You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

264 lines
9.1KB

  1. /* "$Id: fl_utf8.h 8585 2011-04-13 15:43:22Z ianmacarthur $"
  2. *
  3. * Author: Jean-Marc Lienher ( http://oksid.ch )
  4. * Copyright 2000-2010 by O'ksi'D.
  5. *
  6. * This library is free software; you can redistribute it and/or
  7. * modify it under the terms of the GNU Library General Public
  8. * License as published by the Free Software Foundation; either
  9. * version 2 of the License, or (at your option) any later version.
  10. *
  11. * This library is distributed in the hope that it will be useful,
  12. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  14. * Library General Public License for more details.
  15. *
  16. * You should have received a copy of the GNU Library General Public
  17. * License along with this library; if not, write to the Free Software
  18. * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
  19. * USA.
  20. *
  21. * Please report all bugs and problems on the following page:
  22. *
  23. * http://www.fltk.org/str.php
  24. */
  25. /* Merged in some functionality from the fltk-2 version. IMM.
  26. * The following code is an attempt to merge the functions incorporated in FLTK2
  27. * with the functions provided in OksiD's fltk-1.1.6-utf8 port
  28. */
  29. /**
  30. \file fl_utf8.h
  31. \brief header for Unicode and UTF8 chracter handling
  32. */
  33. #ifndef _HAVE_FL_UTF8_HDR_
  34. #define _HAVE_FL_UTF8_HDR_
  35. #include "Fl_Export.H"
  36. #include "fl_types.h"
  37. #include <stdio.h>
  38. #include <string.h>
  39. #include <stdlib.h>
  40. #ifdef WIN32
  41. # include <sys/types.h>
  42. # include <sys/stat.h>
  43. # include <locale.h>
  44. # include <ctype.h>
  45. # define xchar wchar_t
  46. # if !defined(FL_DLL) && !defined(__CYGWIN__)
  47. # undef strdup
  48. # define strdup _strdup
  49. # undef putenv
  50. # define putenv _putenv
  51. # undef stricmp
  52. # define stricmp _stricmp
  53. # undef strnicmp
  54. # define strnicmp _strnicmp
  55. # undef hypot
  56. # define hypot _hypot
  57. # undef chdir
  58. # define chdir _chdir
  59. # endif
  60. #elif defined(__APPLE__)
  61. # include <wchar.h>
  62. # include <sys/stat.h>
  63. # define xchar wchar_t
  64. #else /* X11 */
  65. # include <sys/types.h>
  66. # include <sys/stat.h>
  67. # include "Xutf8.h"
  68. # include <X11/Xlocale.h>
  69. # include <X11/Xlib.h>
  70. # include <locale.h>
  71. # define xchar unsigned short
  72. #endif
  73. # ifdef __cplusplus
  74. extern "C" {
  75. # endif
  76. /** \addtogroup fl_unicode
  77. @{
  78. */
  79. /* F2: comes from FLTK2 */
  80. /* OD: comes from OksiD */
  81. /**
  82. Return the number of bytes needed to encode the given UCS4 character in UTF8.
  83. \param [in] ucs UCS4 encoded character
  84. \return number of bytes required
  85. */
  86. FL_EXPORT int fl_utf8bytes(unsigned ucs);
  87. /* OD: returns the byte length of the first UTF-8 char sequence (returns -1 if not valid) */
  88. FL_EXPORT int fl_utf8len(char c);
  89. /* OD: returns the byte length of the first UTF-8 char sequence (returns +1 if not valid) */
  90. FL_EXPORT int fl_utf8len1(char c);
  91. /* OD: returns the number of Unicode chars in the UTF-8 string */
  92. FL_EXPORT int fl_utf_nb_char(const unsigned char *buf, int len);
  93. /* F2: Convert the next UTF8 char-sequence into a Unicode value (and say how many bytes were used) */
  94. FL_EXPORT unsigned fl_utf8decode(const char* p, const char* end, int* len);
  95. /* F2: Encode a Unicode value into a UTF8 sequence, return the number of bytes used */
  96. FL_EXPORT int fl_utf8encode(unsigned ucs, char* buf);
  97. /* F2: Move forward to the next valid UTF8 sequence start betwen start and end */
  98. FL_EXPORT const char* fl_utf8fwd(const char* p, const char* start, const char* end);
  99. /* F2: Move backward to the previous valid UTF8 sequence start */
  100. FL_EXPORT const char* fl_utf8back(const char* p, const char* start, const char* end);
  101. /* XX: Convert a single 32-bit Unicode value into UTF16 */
  102. FL_EXPORT unsigned fl_ucs_to_Utf16(const unsigned ucs, unsigned short *dst, const unsigned dstlen);
  103. /* F2: Convert a UTF8 string into UTF16 */
  104. FL_EXPORT unsigned fl_utf8toUtf16(const char* src, unsigned srclen, unsigned short* dst, unsigned dstlen);
  105. /* F2: Convert a UTF8 string into a wide character string - makes UTF16 on win32, "UCS4" elsewhere */
  106. FL_EXPORT unsigned fl_utf8towc(const char *src, unsigned srclen, wchar_t *dst, unsigned dstlen);
  107. /* F2: Convert a wide character string to UTF8 - takes in UTF16 on win32, "UCS4" elsewhere */
  108. FL_EXPORT unsigned fl_utf8fromwc(char *dst, unsigned dstlen, const wchar_t *src, unsigned srclen);
  109. /* F2: Convert a UTF8 string into ASCII, eliding untranslatable glyphs */
  110. FL_EXPORT unsigned fl_utf8toa (const char *src, unsigned srclen, char *dst, unsigned dstlen);
  111. /* OD: convert UTF-8 string to latin1 */
  112. /* FL_EXPORT int fl_utf2latin1(const unsigned char *src, int srclen, char *dst); */
  113. /* F2: Convert 8859-1 string to UTF8 */
  114. FL_EXPORT unsigned fl_utf8froma (char *dst, unsigned dstlen, const char *src, unsigned srclen);
  115. /* OD: convert latin1 str to UTF-8 */
  116. /* FL_EXPORT int fl_latin12utf(const unsigned char *src, int srclen, char *dst); */
  117. /* F2: Returns true if the current O/S locale is UTF8 */
  118. FL_EXPORT int fl_utf8locale();
  119. /* F2: Examine the first len characters of src, to determine if the input text is UTF8 or not
  120. * NOTE: The value returned is not simply boolean - it contains information about the probable
  121. * type of the src text. */
  122. FL_EXPORT int fl_utf8test(const char *src, unsigned len);
  123. /* XX: return width of "raw" ucs character in columns.
  124. * for internal use only */
  125. FL_EXPORT int fl_wcwidth_(unsigned int ucs);
  126. /* XX: return width of utf-8 character string in columns.
  127. * NOTE: this may also do C1 control character (0x80 to 0x9f) to CP1252 mapping,
  128. * depending on original build options */
  129. FL_EXPORT int fl_wcwidth(const char *src);
  130. /* OD: Return true if the character is non-spacing */
  131. FL_EXPORT unsigned int fl_nonspacing(unsigned int ucs);
  132. /* F2: Convert UTF8 to a local multi-byte encoding - mainly for win32? */
  133. FL_EXPORT unsigned fl_utf8to_mb(const char *src, unsigned srclen, char *dst, unsigned dstlen);
  134. /* OD: Convert UTF8 to a local multi-byte encoding */
  135. FL_EXPORT char* fl_utf2mbcs(const char *src);
  136. /* F2: Convert a local multi-byte encoding to UTF8 - mainly for win32? */
  137. FL_EXPORT unsigned fl_utf8from_mb(char *dst, unsigned dstlen, const char *src, unsigned srclen);
  138. /* OD: Convert a local multi-byte encoding to UTF8 */
  139. /* FL_EXPORT char* fl_mbcs2utf(const char *src); */
  140. /*****************************************************************************/
  141. #ifdef WIN32
  142. /* OD: Attempt to convert the UTF8 string to the current locale */
  143. FL_EXPORT char *fl_utf8_to_locale(const char *s, int len, unsigned int codepage);
  144. /* OD: Attempt to convert a string in the current locale to UTF8 */
  145. FL_EXPORT char *fl_locale_to_utf8(const char *s, int len, unsigned int codepage);
  146. #endif
  147. /*****************************************************************************
  148. * The following functions are intended to provide portable, UTF8 aware
  149. * versions of standard functions
  150. */
  151. /* OD: UTF8 aware strncasecmp - converts to lower case Unicode and tests */
  152. FL_EXPORT int fl_utf_strncasecmp(const char *s1, const char *s2, int n);
  153. /* OD: UTF8 aware strcasecmp - converts to Unicode and tests */
  154. FL_EXPORT int fl_utf_strcasecmp(const char *s1, const char *s2);
  155. /* OD: return the Unicode lower case value of ucs */
  156. FL_EXPORT int fl_tolower(unsigned int ucs);
  157. /* OD: return the Unicode upper case value of ucs */
  158. FL_EXPORT int fl_toupper(unsigned int ucs);
  159. /* OD: converts the UTF8 string to the lower case equivalent */
  160. FL_EXPORT int fl_utf_tolower(const unsigned char *str, int len, char *buf);
  161. /* OD: converts the UTF8 string to the upper case equivalent */
  162. FL_EXPORT int fl_utf_toupper(const unsigned char *str, int len, char *buf);
  163. /* OD: Portable UTF8 aware chmod wrapper */
  164. FL_EXPORT int fl_chmod(const char* f, int mode);
  165. /* OD: Portable UTF8 aware access wrapper */
  166. FL_EXPORT int fl_access(const char* f, int mode);
  167. /* OD: Portable UTF8 aware stat wrapper */
  168. FL_EXPORT int fl_stat( const char *path, struct stat *buffer );
  169. /* OD: Portable UTF8 aware getcwd wrapper */
  170. FL_EXPORT char* fl_getcwd( char *buf, int maxlen);
  171. /* OD: Portable UTF8 aware fopen wrapper */
  172. FL_EXPORT FILE *fl_fopen(const char *f, const char *mode);
  173. /* OD: Portable UTF8 aware system wrapper */
  174. FL_EXPORT int fl_system(const char* f);
  175. /* OD: Portable UTF8 aware execvp wrapper */
  176. FL_EXPORT int fl_execvp(const char *file, char *const *argv);
  177. /* OD: Portable UTF8 aware open wrapper */
  178. FL_EXPORT int fl_open(const char* f, int o, ...);
  179. /* OD: Portable UTF8 aware unlink wrapper */
  180. FL_EXPORT int fl_unlink(const char *f);
  181. /* OD: Portable UTF8 aware rmdir wrapper */
  182. FL_EXPORT int fl_rmdir(const char *f);
  183. /* OD: Portable UTF8 aware getenv wrapper */
  184. FL_EXPORT char* fl_getenv(const char *name);
  185. /* OD: Portable UTF8 aware execvp wrapper */
  186. FL_EXPORT int fl_mkdir(const char* f, int mode);
  187. /* OD: Portable UTF8 aware rename wrapper */
  188. FL_EXPORT int fl_rename(const char* f, const char *t);
  189. /* OD: Given a full pathname, this will create the directory path needed to hold the file named */
  190. FL_EXPORT void fl_make_path_for_file( const char *path );
  191. /* OD: recursively create a path in the file system */
  192. FL_EXPORT char fl_make_path( const char *path );
  193. /** @} */
  194. /*****************************************************************************/
  195. #ifdef __cplusplus
  196. }
  197. #endif /* __cplusplus */
  198. #endif /* _HAVE_FL_UTF8_HDR_ */
  199. /*
  200. * End of "$Id: fl_utf8.h 8585 2011-04-13 15:43:22Z ianmacarthur $".
  201. */