3 Fixes for searches on Chinese text.
5 Fixes for scan using ICU indexing terms.
9 Fixed some memory violations for ICU enabled indexing that could be
10 triggered by long phrases.
12 Enable ICU on Windows. Thanks to Tumer Garip for makefile patches.
14 Fixed bug #2002: Zebra crashes during merge using ICU indexing.
18 Handle encoding errors for strings passed to ICU. (assert was
21 Added zebraidx command, adelete, which attempts to delete records.
22 It's like 'delete', but does not fail if a record does not exist.
26 Added experimental support for faceted results. This is implemented
27 as a special retrieval on a result set and is triggered using element
28 set name zebra::facet::f1:t1,f2,t2,.. which will return terms and
29 hit counts for field f1, type t1, field f2, type t2, .. The returned
30 data is packed as XML or SUTRS depending on record syntax.
32 Added support for Unicode-based indexing using ICU. This allows
33 locale specific scanning, sorting. The implementation is based on
34 the ICU utility part of YAZ 3.0.16 and later. See the manual as well
35 as the examples/marcxml for an example of the use of ICU.
39 Fixed bug with scan and staticrank enabled. Bug introduced in 2.0.16.
43 Scan now returns a displayTerm which closer to the original word/phrase
46 Added snippet support. Element set name zebra::snippet will make
47 Zebra return an XML record with snippets (terms around matching terms).
49 Fixed bug #1142: Non-indexed but listed attributes issues diagnostic.
51 Fixed bug #1131: Missing value-of data in DOM filter.
53 Fixed bug #1049: zebra.cfg lines with leading space are ignored.
55 Fixed bug #1128: sortmax not honored.
57 Fixed bug #1121: Crash for some searches with customized string.chr.
61 Zebra uses the YAZ 3 API. Note that this changes the Zebra API
62 as well. The register layout is the same however.
64 Fixed bug #1114: scan within set may use excessive CPU.
66 Fixed memory leak that occurred in scan.
68 Fixed memory leak that occurred for each deleted record.
70 Optimize updates of records where content is almost identical to previous
71 version of record. This makes updating of the internal explain database
74 For RPN queries the index type (w,p,..) may be specified verbatim as
75 structure attribute with string value, e.g. @attr 4=w .
77 Changed record update API . It is now handled by function
78 zebra_record_update which does insert/replace/delete/update of records .
79 This function replaces zebra_record_{insert,delete} and
80 zebra_admin_exchange_record. The DOM filter uses this feature and the
81 @type attribute in record element specifies the action
82 insert/replace/delete/update.
84 Added support for multi-record updates (Bug #944). Based on patch from
89 Fixed bug with indexing of attributes for rec.grs-class of filters.
90 If X-Path was enabled xelm a/@b would be ignored.
92 Fixed bug in register system where 'no more space' messages was produced
93 even though there was plenty of space.
95 Fixed bug #884: Entity declarations in input are lost at retrieval time.
97 Implemented new filter 'dom'. See test/xslt/dom-config*xml for examples.
98 This, like alvis, performs indexing and retrieval using XSLT. But Unlike
99 alvis, it allows multiple XSLT steps to be performed and does ISO2709
102 --- 2.0.10 2007/01/24
104 Staticrank indexing is now an index register type defined in default.idx
105 via directive 'staticrank'. The 'staticrank' directive for grs is no longer
106 supported (was only implemented for Zebra 2.0.8).
108 For searches, allow truncmax value to be controlled with attribute 13.
109 If given, that overrides the value of 'truncmax'.
111 For truncations being limited (abort at truncmax terms), zebrasrv returns
112 SearchResponse with resultSetStatus=subset.
114 For estimated hit counts, zebrasrv returns SearchResponse with
115 resultSetStatus=estimate.
119 For searches, do not truncate more than at most 'truncmax' terms in one
120 single term with right-truncation/regular attribute. By default, 'truncmax',
123 Bug #781: Easier tracking of result sets. We only do this when mkstemp
124 is available (most moderun Unixes supports this). The PID is included
125 in filename : tempdir/zrs_PID_XXXXXX.
127 Added support for specification of staticrank for grs-class of filters.
128 This is enabled by using 'staticrank indexname' in .abs. Contents of
129 indexname (elm ... indexname:w) is then used as value for staticrank.
131 Implemented sorting via the ISAMB system. To enable, use sortindex:i in
134 Added special retrieval support for sort keys. These keys can also be
135 fetched using zebra::index:field:s
137 Added support for specification of approximative limits for whole query.
138 This is specified as attribute type 12. Semantics is the same as
139 estimatehits in zebra.cfg.
143 Fixed bug #736: Updates gets slower.
145 Fixed bug #735: Delete fails to process rest of file after a record that
148 Added RPM package for Zebra 2.
150 Improved indexing speed for large documents.
152 Added support for special utility retrieval zebra:: which offers retrieval
153 of fundamental properties for record, such as indexed terms, raw record,
154 and system record ID.
158 Fixed bug #47: Commit needs to check for roll-back.
160 Fixed bug #672: Trailing characters in password are ignored
162 Added extra presence check for tcl.h, because some systems have
163 tclConfig.sh installed even though Tcl C headers are missing.
165 Optimized melm performance.
167 Do not use sync(2) during commit (but rely on sync'd individual files).
169 Fixed bug in ISAMB's utility function decode_ptr. With gcc -O3 on some
170 platforms the src argument was not updated. Problem was compiler
171 optimization due to strict aliasing rules.
175 Bug fixes: #638, #647, #657, #669.
177 Implement skip of 0 hit counts in scan. Useful for scan operations
178 limited by a result set.
180 Honor position attribute, i.e. allow first-in-field search. To
181 enable this, "firstinfield 1" must be given for an index in
182 default.idx. Enabled in tab/default.idx for w. At this stage
183 first-in field is supported for phrase/and-list/or-list searches
185 Common stream reader interface for record filters (struct ZebraRecStream).
187 Debian package fix: packages idzebra-2.0 + libidzebra-2.0-modules did
188 not depend properly on sub packages.
192 New record filter (record type) 'alvis' which uses XSLT transformations
193 to drive both indexing as well as retrieval. See example configuration in the
194 'example/alvis-oai' directory.
196 'isamb' is now the default ISAM system. In Zebra 1.3, the default ISAM was
197 'isamc'. The type used can still be configured with the 'isam' setting
200 Index structure is now 64-bit based, also on 32 bit systems.
201 There are no more 2GB register file limits.
203 Extended search result tuning. Approximate limit for terms can be enabled
204 and specified with attribute 11. The (approx or exact) hit count is returned
205 as part of the search response as in 1.3 series. The subqueryID of a search
206 term hit count can be specified with attribute 10.
208 Zebra uses string attributes for indexing internally. Using set+numeric
209 use attribute can still be used. This is a search-only conversion which
210 inspects '*.att'-set files as indicated using attset-directives in 'zebra.cfg'.
211 'attset' references are no longer required, but when used they deserve
212 as "check" for that the index names used are also present in '*.att'.
214 Zebra record filters (record type handlers) may be built as loadable
215 modules (.so's) on Unix. In particular the Zebra 2.0 Debian package uses
216 separate packages for each of them. This also means that zebra programs
217 such as zebraidx is no longer depending on Tcl/other..
219 Documentation updates, especially on query structure and syntax, SRU,
220 XSLT support, alvis filter module, and many added examples.
222 Improved logging of the 'zebrasrv' and 'zebraidx' binaries.
224 Improved debian package structure.
226 --- 1.3.16 2004/08/16