1 /* $Id: api.h,v 1.51 2007-05-21 11:54:59 adam Exp $
2 Copyright (C) 1995-2007
5 This file is part of the Zebra server.
7 Zebra is free software; you can redistribute it and/or modify it under
8 the terms of the GNU General Public License as published by the Free
9 Software Foundation; either version 2, or (at your option) any later
12 Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
13 WARRANTY; without even the implied warranty of MERCHANTABILITY or
14 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
17 You should have received a copy of the GNU General Public License
18 along with this program; if not, write to the Free Software
19 Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
27 Most functions has return type ZEBRA_RES, where ZEBRA_FAIL indicates
28 failure; ZEBRA_OK indicates success.
35 #include <yaz/proto.h>
36 #include <idzebra/res.h>
37 #include <idzebra/version.h>
38 #include <idzebra/recctrl.h>
49 } ZebraTransactionStatus;
51 /** Retrieval Record Descriptor */
53 int errCode; /* non-zero if error when fetching this */
54 char *errString; /* error string */
55 int position; /* position of record in result set (1,2,..) */
56 char *buf; /* record buffer (void pointer really) */
58 const Odr_oid *format; /* record syntax */
62 } ZebraRetrievalRecord;
64 /** Scan Term Descriptor */
66 zint occurrences; /* scan term occurrences */
67 char *term; /* scan term string */
71 \brief a Zebra Handle - (session)
73 typedef struct zebra_session *ZebraHandle;
76 \brief a Zebra Service handle
78 typedef struct zebra_service *ZebraService;
80 /** \brief Creates a Zebra Service.
81 \param configName name of configuration file
83 This function is a simplified version of zebra_start_res.
86 ZebraService zebra_start(const char *configName
87 ) ZEBRA_GCC_ATTR((warn_unused_result));
89 /** \brief Creates a Zebra service with resources.
90 \param configName name of configuration file
91 \param def_res default resources
92 \param over_res overriding resources
94 This function typically called once in a program. A Zebra Service
95 acts as a factory for Zebra session handles.
98 ZebraService zebra_start_res(const char *configName,
99 Res def_res, Res over_res
100 ) ZEBRA_GCC_ATTR((warn_unused_result));
102 /** \brief stops a Zebra service.
103 \param zs service handle
105 Frees resources used by the service.
108 ZEBRA_RES zebra_stop(ZebraService zs);
110 /** \brief Lists enabled Zebra filters
111 \param zs service handle
112 \param cd callback parameter (opaque)
113 \param cb callback function
116 void zebra_filter_info(ZebraService zs, void *cd,
117 void (*cb)(void *cd, const char *name));
120 /** \brief Creates a Zebra session handle within service.
121 \param zs service handle.
122 \param res resources to be used for the service (NULL for none)
124 There should be one handle for each thread doing something
125 with zebra, be that searching or indexing. In simple apps
126 one handle is sufficient
129 ZebraHandle zebra_open(ZebraService zs, Res res
130 ) ZEBRA_GCC_ATTR((warn_unused_result));
132 /** \brief Destroys Zebra session handle.
133 \param zh zebra session handle.
136 ZEBRA_RES zebra_close(ZebraHandle zh);
138 /** \brief Returns error code for last error
139 \param zh zebra session handle.
142 int zebra_errCode(ZebraHandle zh);
144 /** \brief Returns error string for last error
145 \param zh zebra session handle.
148 const char *zebra_errString(ZebraHandle zh);
150 /** \brief Returns additional info for last error
151 \param zh zebra session handle.
154 char *zebra_errAdd(ZebraHandle zh);
156 /** \brief Returns error code and additional info for last error
157 \param zh zebra session handle.
158 \param code pointer to returned error code
159 \param addinfo pointer to returned additional info
162 void zebra_result(ZebraHandle zh, int *code, char **addinfo);
165 /** \brief Returns character set encoding for session
166 \param zh zebra session handle.
167 \returns encoding name (e.g. "iso-8859-1")
170 const char *zebra_get_encoding(ZebraHandle zh);
172 /** \brief Set limit before Zebra does approx hit count
173 \param zh session handle
174 \param approx_limit the limit
176 Results will be approximiate if hit count is greater than the
177 limit specified. By default there is a high-limit (no limit).
179 ZEBRA_RES zebra_set_approx_limit(ZebraHandle zh, zint approx_limit);
181 /** \brief Search using PQF Query String
182 \param zh session handle
183 \param pqf_query query
184 \param setname name of resultset
185 \param hits of hits is returned
188 ZEBRA_RES zebra_search_PQF(ZebraHandle zh, const char *pqf_query,
189 const char *setname, zint *hits);
191 /** \brief Search using RPN Query structure (from ASN.1)
192 \param zh session handle
194 \param query RPN query using YAZ structure
195 \param setname name of resultset
196 \param hits number of hits is returned
197 \param estimated_hit_count whether hit count is an estimate
198 \param partial_resultset whether result is only partially evaluated
201 ZEBRA_RES zebra_search_RPN_x(ZebraHandle zh, ODR o, Z_RPNQuery *query,
202 const char *setname, zint *hits,
203 int *estimated_hit_count,
204 int *partial_resultset);
207 /** \brief Search using RPN Query structure (from ASN.1)
208 \param zh session handle
210 \param query RPN query using YAZ structure
211 \param setname name of resultset
212 \param hits number of hits is returned
215 ZEBRA_RES zebra_search_RPN(ZebraHandle zh, ODR o, Z_RPNQuery *query,
216 const char *setname, zint *hits);
218 /** \brief Retrieve records from result set (after search)
219 \param zh session handle
220 \param stream allocate records returned using this ODR
221 \param setname name of result set to retrieve records from
222 \param comp Z39.50 record composition
223 \param input_format transfer syntax (OID)
224 \param num_recs number of records to retrieve
225 \param recs store records in this structure (size is num_recs)
228 ZEBRA_RES zebra_records_retrieve(ZebraHandle zh, ODR stream,
230 Z_RecordComposition *comp,
231 const Odr_oid *input_format,
233 ZebraRetrievalRecord *recs);
234 /** \brief Deletes one or more resultsets
235 \param zh session handle
236 \param function Z_DeleteResultSetRequest_{list,all}
237 \param num_setnames number of result sets
238 \param setnames result set names
239 \param statuses status result
242 int zebra_deleteResultSet(ZebraHandle zh, int function,
243 int num_setnames, char **setnames,
247 /** \brief returns number of term info terms assocaited with result set
248 \param zh session handle
249 \param setname result set name
250 \param num_terms number of terms returned in this integer
252 This function is used in conjunction with zebra_result_set_term_info.
253 If operation was successful, ZEBRA_OK is returned; otherwise
254 ZEBRA_FAIL is returned (typically non-existing setname)
257 ZEBRA_RES zebra_result_set_term_no(ZebraHandle zh, const char *setname,
260 /** \brief returns information about a term assocated with a result set
261 \param zh session handle
262 \param setname result set name
263 \param no the term we want to know about (0=first, 1=second,..)
264 \param count the number of occurrences of this term, aka hits (output)
265 \param approx about hits: 0=exact,1=approx (output)
266 \param termbuf buffer for term string (intput, output)
267 \param termlen size of termbuf (input=max, output=actual length)
268 \param term_ref_id if non-NULL *term_ref_id holds term reference
270 Returns information about one search term associated with result set.
271 Use zebra_result_set_term_no to read total number of terms associated
272 with result set. If this function can not return information,
273 due to no out of range or bad result set name, ZEBRA_FAIL is
275 The passed termbuf must be able to hold at least *termlen characters.
276 Upon completion, *termlen holds actual length of search term.
279 ZEBRA_RES zebra_result_set_term_info(ZebraHandle zh, const char *setname,
280 int no, zint *count, int *approx,
281 char *termbuf, size_t *termlen,
282 const char **term_ref_id);
285 /** \brief performs Scan (Z39.50 style)
286 \param zh session handle
287 \param stream ODR handle for result
288 \param zapt Attribute plus Term (start term)
289 \param attributeset Attributeset for Attribute plus Term
290 \param position input/output position
291 \param num_entries number of terms requested / returned
292 \param entries list of resulting terms (ODR allocated)
293 \param is_partial upon return 1=partial, 0=complete
294 \param setname limit scan by this set (NULL means no limit)
296 YAZ_EXPORT ZEBRA_RES zebra_scan(ZebraHandle zh, ODR stream,
297 Z_AttributesPlusTerm *zapt,
298 const Odr_oid *attributeset,
299 int *position, int *num_entries,
300 ZebraScanEntry **entries,
302 const char *setname);
304 /** \brief performs Scan (taking PQF string)
305 \param zh session handle
306 \param stream ODR handle for result
307 \param query PQF scan query
308 \param position input/output position
309 \param num_entries number of terms requested / returned
310 \param entries list of resulting terms (ODR allocated)
311 \param is_partial upon return 1=partial, 0=complete
312 \param setname limit scan by this set (NULL means no limit)
315 ZEBRA_RES zebra_scan_PQF(ZebraHandle zh, ODR stream, const char *query,
316 int *position, int *num_entries,
317 ZebraScanEntry **entries,
318 int *is_partial, const char *setname);
320 /** \brief authenticate user. Returns 0 if OK, != 0 on failure
321 \param zh session handle
322 \param user user name
326 ZEBRA_RES zebra_auth(ZebraHandle zh, const char *user, const char *pass);
328 /** \brief Normalize zebra term for register (subject to change!)
329 \param zh session handle
330 \param reg_id register ID, 'w', 'p',..
331 \param input_str input string buffer
332 \param input_len input string length
333 \param output_str output string buffer
334 \param output_len output string length
337 int zebra_string_norm(ZebraHandle zh, unsigned reg_id, const char *input_str,
338 int input_len, char *output_str, int output_len);
340 /** \brief Creates a database
341 \param zh session handle
342 \param db database to be created
345 ZEBRA_RES zebra_create_database(ZebraHandle zh, const char *db);
347 /** \brief Deletes a database (drop)
348 \param zh session handle
349 \param db database to be deleted
352 ZEBRA_RES zebra_drop_database(ZebraHandle zh, const char *db);
355 ZEBRA_RES zebra_admin_shutdown(ZebraHandle zh);
358 ZEBRA_RES zebra_admin_start(ZebraHandle zh);
361 ZEBRA_RES zebra_shutdown(ZebraService zs);
364 ZEBRA_RES zebra_admin_import_begin(ZebraHandle zh, const char *database,
365 const char *record_type);
368 ZEBRA_RES zebra_admin_import_segment(ZebraHandle zh,
372 ZEBRA_RES zebra_admin_import_end(ZebraHandle zh);
375 ZEBRA_RES zebra_begin_trans(ZebraHandle zh, int rw
376 ) ZEBRA_GCC_ATTR((warn_unused_result));
379 ZEBRA_RES zebra_end_trans(ZebraHandle zh
380 ) ZEBRA_GCC_ATTR((warn_unused_result));
383 ZEBRA_RES zebra_end_transaction(ZebraHandle zh,
384 ZebraTransactionStatus *stat);
387 ZEBRA_RES zebra_commit(ZebraHandle zh);
390 ZEBRA_RES zebra_clean(ZebraHandle zh);
393 ZEBRA_RES zebra_init(ZebraHandle zh);
396 ZEBRA_RES zebra_compact(ZebraHandle zh);
399 ZEBRA_RES zebra_repository_update(ZebraHandle zh, const char *path);
402 ZEBRA_RES zebra_repository_delete(ZebraHandle zh, const char *path);
405 ZEBRA_RES zebra_repository_show(ZebraHandle zh, const char *path);
407 /** \brief Simple update record
408 \param zh session handle
409 \param buf record buffer
410 \param buf_size record buffer size
412 This function is a simple wrapper or zebra_update_record with
413 action=action_update (insert or replace) .
416 ZEBRA_RES zebra_add_record(ZebraHandle zh, const char *buf, int buf_size);
418 /** \brief Updates record
419 \param zh session handle
420 \param action (insert,replace,delete or update (replace/insert)
421 \param recordType filter type (0 indicates default)
422 \param sysno system id (0 may be passed for no known id)
423 \param match match criteria (0 may be passed for no known criteria)
424 \param fname filename to be printed for logging (0 may be passed)
425 \param buf record buffer
426 \param buf_size record buffer size
429 ZEBRA_RES zebra_update_record(ZebraHandle zh,
430 enum zebra_recctrl_action_t action,
431 const char *recordType,
432 zint *sysno, const char *match,
434 const char *buf, int buf_size);
437 ZEBRA_RES zebra_sort(ZebraHandle zh, ODR stream,
438 int num_input_setnames,
439 const char **input_setnames,
440 const char *output_setname,
441 Z_SortKeySpecList *sort_sequence,
443 ) ZEBRA_GCC_ATTR((warn_unused_result));
446 ZEBRA_RES zebra_select_databases(ZebraHandle zh, int num_bases,
447 const char **basenames
448 ) ZEBRA_GCC_ATTR((warn_unused_result));
451 ZEBRA_RES zebra_select_database(ZebraHandle zh, const char *basename
452 ) ZEBRA_GCC_ATTR((warn_unused_result));
455 void zebra_shadow_enable(ZebraHandle zh, int value);
458 int zebra_register_statistics(ZebraHandle zh, int dumpdict);
461 ZEBRA_RES zebra_record_encoding(ZebraHandle zh, const char *encoding);
464 ZEBRA_RES zebra_octet_term_encoding(ZebraHandle zh, const char *encoding);
468 void zebra_set_resource(ZebraHandle zh, const char *name, const char *value);
470 const char *zebra_get_resource(ZebraHandle zh,
471 const char *name, const char *defaultvalue);
475 void zebra_pidfname(ZebraService zs, char *path);
485 ZebraMetaRecord *zebra_meta_records_create(ZebraHandle zh,
487 int num, zint *positions);
491 ZebraMetaRecord *zebra_meta_records_create_range(ZebraHandle zh,
493 zint start, int num);
496 void zebra_meta_records_destroy(ZebraHandle zh, ZebraMetaRecord *records,
500 struct BFiles_struct *zebra_get_bfs(ZebraHandle zh);
503 ZEBRA_RES zebra_set_limit(ZebraHandle zh, int complement_flag, zint *ids);
506 ZEBRA_RES zebra_set_break_handler(ZebraHandle zh,
507 int (*f)(void *client_data),
514 * \section intro_sec Introduction
516 * Zebra is a search engine for structure data, such as XML, MARC
519 * API users should read the api.h for all the public definitions.
521 * The remaining sections briefly describe each of
522 * Zebra major modules/components.
524 * \section util Base Utilities
526 * The Zebra utilities (util.h) defines fundamental types and a few
527 * utilites for Zebra.
529 * \section res Resources
531 * The resources system (res.h) is a manager of configuration
532 * resources. The resources can be viewed as a simple database.
533 * Resources can be read from a configurtion file, they can be
534 * read or written by an application. Resources can also be written,
535 * but that facility is not currently in use.
537 * \section bfile Bfiles
539 * The Bfiles (bfile.h) provides a portable interface to the
540 * local file system. It also provides a facility for safe updates
541 * (shadow updates). All file system access is handle by this module
542 * (except for trival reads of configuration files).
544 * \section dict Dictionary
546 * The Zebra dictionary (dict.h) maps a search term (key) to a value. The
547 * value is a reference to the list of records identifers in which
548 * the term occurs. Zebra uses an ISAM data structure for the list
549 * of term occurrences. The Dictionary uses \ref bfile.
553 * Zebra maintains an ISAM for each term where each ISAM is a list
554 * of record identifiers corresponding to the records in which the
555 * term occur. Unlike traditional ISAM systems, the Zebra ISAM
556 * is compressed. The ISAM system uses \ref bfile.
558 * Zebra has more than one ISAM system. The old and stable ISAM system
559 * is named isamc (see isamc.h). Another version isams is a write-once
560 * isam system that is quite compact - suitable for CD-ROMs (isams.h).
561 * The newest ISAM system, isamb, is implemented as a B-Tree (see isamb.h).
563 * \section data1 Data-1
565 * The data1 (data1.h) module deals with structured documents. The module can
566 * can read, modify and write documents. The document structure was
567 * originally based on GRS-1 - a Z39.50 v3 structure that predates
568 * DOM. These days the data1 structure may describe XML/SGML as well.
569 * The data1, like DOM, is a tree structure. Each node in the tree
570 * can be of type element, text (cdata), preprocessing instruction,
571 * comment. Element nodes can point to attribute nodes.
573 * \section recctrl Record Control
575 * The record control module (recctrl.h) is responsible for
576 * managing the various record types ("classes" or filters).
578 * \section rset Result-Set
580 * The Result-Set module (rset.h) defines an interface that all
581 * Zebra Search Results must implement. Each operation (AND, OR, ..)
582 * correspond to an implementation of that interface.
586 * DFA (dfa.h) Deterministic Finite Automa is a regular expression engine.
587 * The module compiles a regular expression to a DFA. The DFA can then
588 * be used in various application to perform fast match against the
589 * origianl expression. The \ref Dict uses DFA to perform lookup
590 * using regular expressions.
597 * indent-tabs-mode: nil
599 * vim: shiftwidth=4 tabstop=8 expandtab