1 Release candidate Zebra version 2.0.0 prepared.
3 Zebra uses string attributes for indexing internally. Using set+numeric
4 use attribute can still be used. But that is a search-only conversion which
5 inspects .att-set files as indicated using attset-directives in zebra.cfg.
6 attset references are no longer required, but when used they deserve
7 as "check" for that the index names used are also present in .att.
9 Implemented bug #588: allrecords search. Using @attr 1=allrecords ""
10 matches all records in a database.
12 Zebra reads (non-standard) attribute 10 for RPN queries. The value, when
13 specified, is returned as subqueryId in SearchResult-1. It un-specified
14 no subqueryId is part of SearchResult-1.
16 Added 'dbperms' setting to zebra.cfg. Names file with lines of
17 format username: db1+...dbn, listing databases that user has access to.
18 Database-specific access control only happens for searching, at present.
20 Added zebra_set_approx_limit for a ZebraHandle (session). Results
21 will be approximate if hit count is greater than the limit specified.
23 Added support for term hit counts. This was not in place for earlier
24 1.4 versions, but is present in the 1.3 series. Bug #124.
26 Implemented the 'equivalent' directive for .chr-files.
28 Added zebra_get_bfs to get BFile handle for session.
30 Added 'melm' directive to absyn format to simplify config files
31 for MARC-style databases. See tab/marc21.abs for an example.
33 Added bath-compliant definition file marc21.abs for use with MARC databases.
35 Added autoconf utility idzebra.m4 which defines --with-idzebra .
37 Added mechanism to ignore leading articles when doing full-field indexing,
38 based on the character map files. See the manual for further discussion.
40 Fixed bug in record management. Releasing blocks could result in
43 Fixed bug in isam:b. A tree split could result in a lost item.
45 Remove isamd. It's not been in use for a long time and isamb is better
48 Change SYSNO to be zint. Change pointers in isamc and isamb to zint.
49 Change block number in bfile/cfile to zint. zint is a long integer
50 (64-bit). This change practially removes register limits for Zebra.
52 Implement int-list encoding for ISAMs.
54 Added facility to make attibutes in grs.regx and grs.tcl filter using the
55 data command with argument -attribute <name> . The content of data is
56 the value of the attribute. This command should be used inside a
57 begin element , end element section.
59 Update zebra.nsi to NSIS 2.
61 Added a new 'cut' directive to charmaps (.chr files) which specifies that
62 only characters after the cutting char should be indexed.
64 Update Perl internals so that it matches the current Zebra API.
65 The recordGroup structure is no longer available. A group of resources
66 can still be referenced by setting groupName=>.. in various methods.
68 Maximum number of records to be sorted in a result set can be
69 specified by setting "sortmax". Default is 1000.
71 Allow use of string use attributes for regular attribute sets. The
72 name matches the name given in the attribute set file. All strings
73 starting with / are considered X-Path as usual.
75 Fixed bug in grs.regx. filter . 'end element' could pop off top tag
76 element for XML tree. It may only pop off if -record is given.
78 Added grs.danbib filter - for Danish Bibliographic Centre.
80 Rename CHANGELOG to NEWS.
82 For text filter, return only header if elementSetName=H . elementSetName=R
83 returns contents only. Other elementSetName returns both header+content.
85 Added test for charmap and rusmarc.
87 Added feature charmaps (.chr) so that characters may be specified in
90 Fixed problem with encoding directive for charmap(.chr) files.
92 Allow Remote insert/delete/replace/update with record, recordIdNumber
93 (sysno) and/or recordIdOpaque(user supplied record Id). If both
94 IDs are omitted internal record ID match is assumed (recordId: - in
99 Fix bug. X-Path attribute expressions with spaces in them now works.
101 Fix base address for MARC output.
103 --- 1.3.14 2003/11/29
105 Fix bug with shadow and result set handling.
107 Implement MARCXML to ISO2709 conversion.
109 --- 1.3.13 2003/09/26
111 Add missing examples for Windows install.
113 Fix bug in regx filter to make it "greedy" again. This bug appeared
118 --- 1.3.12 2003/09/08
120 Fix XML error handling. Stop XML parse immediately if XML parse error
121 occur (i.e. produce one error only).
123 Zebra ignores "unsupported use attribute" for individual databases
124 when search multiple databases (unless all databases fail).
126 New filter grs.marcxml which works like grs.marc but produces MARCXML.
128 Added support for database deletion. It is possible to create/drop
129 a database from zebraidx utility. Note: only for isam:b.
131 Write zebrasrv.pid to lockdir.
133 Bug fix: result sets were not recovered correctly. Had to
134 add ODR handle for zebra_search_RPN in order to make it work.
136 Fixed a bug in regx filters that didn't do anchors (^) correctly.
138 Fixed a bug in searches with X-Path searches sometimes giving "extra"
141 Zebra server checks for zebrasrv.pid and refuses to start if it is already
142 locked by another (running) zebrasrv.
144 Fixed a bug with text being chunked in pieces for the grs.xml filter.
146 --- 1.3.11 2003/04/25
148 xelm code updates. xelm works regardless state of 'xpath enable/disable'
149 Avoid -L/usr/lib since that is already default library path.
151 Allow multiple updates within one transaction.
153 Fixed a bug with >2GB files (overflow in integer expression).
155 --- 1.3.10 2003/04/01
157 Fix linker error for Perl module.
159 Fix bug in and operation which in some cases could result in "extra"
160 hits. Bug was introduced in 1.3.5.
162 Fix bug in handling of schema conversion when producing numeric tags.
168 Add missing files doc/zvrank.txt and doc/marc_indexing.xml.
172 Zvrank: an experimental ranking algorithm. See doc/zvrank.txt and
173 source in index/zvrank.c. Enable this by using rank: zvrank in zebra.cfg.
174 Contributed by Johannes Leveling <Johannes.Leveling at fernuni-hagen.de>
176 livrank: another experimental ranking algorithm. Source in livcode.c.
177 Enable this by using rank: livrank in zebra.cfg and use -DLIV_CODE=1
179 Contributed by Pete Mallinson, University of Liverpool.
181 Advanced MARC indexing. See doc/marc_indexing.xml
182 Oleg Kolobov <oleg at lib.tpu.ru>
184 Perl API updates and fixes.
185 Peter Popovics <pop at technomat.hu>
187 Fixed 'zebraidx delete'.
189 Implemented 'zebraidx clean'.
191 64-bit offsets for register files on WIN32 (no 2 GB limit).
193 Fixed a few memory leaks WRT sorting.
197 Fixed error handling : error code was not properly returned.
199 Support Truncation 104 (CCL).
203 Added missing source files for perl extension.
207 Implemented xelm directive.
209 Updated for newer version of YAZ (introduction of string schema).
211 Directory examples/zthes now part of distribution (was missing
212 in previous release).
214 New .abs directive, systag, that control where to put retrieval
215 information. The directive takes two arguments: system tag, element name.
216 System tag is one of : rank, sysno, size.
220 Perl Filter and Perl API. By Peter Popovics.
222 For zebra.cfg, if no profilePath is specified, directory
223 (prefix)/share/idzebra/tab
226 Zebra Examples in examples . Zebra tests in test.
228 Bug fix: sort index was not properly modified on
229 record updates/deletes.
231 Fix handling of character entities for sgml filter.
233 Move data1 to Zebra (used to be part of YAZ).
237 Fix character encoding of scan response terms.
239 Fix character decoding of scan request terms.
241 Fix ESpec handling (requires YAZ 1.9.1)
243 Fix searches for complete fields.
247 When name zebra is used in a filename or directory 'idzebra' is used
248 instead to avoid confusion with GNU zebra (routing software).
250 Zebra server stops with a fatal error if config file cannot be read.
252 New config setting, followLinks, that controls whether update of files
253 should follow symbolic. Set it to 1 (for enable) or 0 (to disable).
254 By default symbolic links are followed.
256 Fix MARC transfer . MARC fields had wrong data for multiple fields.
258 XML record reader moved from YAZ to Zebra, to make YAZ less
259 dependant on external libraries.
261 Zebra uses yaz_iconv which is mini iconv library supporting UTF-8,
262 UCS4, ISO-8859-1. This means that Zebra does UNICODE even
263 on systems that doesn't offer iconv.
265 XML record reader supports external system entities.
269 New .abs-directive "xpath" that takes one argument: "enable"
270 or "disable" to enable and disable XPath -indexing. If no "xpath"
271 direcive is found in .abs-file , XPath-indexing is disabled to ensure
272 backwards compatibility. For missing .abs-files XPath-indexing is
273 enabled so that such records are searchable.
275 Zebra warns about missing .abs-file only once (for each type).
277 Fixed a bug in file update where already-inserted files could
282 Zebra license changed to GNU GPL.
284 XPath-like queries used when RPN string attributes are used, eg.
285 @attr 1=/portal/title sometitle
286 @attr 1=/portal/title[@xml:lang=da] danishtitle
287 @attr 1=/portal/title/@xml:lang da
288 @attr 1=//title sometitle
290 Zebra uses UTF-8 internally:
291 1) New setting "encoding" for zebra.cfg that specifies encoding for
292 OCTET terms in queries and record encoding for most transfer syntaxes
293 (except those that use International Strings, such as GRS-1).
294 2) The encoding of International strings is UTF-8 by default. It
295 may be changed by character set negotiation. If character set
296 negotiation is in effect and if records are selected for conversion
297 these'll be converted to the selected character set - thus overriding
298 the encoding setting in zebra.cfg.
299 3) New directive "encoding" in .abs-files. This specifies the external
300 character encoding for files indexed by zebra. However, if records
301 themselves have an XML header that specifies and encoding that'll be used
304 XML filter (-t grs.xml).
306 Multiple registers. New setting in resource 'root' that holds base
307 directory for register(s). A group a databases may be put in separate
308 register in directory root/reg by using db name 'reg/db1' ... 'reg/dbN'.
312 Fixes for Digital Unix
314 Implemented hits per term using USR:SearchResult-1.
316 New Zebra API. Locking system re-implemented.
318 --- 1.1.stable 2002/02/20
320 Rank weight can be controlled with attribute type 9. Default
321 value is 34. Recommended values between 1-36.
325 Updated for YAZ version 1.8.
327 Added support for termsets - a result set of terms matching
328 a given query. For @attr 8=<set> creates termset named <set>.
330 Added support for raw retrieval. Element Set Name R forces the
331 text filter which returns the record in its original form.
333 Added numerical sort - triggered by structure=numeric (4=109).
335 Remote record import using Z39.50 Extended Services and Segments.
337 Fixed bug where updating a database with user-defined attributes
338 could corrupt the register (bad storeKeys).
340 Multi-threaded version.
342 Fixed bug regarding proximity.
344 Documentation updates.
346 Fixed bug in record retrieval module that occured on 64-bit OSF
351 Fixed bug in makefile for WIN32.
353 Fixed bug in configure script - used bash-specific features.
357 Added support for multiple records in one file for filter grs.sgml.
359 Changed record index structure. New layout is incompatible with
360 previous releases. Added setting "recordcompression" to control
361 compression of records. Possible values are "none" (no
362 compression) and bzip2 (compression using libbz2).
364 Added XML transfer syntax support for retrieval of structured records.
365 Schema in CompSpec is recognised in retrieval of structured records.
367 Changed Tcl record filter so that it attemps to read <filt>.tflt. If
368 that fails, the filter reads the file <filt>.flt (regx style filter).
370 Implemented new Tcl record filter - use grs.tcl.<filter> to enable it.
371 Zebra's configure script automatically attempts to locate Tcl. For
372 manual Tcl configuration use option --with-tclconfig=<path> to specify
373 where Tcl's library files are located.
375 Implemented "compression" of Dictionary and ISAM system. Dictionary
378 Added "tagsysno" directive to zebra.cfg to control under which tag the
379 system ID is placed. Use tagsysno: 0 to disable Zebra's system number
382 Added "tagrank" as above.
384 Changed file naming scheme for register files from <name>.mf.<no> to
387 Implemented "position"-flag for register type (as defined in
388 default.idx). When set to zero no position (or seqence number) is
389 saved in register for each word occurrence, thus saving some register
392 Implemented database mapping. Using mapdb one can specify a database
393 to be mapped to one or more physical databases. Usage:
394 mapdb <fromdb> <todb> ..
396 Added SOIF-filter. Thanks to Peter Valkenburg.
398 For the regx-filter "end element -record" may trigger a mark-of-record
399 if outer level is reached.
401 Tag sets may be typed in the reference to it. From the .abs-file the
402 "tagset" directive takes a third optional integer type for the tag set
403 referenced. From a .tag-file the "include" directive takes a third
404 optional type as well. The old "type" directive in the tag set itself
405 is still recognized but acts as the default type for the tag set.
407 Zebra supports the specification of arbitrary attributes sets, schemas
408 and tag sets, because of the change in YAZ' OID management system.
410 Fixed bug in Sort that caused it NOT to use character mapping as it
413 Zebra now uses GNU configure to generate Makefile(s).
415 Added un-optimised support for left and left/right truncation attributes.
417 Added support for relational operators on text when using RPN queries.
419 Added support for sort specifications in RPN queries. Type 7 specifies
420 'sort' where value 1=ascending, value 2=descending. The use attribute
421 specifies the field criteria as usual. The term specifies priority
422 where 0=first, 1=second, ...
424 Changed the way use attributes are specified in the recordId
427 Maximum number of databases in one Zebra register increased.
429 New setting, databasePath, which specifies that first directory during
430 update traversal is the database name (instead of a fixed one).
432 New setting, explainDatabase, which specifies that databases are
435 Modified Zebra so that it works with ASN.1 compiled code for YAZ.
437 Implemented EXPLAIN database maintenance. Zebra automatically
438 generate - and update CategoryList, TargetInfo, DatabaseInfo,
439 AttributeSetInfo and AttributeDetails records at this stage. The
440 records may be transferred as GRS-1, SUTRS or Explain.
442 Fixed register spec so that colon isn't treated as size separator
443 unless followed by [0-9+-] in order to allow DOS drive specifications.
445 Fixed two bugs in ISAMC system.
447 Changed the way Zebra keeps its maintenance information about attribute
448 sets, available attributes, etc.. Records in "SGML" notation using an
449 EXPLAIN schema is now used when appropriate.
451 Bug fix: Index didn't handle update/insert/delete of the same record
452 (i.e. same recordId) in one run (one invocation of zebraidx). Only the
453 first occurence of a record is considered.
455 Most searches now return correct number of hits.
457 New modular ranking system. Interested programmers are encouraged to
458 inspect rank1.c and improve the algorithm.
460 Bug fix: Lock files weren't removed as they should on NT.
462 Implemented Z39.50 Sort. Zebra's sort handler uses use attributes to
463 specify a "sort register". Refer to the gils sample records which refer
464 to index type "s" which is specified as "sort" in the default.idx file.
465 Each sort criteria can either be Ascending or Descending and at most
466 three sort elements can be specified.
468 Bug fix: Character mapping didn't work for text files.
472 Simple ranked searches now return correct number of hits.
474 The test option (-s) only makes a read-lock on the index as well
475 as using read-only operations anywhere.
477 Moved towards generic character mapping. Configuration file default.idx
478 specifies character map files for register types w, p, u, etc.
480 Implemented "begin variant" for the sgml.regx - filter.
482 Fixed a few memory leaks.
484 Added support for C++, headers uses extern "C" for public definitions.
486 Bug fix: The show records facility (-s) only displayed information for
487 the first record in a file (and not for every record in the file).
489 Added option "-f <n>" to limit the logging of record operations. After
490 <n> records has been processed no logging is performed (unless errors
493 Bug fix: the compressed ISAM system didn't handle update operations
496 Added setting, "maxResultSetSize", to hold the number of records to
497 save in a result set.
499 Bug fix: Complete phrase did't work for search operations.
501 Bug fix: temporary result sets weren't deleted.
503 Reduced disk space for saved keys (storeKeys = 1).
505 Added optional, physical ANY (key replication)
507 Implemented proximity operator in search.
509 Bug fix: the path name buffers used by file match traversal routines
510 have been extended to support long file names.
512 New C(ompressed) ISAM system. To enable it, specify "isam: c" in the
513 configuration file. The resulting register without "storeKeys" is about
514 half the size, and the memory used by zebraidx during phase 2 (merge) is
515 reduced to a minimum.
517 Reworked the way Regexp-2 queries with error tolerance are handled and
518 specified. The documentation has been updated accordingly.
520 Bug fix: Zebrasrv didn't search correctly when queries contained masking
521 characters. This bug was introduced in 1.0a8.
523 Zebrasrv now tag records with the proper database name.
525 New settings, memMax and keyTmpDir.
527 Changed name of setting lockDir (previously called lockPath) and
528 setTmpDir (previously called tempSetPath).
530 Generalized and changed record type specifications. In short, there are:
532 grs.sgml structured, "SGML-like" syntax
533 grs.regx.<filter> structured, Regular expression filter
534 grs.marc.<abs> Reads *MARC records in the ISO2709 format. <abs>
535 is the name of an abstract syntax file.
536 Bug fix: Result sets weren't sorted in operations involving boolean
537 operations with "ranked" operands.
541 Added national character-handling subsystem.
545 Small modifications to input filters and profiles.
547 Added support for SOIF syntax (with private OID).
551 Fixed buffer-size problem in indexing.
553 Added compression to temporary files for updating.
555 Added phrase registers.
557 Added dynamic mapping of search attribute to multiple termlists (ANY).
559 Scan support in multiple databases/registers.
561 Configuration settings are case-insensitive and single dash (-)
562 characters are ignored in comparisons.
564 The index processing ignores empty files - warning given.
566 New option to zebraidx (-V) displays version information.
570 Fixed problem in file-update system.
572 Fixed problem in shadow system; register was sometimes corrupted after
577 Fixed problems in the ISAM subsystem. Caused difficulties when updating
580 Fixed small problem in SUTRS-filter. A newline was sometimes inserted before
581 the rank and record number.
583 Fixed bug in the isam subsystem - caused a malfunction when accessing
584 words which occurred more than 10000 times.
586 Distribution should now include YAZ (Z39.50 protocol stack) to simplify
589 Server can now run under inetd. Use option -i, and -w <directory> to
590 set working directory to desired location.
592 New zebraidx command: clean - removes temporary shadow files.
594 Fixed bug in ISAM system. Occurred rarely during register updates.
596 Logging during index merge phase is improved. The remaining running
599 Temporary files generated by zebraidx are removed after each run.
601 Bug fix: Dictionary didn't handle 8-bit characters correctly; was obvious
602 when doing scan operations in dictionaries with European characters.
606 A whole slew of updates, to make the first publicized release. Get the doc
611 Memory-problems in ISAM fixed. More blocktypes added to the default setup
612 to increase performance on larger databases.
614 Various minor changes in data management system.
618 A couple of portability-problems resolved.
620 Changed some malloc() to xmalloc().