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<std::decay_t<Arg>>::value, 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 const& in);
184 std::wstring FZ_PUBLIC_SYMBOL to_wstring_from_utf8(char const* s, size_t len);
185 
190 std::string FZ_PUBLIC_SYMBOL to_string(std::wstring_view const& in);
191 inline std::string FZ_PUBLIC_SYMBOL to_string(std::string_view const& in) { return std::string(in); }
192 
197 template<typename T, typename std::enable_if_t<std::is_same_v<std::string, typename std::decay_t<T>>, int> = 0>
198 inline std::string to_string(T const& in) {
199  return in;
200 }
201 
202 
204 template<typename Arg>
205 inline typename std::enable_if<std::is_arithmetic<std::decay_t<Arg>>::value, std::string>::type to_string(Arg && arg)
206 {
207  return std::to_string(std::forward<Arg>(arg));
208 }
209 
210 
212 template<typename Char>
213 size_t strlen(Char const* str) {
214  return std::char_traits<Char>::length(str);
215 }
216 
217 
224 std::string FZ_PUBLIC_SYMBOL to_utf8(std::string_view const& in);
225 
232 std::string FZ_PUBLIC_SYMBOL to_utf8(std::wstring_view const& in);
233 
235 template<typename String, typename Arg>
236 inline auto toString(Arg&& arg) -> typename std::enable_if<std::is_same<String, std::string>::value, decltype(to_string(std::forward<Arg>(arg)))>::type
237 {
238  return to_string(std::forward<Arg>(arg));
239 }
240 
241 template<typename String, typename Arg>
242 inline auto toString(Arg&& arg) -> typename std::enable_if<std::is_same<String, std::wstring>::value, decltype(to_wstring(std::forward<Arg>(arg)))>::type
243 {
244  return to_wstring(std::forward<Arg>(arg));
245 }
246 
247 #if !defined(fzT) || defined(DOXYGEN)
248 #ifdef FZ_WINDOWS
249 
253 #define fzT(x) L ## x
254 #else
255 
259 #define fzT(x) x
260 #endif
261 #endif
262 
264 template<typename Char>
265 Char const* choose_string(char const* c, wchar_t const* w);
266 
267 template<> inline char const* choose_string(char const* c, wchar_t const*) { return c; }
268 template<> inline wchar_t const* choose_string(char const*, wchar_t const* w) { return w; }
269 
270 #if !defined(fzS) || defined(DOXYGEN)
271 
282 #define fzS(Char, s) fz::choose_string<Char>(s, L ## s)
283 #endif
284 
286 std::string FZ_PUBLIC_SYMBOL replaced_substrings(std::string const& in, std::string const& find, std::string const& replacement);
287 std::wstring FZ_PUBLIC_SYMBOL replaced_substrings(std::wstring const& in, std::wstring const& find, std::wstring const& replacement);
288 
290 bool FZ_PUBLIC_SYMBOL replace_substrings(std::string& in, std::string const& find, std::string const& replacement);
291 bool FZ_PUBLIC_SYMBOL replace_substrings(std::wstring& in, std::wstring const& find, std::wstring const& replacement);
292 
299 std::vector<std::string> FZ_PUBLIC_SYMBOL strtok(std::string_view const& tokens, std::string_view const& delims, bool const ignore_empty = true);
300 std::vector<std::wstring> FZ_PUBLIC_SYMBOL strtok(std::wstring_view const& tokens, std::wstring_view const& delims, bool const ignore_empty = true);
301 inline auto FZ_PUBLIC_SYMBOL strtok(std::string_view const& tokens, char const delim, bool const ignore_empty = true) {
302  return strtok(tokens, std::string_view(&delim, 1), ignore_empty);
303 }
304 inline auto FZ_PUBLIC_SYMBOL strtok(std::wstring_view const& tokens, wchar_t const delim, bool const ignore_empty = true) {
305  return strtok(tokens, std::wstring_view(&delim, 1), ignore_empty);
306 }
307 
316 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);
317 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);
318 inline auto FZ_PUBLIC_SYMBOL strtok_view(std::string_view const& tokens, char const delim, bool const ignore_empty = true) {
319  return strtok_view(tokens, std::string_view(&delim, 1), ignore_empty);
320 }
321 inline auto FZ_PUBLIC_SYMBOL strtok_view(std::wstring_view const& tokens, wchar_t const delim, bool const ignore_empty = true) {
322  return strtok_view(tokens, std::wstring_view(&delim, 1), ignore_empty);
323 }
324 
326 template<typename T, typename String>
327 T to_integral_impl(String const& s, T const errorval = T())
328 {
329  if constexpr (std::is_enum_v<T>) {
330  return static_cast<T>(to_integral_impl<std::underlying_type_t<T>>(s, static_cast<std::underlying_type_t<T>>(errorval)));
331  }
332  else {
333  T ret{};
334  auto it = s.cbegin();
335  if (it != s.cend() && (*it == '-' || *it == '+')) {
336  ++it;
337  }
338 
339  if (it == s.cend()) {
340  return errorval;
341  }
342 
343  for (; it != s.cend(); ++it) {
344  auto const& c = *it;
345  if (c < '0' || c > '9') {
346  return errorval;
347  }
348  ret *= 10;
349  ret += c - '0';
350  }
351 
352  if (!s.empty() && s.front() == '-') {
353  ret *= static_cast<T>(-1);
354  }
355  return ret;
356  }
357 }
358 
360 template<typename T>
361 T to_integral(std::string_view const& s, T const errorval = T()) {
362  return to_integral_impl<T>(s, errorval);
363 }
364 
365 template<typename T>
366 T to_integral(std::wstring_view const& s, T const errorval = T()) {
367  return to_integral_impl<T>(s, errorval);
368 }
369 
370 template<typename T, typename StringType>
371 T to_integral(std::basic_string_view<StringType> const& s, T const errorval = T()) {
372  return to_integral_impl<T>(s, errorval);
373 }
374 
375 
377 template<typename String>
378 bool str_is_ascii(String const& s) {
379  for (auto const& c : s) {
380  if (static_cast<std::make_unsigned_t<typename String::value_type>>(c) > 127) {
381  return false;
382  }
383  }
384 
385  return true;
386 }
387 
389 template<typename String, typename Chars>
390 void trim_impl(String & s, Chars const& chars, bool fromLeft, bool fromRight) {
391  size_t const first = fromLeft ? s.find_first_not_of(chars) : 0;
392  if (first == String::npos) {
393  s = String();
394  return;
395  }
396 
397  size_t const last = fromRight ? s.find_last_not_of(chars) : s.size();
398  if (last == String::npos) {
399  s = String();
400  return;
401  }
402 
403  // Invariant: If first exists, then last >= first
404  s = s.substr(first, last - first + 1);
405 }
406 
408 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)
409 {
410  trim_impl(s, chars, fromLeft, fromRight);
411  return std::string(s);
412 }
413 
414 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)
415 {
416  trim_impl(s, chars, fromLeft, fromRight);
417  return std::wstring(s);
418 }
419 
420 inline std::string FZ_PUBLIC_SYMBOL ltrimmed(std::string_view s, std::string_view const& chars = " \r\n\t")
421 {
422  trim_impl(s, chars, true, false);
423  return std::string(s);
424 }
425 
426 inline std::wstring FZ_PUBLIC_SYMBOL ltrimmed(std::wstring_view s, std::wstring_view const& chars = L" \r\n\t")
427 {
428  trim_impl(s, chars, true, false);
429  return std::wstring(s);
430 }
431 
432 inline std::string FZ_PUBLIC_SYMBOL rtrimmed(std::string_view s, std::string_view const& chars = " \r\n\t")
433 {
434  trim_impl(s, chars, false, true);
435  return std::string(s);
436 }
437 
438 inline std::wstring FZ_PUBLIC_SYMBOL rtrimmed(std::wstring_view s, std::wstring_view const& chars = L" \r\n\t")
439 {
440  trim_impl(s, chars, false, true);
441  return std::wstring(s);
442 }
443 
444 
446 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
447 inline void trim(String & s, std::string_view const& chars = " \r\n\t", bool fromLeft = true, bool fromRight = true)
448 {
449  trim_impl(s, chars, fromLeft, fromRight);
450 }
451 
452 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
453 inline void trim(String & s, std::wstring_view const& chars = L" \r\n\t", bool fromLeft = true, bool fromRight = true)
454 {
455  trim_impl(s, chars, fromLeft, fromRight);
456 }
457 
458 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
459 inline void ltrim(String& s, std::string_view const& chars = " \r\n\t")
460 {
461  trim_impl(s, chars, true, false);
462 }
463 
464 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
465 inline void ltrim(String& s, std::wstring_view const& chars = L" \r\n\t")
466 {
467  trim_impl(s, chars, true, false);
468 }
469 
470 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, char>, int> = 0>
471 inline void rtrim(String& s, std::string_view const& chars = " \r\n\t")
472 {
473  trim_impl(s, chars, false, true);
474 }
475 
476 template<typename String, typename std::enable_if_t<std::is_same_v<typename String::value_type, wchar_t>, int> = 0>
477 inline void rtrim(String & s, std::wstring_view const& chars = L" \r\n\t")
478 {
479  trim_impl(s, chars, false, true);
480 }
481 
486 template<bool insensitive_ascii = false, typename String>
487 bool starts_with(String const& s, String const& beginning)
488 {
489  if (beginning.size() > s.size()) {
490  return false;
491  }
492  if constexpr (insensitive_ascii) {
493  return std::equal(beginning.begin(), beginning.end(), s.begin(), [](typename String::value_type const& a, typename String::value_type const& b) {
494  return tolower_ascii(a) == tolower_ascii(b);
495  });
496  }
497  else {
498  return std::equal(beginning.begin(), beginning.end(), s.begin());
499  }
500 }
501 
506 template<bool insensitive_ascii = false, typename String>
507 bool ends_with(String const& s, String const& ending)
508 {
509  if (ending.size() > s.size()) {
510  return false;
511  }
512 
513  if constexpr (insensitive_ascii) {
514  return std::equal(ending.rbegin(), ending.rend(), s.rbegin(), [](typename String::value_type const& a, typename String::value_type const& b) {
515  return tolower_ascii(a) == tolower_ascii(b);
516  });
517  }
518  else {
519  return std::equal(ending.rbegin(), ending.rend(), s.rbegin());
520  }
521 }
522 
523 }
524 
525 #endif
fz::strtok_view
std::vector< std::string_view > strtok_view(std::string_view const &tokens, std::string_view const &delims, bool const ignore_empty=true)
Tokenizes string.
fz::str_is_ascii
bool str_is_ascii(String const &s)
Returns true iff the string only has characters in the 7-bit ASCII range.
Definition: string.hpp:378
fz::less_insensitive_ascii
Comparator to be used for std::map for case-insensitive keys.
Definition: string.hpp:123
fz::to_wstring
std::enable_if< std::is_arithmetic< std::decay_t< Arg > >::value, std::wstring >::type to_wstring(Arg &&arg)
Converts from arithmetic type to std::wstring.
Definition: string.hpp:173
fz::to_native
native_string to_native(std::string_view const &in)
Converts std::string to native_string.
fz::to_utf8
std::string to_utf8(std::string_view const &in)
Converts from std::string in native encoding into std::string in UTF-8.
fz::to_integral
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:361
fz::tolower_ascii
Char tolower_ascii(Char c)
Converts ASCII uppercase characters to lowercase as if C-locale is used.
Definition: string.hpp:86
fz::replace_substrings
bool replace_substrings(std::string &in, std::string const &find, std::string const &replacement)
Modifies in, replacing all occurrences of find with replacement.
fz::starts_with
bool starts_with(String const &s, String const &beginning)
Tests whether the first string starts with the second string.
Definition: string.hpp:487
fz::equal_insensitive_ascii
bool equal_insensitive_ascii(std::string_view a, std::string_view b)
Locale-insensitive stricmp.
Definition: string.hpp:138
fz::strlen
size_t strlen(Char const *str)
Returns length of 0-terminated character sequence. Works with both narrow and wide-characters.
Definition: string.hpp:213
fz::native_string
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
fz::to_wstring
std::wstring to_wstring(std::string_view const &in)
Converts from std::string in system encoding into std::wstring.
fz::to_string
std::string to_string(std::wstring_view const &in)
Converts from std::wstring into std::string in system encoding.
fz::to_string
std::enable_if< std::is_arithmetic< std::decay_t< Arg > >::value, std::string >::type to_string(Arg &&arg)
Converts from arithmetic type to std::string.
Definition: string.hpp:205
fz::trimmed
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:408
fz::toString
auto toString(Arg &&arg) -> typename std::enable_if< std::is_same< String, std::string >::value, 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:236
fz::to_wstring_from_utf8
std::wstring to_wstring_from_utf8(std::string const &in)
Converts from std::string in UTF-8 into std::wstring.
fz::trim
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:447
fz::choose_string
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:267
fz::stricmp
int stricmp(std::string_view const &a, std::string_view const &b)
Locale-sensitive stricmp.
fz::ends_with
bool ends_with(String const &s, String const &ending)
Tests whether the first string ends with the second string.
Definition: string.hpp:507
fz::toupper_ascii
Char toupper_ascii(Char c)
Converts ASCII lowercase characters to uppercase as if C-locale is used.
Definition: string.hpp:98
libfilezilla.hpp
Sets some global macros and further includes string.hpp.
fz::strtok
std::vector< std::string > strtok(std::string_view const &tokens, std::string_view const &delims, bool const ignore_empty=true)
Tokenizes string.
fz::str_tolower_ascii
std::string str_tolower_ascii(std::string_view const &s)
tr_tolower_ascii does for strings what tolower_ascii does for individual characters
fz
The namespace used by libfilezilla.
Definition: apply.hpp:17
fz::replaced_substrings
std::string replaced_substrings(std::string const &in, std::string const &find, std::string const &replacement)
Returns in with all occurrences of find in the input string replaced with replacement.