2 Zvrank: an experimental ranking algorithm. See doc/zvrank.txt and
3 source in index/zvrank.c. Enable this by using rank: zvrank in zebra.cfg.
4 Contributed by Johannes Leveling <Johannes.Leveling at fernuni-hagen.de>
6 livrank: another experimental ranking algorithm. Source in livcode.c.
7 Enable this by using rank: livrank in zebra.cfg and use -DLIV_CODE=1
9 Contributed by Pete Mallinson, University of Liverpool.
11 Advanced MARC indexing. See doc/marc_indexing.xml
12 Oleg Kolobov <oleg at lib.tpu.ru>
14 Perl API updates and fixes.
15 Peter Popovics <pop at technomat.hu>
17 Fixed 'zebraidx delete'.
19 Implemented 'zebraidx clean'.
21 64-bit offsets for register files on WIN32 (no 2 GB limit).
23 Fixed a few memory leaks WRT sorting.
27 Fixed error handling : error code was not properly returned.
29 Support Truncation 104 (CCL).
33 Added missing source files for perl extension.
37 Implemented xelm directive.
39 Updated for newer version of YAZ (introduction of string schema).
41 Directory examples/zthes now part of distribution (was missing
44 New .abs directive, systag, that control where to put retrieval
45 information. The directive takes two arguments: system tag, element name.
46 System tag is one of : rank, sysno, size.
50 Perl Filter and Perl API. By Peter Popovics.
52 For zebra.cfg, if no profilePath is specified, directory
53 (prefix)/share/idzebra/tab
56 Zebra Examples in examples . Zebra tests in test.
58 Bug fix: sort index was not properly modified on
59 record updates/deletes.
61 Fix handling of character entities for sgml filter.
63 Move data1 to Zebra (used to be part of YAZ).
67 Fix character encoding of scan response terms.
69 Fix character decoding of scan request terms.
71 Fix ESpec handling (requires YAZ 1.9.1)
73 Fix searches for complete fields.
77 When name zebra is used in a filename or directory 'idzebra' is used
78 instead to avoid confusion with GNU zebra (routing software).
80 Zebra server stops with a fatal error if config file cannot be read.
82 New config setting, followLinks, that controls whether update of files
83 should follow symbolic. Set it to 1 (for enable) or 0 (to disable).
84 By default symbolic links are followed.
86 Fix MARC transfer . MARC fields had wrong data for multiple fields.
88 XML record reader moved from YAZ to Zebra, to make YAZ less
89 dependant on external libraries.
91 Zebra uses yaz_iconv which is mini iconv library supporting UTF-8,
92 UCS4, ISO-8859-1. This means that Zebra does UNICODE even
93 on systems that doesn't offer iconv.
95 XML record reader supports external system entities.
99 New .abs-directive "xpath" that takes one argument: "enable"
100 or "disable" to enable and disable XPath -indexing. If no "xpath"
101 direcive is found in .abs-file , XPath-indexing is disabled to ensure
102 backwards compatibility. For missing .abs-files XPath-indexing is
103 enabled so that such records are searchable.
105 Zebra warns about missing .abs-file only once (for each type).
107 Fixed a bug in file update where already-inserted files could
112 Zebra license changed to GNU GPL.
114 XPath-like queries used when RPN string attributes are used, eg.
115 @attr 1=/portal/title sometitle
116 @attr 1=/portal/title[@xml:lang=da] danishtitle
117 @attr 1=/portal/title/@xml:lang da
118 @attr 1=//title sometitle
120 Zebra uses UTF-8 internally:
121 1) New setting "encoding" for zebra.cfg that specifies encoding for
122 OCTET terms in queries and record encoding for most transfer syntaxes
123 (except those that use International Strings, such as GRS-1).
124 2) The encoding of International strings is UTF-8 by default. It
125 may be changed by character set negotiation. If character set
126 negotiation is in effect and if records are selected for conversion
127 these'll be converted to the selected character set - thus overriding
128 the encoding setting in zebra.cfg.
129 3) New directive "encoding" in .abs-files. This specifies the external
130 character encoding for files indexed by zebra. However, if records
131 themselves have an XML header that specifies and encoding that'll be used
134 XML filter (-t grs.xml).
136 Multiple registers. New setting in resource 'root' that holds base
137 directory for register(s). A group a databases may be put in separate
138 register in directory root/reg by using db name 'reg/db1' ... 'reg/dbN'.
142 Fixes for Digital Unix
144 Implemented hits per term using USR:SearchResult-1.
146 New Zebra API. Locking system re-implemented.
148 --- 1.1.stable 2002/02/20
150 Rank weight can be controlled with attribute type 9. Default
151 value is 34. Recommended values between 1-36.
155 Updated for YAZ version 1.8.
157 Added support for termsets - a result set of terms matching
158 a given query. For @attr 8=<set> creates termset named <set>.
160 Added support for raw retrieval. Element Set Name R forces the
161 text filter which returns the record in its original form.
163 Added numerical sort - triggered by structure=numeric (4=109).
165 Remote record import using Z39.50 Extended Services and Segments.
167 Fixed bug where updating a database with user-defined attributes
168 could corrupt the register (bad storeKeys).
170 Multi-threaded version.
172 Fixed bug regarding proximity.
174 Documentation updates.
176 Fixed bug in record retrieval module that occured on 64-bit OSF
181 Fixed bug in makefile for WIN32.
183 Fixed bug in configure script - used bash-specific features.
187 Added support for multiple records in one file for filter grs.sgml.
189 Changed record index structure. New layout is incompatible with
190 previous releases. Added setting "recordcompression" to control
191 compression of records. Possible values are "none" (no
192 compression) and bzip2 (compression using libbz2).
194 Added XML transfer syntax support for retrieval of structured records.
195 Schema in CompSpec is recognised in retrieval of structured records.
197 Changed Tcl record filter so that it attemps to read <filt>.tflt. If
198 that fails, the filter reads the file <filt>.flt (regx style filter).
200 Implemented new Tcl record filter - use grs.tcl.<filter> to enable it.
201 Zebra's configure script automatically attempts to locate Tcl. For
202 manual Tcl configuration use option --with-tclconfig=<path> to specify
203 where Tcl's library files are located.
205 Implemented "compression" of Dictionary and ISAM system. Dictionary
208 Added "tagsysno" directive to zebra.cfg to control under which tag the
209 system ID is placed. Use tagsysno: 0 to disable Zebra's system number
212 Added "tagrank" as above.
214 Changed file naming scheme for register files from <name>.mf.<no> to
217 Implemented "position"-flag for register type (as defined in
218 default.idx). When set to zero no position (or seqence number) is
219 saved in register for each word occurrence, thus saving some register
222 Implemented database mapping. Using mapdb one can specify a database
223 to be mapped to one or more physical databases. Usage:
224 mapdb <fromdb> <todb> ..
226 Added SOIF-filter. Thanks to Peter Valkenburg.
228 For the regx-filter "end element -record" may trigger a mark-of-record
229 if outer level is reached.
231 Tag sets may be typed in the reference to it. From the .abs-file the
232 "tagset" directive takes a third optional integer type for the tag set
233 referenced. From a .tag-file the "include" directive takes a third
234 optional type as well. The old "type" directive in the tag set itself
235 is still recognized but acts as the default type for the tag set.
237 Zebra supports the specification of arbitrary attributes sets, schemas
238 and tag sets, because of the change in YAZ' OID management system.
240 Fixed bug in Sort that caused it NOT to use character mapping as it
243 Zebra now uses GNU configure to generate Makefile(s).
245 Added un-optimised support for left and left/right truncation attributes.
247 Added support for relational operators on text when using RPN queries.
249 Added support for sort specifications in RPN queries. Type 7 specifies
250 'sort' where value 1=ascending, value 2=descending. The use attribute
251 specifies the field criteria as usual. The term specifies priority
252 where 0=first, 1=second, ...
254 Changed the way use attributes are specified in the recordId
257 Maximum number of databases in one Zebra register increased.
259 New setting, databasePath, which specifies that first directory during
260 update traversal is the database name (instead of a fixed one).
262 New setting, explainDatabase, which specifies that databases are
265 Modified Zebra so that it works with ASN.1 compiled code for YAZ.
267 Implemented EXPLAIN database maintenance. Zebra automatically
268 generate - and update CategoryList, TargetInfo, DatabaseInfo,
269 AttributeSetInfo and AttributeDetails records at this stage. The
270 records may be transferred as GRS-1, SUTRS or Explain.
272 Fixed register spec so that colon isn't treated as size separator
273 unless followed by [0-9+-] in order to allow DOS drive specifications.
275 Fixed two bugs in ISAMC system.
277 Changed the way Zebra keeps its maintenance information about attribute
278 sets, available attributes, etc.. Records in "SGML" notation using an
279 EXPLAIN schema is now used when appropriate.
281 Bug fix: Index didn't handle update/insert/delete of the same record
282 (i.e. same recordId) in one run (one invocation of zebraidx). Only the
283 first occurence of a record is considered.
285 Most searches now return correct number of hits.
287 New modular ranking system. Interested programmers are encouraged to
288 inspect rank1.c and improve the algorithm.
290 Bug fix: Lock files weren't removed as they should on NT.
292 Implemented Z39.50 Sort. Zebra's sort handler uses use attributes to
293 specify a "sort register". Refer to the gils sample records which refer
294 to index type "s" which is specified as "sort" in the default.idx file.
295 Each sort criteria can either be Ascending or Descending and at most
296 three sort elements can be specified.
298 Bug fix: Character mapping didn't work for text files.
302 Simple ranked searches now return correct number of hits.
304 The test option (-s) only makes a read-lock on the index as well
305 as using read-only operations anywhere.
307 Moved towards generic character mapping. Configuration file default.idx
308 specifies character map files for register types w, p, u, etc.
310 Implemented "begin variant" for the sgml.regx - filter.
312 Fixed a few memory leaks.
314 Added support for C++, headers uses extern "C" for public definitions.
316 Bug fix: The show records facility (-s) only displayed information for
317 the first record in a file (and not for every record in the file).
319 Added option "-f <n>" to limit the logging of record operations. After
320 <n> records has been processed no logging is performed (unless errors
323 Bug fix: the compressed ISAM system didn't handle update operations
326 Added setting, "maxResultSetSize", to hold the number of records to
327 save in a result set.
329 Bug fix: Complete phrase did't work for search operations.
331 Bug fix: temporary result sets weren't deleted.
333 Reduced disk space for saved keys (storeKeys = 1).
335 Added optional, physical ANY (key replication)
337 Implemented proximity operator in search.
339 Bug fix: the path name buffers used by file match traversal routines
340 have been extended to support long file names.
342 New C(ompressed) ISAM system. To enable it, specify "isam: c" in the
343 configuration file. The resulting register without "storeKeys" is about
344 half the size, and the memory used by zebraidx during phase 2 (merge) is
345 reduced to a minimum.
347 Reworked the way Regexp-2 queries with error tolerance are handled and
348 specified. The documentation has been updated accordingly.
350 Bug fix: Zebrasrv didn't search correctly when queries contained masking
351 characters. This bug was introduced in 1.0a8.
353 Zebrasrv now tag records with the proper database name.
355 New settings, memMax and keyTmpDir.
357 Changed name of setting lockDir (previously called lockPath) and
358 setTmpDir (previously called tempSetPath).
360 Generalized and changed record type specifications. In short, there are:
362 grs.sgml structured, "SGML-like" syntax
363 grs.regx.<filter> structured, Regular expression filter
364 grs.marc.<abs> Reads *MARC records in the ISO2709 format. <abs>
365 is the name of an abstract syntax file.
366 Bug fix: Result sets weren't sorted in operations involving boolean
367 operations with "ranked" operands.
371 Added national character-handling subsystem.
375 Small modifications to input filters and profiles.
377 Added support for SOIF syntax (with private OID).
381 Fixed buffer-size problem in indexing.
383 Added compression to temporary files for updating.
385 Added phrase registers.
387 Added dynamic mapping of search attribute to multiple termlists (ANY).
389 Scan support in multiple databases/registers.
391 Configuration settings are case-insensitive and single dash (-)
392 characters are ignored in comparisons.
394 The index processing ignores empty files - warning given.
396 New option to zebraidx (-V) displays version information.
400 Fixed problem in file-update system.
402 Fixed problem in shadow system; register was sometimes corrupted after
407 Fixed problems in the ISAM subsystem. Caused difficulties when updating
410 Fixed small problem in SUTRS-filter. A newline was sometimes inserted before
411 the rank and record number.
413 Fixed bug in the isam subsystem - caused a malfunction when accessing
414 words which occurred more than 10000 times.
416 Distribution should now include YAZ (Z39.50 protocol stack) to simplify
419 Server can now run under inetd. Use option -i, and -w <directory> to
420 set working directory to desired location.
422 New zebraidx command: clean - removes temporary shadow files.
424 Fixed bug in ISAM system. Occurred rarely during register updates.
426 Logging during index merge phase is improved. The remaining running
429 Temporary files generated by zebraidx are removed after each run.
431 Bug fix: Dictionary didn't handle 8-bit characters correctly; was obvious
432 when doing scan operations in dictionaries with European characters.
436 A whole slew of updates, to make the first publicized release. Get the doc
441 Memory-problems in ISAM fixed. More blocktypes added to the default setup
442 to increase performance on larger databases.
444 Various minor changes in data management system.
448 A couple of portability-problems resolved.
450 Changed some malloc() to xmalloc().