ferret 0.9.6 → 0.10.0
Sign up to get free protection for your applications and to get access to all the features.
- data/MIT-LICENSE +1 -1
- data/README +12 -24
- data/Rakefile +38 -54
- data/TODO +14 -17
- data/ext/analysis.c +982 -823
- data/ext/analysis.h +133 -76
- data/ext/array.c +96 -58
- data/ext/array.h +40 -13
- data/ext/bitvector.c +476 -118
- data/ext/bitvector.h +264 -22
- data/ext/compound_io.c +217 -229
- data/ext/defines.h +49 -0
- data/ext/document.c +107 -317
- data/ext/document.h +31 -65
- data/ext/except.c +81 -36
- data/ext/except.h +117 -55
- data/ext/extconf.rb +2 -9
- data/ext/ferret.c +211 -104
- data/ext/ferret.h +22 -11
- data/ext/filter.c +97 -82
- data/ext/fs_store.c +348 -367
- data/ext/global.c +226 -188
- data/ext/global.h +44 -26
- data/ext/hash.c +474 -391
- data/ext/hash.h +441 -68
- data/ext/hashset.c +124 -96
- data/ext/hashset.h +169 -20
- data/ext/helper.c +56 -5
- data/ext/helper.h +7 -0
- data/ext/inc/lang.h +29 -49
- data/ext/inc/threading.h +31 -0
- data/ext/ind.c +288 -278
- data/ext/ind.h +68 -0
- data/ext/index.c +5688 -0
- data/ext/index.h +663 -616
- data/ext/lang.h +29 -49
- data/ext/libstemmer.c +3 -3
- data/ext/mem_pool.c +84 -0
- data/ext/mem_pool.h +35 -0
- data/ext/posh.c +1006 -0
- data/ext/posh.h +1007 -0
- data/ext/priorityqueue.c +117 -194
- data/ext/priorityqueue.h +135 -39
- data/ext/q_boolean.c +1305 -1108
- data/ext/q_const_score.c +106 -93
- data/ext/q_filtered_query.c +138 -135
- data/ext/q_fuzzy.c +206 -242
- data/ext/q_match_all.c +94 -80
- data/ext/q_multi_term.c +663 -0
- data/ext/q_parser.c +667 -593
- data/ext/q_phrase.c +992 -555
- data/ext/q_prefix.c +72 -61
- data/ext/q_range.c +235 -210
- data/ext/q_span.c +1480 -1166
- data/ext/q_term.c +273 -246
- data/ext/q_wildcard.c +127 -114
- data/ext/r_analysis.c +1720 -711
- data/ext/r_index.c +3049 -0
- data/ext/r_qparser.c +433 -146
- data/ext/r_search.c +2934 -1993
- data/ext/r_store.c +372 -143
- data/ext/r_utils.c +941 -0
- data/ext/ram_store.c +330 -326
- data/ext/search.c +1291 -668
- data/ext/search.h +403 -702
- data/ext/similarity.c +91 -113
- data/ext/similarity.h +45 -30
- data/ext/sort.c +721 -484
- data/ext/stopwords.c +361 -273
- data/ext/store.c +556 -58
- data/ext/store.h +706 -126
- data/ext/tags +3578 -2780
- data/ext/term_vectors.c +352 -0
- data/ext/threading.h +31 -0
- data/ext/win32.h +54 -0
- data/lib/ferret.rb +5 -17
- data/lib/ferret/document.rb +130 -2
- data/lib/ferret/index.rb +577 -26
- data/lib/ferret/number_tools.rb +157 -0
- data/lib/ferret_version.rb +3 -0
- data/test/test_helper.rb +5 -13
- data/test/unit/analysis/tc_analyzer.rb +513 -1
- data/test/unit/analysis/{ctc_tokenstream.rb → tc_token_stream.rb} +23 -0
- data/test/unit/index/tc_index.rb +183 -240
- data/test/unit/index/tc_index_reader.rb +312 -479
- data/test/unit/index/tc_index_writer.rb +397 -13
- data/test/unit/index/th_doc.rb +269 -206
- data/test/unit/query_parser/tc_query_parser.rb +40 -33
- data/test/unit/search/tc_filter.rb +59 -71
- data/test/unit/search/tc_fuzzy_query.rb +24 -16
- data/test/unit/search/tc_index_searcher.rb +23 -201
- data/test/unit/search/tc_multi_searcher.rb +78 -226
- data/test/unit/search/tc_search_and_sort.rb +93 -81
- data/test/unit/search/tc_sort.rb +23 -23
- data/test/unit/search/tc_sort_field.rb +7 -7
- data/test/unit/search/tc_spans.rb +51 -47
- data/test/unit/search/tm_searcher.rb +339 -0
- data/test/unit/store/tc_fs_store.rb +1 -1
- data/test/unit/store/tm_store_lock.rb +3 -3
- data/test/unit/tc_document.rb +81 -0
- data/test/unit/ts_analysis.rb +1 -1
- data/test/unit/ts_utils.rb +1 -1
- data/test/unit/utils/tc_bit_vector.rb +288 -0
- data/test/unit/utils/tc_number_tools.rb +117 -0
- data/test/unit/utils/tc_priority_queue.rb +106 -0
- metadata +140 -301
- data/CHANGELOG +0 -9
- data/ext/dummy.exe +0 -0
- data/ext/field.c +0 -408
- data/ext/frtio.h +0 -13
- data/ext/inc/except.h +0 -90
- data/ext/index_io.c +0 -382
- data/ext/index_rw.c +0 -2658
- data/ext/lang.c +0 -41
- data/ext/nix_io.c +0 -134
- data/ext/q_multi_phrase.c +0 -380
- data/ext/r_doc.c +0 -582
- data/ext/r_index_io.c +0 -1021
- data/ext/r_term.c +0 -219
- data/ext/term.c +0 -820
- data/ext/termdocs.c +0 -611
- data/ext/vector.c +0 -637
- data/ext/w32_io.c +0 -150
- data/lib/ferret/analysis.rb +0 -11
- data/lib/ferret/analysis/analyzers.rb +0 -112
- data/lib/ferret/analysis/standard_tokenizer.rb +0 -71
- data/lib/ferret/analysis/token.rb +0 -100
- data/lib/ferret/analysis/token_filters.rb +0 -86
- data/lib/ferret/analysis/token_stream.rb +0 -26
- data/lib/ferret/analysis/tokenizers.rb +0 -112
- data/lib/ferret/analysis/word_list_loader.rb +0 -27
- data/lib/ferret/document/document.rb +0 -152
- data/lib/ferret/document/field.rb +0 -312
- data/lib/ferret/index/compound_file_io.rb +0 -338
- data/lib/ferret/index/document_writer.rb +0 -289
- data/lib/ferret/index/field_infos.rb +0 -279
- data/lib/ferret/index/fields_io.rb +0 -181
- data/lib/ferret/index/index.rb +0 -675
- data/lib/ferret/index/index_file_names.rb +0 -33
- data/lib/ferret/index/index_reader.rb +0 -503
- data/lib/ferret/index/index_writer.rb +0 -534
- data/lib/ferret/index/multi_reader.rb +0 -377
- data/lib/ferret/index/multiple_term_doc_pos_enum.rb +0 -98
- data/lib/ferret/index/segment_infos.rb +0 -130
- data/lib/ferret/index/segment_merge_info.rb +0 -49
- data/lib/ferret/index/segment_merge_queue.rb +0 -16
- data/lib/ferret/index/segment_merger.rb +0 -358
- data/lib/ferret/index/segment_reader.rb +0 -412
- data/lib/ferret/index/segment_term_enum.rb +0 -169
- data/lib/ferret/index/segment_term_vector.rb +0 -58
- data/lib/ferret/index/term.rb +0 -53
- data/lib/ferret/index/term_buffer.rb +0 -83
- data/lib/ferret/index/term_doc_enum.rb +0 -291
- data/lib/ferret/index/term_enum.rb +0 -52
- data/lib/ferret/index/term_info.rb +0 -37
- data/lib/ferret/index/term_infos_io.rb +0 -321
- data/lib/ferret/index/term_vector_offset_info.rb +0 -20
- data/lib/ferret/index/term_vectors_io.rb +0 -553
- data/lib/ferret/query_parser.rb +0 -312
- data/lib/ferret/query_parser/query_parser.tab.rb +0 -928
- data/lib/ferret/search.rb +0 -50
- data/lib/ferret/search/boolean_clause.rb +0 -100
- data/lib/ferret/search/boolean_query.rb +0 -299
- data/lib/ferret/search/boolean_scorer.rb +0 -294
- data/lib/ferret/search/caching_wrapper_filter.rb +0 -40
- data/lib/ferret/search/conjunction_scorer.rb +0 -99
- data/lib/ferret/search/disjunction_sum_scorer.rb +0 -205
- data/lib/ferret/search/exact_phrase_scorer.rb +0 -32
- data/lib/ferret/search/explanation.rb +0 -41
- data/lib/ferret/search/field_cache.rb +0 -215
- data/lib/ferret/search/field_doc.rb +0 -31
- data/lib/ferret/search/field_sorted_hit_queue.rb +0 -184
- data/lib/ferret/search/filter.rb +0 -11
- data/lib/ferret/search/filtered_query.rb +0 -130
- data/lib/ferret/search/filtered_term_enum.rb +0 -79
- data/lib/ferret/search/fuzzy_query.rb +0 -154
- data/lib/ferret/search/fuzzy_term_enum.rb +0 -247
- data/lib/ferret/search/hit_collector.rb +0 -34
- data/lib/ferret/search/hit_queue.rb +0 -11
- data/lib/ferret/search/index_searcher.rb +0 -200
- data/lib/ferret/search/match_all_query.rb +0 -104
- data/lib/ferret/search/multi_phrase_query.rb +0 -216
- data/lib/ferret/search/multi_searcher.rb +0 -261
- data/lib/ferret/search/multi_term_query.rb +0 -65
- data/lib/ferret/search/non_matching_scorer.rb +0 -22
- data/lib/ferret/search/phrase_positions.rb +0 -55
- data/lib/ferret/search/phrase_query.rb +0 -214
- data/lib/ferret/search/phrase_scorer.rb +0 -152
- data/lib/ferret/search/prefix_query.rb +0 -54
- data/lib/ferret/search/query.rb +0 -140
- data/lib/ferret/search/query_filter.rb +0 -51
- data/lib/ferret/search/range_filter.rb +0 -103
- data/lib/ferret/search/range_query.rb +0 -139
- data/lib/ferret/search/req_excl_scorer.rb +0 -125
- data/lib/ferret/search/req_opt_sum_scorer.rb +0 -70
- data/lib/ferret/search/score_doc.rb +0 -38
- data/lib/ferret/search/score_doc_comparator.rb +0 -114
- data/lib/ferret/search/scorer.rb +0 -91
- data/lib/ferret/search/similarity.rb +0 -278
- data/lib/ferret/search/sloppy_phrase_scorer.rb +0 -47
- data/lib/ferret/search/sort.rb +0 -112
- data/lib/ferret/search/sort_comparator.rb +0 -60
- data/lib/ferret/search/sort_field.rb +0 -91
- data/lib/ferret/search/spans.rb +0 -12
- data/lib/ferret/search/spans/near_spans_enum.rb +0 -304
- data/lib/ferret/search/spans/span_first_query.rb +0 -79
- data/lib/ferret/search/spans/span_near_query.rb +0 -108
- data/lib/ferret/search/spans/span_not_query.rb +0 -130
- data/lib/ferret/search/spans/span_or_query.rb +0 -176
- data/lib/ferret/search/spans/span_query.rb +0 -25
- data/lib/ferret/search/spans/span_scorer.rb +0 -74
- data/lib/ferret/search/spans/span_term_query.rb +0 -105
- data/lib/ferret/search/spans/span_weight.rb +0 -84
- data/lib/ferret/search/spans/spans_enum.rb +0 -44
- data/lib/ferret/search/term_query.rb +0 -128
- data/lib/ferret/search/term_scorer.rb +0 -183
- data/lib/ferret/search/top_docs.rb +0 -36
- data/lib/ferret/search/top_field_docs.rb +0 -17
- data/lib/ferret/search/weight.rb +0 -54
- data/lib/ferret/search/wildcard_query.rb +0 -26
- data/lib/ferret/search/wildcard_term_enum.rb +0 -61
- data/lib/ferret/stemmers.rb +0 -1
- data/lib/ferret/stemmers/porter_stemmer.rb +0 -218
- data/lib/ferret/store.rb +0 -5
- data/lib/ferret/store/buffered_index_io.rb +0 -190
- data/lib/ferret/store/directory.rb +0 -141
- data/lib/ferret/store/fs_store.rb +0 -381
- data/lib/ferret/store/index_io.rb +0 -245
- data/lib/ferret/store/ram_store.rb +0 -286
- data/lib/ferret/utils.rb +0 -8
- data/lib/ferret/utils/bit_vector.rb +0 -123
- data/lib/ferret/utils/date_tools.rb +0 -138
- data/lib/ferret/utils/number_tools.rb +0 -91
- data/lib/ferret/utils/parameter.rb +0 -41
- data/lib/ferret/utils/priority_queue.rb +0 -120
- data/lib/ferret/utils/string_helper.rb +0 -47
- data/lib/ferret/utils/thread_local.rb +0 -28
- data/lib/ferret/utils/weak_key_hash.rb +0 -60
- data/lib/rferret.rb +0 -37
- data/rake_utils/code_statistics.rb +0 -106
- data/test/benchmark/tb_ram_store.rb +0 -76
- data/test/benchmark/tb_rw_vint.rb +0 -26
- data/test/functional/thread_safety_index_test.rb +0 -81
- data/test/functional/thread_safety_test.rb +0 -137
- data/test/longrunning/tc_numbertools.rb +0 -60
- data/test/longrunning/tm_store.rb +0 -19
- data/test/unit/analysis/ctc_analyzer.rb +0 -532
- data/test/unit/analysis/data/wordfile +0 -6
- data/test/unit/analysis/rtc_letter_tokenizer.rb +0 -20
- data/test/unit/analysis/rtc_lower_case_filter.rb +0 -20
- data/test/unit/analysis/rtc_lower_case_tokenizer.rb +0 -27
- data/test/unit/analysis/rtc_per_field_analyzer_wrapper.rb +0 -39
- data/test/unit/analysis/rtc_porter_stem_filter.rb +0 -16
- data/test/unit/analysis/rtc_standard_analyzer.rb +0 -20
- data/test/unit/analysis/rtc_standard_tokenizer.rb +0 -20
- data/test/unit/analysis/rtc_stop_analyzer.rb +0 -20
- data/test/unit/analysis/rtc_stop_filter.rb +0 -14
- data/test/unit/analysis/rtc_white_space_analyzer.rb +0 -21
- data/test/unit/analysis/rtc_white_space_tokenizer.rb +0 -20
- data/test/unit/analysis/rtc_word_list_loader.rb +0 -32
- data/test/unit/analysis/tc_token.rb +0 -25
- data/test/unit/document/rtc_field.rb +0 -28
- data/test/unit/document/tc_document.rb +0 -47
- data/test/unit/document/tc_field.rb +0 -98
- data/test/unit/index/rtc_compound_file_io.rb +0 -107
- data/test/unit/index/rtc_field_infos.rb +0 -127
- data/test/unit/index/rtc_fields_io.rb +0 -167
- data/test/unit/index/rtc_multiple_term_doc_pos_enum.rb +0 -83
- data/test/unit/index/rtc_segment_infos.rb +0 -74
- data/test/unit/index/rtc_segment_term_docs.rb +0 -17
- data/test/unit/index/rtc_segment_term_enum.rb +0 -60
- data/test/unit/index/rtc_segment_term_vector.rb +0 -71
- data/test/unit/index/rtc_term_buffer.rb +0 -57
- data/test/unit/index/rtc_term_info.rb +0 -19
- data/test/unit/index/rtc_term_infos_io.rb +0 -192
- data/test/unit/index/rtc_term_vectors_io.rb +0 -108
- data/test/unit/index/tc_term.rb +0 -27
- data/test/unit/index/tc_term_voi.rb +0 -18
- data/test/unit/search/rtc_similarity.rb +0 -37
- data/test/unit/search/rtc_sort_field.rb +0 -14
- data/test/unit/search/tc_multi_searcher2.rb +0 -126
- data/test/unit/store/rtc_fs_store.rb +0 -62
- data/test/unit/store/rtc_ram_store.rb +0 -15
- data/test/unit/store/rtm_store.rb +0 -150
- data/test/unit/store/rtm_store_lock.rb +0 -2
- data/test/unit/ts_document.rb +0 -2
- data/test/unit/utils/rtc_bit_vector.rb +0 -73
- data/test/unit/utils/rtc_date_tools.rb +0 -50
- data/test/unit/utils/rtc_number_tools.rb +0 -59
- data/test/unit/utils/rtc_parameter.rb +0 -40
- data/test/unit/utils/rtc_priority_queue.rb +0 -62
- data/test/unit/utils/rtc_string_helper.rb +0 -21
- data/test/unit/utils/rtc_thread.rb +0 -61
- data/test/unit/utils/rtc_weak_key_hash.rb +0 -25
- data/test/utils/number_to_spoken.rb +0 -132
data/lib/ferret/utils.rb
DELETED
@@ -1,8 +0,0 @@
|
|
1
|
-
require 'ferret/utils/string_helper'
|
2
|
-
require 'ferret/utils/parameter'
|
3
|
-
require 'ferret/utils/priority_queue'
|
4
|
-
require 'ferret/utils/bit_vector'
|
5
|
-
require 'ferret/utils/date_tools'
|
6
|
-
require 'ferret/utils/number_tools'
|
7
|
-
require 'ferret/utils/weak_key_hash'
|
8
|
-
require 'ferret/utils/thread_local'
|
@@ -1,123 +0,0 @@
|
|
1
|
-
module Ferret::Utils
|
2
|
-
# Optimized implementation of a vector of bits.
|
3
|
-
#
|
4
|
-
# * a count() method, which efficiently computes the number of one bits
|
5
|
-
# * optimized read from and write to disk
|
6
|
-
# * inlinable get() method
|
7
|
-
class BitVector
|
8
|
-
attr_reader :size
|
9
|
-
attr_accessor :bits
|
10
|
-
|
11
|
-
def initialize
|
12
|
-
@bits = 0
|
13
|
-
@count = -1
|
14
|
-
end
|
15
|
-
|
16
|
-
# Sets the value of _bit_ to one.
|
17
|
-
def set(bit)
|
18
|
-
@bits |= 1 << bit
|
19
|
-
@count = -1
|
20
|
-
end
|
21
|
-
|
22
|
-
# Sets the value of _bit_ to zero.
|
23
|
-
def clear(bit)
|
24
|
-
@bits &= ~(1 << bit)
|
25
|
-
@count = -1
|
26
|
-
end
|
27
|
-
|
28
|
-
# Returns _true_ if _bit_ is one and
|
29
|
-
# _false_ if it is zero.
|
30
|
-
def get(bit)
|
31
|
-
return (@bits & (1 << bit)) != 0
|
32
|
-
end
|
33
|
-
alias :[] :get
|
34
|
-
|
35
|
-
# Returns the total number of one bits in this vector. This is
|
36
|
-
# efficiently computed and cached, so that, if the vector is not
|
37
|
-
# changed, no recomputation is done for repeated calls.
|
38
|
-
def count()
|
39
|
-
# if the vector has been modified
|
40
|
-
if (@count == -1)
|
41
|
-
c = 0
|
42
|
-
tmp = @bits
|
43
|
-
while tmp > 0
|
44
|
-
c += BYTE_COUNTS[tmp & 0xFF] # sum bits per byte
|
45
|
-
tmp >>= 8
|
46
|
-
end
|
47
|
-
@count = c
|
48
|
-
end
|
49
|
-
return @count
|
50
|
-
end
|
51
|
-
|
52
|
-
BYTE_COUNTS = [ # table of bits/byte
|
53
|
-
0, 1, 1, 2, 1, 2, 2, 3, 1, 2, 2, 3, 2, 3, 3, 4,
|
54
|
-
1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,
|
55
|
-
1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,
|
56
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
57
|
-
1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,
|
58
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
59
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
60
|
-
3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,
|
61
|
-
1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,
|
62
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
63
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
64
|
-
3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,
|
65
|
-
2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,
|
66
|
-
3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,
|
67
|
-
3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,
|
68
|
-
4, 5, 5, 6, 5, 6, 6, 7, 5, 6, 6, 7, 6, 7, 7, 8
|
69
|
-
]
|
70
|
-
|
71
|
-
|
72
|
-
# Writes this vector to the file _name_ in Directory _d_, in a format
|
73
|
-
# that can be read by the constructor
|
74
|
-
def write(d, name)
|
75
|
-
output = d.create_output(name)
|
76
|
-
begin
|
77
|
-
output.write_string(self.class.bignum_to_string(@bits))
|
78
|
-
ensure
|
79
|
-
output.close()
|
80
|
-
end
|
81
|
-
end
|
82
|
-
|
83
|
-
# Constructs a bit vector from the file _name_ in Directory _d_, as
|
84
|
-
# written by the @link #writeendmethod.
|
85
|
-
def BitVector.read(d, name)
|
86
|
-
bv = BitVector.new
|
87
|
-
input = d.open_input(name)
|
88
|
-
begin
|
89
|
-
bv.bits = string_to_bignum(input.read_string())
|
90
|
-
ensure
|
91
|
-
input.close()
|
92
|
-
end
|
93
|
-
return bv
|
94
|
-
end
|
95
|
-
|
96
|
-
def to_s
|
97
|
-
i = @bits
|
98
|
-
while i > 0
|
99
|
-
print(i&1)
|
100
|
-
i >>= 1
|
101
|
-
end
|
102
|
-
puts ""
|
103
|
-
end
|
104
|
-
|
105
|
-
# converts a BigNum into a string
|
106
|
-
def BitVector.bignum_to_string(num)
|
107
|
-
str = []
|
108
|
-
while (num > 0)
|
109
|
-
str << (num & 0xff)
|
110
|
-
num >>= 8
|
111
|
-
end
|
112
|
-
return str.pack("C*")
|
113
|
-
end
|
114
|
-
|
115
|
-
# converts a string into a bignum
|
116
|
-
def BitVector.string_to_bignum(str)
|
117
|
-
str = str.unpack("C*")
|
118
|
-
num = 0
|
119
|
-
str.reverse.each {|c| num = ((num << 8) | c) }
|
120
|
-
return num
|
121
|
-
end
|
122
|
-
end
|
123
|
-
end
|
@@ -1,138 +0,0 @@
|
|
1
|
-
require 'date'
|
2
|
-
module Ferret::Utils
|
3
|
-
# Provides support for converting dates to strings and vice-versa. The
|
4
|
-
# strings are structured so that lexicographic sorting orders them by
|
5
|
-
# date, which makes them suitable for use as field values and search
|
6
|
-
# terms.
|
7
|
-
#
|
8
|
-
# This class also helps you to limit the resolution of your dates. Do not
|
9
|
-
# save dates with a finer resolution than you really need, as then
|
10
|
-
# RangeQuery and PrefixQuery will require more memory and become slower.
|
11
|
-
#
|
12
|
-
# Compared to the serialize methods the strings generated by the to_s
|
13
|
-
# methods in this class take slightly more space, unless your selected
|
14
|
-
# resolution is set to _Resolution.DAY_ or lower.
|
15
|
-
|
16
|
-
# Provides support for converting dates to strings and vice-versa. The
|
17
|
-
# strings are structured so that lexicographic sorting orders by date,
|
18
|
-
# which makes them suitable for use as field values and search terms.
|
19
|
-
#
|
20
|
-
# Note:: dates before 1970 cannot be used, and therefore cannot be indexed
|
21
|
-
# when using this class.
|
22
|
-
module DateTools
|
23
|
-
# make date strings long enough to last a millenium
|
24
|
-
SERIALIZED_DATE_LEN = (1000*365*24*60*60*1000).to_s(36).length
|
25
|
-
|
26
|
-
# The latest date that can be stored in this format
|
27
|
-
MAX_SERIALIZED_DATE_STRING = Array.new(SERIALIZED_DATE_LEN, "z").to_s.to_i(36)
|
28
|
-
|
29
|
-
# Converts a Date to a string suitable for indexing. Throws Exception
|
30
|
-
# if the date specified in the method argument is before 1970 This
|
31
|
-
# method is unsupported. Please use Time instead of Date
|
32
|
-
def DateTools.serialize_date(date)
|
33
|
-
return serialize_time(Time.parse(date))
|
34
|
-
end
|
35
|
-
|
36
|
-
# Converts a millisecond time to a string suitable for indexing.
|
37
|
-
# Accepts a Time object or a time in milliseconds.
|
38
|
-
#
|
39
|
-
# Throws Exception if the time specified in the method argument is
|
40
|
-
# negative, that is, before 1970 It is recommended that you store the
|
41
|
-
# date as a string if you don't need the time to the nearest
|
42
|
-
# millisecond. That makes things a lot easier.
|
43
|
-
def DateTools.serialize_time(time)
|
44
|
-
if time.instance_of?(Time) then time = time.to_i end
|
45
|
-
|
46
|
-
if (time < 0) then raise("time too early") end
|
47
|
-
|
48
|
-
# convert to milliseconds before serialization
|
49
|
-
s = (time*1000).to_s(36)
|
50
|
-
|
51
|
-
if (s.length() > SERIALIZED_DATE_LEN) then raise("time too late") end
|
52
|
-
|
53
|
-
# pad to 16 charactors
|
54
|
-
s = "0" + s while (s.length() < SERIALIZED_DATE_LEN)
|
55
|
-
|
56
|
-
return s
|
57
|
-
end
|
58
|
-
|
59
|
-
# The earliest date that can be stored in this format.
|
60
|
-
MIN_SERIALIZED_DATE_STRING = DateTools.serialize_time(0)
|
61
|
-
|
62
|
-
# Converts a string-encoded date into a millisecond time.
|
63
|
-
def DateTools.deserialize_time(s)
|
64
|
-
# remember to convert back to seconds
|
65
|
-
return Time.at(s.to_i(36)/1000)
|
66
|
-
end
|
67
|
-
|
68
|
-
def DateTools.date_to_s(date, resolution = Resolution::MILLISECOND)
|
69
|
-
return time_to_s(Time.parse(date), resolution)
|
70
|
-
end
|
71
|
-
|
72
|
-
|
73
|
-
# Converts a millisecond time to a string suitable for indexing.
|
74
|
-
#
|
75
|
-
# time:: the date expressed as milliseconds since January 1, 1970,
|
76
|
-
# 00:00:00 GMT resolution:: the desired resolution, see
|
77
|
-
# #round(long, DateTools.Resolution)
|
78
|
-
# return:: a string in format _%Y%m%d%H%M%SSSS_ or shorter,
|
79
|
-
# depending on _resolution_
|
80
|
-
def DateTools.time_to_s(time, resolution = Resolution::MILLISECOND)
|
81
|
-
if time.instance_of?(Date) then time = Time.parse(time) end
|
82
|
-
suffix = ""
|
83
|
-
if (resolution == Resolution::MILLISECOND)
|
84
|
-
# the suffix is the number of milliseconds if needed.
|
85
|
-
suffix = ((time.to_f-time.to_f.floor)*1000).round.to_s
|
86
|
-
end
|
87
|
-
return time.strftime(resolution.format) + suffix
|
88
|
-
end
|
89
|
-
|
90
|
-
# Converts a string produced by _time_to_s_ or _date_to_s_ back to a
|
91
|
-
# time, represented as the number of milliseconds since January 1, 1970,
|
92
|
-
# 00:00:00 GMT.
|
93
|
-
#
|
94
|
-
# str:: the date string to be converted
|
95
|
-
# return:: the number of milliseconds since January 1, 1970, 00:00:00GMT
|
96
|
-
def DateTools.s_to_time(str)
|
97
|
-
year = str.size >= 4 ? str[ 0.. 3].to_i : nil
|
98
|
-
month = str.size >= 6 ? str[ 4.. 5].to_i : nil
|
99
|
-
day = str.size >= 8 ? str[ 6.. 7].to_i : nil
|
100
|
-
hour = str.size >= 10 ? str[ 8.. 9].to_i : nil
|
101
|
-
minute = str.size >= 12 ? str[10..11].to_i : nil
|
102
|
-
second = str.size >= 14 ? str[12..13].to_i : nil
|
103
|
-
microsecond = str.size >= 17 ? str[14..17].to_i*1000 : nil
|
104
|
-
return Time.mktime(year, month, day, hour, minute, second, microsecond)
|
105
|
-
end
|
106
|
-
|
107
|
-
# Limit a date's resolution. For example, the date _2004-09-21 13:50:11_
|
108
|
-
# will be changed to _2004-09-01 00:00:00_ when using
|
109
|
-
# _Resolution.MONTH_.
|
110
|
-
#
|
111
|
-
# resolution:: The desired resolution of the date to be returned
|
112
|
-
# return:: the date with all values more precise than _resolution_
|
113
|
-
# set to 0 or 1
|
114
|
-
def DateTools.round(time, resolution)
|
115
|
-
return s_to_time(time_to_s(time, resolution))
|
116
|
-
end
|
117
|
-
|
118
|
-
class Resolution < Parameter
|
119
|
-
attr_accessor :format
|
120
|
-
|
121
|
-
private :initialize
|
122
|
-
|
123
|
-
def initialize(name, format)
|
124
|
-
super(name)
|
125
|
-
@format = format
|
126
|
-
end
|
127
|
-
|
128
|
-
YEAR = Resolution.new("year", "%Y")
|
129
|
-
MONTH = Resolution.new("month", "%Y%m")
|
130
|
-
DAY = Resolution.new("day", "%Y%m%d")
|
131
|
-
HOUR = Resolution.new("hour", "%Y%m%d%H")
|
132
|
-
MINUTE = Resolution.new("minute", "%Y%m%d%H%M")
|
133
|
-
SECOND = Resolution.new("second", "%Y%m%d%H%M%S")
|
134
|
-
MILLISECOND = Resolution.new("millisecond", "%Y%m%d%H%M%S")
|
135
|
-
|
136
|
-
end
|
137
|
-
end
|
138
|
-
end
|
@@ -1,91 +0,0 @@
|
|
1
|
-
class Float
|
2
|
-
def =~(o)
|
3
|
-
return (1 - self/o).abs < 0.0000000001
|
4
|
-
end
|
5
|
-
end
|
6
|
-
|
7
|
-
module Ferret::Utils
|
8
|
-
# Provides support for converting longs to Strings, and back again. The
|
9
|
-
# strings are structured so that lexicographic sorting order is preserved.
|
10
|
-
#
|
11
|
-
# That is, if long1 is less than long2 for any two longs long1 and long2,
|
12
|
-
# then NumberTools.long_to_s(long1) is lexicographically less than
|
13
|
-
# NumberTools.long_to_s(long2). (Similarly for "greater than" and "equals".)
|
14
|
-
#
|
15
|
-
# This class handles all long values
|
16
|
-
module NumberTools
|
17
|
-
RADIX = 36
|
18
|
-
NEGATIVE_PREFIX = '-'
|
19
|
-
|
20
|
-
# NB: NEGATIVE_PREFIX must be < POSITIVE_PREFIX
|
21
|
-
POSITIVE_PREFIX = '0'
|
22
|
-
|
23
|
-
# The following constants are from Java
|
24
|
-
LONG_MAX_VALUE = 9223372036854775807
|
25
|
-
LONG_MIN_VALUE = -9223372036854775808
|
26
|
-
|
27
|
-
# NB: This function is used to match the java equivalent. Actually
|
28
|
-
# ruby allows much larger numbers than Java so this is just so that we
|
29
|
-
# can read the Java Lucene created indexes.
|
30
|
-
MIN_STRING_VALUE = NEGATIVE_PREFIX + "0000000000000"
|
31
|
-
MAX_STRING_VALUE = POSITIVE_PREFIX + "1y2p0ij32e8e7"
|
32
|
-
|
33
|
-
# The length of the long field
|
34
|
-
STR_SIZE = MIN_STRING_VALUE.length()
|
35
|
-
|
36
|
-
# Converts a long to a String suitable for indexing.
|
37
|
-
def NumberTools.long_to_s(l)
|
38
|
-
if (l == LONG_MIN_VALUE)
|
39
|
-
# special case, because long is not symetric around zero
|
40
|
-
return MIN_STRING_VALUE;
|
41
|
-
end
|
42
|
-
|
43
|
-
s = ""
|
44
|
-
if (l < 0)
|
45
|
-
s << NEGATIVE_PREFIX
|
46
|
-
l = LONG_MAX_VALUE + l + 1
|
47
|
-
else
|
48
|
-
s << POSITIVE_PREFIX
|
49
|
-
end
|
50
|
-
num = l.to_s(RADIX)
|
51
|
-
|
52
|
-
pad_len = STR_SIZE - num.length() - s.length()
|
53
|
-
while ((pad_len -= 1) >= 0)
|
54
|
-
s << '0'
|
55
|
-
end
|
56
|
-
s << num
|
57
|
-
|
58
|
-
return s
|
59
|
-
end
|
60
|
-
|
61
|
-
# Converts a String that was returned by #long_to_s back to a long.
|
62
|
-
#
|
63
|
-
# Throws:: ArgumentError if the input is nil
|
64
|
-
def NumberTools.s_to_long(s)
|
65
|
-
if (s == nil)
|
66
|
-
raise ArgumentError, "string cannot be nil"
|
67
|
-
end
|
68
|
-
if (s.length() != STR_SIZE)
|
69
|
-
raise ArgumentError, "string is the wrong size"
|
70
|
-
end
|
71
|
-
|
72
|
-
if (s == MIN_STRING_VALUE)
|
73
|
-
return LONG_MIN_VALUE
|
74
|
-
end
|
75
|
-
|
76
|
-
prefix = s[0,1]
|
77
|
-
l = s[1..-1].to_i(36)
|
78
|
-
|
79
|
-
if (prefix == POSITIVE_PREFIX)
|
80
|
-
# nop
|
81
|
-
elsif (prefix == NEGATIVE_PREFIX)
|
82
|
-
l = l - LONG_MAX_VALUE - 1
|
83
|
-
else
|
84
|
-
raise ArgumentError, "string <" + prefix +
|
85
|
-
"> does not begin with the correct prefix"
|
86
|
-
end
|
87
|
-
|
88
|
-
return l
|
89
|
-
end
|
90
|
-
end
|
91
|
-
end
|
@@ -1,41 +0,0 @@
|
|
1
|
-
module Ferret::Utils
|
2
|
-
class Parameter
|
3
|
-
def to_s() return @name end
|
4
|
-
|
5
|
-
def _dump(arg)
|
6
|
-
@name
|
7
|
-
end
|
8
|
-
|
9
|
-
def Parameter._load(var)
|
10
|
-
name = var
|
11
|
-
key = make_key(name)
|
12
|
-
if (@@all_parameters.has_key?(key))
|
13
|
-
return @@all_parameters[key]
|
14
|
-
else
|
15
|
-
return self.new(name)
|
16
|
-
end
|
17
|
-
end
|
18
|
-
|
19
|
-
def hash
|
20
|
-
return self.class.make_key(@name).hash
|
21
|
-
end
|
22
|
-
|
23
|
-
protected
|
24
|
-
@@all_parameters = {}
|
25
|
-
|
26
|
-
def initialize(name)
|
27
|
-
@name = name
|
28
|
-
key = self.class.make_key(name)
|
29
|
-
|
30
|
-
if (@@all_parameters.has_key?(key))
|
31
|
-
raise ArgumentError, "key already in use"
|
32
|
-
end
|
33
|
-
|
34
|
-
@@all_parameters[key] = self
|
35
|
-
end
|
36
|
-
|
37
|
-
def Parameter.make_key(name)
|
38
|
-
return self.to_s + " " + name
|
39
|
-
end
|
40
|
-
end
|
41
|
-
end
|
@@ -1,120 +0,0 @@
|
|
1
|
-
module Ferret::Utils
|
2
|
-
# A PriorityQueue maintains a partial ordering of its objects such that
|
3
|
-
# the least object can always be found in constant time. push()'s and
|
4
|
-
# pop()'s require log(size) time. The objects in this priority queue must
|
5
|
-
# be Comparable
|
6
|
-
class PriorityQueue
|
7
|
-
attr_reader :size
|
8
|
-
|
9
|
-
def less_than(a, b)
|
10
|
-
a < b
|
11
|
-
end
|
12
|
-
|
13
|
-
# Subclass constructors must call this.
|
14
|
-
def initialize(max_size)
|
15
|
-
@size = 0
|
16
|
-
@heap = Array.new(max_size + 1)
|
17
|
-
@max_size = max_size
|
18
|
-
end
|
19
|
-
|
20
|
-
# Adds an Object to a PriorityQueue in log(size) time.
|
21
|
-
#
|
22
|
-
# If one tries to add more objects than max_size from initialize a
|
23
|
-
# RuntimeException (ArrayIndexOutOfBound) is thrown.
|
24
|
-
def push(object)
|
25
|
-
@size += 1
|
26
|
-
@heap[@size] = object
|
27
|
-
up_heap()
|
28
|
-
end
|
29
|
-
alias :<< :push
|
30
|
-
|
31
|
-
# Adds object to the PriorityQueue in log(size) time if either the
|
32
|
-
# PriorityQueue is not full, or not less_than(object, top()).
|
33
|
-
#
|
34
|
-
# object:: the object to be inserted
|
35
|
-
# return true if object is added, false otherwise.
|
36
|
-
def insert(object)
|
37
|
-
if(@size < @max_size)
|
38
|
-
push(object)
|
39
|
-
return true
|
40
|
-
elsif (@size > 0 and less_than(top, object))
|
41
|
-
@heap[1] = object
|
42
|
-
down_heap()
|
43
|
-
return true
|
44
|
-
else
|
45
|
-
return false
|
46
|
-
end
|
47
|
-
end
|
48
|
-
|
49
|
-
# Returns the least object of the PriorityQueue in constant time.
|
50
|
-
def top
|
51
|
-
return @heap[1]
|
52
|
-
end
|
53
|
-
|
54
|
-
# Removes and returns the least object of the PriorityQueue in log(size)
|
55
|
-
# time.
|
56
|
-
def pop()
|
57
|
-
if (@size > 0)
|
58
|
-
result = @heap[1] # save first value
|
59
|
-
@heap[1] = @heap[@size] # move last to first
|
60
|
-
@heap[@size] = nil; # permit GC of objects
|
61
|
-
@size -= 1
|
62
|
-
down_heap() # adjust heap
|
63
|
-
return result
|
64
|
-
else
|
65
|
-
return nil
|
66
|
-
end
|
67
|
-
end
|
68
|
-
|
69
|
-
# Removes all entries from the PriorityQueue.
|
70
|
-
def clear()
|
71
|
-
(1..@size).each do |i|
|
72
|
-
@heap[i] = nil
|
73
|
-
end
|
74
|
-
@size = 0
|
75
|
-
end
|
76
|
-
|
77
|
-
def put_heap
|
78
|
-
puts @heap
|
79
|
-
end
|
80
|
-
|
81
|
-
# resets the queue after the top has been changed
|
82
|
-
def adjust_top()
|
83
|
-
down_heap()
|
84
|
-
end
|
85
|
-
|
86
|
-
private
|
87
|
-
|
88
|
-
def up_heap()
|
89
|
-
i = @size
|
90
|
-
node = @heap[i] # save bottom node
|
91
|
-
j = i >> 1
|
92
|
-
while (j > 0 and less_than(node, @heap[j]))
|
93
|
-
@heap[i] = @heap[j]; # shift parents down
|
94
|
-
i = j
|
95
|
-
j = j >> 1
|
96
|
-
end
|
97
|
-
@heap[i] = node; # install saved node
|
98
|
-
end
|
99
|
-
|
100
|
-
def down_heap()
|
101
|
-
i = 1
|
102
|
-
node = @heap[i] # save top node
|
103
|
-
j = i << 1 # find smaller child
|
104
|
-
k = j + 1
|
105
|
-
if k <= @size and less_than(@heap[k], @heap[j])
|
106
|
-
j = k
|
107
|
-
end
|
108
|
-
while (j <= @size and less_than(@heap[j], node))
|
109
|
-
@heap[i] = @heap[j] # shift up child
|
110
|
-
i = j
|
111
|
-
j = i << 1
|
112
|
-
k = j + 1
|
113
|
-
if k <= @size and less_than(@heap[k], @heap[j])
|
114
|
-
j = k
|
115
|
-
end
|
116
|
-
end
|
117
|
-
@heap[i] = node; # install saved node
|
118
|
-
end
|
119
|
-
end
|
120
|
-
end
|