1/* Word breaks in strings. 2 Copyright (C) 2001-2003, 2006-2010 Free Software Foundation, Inc. 3 Written by Bruno Haible <bruno@clisp.org>, 2009. 4 5 This program is free software: you can redistribute it and/or modify it 6 under the terms of the GNU Lesser General Public License as published 7 by the Free Software Foundation; either version 3 of the License, or 8 (at your option) any later version. 9 10 This program is distributed in the hope that it will be useful, 11 but WITHOUT ANY WARRANTY; without even the implied warranty of 12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 Lesser General Public License for more details. 14 15 You should have received a copy of the GNU Lesser General Public License 16 along with this program. If not, see <http://www.gnu.org/licenses/>. */ 17 18#include <config.h> 19 20/* Specification. */ 21#include "uniwbrk.h" 22 23#include <stdlib.h> 24#include <string.h> 25 26#include "c-ctype.h" 27#include "localcharset.h" 28#include "uniconv.h" 29#include "unilbrk/ulc-common.h" 30 31/* Word breaks of a string in an arbitrary encoding. 32 33 We convert the input string to Unicode. 34 35 The standardized Unicode encodings are UTF-8, UCS-2, UCS-4, UTF-16, 36 UTF-16BE, UTF-16LE, UTF-7. UCS-2 supports only characters up to 37 \U0000FFFF. UTF-16 and variants support only characters up to 38 \U0010FFFF. UTF-7 is way too complex and not supported by glibc-2.1. 39 UCS-4 specification leaves doubts about endianness and byte order mark. 40 glibc currently interprets it as big endian without byte order mark, 41 but this is not backed by an RFC. So we use UTF-8. It supports 42 characters up to \U7FFFFFFF and is unambiguously defined. */ 43 44void 45ulc_wordbreaks (const char *s, size_t n, char *p) 46{ 47 if (n > 0) 48 { 49 const char *encoding = locale_charset (); 50 51 if (is_utf8_encoding (encoding)) 52 u8_wordbreaks ((const uint8_t *) s, n, p); 53 else 54 { 55 /* Convert the string to UTF-8 and build a translation table 56 from offsets into s to offsets into the translated string. */ 57 size_t *offsets = (size_t *) malloc (n * sizeof (size_t)); 58 59 if (offsets != NULL) 60 { 61 uint8_t *t; 62 size_t m; 63 64 t = u8_conv_from_encoding (encoding, iconveh_question_mark, 65 s, n, offsets, NULL, &m); 66 if (t != NULL) 67 { 68 char *q = (char *) (m > 0 ? malloc (m) : NULL); 69 70 if (m == 0 || q != NULL) 71 { 72 size_t i; 73 74 /* Determine the word breaks of the UTF-8 string. */ 75 u8_wordbreaks (t, m, q); 76 77 /* Translate the result back to the original string. */ 78 memset (p, 0, n); 79 for (i = 0; i < n; i++) 80 if (offsets[i] != (size_t)(-1)) 81 p[i] = q[offsets[i]]; 82 83 free (q); 84 free (t); 85 free (offsets); 86 return; 87 } 88 free (t); 89 } 90 free (offsets); 91 } 92 93 /* Impossible to convert. */ 94#if C_CTYPE_ASCII 95 if (is_all_ascii (s, n)) 96 { 97 /* ASCII is a subset of UTF-8. */ 98 u8_wordbreaks ((const uint8_t *) s, n, p); 99 return; 100 } 101#endif 102 /* We have a non-ASCII string and cannot convert it. 103 Don't produce any word breaks. */ 104 memset (p, 0, n); 105 } 106 } 107} 108 109 110#ifdef TEST 111 112#include <locale.h> 113#include <stdio.h> 114#include <stdlib.h> 115 116/* Read the contents of an input stream, and return it, terminated with a NUL 117 byte. */ 118char * 119read_file (FILE *stream) 120{ 121#define BUFSIZE 4096 122 char *buf = NULL; 123 int alloc = 0; 124 int size = 0; 125 int count; 126 127 while (! feof (stream)) 128 { 129 if (size + BUFSIZE > alloc) 130 { 131 alloc = alloc + alloc / 2; 132 if (alloc < size + BUFSIZE) 133 alloc = size + BUFSIZE; 134 buf = realloc (buf, alloc); 135 if (buf == NULL) 136 { 137 fprintf (stderr, "out of memory\n"); 138 exit (1); 139 } 140 } 141 count = fread (buf + size, 1, BUFSIZE, stream); 142 if (count == 0) 143 { 144 if (ferror (stream)) 145 { 146 perror ("fread"); 147 exit (1); 148 } 149 } 150 else 151 size += count; 152 } 153 buf = realloc (buf, size + 1); 154 if (buf == NULL) 155 { 156 fprintf (stderr, "out of memory\n"); 157 exit (1); 158 } 159 buf[size] = '\0'; 160 return buf; 161#undef BUFSIZE 162} 163 164int 165main (int argc, char * argv[]) 166{ 167 setlocale (LC_CTYPE, ""); 168 if (argc == 1) 169 { 170 /* Display all the word breaks in the input string. */ 171 char *input = read_file (stdin); 172 int length = strlen (input); 173 char *breaks = malloc (length); 174 int i; 175 176 ulc_wordbreaks (input, length, breaks); 177 178 for (i = 0; i < length; i++) 179 { 180 switch (breaks[i]) 181 { 182 case 1: 183 putc ('|', stdout); 184 break; 185 case 0: 186 break; 187 default: 188 abort (); 189 } 190 putc (input[i], stdout); 191 } 192 193 free (breaks); 194 195 return 0; 196 } 197 else 198 return 1; 199} 200 201#endif /* TEST */ 202