2 Fix XML error handling. Stop XML parse immediately if XML parse error
3 occur (i.e. produce one error only).
5 Zebra ignores "unsupported use attribute" for individual databases
6 when search multiple databases (unless all databases fail).
8 New filter grs.marcxml which works like grs.marc but produces MARCXML.
10 Added support for database deletion. It is possible to create/drop
11 a database from zebraidx utility.
13 Write zebrasrv.pid to lockdir.
15 Bug fix: result sets were not recovered correctly. Had to
16 add ODR handle for zebra_search_RPN in order to make it work.
18 Fixed a bug in regx filters that didn't do anchors (^) correctly.
20 Fixed a bug in searches with X-Path searches sometimes giving "extra"
23 Zebra server checks for zebrasrv.pid and refuses to start if it is already
24 locked by another (running) zebrasrv.
26 Fixed a bug with text being chunked in pieces for the grs.xml filter.
30 xelm code updates. xelm works regardless state of 'xpath enable/disable'
31 Avoid -L/usr/lib since that is already default library path.
33 Allow multiple updates within one transaction.
35 Fixed a bug with >2GB files (overflow in integer expression).
39 Fix linker error for Perl module.
41 Fix bug in and operation which in some cases could result in "extra"
42 hits. Bug was introduced in 1.3.5.
44 Fix bug in handling of schema conversion when producing numeric tags.
50 Add missing files doc/zvrank.txt and doc/marc_indexing.xml.
54 Zvrank: an experimental ranking algorithm. See doc/zvrank.txt and
55 source in index/zvrank.c. Enable this by using rank: zvrank in zebra.cfg.
56 Contributed by Johannes Leveling <Johannes.Leveling at fernuni-hagen.de>
58 livrank: another experimental ranking algorithm. Source in livcode.c.
59 Enable this by using rank: livrank in zebra.cfg and use -DLIV_CODE=1
61 Contributed by Pete Mallinson, University of Liverpool.
63 Advanced MARC indexing. See doc/marc_indexing.xml
64 Oleg Kolobov <oleg at lib.tpu.ru>
66 Perl API updates and fixes.
67 Peter Popovics <pop at technomat.hu>
69 Fixed 'zebraidx delete'.
71 Implemented 'zebraidx clean'.
73 64-bit offsets for register files on WIN32 (no 2 GB limit).
75 Fixed a few memory leaks WRT sorting.
79 Fixed error handling : error code was not properly returned.
81 Support Truncation 104 (CCL).
85 Added missing source files for perl extension.
89 Implemented xelm directive.
91 Updated for newer version of YAZ (introduction of string schema).
93 Directory examples/zthes now part of distribution (was missing
96 New .abs directive, systag, that control where to put retrieval
97 information. The directive takes two arguments: system tag, element name.
98 System tag is one of : rank, sysno, size.
102 Perl Filter and Perl API. By Peter Popovics.
104 For zebra.cfg, if no profilePath is specified, directory
105 (prefix)/share/idzebra/tab
108 Zebra Examples in examples . Zebra tests in test.
110 Bug fix: sort index was not properly modified on
111 record updates/deletes.
113 Fix handling of character entities for sgml filter.
115 Move data1 to Zebra (used to be part of YAZ).
119 Fix character encoding of scan response terms.
121 Fix character decoding of scan request terms.
123 Fix ESpec handling (requires YAZ 1.9.1)
125 Fix searches for complete fields.
129 When name zebra is used in a filename or directory 'idzebra' is used
130 instead to avoid confusion with GNU zebra (routing software).
132 Zebra server stops with a fatal error if config file cannot be read.
134 New config setting, followLinks, that controls whether update of files
135 should follow symbolic. Set it to 1 (for enable) or 0 (to disable).
136 By default symbolic links are followed.
138 Fix MARC transfer . MARC fields had wrong data for multiple fields.
140 XML record reader moved from YAZ to Zebra, to make YAZ less
141 dependant on external libraries.
143 Zebra uses yaz_iconv which is mini iconv library supporting UTF-8,
144 UCS4, ISO-8859-1. This means that Zebra does UNICODE even
145 on systems that doesn't offer iconv.
147 XML record reader supports external system entities.
151 New .abs-directive "xpath" that takes one argument: "enable"
152 or "disable" to enable and disable XPath -indexing. If no "xpath"
153 direcive is found in .abs-file , XPath-indexing is disabled to ensure
154 backwards compatibility. For missing .abs-files XPath-indexing is
155 enabled so that such records are searchable.
157 Zebra warns about missing .abs-file only once (for each type).
159 Fixed a bug in file update where already-inserted files could
164 Zebra license changed to GNU GPL.
166 XPath-like queries used when RPN string attributes are used, eg.
167 @attr 1=/portal/title sometitle
168 @attr 1=/portal/title[@xml:lang=da] danishtitle
169 @attr 1=/portal/title/@xml:lang da
170 @attr 1=//title sometitle
172 Zebra uses UTF-8 internally:
173 1) New setting "encoding" for zebra.cfg that specifies encoding for
174 OCTET terms in queries and record encoding for most transfer syntaxes
175 (except those that use International Strings, such as GRS-1).
176 2) The encoding of International strings is UTF-8 by default. It
177 may be changed by character set negotiation. If character set
178 negotiation is in effect and if records are selected for conversion
179 these'll be converted to the selected character set - thus overriding
180 the encoding setting in zebra.cfg.
181 3) New directive "encoding" in .abs-files. This specifies the external
182 character encoding for files indexed by zebra. However, if records
183 themselves have an XML header that specifies and encoding that'll be used
186 XML filter (-t grs.xml).
188 Multiple registers. New setting in resource 'root' that holds base
189 directory for register(s). A group a databases may be put in separate
190 register in directory root/reg by using db name 'reg/db1' ... 'reg/dbN'.
194 Fixes for Digital Unix
196 Implemented hits per term using USR:SearchResult-1.
198 New Zebra API. Locking system re-implemented.
200 --- 1.1.stable 2002/02/20
202 Rank weight can be controlled with attribute type 9. Default
203 value is 34. Recommended values between 1-36.
207 Updated for YAZ version 1.8.
209 Added support for termsets - a result set of terms matching
210 a given query. For @attr 8=<set> creates termset named <set>.
212 Added support for raw retrieval. Element Set Name R forces the
213 text filter which returns the record in its original form.
215 Added numerical sort - triggered by structure=numeric (4=109).
217 Remote record import using Z39.50 Extended Services and Segments.
219 Fixed bug where updating a database with user-defined attributes
220 could corrupt the register (bad storeKeys).
222 Multi-threaded version.
224 Fixed bug regarding proximity.
226 Documentation updates.
228 Fixed bug in record retrieval module that occured on 64-bit OSF
233 Fixed bug in makefile for WIN32.
235 Fixed bug in configure script - used bash-specific features.
239 Added support for multiple records in one file for filter grs.sgml.
241 Changed record index structure. New layout is incompatible with
242 previous releases. Added setting "recordcompression" to control
243 compression of records. Possible values are "none" (no
244 compression) and bzip2 (compression using libbz2).
246 Added XML transfer syntax support for retrieval of structured records.
247 Schema in CompSpec is recognised in retrieval of structured records.
249 Changed Tcl record filter so that it attemps to read <filt>.tflt. If
250 that fails, the filter reads the file <filt>.flt (regx style filter).
252 Implemented new Tcl record filter - use grs.tcl.<filter> to enable it.
253 Zebra's configure script automatically attempts to locate Tcl. For
254 manual Tcl configuration use option --with-tclconfig=<path> to specify
255 where Tcl's library files are located.
257 Implemented "compression" of Dictionary and ISAM system. Dictionary
260 Added "tagsysno" directive to zebra.cfg to control under which tag the
261 system ID is placed. Use tagsysno: 0 to disable Zebra's system number
264 Added "tagrank" as above.
266 Changed file naming scheme for register files from <name>.mf.<no> to
269 Implemented "position"-flag for register type (as defined in
270 default.idx). When set to zero no position (or seqence number) is
271 saved in register for each word occurrence, thus saving some register
274 Implemented database mapping. Using mapdb one can specify a database
275 to be mapped to one or more physical databases. Usage:
276 mapdb <fromdb> <todb> ..
278 Added SOIF-filter. Thanks to Peter Valkenburg.
280 For the regx-filter "end element -record" may trigger a mark-of-record
281 if outer level is reached.
283 Tag sets may be typed in the reference to it. From the .abs-file the
284 "tagset" directive takes a third optional integer type for the tag set
285 referenced. From a .tag-file the "include" directive takes a third
286 optional type as well. The old "type" directive in the tag set itself
287 is still recognized but acts as the default type for the tag set.
289 Zebra supports the specification of arbitrary attributes sets, schemas
290 and tag sets, because of the change in YAZ' OID management system.
292 Fixed bug in Sort that caused it NOT to use character mapping as it
295 Zebra now uses GNU configure to generate Makefile(s).
297 Added un-optimised support for left and left/right truncation attributes.
299 Added support for relational operators on text when using RPN queries.
301 Added support for sort specifications in RPN queries. Type 7 specifies
302 'sort' where value 1=ascending, value 2=descending. The use attribute
303 specifies the field criteria as usual. The term specifies priority
304 where 0=first, 1=second, ...
306 Changed the way use attributes are specified in the recordId
309 Maximum number of databases in one Zebra register increased.
311 New setting, databasePath, which specifies that first directory during
312 update traversal is the database name (instead of a fixed one).
314 New setting, explainDatabase, which specifies that databases are
317 Modified Zebra so that it works with ASN.1 compiled code for YAZ.
319 Implemented EXPLAIN database maintenance. Zebra automatically
320 generate - and update CategoryList, TargetInfo, DatabaseInfo,
321 AttributeSetInfo and AttributeDetails records at this stage. The
322 records may be transferred as GRS-1, SUTRS or Explain.
324 Fixed register spec so that colon isn't treated as size separator
325 unless followed by [0-9+-] in order to allow DOS drive specifications.
327 Fixed two bugs in ISAMC system.
329 Changed the way Zebra keeps its maintenance information about attribute
330 sets, available attributes, etc.. Records in "SGML" notation using an
331 EXPLAIN schema is now used when appropriate.
333 Bug fix: Index didn't handle update/insert/delete of the same record
334 (i.e. same recordId) in one run (one invocation of zebraidx). Only the
335 first occurence of a record is considered.
337 Most searches now return correct number of hits.
339 New modular ranking system. Interested programmers are encouraged to
340 inspect rank1.c and improve the algorithm.
342 Bug fix: Lock files weren't removed as they should on NT.
344 Implemented Z39.50 Sort. Zebra's sort handler uses use attributes to
345 specify a "sort register". Refer to the gils sample records which refer
346 to index type "s" which is specified as "sort" in the default.idx file.
347 Each sort criteria can either be Ascending or Descending and at most
348 three sort elements can be specified.
350 Bug fix: Character mapping didn't work for text files.
354 Simple ranked searches now return correct number of hits.
356 The test option (-s) only makes a read-lock on the index as well
357 as using read-only operations anywhere.
359 Moved towards generic character mapping. Configuration file default.idx
360 specifies character map files for register types w, p, u, etc.
362 Implemented "begin variant" for the sgml.regx - filter.
364 Fixed a few memory leaks.
366 Added support for C++, headers uses extern "C" for public definitions.
368 Bug fix: The show records facility (-s) only displayed information for
369 the first record in a file (and not for every record in the file).
371 Added option "-f <n>" to limit the logging of record operations. After
372 <n> records has been processed no logging is performed (unless errors
375 Bug fix: the compressed ISAM system didn't handle update operations
378 Added setting, "maxResultSetSize", to hold the number of records to
379 save in a result set.
381 Bug fix: Complete phrase did't work for search operations.
383 Bug fix: temporary result sets weren't deleted.
385 Reduced disk space for saved keys (storeKeys = 1).
387 Added optional, physical ANY (key replication)
389 Implemented proximity operator in search.
391 Bug fix: the path name buffers used by file match traversal routines
392 have been extended to support long file names.
394 New C(ompressed) ISAM system. To enable it, specify "isam: c" in the
395 configuration file. The resulting register without "storeKeys" is about
396 half the size, and the memory used by zebraidx during phase 2 (merge) is
397 reduced to a minimum.
399 Reworked the way Regexp-2 queries with error tolerance are handled and
400 specified. The documentation has been updated accordingly.
402 Bug fix: Zebrasrv didn't search correctly when queries contained masking
403 characters. This bug was introduced in 1.0a8.
405 Zebrasrv now tag records with the proper database name.
407 New settings, memMax and keyTmpDir.
409 Changed name of setting lockDir (previously called lockPath) and
410 setTmpDir (previously called tempSetPath).
412 Generalized and changed record type specifications. In short, there are:
414 grs.sgml structured, "SGML-like" syntax
415 grs.regx.<filter> structured, Regular expression filter
416 grs.marc.<abs> Reads *MARC records in the ISO2709 format. <abs>
417 is the name of an abstract syntax file.
418 Bug fix: Result sets weren't sorted in operations involving boolean
419 operations with "ranked" operands.
423 Added national character-handling subsystem.
427 Small modifications to input filters and profiles.
429 Added support for SOIF syntax (with private OID).
433 Fixed buffer-size problem in indexing.
435 Added compression to temporary files for updating.
437 Added phrase registers.
439 Added dynamic mapping of search attribute to multiple termlists (ANY).
441 Scan support in multiple databases/registers.
443 Configuration settings are case-insensitive and single dash (-)
444 characters are ignored in comparisons.
446 The index processing ignores empty files - warning given.
448 New option to zebraidx (-V) displays version information.
452 Fixed problem in file-update system.
454 Fixed problem in shadow system; register was sometimes corrupted after
459 Fixed problems in the ISAM subsystem. Caused difficulties when updating
462 Fixed small problem in SUTRS-filter. A newline was sometimes inserted before
463 the rank and record number.
465 Fixed bug in the isam subsystem - caused a malfunction when accessing
466 words which occurred more than 10000 times.
468 Distribution should now include YAZ (Z39.50 protocol stack) to simplify
471 Server can now run under inetd. Use option -i, and -w <directory> to
472 set working directory to desired location.
474 New zebraidx command: clean - removes temporary shadow files.
476 Fixed bug in ISAM system. Occurred rarely during register updates.
478 Logging during index merge phase is improved. The remaining running
481 Temporary files generated by zebraidx are removed after each run.
483 Bug fix: Dictionary didn't handle 8-bit characters correctly; was obvious
484 when doing scan operations in dictionaries with European characters.
488 A whole slew of updates, to make the first publicized release. Get the doc
493 Memory-problems in ISAM fixed. More blocktypes added to the default setup
494 to increase performance on larger databases.
496 Various minor changes in data management system.
500 A couple of portability-problems resolved.
502 Changed some malloc() to xmalloc().