libfilezilla
string.hpp
Go to the documentation of this file.
1 #ifndef LIBFILEZILLA_STRING_HEADER
2 #define LIBFILEZILLA_STRING_HEADER
3 
4 #include "libfilezilla.hpp"
5 
6 #include <algorithm>
7 #include <string>
8 #include <string_view>
9 #include <vector>
10 
18 namespace fz {
19 
32 #ifdef FZ_WINDOWS
33 typedef std::wstring native_string;
34 typedef std::wstring_view native_string_view;
35 #endif
36 #if defined(FZ_UNIX) || defined(FZ_MAC)
37 typedef std::string native_string;
38 typedef std::string_view native_string_view;
39 #endif
40 
45 native_string FZ_PUBLIC_SYMBOL to_native(std::string_view const& in);
46 
51 native_string FZ_PUBLIC_SYMBOL to_native(std::wstring_view const& in);
52 
54 template<typename T, typename std::enable_if_t<std::is_same_v<native_string, typename std::decay_t<T>>, int> = 0>
55 inline native_string to_native(T const& in) {
56  return in;
57 }
58 
65 int FZ_PUBLIC_SYMBOL stricmp(std::string_view const& a, std::string_view const& b);
66 int FZ_PUBLIC_SYMBOL stricmp(std::wstring_view const& a, std::wstring_view const& b);
67 
85 template<typename Char>
86 Char tolower_ascii(Char c) {
87  if (c >= 'A' && c <= 'Z') {
88  return c + ('a' - 'A');
89  }
90  return c;
91 }
92 
93 template<>
94 std::wstring::value_type FZ_PUBLIC_SYMBOL tolower_ascii(std::wstring::value_type c);
95 
97 template<typename Char>
98 Char toupper_ascii(Char c) {
99  if (c >= 'a' && c <= 'z') {
100  return c + ('A' - 'a');
101  }
102  return c;
103 }
104 
105 template<>
106 std::wstring::value_type FZ_PUBLIC_SYMBOL toupper_ascii(std::wstring::value_type c);
107 
110  // Note: For UTF-8 strings it works on individual octets!
111 std::string FZ_PUBLIC_SYMBOL str_tolower_ascii(std::string_view const& s);
112 std::wstring FZ_PUBLIC_SYMBOL str_tolower_ascii(std::wstring_view const& s);
113 
114 std::string FZ_PUBLIC_SYMBOL str_toupper_ascii(std::string_view const& s);
115 std::wstring FZ_PUBLIC_SYMBOL str_toupper_ascii(std::wstring_view const& s);
116 
122 struct FZ_PUBLIC_SYMBOL less_insensitive_ascii final
123 {
124  template<typename T>
125  bool operator()(T const& lhs, T const& rhs) const {
126  return std::lexicographical_compare(lhs.cbegin(), lhs.cend(), rhs.cbegin(), rhs.cend(),
127  [](typename T::value_type const& a, typename T::value_type const& b) {
128  return tolower_ascii(a) < tolower_ascii(b);
129  }
130  );
131  }
132 };
133 
138 inline bool equal_insensitive_ascii(std::string_view a, std::string_view b)
139 {
140  return std::equal(a.cbegin(), a.cend(), b.cbegin(), b.cend(),
141  [](auto const& a, auto const& b) {
142  return tolower_ascii(a) == tolower_ascii(b);
143  }
144  );
145 }
146 inline bool equal_insensitive_ascii(std::wstring_view a, std::wstring_view b)
147 {
148  return std::equal(a.cbegin(), a.cend(), b.cbegin(), b.cend(),
149  [](auto const& a, auto const& b) {
150  return tolower_ascii(a) == tolower_ascii(b);
151  }
152  );
153 }
154 
159 std::wstring FZ_PUBLIC_SYMBOL to_wstring(std::string_view const& in);
160 inline std::wstring FZ_PUBLIC_SYMBOL to_wstring(std::wstring_view const& in) { return std::wstring(in); }
161 
166 template<typename T, typename std::enable_if_t<std::is_same_v<std::wstring, typename std::decay_t<T>>, int> = 0>
167 inline std::wstring to_wstring(T const& in) {
168  return in;
169 }
170 
172 template<typename Arg>
173 inline typename std::enable_if<std::is_arithmetic_v<std::decay_t<Arg>>, std::wstring>::type to_wstring(Arg && arg)
174 {
175  return std::to_wstring(std::forward<Arg>(arg));
176 }
177 
178 
183 std::wstring FZ_PUBLIC_SYMBOL to_wstring_from_utf8(std::string_view const& in);
184 std::wstring FZ_PUBLIC_SYMBOL to_wstring_from_utf8(char const* s, size_t len);
185 
186 class buffer;
187 std::wstring FZ_PUBLIC_SYMBOL to_wstring_from_utf8(fz::buffer const& in);
188 
193 std::string FZ_PUBLIC_SYMBOL to_string(std::wstring_view const& in);
194 inline std::string FZ_PUBLIC_SYMBOL to_string(std::string_view const& in) { return std::string(in); }
195 
200 template<typename T, typename std::enable_if_t<std::is_same_v<std::string, typename std::decay_t<T>>, int> = 0>
201 inline std::string to_string(T const& in) {
202  return in;
203 }
204 
205 
207 template<typename Arg>
208 inline typename std::enable_if<std::is_arithmetic_v<std::decay_t<Arg>>, std::string>::type to_string(Arg && arg)
209 {
210  return std::to_string(std::forward<Arg>(arg));
211 }
212 
213 
215 template<typename Char>
216 size_t strlen(Char const* str) {
217  return std::char_traits<Char>::length(str);
218 }
219 
220 
227 std::string FZ_PUBLIC_SYMBOL to_utf8(std::string_view const& in);
228 
235 std::string FZ_PUBLIC_SYMBOL to_utf8(std::wstring_view const& in);
236 
238 template<typename String, typename Arg>
239 inline auto toString(Arg&& arg) -> typename std::enable_if<std::is_same_v<String, std::string>, decltype(to_string(std::forward<Arg>(arg)))>::type
240 {
241  return to_string(std::forward<Arg>(arg));
242 }
243 
244 template<typename String, typename Arg>
245 inline auto toString(Arg&& arg) -> typename std::enable_if<std::is_same_v<String, std::wstring>, decltype(to_wstring(std::forward<Arg>(arg)))>::type
246 {
247  return to_wstring(std::forward<Arg>(arg));
248 }
249 
250 #if !defined(fzT) || defined(DOXYGEN)
251 #ifdef FZ_WINDOWS
252 
256 #define fzT(x) L ## x
257 #else
258 
262 #define fzT(x) x
263 #endif
264 #endif
265 
267 template<typename Char>
268 Char const* choose_string(char const* c, wchar_t const* w);
269 
270 template<> inline char const* choose_string(char const* c, wchar_t const*) { return c; }
271 template<> inline wchar_t const* choose_string(char const*, wchar_t const* w) { return w; }
272 
273 #if !defined(fzS) || defined(DOXYGEN)
274 
285 #define fzS(Char, s) fz::choose_string<Char>(s, L ## s)
286 #endif
287 
292 std::string FZ_PUBLIC_SYMBOL replaced_substrings(std::string_view const& in, std::string_view const& find, std::string_view const& replacement);
293 std::wstring FZ_PUBLIC_SYMBOL replaced_substrings(std::wstring_view const& in, std::wstring_view const& find, std::wstring_view const& replacement);
294 
296 std::string FZ_PUBLIC_SYMBOL replaced_substrings(std::string_view const& in, char find, char replacement);
297 std::wstring FZ_PUBLIC_SYMBOL replaced_substrings(std::wstring_view const& in, wchar_t find, wchar_t replacement);
298 
303 bool FZ_PUBLIC_SYMBOL replace_substrings(std::string& in, std::string_view const& find, std::string_view const& replacement);
304 bool FZ_PUBLIC_SYMBOL replace_substrings(std::wstring& in, std::wstring_view const& find, std::wstring_view const& replacement);
305 
307 bool FZ_PUBLIC_SYMBOL replace_substrings(std::string& in, char find, char replacement);
308 bool FZ_PUBLIC_SYMBOL replace_substrings(std::wstring& in, wchar_t find, wchar_t replacement);
309 
316 std::vector<std::string> FZ_PUBLIC_SYMBOL strtok(std::string_view const& tokens, std::string_view const& delims, bool const ignore_empty = true);
317 std::vector<std::wstring> FZ_PUBLIC_SYMBOL strtok(std::wstring_view const& tokens, std::wstring_view const& delims, bool const ignore_empty = true);
318 inline auto FZ_PUBLIC_SYMBOL strtok(std::string_view const& tokens, char const delim, bool const ignore_empty = true) {
319  return strtok(tokens, std::string_view(&delim, 1), ignore_empty);
320 }
321 inline auto FZ_PUBLIC_SYMBOL strtok(std::wstring_view const& tokens, wchar_t const delim, bool const ignore_empty = true) {
322  return strtok(tokens, std::wstring_view(&delim, 1), ignore_empty);
323 }
324 
333 std::vector<std::string_view> FZ_PUBLIC_SYMBOL strtok_view(std::string_view const& tokens, std::string_view const& delims, bool const ignore_empty = true);
334 std::vector<std::wstring_view> FZ_PUBLIC_SYMBOL strtok_view(std::wstring_view const& tokens, std::wstring_view const& delims, bool const ignore_empty = true);
335 inline auto FZ_PUBLIC_SYMBOL strtok_view(std::string_view const& tokens, char const delim, bool const ignore_empty = true) {
336  return strtok_view(tokens, std::string_view(&delim, 1), ignore_empty);
337 }
338 inline auto FZ_PUBLIC_SYMBOL strtok_view(std::wstring_view const& tokens, wchar_t const delim, bool const ignore_empty = true) {
339  return strtok_view(tokens, std::wstring_view(&delim, 1), ignore_empty);
340 }
341 
343 template<typename T, typename String>
344 T to_integral_impl(String const& s, T const errorval = T())
345 {
346  if constexpr (std::is_same_v<T, bool>) {
347  return static_cast<T>(to_integral_impl<unsigned int>(s, static_cast<unsigned int>(errorval))) != 0;
348  }
349  else if constexpr (std::is_enum_v<T>) {
350  return static_cast<T>(to_integral_impl<std::underlying_type_t<T>>(s, static_cast<std::underlying_type_t<T>>(errorval)));
351  }
352  else {
353  T ret{};
354  auto it = s.cbegin();
355  if (it != s.cend() && (*it == '-' || *it == '+')) {
356  ++it;
357  }
358 
359  if (it == s.cend()) {
360  return errorval;
361  }
362 
363  for (; it != s.cend(); ++it) {
364  auto const& c = *it;
365  if (c < '0' || c > '9') {
366  return errorval;
367  }
368  ret *= 10;
369  ret += c - '0';
370  }
371 
372  if (!s.empty() && s.front() == '-') {
373  ret *= static_cast<T>(-1);
374  }
375  return ret;
376  }
377 }
378 
380 template<typename T>
381 T to_integral(std::string_view const& s, T const errorval = T()) {
382  return to_integral_impl<T>(s, errorval);
383 }
384 
385 template<typename T>
386 T to_integral(std::wstring_view const& s, T const errorval = T()) {
387  return to_integral_impl<T>(s, errorval);
388 }
389 
390 template<typename T, typename StringType>
391 T to_integral(std::basic_string_view<StringType> const& s, T const errorval = T()) {
392  return to_integral_impl<T>(s, errorval);
393 }
394 
395 
397 template<typename String>
398 bool str_is_ascii(String const& s) {
399  for (auto const& c : s) {
400  if (static_cast<std::make_unsigned_t<typename String::value_type>>(c) > 127) {
401  return false;
402  }
403  }
404 
405  return true;
406 }
407 
409 template<typename String, typename Chars>
410 void trim_impl(String & s, Chars const& chars, bool fromLeft, bool fromRight) {
411  size_t const first = fromLeft ? s.find_first_not_of(chars) : 0;
412  if (first == String::npos) {
413  s = String();
414  return;
415  }
416 
417  size_t const last = fromRight ? s.find_last_not_of(chars) : s.size();
418  if (last == String::npos) {
419  s = String();
420  return;
421  }
422 
423  // Invariant: If first exists, then last >= first
424  s = s.substr(first, last - first + 1);
425 }
426 
428 inline std::string FZ_PUBLIC_SYMBOL trimmed(std::string_view s, std::string_view const& chars = " \r\n\t", bool fromLeft = true, bool fromRight = true)
429 {
430  trim_impl(s, chars, fromLeft, fromRight);
431  return std::string(s);
432 }
433 
434 inline std::wstring FZ_PUBLIC_SYMBOL trimmed(std::wstring_view s, std::wstring_view const& chars = L" \r\n\t", bool fromLeft = true, bool fromRight = true)
435 {
436  trim_impl(s, chars, fromLeft, fromRight);
437  return std::wstring(s);
438 }
439 
440 inline std::string FZ_PUBLIC_SYMBOL ltrimmed(std::string_view s, std::string_view const& chars = " \r\n\t")
441 {
442  trim_impl(s, chars, true, false);
443  return std::string(s);
444 }
445 
446 inline std::wstring FZ_PUBLIC_SYMBOL ltrimmed(std::wstring_view s, std::wstring_view const& chars = L" \r\n\t")
447 {
448  trim_impl(s, chars, true, false);
449  return std::wstring(s);
450 }
451 
452 inline std::string FZ_PUBLIC_SYMBOL rtrimmed(std::string_view s, std::string_view const& chars = " \r\n\t")
453 {
454  trim_impl(s, chars, false, true);
455  return std::string(s);
456 }
457 
458 inline std::wstring FZ_PUBLIC_SYMBOL rtrimmed(std::wstring_view s, std::wstring_view const& chars = L" \r\n\t")
459 {
460  trim_impl(s, chars, false, true);
461  return std::wstring(s);
462 }
463 
464 
466 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
467 inline void trim(String & s, std::string_view const& chars = " \r\n\t", bool fromLeft = true, bool fromRight = true)
468 {
469  trim_impl(s, chars, fromLeft, fromRight);
470 }
471 
472 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
473 inline void trim(String & s, std::wstring_view const& chars = L" \r\n\t", bool fromLeft = true, bool fromRight = true)
474 {
475  trim_impl(s, chars, fromLeft, fromRight);
476 }
477 
478 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
479 inline void ltrim(String& s, std::string_view const& chars = " \r\n\t")
480 {
481  trim_impl(s, chars, true, false);
482 }
483 
484 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
485 inline void ltrim(String& s, std::wstring_view const& chars = L" \r\n\t")
486 {
487  trim_impl(s, chars, true, false);
488 }
489 
490 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
491 inline void rtrim(String& s, std::string_view const& chars = " \r\n\t")
492 {
493  trim_impl(s, chars, false, true);
494 }
495 
496 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
497 inline void rtrim(String & s, std::wstring_view const& chars = L" \r\n\t")
498 {
499  trim_impl(s, chars, false, true);
500 }
501 
506 template<bool insensitive_ascii = false, typename String>
507 bool starts_with(String const& s, String const& beginning)
508 {
509  if (beginning.size() > s.size()) {
510  return false;
511  }
512  if constexpr (insensitive_ascii) {
513  return std::equal(beginning.begin(), beginning.end(), s.begin(), [](typename String::value_type const& a, typename String::value_type const& b) {
514  return tolower_ascii(a) == tolower_ascii(b);
515  });
516  }
517  else {
518  return std::equal(beginning.begin(), beginning.end(), s.begin());
519  }
520 }
521 
526 template<bool insensitive_ascii = false, typename String>
527 bool ends_with(String const& s, String const& ending)
528 {
529  if (ending.size() > s.size()) {
530  return false;
531  }
532 
533  if constexpr (insensitive_ascii) {
534  return std::equal(ending.rbegin(), ending.rend(), s.rbegin(), [](typename String::value_type const& a, typename String::value_type const& b) {
535  return tolower_ascii(a) == tolower_ascii(b);
536  });
537  }
538  else {
539  return std::equal(ending.rbegin(), ending.rend(), s.rbegin());
540  }
541 }
542 
548 std::string FZ_PUBLIC_SYMBOL normalize_hyphens(std::string_view const& in);
549 std::wstring FZ_PUBLIC_SYMBOL normalize_hyphens(std::wstring_view const& in);
550 
551 }
552 
553 #endif
auto toString(Arg &&arg) -> typename std::enable_if< std::is_same_v< String, std::string >, decltype(to_string(std::forward< Arg >(arg)))>::type
Calls either fz::to_string or fz::to_wstring depending on the passed template argument.
Definition: string.hpp:239
T to_integral(std::string_view const &s, T const errorval=T())
Converts string to integral type T. If string is not convertible, errorval is returned.
Definition: string.hpp:381
Comparator to be used for std::map for case-insensitive keys.
Definition: string.hpp:122
bool starts_with(String const &s, String const &beginning)
Tests whether the first string starts with the second string.
Definition: string.hpp:507
bool replace_substrings(std::string &in, std::string_view const &find, std::string_view const &replacement)
Modifies in, replacing all occurrences of find with replacement.
std::wstring to_wstring(std::string_view const &in)
Converts from std::string in system encoding into std::wstring.
Char toupper_ascii(Char c)
Converts ASCII lowercase characters to uppercase as if C-locale is used.
Definition: string.hpp:98
native_string to_native(std::string_view const &in)
Converts std::string to native_string.
std::string to_utf8(std::string_view const &in)
Converts from std::string in native encoding into std::string in UTF-8.
std::vector< std::string_view > strtok_view(std::string_view const &tokens, std::string_view const &delims, bool const ignore_empty=true)
Tokenizes string.
std::string trimmed(std::string_view s, std::string_view const &chars=" \r\n\t", bool fromLeft=true, bool fromRight=true)
Return passed string with all leading and trailing whitespace removed.
Definition: string.hpp:428
bool equal_insensitive_ascii(std::string_view a, std::string_view b)
Locale-insensitive stricmp.
Definition: string.hpp:138
std::string replaced_substrings(std::string_view const &in, std::string_view const &find, std::string_view const &replacement)
Returns in with all occurrences of find in the input string replaced with replacement.
Char tolower_ascii(Char c)
Converts ASCII uppercase characters to lowercase as if C-locale is used.
Definition: string.hpp:86
std::enable_if< std::is_arithmetic_v< std::decay_t< Arg > >, std::string >::type to_string(Arg &&arg)
Converts from arithmetic type to std::string.
Definition: string.hpp:208
std::wstring to_wstring_from_utf8(std::string_view const &in)
Converts from std::string in UTF-8 into std::wstring.
size_t strlen(Char const *str)
Returns length of 0-terminated character sequence. Works with both narrow and wide-characters.
Definition: string.hpp:216
std::string normalize_hyphens(std::string_view const &in)
bool str_is_ascii(String const &s)
Returns true iff the string only has characters in the 7-bit ASCII range.
Definition: string.hpp:398
std::string str_tolower_ascii(std::string_view const &s)
tr_tolower_ascii does for strings what tolower_ascii does for individual characters ...
bool ends_with(String const &s, String const &ending)
Tests whether the first string ends with the second string.
Definition: string.hpp:527
std::wstring native_string
A string in the system's native character type and encoding. Note: This typedef changes depending on...
Definition: string.hpp:33
void trim(String &s, std::string_view const &chars=" \r\n\t", bool fromLeft=true, bool fromRight=true)
Remove all leading and trailing whitespace from string.
Definition: string.hpp:467
The namespace used by libfilezilla.
Definition: apply.hpp:17
Char const * choose_string(char const *c, wchar_t const *w)
Returns the function argument of the type matching the template argument.
Definition: string.hpp:270
std::vector< std::string > strtok(std::string_view const &tokens, std::string_view const &delims, bool const ignore_empty=true)
Tokenizes string.
Sets some global macros and further includes string.hpp.
int stricmp(std::string_view const &a, std::string_view const &b)
Locale-sensitive stricmp.
The buffer class is a simple buffer where data can be appended at the end and consumed at the front...
Definition: buffer.hpp:25
std::enable_if< std::is_arithmetic_v< std::decay_t< Arg > >, std::wstring >::type to_wstring(Arg &&arg)
Converts from arithmetic type to std::wstring.
Definition: string.hpp:173
std::string to_string(std::wstring_view const &in)
Converts from std::wstring into std::string in system encoding.