source/texk/web2c/luatexdir/unilib/ustring.h

   1 /* Copyright (C) 2000-2012 by George Williams */
   2 /*
   3  * Redistribution and use in source and binary forms, with or without
   4  * modification, are permitted provided that the following conditions are met:
   5
   6  * Redistributions of source code must retain the above copyright notice, this
   7  * list of conditions and the following disclaimer.
   8
   9  * Redistributions in binary form must reproduce the above copyright notice,
  10  * this list of conditions and the following disclaimer in the documentation
  11  * and/or other materials provided with the distribution.
  12
  13  * The name of the author may not be used to endorse or promote products
  14  * derived from this software without specific prior written permission.
  15
  16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
  17  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
  18  * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
  19  * EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  20  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  21  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
  22  * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
  23  * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
  24  * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
  25  * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  26  */
  27 #ifndef _UCHAR_H
  28 # define _UCHAR_H
  29 #include <unibasics.h>
  30 #include <stdarg.h>
  31 #include <string.h>
  32 #include <memory.h>
  33 #include <charset.h>
  34
  35 extern unichar_t *u_copy(const unichar_t*);
  36 extern unichar_t *u_copyn(const unichar_t*, long);
  37 extern unichar_t *u_copynallocm(const unichar_t *pt, long n, long m);
  38 extern unichar_t *uc_copyn(const char *, int);
  39 extern unichar_t *uc_copy(const char*);
  40 extern unichar_t *u_concat(const unichar_t*,const unichar_t*);
  41 extern char      *cu_copyn(const unichar_t *pt,int len);
  42 extern char      *cu_copy(const unichar_t*);
  43
  44 extern long uc_strcmp(const unichar_t *,const char *);
  45 extern long u_strcmp(const unichar_t *, const unichar_t *);
  46 extern long uc_strncmp(const unichar_t *,const char *,int);
  47 extern long u_strncmp(const unichar_t *, const unichar_t *,int);
  48 extern long uc_strmatch(const unichar_t *,const char *);
  49 extern long uc_strnmatch(const unichar_t *,const char *,int);
  50 extern long u_strnmatch(const unichar_t *str1, const unichar_t *str2, int len);
  51 extern long u_strmatch(const unichar_t *, const unichar_t *);
  52 extern int    strmatch(const char *,const char *);
  53 extern int    strnmatch(const char *str1, const char *str2, int n);
  54 extern void uc_strcpy(unichar_t *, const char *);
  55 extern void cu_strcpy(char *, const unichar_t *);
  56 extern void u_strcpy(unichar_t *, const unichar_t *);
  57 extern void u_strncpy(unichar_t *, const unichar_t *,int);
  58 extern void cu_strncpy(char *to, const unichar_t *from, int len);
  59 extern void uc_strncpy(unichar_t *to, const char *from, int len);
  60 extern void uc_strcat(unichar_t *, const char *);
  61 extern void uc_strncat(unichar_t *, const char *,int len);
  62 extern void cu_strcat(char *, const unichar_t *);
  63 extern void cu_strncat(char *, const unichar_t *,int len);
  64 extern void u_strcat(unichar_t *, const unichar_t *);
  65 extern void u_strncat(unichar_t *, const unichar_t *, int len);
  66 extern int  u_strlen(const unichar_t *);
  67 extern unichar_t *u_strchr(const unichar_t *,unichar_t);
  68 extern unichar_t *u_strrchr(const unichar_t *,unichar_t);
  69 extern unichar_t *uc_strstr(const unichar_t *,const char *);
  70 extern unichar_t *u_strstr(const unichar_t *,const unichar_t *);
  71 extern unichar_t *uc_strstrmatch(const unichar_t *,const char *);
  72 extern unichar_t *u_strstrmatch(const unichar_t *,const unichar_t *);
  73 extern char      *  strstrmatch(const char *,const char *);
  74
  75 extern char *u_to_c(const unichar_t *);
  76 extern unichar_t *c_to_u(const char *);
  77
  78 extern unsigned long u_strtoul(const unichar_t *,unichar_t **,int);
  79 extern long   u_strtol(const unichar_t *,unichar_t **,int);
  80 extern double u_strtod(const unichar_t *,unichar_t **);
  81
  82 extern char *strstart(const char *initial,const char *full);
  83 extern char *strstartmatch(const char *initial,const char *full);
  84 extern unichar_t *u_strstartmatch(const unichar_t *initial, const unichar_t *full);
  85 extern unichar_t *cu_strstartmatch(const char *initial, const unichar_t *full);
  86
  87 #define utf82u_strncpy utf82U_strncpy
  88 extern int32 utf8_ildb(const char **utf8_text);
  89 extern char *utf8_idpb(char *utf8_text,uint32 ch);
  90 extern char *utf8_db(char *utf8_text);
  91 extern char *utf8_ib(char *utf8_text);
  92 extern int utf8_valid(const char *str);
  93 extern void utf8_truncatevalid(char *str);
  94 extern char *latin1_2_utf8_strcpy(char *utf8buf,const char *lbuf);
  95 extern char *latin1_2_utf8_copy(const char *lbuf);
  96 extern char *utf8_2_latin1_copy(const char *utf8buf);
  97 extern int utf8_strlen(const char *utf8_str); /* how many characters in the string */
  98 extern int utf82u_strlen(const char *utf8_str); /* how many long would this be in shorts (UCS2) */
  99 extern void utf8_strncpy(register char *to, const char *from, int len); /* copy n characters NOT bytes */
 100 extern char *def2utf8_copy(const char *from);
 101 extern char *utf82def_copy(const char *ufrom);
 102 extern char *utf8_strchr(const char *utf8_str, int search_char);
 103
 104 extern unichar_t *utf82u_strncpy(unichar_t *ubuf,const char *utf8buf,int len);
 105 extern unichar_t *utf82u_strcpy(unichar_t *ubuf,const char *utf8buf);
 106 extern void       utf82u_strcat(unichar_t *ubuf,const char *utf8buf);
 107 extern unichar_t *utf82u_copyn(const char *utf8buf,int len);
 108 extern unichar_t *utf82u_copy(const char *utf8buf);
 109 extern char *u2utf8_strcpy(char *utf8buf,const unichar_t *ubuf);
 110 extern char *u2utf8_copy(const unichar_t *ubuf);
 111 extern char *u2utf8_copyn(const unichar_t *ubuf,int len);
 112 extern unichar_t *encoding2u_strncpy(unichar_t *uto, const char *from, int n, enum encoding cs);
 113 extern char *u2encoding_strncpy(char *to, const unichar_t *ufrom, int n, enum encoding cs);
 114 extern unichar_t *def2u_strncpy(unichar_t *uto, const char *from, int n);
 115 extern char *u2def_strncpy(char *to, const unichar_t *ufrom, int n);
 116 extern unichar_t *def2u_copy(const char *from);
 117 extern char *u2def_copy(const unichar_t *ufrom);
 118
 119 extern int u_sprintf(unichar_t *str, const unichar_t *format, ... );
 120 extern int u_snprintf(unichar_t *str, int len, const unichar_t *format, ... );
 121 extern int u_vsnprintf(unichar_t *str, int len, const unichar_t *format, va_list ap );
 122
 123 extern int uAllAscii(const unichar_t *str);
 124 extern int AllAscii(const char *);
 125 extern char *StripToASCII(const char *utf8_str);
 126
 127 extern char *copytolower(const char *);
 128 extern int endswith(const char *haystack,const char *needle);
 129 extern int endswithi(const char *haystack,const char *needle);
 130 extern int endswithi_partialExtension( const char *haystack,const char *needle);
 131
 132 /**
 133  * Remove trailing \n or \r from the given string. No memory
 134  * allocations are performed, null is injected over these terminators
 135  * to trim the string.
 136  *
 137  * This function is designed to be impotent if called with a string
 138  * that does not end with \n or \r. ie, you don't need to redundantly
 139  * check if there is a newline at the end of string and not call here
 140  * if there is no newline. You can just call here with any string and
 141  * be assured that afterwards there will be no trailing newline or
 142  * carrage return character found at the end of the string pointed to
 143  * by 'p'.
 144  */
 145 extern char* chomp( char* p );
 146
 147 /**
 148  * Return true if the haystack plain string ends with the string
 149  * needle. Return 0 otherwise.
 150  *
 151  * Needles which are larger than the haystack are handled.
 152  *
 153  * No new strings are allocated, freed, or returned.
 154  */
 155 int endswith(const char *haystack,const char *needle);
 156
 157 /**
 158  * Return true if the haystack unicode string ends with the string needle.
 159  * Return 0 otherwise.
 160  *
 161  * Needles which are larger than the haystack are handled.
 162  *
 163  * No new strings are allocated, freed, or returned.
 164  */
 165 extern int u_endswith(const unichar_t *haystack,const unichar_t *needle);
 166
 167 #endif