4 Fix MARC transfer . MARC fields had wrong data for multiple fields.
6 XML record reader moved from YAZ to Zebra, to make YAZ less
7 dependant on external libraries.
9 Zebra uses yaz_iconv which is mini iconv library supporting UTF-8,
10 UCS4, ISO-8859-1. This means that Zebra does UNICODE even
11 on systems that doesn't offer iconv.
13 XML record reader supports external system entities.
17 New .abs-directive "xpath" that takes one argument: "enable"
18 or "disable" to enable and disable XPath -indexing. If no "xpath"
19 direcive is found in .abs-file , XPath-indexing is disabled to ensure
20 backwards compatibility. For missing .abs-files XPath-indexing is
21 enabled so that such records are searchable.
23 Zebra warns about missing .abs-file only once (for each type).
25 Fixed a bug in file update where already-inserted files could
30 Zebra license changed to GNU GPL.
32 XPath-like queries used when RPN string attributes are used, eg.
33 @attr 1=/portal/title sometitle
34 @attr 1=/portal/title[@xml:lang=da] danishtitle
35 @attr 1=/portal/title/@xml:lang da
36 @attr 1=//title sometitle
38 Zebra uses UTF-8 internally:
39 1) New setting "encoding" for zebra.cfg that specifies encoding for
40 OCTET terms in queries and record encoding for most transfer syntaxes
41 (except those that use International Strings, such as GRS-1).
42 2) The encoding of International strings is UTF-8 by default. It
43 may be changed by character set negotiation. If character set
44 negotiation is in effect and if records are selected for conversion
45 these'll be converted to the selected character set - thus overriding
46 the encoding setting in zebra.cfg.
47 3) New directive "encoding" in .abs-files. This specifies the external
48 character encoding for files indexed by zebra. However, if records
49 themselves have an XML header that specifies and encoding that'll be used
52 XML filter (-t grs.xml).
54 Multiple registers. New setting in resource 'root' that holds base
55 directory for register(s). A group a databases may be put in separate
56 register in directory root/reg by using db name 'reg/db1' ... 'reg/dbN'.
60 Fixes for Digital Unix
62 Implemented hits per term using USR:SearchResult-1.
64 New Zebra API. Locking system re-implemented.
66 --- 1.1.stable 2002/02/20
68 Rank weight can be controlled with attribute type 9. Default
69 value is 34. Recommended values between 1-36.
73 Updated for YAZ version 1.8.
75 Added support for termsets - a result set of terms matching
76 a given query. For @attr 8=<set> creates termset named <set>.
78 Added support for raw retrieval. Element Set Name R forces the
79 text filter which returns the record in its original form.
81 Added numerical sort - triggered by structure=numeric (4=109).
83 Remote record import using Z39.50 Extended Services and Segments.
85 Fixed bug where updating a database with user-defined attributes
86 could corrupt the register (bad storeKeys).
88 Multi-threaded version.
90 Fixed bug regarding proximity.
92 Documentation updates.
94 Fixed bug in record retrieval module that occured on 64-bit OSF
99 Fixed bug in makefile for WIN32.
101 Fixed bug in configure script - used bash-specific features.
105 Added support for multiple records in one file for filter grs.sgml.
107 Changed record index structure. New layout is incompatible with
108 previous releases. Added setting "recordcompression" to control
109 compression of records. Possible values are "none" (no
110 compression) and bzip2 (compression using libbz2).
112 Added XML transfer syntax support for retrieval of structured records.
113 Schema in CompSpec is recognised in retrieval of structured records.
115 Changed Tcl record filter so that it attemps to read <filt>.tflt. If
116 that fails, the filter reads the file <filt>.flt (regx style filter).
118 Implemented new Tcl record filter - use grs.tcl.<filter> to enable it.
119 Zebra's configure script automatically attempts to locate Tcl. For
120 manual Tcl configuration use option --with-tclconfig=<path> to specify
121 where Tcl's library files are located.
123 Implemented "compression" of Dictionary and ISAM system. Dictionary
126 Added "tagsysno" directive to zebra.cfg to control under which tag the
127 system ID is placed. Use tagsysno: 0 to disable Zebra's system number
130 Added "tagrank" as above.
132 Changed file naming scheme for register files from <name>.mf.<no> to
135 Implemented "position"-flag for register type (as defined in
136 default.idx). When set to zero no position (or seqence number) is
137 saved in register for each word occurrence, thus saving some register
140 Implemented database mapping. Using mapdb one can specify a database
141 to be mapped to one or more physical databases. Usage:
142 mapdb <fromdb> <todb> ..
144 Added SOIF-filter. Thanks to Peter Valkenburg.
146 For the regx-filter "end element -record" may trigger a mark-of-record
147 if outer level is reached.
149 Tag sets may be typed in the reference to it. From the .abs-file the
150 "tagset" directive takes a third optional integer type for the tag set
151 referenced. From a .tag-file the "include" directive takes a third
152 optional type as well. The old "type" directive in the tag set itself
153 is still recognized but acts as the default type for the tag set.
155 Zebra supports the specification of arbitrary attributes sets, schemas
156 and tag sets, because of the change in YAZ' OID management system.
158 Fixed bug in Sort that caused it NOT to use character mapping as it
161 Zebra now uses GNU configure to generate Makefile(s).
163 Added un-optimised support for left and left/right truncation attributes.
165 Added support for relational operators on text when using RPN queries.
167 Added support for sort specifications in RPN queries. Type 7 specifies
168 'sort' where value 1=ascending, value 2=descending. The use attribute
169 specifies the field criteria as usual. The term specifies priority
170 where 0=first, 1=second, ...
172 Changed the way use attributes are specified in the recordId
175 Maximum number of databases in one Zebra register increased.
177 New setting, databasePath, which specifies that first directory during
178 update traversal is the database name (instead of a fixed one).
180 New setting, explainDatabase, which specifies that databases are
183 Modified Zebra so that it works with ASN.1 compiled code for YAZ.
185 Implemented EXPLAIN database maintenance. Zebra automatically
186 generate - and update CategoryList, TargetInfo, DatabaseInfo,
187 AttributeSetInfo and AttributeDetails records at this stage. The
188 records may be transferred as GRS-1, SUTRS or Explain.
190 Fixed register spec so that colon isn't treated as size separator
191 unless followed by [0-9+-] in order to allow DOS drive specifications.
193 Fixed two bugs in ISAMC system.
195 Changed the way Zebra keeps its maintenance information about attribute
196 sets, available attributes, etc.. Records in "SGML" notation using an
197 EXPLAIN schema is now used when appropriate.
199 Bug fix: Index didn't handle update/insert/delete of the same record
200 (i.e. same recordId) in one run (one invocation of zebraidx). Only the
201 first occurence of a record is considered.
203 Most searches now return correct number of hits.
205 New modular ranking system. Interested programmers are encouraged to
206 inspect rank1.c and improve the algorithm.
208 Bug fix: Lock files weren't removed as they should on NT.
210 Implemented Z39.50 Sort. Zebra's sort handler uses use attributes to
211 specify a "sort register". Refer to the gils sample records which refer
212 to index type "s" which is specified as "sort" in the default.idx file.
213 Each sort criteria can either be Ascending or Descending and at most
214 three sort elements can be specified.
216 Bug fix: Character mapping didn't work for text files.
220 Simple ranked searches now return correct number of hits.
222 The test option (-s) only makes a read-lock on the index as well
223 as using read-only operations anywhere.
225 Moved towards generic character mapping. Configuration file default.idx
226 specifies character map files for register types w, p, u, etc.
228 Implemented "begin variant" for the sgml.regx - filter.
230 Fixed a few memory leaks.
232 Added support for C++, headers uses extern "C" for public definitions.
234 Bug fix: The show records facility (-s) only displayed information for
235 the first record in a file (and not for every record in the file).
237 Added option "-f <n>" to limit the logging of record operations. After
238 <n> records has been processed no logging is performed (unless errors
241 Bug fix: the compressed ISAM system didn't handle update operations
244 Added setting, "maxResultSetSize", to hold the number of records to
245 save in a result set.
247 Bug fix: Complete phrase did't work for search operations.
249 Bug fix: temporary result sets weren't deleted.
251 Reduced disk space for saved keys (storeKeys = 1).
253 Added optional, physical ANY (key replication)
255 Implemented proximity operator in search.
257 Bug fix: the path name buffers used by file match traversal routines
258 have been extended to support long file names.
260 New C(ompressed) ISAM system. To enable it, specify "isam: c" in the
261 configuration file. The resulting register without "storeKeys" is about
262 half the size, and the memory used by zebraidx during phase 2 (merge) is
263 reduced to a minimum.
265 Reworked the way Regexp-2 queries with error tolerance are handled and
266 specified. The documentation has been updated accordingly.
268 Bug fix: Zebrasrv didn't search correctly when queries contained masking
269 characters. This bug was introduced in 1.0a8.
271 Zebrasrv now tag records with the proper database name.
273 New settings, memMax and keyTmpDir.
275 Changed name of setting lockDir (previously called lockPath) and
276 setTmpDir (previously called tempSetPath).
278 Generalized and changed record type specifications. In short, there are:
280 grs.sgml structured, "SGML-like" syntax
281 grs.regx.<filter> structured, Regular expression filter
282 grs.marc.<abs> Reads *MARC records in the ISO2709 format. <abs>
283 is the name of an abstract syntax file.
284 Bug fix: Result sets weren't sorted in operations involving boolean
285 operations with "ranked" operands.
289 Added national character-handling subsystem.
293 Small modifications to input filters and profiles.
295 Added support for SOIF syntax (with private OID).
299 Fixed buffer-size problem in indexing.
301 Added compression to temporary files for updating.
303 Added phrase registers.
305 Added dynamic mapping of search attribute to multiple termlists (ANY).
307 Scan support in multiple databases/registers.
309 Configuration settings are case-insensitive and single dash (-)
310 characters are ignored in comparisons.
312 The index processing ignores empty files - warning given.
314 New option to zebraidx (-V) displays version information.
318 Fixed problem in file-update system.
320 Fixed problem in shadow system; register was sometimes corrupted after
325 Fixed problems in the ISAM subsystem. Caused difficulties when updating
328 Fixed small problem in SUTRS-filter. A newline was sometimes inserted before
329 the rank and record number.
331 Fixed bug in the isam subsystem - caused a malfunction when accessing
332 words which occurred more than 10000 times.
334 Distribution should now include YAZ (Z39.50 protocol stack) to simplify
337 Server can now run under inetd. Use option -i, and -w <directory> to
338 set working directory to desired location.
340 New zebraidx command: clean - removes temporary shadow files.
342 Fixed bug in ISAM system. Occurred rarely during register updates.
344 Logging during index merge phase is improved. The remaining running
347 Temporary files generated by zebraidx are removed after each run.
349 Bug fix: Dictionary didn't handle 8-bit characters correctly; was obvious
350 when doing scan operations in dictionaries with European characters.
354 A whole slew of updates, to make the first publicized release. Get the doc
359 Memory-problems in ISAM fixed. More blocktypes added to the default setup
360 to increase performance on larger databases.
362 Various minor changes in data management system.
366 A couple of portability-problems resolved.
368 Changed some malloc() to xmalloc().