Adam Dickmeiss [Thu, 31 May 2012 09:02:17 +0000 (11:02 +0200)]
Increase max bytes per term from 256 to 512
That's IT_MAX_WORD.
Adam Dickmeiss [Mon, 21 May 2012 14:21:01 +0000 (16:21 +0200)]
Bounds check terms based in ICU norm
Adam Dickmeiss [Mon, 21 May 2012 14:09:45 +0000 (16:09 +0200)]
Prevent memcpy overlap
Happens inside zebra_replace. Simply don't call it here. It has
no effect - except for copying it to WRBUF.
Adam Dickmeiss [Mon, 21 May 2012 14:05:30 +0000 (16:05 +0200)]
Extra wheezy release
Adam Dickmeiss [Tue, 1 May 2012 10:23:02 +0000 (12:23 +0200)]
Build for Ubuntu precise, no build for maverick
Adam Dickmeiss [Tue, 24 Apr 2012 12:29:13 +0000 (14:29 +0200)]
Do not build for Debian lenny anymore
Adam Dickmeiss [Fri, 17 Feb 2012 14:33:30 +0000 (15:33 +0100)]
Update doc/common
Adam Dickmeiss [Wed, 4 Jan 2012 12:30:08 +0000 (13:30 +0100)]
Remove obsolete references
Adam Dickmeiss [Wed, 4 Jan 2012 12:19:45 +0000 (13:19 +0100)]
Remove Emilda section
Adam Dickmeiss [Thu, 15 Dec 2011 09:27:10 +0000 (10:27 +0100)]
IDMETA and mk_version.tcl part of Windows package
Adam Dickmeiss [Thu, 15 Dec 2011 09:21:28 +0000 (10:21 +0100)]
Merge branch 'master' of ssh://git.indexdata.com/home/git/pub/idzebra
Adam Dickmeiss [Thu, 15 Dec 2011 09:21:13 +0000 (10:21 +0100)]
mk_version.tcl: read version from IDMETA
Adam Dickmeiss [Fri, 25 Nov 2011 13:37:19 +0000 (14:37 +0100)]
buildconf removes autom4te.cache
This is to ensure that software version is not cached but taken
from IDMETA as it should.
Adam Dickmeiss [Wed, 23 Nov 2011 20:34:05 +0000 (21:34 +0100)]
Remove Ubuntu karmic from build
Adam Dickmeiss [Tue, 15 Nov 2011 10:01:07 +0000 (11:01 +0100)]
RPM: get version from IDMETA
So we don't need to update the spec file for each release anymore.
Adam Dickmeiss [Tue, 15 Nov 2011 09:59:48 +0000 (10:59 +0100)]
Read version from IDMETA rather than configure.ac.
Adam Dickmeiss [Thu, 10 Nov 2011 11:50:30 +0000 (12:50 +0100)]
Update m4
Adam Dickmeiss [Tue, 1 Nov 2011 13:10:59 +0000 (14:10 +0100)]
Version 2.0.50
Adam Dickmeiss [Tue, 1 Nov 2011 13:03:14 +0000 (14:03 +0100)]
Report 'too many characters in search..'
Make proper diagnostic for too many characters in search
term. And make a test case for it.
Adam Dickmeiss [Tue, 1 Nov 2011 12:43:09 +0000 (13:43 +0100)]
Fix uninit variable (hits_limit)
Adam Dickmeiss [Tue, 1 Nov 2011 09:18:38 +0000 (10:18 +0100)]
Ignore config.h stuff
Adam Dickmeiss [Tue, 1 Nov 2011 09:16:19 +0000 (10:16 +0100)]
Ignore idzebra.pdf
Adam Dickmeiss [Mon, 31 Oct 2011 14:27:05 +0000 (15:27 +0100)]
Refactor if stmt a bit
Adam Dickmeiss [Mon, 31 Oct 2011 14:17:09 +0000 (15:17 +0100)]
Don't reset reg state if register can not be opened
If zebra_register_open fails, do not set state back to 'o', since
it may be dirty (half transaction).
Adam Dickmeiss [Mon, 31 Oct 2011 10:22:32 +0000 (11:22 +0100)]
exit if state file cannot be written
Adam Dickmeiss [Fri, 14 Oct 2011 07:52:56 +0000 (09:52 +0200)]
Proper use of setting:value in zebra conf doc
Adam Dickmeiss [Fri, 14 Oct 2011 07:50:35 +0000 (09:50 +0200)]
Document store{Keys,Data}
Adam Dickmeiss [Wed, 28 Sep 2011 12:34:12 +0000 (14:34 +0200)]
Remove paragraph about commercial support
Because, apparently, www.indexdata.com/support no longer works.
Adam Dickmeiss [Fri, 2 Sep 2011 08:29:14 +0000 (10:29 +0200)]
Version 2.0.49
Adam Dickmeiss [Thu, 1 Sep 2011 10:51:58 +0000 (12:51 +0200)]
Scan: allow preferred position > number + 1
Add a few scan test cases as well.
Adam Dickmeiss [Wed, 31 Aug 2011 17:45:16 +0000 (19:45 +0200)]
scantest: more test cases
Including test case for bug #4592.
Adam Dickmeiss [Wed, 31 Aug 2011 17:40:01 +0000 (19:40 +0200)]
Fix bug #4592: dict_scan misses
Adam Dickmeiss [Wed, 31 Aug 2011 14:23:43 +0000 (16:23 +0200)]
Add test for bug 4592: dict_scan misses
The test is commented out, though (#if 0, #endif).
Adam Dickmeiss [Tue, 30 Aug 2011 11:56:34 +0000 (13:56 +0200)]
Remove unused variable
Adam Dickmeiss [Tue, 30 Aug 2011 11:56:25 +0000 (13:56 +0200)]
Remove unused variable
Adam Dickmeiss [Tue, 30 Aug 2011 11:56:13 +0000 (13:56 +0200)]
Remove unused variable
Adam Dickmeiss [Tue, 30 Aug 2011 11:55:26 +0000 (13:55 +0200)]
Simplify snippet_add_incomplete_field
Same functionality.
Adam Dickmeiss [Tue, 30 Aug 2011 11:31:30 +0000 (13:31 +0200)]
Fix snippets for complete fields, bug #4590
The snippet_add_complete_fields (chr-system) now takes the
indexing term verbatim and makes that the display term.
Adam Dickmeiss [Mon, 29 Aug 2011 08:20:13 +0000 (10:20 +0200)]
Fix comment about term_104, term_105 functions
Adam Dickmeiss [Wed, 29 Jun 2011 16:08:08 +0000 (18:08 +0200)]
Man pages: add refmiscinfo and orgname
Adam Dickmeiss [Tue, 28 Jun 2011 16:24:35 +0000 (16:24 +0000)]
Version 2.0.48
Adam Dickmeiss [Tue, 28 Jun 2011 16:24:25 +0000 (16:24 +0000)]
install idzebra-abs2dom.1
Adam Dickmeiss [Tue, 28 Jun 2011 16:15:47 +0000 (16:15 +0000)]
idzebra-abs2dom part of DIST
Adam Dickmeiss [Tue, 28 Jun 2011 15:16:53 +0000 (17:16 +0200)]
Merge branch 'master' of ssh://git.indexdata.com/home/git/pub/idzebra
Adam Dickmeiss [Tue, 28 Jun 2011 15:15:52 +0000 (17:15 +0200)]
idzebra-abs2dom part of dist and packages
Solves bug #2685.
Adam Dickmeiss [Thu, 23 Jun 2011 09:27:47 +0000 (11:27 +0200)]
Quote m4 arguments
Adam Dickmeiss [Sun, 5 Jun 2011 20:23:47 +0000 (22:23 +0200)]
Produce error if )-char is missing in charmap spec
Adam Dickmeiss [Sun, 5 Jun 2011 20:22:22 +0000 (22:22 +0200)]
Increment seqno for complete fields (non-ICU)
Adam Dickmeiss [Thu, 26 May 2011 14:22:55 +0000 (16:22 +0200)]
zebrasrv: sortkeys args are optional
Adam Dickmeiss [Tue, 24 May 2011 13:55:12 +0000 (13:55 +0000)]
Version 2.0.47
Adam Dickmeiss [Tue, 24 May 2011 13:52:06 +0000 (13:52 +0000)]
RPM: store libs in %{_libdir}
Libraries and modules are stored in %{_libdir}, which is /usr/lib64 on
64-bit platforms.
Adam Dickmeiss [Tue, 24 May 2011 12:03:36 +0000 (12:03 +0000)]
Remove last file of xelm test
Adam Dickmeiss [Tue, 24 May 2011 11:57:43 +0000 (11:57 +0000)]
Remove test not in use: test/xelm
Adam Dickmeiss [Tue, 24 May 2011 11:04:47 +0000 (13:04 +0200)]
Require libyaz4-devel >= 4.2.0
Adam Dickmeiss [Tue, 24 May 2011 11:04:15 +0000 (13:04 +0200)]
Require libyaz4-dev >= 4.2.0
Adam Dickmeiss [Tue, 24 May 2011 11:04:00 +0000 (13:04 +0200)]
Make Zebra compile with YAZ 3 sources again
Adam Dickmeiss [Thu, 19 May 2011 13:27:47 +0000 (15:27 +0200)]
configure produces config.h
Adam Dickmeiss [Thu, 19 May 2011 11:20:27 +0000 (13:20 +0200)]
Fix null ptr WRT sortkeys check
Adam Dickmeiss [Mon, 16 May 2011 11:42:22 +0000 (13:42 +0200)]
buildconf.sh: update package list for compiling with sources
Adam Dickmeiss [Mon, 16 May 2011 10:21:52 +0000 (12:21 +0200)]
Zebra honors SRU sort, bug 4211
The srw_sortKeys are converted to PQF with sorting (type-1).
At this point the code relies on new YAZ function nmem_strsplitx.
Adam Dickmeiss [Fri, 13 May 2011 09:15:29 +0000 (11:15 +0200)]
Avoid truncating hit counts because Odr_int is 64-bit
Hit counts were represented as an int in YAZ 3 and earlier; no longer
the case.
Adam Dickmeiss [Wed, 11 May 2011 11:25:19 +0000 (13:25 +0200)]
Simplify gitignore
Adam Dickmeiss [Wed, 11 May 2011 11:25:03 +0000 (13:25 +0200)]
Update comment about version update
Adam Dickmeiss [Wed, 11 May 2011 08:57:22 +0000 (10:57 +0200)]
scan: Fix first-in-field term skipping
Different check for first-in-feild term. The check for first-in-field
term could skip CJK entries for ICU normalized sort keys. In theory it
can still happen if CJK entries would map to exactly the same sequence
as first-in-field entries (highly unlikely). There was also a problem
with scan for the empty string. Before this fix, it would result in
scan failure (no entries returned), because first entry would hit the
the first-in-field term. Fixes bug #4207.
Adam Dickmeiss [Wed, 11 May 2011 08:56:52 +0000 (10:56 +0200)]
marcxml example: utf-8 encoding for queries, etc
Adam Dickmeiss [Wed, 4 May 2011 15:24:30 +0000 (17:24 +0200)]
usmarc test: run init first
Adam Dickmeiss [Fri, 29 Apr 2011 11:23:46 +0000 (11:23 +0000)]
Add Debian wheezy, Ubuntu natty
Adam Dickmeiss [Wed, 23 Mar 2011 19:16:12 +0000 (20:16 +0100)]
Version 2.0.46
Adam Dickmeiss [Wed, 23 Mar 2011 19:07:16 +0000 (20:07 +0100)]
Improve estimate hit counts for rsprox sets.
The method used now is similar to and-sets.
Adam Dickmeiss [Wed, 23 Mar 2011 17:45:25 +0000 (18:45 +0100)]
Fix dict_delete that could delete wrong entry
dict_delete could delete entry X if entries XY (Y suffixes) were
all removed. If prefix entry X was a real entry (not just a subptr
entry) it would be removed too. dict_del_string now checks if
X is a real entry or not before removing.. In case of real entry
subptr is just set to 0.
Adam Dickmeiss [Wed, 23 Mar 2011 16:48:45 +0000 (17:48 +0100)]
Update m4.
Adam Dickmeiss [Tue, 22 Mar 2011 13:44:45 +0000 (13:44 +0000)]
Merge branch 'master' of ssh://git.indexdata.com/home/git/pub/idzebra
Adam Dickmeiss [Tue, 22 Mar 2011 13:44:35 +0000 (13:44 +0000)]
Fix comp for FLUSH2==0
Adam Dickmeiss [Tue, 22 Mar 2011 13:43:29 +0000 (13:43 +0000)]
Fix comp for FLUSH2==0
Adam Dickmeiss [Mon, 21 Mar 2011 11:40:49 +0000 (12:40 +0100)]
Display command check in usage
Adam Dickmeiss [Mon, 28 Feb 2011 14:02:51 +0000 (15:02 +0100)]
Introduce spec for zebra_register_check
Adam Dickmeiss [Mon, 28 Feb 2011 13:15:36 +0000 (14:15 +0100)]
Fix warning and code WRT uninit var (sort)
Adam Dickmeiss [Mon, 28 Feb 2011 12:32:49 +0000 (13:32 +0100)]
zebra_register_check fixes
Log at most 20 entries for each error type during check. Allow
up to IT_KEY_LEVEL_MAX (5) , not 4 in check for each key.
Adam Dickmeiss [Tue, 22 Feb 2011 14:29:13 +0000 (15:29 +0100)]
Zebra modules: add YAZ libs
Add YAZLALIB to Zebra filter modules, because they use YAZ symbols
directly.
Adam Dickmeiss [Wed, 9 Feb 2011 09:31:25 +0000 (10:31 +0100)]
Happy new year.
Adam Dickmeiss [Wed, 9 Feb 2011 09:20:29 +0000 (09:20 +0000)]
Summarize type of failures for register check
Adam Dickmeiss [Mon, 7 Feb 2011 13:38:55 +0000 (13:38 +0000)]
zebra_record_check checks isamb too.
Adam Dickmeiss [Mon, 7 Feb 2011 13:28:44 +0000 (13:28 +0000)]
Fix dirty marker for CategoryList (IR-Explain-1)
Register check showed that keys for it is IR-Explain-1's category
list is inconsistent. NO longer the case with this fix.
Adam Dickmeiss [Mon, 7 Feb 2011 11:14:33 +0000 (11:14 +0000)]
Verbose level for register check
Adam Dickmeiss [Fri, 4 Feb 2011 15:29:17 +0000 (16:29 +0100)]
Register check checks dictionary (dict_lookup)
Adam Dickmeiss [Tue, 1 Feb 2011 12:06:21 +0000 (12:06 +0000)]
Beginnings of zebra_register_check
Adam Dickmeiss [Tue, 1 Feb 2011 12:05:54 +0000 (12:05 +0000)]
Add rec_get_next
Adam Dickmeiss [Tue, 1 Feb 2011 12:05:24 +0000 (12:05 +0000)]
Restructure print key debuggin
Adam Dickmeiss [Thu, 23 Dec 2010 10:20:24 +0000 (11:20 +0100)]
Remove idzebra.spec.in
Adam Dickmeiss [Thu, 23 Dec 2010 10:17:52 +0000 (11:17 +0100)]
Update m4
Adam Dickmeiss [Wed, 22 Dec 2010 13:29:26 +0000 (14:29 +0100)]
Merge branch 'master' of ssh://git.indexdata.com/home/git/pub/idzebra
Adam Dickmeiss [Wed, 22 Dec 2010 13:08:15 +0000 (14:08 +0100)]
Version 2.0.45
Adam Dickmeiss [Wed, 22 Dec 2010 13:03:31 +0000 (14:03 +0100)]
RPN: avoid \-handling for trunctation=105/106.
Queries with trunction=105/106 that holds \-characters will now search
for literal \ instead of mark a backslash sequence (such as known
from regular expressions).
Adam Dickmeiss [Wed, 22 Dec 2010 12:57:18 +0000 (13:57 +0100)]
rpn: Disable \-handling for struct=105/106
Adam Dickmeiss [Fri, 15 Oct 2010 12:51:27 +0000 (14:51 +0200)]
Add Ubuntu maverick
Adam Dickmeiss [Wed, 6 Oct 2010 11:31:36 +0000 (13:31 +0200)]
Doc: update to Docbook 4.4
Adam Dickmeiss [Wed, 30 Jun 2010 14:45:50 +0000 (16:45 +0200)]
Produce PDf idzebra.pdf (was zebra.pdf)
Include other formatting changes as well. acronym/literal inside
screen not in use anymore.
Adam Dickmeiss [Wed, 30 Jun 2010 14:44:17 +0000 (16:44 +0200)]
Remove newline in <filename> content and other
Having a newline in <filename> CDATA breaks conversion to PDF.
Only took two hours to find out.
Adam Dickmeiss [Wed, 30 Jun 2010 13:36:07 +0000 (15:36 +0200)]
Remove unused definitions
Adam Dickmeiss [Wed, 30 Jun 2010 13:22:44 +0000 (15:22 +0200)]
Remove etch, jaunty, intrepid, hardy