ferret 0.9.6 → 0.10.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (295) hide show
  1. data/MIT-LICENSE +1 -1
  2. data/README +12 -24
  3. data/Rakefile +38 -54
  4. data/TODO +14 -17
  5. data/ext/analysis.c +982 -823
  6. data/ext/analysis.h +133 -76
  7. data/ext/array.c +96 -58
  8. data/ext/array.h +40 -13
  9. data/ext/bitvector.c +476 -118
  10. data/ext/bitvector.h +264 -22
  11. data/ext/compound_io.c +217 -229
  12. data/ext/defines.h +49 -0
  13. data/ext/document.c +107 -317
  14. data/ext/document.h +31 -65
  15. data/ext/except.c +81 -36
  16. data/ext/except.h +117 -55
  17. data/ext/extconf.rb +2 -9
  18. data/ext/ferret.c +211 -104
  19. data/ext/ferret.h +22 -11
  20. data/ext/filter.c +97 -82
  21. data/ext/fs_store.c +348 -367
  22. data/ext/global.c +226 -188
  23. data/ext/global.h +44 -26
  24. data/ext/hash.c +474 -391
  25. data/ext/hash.h +441 -68
  26. data/ext/hashset.c +124 -96
  27. data/ext/hashset.h +169 -20
  28. data/ext/helper.c +56 -5
  29. data/ext/helper.h +7 -0
  30. data/ext/inc/lang.h +29 -49
  31. data/ext/inc/threading.h +31 -0
  32. data/ext/ind.c +288 -278
  33. data/ext/ind.h +68 -0
  34. data/ext/index.c +5688 -0
  35. data/ext/index.h +663 -616
  36. data/ext/lang.h +29 -49
  37. data/ext/libstemmer.c +3 -3
  38. data/ext/mem_pool.c +84 -0
  39. data/ext/mem_pool.h +35 -0
  40. data/ext/posh.c +1006 -0
  41. data/ext/posh.h +1007 -0
  42. data/ext/priorityqueue.c +117 -194
  43. data/ext/priorityqueue.h +135 -39
  44. data/ext/q_boolean.c +1305 -1108
  45. data/ext/q_const_score.c +106 -93
  46. data/ext/q_filtered_query.c +138 -135
  47. data/ext/q_fuzzy.c +206 -242
  48. data/ext/q_match_all.c +94 -80
  49. data/ext/q_multi_term.c +663 -0
  50. data/ext/q_parser.c +667 -593
  51. data/ext/q_phrase.c +992 -555
  52. data/ext/q_prefix.c +72 -61
  53. data/ext/q_range.c +235 -210
  54. data/ext/q_span.c +1480 -1166
  55. data/ext/q_term.c +273 -246
  56. data/ext/q_wildcard.c +127 -114
  57. data/ext/r_analysis.c +1720 -711
  58. data/ext/r_index.c +3049 -0
  59. data/ext/r_qparser.c +433 -146
  60. data/ext/r_search.c +2934 -1993
  61. data/ext/r_store.c +372 -143
  62. data/ext/r_utils.c +941 -0
  63. data/ext/ram_store.c +330 -326
  64. data/ext/search.c +1291 -668
  65. data/ext/search.h +403 -702
  66. data/ext/similarity.c +91 -113
  67. data/ext/similarity.h +45 -30
  68. data/ext/sort.c +721 -484
  69. data/ext/stopwords.c +361 -273
  70. data/ext/store.c +556 -58
  71. data/ext/store.h +706 -126
  72. data/ext/tags +3578 -2780
  73. data/ext/term_vectors.c +352 -0
  74. data/ext/threading.h +31 -0
  75. data/ext/win32.h +54 -0
  76. data/lib/ferret.rb +5 -17
  77. data/lib/ferret/document.rb +130 -2
  78. data/lib/ferret/index.rb +577 -26
  79. data/lib/ferret/number_tools.rb +157 -0
  80. data/lib/ferret_version.rb +3 -0
  81. data/test/test_helper.rb +5 -13
  82. data/test/unit/analysis/tc_analyzer.rb +513 -1
  83. data/test/unit/analysis/{ctc_tokenstream.rb → tc_token_stream.rb} +23 -0
  84. data/test/unit/index/tc_index.rb +183 -240
  85. data/test/unit/index/tc_index_reader.rb +312 -479
  86. data/test/unit/index/tc_index_writer.rb +397 -13
  87. data/test/unit/index/th_doc.rb +269 -206
  88. data/test/unit/query_parser/tc_query_parser.rb +40 -33
  89. data/test/unit/search/tc_filter.rb +59 -71
  90. data/test/unit/search/tc_fuzzy_query.rb +24 -16
  91. data/test/unit/search/tc_index_searcher.rb +23 -201
  92. data/test/unit/search/tc_multi_searcher.rb +78 -226
  93. data/test/unit/search/tc_search_and_sort.rb +93 -81
  94. data/test/unit/search/tc_sort.rb +23 -23
  95. data/test/unit/search/tc_sort_field.rb +7 -7
  96. data/test/unit/search/tc_spans.rb +51 -47
  97. data/test/unit/search/tm_searcher.rb +339 -0
  98. data/test/unit/store/tc_fs_store.rb +1 -1
  99. data/test/unit/store/tm_store_lock.rb +3 -3
  100. data/test/unit/tc_document.rb +81 -0
  101. data/test/unit/ts_analysis.rb +1 -1
  102. data/test/unit/ts_utils.rb +1 -1
  103. data/test/unit/utils/tc_bit_vector.rb +288 -0
  104. data/test/unit/utils/tc_number_tools.rb +117 -0
  105. data/test/unit/utils/tc_priority_queue.rb +106 -0
  106. metadata +140 -301
  107. data/CHANGELOG +0 -9
  108. data/ext/dummy.exe +0 -0
  109. data/ext/field.c +0 -408
  110. data/ext/frtio.h +0 -13
  111. data/ext/inc/except.h +0 -90
  112. data/ext/index_io.c +0 -382
  113. data/ext/index_rw.c +0 -2658
  114. data/ext/lang.c +0 -41
  115. data/ext/nix_io.c +0 -134
  116. data/ext/q_multi_phrase.c +0 -380
  117. data/ext/r_doc.c +0 -582
  118. data/ext/r_index_io.c +0 -1021
  119. data/ext/r_term.c +0 -219
  120. data/ext/term.c +0 -820
  121. data/ext/termdocs.c +0 -611
  122. data/ext/vector.c +0 -637
  123. data/ext/w32_io.c +0 -150
  124. data/lib/ferret/analysis.rb +0 -11
  125. data/lib/ferret/analysis/analyzers.rb +0 -112
  126. data/lib/ferret/analysis/standard_tokenizer.rb +0 -71
  127. data/lib/ferret/analysis/token.rb +0 -100
  128. data/lib/ferret/analysis/token_filters.rb +0 -86
  129. data/lib/ferret/analysis/token_stream.rb +0 -26
  130. data/lib/ferret/analysis/tokenizers.rb +0 -112
  131. data/lib/ferret/analysis/word_list_loader.rb +0 -27
  132. data/lib/ferret/document/document.rb +0 -152
  133. data/lib/ferret/document/field.rb +0 -312
  134. data/lib/ferret/index/compound_file_io.rb +0 -338
  135. data/lib/ferret/index/document_writer.rb +0 -289
  136. data/lib/ferret/index/field_infos.rb +0 -279
  137. data/lib/ferret/index/fields_io.rb +0 -181
  138. data/lib/ferret/index/index.rb +0 -675
  139. data/lib/ferret/index/index_file_names.rb +0 -33
  140. data/lib/ferret/index/index_reader.rb +0 -503
  141. data/lib/ferret/index/index_writer.rb +0 -534
  142. data/lib/ferret/index/multi_reader.rb +0 -377
  143. data/lib/ferret/index/multiple_term_doc_pos_enum.rb +0 -98
  144. data/lib/ferret/index/segment_infos.rb +0 -130
  145. data/lib/ferret/index/segment_merge_info.rb +0 -49
  146. data/lib/ferret/index/segment_merge_queue.rb +0 -16
  147. data/lib/ferret/index/segment_merger.rb +0 -358
  148. data/lib/ferret/index/segment_reader.rb +0 -412
  149. data/lib/ferret/index/segment_term_enum.rb +0 -169
  150. data/lib/ferret/index/segment_term_vector.rb +0 -58
  151. data/lib/ferret/index/term.rb +0 -53
  152. data/lib/ferret/index/term_buffer.rb +0 -83
  153. data/lib/ferret/index/term_doc_enum.rb +0 -291
  154. data/lib/ferret/index/term_enum.rb +0 -52
  155. data/lib/ferret/index/term_info.rb +0 -37
  156. data/lib/ferret/index/term_infos_io.rb +0 -321
  157. data/lib/ferret/index/term_vector_offset_info.rb +0 -20
  158. data/lib/ferret/index/term_vectors_io.rb +0 -553
  159. data/lib/ferret/query_parser.rb +0 -312
  160. data/lib/ferret/query_parser/query_parser.tab.rb +0 -928
  161. data/lib/ferret/search.rb +0 -50
  162. data/lib/ferret/search/boolean_clause.rb +0 -100
  163. data/lib/ferret/search/boolean_query.rb +0 -299
  164. data/lib/ferret/search/boolean_scorer.rb +0 -294
  165. data/lib/ferret/search/caching_wrapper_filter.rb +0 -40
  166. data/lib/ferret/search/conjunction_scorer.rb +0 -99
  167. data/lib/ferret/search/disjunction_sum_scorer.rb +0 -205
  168. data/lib/ferret/search/exact_phrase_scorer.rb +0 -32
  169. data/lib/ferret/search/explanation.rb +0 -41
  170. data/lib/ferret/search/field_cache.rb +0 -215
  171. data/lib/ferret/search/field_doc.rb +0 -31
  172. data/lib/ferret/search/field_sorted_hit_queue.rb +0 -184
  173. data/lib/ferret/search/filter.rb +0 -11
  174. data/lib/ferret/search/filtered_query.rb +0 -130
  175. data/lib/ferret/search/filtered_term_enum.rb +0 -79
  176. data/lib/ferret/search/fuzzy_query.rb +0 -154
  177. data/lib/ferret/search/fuzzy_term_enum.rb +0 -247
  178. data/lib/ferret/search/hit_collector.rb +0 -34
  179. data/lib/ferret/search/hit_queue.rb +0 -11
  180. data/lib/ferret/search/index_searcher.rb +0 -200
  181. data/lib/ferret/search/match_all_query.rb +0 -104
  182. data/lib/ferret/search/multi_phrase_query.rb +0 -216
  183. data/lib/ferret/search/multi_searcher.rb +0 -261
  184. data/lib/ferret/search/multi_term_query.rb +0 -65
  185. data/lib/ferret/search/non_matching_scorer.rb +0 -22
  186. data/lib/ferret/search/phrase_positions.rb +0 -55
  187. data/lib/ferret/search/phrase_query.rb +0 -214
  188. data/lib/ferret/search/phrase_scorer.rb +0 -152
  189. data/lib/ferret/search/prefix_query.rb +0 -54
  190. data/lib/ferret/search/query.rb +0 -140
  191. data/lib/ferret/search/query_filter.rb +0 -51
  192. data/lib/ferret/search/range_filter.rb +0 -103
  193. data/lib/ferret/search/range_query.rb +0 -139
  194. data/lib/ferret/search/req_excl_scorer.rb +0 -125
  195. data/lib/ferret/search/req_opt_sum_scorer.rb +0 -70
  196. data/lib/ferret/search/score_doc.rb +0 -38
  197. data/lib/ferret/search/score_doc_comparator.rb +0 -114
  198. data/lib/ferret/search/scorer.rb +0 -91
  199. data/lib/ferret/search/similarity.rb +0 -278
  200. data/lib/ferret/search/sloppy_phrase_scorer.rb +0 -47
  201. data/lib/ferret/search/sort.rb +0 -112
  202. data/lib/ferret/search/sort_comparator.rb +0 -60
  203. data/lib/ferret/search/sort_field.rb +0 -91
  204. data/lib/ferret/search/spans.rb +0 -12
  205. data/lib/ferret/search/spans/near_spans_enum.rb +0 -304
  206. data/lib/ferret/search/spans/span_first_query.rb +0 -79
  207. data/lib/ferret/search/spans/span_near_query.rb +0 -108
  208. data/lib/ferret/search/spans/span_not_query.rb +0 -130
  209. data/lib/ferret/search/spans/span_or_query.rb +0 -176
  210. data/lib/ferret/search/spans/span_query.rb +0 -25
  211. data/lib/ferret/search/spans/span_scorer.rb +0 -74
  212. data/lib/ferret/search/spans/span_term_query.rb +0 -105
  213. data/lib/ferret/search/spans/span_weight.rb +0 -84
  214. data/lib/ferret/search/spans/spans_enum.rb +0 -44
  215. data/lib/ferret/search/term_query.rb +0 -128
  216. data/lib/ferret/search/term_scorer.rb +0 -183
  217. data/lib/ferret/search/top_docs.rb +0 -36
  218. data/lib/ferret/search/top_field_docs.rb +0 -17
  219. data/lib/ferret/search/weight.rb +0 -54
  220. data/lib/ferret/search/wildcard_query.rb +0 -26
  221. data/lib/ferret/search/wildcard_term_enum.rb +0 -61
  222. data/lib/ferret/stemmers.rb +0 -1
  223. data/lib/ferret/stemmers/porter_stemmer.rb +0 -218
  224. data/lib/ferret/store.rb +0 -5
  225. data/lib/ferret/store/buffered_index_io.rb +0 -190
  226. data/lib/ferret/store/directory.rb +0 -141
  227. data/lib/ferret/store/fs_store.rb +0 -381
  228. data/lib/ferret/store/index_io.rb +0 -245
  229. data/lib/ferret/store/ram_store.rb +0 -286
  230. data/lib/ferret/utils.rb +0 -8
  231. data/lib/ferret/utils/bit_vector.rb +0 -123
  232. data/lib/ferret/utils/date_tools.rb +0 -138
  233. data/lib/ferret/utils/number_tools.rb +0 -91
  234. data/lib/ferret/utils/parameter.rb +0 -41
  235. data/lib/ferret/utils/priority_queue.rb +0 -120
  236. data/lib/ferret/utils/string_helper.rb +0 -47
  237. data/lib/ferret/utils/thread_local.rb +0 -28
  238. data/lib/ferret/utils/weak_key_hash.rb +0 -60
  239. data/lib/rferret.rb +0 -37
  240. data/rake_utils/code_statistics.rb +0 -106
  241. data/test/benchmark/tb_ram_store.rb +0 -76
  242. data/test/benchmark/tb_rw_vint.rb +0 -26
  243. data/test/functional/thread_safety_index_test.rb +0 -81
  244. data/test/functional/thread_safety_test.rb +0 -137
  245. data/test/longrunning/tc_numbertools.rb +0 -60
  246. data/test/longrunning/tm_store.rb +0 -19
  247. data/test/unit/analysis/ctc_analyzer.rb +0 -532
  248. data/test/unit/analysis/data/wordfile +0 -6
  249. data/test/unit/analysis/rtc_letter_tokenizer.rb +0 -20
  250. data/test/unit/analysis/rtc_lower_case_filter.rb +0 -20
  251. data/test/unit/analysis/rtc_lower_case_tokenizer.rb +0 -27
  252. data/test/unit/analysis/rtc_per_field_analyzer_wrapper.rb +0 -39
  253. data/test/unit/analysis/rtc_porter_stem_filter.rb +0 -16
  254. data/test/unit/analysis/rtc_standard_analyzer.rb +0 -20
  255. data/test/unit/analysis/rtc_standard_tokenizer.rb +0 -20
  256. data/test/unit/analysis/rtc_stop_analyzer.rb +0 -20
  257. data/test/unit/analysis/rtc_stop_filter.rb +0 -14
  258. data/test/unit/analysis/rtc_white_space_analyzer.rb +0 -21
  259. data/test/unit/analysis/rtc_white_space_tokenizer.rb +0 -20
  260. data/test/unit/analysis/rtc_word_list_loader.rb +0 -32
  261. data/test/unit/analysis/tc_token.rb +0 -25
  262. data/test/unit/document/rtc_field.rb +0 -28
  263. data/test/unit/document/tc_document.rb +0 -47
  264. data/test/unit/document/tc_field.rb +0 -98
  265. data/test/unit/index/rtc_compound_file_io.rb +0 -107
  266. data/test/unit/index/rtc_field_infos.rb +0 -127
  267. data/test/unit/index/rtc_fields_io.rb +0 -167
  268. data/test/unit/index/rtc_multiple_term_doc_pos_enum.rb +0 -83
  269. data/test/unit/index/rtc_segment_infos.rb +0 -74
  270. data/test/unit/index/rtc_segment_term_docs.rb +0 -17
  271. data/test/unit/index/rtc_segment_term_enum.rb +0 -60
  272. data/test/unit/index/rtc_segment_term_vector.rb +0 -71
  273. data/test/unit/index/rtc_term_buffer.rb +0 -57
  274. data/test/unit/index/rtc_term_info.rb +0 -19
  275. data/test/unit/index/rtc_term_infos_io.rb +0 -192
  276. data/test/unit/index/rtc_term_vectors_io.rb +0 -108
  277. data/test/unit/index/tc_term.rb +0 -27
  278. data/test/unit/index/tc_term_voi.rb +0 -18
  279. data/test/unit/search/rtc_similarity.rb +0 -37
  280. data/test/unit/search/rtc_sort_field.rb +0 -14
  281. data/test/unit/search/tc_multi_searcher2.rb +0 -126
  282. data/test/unit/store/rtc_fs_store.rb +0 -62
  283. data/test/unit/store/rtc_ram_store.rb +0 -15
  284. data/test/unit/store/rtm_store.rb +0 -150
  285. data/test/unit/store/rtm_store_lock.rb +0 -2
  286. data/test/unit/ts_document.rb +0 -2
  287. data/test/unit/utils/rtc_bit_vector.rb +0 -73
  288. data/test/unit/utils/rtc_date_tools.rb +0 -50
  289. data/test/unit/utils/rtc_number_tools.rb +0 -59
  290. data/test/unit/utils/rtc_parameter.rb +0 -40
  291. data/test/unit/utils/rtc_priority_queue.rb +0 -62
  292. data/test/unit/utils/rtc_string_helper.rb +0 -21
  293. data/test/unit/utils/rtc_thread.rb +0 -61
  294. data/test/unit/utils/rtc_weak_key_hash.rb +0 -25
  295. data/test/utils/number_to_spoken.rb +0 -132
data/ext/lang.c DELETED
@@ -1,41 +0,0 @@
1
- #include <stdarg.h>
2
- #include <string.h>
3
- #include <stdio.h>
4
- #include <errno.h>
5
- #include "global.h"
6
-
7
- void ft_raise(char *file, int line_num, VALUE etype, const char *fmt, ...)
8
- {
9
- va_list args;
10
- char buf[MAX_ERROR_LEN];
11
- char *buf_ptr = buf;
12
-
13
- if (progname() != NULL) {
14
- sprintf(buf_ptr, "%s: ", progname());
15
- buf_ptr += strlen(buf_ptr);
16
- }
17
-
18
- sprintf(buf_ptr, "Error occured at <%s>:%d\n", file, line_num);
19
- buf_ptr += strlen(buf_ptr);
20
- va_start(args, fmt);
21
- vsprintf(buf_ptr, fmt, args);
22
- buf_ptr += strlen(buf_ptr);
23
- va_end(args);
24
-
25
- if (fmt[0] != '\0' && fmt[strlen(fmt)-1] == ':') {
26
- sprintf(buf_ptr, " %s", strerror(errno));
27
- buf_ptr += strlen(buf_ptr);
28
- }
29
- sprintf(buf_ptr, "\n");
30
- rb_raise(etype, buf); /* conventional value for failed execution */
31
- }
32
-
33
- #ifdef WIN32
34
- void eprintf(VALUE etype, const char *fmt, ...)
35
- {
36
- va_list args;
37
- va_start(args, fmt);
38
- ft_raise("Windows", -1, etype, fmt, args);
39
- va_end(args);
40
- }
41
- #endif
data/ext/nix_io.c DELETED
@@ -1,134 +0,0 @@
1
- #ifndef WIN32
2
-
3
- #include "global.h"
4
- #include "store.h"
5
- #include <dirent.h>
6
- #include <unistd.h>
7
- #include <string.h>
8
- #include <errno.h>
9
- #include <fcntl.h>
10
-
11
- /**
12
- * Create a filepath for a file in the store using the operating systems
13
- * default file seperator.
14
- */
15
- char *join_path(char *buf, const char *base, const char *filename)
16
- {
17
- sprintf(buf, "%s/%s", base, filename);
18
- return buf;
19
- }
20
-
21
- bool exists(char *path)
22
- {
23
- int fd = open(path, 0);
24
- if (fd < 0) {
25
- if (errno != ENOENT) {
26
- RAISE(IO_ERROR, strerror(errno));
27
- }
28
- return false;
29
- }
30
- close(fd);
31
- return true;
32
- }
33
-
34
- int fcount(char *path)
35
- {
36
- int cnt = 0;
37
- struct dirent *de;
38
- DIR *d = opendir(path);
39
-
40
- if (!d) RAISE(IO_ERROR, strerror(errno));
41
-
42
- while ((de = readdir(d)) != NULL) {
43
- if (de->d_name[0] != '.') {
44
- cnt++;
45
- }
46
- }
47
- closedir(d);
48
-
49
- return cnt;
50
- }
51
-
52
- void dir_each(char *path, void (*func)(char *fname, void *arg), void *arg)
53
- {
54
- struct dirent *de;
55
- DIR *d = opendir(path);
56
-
57
- if (!d) RAISE(IO_ERROR, strerror(errno));
58
-
59
- while ((de = readdir(d)) != NULL) {
60
- if (de->d_name[0] != '.' && !file_is_lock(de->d_name)) {
61
- func(de->d_name, arg);
62
- }
63
- }
64
- closedir(d);
65
- }
66
-
67
- /**
68
- * Clear all the locks in the store.
69
- *
70
- * @param store the store to clear the locks from
71
- * @throws IO_ERROR if there is an error opening the directory
72
- */
73
- void fs_clear_locks(Store *store)
74
- {
75
- struct dirent *de;
76
- DIR *d = opendir(store->dir.path);
77
-
78
- if (!d) RAISE(IO_ERROR, strerror(errno));
79
-
80
- while ((de = readdir(d)) != NULL) {
81
- if (file_is_lock(de->d_name)) {
82
- char buf[MAX_FILE_PATH];
83
- remove(join_path(buf, store->dir.path, de->d_name));
84
- }
85
- }
86
- closedir(d);
87
- }
88
-
89
- /**
90
- * Clear all files from the store except the lock files.
91
- *
92
- * @param store the store to clear all the files from
93
- * @throws IO_ERROR if there is an error deleting the files
94
- */
95
- void fs_clear(Store *store)
96
- {
97
- struct dirent *de;
98
- DIR *d = opendir(store->dir.path);
99
-
100
- if (!d) RAISE(IO_ERROR, strerror(errno));
101
-
102
- while ((de = readdir(d)) != NULL) {
103
- if (de->d_name[0] != '.' && !file_is_lock(de->d_name)) {
104
- char buf[MAX_FILE_PATH];
105
- remove(join_path(buf, store->dir.path, de->d_name));
106
- }
107
- }
108
- closedir(d);
109
- }
110
-
111
- /**
112
- * Clear all files from the store including the lock files.
113
- *
114
- * @param store the store to clear all the files from
115
- * @throws IO_ERROR if there is an error deleting the files
116
- */
117
- void fs_clear_all(Store *store)
118
- {
119
- struct dirent *de;
120
- DIR *d = opendir(store->dir.path);
121
-
122
- if (!d) RAISE(IO_ERROR, strerror(errno));
123
-
124
- while ((de = readdir(d)) != NULL) {
125
- if (de->d_name[0] != '.') {
126
- char buf[MAX_FILE_PATH];
127
- remove(join_path(buf, store->dir.path, de->d_name));
128
- }
129
- }
130
- closedir(d);
131
- }
132
-
133
- #endif
134
-
data/ext/q_multi_phrase.c DELETED
@@ -1,380 +0,0 @@
1
- #include "search.h"
2
- #include <string.h>
3
-
4
- static char * const FIELD_CHANGE_ERROR_MSG = "All phrase terms must be in the same field.";
5
-
6
- /***************************************************************************
7
- *
8
- * MultiPhraseWeight
9
- *
10
- ***************************************************************************/
11
-
12
- char *mphw_to_s(Weight *self)
13
- {
14
- return strfmt("MultiPhraseWeight(%f)", self->value);
15
- }
16
-
17
- Scorer *mphw_scorer(Weight *self, IndexReader *ir)
18
- {
19
- Scorer *phsc;
20
- MultiPhraseQuery *mphq = (MultiPhraseQuery *)self->query->data;
21
- int i;
22
- TermDocEnum **tps;
23
-
24
- if (mphq->t_cnt == 0) {
25
- return NULL; /* optimize zero-term case */
26
- }
27
-
28
- tps = ALLOC_N(TermDocEnum *, mphq->t_cnt);
29
-
30
- for (i = 0; i < mphq->t_cnt; i++) {
31
- if (mphq->pt_cnt[i] == 1) {
32
- tps[i] = ir_term_positions_for(ir, mphq->terms[i][0]);
33
- } else {
34
- tps[i] = mtdpe_create(ir, mphq->terms[i], mphq->pt_cnt[i]);
35
- }
36
- if (tps[i] == NULL) {
37
- /* free everything we just created and return NULL */
38
- int j;
39
- for (j = 0; j < i; j++) {
40
- tps[i]->close(tps[i]);
41
- }
42
- free(tps);
43
- return NULL;
44
- }
45
- }
46
-
47
- if (mphq->slop == 0) { /* optimize exact case */
48
- phsc = exact_phrase_scorer_create(self, tps, mphq->positions, mphq->t_cnt,
49
- self->similarity,
50
- ir->get_norms(ir, mphq->field));
51
- } else {
52
- phsc = sloppy_phrase_scorer_create(self, tps, mphq->positions, mphq->t_cnt,
53
- self->similarity,
54
- mphq->slop,
55
- ir->get_norms(ir, mphq->field));
56
- }
57
- free(tps);
58
- return phsc;
59
- }
60
-
61
- Explanation *mphw_explain(Weight *self, IndexReader *ir, int doc_num)
62
- {
63
- Explanation *idf_expl1;
64
- Explanation *idf_expl2;
65
- Explanation *query_expl;
66
- Explanation *qnorm_expl;
67
- Explanation *field_expl;
68
- Explanation *tf_expl;
69
- Scorer *scorer;
70
- uchar *field_norms;
71
- float field_norm;
72
- Explanation *field_norm_expl;
73
-
74
- char *query_str = self->query->to_s(self->query, "");
75
- MultiPhraseQuery *mphq = (MultiPhraseQuery *)self->query->data;
76
- int i, j;
77
- char *doc_freqs = NULL;
78
- int len = 0, pos = 0;
79
-
80
- Explanation *expl = expl_create(0.0,
81
- strfmt("weight(%s in %d), product of:", query_str, doc_num));
82
-
83
- for (i = 0; i < mphq->t_cnt; i++) {
84
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
85
- len += (int)strlen(mphq->terms[i][j]->text) + 30;
86
- }
87
- }
88
- doc_freqs = ALLOC_N(char, len);
89
- for (i = 0; i < mphq->t_cnt; i++) {
90
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
91
- Term *term = mphq->terms[i][j];
92
- sprintf(doc_freqs + pos, "%s=%d, ", term->text, ir->doc_freq(ir, term));
93
- pos += (int)strlen(doc_freqs + pos);
94
- }
95
- }
96
- pos -= 2; // remove ", " from the end
97
- doc_freqs[pos] = 0;
98
-
99
- idf_expl1 = expl_create(self->idf,
100
- strfmt("idf(%s:<%s>)", mphq->field, doc_freqs));
101
- idf_expl2 = expl_create(self->idf,
102
- strfmt("idf(%s:<%s>)", mphq->field, doc_freqs));
103
- free(doc_freqs);
104
-
105
- /* explain query weight */
106
- query_expl = expl_create(0.0,
107
- strfmt("query_weight(%s), product of:", query_str));
108
-
109
- if (self->query->boost != 1.0) {
110
- expl_add_detail(query_expl, expl_create(self->query->boost, estrdup("boost")));
111
- }
112
- expl_add_detail(query_expl, idf_expl1);
113
-
114
- qnorm_expl = expl_create(self->qnorm, estrdup("query_norm"));
115
- expl_add_detail(query_expl, qnorm_expl);
116
-
117
- query_expl->value = self->query->boost * self->idf * self->qnorm;
118
-
119
- expl_add_detail(expl, query_expl);
120
-
121
- /* explain field weight */
122
- field_expl = expl_create(0.0,
123
- strfmt("field_weight(%s in %d), product of:", query_str, doc_num));
124
- free(query_str);
125
-
126
- scorer = self->scorer(self, ir);
127
- tf_expl = scorer->explain(scorer, doc_num);
128
- scorer->destroy(scorer);
129
- expl_add_detail(field_expl, tf_expl);
130
- expl_add_detail(field_expl, idf_expl2);
131
-
132
- field_norms = ir->get_norms(ir, mphq->field);
133
- field_norm = (field_norms != NULL)
134
- ? sim_decode_norm(self->similarity, field_norms[doc_num])
135
- : (float)0.0;
136
- field_norm_expl = expl_create(field_norm,
137
- strfmt("field_norm(field=%s, doc=%d)", mphq->field, doc_num));
138
-
139
- expl_add_detail(field_expl, field_norm_expl);
140
-
141
- field_expl->value = tf_expl->value * self->idf * field_norm;
142
-
143
- /* combine them */
144
- if (query_expl->value == 1.0) {
145
- expl_destoy(expl);
146
- return field_expl;
147
- } else {
148
- expl->value = (query_expl->value * field_expl->value);
149
- expl_add_detail(expl, field_expl);
150
- return expl;
151
- }
152
- }
153
-
154
- Weight *mphw_create(Query *query, Searcher *searcher)
155
- {
156
- Weight *self = w_create(query);
157
-
158
- MultiPhraseQuery *mphq = (MultiPhraseQuery *)query->data;
159
- int i, j;
160
-
161
- self->scorer = &mphw_scorer;
162
- self->explain = &mphw_explain;
163
- self->to_s = &mphw_to_s;
164
- self->sum_of_squared_weights = &w_sum_of_squared_weights;
165
-
166
- self->similarity = query->get_similarity(query, searcher);
167
- self->query = query;
168
- self->value = query->boost;
169
- self->idf = 0.0;
170
-
171
- for (i = 0; i < mphq->t_cnt; i++) {
172
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
173
- self->idf += sim_idf_term(self->similarity, mphq->terms[i][j], searcher);
174
- }
175
- }
176
-
177
- return self;
178
- }
179
-
180
- /***************************************************************************
181
- *
182
- * MultiPhraseQuery
183
- *
184
- ***************************************************************************/
185
-
186
- #define GET_MPHQ MultiPhraseQuery *mphq = (MultiPhraseQuery *)self->data
187
-
188
- /**
189
- * NOTE: terms must be allocated and it will be freed when the query is
190
- * destroyed.
191
- */
192
- void mphq_add_terms(Query *self, Term **terms, int t_cnt, int pos_inc)
193
- {
194
- GET_MPHQ;
195
- int position, index = mphq->t_cnt;
196
- if (index >= mphq->t_capa) {
197
- mphq->t_capa <<= 1;
198
- REALLOC_N(mphq->terms, Term **, mphq->t_capa);
199
- REALLOC_N(mphq->positions, int, mphq->t_capa);
200
- REALLOC_N(mphq->pt_cnt, int, mphq->t_capa);
201
- }
202
- if (index == 0) {
203
- position = 0;
204
- mphq->field = terms[0]->field;
205
- } else {
206
- int i;
207
- position = mphq->positions[index - 1] + pos_inc;
208
- for (i = 0; i < t_cnt; i++) {
209
- if (strcmp(terms[i]->field, mphq->field) != 0) {
210
- RAISE(ARG_ERROR, FIELD_CHANGE_ERROR_MSG);
211
- }
212
- }
213
- }
214
-
215
- mphq->terms[index] = terms;
216
- mphq->pt_cnt[index] = t_cnt;
217
- mphq->positions[index] = position;
218
- mphq->t_cnt++;
219
- }
220
-
221
- void mphq_destroy(Query *self)
222
- {
223
- GET_MPHQ;
224
- int i, j;
225
- if (self->destroy_all) {
226
- for (i = 0; i < mphq->t_cnt; i++) {
227
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
228
- term_destroy(mphq->terms[i][j]);
229
- }
230
- free(mphq->terms[i]);
231
- }
232
- }
233
- free(mphq->terms);
234
- free(mphq->positions);
235
- free(mphq->pt_cnt);
236
- free(mphq);
237
-
238
- q_destroy_i(self);
239
- }
240
-
241
- void mphq_extract_terms(Query *self, HashSet *terms)
242
- {
243
- GET_MPHQ;
244
- int i, j;
245
- for (i = 0; i < mphq->t_cnt; i++) {
246
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
247
- hs_add(terms, term_clone(mphq->terms[i][j]));
248
- }
249
- }
250
- }
251
-
252
- char *mphq_to_s(Query *self, char *field)
253
- {
254
- GET_MPHQ;
255
- int i, j, buf_index = 0, len = 0, pos, last_pos = -1;
256
- char *buffer;
257
- if (!mphq->t_cnt) return NULL;
258
- len = (int)strlen(mphq->field) + 1;
259
- for (i = 0; i < mphq->t_cnt; i++) {
260
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
261
- len += (int)strlen(mphq->terms[i][j]->text) + 1;
262
- }
263
- }
264
-
265
- // add space for extra characters and boost and slop
266
- len += 100 + 3 * mphq->positions[mphq->t_cnt - 1];
267
-
268
- buffer = ALLOC_N(char, len);
269
-
270
- if (strcmp(field, mphq->field) != 0) {
271
- len = (int)strlen(mphq->field);
272
- memcpy(buffer, mphq->field, len);
273
- buffer[len] = ':';
274
- buf_index += len + 1;
275
- }
276
- buffer[buf_index++] = '"';
277
-
278
- for (i = 0; i < mphq->t_cnt; i++) {
279
- pos = mphq->positions[i];
280
- for (j = last_pos; j < pos - 1; j++) {
281
- memcpy(buffer + buf_index, "<> ", 3);
282
- buf_index += 3;
283
- }
284
- last_pos = pos;
285
-
286
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
287
- Term *term = mphq->terms[i][j];
288
- len = (int)strlen(term->text);
289
- memcpy(buffer + buf_index, term->text, len);
290
- buf_index += len;
291
- buffer[buf_index++] = '|';
292
- }
293
- buffer[buf_index-1] = ' '; /* change last '|' to ' ' */
294
- }
295
-
296
- if (buffer[buf_index-1] == ' ') buf_index--;
297
- buffer[buf_index++] = '"';
298
- buffer[buf_index] = 0;
299
- if (mphq->slop != 0) {
300
- sprintf(buffer + buf_index, "~%d", mphq->slop);
301
- buf_index += (int)strlen(buffer + buf_index);
302
- }
303
- if (self->boost != 1.0) {
304
- buffer[buf_index] = '^';
305
- dbl_to_s(buffer + buf_index + 1, self->boost);
306
- }
307
- return buffer;
308
- }
309
-
310
- Query *mphq_rewrite(Query *self, IndexReader *ir)
311
- {
312
- GET_MPHQ;
313
- if (mphq->t_cnt == 1) { // optimize one-term case
314
- Term **terms = mphq->terms[0];
315
- Query *bq = bq_create(true);
316
- int i;
317
- for (i = 0; i < mphq->pt_cnt[0]; i++) {
318
- bq_add_query(bq, tq_create(term_clone(terms[i])), BC_SHOULD);
319
- }
320
- bq->boost = self->boost;
321
- return bq;
322
- } else {
323
- self->ref_cnt++;
324
- return self;
325
- }
326
- }
327
-
328
- static uint mphq_hash(Query *self)
329
- {
330
- int i, j;
331
- uint hash = 0;
332
- MultiPhraseQuery *mphq = (MultiPhraseQuery *)self->data;
333
- for (i = 0; i < mphq->t_cnt; i++) {
334
- for (j = 0; j < mphq->pt_cnt[i]; j++) {
335
- hash ^= (term_hash(mphq->terms[i][j]) ^ mphq->positions[i]);
336
- }
337
- hash <<= 1;
338
- }
339
- return (hash ^ mphq->slop);
340
- }
341
-
342
- static int mphq_eq(Query *self, Query *o)
343
- {
344
- int i, j;
345
- MultiPhraseQuery *mphq1 = (MultiPhraseQuery *)self->data;
346
- MultiPhraseQuery *mphq2 = (MultiPhraseQuery *)o->data;
347
- if (mphq1->slop != mphq2->slop) return false;
348
- for (i = 0; i < mphq1->t_cnt; i++) {
349
- if ((mphq1->pt_cnt[i] != mphq2->pt_cnt[i]) ||
350
- (mphq1->positions[i] != mphq2->positions[i])) return false;
351
-
352
- for (j = 0; j < mphq1->pt_cnt[i]; j++) {
353
- if (!term_eq(mphq1->terms[i][j], mphq2->terms[i][j])) return false;
354
- }
355
- }
356
- return true;
357
- }
358
-
359
- Query *mphq_create()
360
- {
361
- Query *self = q_create();
362
-
363
- MultiPhraseQuery *mphq = ALLOC(MultiPhraseQuery);
364
- ZEROSET(mphq, MultiPhraseQuery, 1);
365
- mphq->t_capa = PHQ_INIT_CAPA;
366
- mphq->terms = ALLOC_N(Term **, PHQ_INIT_CAPA);
367
- mphq->positions = ALLOC_N(int, PHQ_INIT_CAPA);
368
- mphq->pt_cnt = ALLOC_N(int, PHQ_INIT_CAPA);
369
- self->data = mphq;
370
-
371
- self->type = MULTI_PHRASE_QUERY;
372
- self->rewrite = &mphq_rewrite;
373
- self->extract_terms = &mphq_extract_terms;
374
- self->to_s = &mphq_to_s;
375
- self->hash = &mphq_hash;
376
- self->eq = &mphq_eq;
377
- self->destroy_i = &mphq_destroy;
378
- self->create_weight_i = &mphw_create;
379
- return self;
380
- }