1 /* This file is part of the YAZ toolkit.
2 * Copyright (C) Index Data.
4 * Redistribution and use in source and binary forms, with or without
5 * modification, are permitted provided that the following conditions are met:
7 * * Redistributions of source code must retain the above copyright
8 * notice, this list of conditions and the following disclaimer.
9 * * Redistributions in binary form must reproduce the above copyright
10 * notice, this list of conditions and the following disclaimer in the
11 * documentation and/or other materials provided with the distribution.
12 * * Neither the name of Index Data nor the names of its contributors
13 * may be used to endorse or promote products derived from this
14 * software without specific prior written permission.
16 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND ANY
17 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
18 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
19 * DISCLAIMED. IN NO EVENT SHALL THE REGENTS AND CONTRIBUTORS BE LIABLE FOR ANY
20 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
21 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
22 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
23 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
25 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
36 #include <yaz/yconfig.h>
38 #include <yaz/xmltypes.h>
40 #include <unicode/utypes.h>
44 /** \brief opaque ICU chain */
45 typedef struct icu_chain *yaz_icu_chain_t;
47 /** \brief destroys ICU chain */
48 YAZ_EXPORT void icu_chain_destroy(yaz_icu_chain_t chain);
50 /** \brief constructs ICU chain from XML specification
51 \param xml_node icu_chain XML node - with attribute locale in it
52 \param sort 1 if ICU chain is to deal with sort keys; 0 otherwise
53 \param status May include ICU error code on failure
54 \returns chain ptr or NULL on failure in which case status may hold info
56 YAZ_EXPORT yaz_icu_chain_t icu_chain_xml_config(const xmlNode *xml_node,
59 /** \brief pass string to ICU for parsing/tokenization/etc
60 \param chain ICU chain to be used for parsing
61 \param src8cstr input C string (null-terminated)
62 \param status may include ICU error on failure
66 YAZ_EXPORT int icu_chain_assign_cstr(yaz_icu_chain_t chain,
70 /** \brief returns one token (if any)
71 \param chain ICU chain
72 \param status may include ICU error on failure
73 \retval 0 error or end-of-tokens (no more tokens)
74 \retval >0 token number (1, 3, 3, ..)
76 This function tries to move to "next" token in assigned
77 C-string .. Or returns 0 if no more is to be found
79 YAZ_EXPORT int icu_chain_next_token(yaz_icu_chain_t chain,
82 /** \brief returns token number of last token processed
83 \brief chain ICU chain
84 \returns token number (numbered from 1)
86 YAZ_EXPORT int icu_chain_token_number(yaz_icu_chain_t chain);
88 /** \brief returns display token of last token processed
89 \param chain ICU chain
90 \returns display token string (C string)
91 This function returns display string for last token returned
92 by icu_chain_next_token.
94 YAZ_EXPORT const char * icu_chain_token_display(yaz_icu_chain_t chain);
96 /** \brief returns normalized token of last token processed
97 \param chain ICU chain
98 \returns normalized token string (C string)
99 This function returns normalized string for last token returned
100 by icu_chain_next_token.
102 YAZ_EXPORT const char * icu_chain_token_norm(yaz_icu_chain_t chain);
104 /** \brief returns sortkey token of last token processed
105 \param chain ICU chain
106 \returns sortkey token string (C string)
107 This function returns sortkey string for last token returned
108 by icu_chain_next_token.
110 YAZ_EXPORT const char * icu_chain_token_sortkey(yaz_icu_chain_t chain);
112 /** \brief returns token as it relates to original text (legacy)
113 \param chain ICU chain
114 \param start offset in original text
115 \param len number of uchars in original text
117 YAZ_EXPORT void icu_chain_get_org_info(yaz_icu_chain_t chain,
118 size_t *start, size_t *len);
120 /** \brief returns token as it relates to original text (2nd version)
121 \param chain ICU chain
122 \param start offset in original text
123 \param len number of uchars in original text
124 \param cstr if not-null, holds original string in there
126 YAZ_EXPORT void icu_chain_get_org_info2(yaz_icu_chain_t chain,
127 size_t *start, size_t *len,
130 /** \brief ICU tokenizer iterator type (opaque) */
131 typedef struct icu_iter *yaz_icu_iter_t;
133 /** \brief create ICU tokenizer iterator from chain
134 \param chain ICU chain
135 \returns ICU iterator
138 yaz_icu_iter_t icu_iter_create(struct icu_chain *chain);
140 /** \brief starts iteration over string
141 \param iter ICU tokenizer iterator
142 \param src8cstr input string (0-terminated)
144 Call icu_iter_next to iterate over each token.
147 void icu_iter_first(yaz_icu_iter_t iter, const char *src8cstr);
149 /** \brief iterates over one token
150 \param iter ICU tokenizer iterator
151 \retval 0 no more tokens (EOF)
152 \retval 1 got one token (use icu_iter_get..-functions)
155 int icu_iter_next(yaz_icu_iter_t iter);
157 /** \brief destroy ICU tokenizer iterator
158 \param iter ICU tokenizer iterator
161 void icu_iter_destroy(yaz_icu_iter_t iter);
163 /** \brief returns ICU normalized token
164 \param iter ICU tokenizer iterator
165 \returns string (0-terminated)
168 const char *icu_iter_get_norm(yaz_icu_iter_t iter);
170 /** \brief returns ICU sortkey string
171 \param iter ICU tokenizer iterator
172 \returns string (0-terminated)
175 const char *icu_iter_get_sortkey(yaz_icu_iter_t iter);
177 /** \brief returns ICU display string
178 \param iter ICU tokenizer iterator
179 \returns string (0-terminated)
182 const char *icu_iter_get_display(yaz_icu_iter_t iter);
184 /** \brief returns ICU token count for iterator
185 \param iter ICU tokenizer iterator
186 \returns token count (1, 2, 3...)
189 int icu_iter_get_token_number(yaz_icu_iter_t iter);
191 /** \brief returns ICU original token start (offset) and length (legacy)
192 \param iter ICU tokenizer iterator
193 \param start offset of last token in original text
194 \param len length of last token in original text
197 void icu_iter_get_org_info(yaz_icu_iter_t iter, size_t *start, size_t *len);
199 /** \brief returns ICU original token start (offset) and length
200 \param iter ICU tokenizer iterator
201 \param start offset of last token in original text
202 \param len length of last token in original text
203 \param cstr if non-null: original string
206 void icu_iter_get_org_info2(yaz_icu_iter_t iter, size_t *start, size_t *len,
211 #endif /* YAZ_ICU_H */
216 * c-file-style: "Stroustrup"
217 * indent-tabs-mode: nil
219 * vim: shiftwidth=4 tabstop=8 expandtab