1 | //===- llvm/ADT/StringExtras.h - Useful string functions --------*- C++ -*-===// |
2 | // |
3 | // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
4 | // See https://llvm.org/LICENSE.txt for license information. |
5 | // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
6 | // |
7 | //===----------------------------------------------------------------------===// |
8 | // |
9 | // This file contains some functions that are useful when dealing with strings. |
10 | // |
11 | //===----------------------------------------------------------------------===// |
12 | |
13 | #ifndef LLVM_ADT_STRINGEXTRAS_H |
14 | #define |
15 | |
16 | #include "llvm/ADT/ArrayRef.h" |
17 | #include "llvm/ADT/SmallString.h" |
18 | #include "llvm/ADT/StringRef.h" |
19 | #include "llvm/ADT/Twine.h" |
20 | #include <cassert> |
21 | #include <cstddef> |
22 | #include <cstdint> |
23 | #include <cstdlib> |
24 | #include <cstring> |
25 | #include <iterator> |
26 | #include <string> |
27 | #include <utility> |
28 | |
29 | namespace llvm { |
30 | |
31 | template<typename T> class SmallVectorImpl; |
32 | class raw_ostream; |
33 | |
34 | /// hexdigit - Return the hexadecimal character for the |
35 | /// given number \p X (which should be less than 16). |
36 | inline char hexdigit(unsigned X, bool LowerCase = false) { |
37 | const char HexChar = LowerCase ? 'a' : 'A'; |
38 | return X < 10 ? '0' + X : HexChar + X - 10; |
39 | } |
40 | |
41 | /// Given an array of c-style strings terminated by a null pointer, construct |
42 | /// a vector of StringRefs representing the same strings without the terminating |
43 | /// null string. |
44 | inline std::vector<StringRef> toStringRefArray(const char *const *Strings) { |
45 | std::vector<StringRef> Result; |
46 | while (*Strings) |
47 | Result.push_back(*Strings++); |
48 | return Result; |
49 | } |
50 | |
51 | /// Construct a string ref from a boolean. |
52 | inline StringRef toStringRef(bool B) { return StringRef(B ? "true" : "false" ); } |
53 | |
54 | /// Construct a string ref from an array ref of unsigned chars. |
55 | inline StringRef toStringRef(ArrayRef<uint8_t> Input) { |
56 | return StringRef(reinterpret_cast<const char *>(Input.begin()), Input.size()); |
57 | } |
58 | |
59 | /// Construct a string ref from an array ref of unsigned chars. |
60 | inline ArrayRef<uint8_t> arrayRefFromStringRef(StringRef Input) { |
61 | return {Input.bytes_begin(), Input.bytes_end()}; |
62 | } |
63 | |
64 | /// Interpret the given character \p C as a hexadecimal digit and return its |
65 | /// value. |
66 | /// |
67 | /// If \p C is not a valid hex digit, -1U is returned. |
68 | inline unsigned hexDigitValue(char C) { |
69 | struct HexTable { |
70 | unsigned LUT[255] = {}; |
71 | constexpr HexTable() { |
72 | // Default initialize everything to invalid. |
73 | for (int i = 0; i < 255; ++i) |
74 | LUT[i] = ~0U; |
75 | // Initialize `0`-`9`. |
76 | for (int i = 0; i < 10; ++i) |
77 | LUT['0' + i] = i; |
78 | // Initialize `A`-`F` and `a`-`f`. |
79 | for (int i = 0; i < 6; ++i) |
80 | LUT['A' + i] = LUT['a' + i] = 10 + i; |
81 | } |
82 | }; |
83 | constexpr HexTable Table; |
84 | return Table.LUT[static_cast<unsigned char>(C)]; |
85 | } |
86 | |
87 | /// Checks if character \p C is one of the 10 decimal digits. |
88 | inline bool isDigit(char C) { return C >= '0' && C <= '9'; } |
89 | |
90 | /// Checks if character \p C is a hexadecimal numeric character. |
91 | inline bool isHexDigit(char C) { return hexDigitValue(C) != ~0U; } |
92 | |
93 | /// Checks if character \p C is a valid letter as classified by "C" locale. |
94 | inline bool isAlpha(char C) { |
95 | return ('a' <= C && C <= 'z') || ('A' <= C && C <= 'Z'); |
96 | } |
97 | |
98 | /// Checks whether character \p C is either a decimal digit or an uppercase or |
99 | /// lowercase letter as classified by "C" locale. |
100 | inline bool isAlnum(char C) { return isAlpha(C) || isDigit(C); } |
101 | |
102 | /// Checks whether character \p C is valid ASCII (high bit is zero). |
103 | inline bool isASCII(char C) { return static_cast<unsigned char>(C) <= 127; } |
104 | |
105 | /// Checks whether all characters in S are ASCII. |
106 | inline bool isASCII(llvm::StringRef S) { |
107 | for (char C : S) |
108 | if (LLVM_UNLIKELY(!isASCII(C))) |
109 | return false; |
110 | return true; |
111 | } |
112 | |
113 | /// Checks whether character \p C is printable. |
114 | /// |
115 | /// Locale-independent version of the C standard library isprint whose results |
116 | /// may differ on different platforms. |
117 | inline bool isPrint(char C) { |
118 | unsigned char UC = static_cast<unsigned char>(C); |
119 | return (0x20 <= UC) && (UC <= 0x7E); |
120 | } |
121 | |
122 | /// Checks whether character \p C is whitespace in the "C" locale. |
123 | /// |
124 | /// Locale-independent version of the C standard library isspace. |
125 | inline bool isSpace(char C) { |
126 | return C == ' ' || C == '\f' || C == '\n' || C == '\r' || C == '\t' || |
127 | C == '\v'; |
128 | } |
129 | |
130 | /// Returns the corresponding lowercase character if \p x is uppercase. |
131 | inline char toLower(char x) { |
132 | if (x >= 'A' && x <= 'Z') |
133 | return x - 'A' + 'a'; |
134 | return x; |
135 | } |
136 | |
137 | /// Returns the corresponding uppercase character if \p x is lowercase. |
138 | inline char toUpper(char x) { |
139 | if (x >= 'a' && x <= 'z') |
140 | return x - 'a' + 'A'; |
141 | return x; |
142 | } |
143 | |
144 | inline std::string utohexstr(uint64_t X, bool LowerCase = false) { |
145 | char Buffer[17]; |
146 | char *BufPtr = std::end(Buffer); |
147 | |
148 | if (X == 0) *--BufPtr = '0'; |
149 | |
150 | while (X) { |
151 | unsigned char Mod = static_cast<unsigned char>(X) & 15; |
152 | *--BufPtr = hexdigit(Mod, LowerCase); |
153 | X >>= 4; |
154 | } |
155 | |
156 | return std::string(BufPtr, std::end(Buffer)); |
157 | } |
158 | |
159 | /// Convert buffer \p Input to its hexadecimal representation. |
160 | /// The returned string is double the size of \p Input. |
161 | inline std::string toHex(StringRef Input, bool LowerCase = false) { |
162 | static const char *const LUT = "0123456789ABCDEF" ; |
163 | const uint8_t Offset = LowerCase ? 32 : 0; |
164 | size_t Length = Input.size(); |
165 | |
166 | std::string Output; |
167 | Output.reserve(2 * Length); |
168 | for (size_t i = 0; i < Length; ++i) { |
169 | const unsigned char c = Input[i]; |
170 | Output.push_back(LUT[c >> 4] | Offset); |
171 | Output.push_back(LUT[c & 15] | Offset); |
172 | } |
173 | return Output; |
174 | } |
175 | |
176 | inline std::string toHex(ArrayRef<uint8_t> Input, bool LowerCase = false) { |
177 | return toHex(toStringRef(Input), LowerCase); |
178 | } |
179 | |
180 | /// Store the binary representation of the two provided values, \p MSB and |
181 | /// \p LSB, that make up the nibbles of a hexadecimal digit. If \p MSB or \p LSB |
182 | /// do not correspond to proper nibbles of a hexadecimal digit, this method |
183 | /// returns false. Otherwise, returns true. |
184 | inline bool tryGetHexFromNibbles(char MSB, char LSB, uint8_t &Hex) { |
185 | unsigned U1 = hexDigitValue(MSB); |
186 | unsigned U2 = hexDigitValue(LSB); |
187 | if (U1 == ~0U || U2 == ~0U) |
188 | return false; |
189 | |
190 | Hex = static_cast<uint8_t>((U1 << 4) | U2); |
191 | return true; |
192 | } |
193 | |
194 | /// Return the binary representation of the two provided values, \p MSB and |
195 | /// \p LSB, that make up the nibbles of a hexadecimal digit. |
196 | inline uint8_t hexFromNibbles(char MSB, char LSB) { |
197 | uint8_t Hex = 0; |
198 | bool GotHex = tryGetHexFromNibbles(MSB, LSB, Hex); |
199 | (void)GotHex; |
200 | assert(GotHex && "MSB and/or LSB do not correspond to hex digits" ); |
201 | return Hex; |
202 | } |
203 | |
204 | /// Convert hexadecimal string \p Input to its binary representation and store |
205 | /// the result in \p Output. Returns true if the binary representation could be |
206 | /// converted from the hexadecimal string. Returns false if \p Input contains |
207 | /// non-hexadecimal digits. The output string is half the size of \p Input. |
208 | inline bool tryGetFromHex(StringRef Input, std::string &Output) { |
209 | if (Input.empty()) |
210 | return true; |
211 | |
212 | Output.reserve((Input.size() + 1) / 2); |
213 | if (Input.size() % 2 == 1) { |
214 | uint8_t Hex = 0; |
215 | if (!tryGetHexFromNibbles('0', Input.front(), Hex)) |
216 | return false; |
217 | |
218 | Output.push_back(Hex); |
219 | Input = Input.drop_front(); |
220 | } |
221 | |
222 | assert(Input.size() % 2 == 0); |
223 | while (!Input.empty()) { |
224 | uint8_t Hex = 0; |
225 | if (!tryGetHexFromNibbles(Input[0], Input[1], Hex)) |
226 | return false; |
227 | |
228 | Output.push_back(Hex); |
229 | Input = Input.drop_front(2); |
230 | } |
231 | return true; |
232 | } |
233 | |
234 | /// Convert hexadecimal string \p Input to its binary representation. |
235 | /// The return string is half the size of \p Input. |
236 | inline std::string fromHex(StringRef Input) { |
237 | std::string Hex; |
238 | bool GotHex = tryGetFromHex(Input, Hex); |
239 | (void)GotHex; |
240 | assert(GotHex && "Input contains non hex digits" ); |
241 | return Hex; |
242 | } |
243 | |
244 | /// Convert the string \p S to an integer of the specified type using |
245 | /// the radix \p Base. If \p Base is 0, auto-detects the radix. |
246 | /// Returns true if the number was successfully converted, false otherwise. |
247 | template <typename N> bool to_integer(StringRef S, N &Num, unsigned Base = 0) { |
248 | return !S.getAsInteger(Base, Num); |
249 | } |
250 | |
251 | namespace detail { |
252 | template <typename N> |
253 | inline bool to_float(const Twine &T, N &Num, N (*StrTo)(const char *, char **)) { |
254 | SmallString<32> Storage; |
255 | StringRef S = T.toNullTerminatedStringRef(Storage); |
256 | char *End; |
257 | N Temp = StrTo(S.data(), &End); |
258 | if (*End != '\0') |
259 | return false; |
260 | Num = Temp; |
261 | return true; |
262 | } |
263 | } |
264 | |
265 | inline bool to_float(const Twine &T, float &Num) { |
266 | return detail::to_float(T, Num, strtof); |
267 | } |
268 | |
269 | inline bool to_float(const Twine &T, double &Num) { |
270 | return detail::to_float(T, Num, strtod); |
271 | } |
272 | |
273 | inline bool to_float(const Twine &T, long double &Num) { |
274 | return detail::to_float(T, Num, strtold); |
275 | } |
276 | |
277 | inline std::string utostr(uint64_t X, bool isNeg = false) { |
278 | char Buffer[21]; |
279 | char *BufPtr = std::end(Buffer); |
280 | |
281 | if (X == 0) *--BufPtr = '0'; // Handle special case... |
282 | |
283 | while (X) { |
284 | *--BufPtr = '0' + char(X % 10); |
285 | X /= 10; |
286 | } |
287 | |
288 | if (isNeg) *--BufPtr = '-'; // Add negative sign... |
289 | return std::string(BufPtr, std::end(Buffer)); |
290 | } |
291 | |
292 | inline std::string itostr(int64_t X) { |
293 | if (X < 0) |
294 | return utostr(static_cast<uint64_t>(1) + ~static_cast<uint64_t>(X), true); |
295 | else |
296 | return utostr(static_cast<uint64_t>(X)); |
297 | } |
298 | |
299 | /// StrInStrNoCase - Portable version of strcasestr. Locates the first |
300 | /// occurrence of string 's1' in string 's2', ignoring case. Returns |
301 | /// the offset of s2 in s1 or npos if s2 cannot be found. |
302 | StringRef::size_type StrInStrNoCase(StringRef s1, StringRef s2); |
303 | |
304 | /// getToken - This function extracts one token from source, ignoring any |
305 | /// leading characters that appear in the Delimiters string, and ending the |
306 | /// token at any of the characters that appear in the Delimiters string. If |
307 | /// there are no tokens in the source string, an empty string is returned. |
308 | /// The function returns a pair containing the extracted token and the |
309 | /// remaining tail string. |
310 | std::pair<StringRef, StringRef> getToken(StringRef Source, |
311 | StringRef Delimiters = " \t\n\v\f\r" ); |
312 | |
313 | /// SplitString - Split up the specified string according to the specified |
314 | /// delimiters, appending the result fragments to the output list. |
315 | void SplitString(StringRef Source, |
316 | SmallVectorImpl<StringRef> &OutFragments, |
317 | StringRef Delimiters = " \t\n\v\f\r" ); |
318 | |
319 | /// Returns the English suffix for an ordinal integer (-st, -nd, -rd, -th). |
320 | inline StringRef getOrdinalSuffix(unsigned Val) { |
321 | // It is critically important that we do this perfectly for |
322 | // user-written sequences with over 100 elements. |
323 | switch (Val % 100) { |
324 | case 11: |
325 | case 12: |
326 | case 13: |
327 | return "th" ; |
328 | default: |
329 | switch (Val % 10) { |
330 | case 1: return "st" ; |
331 | case 2: return "nd" ; |
332 | case 3: return "rd" ; |
333 | default: return "th" ; |
334 | } |
335 | } |
336 | } |
337 | |
338 | /// Print each character of the specified string, escaping it if it is not |
339 | /// printable or if it is an escape char. |
340 | void printEscapedString(StringRef Name, raw_ostream &Out); |
341 | |
342 | /// Print each character of the specified string, escaping HTML special |
343 | /// characters. |
344 | void printHTMLEscaped(StringRef String, raw_ostream &Out); |
345 | |
346 | /// printLowerCase - Print each character as lowercase if it is uppercase. |
347 | void printLowerCase(StringRef String, raw_ostream &Out); |
348 | |
349 | /// Converts a string from camel-case to snake-case by replacing all uppercase |
350 | /// letters with '_' followed by the letter in lowercase, except if the |
351 | /// uppercase letter is the first character of the string. |
352 | std::string convertToSnakeFromCamelCase(StringRef input); |
353 | |
354 | /// Converts a string from snake-case to camel-case by replacing all occurrences |
355 | /// of '_' followed by a lowercase letter with the letter in uppercase. |
356 | /// Optionally allow capitalization of the first letter (if it is a lowercase |
357 | /// letter) |
358 | std::string convertToCamelFromSnakeCase(StringRef input, |
359 | bool capitalizeFirst = false); |
360 | |
361 | namespace detail { |
362 | |
363 | template <typename IteratorT> |
364 | inline std::string join_impl(IteratorT Begin, IteratorT End, |
365 | StringRef Separator, std::input_iterator_tag) { |
366 | std::string S; |
367 | if (Begin == End) |
368 | return S; |
369 | |
370 | S += (*Begin); |
371 | while (++Begin != End) { |
372 | S += Separator; |
373 | S += (*Begin); |
374 | } |
375 | return S; |
376 | } |
377 | |
378 | template <typename IteratorT> |
379 | inline std::string join_impl(IteratorT Begin, IteratorT End, |
380 | StringRef Separator, std::forward_iterator_tag) { |
381 | std::string S; |
382 | if (Begin == End) |
383 | return S; |
384 | |
385 | size_t Len = (std::distance(Begin, End) - 1) * Separator.size(); |
386 | for (IteratorT I = Begin; I != End; ++I) |
387 | Len += (*I).size(); |
388 | S.reserve(Len); |
389 | size_t PrevCapacity = S.capacity(); |
390 | (void)PrevCapacity; |
391 | S += (*Begin); |
392 | while (++Begin != End) { |
393 | S += Separator; |
394 | S += (*Begin); |
395 | } |
396 | assert(PrevCapacity == S.capacity() && "String grew during building" ); |
397 | return S; |
398 | } |
399 | |
400 | template <typename Sep> |
401 | inline void join_items_impl(std::string &Result, Sep Separator) {} |
402 | |
403 | template <typename Sep, typename Arg> |
404 | inline void join_items_impl(std::string &Result, Sep Separator, |
405 | const Arg &Item) { |
406 | Result += Item; |
407 | } |
408 | |
409 | template <typename Sep, typename Arg1, typename... Args> |
410 | inline void join_items_impl(std::string &Result, Sep Separator, const Arg1 &A1, |
411 | Args &&... Items) { |
412 | Result += A1; |
413 | Result += Separator; |
414 | join_items_impl(Result, Separator, std::forward<Args>(Items)...); |
415 | } |
416 | |
417 | inline size_t join_one_item_size(char) { return 1; } |
418 | inline size_t join_one_item_size(const char *S) { return S ? ::strlen(S) : 0; } |
419 | |
420 | template <typename T> inline size_t join_one_item_size(const T &Str) { |
421 | return Str.size(); |
422 | } |
423 | |
424 | inline size_t join_items_size() { return 0; } |
425 | |
426 | template <typename A1> inline size_t join_items_size(const A1 &A) { |
427 | return join_one_item_size(A); |
428 | } |
429 | template <typename A1, typename... Args> |
430 | inline size_t join_items_size(const A1 &A, Args &&... Items) { |
431 | return join_one_item_size(A) + join_items_size(std::forward<Args>(Items)...); |
432 | } |
433 | |
434 | } // end namespace detail |
435 | |
436 | /// Joins the strings in the range [Begin, End), adding Separator between |
437 | /// the elements. |
438 | template <typename IteratorT> |
439 | inline std::string join(IteratorT Begin, IteratorT End, StringRef Separator) { |
440 | using tag = typename std::iterator_traits<IteratorT>::iterator_category; |
441 | return detail::join_impl(Begin, End, Separator, tag()); |
442 | } |
443 | |
444 | /// Joins the strings in the range [R.begin(), R.end()), adding Separator |
445 | /// between the elements. |
446 | template <typename Range> |
447 | inline std::string join(Range &&R, StringRef Separator) { |
448 | return join(R.begin(), R.end(), Separator); |
449 | } |
450 | |
451 | /// Joins the strings in the parameter pack \p Items, adding \p Separator |
452 | /// between the elements. All arguments must be implicitly convertible to |
453 | /// std::string, or there should be an overload of std::string::operator+=() |
454 | /// that accepts the argument explicitly. |
455 | template <typename Sep, typename... Args> |
456 | inline std::string join_items(Sep Separator, Args &&... Items) { |
457 | std::string Result; |
458 | if (sizeof...(Items) == 0) |
459 | return Result; |
460 | |
461 | size_t NS = detail::join_one_item_size(Separator); |
462 | size_t NI = detail::join_items_size(std::forward<Args>(Items)...); |
463 | Result.reserve(NI + (sizeof...(Items) - 1) * NS + 1); |
464 | detail::join_items_impl(Result, Separator, std::forward<Args>(Items)...); |
465 | return Result; |
466 | } |
467 | |
468 | /// A helper class to return the specified delimiter string after the first |
469 | /// invocation of operator StringRef(). Used to generate a comma-separated |
470 | /// list from a loop like so: |
471 | /// |
472 | /// \code |
473 | /// ListSeparator LS; |
474 | /// for (auto &I : C) |
475 | /// OS << LS << I.getName(); |
476 | /// \end |
477 | class ListSeparator { |
478 | bool First = true; |
479 | StringRef Separator; |
480 | |
481 | public: |
482 | ListSeparator(StringRef Separator = ", " ) : Separator(Separator) {} |
483 | operator StringRef() { |
484 | if (First) { |
485 | First = false; |
486 | return {}; |
487 | } |
488 | return Separator; |
489 | } |
490 | }; |
491 | |
492 | } // end namespace llvm |
493 | |
494 | #endif // LLVM_ADT_STRINGEXTRAS_H |
495 | |