1 Maximum number of databases in one Zebra register increased.
3 New setting, databasePath, which specifies that first directory
4 during update traversal is the database name (instead of
7 New setting, explainDatabase, which specifies that databases
10 Modified Zebra so that it works with ASN.1 compiled code for YAZ.
12 Implemented EXPLAIN database maintenance. Zebra automatically
13 generate - and update TargetInfo, DatabaseInfo, AttributeSetInfo
14 and AttributeDetails records at this stage. The records may
15 be transferred as GRS-1, SUTRS and Explain.
17 Fixed register spec so that colon isn't treated as size separator
18 unless followed by [0-9+-] in order to allow DOS drive specifications.
20 Fixed two bugs in ISAMC system.
22 Changed the way Zebra keeps its maintenance information. Records
23 in "SGML" notation using an EXPLAIN schema is now used when
26 Bug fix: Index didn't handle update/insert/delete of the same
27 record (i.e. same recordId) in one run (one invocation of zebraidx).
28 Only the first occurence of a record is considered.
30 Most searches now return correct number of hits.
32 New modular ranking system. Interested programmers are encouraged to
33 inspect rank1.c and improve the algorithm.
35 Bug fix: Lock files weren't removed as they should when using NT.
37 Implemented Z39.50 Sort. Zebra's sort handler uses use attributes to
38 specify a "sort register". Refer to the gils sample records which
39 refer to index type "s" which is specified as "sort" in the default.idx
40 file. Each sort criteria can either be Ascending or Descending and
41 at most three sort elements can be specified.
43 Bug fix: Character mapping didn't work for text files.
47 Simple ranked searches now return correct number of hits.
49 The test option (-s) only makes a read-lock on the index as well
50 as using read-only operations anywhere.
52 Moved towards generic character mapping. Configuration file default.idx
53 specifies character map files for register types w, p, u, etc.
55 Implemented "begin variant" for the sgml.regx - filter.
57 Fixed a few memory leaks.
59 Added support for C++, headers uses extern "C" for public definitions.
61 Bug fix: The show records facility (-s) only displayed information for
62 the first record in a file (and not for every record in the file).
64 Added option "-f <n>" to limit the logging of record operations. After
65 <n> records has been processed no logging is performed (unless errors
68 Bug fix: the compressed ISAM system didn't handle update operations
71 Added setting, "maxResultSetSize", to hold the number of records to
74 Bug fix: Complete phrase did't work for search operations.
76 Bug fix: temporary result sets weren't deleted.
78 Reduced disk space for saved keys (storeKeys = 1).
80 Added optional, physical ANY (key replication)
82 Implemented proximity operator in search.
84 Bug fix: the path name buffers used by file match traversal routines
85 have been extended to support long file names.
87 New C(ompressed) ISAM system. To enable it, specify "isam: c" in the
88 configuration file. The resulting register without "storeKeys" is about
89 half the size, and the memory used by zebraidx during phase 2 (merge) is
92 Reworked the way Regexp-2 queries with error tolerance are handled and
93 specified. The documentation has been updated accordingly.
95 Bug fix: Zebrasrv didn't search correctly when queries contained masking
96 characters. This bug was introduced in 1.0a8.
98 Zebrasrv now tag records with the proper database name.
100 New settings, memMax and keyTmpDir.
102 Changed name of setting lockDir (previously called lockPath) and
103 setTmpDir (previously called tempSetPath).
105 Generalized and changed record type specifications. In short, there are:
107 grs.sgml structured, "SGML-like" syntax
108 grs.regx.<filter> structured, Regular expression filter
109 grs.marc.<abs> Reads *MARC records in the ISO2709 format. <abs>
110 is the name of an abstract syntax file.
111 Bug fix: Result sets weren't sorted in operations involving boolean
112 operations with "ranked" operands.
116 Added national character-handling subsystem.
120 Small modifications to input filters and profiles.
122 Added support for SOIF syntax (with private OID).
126 Fixed buffer-size problem in indexing.
128 Added compression to temporary files for updating.
130 Added phrase registers.
132 Added dynamic mapping of search attribute to multiple termlists (ANY).
134 Scan support in multiple databases/registers.
136 Configuration settings are case-insensitive and single dash (-)
137 characters are ignored in comparisons.
139 The index processing ignores empty files - warning given.
141 New option to zebraidx (-V) displays version information.
145 Fixed problem in file-update system.
147 Fixed problem in shadow system; register was sometimes corrupted after
152 Fixed problems in the ISAM subsystem. Caused difficulties when updating
155 Fixed small problem in SUTRS-filter. A newline was sometimes inserted before
156 the rank and record number.
158 Fixed bug in the isam subsystem - caused a malfunction when accessing
159 words which occurred more than 10000 times.
161 Distribution should now include YAZ (Z39.50 protocol stack) to simplify
164 Server can now run under inetd. Use option -i, and -w <directory> to
165 set working directory to desired location.
167 New zebraidx command: clean - removes temporary shadow files.
169 Fixed bug in ISAM system. Occurred rarely during register updates.
171 Logging during index merge phase is improved. The remaining running
174 Temporary files generated by zebraidx are removed after each run.
176 Bug fix: Dictionary didn't handle 8-bit characters correctly; was obvious
177 when doing scan operations in dictionaries with European characters.
181 A whole slew of updates, to make the first publicized release. Get the doc
186 Memory-problems in ISAM fixed. More blocktypes added to the default setup
187 to increase performance on larger databases.
189 Various minor changes in data management system.
193 A couple of portability-problems resolved.
195 Changed some malloc() to xmalloc().