3 Fixed bug with scan and staticrank enabled. Bug introduced in 2.0.16.
7 Scan now returns a displayTerm which closer to the original word/phrase
10 Added snippet support. Element set name zebra::snippet will make
11 Zebra return an XML record with snippets (terms around matching terms).
13 Fixed bug #1142: Non-indexed but listed attributes issues diagnostic.
15 Fixed bug #1131: Missing value-of data in DOM filter.
17 Fixed bug #1049: zebra.cfg lines with leading space are ignored.
19 Fixed bug #1128: sortmax not honored.
21 Fixed bug #1121: Crash for some searches with customized string.chr.
25 Zebra uses the YAZ 3 API. Note that this changes the Zebra API
26 as well. The register layout is the same however.
28 Fixed bug #1114: scan within set may use excessive CPU.
30 Fixed memory leak that occurred in scan.
32 Fixed memory leak that occurred for each deleted record.
34 Optimize updates of records where content is almost identical to previous
35 version of record. This makes updating of the internal explain database
38 For RPN queries the index type (w,p,..) may be specified verbatim as
39 structure attribute with string value, e.g. @attr 4=w .
41 Changed record update API . It is now handled by function
42 zebra_record_update which does insert/replace/delete/update of records .
43 This function replaces zebra_record_{insert,delete} and
44 zebra_admin_exchange_record. The DOM filter uses this feature and the
45 @type attribute in record element specifies the action
46 insert/replace/delete/update.
48 Added support for multi-record updates (Bug #944). Based on patch from
53 Fixed bug with indexing of attributes for rec.grs-class of filters.
54 If X-Path was enabled xelm a/@b would be ignored.
56 Fixed bug in register system where 'no more space' messages was produced
57 even though there was plenty of space.
59 Fixed bug #884: Entity declarations in input are lost at retrieval time.
61 Implemented new filter 'dom'. See test/xslt/dom-config*xml for examples.
62 This, like alvis, performs indexing and retrieval using XSLT. But Unlike
63 alvis, it allows multiple XSLT steps to be performed and does ISO2709
68 Staticrank indexing is now an index register type defined in default.idx
69 via directive 'staticrank'. The 'staticrank' directive for grs is no longer
70 supported (was only implemented for Zebra 2.0.8).
72 For searches, allow truncmax value to be controlled with attribute 13.
73 If given, that overrides the value of 'truncmax'.
75 For truncations being limited (abort at truncmax terms), zebrasrv returns
76 SearchResponse with resultSetStatus=subset.
78 For estimated hit counts, zebrasrv returns SearchResponse with
79 resultSetStatus=estimate.
83 For searches, do not truncate more than at most 'truncmax' terms in one
84 single term with right-truncation/regular attribute. By default, 'truncmax',
87 Bug #781: Easier tracking of result sets. We only do this when mkstemp
88 is available (most moderun Unixes supports this). The PID is included
89 in filename : tempdir/zrs_PID_XXXXXX.
91 Added support for specification of staticrank for grs-class of filters.
92 This is enabled by using 'staticrank indexname' in .abs. Contents of
93 indexname (elm ... indexname:w) is then used as value for staticrank.
95 Implemented sorting via the ISAMB system. To enable, use sortindex:i in
98 Added special retrieval support for sort keys. These keys can also be
99 fetched using zebra::index:field:s
101 Added support for specification of approximative limits for whole query.
102 This is specified as attribute type 12. Semantics is the same as
103 estimatehits in zebra.cfg.
107 Fixed bug #736: Updates gets slower.
109 Fixed bug #735: Delete fails to process rest of file after a record that
112 Added RPM package for Zebra 2.
114 Improved indexing speed for large documents.
116 Added support for special utility retrieval zebra:: which offers retrieval
117 of fundamental properties for record, such as indexed terms, raw record,
118 and system record ID.
122 Fixed bug #47: Commit needs to check for roll-back.
124 Fixed bug #672: Trailing characters in password are ignored
126 Added extra presence check for tcl.h, because some systems have
127 tclConfig.sh installed even though Tcl C headers are missing.
129 Optimized melm performance.
131 Do not use sync(2) during commit (but rely on sync'd individual files).
133 Fixed bug in ISAMB's utility function decode_ptr. With gcc -O3 on some
134 platforms the src argument was not updated. Problem was compiler
135 optimization due to strict aliasing rules.
139 Bug fixes: #638, #647, #657, #669.
141 Implement skip of 0 hit counts in scan. Useful for scan operations
142 limited by a result set.
144 Honor position attribute, i.e. allow first-in-field search. To
145 enable this, "firstinfield 1" must be given for an index in
146 default.idx. Enabled in tab/default.idx for w. At this stage
147 first-in field is supported for phrase/and-list/or-list searches
149 Common stream reader interface for record filters (struct ZebraRecStream).
151 Debian package fix: packages idzebra-2.0 + libidzebra-2.0-modules did
152 not depend properly on sub packages.
156 New record filter (record type) 'alvis' which uses XSLT transformations
157 to drive both indexing as well as retrieval. See example configuration in the
158 'example/alvis-oai' directory.
160 'isamb' is now the default ISAM system. In Zebra 1.3, the default ISAM was
161 'isamc'. The type used can still be configured with the 'isam' setting
164 Index structure is now 64-bit based, also on 32 bit systems.
165 There are no more 2GB register file limits.
167 Extended search result tuning. Approximate limit for terms can be enabled
168 and specified with attribute 11. The (approx or exact) hit count is returned
169 as part of the search response as in 1.3 series. The subqueryID of a search
170 term hit count can be specified with attribute 10.
172 Zebra uses string attributes for indexing internally. Using set+numeric
173 use attribute can still be used. This is a search-only conversion which
174 inspects '*.att'-set files as indicated using attset-directives in 'zebra.cfg'.
175 'attset' references are no longer required, but when used they deserve
176 as "check" for that the index names used are also present in '*.att'.
178 Zebra record filters (record type handlers) may be built as loadable
179 modules (.so's) on Unix. In particular the Zebra 2.0 Debian package uses
180 separate packages for each of them. This also means that zebra programs
181 such as zebraidx is no longer depending on Tcl/other..
183 Documentation updates, especially on query structure and syntax, SRU,
184 XSLT support, alvis filter module, and many added examples.
186 Improved logging of the 'zebrasrv' and 'zebraidx' binaries.
188 Improved debian package structure.
190 --- 1.3.16 2004/08/16