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 LLVM_ADT_STRINGEXTRAS_H
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.
toStringRefArray(const char * const * Strings)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.
toStringRef(bool B)52 inline StringRef toStringRef(bool B) { return StringRef(B ? "true" : "false"); }
53
54 /// Construct a string ref from an array ref of unsigned chars.
toStringRef(ArrayRef<uint8_t> Input)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.
arrayRefFromStringRef(StringRef Input)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.
hexDigitValue(char C)68 inline unsigned hexDigitValue(char C) {
69 if (C >= '0' && C <= '9') return C-'0';
70 if (C >= 'a' && C <= 'f') return C-'a'+10U;
71 if (C >= 'A' && C <= 'F') return C-'A'+10U;
72 return -1U;
73 }
74
75 /// Checks if character \p C is one of the 10 decimal digits.
isDigit(char C)76 inline bool isDigit(char C) { return C >= '0' && C <= '9'; }
77
78 /// Checks if character \p C is a hexadecimal numeric character.
isHexDigit(char C)79 inline bool isHexDigit(char C) { return hexDigitValue(C) != -1U; }
80
81 /// Checks if character \p C is a valid letter as classified by "C" locale.
isAlpha(char C)82 inline bool isAlpha(char C) {
83 return ('a' <= C && C <= 'z') || ('A' <= C && C <= 'Z');
84 }
85
86 /// Checks whether character \p C is either a decimal digit or an uppercase or
87 /// lowercase letter as classified by "C" locale.
isAlnum(char C)88 inline bool isAlnum(char C) { return isAlpha(C) || isDigit(C); }
89
90 /// Checks whether character \p C is valid ASCII (high bit is zero).
isASCII(char C)91 inline bool isASCII(char C) { return static_cast<unsigned char>(C) <= 127; }
92
93 /// Checks whether all characters in S are ASCII.
isASCII(llvm::StringRef S)94 inline bool isASCII(llvm::StringRef S) {
95 for (char C : S)
96 if (LLVM_UNLIKELY(!isASCII(C)))
97 return false;
98 return true;
99 }
100
101 /// Checks whether character \p C is printable.
102 ///
103 /// Locale-independent version of the C standard library isprint whose results
104 /// may differ on different platforms.
isPrint(char C)105 inline bool isPrint(char C) {
106 unsigned char UC = static_cast<unsigned char>(C);
107 return (0x20 <= UC) && (UC <= 0x7E);
108 }
109
110 /// Checks whether character \p C is whitespace in the "C" locale.
111 ///
112 /// Locale-independent version of the C standard library isspace.
isSpace(char C)113 inline bool isSpace(char C) {
114 return C == ' ' || C == '\f' || C == '\n' || C == '\r' || C == '\t' ||
115 C == '\v';
116 }
117
118 /// Returns the corresponding lowercase character if \p x is uppercase.
toLower(char x)119 inline char toLower(char x) {
120 if (x >= 'A' && x <= 'Z')
121 return x - 'A' + 'a';
122 return x;
123 }
124
125 /// Returns the corresponding uppercase character if \p x is lowercase.
toUpper(char x)126 inline char toUpper(char x) {
127 if (x >= 'a' && x <= 'z')
128 return x - 'a' + 'A';
129 return x;
130 }
131
132 inline std::string utohexstr(uint64_t X, bool LowerCase = false) {
133 char Buffer[17];
134 char *BufPtr = std::end(Buffer);
135
136 if (X == 0) *--BufPtr = '0';
137
138 while (X) {
139 unsigned char Mod = static_cast<unsigned char>(X) & 15;
140 *--BufPtr = hexdigit(Mod, LowerCase);
141 X >>= 4;
142 }
143
144 return std::string(BufPtr, std::end(Buffer));
145 }
146
147 /// Convert buffer \p Input to its hexadecimal representation.
148 /// The returned string is double the size of \p Input.
149 inline std::string toHex(StringRef Input, bool LowerCase = false) {
150 static const char *const LUT = "0123456789ABCDEF";
151 const uint8_t Offset = LowerCase ? 32 : 0;
152 size_t Length = Input.size();
153
154 std::string Output;
155 Output.reserve(2 * Length);
156 for (size_t i = 0; i < Length; ++i) {
157 const unsigned char c = Input[i];
158 Output.push_back(LUT[c >> 4] | Offset);
159 Output.push_back(LUT[c & 15] | Offset);
160 }
161 return Output;
162 }
163
164 inline std::string toHex(ArrayRef<uint8_t> Input, bool LowerCase = false) {
165 return toHex(toStringRef(Input), LowerCase);
166 }
167
hexFromNibbles(char MSB,char LSB)168 inline uint8_t hexFromNibbles(char MSB, char LSB) {
169 unsigned U1 = hexDigitValue(MSB);
170 unsigned U2 = hexDigitValue(LSB);
171 assert(U1 != -1U && U2 != -1U);
172
173 return static_cast<uint8_t>((U1 << 4) | U2);
174 }
175
176 /// Convert hexadecimal string \p Input to its binary representation.
177 /// The return string is half the size of \p Input.
fromHex(StringRef Input)178 inline std::string fromHex(StringRef Input) {
179 if (Input.empty())
180 return std::string();
181
182 std::string Output;
183 Output.reserve((Input.size() + 1) / 2);
184 if (Input.size() % 2 == 1) {
185 Output.push_back(hexFromNibbles('0', Input.front()));
186 Input = Input.drop_front();
187 }
188
189 assert(Input.size() % 2 == 0);
190 while (!Input.empty()) {
191 uint8_t Hex = hexFromNibbles(Input[0], Input[1]);
192 Output.push_back(Hex);
193 Input = Input.drop_front(2);
194 }
195 return Output;
196 }
197
198 /// Convert the string \p S to an integer of the specified type using
199 /// the radix \p Base. If \p Base is 0, auto-detects the radix.
200 /// Returns true if the number was successfully converted, false otherwise.
201 template <typename N> bool to_integer(StringRef S, N &Num, unsigned Base = 0) {
202 return !S.getAsInteger(Base, Num);
203 }
204
205 namespace detail {
206 template <typename N>
to_float(const Twine & T,N & Num,N (* StrTo)(const char *,char **))207 inline bool to_float(const Twine &T, N &Num, N (*StrTo)(const char *, char **)) {
208 SmallString<32> Storage;
209 StringRef S = T.toNullTerminatedStringRef(Storage);
210 char *End;
211 N Temp = StrTo(S.data(), &End);
212 if (*End != '\0')
213 return false;
214 Num = Temp;
215 return true;
216 }
217 }
218
to_float(const Twine & T,float & Num)219 inline bool to_float(const Twine &T, float &Num) {
220 return detail::to_float(T, Num, strtof);
221 }
222
to_float(const Twine & T,double & Num)223 inline bool to_float(const Twine &T, double &Num) {
224 return detail::to_float(T, Num, strtod);
225 }
226
to_float(const Twine & T,long double & Num)227 inline bool to_float(const Twine &T, long double &Num) {
228 return detail::to_float(T, Num, strtold);
229 }
230
231 inline std::string utostr(uint64_t X, bool isNeg = false) {
232 char Buffer[21];
233 char *BufPtr = std::end(Buffer);
234
235 if (X == 0) *--BufPtr = '0'; // Handle special case...
236
237 while (X) {
238 *--BufPtr = '0' + char(X % 10);
239 X /= 10;
240 }
241
242 if (isNeg) *--BufPtr = '-'; // Add negative sign...
243 return std::string(BufPtr, std::end(Buffer));
244 }
245
itostr(int64_t X)246 inline std::string itostr(int64_t X) {
247 if (X < 0)
248 return utostr(-static_cast<uint64_t>(X), true);
249 else
250 return utostr(static_cast<uint64_t>(X));
251 }
252
253 /// StrInStrNoCase - Portable version of strcasestr. Locates the first
254 /// occurrence of string 's1' in string 's2', ignoring case. Returns
255 /// the offset of s2 in s1 or npos if s2 cannot be found.
256 StringRef::size_type StrInStrNoCase(StringRef s1, StringRef s2);
257
258 /// getToken - This function extracts one token from source, ignoring any
259 /// leading characters that appear in the Delimiters string, and ending the
260 /// token at any of the characters that appear in the Delimiters string. If
261 /// there are no tokens in the source string, an empty string is returned.
262 /// The function returns a pair containing the extracted token and the
263 /// remaining tail string.
264 std::pair<StringRef, StringRef> getToken(StringRef Source,
265 StringRef Delimiters = " \t\n\v\f\r");
266
267 /// SplitString - Split up the specified string according to the specified
268 /// delimiters, appending the result fragments to the output list.
269 void SplitString(StringRef Source,
270 SmallVectorImpl<StringRef> &OutFragments,
271 StringRef Delimiters = " \t\n\v\f\r");
272
273 /// Returns the English suffix for an ordinal integer (-st, -nd, -rd, -th).
getOrdinalSuffix(unsigned Val)274 inline StringRef getOrdinalSuffix(unsigned Val) {
275 // It is critically important that we do this perfectly for
276 // user-written sequences with over 100 elements.
277 switch (Val % 100) {
278 case 11:
279 case 12:
280 case 13:
281 return "th";
282 default:
283 switch (Val % 10) {
284 case 1: return "st";
285 case 2: return "nd";
286 case 3: return "rd";
287 default: return "th";
288 }
289 }
290 }
291
292 /// Print each character of the specified string, escaping it if it is not
293 /// printable or if it is an escape char.
294 void printEscapedString(StringRef Name, raw_ostream &Out);
295
296 /// Print each character of the specified string, escaping HTML special
297 /// characters.
298 void printHTMLEscaped(StringRef String, raw_ostream &Out);
299
300 /// printLowerCase - Print each character as lowercase if it is uppercase.
301 void printLowerCase(StringRef String, raw_ostream &Out);
302
303 /// Converts a string from camel-case to snake-case by replacing all uppercase
304 /// letters with '_' followed by the letter in lowercase, except if the
305 /// uppercase letter is the first character of the string.
306 std::string convertToSnakeFromCamelCase(StringRef input);
307
308 /// Converts a string from snake-case to camel-case by replacing all occurrences
309 /// of '_' followed by a lowercase letter with the letter in uppercase.
310 /// Optionally allow capitalization of the first letter (if it is a lowercase
311 /// letter)
312 std::string convertToCamelFromSnakeCase(StringRef input,
313 bool capitalizeFirst = false);
314
315 namespace detail {
316
317 template <typename IteratorT>
join_impl(IteratorT Begin,IteratorT End,StringRef Separator,std::input_iterator_tag)318 inline std::string join_impl(IteratorT Begin, IteratorT End,
319 StringRef Separator, std::input_iterator_tag) {
320 std::string S;
321 if (Begin == End)
322 return S;
323
324 S += (*Begin);
325 while (++Begin != End) {
326 S += Separator;
327 S += (*Begin);
328 }
329 return S;
330 }
331
332 template <typename IteratorT>
join_impl(IteratorT Begin,IteratorT End,StringRef Separator,std::forward_iterator_tag)333 inline std::string join_impl(IteratorT Begin, IteratorT End,
334 StringRef Separator, std::forward_iterator_tag) {
335 std::string S;
336 if (Begin == End)
337 return S;
338
339 size_t Len = (std::distance(Begin, End) - 1) * Separator.size();
340 for (IteratorT I = Begin; I != End; ++I)
341 Len += (*Begin).size();
342 S.reserve(Len);
343 S += (*Begin);
344 while (++Begin != End) {
345 S += Separator;
346 S += (*Begin);
347 }
348 return S;
349 }
350
351 template <typename Sep>
join_items_impl(std::string & Result,Sep Separator)352 inline void join_items_impl(std::string &Result, Sep Separator) {}
353
354 template <typename Sep, typename Arg>
join_items_impl(std::string & Result,Sep Separator,const Arg & Item)355 inline void join_items_impl(std::string &Result, Sep Separator,
356 const Arg &Item) {
357 Result += Item;
358 }
359
360 template <typename Sep, typename Arg1, typename... Args>
join_items_impl(std::string & Result,Sep Separator,const Arg1 & A1,Args &&...Items)361 inline void join_items_impl(std::string &Result, Sep Separator, const Arg1 &A1,
362 Args &&... Items) {
363 Result += A1;
364 Result += Separator;
365 join_items_impl(Result, Separator, std::forward<Args>(Items)...);
366 }
367
join_one_item_size(char)368 inline size_t join_one_item_size(char) { return 1; }
join_one_item_size(const char * S)369 inline size_t join_one_item_size(const char *S) { return S ? ::strlen(S) : 0; }
370
join_one_item_size(const T & Str)371 template <typename T> inline size_t join_one_item_size(const T &Str) {
372 return Str.size();
373 }
374
join_items_size()375 inline size_t join_items_size() { return 0; }
376
join_items_size(const A1 & A)377 template <typename A1> inline size_t join_items_size(const A1 &A) {
378 return join_one_item_size(A);
379 }
380 template <typename A1, typename... Args>
join_items_size(const A1 & A,Args &&...Items)381 inline size_t join_items_size(const A1 &A, Args &&... Items) {
382 return join_one_item_size(A) + join_items_size(std::forward<Args>(Items)...);
383 }
384
385 } // end namespace detail
386
387 /// Joins the strings in the range [Begin, End), adding Separator between
388 /// the elements.
389 template <typename IteratorT>
join(IteratorT Begin,IteratorT End,StringRef Separator)390 inline std::string join(IteratorT Begin, IteratorT End, StringRef Separator) {
391 using tag = typename std::iterator_traits<IteratorT>::iterator_category;
392 return detail::join_impl(Begin, End, Separator, tag());
393 }
394
395 /// Joins the strings in the range [R.begin(), R.end()), adding Separator
396 /// between the elements.
397 template <typename Range>
join(Range && R,StringRef Separator)398 inline std::string join(Range &&R, StringRef Separator) {
399 return join(R.begin(), R.end(), Separator);
400 }
401
402 /// Joins the strings in the parameter pack \p Items, adding \p Separator
403 /// between the elements. All arguments must be implicitly convertible to
404 /// std::string, or there should be an overload of std::string::operator+=()
405 /// that accepts the argument explicitly.
406 template <typename Sep, typename... Args>
join_items(Sep Separator,Args &&...Items)407 inline std::string join_items(Sep Separator, Args &&... Items) {
408 std::string Result;
409 if (sizeof...(Items) == 0)
410 return Result;
411
412 size_t NS = detail::join_one_item_size(Separator);
413 size_t NI = detail::join_items_size(std::forward<Args>(Items)...);
414 Result.reserve(NI + (sizeof...(Items) - 1) * NS + 1);
415 detail::join_items_impl(Result, Separator, std::forward<Args>(Items)...);
416 return Result;
417 }
418
419 } // end namespace llvm
420
421 #endif // LLVM_ADT_STRINGEXTRAS_H
422