
4692 | String String::get_invalid_node_name_characters() { |
4693 | // Do not use this function for critical validation. |
4694 | String r; |
4695 | const char32_t *c = invalid_node_name_characters; |
4696 | while (*c) { |
4697 | if (c != invalid_node_name_characters) { |
4698 | r += " " ; |
4699 | } |
4700 | r += String::chr(*c); |
4701 | c++; |
4702 | } |
4703 | return r; |
4704 | } |
4705 | |
4706 | String String::validate_node_name() const { |
4707 | // This is a critical validation in node addition, so it must be optimized. |
4708 | const char32_t *cn = ptr(); |
4709 | if (cn == nullptr) { |
4710 | return String(); |
4711 | } |
4712 | bool valid = true; |
4713 | uint32_t idx = 0; |
4714 | while (cn[idx]) { |
4715 | const char32_t *c = invalid_node_name_characters; |
4716 | while (*c) { |
4717 | if (cn[idx] == *c) { |
4718 | valid = false; |
4719 | break; |
4720 | } |
4721 | c++; |
4722 | } |
4723 | if (!valid) { |
4724 | break; |
4725 | } |
4726 | idx++; |
4727 | } |
4728 | |
4729 | if (valid) { |
4730 | return *this; |
4731 | } |
4732 | |
4733 | String validated = *this; |
4734 | char32_t *nn = validated.ptrw(); |
4735 | while (nn[idx]) { |
4736 | const char32_t *c = invalid_node_name_characters; |
4737 | while (*c) { |
4738 | if (nn[idx] == *c) { |
4739 | nn[idx] = '_'; |
4740 | break; |
4741 | } |
4742 | c++; |
4743 | } |
4744 | idx++; |
4745 | } |
4746 | |
4747 | return validated; |
4748 | } |
4749 | |
4750 | String String::get_basename() const { |
4751 | int pos = rfind("." ); |
4752 | if (pos < 0 || pos < MAX(rfind("/" ), rfind("\\" ))) { |
4753 | return *this; |
4754 | } |
4755 | |
4756 | return substr(0, pos); |
4757 | } |
4758 | |
4759 | String itos(int64_t p_val) { |
4760 | return String::num_int64(p_val); |
4761 | } |
4762 | |
4763 | String uitos(uint64_t p_val) { |
4764 | return String::num_uint64(p_val); |
4765 | } |
4766 | |
4767 | String rtos(double p_val) { |
4768 | return String::num(p_val); |
4769 | } |
4770 | |
4771 | String rtoss(double p_val) { |
4772 | return String::num_scientific(p_val); |
4773 | } |
4774 | |
4775 | // Right-pad with a character. |
4776 | String String::rpad(int min_length, const String &character) const { |
4777 | String s = *this; |
4778 | int padding = min_length - s.length(); |
4779 | if (padding > 0) { |
4780 | s += character.repeat(padding); |
4781 | } |
4782 | return s; |
4783 | } |
4784 | |
4785 | // Left-pad with a character. |
4786 | String String::lpad(int min_length, const String &character) const { |
4787 | String s = *this; |
4788 | int padding = min_length - s.length(); |
4789 | if (padding > 0) { |
4790 | s = character.repeat(padding) + s; |
4791 | } |
4792 | return s; |
4793 | } |
4794 | |
4795 | // sprintf is implemented in GDScript via: |
4796 | // "fish %s pie" % "frog" |
4797 | // "fish %s %d pie" % ["frog", 12] |
4798 | // In case of an error, the string returned is the error description and "error" is true. |
4799 | String String::sprintf(const Array &values, bool *error) const { |
4800 | String formatted; |
4801 | char32_t *self = (char32_t *)get_data(); |
4802 | bool in_format = false; |
4803 | int value_index = 0; |
4804 | int min_chars = 0; |
4805 | int min_decimals = 0; |
4806 | bool in_decimals = false; |
4807 | bool pad_with_zeros = false; |
4808 | bool left_justified = false; |
4809 | bool show_sign = false; |
4810 | |
4811 | if (error) { |
4812 | *error = true; |
4813 | } |
4814 | |
4815 | for (; *self; self++) { |
4816 | const char32_t c = *self; |
4817 | |
4818 | if (in_format) { // We have % - let's see what else we get. |
4819 | switch (c) { |
4820 | case '%': { // Replace %% with % |
4821 | formatted += chr(c); |
4822 | in_format = false; |
4823 | break; |
4824 | } |
4825 | case 'd': // Integer (signed) |
4826 | case 'o': // Octal |
4827 | case 'x': // Hexadecimal (lowercase) |
4828 | case 'X': { // Hexadecimal (uppercase) |
4829 | if (value_index >= values.size()) { |
4830 | return "not enough arguments for format string" ; |
4831 | } |
4832 | |
4833 | if (!values[value_index].is_num()) { |
4834 | return "a number is required" ; |
4835 | } |
4836 | |
4837 | int64_t value = values[value_index]; |
4838 | int base = 16; |
4839 | bool capitalize = false; |
4840 | switch (c) { |
4841 | case 'd': |
4842 | base = 10; |
4843 | break; |
4844 | case 'o': |
4845 | base = 8; |
4846 | break; |
4847 | case 'x': |
4848 | break; |
4849 | case 'X': |
4850 | base = 16; |
4851 | capitalize = true; |
4852 | break; |
4853 | } |
4854 | // Get basic number. |
4855 | String str = String::num_int64(ABS(value), base, capitalize); |
4856 | int number_len = str.length(); |
4857 | |
4858 | // Padding. |
4859 | int pad_chars_count = (value < 0 || show_sign) ? min_chars - 1 : min_chars; |
4860 | String pad_char = pad_with_zeros ? String("0" ) : String(" " ); |
4861 | if (left_justified) { |
4862 | str = str.rpad(pad_chars_count, pad_char); |
4863 | } else { |
4864 | str = str.lpad(pad_chars_count, pad_char); |
4865 | } |
4866 | |
4867 | // Sign. |
4868 | if (show_sign || value < 0) { |
4869 | String sign_char = value < 0 ? "-" : "+" ; |
4870 | if (left_justified) { |
4871 | str = str.insert(0, sign_char); |
4872 | } else { |
4873 | str = str.insert(pad_with_zeros ? 0 : str.length() - number_len, sign_char); |
4874 | } |
4875 | } |
4876 | |
4877 | formatted += str; |
4878 | ++value_index; |
4879 | in_format = false; |
4880 | |
4881 | break; |
4882 | } |
4883 | case 'f': { // Float |
4884 | if (value_index >= values.size()) { |
4885 | return "not enough arguments for format string" ; |
4886 | } |
4887 | |
4888 | if (!values[value_index].is_num()) { |
4889 | return "a number is required" ; |
4890 | } |
4891 | |
4892 | double value = values[value_index]; |
4893 | bool is_negative = signbit(value); |
4894 | String str = String::num(Math::abs(value), min_decimals); |
4895 | const bool is_finite = Math::is_finite(value); |
4896 | |
4897 | // Pad decimals out. |
4898 | if (is_finite) { |
4899 | str = str.pad_decimals(min_decimals); |
4900 | } |
4901 | |
4902 | int initial_len = str.length(); |
4903 | |
4904 | // Padding. Leave room for sign later if required. |
4905 | int pad_chars_count = (is_negative || show_sign) ? min_chars - 1 : min_chars; |
4906 | String pad_char = (pad_with_zeros && is_finite) ? String("0" ) : String(" " ); // Never pad NaN or inf with zeros |
4907 | if (left_justified) { |
4908 | str = str.rpad(pad_chars_count, pad_char); |
4909 | } else { |
4910 | str = str.lpad(pad_chars_count, pad_char); |
4911 | } |
4912 | |
4913 | // Add sign if needed. |
4914 | if (show_sign || is_negative) { |
4915 | String sign_char = is_negative ? "-" : "+" ; |
4916 | if (left_justified) { |
4917 | str = str.insert(0, sign_char); |
4918 | } else { |
4919 | str = str.insert(pad_with_zeros ? 0 : str.length() - initial_len, sign_char); |
4920 | } |
4921 | } |
4922 | |
4923 | formatted += str; |
4924 | ++value_index; |
4925 | in_format = false; |
4926 | break; |
4927 | } |
4928 | case 'v': { // Vector2/3/4/2i/3i/4i |
4929 | if (value_index >= values.size()) { |
4930 | return "not enough arguments for format string" ; |
4931 | } |
4932 | |
4933 | int count; |
4934 | switch (values[value_index].get_type()) { |
4935 | case Variant::VECTOR2: |
4936 | case Variant::VECTOR2I: { |
4937 | count = 2; |
4938 | } break; |
4939 | case Variant::VECTOR3: |
4940 | case Variant::VECTOR3I: { |
4941 | count = 3; |
4942 | } break; |
4943 | case Variant::VECTOR4: |
4944 | case Variant::VECTOR4I: { |
4945 | count = 4; |
4946 | } break; |
4947 | default: { |
4948 | return "%v requires a vector type (Vector2/3/4/2i/3i/4i)" ; |
4949 | } |
4950 | } |
4951 | |
4952 | Vector4 vec = values[value_index]; |
4953 | String str = "(" ; |
4954 | for (int i = 0; i < count; i++) { |
4955 | double val = vec[i]; |
4956 | String number_str = String::num(Math::abs(val), min_decimals); |
4957 | const bool is_finite = Math::is_finite(val); |
4958 | |
4959 | // Pad decimals out. |
4960 | if (is_finite) { |
4961 | number_str = number_str.pad_decimals(min_decimals); |
4962 | } |
4963 | |
4964 | int initial_len = number_str.length(); |
4965 | |
4966 | // Padding. Leave room for sign later if required. |
4967 | int pad_chars_count = val < 0 ? min_chars - 1 : min_chars; |
4968 | String pad_char = (pad_with_zeros && is_finite) ? String("0" ) : String(" " ); // Never pad NaN or inf with zeros |
4969 | if (left_justified) { |
4970 | number_str = number_str.rpad(pad_chars_count, pad_char); |
4971 | } else { |
4972 | number_str = number_str.lpad(pad_chars_count, pad_char); |
4973 | } |
4974 | |
4975 | // Add sign if needed. |
4976 | if (val < 0) { |
4977 | if (left_justified) { |
4978 | number_str = number_str.insert(0, "-" ); |
4979 | } else { |
4980 | number_str = number_str.insert(pad_with_zeros ? 0 : number_str.length() - initial_len, "-" ); |
4981 | } |
4982 | } |
4983 | |
4984 | // Add number to combined string |
4985 | str += number_str; |
4986 | |
4987 | if (i < count - 1) { |
4988 | str += ", " ; |
4989 | } |
4990 | } |
4991 | str += ")" ; |
4992 | |
4993 | formatted += str; |
4994 | ++value_index; |
4995 | in_format = false; |
4996 | break; |
4997 | } |
4998 | case 's': { // String |
4999 | if (value_index >= values.size()) { |
5000 | return "not enough arguments for format string" ; |
5001 | } |
5002 | |
5003 | String str = values[value_index]; |
5004 | // Padding. |
5005 | if (left_justified) { |
5006 | str = str.rpad(min_chars); |
5007 | } else { |
5008 | str = str.lpad(min_chars); |
5009 | } |
5010 | |
5011 | formatted += str; |
5012 | ++value_index; |
5013 | in_format = false; |
5014 | break; |
5015 | } |
5016 | case 'c': { |
5017 | if (value_index >= values.size()) { |
5018 | return "not enough arguments for format string" ; |
5019 | } |
5020 | |
5021 | // Convert to character. |
5022 | String str; |
5023 | if (values[value_index].is_num()) { |
5024 | int value = values[value_index]; |
5025 | if (value < 0) { |
5026 | return "unsigned integer is lower than minimum" ; |
5027 | } else if (value >= 0xd800 && value <= 0xdfff) { |
5028 | return "unsigned integer is invalid Unicode character" ; |
5029 | } else if (value > 0x10ffff) { |
5030 | return "unsigned integer is greater than maximum" ; |
5031 | } |
5032 | str = chr(values[value_index]); |
5033 | } else if (values[value_index].get_type() == Variant::STRING) { |
5034 | str = values[value_index]; |
5035 | if (str.length() != 1) { |
5036 | return "%c requires number or single-character string" ; |
5037 | } |
5038 | } else { |
5039 | return "%c requires number or single-character string" ; |
5040 | } |
5041 | |
5042 | // Padding. |
5043 | if (left_justified) { |
5044 | str = str.rpad(min_chars); |
5045 | } else { |
5046 | str = str.lpad(min_chars); |
5047 | } |
5048 | |
5049 | formatted += str; |
5050 | ++value_index; |
5051 | in_format = false; |
5052 | break; |
5053 | } |
5054 | case '-': { // Left justify |
5055 | left_justified = true; |
5056 | break; |
5057 | } |
5058 | case '+': { // Show + if positive. |
5059 | show_sign = true; |
5060 | break; |
5061 | } |
5062 | case '0': |
5063 | case '1': |
5064 | case '2': |
5065 | case '3': |
5066 | case '4': |
5067 | case '5': |
5068 | case '6': |
5069 | case '7': |
5070 | case '8': |
5071 | case '9': { |
5072 | int n = c - '0'; |
5073 | if (in_decimals) { |
5074 | min_decimals *= 10; |
5075 | min_decimals += n; |
5076 | } else { |
5077 | if (c == '0' && min_chars == 0) { |
5078 | if (left_justified) { |
5079 | WARN_PRINT("'0' flag ignored with '-' flag in string format" ); |
5080 | } else { |
5081 | pad_with_zeros = true; |
5082 | } |
5083 | } else { |
5084 | min_chars *= 10; |
5085 | min_chars += n; |
5086 | } |
5087 | } |
5088 | break; |
5089 | } |
5090 | case '.': { // Float/Vector separator. |
5091 | if (in_decimals) { |
5092 | return "too many decimal points in format" ; |
5093 | } |
5094 | in_decimals = true; |
5095 | min_decimals = 0; // We want to add the value manually. |
5096 | break; |
5097 | } |
5098 | |
5099 | case '*': { // Dynamic width, based on value. |
5100 | if (value_index >= values.size()) { |
5101 | return "not enough arguments for format string" ; |
5102 | } |
5103 | |
5104 | Variant::Type value_type = values[value_index].get_type(); |
5105 | if (!values[value_index].is_num() && |
5106 | value_type != Variant::VECTOR2 && value_type != Variant::VECTOR2I && |
5107 | value_type != Variant::VECTOR3 && value_type != Variant::VECTOR3I && |
5108 | value_type != Variant::VECTOR4 && value_type != Variant::VECTOR4I) { |
5109 | return "* wants number or vector" ; |
5110 | } |
5111 | |
5112 | int size = values[value_index]; |
5113 | |
5114 | if (in_decimals) { |
5115 | min_decimals = size; |
5116 | } else { |
5117 | min_chars = size; |
5118 | } |
5119 | |
5120 | ++value_index; |
5121 | break; |
5122 | } |
5123 | |
5124 | default: { |
5125 | return "unsupported format character" ; |
5126 | } |
5127 | } |
5128 | } else { // Not in format string. |
5129 | switch (c) { |
5130 | case '%': |
5131 | in_format = true; |
5132 | // Back to defaults: |
5133 | min_chars = 0; |
5134 | min_decimals = 6; |
5135 | pad_with_zeros = false; |
5136 | left_justified = false; |
5137 | show_sign = false; |
5138 | in_decimals = false; |
5139 | break; |
5140 | default: |
5141 | formatted += chr(c); |
5142 | } |
5143 | } |
5144 | } |
5145 | |
5146 | if (in_format) { |
5147 | return "incomplete format" ; |
5148 | } |
5149 | |
5150 | if (value_index != values.size()) { |
5151 | return "not all arguments converted during string formatting" ; |
5152 | } |
5153 | |
5154 | if (error) { |
5155 | *error = false; |
5156 | } |
5157 | return formatted; |
5158 | } |
5159 | |
5160 | String String::quote(String quotechar) const { |
5161 | return quotechar + *this + quotechar; |
5162 | } |
5163 | |
5164 | String String::unquote() const { |
5165 | if (!is_quoted()) { |
5166 | return *this; |
5167 | } |
5168 | |
5169 | return substr(1, length() - 2); |
5170 | } |
5171 | |
5172 | Vector<uint8_t> String::to_ascii_buffer() const { |
5173 | const String *s = this; |
5174 | if (s->is_empty()) { |
5175 | return Vector<uint8_t>(); |
5176 | } |
5177 | CharString charstr = s->ascii(); |
5178 | |
5179 | Vector<uint8_t> retval; |
5180 | size_t len = charstr.length(); |
5181 | retval.resize(len); |
5182 | uint8_t *w = retval.ptrw(); |
5183 | memcpy(w, charstr.ptr(), len); |
5184 | |
5185 | return retval; |
5186 | } |
5187 | |
5188 | Vector<uint8_t> String::to_utf8_buffer() const { |
5189 | const String *s = this; |
5190 | if (s->is_empty()) { |
5191 | return Vector<uint8_t>(); |
5192 | } |
5193 | CharString charstr = s->utf8(); |
5194 | |
5195 | Vector<uint8_t> retval; |
5196 | size_t len = charstr.length(); |
5197 | retval.resize(len); |
5198 | uint8_t *w = retval.ptrw(); |
5199 | memcpy(w, charstr.ptr(), len); |
5200 | |
5201 | return retval; |
5202 | } |
5203 | |
5204 | Vector<uint8_t> String::to_utf16_buffer() const { |
5205 | const String *s = this; |
5206 | if (s->is_empty()) { |
5207 | return Vector<uint8_t>(); |
5208 | } |
5209 | Char16String charstr = s->utf16(); |
5210 | |
5211 | Vector<uint8_t> retval; |
5212 | size_t len = charstr.length() * sizeof(char16_t); |
5213 | retval.resize(len); |
5214 | uint8_t *w = retval.ptrw(); |
5215 | memcpy(w, (const void *)charstr.ptr(), len); |
5216 | |
5217 | return retval; |
5218 | } |
5219 | |
5220 | Vector<uint8_t> String::to_utf32_buffer() const { |
5221 | const String *s = this; |
5222 | if (s->is_empty()) { |
5223 | return Vector<uint8_t>(); |
5224 | } |
5225 | |
5226 | Vector<uint8_t> retval; |
5227 | size_t len = s->length() * sizeof(char32_t); |
5228 | retval.resize(len); |
5229 | uint8_t *w = retval.ptrw(); |
5230 | memcpy(w, (const void *)s->ptr(), len); |
5231 | |
5232 | return retval; |
5233 | } |
5234 | |
5235 | Vector<uint8_t> String::to_wchar_buffer() const { |
5236 | #ifdef WINDOWS_ENABLED |
5237 | return to_utf16_buffer(); |
5238 | #else |
5239 | return to_utf32_buffer(); |
5240 | #endif |
5241 | } |
5242 | |
5243 | #ifdef TOOLS_ENABLED |
5244 | /** |
5245 | * "Tools TRanslate". Performs string replacement for internationalization |
5246 | * within the editor. A translation context can optionally be specified to |
5247 | * disambiguate between identical source strings in translations. When |
5248 | * placeholders are desired, use `vformat(TTR("Example: %s"), some_string)`. |
5249 | * If a string mentions a quantity (and may therefore need a dynamic plural form), |
5250 | * use `TTRN()` instead of `TTR()`. |
5251 | * |
5252 | * NOTE: Only use `TTR()` in editor-only code (typically within the `editor/` folder). |
5253 | * For translations that can be supplied by exported projects, use `RTR()` instead. |
5254 | */ |
5255 | String TTR(const String &p_text, const String &p_context) { |
5256 | if (TranslationServer::get_singleton()) { |
5257 | return TranslationServer::get_singleton()->tool_translate(p_text, p_context); |
5258 | } |
5259 | |
5260 | return p_text; |
5261 | } |
5262 | |
5263 | /** |
5264 | * "Tools TRanslate for N items". Performs string replacement for |
5265 | * internationalization within the editor. A translation context can optionally |
5266 | * be specified to disambiguate between identical source strings in |
5267 | * translations. Use `TTR()` if the string doesn't need dynamic plural form. |
5268 | * When placeholders are desired, use |
5269 | * `vformat(TTRN("%d item", "%d items", some_integer), some_integer)`. |
5270 | * The placeholder must be present in both strings to avoid run-time warnings in `vformat()`. |
5271 | * |
5272 | * NOTE: Only use `TTRN()` in editor-only code (typically within the `editor/` folder). |
5273 | * For translations that can be supplied by exported projects, use `RTRN()` instead. |
5274 | */ |
5275 | String TTRN(const String &p_text, const String &p_text_plural, int p_n, const String &p_context) { |
5276 | if (TranslationServer::get_singleton()) { |
5277 | return TranslationServer::get_singleton()->tool_translate_plural(p_text, p_text_plural, p_n, p_context); |
5278 | } |
5279 | |
5280 | // Return message based on English plural rule if translation is not possible. |
5281 | if (p_n == 1) { |
5282 | return p_text; |
5283 | } |
5284 | return p_text_plural; |
5285 | } |
5286 | |
5287 | /** |
5288 | * "Docs TRanslate". Used for the editor class reference documentation, |
5289 | * handling descriptions extracted from the XML. |
5290 | * It also replaces `$DOCS_URL` with the actual URL to the documentation's branch, |
5291 | * to allow dehardcoding it in the XML and doing proper substitutions everywhere. |
5292 | */ |
5293 | String DTR(const String &p_text, const String &p_context) { |
5294 | // Comes straight from the XML, so remove indentation and any trailing whitespace. |
5295 | const String text = p_text.dedent().strip_edges(); |
5296 | |
5297 | if (TranslationServer::get_singleton()) { |
5298 | return String(TranslationServer::get_singleton()->doc_translate(text, p_context)).replace("$DOCS_URL" , VERSION_DOCS_URL); |
5299 | } |
5300 | |
5301 | return text.replace("$DOCS_URL" , VERSION_DOCS_URL); |
5302 | } |
5303 | |
5304 | /** |
5305 | * "Docs TRanslate for N items". Used for the editor class reference documentation |
5306 | * (with support for plurals), handling descriptions extracted from the XML. |
5307 | * It also replaces `$DOCS_URL` with the actual URL to the documentation's branch, |
5308 | * to allow dehardcoding it in the XML and doing proper substitutions everywhere. |
5309 | */ |
5310 | String DTRN(const String &p_text, const String &p_text_plural, int p_n, const String &p_context) { |
5311 | const String text = p_text.dedent().strip_edges(); |
5312 | const String text_plural = p_text_plural.dedent().strip_edges(); |
5313 | |
5314 | if (TranslationServer::get_singleton()) { |
5315 | return String(TranslationServer::get_singleton()->doc_translate_plural(text, text_plural, p_n, p_context)).replace("$DOCS_URL" , VERSION_DOCS_URL); |
5316 | } |
5317 | |
5318 | // Return message based on English plural rule if translation is not possible. |
5319 | if (p_n == 1) { |
5320 | return text.replace("$DOCS_URL" , VERSION_DOCS_URL); |
5321 | } |
5322 | return text_plural.replace("$DOCS_URL" , VERSION_DOCS_URL); |
5323 | } |
5324 | #endif |
5325 | |
5326 | /** |
5327 | * "Run-time TRanslate". Performs string replacement for internationalization |
5328 | * within a running project. The translation string must be supplied by the |
5329 | * project, as Godot does not provide built-in translations for `RTR()` strings |
5330 | * to keep binary size low. A translation context can optionally be specified to |
5331 | * disambiguate between identical source strings in translations. When |
5332 | * placeholders are desired, use `vformat(RTR("Example: %s"), some_string)`. |
5333 | * If a string mentions a quantity (and may therefore need a dynamic plural form), |
5334 | * use `RTRN()` instead of `RTR()`. |
5335 | * |
5336 | * NOTE: Do not use `RTR()` in editor-only code (typically within the `editor/` |
5337 | * folder). For editor translations, use `TTR()` instead. |
5338 | */ |
5339 | String RTR(const String &p_text, const String &p_context) { |
5340 | if (TranslationServer::get_singleton()) { |
5341 | String rtr = TranslationServer::get_singleton()->tool_translate(p_text, p_context); |
5342 | if (rtr.is_empty() || rtr == p_text) { |
5343 | return TranslationServer::get_singleton()->translate(p_text, p_context); |
5344 | } else { |
5345 | return rtr; |
5346 | } |
5347 | } |
5348 | |
5349 | return p_text; |
5350 | } |
5351 | |
5352 | /** |
5353 | * "Run-time TRanslate for N items". Performs string replacement for |
5354 | * internationalization within a running project. The translation string must be |
5355 | * supplied by the project, as Godot does not provide built-in translations for |
5356 | * `RTRN()` strings to keep binary size low. A translation context can |
5357 | * optionally be specified to disambiguate between identical source strings in |
5358 | * translations. Use `RTR()` if the string doesn't need dynamic plural form. |
5359 | * When placeholders are desired, use |
5360 | * `vformat(RTRN("%d item", "%d items", some_integer), some_integer)`. |
5361 | * The placeholder must be present in both strings to avoid run-time warnings in `vformat()`. |
5362 | * |
5363 | * NOTE: Do not use `RTRN()` in editor-only code (typically within the `editor/` |
5364 | * folder). For editor translations, use `TTRN()` instead. |
5365 | */ |
5366 | String RTRN(const String &p_text, const String &p_text_plural, int p_n, const String &p_context) { |
5367 | if (TranslationServer::get_singleton()) { |
5368 | String rtr = TranslationServer::get_singleton()->tool_translate_plural(p_text, p_text_plural, p_n, p_context); |
5369 | if (rtr.is_empty() || rtr == p_text || rtr == p_text_plural) { |
5370 | return TranslationServer::get_singleton()->translate_plural(p_text, p_text_plural, p_n, p_context); |
5371 | } else { |
5372 | return rtr; |
5373 | } |
5374 | } |
5375 | |
5376 | // Return message based on English plural rule if translation is not possible. |
5377 | if (p_n == 1) { |
5378 | return p_text; |
5379 | } |
5380 | return p_text_plural; |
5381 | } |
5382 | |