1 Fixed bug with scan and staticrank enabled. Bug introduced in 2.0.16.
5 Scan now returns a displayTerm which closer to the original word/phrase
8 Added snippet support. Element set name zebra::snippet will make
9 Zebra return an XML record with snippets (terms around matching terms).
11 Fixed bug #1142: Non-indexed but listed attributes issues diagnostic.
13 Fixed bug #1131: Missing value-of data in DOM filter.
15 Fixed bug #1049: zebra.cfg lines with leading space are ignored.
17 Fixed bug #1128: sortmax not honored.
19 Fixed bug #1121: Crash for some searches with customized string.chr.
23 Zebra uses the YAZ 3 API. Note that this changes the Zebra API
24 as well. The register layout is the same however.
26 Fixed bug #1114: scan within set may use excessive CPU.
28 Fixed memory leak that occurred in scan.
30 Fixed memory leak that occurred for each deleted record.
32 Optimize updates of records where content is almost identical to previous
33 version of record. This makes updating of the internal explain database
36 For RPN queries the index type (w,p,..) may be specified verbatim as
37 structure attribute with string value, e.g. @attr 4=w .
39 Changed record update API . It is now handled by function
40 zebra_record_update which does insert/replace/delete/update of records .
41 This function replaces zebra_record_{insert,delete} and
42 zebra_admin_exchange_record. The DOM filter uses this feature and the
43 @type attribute in record element specifies the action
44 insert/replace/delete/update.
46 Added support for multi-record updates (Bug #944). Based on patch from
51 Fixed bug with indexing of attributes for rec.grs-class of filters.
52 If X-Path was enabled xelm a/@b would be ignored.
54 Fixed bug in register system where 'no more space' messages was produced
55 even though there was plenty of space.
57 Fixed bug #884: Entity declarations in input are lost at retrieval time.
59 Implemented new filter 'dom'. See test/xslt/dom-config*xml for examples.
60 This, like alvis, performs indexing and retrieval using XSLT. But Unlike
61 alvis, it allows multiple XSLT steps to be performed and does ISO2709
66 Staticrank indexing is now an index register type defined in default.idx
67 via directive 'staticrank'. The 'staticrank' directive for grs is no longer
68 supported (was only implemented for Zebra 2.0.8).
70 For searches, allow truncmax value to be controlled with attribute 13.
71 If given, that overrides the value of 'truncmax'.
73 For truncations being limited (abort at truncmax terms), zebrasrv returns
74 SearchResponse with resultSetStatus=subset.
76 For estimated hit counts, zebrasrv returns SearchResponse with
77 resultSetStatus=estimate.
81 For searches, do not truncate more than at most 'truncmax' terms in one
82 single term with right-truncation/regular attribute. By default, 'truncmax',
85 Bug #781: Easier tracking of result sets. We only do this when mkstemp
86 is available (most moderun Unixes supports this). The PID is included
87 in filename : tempdir/zrs_PID_XXXXXX.
89 Added support for specification of staticrank for grs-class of filters.
90 This is enabled by using 'staticrank indexname' in .abs. Contents of
91 indexname (elm ... indexname:w) is then used as value for staticrank.
93 Implemented sorting via the ISAMB system. To enable, use sortindex:i in
96 Added special retrieval support for sort keys. These keys can also be
97 fetched using zebra::index:field:s
99 Added support for specification of approximative limits for whole query.
100 This is specified as attribute type 12. Semantics is the same as
101 estimatehits in zebra.cfg.
105 Fixed bug #736: Updates gets slower.
107 Fixed bug #735: Delete fails to process rest of file after a record that
110 Added RPM package for Zebra 2.
112 Improved indexing speed for large documents.
114 Added support for special utility retrieval zebra:: which offers retrieval
115 of fundamental properties for record, such as indexed terms, raw record,
116 and system record ID.
120 Fixed bug #47: Commit needs to check for roll-back.
122 Fixed bug #672: Trailing characters in password are ignored
124 Added extra presence check for tcl.h, because some systems have
125 tclConfig.sh installed even though Tcl C headers are missing.
127 Optimized melm performance.
129 Do not use sync(2) during commit (but rely on sync'd individual files).
131 Fixed bug in ISAMB's utility function decode_ptr. With gcc -O3 on some
132 platforms the src argument was not updated. Problem was compiler
133 optimization due to strict aliasing rules.
137 Bug fixes: #638, #647, #657, #669.
139 Implement skip of 0 hit counts in scan. Useful for scan operations
140 limited by a result set.
142 Honor position attribute, i.e. allow first-in-field search. To
143 enable this, "firstinfield 1" must be given for an index in
144 default.idx. Enabled in tab/default.idx for w. At this stage
145 first-in field is supported for phrase/and-list/or-list searches
147 Common stream reader interface for record filters (struct ZebraRecStream).
149 Debian package fix: packages idzebra-2.0 + libidzebra-2.0-modules did
150 not depend properly on sub packages.
154 New record filter (record type) 'alvis' which uses XSLT transformations
155 to drive both indexing as well as retrieval. See example configuration in the
156 'example/alvis-oai' directory.
158 'isamb' is now the default ISAM system. In Zebra 1.3, the default ISAM was
159 'isamc'. The type used can still be configured with the 'isam' setting
162 Index structure is now 64-bit based, also on 32 bit systems.
163 There are no more 2GB register file limits.
165 Extended search result tuning. Approximate limit for terms can be enabled
166 and specified with attribute 11. The (approx or exact) hit count is returned
167 as part of the search response as in 1.3 series. The subqueryID of a search
168 term hit count can be specified with attribute 10.
170 Zebra uses string attributes for indexing internally. Using set+numeric
171 use attribute can still be used. This is a search-only conversion which
172 inspects '*.att'-set files as indicated using attset-directives in 'zebra.cfg'.
173 'attset' references are no longer required, but when used they deserve
174 as "check" for that the index names used are also present in '*.att'.
176 Zebra record filters (record type handlers) may be built as loadable
177 modules (.so's) on Unix. In particular the Zebra 2.0 Debian package uses
178 separate packages for each of them. This also means that zebra programs
179 such as zebraidx is no longer depending on Tcl/other..
181 Documentation updates, especially on query structure and syntax, SRU,
182 XSLT support, alvis filter module, and many added examples.
184 Improved logging of the 'zebrasrv' and 'zebraidx' binaries.
186 Improved debian package structure.
188 --- 1.3.16 2004/08/16