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