1 /* This file is part of the YAZ toolkit.
2 * Copyright (C) 1995-2009 Index Data
3 * See the file LICENSE for details.
8 * \brief UTF-8 string utilities for ICU
16 #include <yaz/xmalloc.h>
18 #include <yaz/icu_I18N.h>
26 #include <unicode/ustring.h> /* some more string fcns*/
27 #include <unicode/uchar.h> /* char names */
29 struct icu_buf_utf8 *icu_buf_utf8_create(size_t capacity)
31 struct icu_buf_utf8 * buf8
32 = (struct icu_buf_utf8 *) xmalloc(sizeof(struct icu_buf_utf8));
40 buf8->utf8 = (uint8_t *) xmalloc(sizeof(uint8_t) * capacity);
41 buf8->utf8[0] = (uint8_t) 0;
42 buf8->utf8_cap = capacity;
47 struct icu_buf_utf8 * icu_buf_utf8_clear(struct icu_buf_utf8 * buf8)
52 buf8->utf8[0] = (uint8_t) 0;
58 struct icu_buf_utf8 * icu_buf_utf8_resize(struct icu_buf_utf8 * buf8,
67 buf8->utf8 = (uint8_t *) xmalloc(sizeof(uint8_t) * capacity);
70 = (uint8_t *) xrealloc(buf8->utf8, sizeof(uint8_t) * capacity);
72 buf8->utf8_cap = capacity;
84 const char *icu_buf_utf8_to_cstr(struct icu_buf_utf8 *src8)
86 if (!src8 || src8->utf8_len == 0)
89 if (src8->utf8_len == src8->utf8_cap)
90 src8 = icu_buf_utf8_resize(src8, src8->utf8_len * 2 + 1);
92 src8->utf8[src8->utf8_len] = '\0';
94 return (const char *) src8->utf8;
97 void icu_buf_utf8_destroy(struct icu_buf_utf8 * buf8)
104 UErrorCode icu_utf16_from_utf8_cstr(struct icu_buf_utf16 * dest16,
105 const char * src8cstr,
108 size_t src8cstr_len = 0;
109 int32_t utf16_len = 0;
111 *status = U_ZERO_ERROR;
112 src8cstr_len = strlen(src8cstr);
114 u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
116 src8cstr, src8cstr_len, status);
118 /* check for buffer overflow, resize and retry */
119 if (*status == U_BUFFER_OVERFLOW_ERROR)
121 icu_buf_utf16_resize(dest16, utf16_len * 2);
122 *status = U_ZERO_ERROR;
123 u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
125 src8cstr, src8cstr_len, status);
128 if (U_SUCCESS(*status)
129 && utf16_len <= dest16->utf16_cap)
130 dest16->utf16_len = utf16_len;
132 icu_buf_utf16_clear(dest16);
137 UErrorCode icu_utf16_to_utf8(struct icu_buf_utf8 *dest8,
138 const struct icu_buf_utf16 *src16,
141 int32_t utf8_len = 0;
143 u_strToUTF8((char *) dest8->utf8, dest8->utf8_cap,
145 src16->utf16, src16->utf16_len, status);
147 /* check for buffer overflow, resize and retry */
148 if (*status == U_BUFFER_OVERFLOW_ERROR)
150 icu_buf_utf8_resize(dest8, utf8_len * 2);
151 *status = U_ZERO_ERROR;
152 u_strToUTF8((char *) dest8->utf8, dest8->utf8_cap,
154 src16->utf16, src16->utf16_len, status);
157 if (U_SUCCESS(*status)
158 && utf8_len <= dest8->utf8_cap)
159 dest8->utf8_len = utf8_len;
161 icu_buf_utf8_clear(dest8);
166 #endif /* YAZ_HAVE_ICU */
171 * c-file-style: "Stroustrup"
172 * indent-tabs-mode: nil
174 * vim: shiftwidth=4 tabstop=8 expandtab