1 /* $Id: api.h,v 1.47 2007-03-14 11:48:31 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.
36 #include <yaz/proto.h>
37 #include <idzebra/res.h>
38 #include <idzebra/version.h>
39 #include <idzebra/recctrl.h>
50 } ZebraTransactionStatus;
52 /** Retrieval Record Descriptor */
54 int errCode; /* non-zero if error when fetching this */
55 char *errString; /* error string */
56 int position; /* position of record in result set (1,2,..) */
57 char *buf; /* record buffer (void pointer really) */
59 oid_value format; /* record syntax */
63 } ZebraRetrievalRecord;
65 /** Scan Term Descriptor */
67 zint occurrences; /* scan term occurrences */
68 char *term; /* scan term string */
72 \brief a Zebra Handle - (session)
74 typedef struct zebra_session *ZebraHandle;
77 \brief a Zebra Service handle
79 typedef struct zebra_service *ZebraService;
81 /** \brief Creates a Zebra Service.
82 \param configName name of configuration file
84 This function is a simplified version of zebra_start_res.
87 ZebraService zebra_start(const char *configName
88 ) ZEBRA_GCC_ATTR((warn_unused_result));
90 /** \brief Creates a Zebra service with resources.
91 \param configName name of configuration file
92 \param def_res default resources
93 \param over_res overriding resources
95 This function typically called once in a program. A Zebra Service
96 acts as a factory for Zebra session handles.
99 ZebraService zebra_start_res(const char *configName,
100 Res def_res, Res over_res
101 ) ZEBRA_GCC_ATTR((warn_unused_result));
103 /** \brief stops a Zebra service.
104 \param zs service handle
106 Frees resources used by the service.
109 ZEBRA_RES zebra_stop(ZebraService zs);
111 /** \brief Lists enabled Zebra filters
112 \param zs service handle
113 \param cd callback parameter (opaque)
114 \param cb callback function
117 void zebra_filter_info(ZebraService zs, void *cd,
118 void (*cb)(void *cd, const char *name));
121 /** \brief Creates a Zebra session handle within service.
122 \param zs service handle.
123 \param res resources to be used for the service (NULL for none)
125 There should be one handle for each thread doing something
126 with zebra, be that searching or indexing. In simple apps
127 one handle is sufficient
130 ZebraHandle zebra_open(ZebraService zs, Res res
131 ) ZEBRA_GCC_ATTR((warn_unused_result));
133 /** \brief Destroys Zebra session handle.
134 \param zh zebra session handle.
137 ZEBRA_RES zebra_close(ZebraHandle zh);
139 /** \brief Returns error code for last error
140 \param zh zebra session handle.
143 int zebra_errCode(ZebraHandle zh);
145 /** \brief Returns error string for last error
146 \param zh zebra session handle.
149 const char *zebra_errString(ZebraHandle zh);
151 /** \brief Returns additional info for last error
152 \param zh zebra session handle.
155 char *zebra_errAdd(ZebraHandle zh);
157 /** \brief Returns error code and additional info for last error
158 \param zh zebra session handle.
159 \param code pointer to returned error code
160 \param addinfo pointer to returned additional info
163 void zebra_result(ZebraHandle zh, int *code, char **addinfo);
165 /** \brief Set limit before Zebra does approx hit count
166 \param zh session handle
167 \param approx_limit the limit
169 Results will be approximiate if hit count is greater than the
170 limit specified. By default there is a high-limit (no limit).
172 ZEBRA_RES zebra_set_approx_limit(ZebraHandle zh, zint approx_limit);
174 /** \brief Search using PQF Query String
175 \param zh session handle
176 \param pqf_query query
177 \param setname name of resultset
178 \param hits of hits is returned
181 ZEBRA_RES zebra_search_PQF(ZebraHandle zh, const char *pqf_query,
182 const char *setname, zint *hits);
184 /** \brief Search using RPN Query structure (from ASN.1)
185 \param zh session handle
187 \param query RPN query using YAZ structure
188 \param setname name of resultset
189 \param hits number of hits is returned
190 \param estimated_hit_count whether hit count is an estimate
191 \param partial_resultset whether result is only partially evaluated
194 ZEBRA_RES zebra_search_RPN_x(ZebraHandle zh, ODR o, Z_RPNQuery *query,
195 const char *setname, zint *hits,
196 int *estimated_hit_count,
197 int *partial_resultset);
200 /** \brief Search using RPN Query structure (from ASN.1)
201 \param zh session handle
203 \param query RPN query using YAZ structure
204 \param setname name of resultset
205 \param hits number of hits is returned
208 ZEBRA_RES zebra_search_RPN(ZebraHandle zh, ODR o, Z_RPNQuery *query,
209 const char *setname, zint *hits);
211 /** \brief Retrieve records from result set (after search)
212 \param zh session handle
213 \param stream allocate records returned using this ODR
214 \param setname name of result set to retrieve records from
215 \param comp Z39.50 record composition
216 \param input_format transfer syntax (OID)
217 \param num_recs number of records to retrieve
218 \param recs store records in this structure (size is num_recs)
221 ZEBRA_RES zebra_records_retrieve(ZebraHandle zh, ODR stream,
223 Z_RecordComposition *comp,
224 oid_value input_format,
226 ZebraRetrievalRecord *recs);
227 /** \brief Deletes one or more resultsets
228 \param zh session handle
229 \param function Z_DeleteResultSetRequest_{list,all}
230 \param num_setnames number of result sets
231 \param setnames result set names
232 \param statuses status result
235 int zebra_deleteResultSet(ZebraHandle zh, int function,
236 int num_setnames, char **setnames,
240 /** \brief returns number of term info terms assocaited with result set
241 \param zh session handle
242 \param setname result set name
243 \param num_terms number of terms returned in this integer
245 This function is used in conjunction with zebra_result_set_term_info.
246 If operation was successful, ZEBRA_OK is returned; otherwise
247 ZEBRA_FAIL is returned (typically non-existing setname)
250 ZEBRA_RES zebra_result_set_term_no(ZebraHandle zh, const char *setname,
253 /** \brief returns information about a term assocated with a result set
254 \param zh session handle
255 \param setname result set name
256 \param no the term we want to know about (0=first, 1=second,..)
257 \param count the number of occurrences of this term, aka hits (output)
258 \param approx about hits: 0=exact,1=approx (output)
259 \param termbuf buffer for term string (intput, output)
260 \param termlen size of termbuf (input=max, output=actual length)
261 \param term_ref_id if non-NULL *term_ref_id holds term reference
263 Returns information about one search term associated with result set.
264 Use zebra_result_set_term_no to read total number of terms associated
265 with result set. If this function can not return information,
266 due to no out of range or bad result set name, ZEBRA_FAIL is
268 The passed termbuf must be able to hold at least *termlen characters.
269 Upon completion, *termlen holds actual length of search term.
272 ZEBRA_RES zebra_result_set_term_info(ZebraHandle zh, const char *setname,
273 int no, zint *count, int *approx,
274 char *termbuf, size_t *termlen,
275 const char **term_ref_id);
278 /** \brief performs Scan (Z39.50 style)
279 \param zh session handle
280 \param stream ODR handle for result
281 \param zapt Attribute plus Term (start term)
282 \param attributeset Attributeset for Attribute plus Term
283 \param position input/output position
284 \param num_entries number of terms requested / returned
285 \param entries list of resulting terms (ODR allocated)
286 \param is_partial upon return 1=partial, 0=complete
287 \param setname limit scan by this set (NULL means no limit)
289 YAZ_EXPORT ZEBRA_RES zebra_scan(ZebraHandle zh, ODR stream,
290 Z_AttributesPlusTerm *zapt,
291 oid_value attributeset,
292 int *position, int *num_entries,
293 ZebraScanEntry **entries,
295 const char *setname);
297 /** \brief performs Scan (taking PQF string)
298 \param zh session handle
299 \param stream ODR handle for result
300 \param query PQF scan query
301 \param position input/output position
302 \param num_entries number of terms requested / returned
303 \param entries list of resulting terms (ODR allocated)
304 \param is_partial upon return 1=partial, 0=complete
305 \param setname limit scan by this set (NULL means no limit)
308 ZEBRA_RES zebra_scan_PQF(ZebraHandle zh, ODR stream, const char *query,
309 int *position, int *num_entries,
310 ZebraScanEntry **entries,
311 int *is_partial, const char *setname);
313 /** \brief authenticate user. Returns 0 if OK, != 0 on failure
314 \param zh session handle
315 \param user user name
319 ZEBRA_RES zebra_auth(ZebraHandle zh, const char *user, const char *pass);
321 /** \brief Normalize zebra term for register (subject to change!)
322 \param zh session handle
323 \param reg_id register ID, 'w', 'p',..
324 \param input_str input string buffer
325 \param input_len input string length
326 \param output_str output string buffer
327 \param output_len output string length
330 int zebra_string_norm(ZebraHandle zh, unsigned reg_id, const char *input_str,
331 int input_len, char *output_str, int output_len);
333 /** \brief Creates a database
334 \param zh session handle
335 \param db database to be created
338 ZEBRA_RES zebra_create_database(ZebraHandle zh, const char *db);
340 /** \brief Deletes a database (drop)
341 \param zh session handle
342 \param db database to be deleted
345 ZEBRA_RES zebra_drop_database(ZebraHandle zh, const char *db);
348 ZEBRA_RES zebra_admin_shutdown(ZebraHandle zh);
351 ZEBRA_RES zebra_admin_start(ZebraHandle zh);
354 ZEBRA_RES zebra_shutdown(ZebraService zs);
357 ZEBRA_RES zebra_admin_import_begin(ZebraHandle zh, const char *database,
358 const char *record_type);
361 ZEBRA_RES zebra_admin_import_segment(ZebraHandle zh,
365 ZEBRA_RES zebra_admin_import_end(ZebraHandle zh);
368 ZEBRA_RES zebra_begin_trans(ZebraHandle zh, int rw
369 ) ZEBRA_GCC_ATTR((warn_unused_result));
372 ZEBRA_RES zebra_end_trans(ZebraHandle zh
373 ) ZEBRA_GCC_ATTR((warn_unused_result));
376 ZEBRA_RES zebra_end_transaction(ZebraHandle zh,
377 ZebraTransactionStatus *stat);
380 ZEBRA_RES zebra_commit(ZebraHandle zh);
383 ZEBRA_RES zebra_clean(ZebraHandle zh);
386 ZEBRA_RES zebra_init(ZebraHandle zh);
389 ZEBRA_RES zebra_compact(ZebraHandle zh);
392 ZEBRA_RES zebra_repository_update(ZebraHandle zh, const char *path);
395 ZEBRA_RES zebra_repository_delete(ZebraHandle zh, const char *path);
398 ZEBRA_RES zebra_repository_show(ZebraHandle zh, const char *path);
400 /** \brief Simple update record
401 \param zh session handle
402 \param buf record buffer
403 \param buf_size record buffer size
405 This function is a simple wrapper or zebra_update_record with
406 action=action_update (insert or replace) .
410 ZEBRA_RES zebra_add_record(ZebraHandle zh, const char *buf, int buf_size);
412 /** \brief Updates record
413 \param zh session handle
414 \param action (insert,replace,delete or update (replace/insert)
415 \param recordType filter type (0 indicates default)
416 \param sysno system id (0 may be passed for no known id)
417 \param match match criteria (0 may be passed for no known criteria)
418 \param fname filename to be printed for logging (0 may be passed)
419 \param buf record buffer
420 \param buf_size record buffer size
423 ZEBRA_RES zebra_update_record(ZebraHandle zh,
424 enum zebra_recctrl_action_t action,
425 const char *recordType,
426 zint *sysno, const char *match,
428 const char *buf, int buf_size);
431 ZEBRA_RES zebra_sort(ZebraHandle zh, ODR stream,
432 int num_input_setnames,
433 const char **input_setnames,
434 const char *output_setname,
435 Z_SortKeySpecList *sort_sequence,
437 ) ZEBRA_GCC_ATTR((warn_unused_result));
440 ZEBRA_RES zebra_select_databases(ZebraHandle zh, int num_bases,
441 const char **basenames
442 ) ZEBRA_GCC_ATTR((warn_unused_result));
445 ZEBRA_RES zebra_select_database(ZebraHandle zh, const char *basename
446 ) ZEBRA_GCC_ATTR((warn_unused_result));
449 void zebra_shadow_enable(ZebraHandle zh, int value);
452 int zebra_register_statistics(ZebraHandle zh, int dumpdict);
455 ZEBRA_RES zebra_record_encoding(ZebraHandle zh, const char *encoding);
458 ZEBRA_RES zebra_octet_term_encoding(ZebraHandle zh, const char *encoding);
462 void zebra_set_resource(ZebraHandle zh, const char *name, const char *value);
464 const char *zebra_get_resource(ZebraHandle zh,
465 const char *name, const char *defaultvalue);
469 void zebra_pidfname(ZebraService zs, char *path);
479 ZebraMetaRecord *zebra_meta_records_create(ZebraHandle zh,
481 int num, zint *positions);
485 ZebraMetaRecord *zebra_meta_records_create_range(ZebraHandle zh,
487 zint start, int num);
490 void zebra_meta_records_destroy(ZebraHandle zh, ZebraMetaRecord *records,
494 struct BFiles_struct *zebra_get_bfs(ZebraHandle zh);
497 ZEBRA_RES zebra_set_limit(ZebraHandle zh, int complement_flag, zint *ids);
500 ZEBRA_RES zebra_set_break_handler(ZebraHandle zh,
501 int (*f)(void *client_data),
508 * \section intro_sec Introduction
510 * Zebra is a search engine for structure data, such as XML, MARC
513 * API users should read the api.h for all the public definitions.
515 * The remaining sections briefly describe each of
516 * Zebra major modules/components.
518 * \section util Base Utilities
520 * The Zebra utilities (util.h) defines fundamental types and a few
521 * utilites for Zebra.
523 * \section res Resources
525 * The resources system (res.h) is a manager of configuration
526 * resources. The resources can be viewed as a simple database.
527 * Resources can be read from a configurtion file, they can be
528 * read or written by an application. Resources can also be written,
529 * but that facility is not currently in use.
531 * \section bfile Bfiles
533 * The Bfiles (bfile.h) provides a portable interface to the
534 * local file system. It also provides a facility for safe updates
535 * (shadow updates). All file system access is handle by this module
536 * (except for trival reads of configuration files).
538 * \section dict Dictionary
540 * The Zebra dictionary (dict.h) maps a search term (key) to a value. The
541 * value is a reference to the list of records identifers in which
542 * the term occurs. Zebra uses an ISAM data structure for the list
543 * of term occurrences. The Dictionary uses \ref bfile.
547 * Zebra maintains an ISAM for each term where each ISAM is a list
548 * of record identifiers corresponding to the records in which the
549 * term occur. Unlike traditional ISAM systems, the Zebra ISAM
550 * is compressed. The ISAM system uses \ref bfile.
552 * Zebra has more than one ISAM system. The old and stable ISAM system
553 * is named isamc (see isamc.h). Another version isams is a write-once
554 * isam system that is quite compact - suitable for CD-ROMs (isams.h).
555 * The newest ISAM system, isamb, is implemented as a B-Tree (see isamb.h).
557 * \section data1 Data-1
559 * The data1 (data1.h) module deals with structured documents. The module can
560 * can read, modify and write documents. The document structure was
561 * originally based on GRS-1 - a Z39.50 v3 structure that predates
562 * DOM. These days the data1 structure may describe XML/SGML as well.
563 * The data1, like DOM, is a tree structure. Each node in the tree
564 * can be of type element, text (cdata), preprocessing instruction,
565 * comment. Element nodes can point to attribute nodes.
567 * \section recctrl Record Control
569 * The record control module (recctrl.h) is responsible for
570 * managing the various record types ("classes" or filters).
572 * \section rset Result-Set
574 * The Result-Set module (rset.h) defines an interface that all
575 * Zebra Search Results must implement. Each operation (AND, OR, ..)
576 * correspond to an implementation of that interface.
580 * DFA (dfa.h) Deterministic Finite Automa is a regular expression engine.
581 * The module compiles a regular expression to a DFA. The DFA can then
582 * be used in various application to perform fast match against the
583 * origianl expression. The \ref Dict uses DFA to perform lookup
584 * using regular expressions.
591 * indent-tabs-mode: nil
593 * vim: shiftwidth=4 tabstop=8 expandtab