1 grs1: avoid term_len < 0 for .abs operator 'range'
3 ICU: fix use of un-init variable.
5 More reasonable approx hits for truncated searches.
7 Updates WRT Odr_int (YAZ 4).
11 Fix hits counts for searches when @attr 12=n is in use. This bug
12 was introduced in 2.0.40.
16 Proximity modifications. The optimized proximity now deals with more
17 cases, ie all prox of the form @prox 0 d 1 1,2,3 k 2 ..
18 That's exlusive==0, any distance, ordered (true), relation (<, =, <=),
21 Fix bug WRT inconsistent registers, bug #3062.
23 Fix install doc rule so it's compatible with Automake 1.11.
27 Fix typos and other errors in the documentation. Thanks to Galen
30 Fix DOM filter so that full record ID (z:id) if given is used verbatim
31 rather than truncated until first blank - in other words use memcpy
32 rather than sscanf. For PI fix record attributes parsing which was
37 Fixed problem in result set handling and segment handling.
41 Improved speed of record fetching by using a different block sizes
44 Added zlib-compression of recd-storage (record data). This is enabled
45 by using recordCompression: zlib in zebra.cfg.
47 Dictionary delete more agressive (dict_del_string).
51 Improved estimated hits for AND operation (bug #2907)
53 Fixed problem with register area growing after "drop DB". (bug #2913).
57 Example with marc21 is now installed and part of packages.
59 Fixed XSL for OAI example.
61 Fixed problem with generic proximity handling.
63 Added new ranking handler: rank-2. This is a combination of rank-1 and
64 static rank.. The staticrank is inverse (lowest value comes first, etc..).
65 rank-2 thus substracts that from score. This allows a system to specify
66 a "base" rank (=staticrank).
70 Added implementation-codes directive for .mar files.
72 Avoid error for missing recordId when zebraidx' option -s is in use.
74 Fixed problem with some disk blocks not beeing freed correctly when
75 'drop database' is used.
79 Facets can now be performed on sort registers (:s), not just regular indexes
80 (:w, :p) etc.. The performance appers to be much better than before.
82 Added a facility to embed metadata for records generated by the DOM XML
83 filter. The meta data step is triggered by a 'process-meta' section in the
84 retrieve section of the dom config, e.g.
85 <retrieve name="myelementset">
86 <xslt stylesheet="my-usual.xsl"/>
89 The meta section substitutes metadata for all elements called 'meta' in
90 namespace http://indexdata.com/zebra-2.0 . The meta element must have
91 exactly one attribute, name, which specifies the special
92 element set name which is equivalent to the suffix to zebra:: for
93 non-embedded special retrievals. This allows snippets, facets etc to be
96 Fixed bug #2202: incorrect sorting order when searching multiple databases.
102 --- 2.0.30 2008/05/14
104 Changed service name for Windows. Bug #978.
106 Fixed the alwaysmatches including _ALLRECORDS for ICU term indexing. Bug #2139.
108 --- 2.0.28 2008/04/03
110 Handle right-truncation for ICU normalized terms.
112 Updated OAI-PMH example.
114 --- 2.0.26 2008/01/28
116 Fixes for searches on Chinese text.
118 Fixes for scan using ICU indexing terms.
120 --- 2.0.24 2008/01/18
122 Fixed some memory violations for ICU enabled indexing that could be
123 triggered by long phrases.
125 Enable ICU on Windows. Thanks to Tumer Garip for makefile patches.
127 Fixed bug #2002: Zebra crashes during merge using ICU indexing.
129 --- 2.0.22 2007/12/20
131 Handle encoding errors for strings passed to ICU. (assert was
134 Added zebraidx command, adelete, which attempts to delete records.
135 It's like 'delete', but does not fail if a record does not exist.
137 --- 2.0.20 2007/12/19
139 Added experimental support for faceted results. This is implemented
140 as a special retrieval on a result set and is triggered using element
141 set name zebra::facet::f1:t1,f2,t2,.. which will return terms and
142 hit counts for field f1, type t1, field f2, type t2, .. The returned
143 data is packed as XML or SUTRS depending on record syntax.
145 Added support for Unicode-based indexing using ICU. This allows
146 locale specific scanning, sorting. The implementation is based on
147 the ICU utility part of YAZ 3.0.16 and later. See the manual as well
148 as the examples/marcxml for an example of the use of ICU.
150 --- 2.0.18 2007/09/19
152 Fixed bug with scan and staticrank enabled. Bug introduced in 2.0.16.
154 --- 2.0.16 2007/09/12
156 Scan now returns a displayTerm which closer to the original word/phrase
157 from original record.
159 Added snippet support. Element set name zebra::snippet will make
160 Zebra return an XML record with snippets (terms around matching terms).
162 Fixed bug #1142: Non-indexed but listed attributes issues diagnostic.
164 Fixed bug #1131: Missing value-of data in DOM filter.
166 Fixed bug #1049: zebra.cfg lines with leading space are ignored.
168 Fixed bug #1128: sortmax not honored.
170 Fixed bug #1121: Crash for some searches with customized string.chr.
172 --- 2.0.14 2007/05/09
174 Zebra uses the YAZ 3 API. Note that this changes the Zebra API
175 as well. The register layout is the same however.
177 Fixed bug #1114: scan within set may use excessive CPU.
179 Fixed memory leak that occurred in scan.
181 Fixed memory leak that occurred for each deleted record.
183 Optimize updates of records where content is almost identical to previous
184 version of record. This makes updating of the internal explain database
187 For RPN queries the index type (w,p,..) may be specified verbatim as
188 structure attribute with string value, e.g. @attr 4=w .
190 Changed record update API . It is now handled by function
191 zebra_record_update which does insert/replace/delete/update of records .
192 This function replaces zebra_record_{insert,delete} and
193 zebra_admin_exchange_record. The DOM filter uses this feature and the
194 @type attribute in record element specifies the action
195 insert/replace/delete/update.
197 Added support for multi-record updates (Bug #944). Based on patch from
200 --- 2.0.12 2007/03/07
202 Fixed bug with indexing of attributes for rec.grs-class of filters.
203 If X-Path was enabled xelm a/@b would be ignored.
205 Fixed bug in register system where 'no more space' messages was produced
206 even though there was plenty of space.
208 Fixed bug #884: Entity declarations in input are lost at retrieval time.
210 Implemented new filter 'dom'. See test/xslt/dom-config*xml for examples.
211 This, like alvis, performs indexing and retrieval using XSLT. But Unlike
212 alvis, it allows multiple XSLT steps to be performed and does ISO2709
215 --- 2.0.10 2007/01/24
217 Staticrank indexing is now an index register type defined in default.idx
218 via directive 'staticrank'. The 'staticrank' directive for grs is no longer
219 supported (was only implemented for Zebra 2.0.8).
221 For searches, allow truncmax value to be controlled with attribute 13.
222 If given, that overrides the value of 'truncmax'.
224 For truncations being limited (abort at truncmax terms), zebrasrv returns
225 SearchResponse with resultSetStatus=subset.
227 For estimated hit counts, zebrasrv returns SearchResponse with
228 resultSetStatus=estimate.
232 For searches, do not truncate more than at most 'truncmax' terms in one
233 single term with right-truncation/regular attribute. By default, 'truncmax',
236 Bug #781: Easier tracking of result sets. We only do this when mkstemp
237 is available (most moderun Unixes supports this). The PID is included
238 in filename : tempdir/zrs_PID_XXXXXX.
240 Added support for specification of staticrank for grs-class of filters.
241 This is enabled by using 'staticrank indexname' in .abs. Contents of
242 indexname (elm ... indexname:w) is then used as value for staticrank.
244 Implemented sorting via the ISAMB system. To enable, use sortindex:i in
247 Added special retrieval support for sort keys. These keys can also be
248 fetched using zebra::index:field:s
250 Added support for specification of approximative limits for whole query.
251 This is specified as attribute type 12. Semantics is the same as
252 estimatehits in zebra.cfg.
256 Fixed bug #736: Updates gets slower.
258 Fixed bug #735: Delete fails to process rest of file after a record that
261 Added RPM package for Zebra 2.
263 Improved indexing speed for large documents.
265 Added support for special utility retrieval zebra:: which offers retrieval
266 of fundamental properties for record, such as indexed terms, raw record,
267 and system record ID.
271 Fixed bug #47: Commit needs to check for roll-back.
273 Fixed bug #672: Trailing characters in password are ignored
275 Added extra presence check for tcl.h, because some systems have
276 tclConfig.sh installed even though Tcl C headers are missing.
278 Optimized melm performance.
280 Do not use sync(2) during commit (but rely on sync'd individual files).
282 Fixed bug in ISAMB's utility function decode_ptr. With gcc -O3 on some
283 platforms the src argument was not updated. Problem was compiler
284 optimization due to strict aliasing rules.
288 Bug fixes: #638, #647, #657, #669.
290 Implement skip of 0 hit counts in scan. Useful for scan operations
291 limited by a result set.
293 Honor position attribute, i.e. allow first-in-field search. To
294 enable this, "firstinfield 1" must be given for an index in
295 default.idx. Enabled in tab/default.idx for w. At this stage
296 first-in field is supported for phrase/and-list/or-list searches
298 Common stream reader interface for record filters (struct ZebraRecStream).
300 Debian package fix: packages idzebra-2.0 + libidzebra-2.0-modules did
301 not depend properly on sub packages.
305 New record filter (record type) 'alvis' which uses XSLT transformations
306 to drive both indexing as well as retrieval. See example configuration in the
307 'example/alvis-oai' directory.
309 'isamb' is now the default ISAM system. In Zebra 1.3, the default ISAM was
310 'isamc'. The type used can still be configured with the 'isam' setting
313 Index structure is now 64-bit based, also on 32 bit systems.
314 There are no more 2GB register file limits.
316 Extended search result tuning. Approximate limit for terms can be enabled
317 and specified with attribute 11. The (approx or exact) hit count is returned
318 as part of the search response as in 1.3 series. The subqueryID of a search
319 term hit count can be specified with attribute 10.
321 Zebra uses string attributes for indexing internally. Using set+numeric
322 use attribute can still be used. This is a search-only conversion which
323 inspects '*.att'-set files as indicated using attset-directives in 'zebra.cfg'.
324 'attset' references are no longer required, but when used they deserve
325 as "check" for that the index names used are also present in '*.att'.
327 Zebra record filters (record type handlers) may be built as loadable
328 modules (.so's) on Unix. In particular the Zebra 2.0 Debian package uses
329 separate packages for each of them. This also means that zebra programs
330 such as zebraidx is no longer depending on Tcl/other..
332 Documentation updates, especially on query structure and syntax, SRU,
333 XSLT support, alvis filter module, and many added examples.
335 Improved logging of the 'zebrasrv' and 'zebraidx' binaries.
337 Improved debian package structure.
339 --- 1.3.16 2004/08/16