3 For mod_dom filter, allow type="adelete", which will not treat a failed
4 delete as an error (such as a record ID that do not exist already).
6 Log XML2/XSLT errors to yaz_log instead of stdout/stderr.
8 Relay \-character in non-regexp searches. Was treated as backslash
13 Improve hit estimate for or-searches (typically truncation)
15 Fix snippets for non-ICU / seqno not updated.
17 Replace key_compare_it with key_compare. This is because key_compare_it
18 throws exception on some CPUs with alignment constraints.
20 ICU: support @attr 5=2, @attr 5=3 in searches. (left truncation and
21 left+right truncation).
25 Fix problem in dictionary due to increase max size of term.
29 Fix buffer overrun for terms using ICU normalization.
31 Increase max size of indexing term from 256 to 512.
35 Fix a bug in zebra_begin_trans where Zebra could mark a transaction
36 complete when it actually failed. This could happen if register files
41 Scan: allow preferred position > number + 1
43 Fix bug #4592: dict_scan misses
45 Fix snippets for complete fields, bug #4590. The
46 snippet_add_complete_fields (chr-system) now takes the indexing term
47 verbatim and makes that the display term.
51 Utility idebra-abs2dom part of distribution and deb+RPM packages.
53 Fix problem with scan and complete subfields (:p) . In some cases,
54 the display scan terms, would be invalid.
58 RPM: Store libraries in /usr/lib64 on 64-bit platforms.
60 Zebra honors SRU sortkeys. If used with YAZ 4.2.0 or later,
61 CQL sortby is honored as well.
63 Fix scan's handling of normalized terms. Some terms were incorrectly
64 ignored in the scan results, CJK terms in particular.
68 Fix bug in dictionary component of Zebra that could lead to
69 inconsistent register assert for large deleteions.
71 Improve estimate hits for phrase searches.
75 Add Debian package for Ubuntu maverick.
77 Update to newer Libxml2/Libxslt for Windows package.
79 Escape-sequences are disabled for RPN-queries with truncation=105/106.
80 These, unlike regular expressions, do not support \-sequences .
84 grs1: avoid term_len < 0 for .abs operator 'range'
86 ICU: fix use of un-init variable.
88 More reasonable approx hits for truncated searches.
90 Updates WRT Odr_int (YAZ 4).
94 Fix hits counts for searches when @attr 12=n is in use. This bug
95 was introduced in 2.0.40.
99 Proximity modifications. The optimized proximity now deals with more
100 cases, ie all prox of the form @prox 0 d 1 1,2,3 k 2 ..
101 That's exlusive==0, any distance, ordered (true), relation (<, =, <=),
104 Fix bug WRT inconsistent registers, bug #3062.
106 Fix install doc rule so it's compatible with Automake 1.11.
108 --- 2.0.41 2009/08/24
110 Fix typos and other errors in the documentation. Thanks to Galen
113 Fix DOM filter so that full record ID (z:id) if given is used verbatim
114 rather than truncated until first blank - in other words use memcpy
115 rather than sscanf. For PI fix record attributes parsing which was
118 --- 2.0.40 2009/07/13
120 Fixed problem in result set handling and segment handling.
122 --- 2.0.39 2009/07/09
124 Improved speed of record fetching by using a different block sizes
127 Added zlib-compression of recd-storage (record data). This is enabled
128 by using recordCompression: zlib in zebra.cfg.
130 Dictionary delete more agressive (dict_del_string).
132 --- 2.0.38 2009/06/19
134 Improved estimated hits for AND operation (bug #2907)
136 Fixed problem with register area growing after "drop DB". (bug #2913).
138 --- 2.0.37 2009/05/18
140 Example with marc21 is now installed and part of packages.
142 Fixed XSL for OAI example.
144 Fixed problem with generic proximity handling.
146 Added new ranking handler: rank-2. This is a combination of rank-1 and
147 static rank.. The staticrank is inverse (lowest value comes first, etc..).
148 rank-2 thus substracts that from score. This allows a system to specify
149 a "base" rank (=staticrank).
151 --- 2.0.36 2009/02/17
153 Added implementation-codes directive for .mar files.
155 Avoid error for missing recordId when zebraidx' option -s is in use.
157 Fixed problem with some disk blocks not beeing freed correctly when
158 'drop database' is used.
160 --- 2.0.34 2008/11/06
162 Facets can now be performed on sort registers (:s), not just regular indexes
163 (:w, :p) etc.. The performance appers to be much better than before.
165 Added a facility to embed metadata for records generated by the DOM XML
166 filter. The meta data step is triggered by a 'process-meta' section in the
167 retrieve section of the dom config, e.g.
168 <retrieve name="myelementset">
169 <xslt stylesheet="my-usual.xsl"/>
172 The meta section substitutes metadata for all elements called 'meta' in
173 namespace http://indexdata.com/zebra-2.0 . The meta element must have
174 exactly one attribute, name, which specifies the special
175 element set name which is equivalent to the suffix to zebra:: for
176 non-embedded special retrievals. This allows snippets, facets etc to be
179 Fixed bug #2202: incorrect sorting order when searching multiple databases.
181 --- 2.0.32 2008/06/06
185 --- 2.0.30 2008/05/14
187 Changed service name for Windows. Bug #978.
189 Fixed the alwaysmatches including _ALLRECORDS for ICU term indexing. Bug #2139.
191 --- 2.0.28 2008/04/03
193 Handle right-truncation for ICU normalized terms.
195 Updated OAI-PMH example.
197 --- 2.0.26 2008/01/28
199 Fixes for searches on Chinese text.
201 Fixes for scan using ICU indexing terms.
203 --- 2.0.24 2008/01/18
205 Fixed some memory violations for ICU enabled indexing that could be
206 triggered by long phrases.
208 Enable ICU on Windows. Thanks to Tumer Garip for makefile patches.
210 Fixed bug #2002: Zebra crashes during merge using ICU indexing.
212 --- 2.0.22 2007/12/20
214 Handle encoding errors for strings passed to ICU. (assert was
217 Added zebraidx command, adelete, which attempts to delete records.
218 It's like 'delete', but does not fail if a record does not exist.
220 --- 2.0.20 2007/12/19
222 Added experimental support for faceted results. This is implemented
223 as a special retrieval on a result set and is triggered using element
224 set name zebra::facet::f1:t1,f2,t2,.. which will return terms and
225 hit counts for field f1, type t1, field f2, type t2, .. The returned
226 data is packed as XML or SUTRS depending on record syntax.
228 Added support for Unicode-based indexing using ICU. This allows
229 locale specific scanning, sorting. The implementation is based on
230 the ICU utility part of YAZ 3.0.16 and later. See the manual as well
231 as the examples/marcxml for an example of the use of ICU.
233 --- 2.0.18 2007/09/19
235 Fixed bug with scan and staticrank enabled. Bug introduced in 2.0.16.
237 --- 2.0.16 2007/09/12
239 Scan now returns a displayTerm which closer to the original word/phrase
240 from original record.
242 Added snippet support. Element set name zebra::snippet will make
243 Zebra return an XML record with snippets (terms around matching terms).
245 Fixed bug #1142: Non-indexed but listed attributes issues diagnostic.
247 Fixed bug #1131: Missing value-of data in DOM filter.
249 Fixed bug #1049: zebra.cfg lines with leading space are ignored.
251 Fixed bug #1128: sortmax not honored.
253 Fixed bug #1121: Crash for some searches with customized string.chr.
255 --- 2.0.14 2007/05/09
257 Zebra uses the YAZ 3 API. Note that this changes the Zebra API
258 as well. The register layout is the same however.
260 Fixed bug #1114: scan within set may use excessive CPU.
262 Fixed memory leak that occurred in scan.
264 Fixed memory leak that occurred for each deleted record.
266 Optimize updates of records where content is almost identical to previous
267 version of record. This makes updating of the internal explain database
270 For RPN queries the index type (w,p,..) may be specified verbatim as
271 structure attribute with string value, e.g. @attr 4=w .
273 Changed record update API . It is now handled by function
274 zebra_record_update which does insert/replace/delete/update of records .
275 This function replaces zebra_record_{insert,delete} and
276 zebra_admin_exchange_record. The DOM filter uses this feature and the
277 @type attribute in record element specifies the action
278 insert/replace/delete/update.
280 Added support for multi-record updates (Bug #944). Based on patch from
283 --- 2.0.12 2007/03/07
285 Fixed bug with indexing of attributes for rec.grs-class of filters.
286 If X-Path was enabled xelm a/@b would be ignored.
288 Fixed bug in register system where 'no more space' messages was produced
289 even though there was plenty of space.
291 Fixed bug #884: Entity declarations in input are lost at retrieval time.
293 Implemented new filter 'dom'. See test/xslt/dom-config*xml for examples.
294 This, like alvis, performs indexing and retrieval using XSLT. But Unlike
295 alvis, it allows multiple XSLT steps to be performed and does ISO2709
298 --- 2.0.10 2007/01/24
300 Staticrank indexing is now an index register type defined in default.idx
301 via directive 'staticrank'. The 'staticrank' directive for grs is no longer
302 supported (was only implemented for Zebra 2.0.8).
304 For searches, allow truncmax value to be controlled with attribute 13.
305 If given, that overrides the value of 'truncmax'.
307 For truncations being limited (abort at truncmax terms), zebrasrv returns
308 SearchResponse with resultSetStatus=subset.
310 For estimated hit counts, zebrasrv returns SearchResponse with
311 resultSetStatus=estimate.
315 For searches, do not truncate more than at most 'truncmax' terms in one
316 single term with right-truncation/regular attribute. By default, 'truncmax',
319 Bug #781: Easier tracking of result sets. We only do this when mkstemp
320 is available (most moderun Unixes supports this). The PID is included
321 in filename : tempdir/zrs_PID_XXXXXX.
323 Added support for specification of staticrank for grs-class of filters.
324 This is enabled by using 'staticrank indexname' in .abs. Contents of
325 indexname (elm ... indexname:w) is then used as value for staticrank.
327 Implemented sorting via the ISAMB system. To enable, use sortindex:i in
330 Added special retrieval support for sort keys. These keys can also be
331 fetched using zebra::index:field:s
333 Added support for specification of approximative limits for whole query.
334 This is specified as attribute type 12. Semantics is the same as
335 estimatehits in zebra.cfg.
339 Fixed bug #736: Updates gets slower.
341 Fixed bug #735: Delete fails to process rest of file after a record that
344 Added RPM package for Zebra 2.
346 Improved indexing speed for large documents.
348 Added support for special utility retrieval zebra:: which offers retrieval
349 of fundamental properties for record, such as indexed terms, raw record,
350 and system record ID.
354 Fixed bug #47: Commit needs to check for roll-back.
356 Fixed bug #672: Trailing characters in password are ignored
358 Added extra presence check for tcl.h, because some systems have
359 tclConfig.sh installed even though Tcl C headers are missing.
361 Optimized melm performance.
363 Do not use sync(2) during commit (but rely on sync'd individual files).
365 Fixed bug in ISAMB's utility function decode_ptr. With gcc -O3 on some
366 platforms the src argument was not updated. Problem was compiler
367 optimization due to strict aliasing rules.
371 Bug fixes: #638, #647, #657, #669.
373 Implement skip of 0 hit counts in scan. Useful for scan operations
374 limited by a result set.
376 Honor position attribute, i.e. allow first-in-field search. To
377 enable this, "firstinfield 1" must be given for an index in
378 default.idx. Enabled in tab/default.idx for w. At this stage
379 first-in field is supported for phrase/and-list/or-list searches
381 Common stream reader interface for record filters (struct ZebraRecStream).
383 Debian package fix: packages idzebra-2.0 + libidzebra-2.0-modules did
384 not depend properly on sub packages.
388 New record filter (record type) 'alvis' which uses XSLT transformations
389 to drive both indexing as well as retrieval. See example configuration in the
390 'example/alvis-oai' directory.
392 'isamb' is now the default ISAM system. In Zebra 1.3, the default ISAM was
393 'isamc'. The type used can still be configured with the 'isam' setting
396 Index structure is now 64-bit based, also on 32 bit systems.
397 There are no more 2GB register file limits.
399 Extended search result tuning. Approximate limit for terms can be enabled
400 and specified with attribute 11. The (approx or exact) hit count is returned
401 as part of the search response as in 1.3 series. The subqueryID of a search
402 term hit count can be specified with attribute 10.
404 Zebra uses string attributes for indexing internally. Using set+numeric
405 use attribute can still be used. This is a search-only conversion which
406 inspects '*.att'-set files as indicated using attset-directives in 'zebra.cfg'.
407 'attset' references are no longer required, but when used they deserve
408 as "check" for that the index names used are also present in '*.att'.
410 Zebra record filters (record type handlers) may be built as loadable
411 modules (.so's) on Unix. In particular the Zebra 2.0 Debian package uses
412 separate packages for each of them. This also means that zebra programs
413 such as zebraidx is no longer depending on Tcl/other..
415 Documentation updates, especially on query structure and syntax, SRU,
416 XSLT support, alvis filter module, and many added examples.
418 Improved logging of the 'zebrasrv' and 'zebraidx' binaries.
420 Improved debian package structure.
422 --- 1.3.16 2004/08/16