Debian Codegraph @

Changes

raw
1
2 1.25 - Mon Sep 5 09:24:00 UTC 2005
3
4 - Stop QueryParser turning 'notice' into 'NOT ice' (RT#14450)
5
6 1.24 - Wed 20 July 2005
7
8 - use pack and unpack to read/write frq and prx files [Minty]
9 - Small doc changes [Minty / TB]
10
11 1.23 - Sun 17 July 2005
12
13 - Stop rolling our own sorting in SegmentMerger/merge_term_infos [Minty]
14 - Replace sequence of read_byte() calls in SegmentReader with one
15 direct call to read() [Aaron Crane]
16 - add a fast clone() method to Plucene::Index::TermInfo [TB]
17 - and use it in preference to copy_in() [Minty]
18 - close the file at DESTROY in Input/Output Stream classes [Minty]
19 - Improve POD for Document::Field [Geoff Richards]
20 - minor POD fixes [Minty / TB]
21 - add indexing benchmark script [TB]
22
23 1.22 - Tue Jun 28 2005
24
25 - perl 5.8 required
26 - use pack and unpack to read/write 'segments' files in SegmentInfos [TB]
27 - Use string xor to find longest common string prefix in
28 TermInfosWriter [Minty / Nicholas Clark]
29 - switch to direct hash access in comparison methods in
30 Plucene::Index::Term [Minty]
31 - remove some indirection in SegmentTermEnum [TB]
32 - remove Plucene::Index::FieldInfo file [TB]
33 - minor speed tweaks to SegmentMergeInfo, SegmentTermPositions [TB]
34 - minor speed tweaks to SegmentsTermEnum (and split out into own file) [TB]
35 - remove Tie::Array::Sorted from SegmentMerger [TB/MP]
36
37 1.21 - Mon 24 Jan 2005
38 - use pack and unpack to read/write FieldInfos files (.fnm). This
39 breaks index compatability with Lucene, as we now use BER
40 compressed integers (but we never really had index compatability
41 anyway). It will also break index compatability with older versions
42 of Plucene if you have documents with more than 127 fields. Hopefully
43 no-one actually does. [MP / TB]
44 - localise $\ in Plucene::Store::OutputStream so -l works [Chris Nandor]
45 - Removed needless new() method from FieldInfo for speed [TB]
46 - Speed up FieldsInfo::_add_internal [TB]
47 - explicitly close lock files immediately after creating them [MP]
48 - remove all remaining catfile() calls [TB]
49 - tweak docs for skipping_doc() [TB]
50 - lots of tiny micro-optimisations [TB]
51
52 1.20 - Wed 15 Dec 2004
53 * Just upped the version number, to fix some issues with CGI::Kwiki
54
55 1.19 - Wed 21 July 2004
56 * Removed accidentally-included experimental patches
57
58 1.18 - Tue 20 July 2004
59 * Downgraded dependencies to work with Perl 5.6.1
60 * Fix to delete_term
61
62 1.17 - Tue 27 Apr 2004
63 * Plucene::Search::Hits::hit_doc should call _get_more_hits not get_more_docs
64 (thanks to Kjetil Fikkan for pointing it out)
65
66 1.16 - Mon 12 Apr 2004
67 * not released
68
69 1.15 - Mon 29 Mar 2004
70 * Remove some needless prereqs.
71 * Removal of Class::HasA completely
72 * Remove use of 'catfile'
73
74 1.14 - thu Mar 25 2004
75 * Put Encode::compat in a BEGIN block in the Store classes.
76 * Change the Build.PL to need Encode::compat if < 5.6
77
78 1.13 - Thu Mar 25 2004
79 * Remove Encode::compat.
80
81 1.12 - Tue Mar 23 2004
82 * Put back Encode::compat
83
84 1.11 - Tue Mar 23 2004
85 * More speed up patches (thanks to Marty Pauley and Tony Bowden)
86
87 1.10 - Thu Mar 18 2004
88 * Include the test left out for the bug in 1.08
89
90 1.09 - Wed Mar 17 2004
91 * Now allows documents of id 0 to be deleted correctly! (Thanks to Steve Rushe for pointing this out)
92
93 1.08 - Wed Mar 17 2004
94 * Some speed ups in InputStream (thanks to Marty Pauley)
95
96 1.07 - Web Feb 18 2004
97 * Allowed the changing of mergefactor
98
99 1.06 - Tue Feb 17 2004
100 * Checked $] in Build.pl to decide if to use Encode::compat
101 * Added create_makefile_pl => 'traditional' to Build.PL
102 * moved from dist_name to module_name to stop warning when running Makefile.PL
103
104 1.05 - Fri Feb 13 2004
105 * Remove commented out line in Plucene::Store::InputStream
106 * Removed hardcoded /tmp index directories, moved to File::Temp
107 * Removed system() calls from tests
108
109 1.04 - Fri Feb 06 2004
110 * Doc change to clarify what optimize() does to a writer
111 * Removed tell call from Plucene::Store::InputStream::read_byte
112
113 1.03 - Thu Feb 05 2004
114
115 * Added in use of Enocde::compat to ensure works with < 5.7
116 (thanks to Simon Cozens)
117 * Added Plucene::Index::IndexSearcher::close()
118
119 1.02 - Wed Feb 04 2004
120
121 * Speed up sort (thanks to Tony Bowden)
122 * Move from Class::Accessor to Class::Accessor::Fast (thank to Tony again)
123 * StandardAnalyzer should be a subclass of Analyzer!
124 * StandardTokenizer now tokenizes correctly
125 * IndexSearcher should call TopDocs with a hashref and named keys if no scorer.
126
127 1.01 - Tue Feb 03 2004
128
129 * Doc fixes to Plucene.pm, bin/dump_index and README
130 * Remove Lucene::QueryParser from the tests
131 * Added this Changes file.
132
133 1.0 - Mon Feb 02 2004
134
135 * Initial release to CPAN