1 /* $Id: icu_I18N.h,v 1.15 2007-05-20 19:00:17 marc Exp $
2 Copyright (c) 2006-2007, Index Data.
4 This file is part of Pazpar2.
6 Pazpar2 is free software; you can redistribute it and/or modify it under
7 the terms of the GNU General Public License as published by the Free
8 Software Foundation; either version 2, or (at your option) any later
11 Pazpar2 is distributed in the hope that it will be useful, but WITHOUT ANY
12 WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
16 You should have received a copy of the GNU General Public License
17 along with Pazpar2; see the file LICENSE. If not, write to the
18 Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
30 #include <unicode/utypes.h> /* Basic ICU data types */
31 #include <unicode/uchar.h> /* char names */
33 //#include <unicode/ustdio.h>
34 #include <unicode/ucol.h>
35 //#include <unicode/ucnv.h> /* C Converter API */
36 //#include <unicode/ustring.h> /* some more string fcns*/
37 //#include <unicode/uloc.h>
38 #include <unicode/ubrk.h>
39 //#include <unicode/unistr.h>
40 #include <unicode/utrans.h>
44 // declared structs and functions
46 int icu_check_status (UErrorCode status);
55 struct icu_buf_utf16 * icu_buf_utf16_create(size_t capacity);
56 struct icu_buf_utf16 * icu_buf_utf16_resize(struct icu_buf_utf16 * buf16,
58 struct icu_buf_utf16 * icu_buf_utf16_copy(struct icu_buf_utf16 * dest16,
59 struct icu_buf_utf16 * src16);
60 void icu_buf_utf16_destroy(struct icu_buf_utf16 * buf16);
71 struct icu_buf_utf8 * icu_buf_utf8_create(size_t capacity);
72 struct icu_buf_utf8 * icu_buf_utf8_resize(struct icu_buf_utf8 * buf8,
74 void icu_buf_utf8_destroy(struct icu_buf_utf8 * buf8);
77 UErrorCode icu_utf16_from_utf8(struct icu_buf_utf16 * dest16,
78 struct icu_buf_utf8 * src8,
81 UErrorCode icu_utf16_from_utf8_cstr(struct icu_buf_utf16 * dest16,
82 const char * src8cstr,
86 UErrorCode icu_utf16_to_utf8(struct icu_buf_utf8 * dest8,
87 struct icu_buf_utf16 * src16,
96 struct icu_casemap * icu_casemap_create(const char *locale, char action,
99 void icu_casemap_destroy(struct icu_casemap * casemap);
101 int icu_casemap_casemap(struct icu_casemap * casemap,
102 struct icu_buf_utf16 * dest16,
103 struct icu_buf_utf16 * src16,
106 int icu_utf16_casemap(struct icu_buf_utf16 * dest16,
107 struct icu_buf_utf16 * src16,
108 const char *locale, char action,
111 UErrorCode icu_sortkey8_from_utf16(UCollator *coll,
112 struct icu_buf_utf8 * dest8,
113 struct icu_buf_utf16 * src16,
114 UErrorCode * status);
121 struct icu_buf_utf16 * buf16;
126 // keep always invariant
129 // <= buf16->utf16_len
131 // 0 <= token_id <= token_count
134 struct icu_tokenizer * icu_tokenizer_create(const char *locale, char action,
137 void icu_tokenizer_destroy(struct icu_tokenizer * tokenizer);
139 int icu_tokenizer_attach(struct icu_tokenizer * tokenizer,
140 struct icu_buf_utf16 * src16, UErrorCode *status);
142 int32_t icu_tokenizer_next_token(struct icu_tokenizer * tokenizer,
143 struct icu_buf_utf16 * tkn16,
146 int32_t icu_tokenizer_token_id(struct icu_tokenizer * tokenizer);
147 int32_t icu_tokenizer_token_start(struct icu_tokenizer * tokenizer);
148 int32_t icu_tokenizer_token_end(struct icu_tokenizer * tokenizer);
149 int32_t icu_tokenizer_token_length(struct icu_tokenizer * tokenizer);
150 int32_t icu_tokenizer_token_count(struct icu_tokenizer * tokenizer);
154 struct icu_normalizer
157 struct icu_buf_utf16 * rules16;
158 UParseError parse_error[256];
159 UTransliterator * trans;
162 struct icu_normalizer * icu_normalizer_create(const char *rules, char action,
166 void icu_normalizer_destroy(struct icu_normalizer * normalizer);
168 int icu_normalizer_normalize(struct icu_normalizer * normalizer,
169 struct icu_buf_utf16 * dest16,
170 struct icu_buf_utf16 * src16,
185 enum icu_chain_step_type {
186 ICU_chain_step_type_none, //
187 ICU_chain_step_type_display, // convert to utf8 display format
188 ICU_chain_step_type_norm, // convert to utf8 norm format
189 ICU_chain_step_type_sort, // convert to utf8 sort format
190 ICU_chain_step_type_casemap, // apply utf16 charmap
191 ICU_chain_step_type_normalize, // apply utf16 normalization
192 ICU_chain_step_type_tokenize // apply utf16 tokenization
197 struct icu_chain_step
199 // type and action object
200 enum icu_chain_step_type type;
202 struct icu_casemap * casemap;
203 struct icu_normalizer * normalizer;
204 struct icu_tokenizer * tokenizer;
206 // temprary post-action utf16 buffer
207 struct icu_buf_utf16 * buf16;
208 struct icu_chain_step * previous;
216 struct icu_chain_step * icu_chain_step_create(struct icu_chain * chain,
217 enum icu_chain_step_type type,
218 const uint8_t * rule,
219 struct icu_buf_utf16 * buf16,
223 void icu_chain_step_destroy(struct icu_chain_step * step);
228 uint8_t identifier[128];
231 // number of tokens returned so far
234 // utf8 output buffers
235 struct icu_buf_utf8 * display8;
236 struct icu_buf_utf8 * norm8;
237 struct icu_buf_utf8 * sort8;
239 // utf16 source buffer
240 struct icu_buf_utf16 * src16;
242 // linked list of chain steps
243 struct icu_chain_step * steps;
246 struct icu_chain * icu_chain_create(const uint8_t * identifier,
247 const uint8_t * locale);
248 void icu_chain_destroy(struct icu_chain * chain);
250 struct icu_chain_step * icu_chain_insert_step(struct icu_chain * chain,
251 enum icu_chain_step_type type,
252 const uint8_t * rule,
256 int icu_chain_step_next_token(struct icu_chain * chain,
257 struct icu_chain_step * step,
260 int icu_chain_assign_cstr(struct icu_chain * chain,
261 const char * src8cstr,
264 int icu_chain_next_token(struct icu_chain * chain,
267 int icu_chain_get_token_count(struct icu_chain * chain);
269 const char * icu_chain_get_display(struct icu_chain * chain);
271 const char * icu_chain_get_norm(struct icu_chain * chain);
273 const char * icu_chain_get_sort(struct icu_chain * chain);
280 #endif // ICU_I18NL_H