paul@0 | 1 | #!/usr/bin/env python |
paul@0 | 2 | |
paul@44 | 3 | from iixr.files import * |
paul@44 | 4 | from iixr.fields import * |
paul@44 | 5 | from iixr.terms import * |
paul@44 | 6 | from iixr.positions import * |
paul@44 | 7 | from iixr.index import * |
paul@59 | 8 | import os, sys |
paul@18 | 9 | |
paul@18 | 10 | # Remove old test files. |
paul@18 | 11 | |
paul@74 | 12 | for filename in ("test", "testMS", "testNMS", "testF", "testFI", "testI", "testP", "testP2", "testPI"): |
paul@18 | 13 | try: |
paul@18 | 14 | os.remove(filename) |
paul@18 | 15 | except OSError: |
paul@18 | 16 | pass |
paul@18 | 17 | |
paul@18 | 18 | try: |
paul@77 | 19 | for dirname in ("test_index", "test_index2", "test_index3", "test_indexT"): |
paul@59 | 20 | for filename in os.listdir(dirname): |
paul@59 | 21 | os.remove(os.path.join(dirname, filename)) |
paul@59 | 22 | os.rmdir(dirname) |
paul@18 | 23 | except OSError: |
paul@18 | 24 | pass |
paul@0 | 25 | |
paul@59 | 26 | if "clean" in sys.argv: |
paul@59 | 27 | sys.exit(0) |
paul@59 | 28 | |
paul@69 | 29 | print "- Test basic data types." |
paul@9 | 30 | |
paul@5 | 31 | numbers = [12345678, 0, 1, 127, 128, 255, 256] |
paul@0 | 32 | |
paul@0 | 33 | f = open("test", "wb") |
paul@44 | 34 | w = FileWriter(f) |
paul@89 | 35 | w.begin_record() |
paul@0 | 36 | for number in numbers: |
paul@0 | 37 | w.write_number(number) |
paul@89 | 38 | w.end_record() |
paul@0 | 39 | w.close() |
paul@0 | 40 | |
paul@3 | 41 | f = open("test", "rb") |
paul@44 | 42 | r = FileReader(f) |
paul@89 | 43 | r.begin_record() |
paul@0 | 44 | for number in numbers: |
paul@0 | 45 | n = r.read_number() |
paul@0 | 46 | print number == n, number, n |
paul@89 | 47 | r.end_record() |
paul@0 | 48 | r.close() |
paul@0 | 49 | |
paul@74 | 50 | tuples = [(0, 0), (1, 3), (2, 5), (3, 9)] |
paul@74 | 51 | |
paul@74 | 52 | f = open("testMS", "wb") |
paul@74 | 53 | w = FileWriter(f) |
paul@89 | 54 | w.begin_record() |
paul@91 | 55 | w.write_monotonic_sequence(tuples, 2) |
paul@89 | 56 | w.end_record() |
paul@74 | 57 | w.close() |
paul@74 | 58 | |
paul@74 | 59 | f = open("testMS", "rb") |
paul@74 | 60 | r = FileReader(f) |
paul@89 | 61 | r.begin_record() |
paul@91 | 62 | for t, t2 in zip(r.read_monotonic_sequence(2), tuples): |
paul@74 | 63 | print t == t2, t, t2 |
paul@89 | 64 | r.end_record() |
paul@74 | 65 | r.close() |
paul@74 | 66 | |
paul@74 | 67 | tuples2 = [(0, 0), (1, 3), (2, 1), (3, 2), (4, 0)] |
paul@74 | 68 | |
paul@74 | 69 | f = open("testNMS", "wb") |
paul@74 | 70 | w = FileWriter(f) |
paul@89 | 71 | w.begin_record() |
paul@91 | 72 | w.write_delta_sequence(tuples2, 2) |
paul@89 | 73 | w.end_record() |
paul@74 | 74 | w.close() |
paul@74 | 75 | |
paul@74 | 76 | f = open("testNMS", "rb") |
paul@74 | 77 | r = FileReader(f) |
paul@89 | 78 | r.begin_record() |
paul@91 | 79 | for t, t2 in zip(r.read_delta_sequence(2), tuples2): |
paul@74 | 80 | print t == t2, t, t2 |
paul@89 | 81 | r.end_record() |
paul@74 | 82 | r.close() |
paul@74 | 83 | |
paul@69 | 84 | print "- Test positions." |
paul@9 | 85 | |
paul@0 | 86 | all_doc_positions = [ |
paul@0 | 87 | [ |
paul@0 | 88 | (123, [1, 3, 5, 15, 25]), |
paul@19 | 89 | (124, [0, 100]), |
paul@19 | 90 | (125, [11, 99, 199]), |
paul@19 | 91 | (130, [77, 78, 80, 82, 89]) |
paul@0 | 92 | ], |
paul@0 | 93 | [ |
paul@0 | 94 | (78, [9]), |
paul@19 | 95 | (196, [10, 11]), |
paul@19 | 96 | (197, [17, 21, 30]) |
paul@0 | 97 | ] |
paul@0 | 98 | ] |
paul@0 | 99 | |
paul@19 | 100 | f = open("testP", "wb") |
paul@44 | 101 | w = PositionWriter(f) |
paul@91 | 102 | w.begin(0, 0) |
paul@0 | 103 | for doc_positions in all_doc_positions: |
paul@91 | 104 | w.reset() |
paul@0 | 105 | for docnum, positions in doc_positions: |
paul@0 | 106 | w.write_positions(docnum, positions) |
paul@0 | 107 | w.close() |
paul@0 | 108 | |
paul@19 | 109 | f = open("testP", "rb") |
paul@68 | 110 | r = PositionReader(f) |
paul@0 | 111 | for doc_positions in all_doc_positions: |
paul@91 | 112 | r.reset() |
paul@0 | 113 | for docnum, positions in doc_positions: |
paul@0 | 114 | d, p = r.read_positions() |
paul@0 | 115 | print docnum == d, docnum, d |
paul@0 | 116 | print positions == p, positions, p |
paul@0 | 117 | r.close() |
paul@0 | 118 | |
paul@74 | 119 | all_doc_positions_seq = [ |
paul@74 | 120 | [ |
paul@74 | 121 | ((123, 0), [(1, 5), (3, 9), (5, 15), (15, 45), (25, 70)]), |
paul@74 | 122 | ((124, 1), [(0, 0), (100, 350)]), |
paul@74 | 123 | ((124, 2), [(11, 38), (99, 379), (199, 720)]), |
paul@74 | 124 | ((130, 0), [(77, 286), (78, 290), (80, 300), (82, 304), (89, 316)]) |
paul@74 | 125 | ], |
paul@74 | 126 | [ |
paul@74 | 127 | ((78, 1), [(9, 19)]), |
paul@74 | 128 | ((196, 0), [(10, 27), (11, 29)]), |
paul@74 | 129 | ((196, 1), [(17, 46), (21, 52), (30, 60)]) |
paul@74 | 130 | ] |
paul@74 | 131 | ] |
paul@74 | 132 | |
paul@74 | 133 | f = open("testP2", "wb") |
paul@74 | 134 | w = PositionWriter(f) |
paul@91 | 135 | w.begin(2, 2) |
paul@74 | 136 | for doc_positions in all_doc_positions_seq: |
paul@91 | 137 | w.reset() |
paul@74 | 138 | for docnum, positions in doc_positions: |
paul@74 | 139 | w.write_positions(docnum, positions) |
paul@74 | 140 | w.close() |
paul@74 | 141 | |
paul@74 | 142 | f = open("testP2", "rb") |
paul@74 | 143 | r = PositionReader(f) |
paul@74 | 144 | for doc_positions in all_doc_positions_seq: |
paul@91 | 145 | r.reset() |
paul@74 | 146 | for docnum, positions in doc_positions: |
paul@74 | 147 | d, p = r.read_positions() |
paul@89 | 148 | print docnum == d, docnum, d |
paul@89 | 149 | print positions == p, positions, p |
paul@74 | 150 | r.close() |
paul@74 | 151 | |
paul@69 | 152 | print "- Test position index files." |
paul@19 | 153 | |
paul@19 | 154 | indexed_positions = [ |
paul@19 | 155 | [ |
paul@19 | 156 | (1234, 0, 100), |
paul@19 | 157 | (2345, 700, 100), |
paul@19 | 158 | (3456, 1900, 50) |
paul@19 | 159 | ], |
paul@19 | 160 | [ |
paul@19 | 161 | (4567, 2800, 20) |
paul@19 | 162 | ] |
paul@19 | 163 | ] |
paul@19 | 164 | |
paul@19 | 165 | offsets = [] |
paul@19 | 166 | f = open("testPI", "wb") |
paul@44 | 167 | w = PositionIndexWriter(f) |
paul@91 | 168 | w.begin(0) |
paul@19 | 169 | for term_positions in indexed_positions: |
paul@19 | 170 | offset = None |
paul@19 | 171 | doc_frequency = 0 |
paul@19 | 172 | w.reset() |
paul@19 | 173 | for docnum, pos_offset, count in term_positions: |
paul@19 | 174 | if offset is None: |
paul@88 | 175 | offset = w.tell() |
paul@55 | 176 | w.write_positions(docnum, pos_offset, count) |
paul@19 | 177 | doc_frequency += count |
paul@19 | 178 | offsets.append((offset, doc_frequency)) |
paul@19 | 179 | w.close() |
paul@19 | 180 | |
paul@69 | 181 | r = PositionIndexIterator(PositionIndexReader(open("testPI", "rb"))) |
paul@19 | 182 | offsets.reverse() |
paul@19 | 183 | indexed_positions.reverse() |
paul@19 | 184 | for (offset, doc_frequency), term_positions in zip(offsets, indexed_positions): |
paul@68 | 185 | r.seek(offset, doc_frequency) |
paul@68 | 186 | for (docnum, pos_offset, count), (dn, po, c) in zip(term_positions, r): |
paul@19 | 187 | print docnum == dn, docnum, dn |
paul@19 | 188 | print pos_offset == po, pos_offset, po |
paul@19 | 189 | print count == c, count, c |
paul@69 | 190 | r.reader.close() |
paul@19 | 191 | |
paul@69 | 192 | print "- Test position dictionaries." |
paul@19 | 193 | |
paul@19 | 194 | f = open("testP", "wb") |
paul@44 | 195 | w = PositionWriter(f) |
paul@19 | 196 | f2 = open("testPI", "wb") |
paul@44 | 197 | w2 = PositionIndexWriter(f2) |
paul@44 | 198 | wd = PositionDictionaryWriter(w, w2, 2) |
paul@0 | 199 | offsets = [] |
paul@0 | 200 | for doc_positions in all_doc_positions: |
paul@19 | 201 | offset, frequency, doc_frequency = wd.write_term_positions(doc_positions) |
paul@19 | 202 | offsets.append((offset, doc_frequency)) |
paul@20 | 203 | wd.close() |
paul@0 | 204 | |
paul@68 | 205 | r = PositionReader(open("testP", "rb")) |
paul@68 | 206 | r2 = PositionIndexReader(open("testPI", "rb")) |
paul@44 | 207 | rd = PositionDictionaryReader(r, r2) |
paul@0 | 208 | offsets.reverse() |
paul@0 | 209 | all_doc_positions.reverse() |
paul@19 | 210 | for (offset, doc_frequency), doc_positions in zip(offsets, all_doc_positions): |
paul@69 | 211 | it = rd.read_term_positions(offset, doc_frequency) |
paul@69 | 212 | dp = list(it) |
paul@0 | 213 | print doc_positions == dp, doc_positions, dp |
paul@20 | 214 | rd.close() |
paul@0 | 215 | |
paul@69 | 216 | print "- Test fields." |
paul@9 | 217 | |
paul@8 | 218 | doc_fields = [ |
paul@9 | 219 | (123, ["testing", "fields", "stored", "compressed"]), |
paul@9 | 220 | (456, ["fields", "for a second", "document"]), |
paul@9 | 221 | (789, ["field value"]), |
paul@9 | 222 | (1234, []), |
paul@9 | 223 | (2345, ["abc", "def"]), |
paul@9 | 224 | (3456, ["apple", "banana", "cherry"]), |
paul@9 | 225 | (4567, ["drue", "eple"]) |
paul@8 | 226 | ] |
paul@8 | 227 | |
paul@8 | 228 | f = open("testF", "wb") |
paul@44 | 229 | w = FieldWriter(f) |
paul@91 | 230 | w.begin(0) |
paul@91 | 231 | w.reset() |
paul@9 | 232 | for docnum, fields in doc_fields: |
paul@13 | 233 | w.write_fields(docnum, list(enumerate(fields))) |
paul@8 | 234 | w.close() |
paul@8 | 235 | |
paul@8 | 236 | f = open("testF", "rb") |
paul@44 | 237 | r = FieldReader(f) |
paul@91 | 238 | r.reset() |
paul@9 | 239 | for docnum, fields in doc_fields: |
paul@9 | 240 | dn, df = r.read_fields() |
paul@9 | 241 | print docnum == dn, docnum, dn |
paul@13 | 242 | print list(enumerate(fields)) == df, list(enumerate(fields)), df |
paul@8 | 243 | r.close() |
paul@8 | 244 | |
paul@69 | 245 | print "- Test field index files." |
paul@9 | 246 | |
paul@9 | 247 | indexed_docs = [ |
paul@9 | 248 | (123, 100000987), |
paul@9 | 249 | (456, 100004321), |
paul@9 | 250 | (789, 100008765) |
paul@9 | 251 | ] |
paul@9 | 252 | |
paul@9 | 253 | f = open("testFI", "wb") |
paul@44 | 254 | w = FieldIndexWriter(f) |
paul@91 | 255 | w.begin(0) |
paul@91 | 256 | w.reset() |
paul@9 | 257 | for docnum, offset in indexed_docs: |
paul@9 | 258 | w.write_document(docnum, offset) |
paul@9 | 259 | w.close() |
paul@9 | 260 | |
paul@9 | 261 | f = open("testFI", "rb") |
paul@44 | 262 | r = FieldIndexReader(f) |
paul@91 | 263 | r.reset() |
paul@9 | 264 | for docnum, offset in indexed_docs: |
paul@9 | 265 | dn, o = r.read_document() |
paul@9 | 266 | print docnum == dn, docnum, dn |
paul@9 | 267 | print offset == o, offset, o |
paul@9 | 268 | r.close() |
paul@9 | 269 | |
paul@69 | 270 | print "- Test field dictionaries." |
paul@9 | 271 | |
paul@9 | 272 | f = open("testF", "wb") |
paul@44 | 273 | w = FieldWriter(f) |
paul@9 | 274 | f2 = open("testFI", "wb") |
paul@44 | 275 | w2 = FieldIndexWriter(f2) |
paul@44 | 276 | wd = FieldDictionaryWriter(w, w2, 3) |
paul@9 | 277 | for docnum, fields in doc_fields: |
paul@13 | 278 | wd.write_fields(docnum, list(enumerate(fields))) |
paul@9 | 279 | wd.close() |
paul@9 | 280 | |
paul@9 | 281 | f = open("testF", "rb") |
paul@44 | 282 | r = FieldReader(f) |
paul@9 | 283 | f2 = open("testFI", "rb") |
paul@44 | 284 | r2 = FieldIndexReader(f2) |
paul@44 | 285 | rd = FieldDictionaryReader(r, r2) |
paul@9 | 286 | doc_fields_reversed = doc_fields[:] |
paul@9 | 287 | doc_fields_reversed.reverse() |
paul@9 | 288 | for docnum, fields in doc_fields_reversed: |
paul@25 | 289 | df = dict(rd.get_fields(docnum)) |
paul@25 | 290 | print dict(enumerate(fields)) == df, dict(enumerate(fields)), df |
paul@9 | 291 | for docnum in (13579, 246810): |
paul@13 | 292 | df = rd.get_fields(docnum) |
paul@9 | 293 | print df is None, df |
paul@13 | 294 | |
paul@69 | 295 | print "- (Test sequential access.)" |
paul@13 | 296 | |
paul@13 | 297 | rd.rewind() |
paul@13 | 298 | for docnum, fields in doc_fields: |
paul@13 | 299 | dn, df = rd.read_fields() |
paul@13 | 300 | print docnum == dn, docnum, dn |
paul@13 | 301 | print list(enumerate(fields)) == df, list(enumerate(fields)), df |
paul@9 | 302 | rd.close() |
paul@9 | 303 | |
paul@69 | 304 | print "- Test terms." |
paul@9 | 305 | |
paul@2 | 306 | terms = [ |
paul@19 | 307 | # term offset frequency doc_frequency |
paul@19 | 308 | ("aardvark", 100000123, 1, 1), |
paul@19 | 309 | ("anteater", 100000456, 2, 1), |
paul@19 | 310 | ("badger", 100000789, 13, 7), |
paul@19 | 311 | ("bull", 1000001234, 59, 17), |
paul@19 | 312 | ("bulldog", 1000002345, 99, 80), |
paul@19 | 313 | ("cat", 1000003456, 89, 28) |
paul@2 | 314 | ] |
paul@2 | 315 | |
paul@2 | 316 | f = open("test", "wb") |
paul@44 | 317 | w = TermWriter(f) |
paul@91 | 318 | w.reset() |
paul@19 | 319 | for term, offset, frequency, doc_frequency in terms: |
paul@19 | 320 | w.write_term(term, offset, frequency, doc_frequency) |
paul@2 | 321 | w.close() |
paul@2 | 322 | |
paul@3 | 323 | f = open("test", "rb") |
paul@44 | 324 | r = TermReader(f) |
paul@91 | 325 | r.reset() |
paul@19 | 326 | for term, offset, frequency, doc_frequency in terms: |
paul@19 | 327 | t, o, fr, df = r.read_term() |
paul@2 | 328 | print term == t, term, t |
paul@2 | 329 | print offset == o, offset, o |
paul@11 | 330 | print frequency == fr, frequency, fr |
paul@19 | 331 | print doc_frequency == df, doc_frequency, df |
paul@2 | 332 | r.close() |
paul@2 | 333 | |
paul@69 | 334 | print "- Test terms in index files." |
paul@9 | 335 | |
paul@3 | 336 | indexed_terms = [ |
paul@19 | 337 | # term offset frequency doc_frequency info_offset |
paul@19 | 338 | ("aardvark", 100000123, 1, 1, 200000321), |
paul@19 | 339 | ("anteater", 100000456, 2, 1, 200000654), |
paul@19 | 340 | ("badger", 100000789, 13, 7, 200000987), |
paul@19 | 341 | ("bull", 1000001234, 59, 17, 200004321), |
paul@19 | 342 | ("bulldog", 1000002345, 99, 80, 200005432), |
paul@19 | 343 | ("cat", 1000003456, 89, 28, 200006543) |
paul@3 | 344 | ] |
paul@3 | 345 | |
paul@3 | 346 | f = open("test", "wb") |
paul@44 | 347 | w = TermIndexWriter(f) |
paul@91 | 348 | w.reset() |
paul@19 | 349 | for term, offset, frequency, doc_frequency, info_offset in indexed_terms: |
paul@19 | 350 | w.write_term(term, offset, frequency, doc_frequency, info_offset) |
paul@3 | 351 | w.close() |
paul@3 | 352 | |
paul@3 | 353 | f = open("test", "rb") |
paul@44 | 354 | r = TermIndexReader(f) |
paul@91 | 355 | r.reset() |
paul@19 | 356 | for term, offset, frequency, doc_frequency, info_offset in indexed_terms: |
paul@19 | 357 | t, o, fr, df, i = r.read_term() |
paul@3 | 358 | print term == t, term, t |
paul@3 | 359 | print offset == o, offset, o |
paul@11 | 360 | print frequency == fr, frequency, fr |
paul@19 | 361 | print doc_frequency == df, doc_frequency, df |
paul@3 | 362 | print info_offset == i, info_offset, i |
paul@3 | 363 | r.close() |
paul@3 | 364 | |
paul@69 | 365 | print "- Test dictionaries with only term data." |
paul@9 | 366 | |
paul@3 | 367 | f = open("test", "wb") |
paul@44 | 368 | w = TermWriter(f) |
paul@3 | 369 | f2 = open("testI", "wb") |
paul@44 | 370 | w2 = TermIndexWriter(f2) |
paul@20 | 371 | f3 = open("testP", "wb") |
paul@44 | 372 | w3 = PositionWriter(f3) |
paul@20 | 373 | f4 = open("testPI", "wb") |
paul@44 | 374 | w4 = PositionIndexWriter(f4) |
paul@44 | 375 | wp = PositionDictionaryWriter(w3, w4, 2) |
paul@44 | 376 | wd = TermDictionaryWriter(w, w2, wp, 3) |
paul@19 | 377 | for term, offset, frequency, doc_frequency in terms: |
paul@19 | 378 | wd._write_term(term, offset, frequency, doc_frequency) |
paul@5 | 379 | wd.close() |
paul@3 | 380 | |
paul@3 | 381 | f = open("test", "rb") |
paul@44 | 382 | r = TermReader(f) |
paul@3 | 383 | f2 = open("testI", "rb") |
paul@44 | 384 | r2 = TermIndexReader(f2) |
paul@68 | 385 | r3 = PositionReader(open("testP", "rb")) |
paul@68 | 386 | r4 = PositionIndexReader(open("testPI", "rb")) |
paul@44 | 387 | rp = PositionDictionaryReader(r3, r4) |
paul@44 | 388 | rd = TermDictionaryReader(r, r2, rp) |
paul@3 | 389 | terms_reversed = terms[:] |
paul@3 | 390 | terms_reversed.reverse() |
paul@19 | 391 | for term, offset, frequency, doc_frequency in terms_reversed: |
paul@19 | 392 | o, fr, df = rd._find_term(term) |
paul@3 | 393 | print offset == o, offset, o |
paul@11 | 394 | print frequency == fr, frequency, fr |
paul@19 | 395 | print doc_frequency == df, doc_frequency, df |
paul@3 | 396 | for term in ("dog", "dingo"): |
paul@11 | 397 | t = rd._find_term(term) |
paul@11 | 398 | print t is None, t |
paul@25 | 399 | |
paul@69 | 400 | print "- (Test term prefix searching.)" |
paul@25 | 401 | |
paul@25 | 402 | print rd.find_terms("a") == ["aardvark", "anteater"], rd.find_terms("a"), ["aardvark", "anteater"] |
paul@25 | 403 | print rd.find_terms("bu") == ["bull", "bulldog"], rd.find_terms("bu"), ["bull", "bulldog"] |
paul@25 | 404 | print rd.find_terms("c") == ["cat"], rd.find_terms("c"), ["cat"] |
paul@25 | 405 | print rd.find_terms("d") == [], rd.find_terms("d"), [] |
paul@5 | 406 | rd.close() |
paul@5 | 407 | |
paul@69 | 408 | print "- Test dictionaries with term and position data." |
paul@9 | 409 | |
paul@5 | 410 | terms_with_positions = [ |
paul@5 | 411 | ("aardvark", [(1, [2, 45, 96]), (20, [13])]), |
paul@5 | 412 | ("anteater", [(1, [43, 44])]), |
paul@5 | 413 | ("badger", [(7, [2, 22, 196]), (19, [55, 1333]), (21, [0])]), |
paul@19 | 414 | ("bull", [(6, [128]), (16, [12]), (26, [1, 3, 5, 7, 9]), (36, [2, 4, 6, 8, 10])]), |
paul@5 | 415 | ("bulldog", [(43, [17, 19, 256, 512])]), |
paul@5 | 416 | ("cat", [(123, [12, 145, 196]), (1200, [113])]) |
paul@5 | 417 | ] |
paul@5 | 418 | |
paul@22 | 419 | position_dict_tests = [ |
paul@22 | 420 | ("badger", 19, [55, 1333]), |
paul@22 | 421 | ("badger", 20, None), |
paul@22 | 422 | ("bull", 6, [128]), |
paul@22 | 423 | ("bull", 26, [1, 3, 5, 7, 9]), |
paul@22 | 424 | ("cat", 111, None), |
paul@22 | 425 | ("cat", 123, [12, 145, 196]), |
paul@22 | 426 | ("cat", 1234, None) |
paul@22 | 427 | ] |
paul@22 | 428 | |
paul@5 | 429 | f = open("test", "wb") |
paul@44 | 430 | w = TermWriter(f) |
paul@5 | 431 | f2 = open("testI", "wb") |
paul@44 | 432 | w2 = TermIndexWriter(f2) |
paul@5 | 433 | f3 = open("testP", "wb") |
paul@44 | 434 | w3 = PositionWriter(f3) |
paul@19 | 435 | f4 = open("testPI", "wb") |
paul@44 | 436 | w4 = PositionIndexWriter(f4) |
paul@44 | 437 | wp = PositionDictionaryWriter(w3, w4, 2) |
paul@44 | 438 | wd = TermDictionaryWriter(w, w2, wp, 3) |
paul@5 | 439 | for term, doc_positions in terms_with_positions: |
paul@5 | 440 | wd.write_term_positions(term, doc_positions) |
paul@5 | 441 | wd.close() |
paul@5 | 442 | |
paul@5 | 443 | f = open("test", "rb") |
paul@44 | 444 | r = TermReader(f) |
paul@5 | 445 | f2 = open("testI", "rb") |
paul@44 | 446 | r2 = TermIndexReader(f2) |
paul@68 | 447 | r3 = PositionReader(open("testP", "rb")) |
paul@68 | 448 | r4 = PositionIndexReader(open("testPI", "rb")) |
paul@44 | 449 | rp = PositionDictionaryReader(r3, r4) |
paul@44 | 450 | rd = TermDictionaryReader(r, r2, rp) |
paul@5 | 451 | terms_reversed = terms_with_positions[:] |
paul@5 | 452 | terms_reversed.reverse() |
paul@5 | 453 | for term, doc_positions in terms_reversed: |
paul@18 | 454 | dp = list(rd.find_positions(term)) |
paul@5 | 455 | print doc_positions == dp, doc_positions, dp |
paul@25 | 456 | for term in ("aaa", "dog", "dingo"): |
paul@5 | 457 | dp = rd.find_positions(term) |
paul@61 | 458 | print dp == [], dp |
paul@12 | 459 | |
paul@69 | 460 | print "- (Test iterators.)" |
paul@22 | 461 | |
paul@22 | 462 | for term, docnum, positions in position_dict_tests: |
paul@22 | 463 | dp = rd.find_positions(term) |
paul@22 | 464 | pos = dp.from_document(docnum) |
paul@22 | 465 | print positions is None and pos is None or pos is not None and positions == list(pos), positions, pos |
paul@22 | 466 | |
paul@69 | 467 | print "- (Test sequential access.)" |
paul@12 | 468 | |
paul@12 | 469 | rd.rewind() |
paul@12 | 470 | for term, doc_positions in terms_with_positions: |
paul@19 | 471 | t, fr, df, dp = rd.read_term() |
paul@18 | 472 | dp = list(dp) |
paul@12 | 473 | print term == t, term, t |
paul@12 | 474 | print doc_positions == dp, doc_positions, dp |
paul@5 | 475 | rd.close() |
paul@3 | 476 | |
paul@69 | 477 | print "- Test high-level index operations (including merging)." |
paul@9 | 478 | |
paul@6 | 479 | docs = [ |
paul@6 | 480 | (1, "The cat sat on the mat"), |
paul@6 | 481 | (2, "Every good boy deserves football"), |
paul@6 | 482 | (13, "One good turn deserves another"), |
paul@6 | 483 | (14, "Every man for himself"), |
paul@6 | 484 | (25, "Red sky at night shepherd's delight"), |
paul@6 | 485 | (36, "She sells sea shells on the sea shore") |
paul@6 | 486 | ] |
paul@6 | 487 | |
paul@6 | 488 | doc_tests = [ |
paul@11 | 489 | ("Every", 2, [(2, [0]), (14, [0])]), |
paul@11 | 490 | ("good", 2, [(2, [1]), (13, [1])]), |
paul@11 | 491 | ("deserves", 2, [(2, [3]), (13, [3])]), |
paul@11 | 492 | ("sea", 2, [(36, [2, 6])]) |
paul@6 | 493 | ] |
paul@6 | 494 | |
paul@21 | 495 | position_tests = [ |
paul@21 | 496 | ("Every", 14, [0]), |
paul@21 | 497 | ("sea", 36, [2, 6]), |
paul@22 | 498 | ("shells", 1, None), |
paul@22 | 499 | ("shells", 37, None) |
paul@21 | 500 | ] |
paul@21 | 501 | |
paul@60 | 502 | phrase_tests = [ |
paul@62 | 503 | (["good", "boy"], [(2, [1, 2])]), |
paul@62 | 504 | (["on", "the"], [(1, [3, 4]), (36, [4, 5])]), |
paul@62 | 505 | (["sea", "shore"], [(36, [6, 7])]) |
paul@60 | 506 | ] |
paul@60 | 507 | |
paul@64 | 508 | index = Index("test_index", 3, 2, 3, 6) |
paul@64 | 509 | wi = index.get_writer() |
paul@6 | 510 | for docnum, text in docs: |
paul@44 | 511 | doc = Document(docnum) |
paul@6 | 512 | for position, term in enumerate(text.split()): |
paul@28 | 513 | doc.add_position(term, position) |
paul@28 | 514 | doc.add_field(123, text) |
paul@28 | 515 | wi.add_document(doc) |
paul@6 | 516 | wi.close() |
paul@6 | 517 | |
paul@7 | 518 | rd = index.get_reader() |
paul@60 | 519 | |
paul@69 | 520 | print "- (Test searching.)" |
paul@60 | 521 | |
paul@11 | 522 | for term, frequency, doc_positions in doc_tests: |
paul@18 | 523 | dp = list(rd.find_positions(term)) |
paul@6 | 524 | print doc_positions == dp, doc_positions, dp |
paul@11 | 525 | fr = rd.get_frequency(term) |
paul@11 | 526 | print frequency == fr, frequency, fr |
paul@60 | 527 | |
paul@69 | 528 | print "- (Test fields.)" |
paul@60 | 529 | |
paul@10 | 530 | for docnum, text in docs: |
paul@25 | 531 | df = dict(rd.get_fields(docnum)) |
paul@25 | 532 | print df[123] == text, text, df[123] |
paul@60 | 533 | |
paul@69 | 534 | print "- (Test navigation.)" |
paul@60 | 535 | |
paul@21 | 536 | for term, docnum, positions in position_tests: |
paul@21 | 537 | dp = rd.find_positions(term) |
paul@22 | 538 | pos = dp.from_document(docnum) |
paul@22 | 539 | print positions is None and pos is None or pos is not None and positions == list(pos), positions, pos |
paul@60 | 540 | |
paul@69 | 541 | print "- (Test phrases.)" |
paul@60 | 542 | |
paul@60 | 543 | for terms, results in phrase_tests: |
paul@60 | 544 | res = list(rd.find_common_positions(terms)) |
paul@60 | 545 | print results == res, results, res |
paul@60 | 546 | |
paul@7 | 547 | index.close() |
paul@6 | 548 | |
paul@77 | 549 | docs2 = [ |
paul@77 | 550 | ((1, 0), "The cat sat on the mat"), |
paul@77 | 551 | ((1, 2), "Every good boy deserves football"), |
paul@77 | 552 | ((13, 1), "One good turn deserves another"), |
paul@77 | 553 | ((14, 0), "Every man for himself"), |
paul@77 | 554 | ((14, 25), "Red sky at night shepherd's delight"), |
paul@77 | 555 | ((36, 12), "She sells sea shells on the sea shore") |
paul@77 | 556 | ] |
paul@77 | 557 | |
paul@77 | 558 | doc_tests2 = [ |
paul@77 | 559 | ("Every", 2, [((1, 2), [(0, 0)]), ((14, 0), [(0, 0)])]), |
paul@77 | 560 | ("good", 2, [((1, 2), [(1, 6)]), ((13, 1), [(1, 4)])]), |
paul@77 | 561 | ("deserves", 2, [((1, 2), [(3, 15)]), ((13, 1), [(3, 14)])]), |
paul@77 | 562 | ("sea", 2, [((36, 12), [(2, 10), (6, 28)])]) |
paul@77 | 563 | ] |
paul@77 | 564 | |
paul@77 | 565 | position_tests2 = [ |
paul@77 | 566 | ("Every", (14, 0), [(0, 0)]), |
paul@77 | 567 | ("sea", (36, 12), [(2, 10), (6, 28)]), |
paul@77 | 568 | ("shells", (1, 0), None), |
paul@77 | 569 | ("shells", (37, 0), None) |
paul@77 | 570 | ] |
paul@77 | 571 | |
paul@77 | 572 | phrase_tests2 = [ |
paul@77 | 573 | (["good", "boy"], [((1, 2), [(1, 6), (2, 11)])]), |
paul@77 | 574 | (["on", "the"], [((1, 0), [(3, 12), (4, 15)]), ((36, 12), [(4, 21), (5, 24)])]), |
paul@77 | 575 | (["sea", "shore"], [((36, 12), [(6, 28), (7, 32)])]) |
paul@77 | 576 | ] |
paul@77 | 577 | |
paul@77 | 578 | index = Index("test_indexT", 3, 2, 3, 6) |
paul@77 | 579 | wi = index.get_writer() |
paul@77 | 580 | for docnum, text in docs2: |
paul@77 | 581 | doc = Document(docnum) |
paul@77 | 582 | offset = 0 |
paul@77 | 583 | for position, term in enumerate(text.split()): |
paul@77 | 584 | doc.add_position(term, (position, offset)) |
paul@77 | 585 | offset += len(term) + 1 # assume one space after the term |
paul@77 | 586 | doc.add_field(123, text) |
paul@77 | 587 | wi.add_document(doc) |
paul@77 | 588 | wi.close() |
paul@77 | 589 | |
paul@77 | 590 | rd = index.get_reader() |
paul@77 | 591 | |
paul@77 | 592 | print "- (Test searching.)" |
paul@77 | 593 | |
paul@77 | 594 | for term, frequency, doc_positions in doc_tests2: |
paul@77 | 595 | dp = list(rd.find_positions(term)) |
paul@77 | 596 | print doc_positions == dp, doc_positions, dp |
paul@77 | 597 | fr = rd.get_frequency(term) |
paul@77 | 598 | print frequency == fr, frequency, fr |
paul@77 | 599 | |
paul@77 | 600 | print "- (Test fields.)" |
paul@77 | 601 | |
paul@77 | 602 | for docnum, text in docs2: |
paul@77 | 603 | df = dict(rd.get_fields(docnum)) |
paul@77 | 604 | print df[123] == text, text, df[123] |
paul@77 | 605 | |
paul@77 | 606 | print "- (Test navigation.)" |
paul@77 | 607 | |
paul@77 | 608 | for term, docnum, positions in position_tests2: |
paul@77 | 609 | dp = rd.find_positions(term) |
paul@77 | 610 | pos = dp.from_document(docnum) |
paul@77 | 611 | print positions is None and pos is None or pos is not None and positions == list(pos), positions, pos |
paul@77 | 612 | |
paul@77 | 613 | print "- (Test phrases.)" |
paul@77 | 614 | |
paul@77 | 615 | for terms, results in phrase_tests2: |
paul@77 | 616 | res = list(rd.find_common_positions(terms)) |
paul@77 | 617 | print results == res, results, res |
paul@77 | 618 | |
paul@77 | 619 | index.close() |
paul@77 | 620 | |
paul@69 | 621 | print "- Test index updates." |
paul@58 | 622 | |
paul@58 | 623 | index = Index("test_index") |
paul@64 | 624 | index2 = Index("test_index2", 3, 2, 3, 6) |
paul@64 | 625 | wi = index2.get_writer() |
paul@58 | 626 | for docnum, text in docs: |
paul@58 | 627 | |
paul@58 | 628 | # Add the same documents but with different numbers. |
paul@58 | 629 | |
paul@58 | 630 | doc = Document(docnum + 100) |
paul@58 | 631 | for position, term in enumerate(text.split()): |
paul@58 | 632 | doc.add_position(term, position) |
paul@58 | 633 | doc.add_field(123, text) |
paul@58 | 634 | wi.add_document(doc) |
paul@58 | 635 | wi.close() |
paul@58 | 636 | |
paul@58 | 637 | index2.update([index]) |
paul@58 | 638 | index.close() |
paul@58 | 639 | |
paul@58 | 640 | rd = index2.get_reader() |
paul@58 | 641 | for term, frequency, doc_positions in doc_tests: |
paul@58 | 642 | |
paul@58 | 643 | # Add the extra documents to the expected result. |
paul@58 | 644 | |
paul@59 | 645 | orig_doc_positions = doc_positions |
paul@59 | 646 | doc_positions = doc_positions[:] |
paul@59 | 647 | |
paul@59 | 648 | for docnum, positions in orig_doc_positions: |
paul@58 | 649 | doc_positions.append((docnum + 100, positions)) |
paul@58 | 650 | frequency *= 2 |
paul@58 | 651 | |
paul@58 | 652 | dp = list(rd.find_positions(term)) |
paul@58 | 653 | print doc_positions == dp, doc_positions, dp |
paul@58 | 654 | fr = rd.get_frequency(term) |
paul@58 | 655 | print frequency == fr, frequency, fr |
paul@58 | 656 | index2.close() |
paul@58 | 657 | |
paul@69 | 658 | print "- (Test update of an empty index.)" |
paul@59 | 659 | |
paul@59 | 660 | index = Index("test_index") |
paul@59 | 661 | index3 = Index("test_index3") |
paul@59 | 662 | index3.update([index]) |
paul@59 | 663 | index.close() |
paul@59 | 664 | |
paul@59 | 665 | rd = index3.get_reader() |
paul@59 | 666 | for term, frequency, doc_positions in doc_tests: |
paul@59 | 667 | dp = list(rd.find_positions(term)) |
paul@59 | 668 | print doc_positions == dp, doc_positions, dp |
paul@59 | 669 | fr = rd.get_frequency(term) |
paul@59 | 670 | print frequency == fr, frequency, fr |
paul@59 | 671 | index3.close() |
paul@59 | 672 | |
paul@0 | 673 | # vim: tabstop=4 expandtab shiftwidth=4 |