| 1 |
/* |
| 2 |
** |
| 3 |
** This program and library is free software; you can redistribute it and/or |
| 4 |
** modify it under the terms of the GNU (Library) General Public License |
| 5 |
** as published by the Free Software Foundation; either version 2 |
| 6 |
** of the License, or any later version. |
| 7 |
** |
| 8 |
** This program is distributed in the hope that it will be useful, |
| 9 |
** but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 |
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 11 |
** GNU (Library) General Public License for more details. |
| 12 |
** |
| 13 |
** You should have received a copy of the GNU (Library) General Public License |
| 14 |
** along with this program; if not, write to the Free Software |
| 15 |
** Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. |
| 16 |
** |
| 17 |
** |
| 18 |
** |
| 19 |
** 2001-05-07 jmruiz init coding |
| 20 |
** |
| 21 |
*/ |
| 22 |
|
| 23 |
#include "swish.h" |
| 24 |
#include "merge.h" |
| 25 |
#include "docprop.h" |
| 26 |
#include "hash.h" |
| 27 |
#include "string.h" |
| 28 |
#include "mem.h" |
| 29 |
#include "db.h" |
| 30 |
#include "compress.h" |
| 31 |
#include "index.h" |
| 32 |
#include "search.h" |
| 33 |
#include "result_output.h" |
| 34 |
#include "metanames.h" |
| 35 |
#include "dump.h" |
| 36 |
|
| 37 |
|
| 38 |
|
| 39 |
|
| 40 |
void dump_index_file_list( SWISH *sw, IndexFILE *indexf ) |
| 41 |
{ |
| 42 |
int i; |
| 43 |
int end = indexf->header.totalfiles; |
| 44 |
|
| 45 |
i = sw->Search->beginhits ? sw->Search->beginhits - 1 : 0; |
| 46 |
|
| 47 |
if ( i >= indexf->header.totalfiles ) |
| 48 |
{ |
| 49 |
printf("Hey, there are only %d files\n", indexf->header.totalfiles ); |
| 50 |
exit(-1); |
| 51 |
} |
| 52 |
|
| 53 |
end = indexf->header.totalfiles; |
| 54 |
|
| 55 |
if ( sw->Search->maxhits > 0 ) |
| 56 |
{ |
| 57 |
end = i + sw->Search->maxhits; |
| 58 |
if ( end > indexf->header.totalfiles ) |
| 59 |
end = indexf->header.totalfiles; |
| 60 |
} |
| 61 |
|
| 62 |
|
| 63 |
printf("\n\n-----> FILES in index %s <-----\n", indexf->line ); |
| 64 |
|
| 65 |
for (; i < end; i++) |
| 66 |
{ |
| 67 |
FileRec fi; |
| 68 |
|
| 69 |
memset( &fi, 0, sizeof( FileRec ) ); |
| 70 |
|
| 71 |
fi.filenum = i+1; |
| 72 |
|
| 73 |
fflush(stdout); |
| 74 |
printf("Dumping File Properties for File Number: %d\n", i+1); |
| 75 |
|
| 76 |
|
| 77 |
dump_file_properties( indexf, &fi ); |
| 78 |
printf("\n"); |
| 79 |
|
| 80 |
|
| 81 |
printf("ReadAllDocProperties:\n"); |
| 82 |
fi.docProperties = ReadAllDocPropertiesFromDisk( sw, indexf, i+1 ); |
| 83 |
dump_file_properties( indexf, &fi ); |
| 84 |
freefileinfo( &fi ); |
| 85 |
|
| 86 |
printf("\n"); |
| 87 |
|
| 88 |
|
| 89 |
/* dump one at a time */ |
| 90 |
{ |
| 91 |
propEntry *p; |
| 92 |
int j; |
| 93 |
struct metaEntry *meta_entry; |
| 94 |
INDEXDATAHEADER *header = &indexf->header; |
| 95 |
int count = header->property_count; |
| 96 |
|
| 97 |
printf("ReadSingleDocPropertiesFromDisk:\n"); |
| 98 |
|
| 99 |
for (j=0; j< count; j++) // just for testing |
| 100 |
{ |
| 101 |
int metaID = header->propIDX_to_metaID[j]; |
| 102 |
|
| 103 |
if ( !(p = ReadSingleDocPropertiesFromDisk(sw, indexf, &fi, metaID, 0 )) ) |
| 104 |
continue; |
| 105 |
|
| 106 |
meta_entry = getPropNameByID( &indexf->header, metaID ); |
| 107 |
dump_single_property( p, meta_entry ); |
| 108 |
|
| 109 |
{ // show compression |
| 110 |
char *buffer; |
| 111 |
int uncompressed_len; |
| 112 |
int buf_len; |
| 113 |
|
| 114 |
if ( (buffer = DB_ReadProperty( sw, indexf, &fi, meta_entry->metaID, &buf_len, &uncompressed_len, indexf->DB ))) |
| 115 |
{ |
| 116 |
if ( uncompressed_len ) |
| 117 |
printf(" %20s: %d -> %d (%4.2f%%)\n", "**Compressed**", uncompressed_len , buf_len, (float)buf_len/(float)uncompressed_len * 100.00f ); |
| 118 |
|
| 119 |
efree(buffer); |
| 120 |
} |
| 121 |
} |
| 122 |
|
| 123 |
|
| 124 |
|
| 125 |
freeProperty( p ); |
| 126 |
} |
| 127 |
} |
| 128 |
printf("\n"); |
| 129 |
|
| 130 |
|
| 131 |
freefileinfo(&fi); |
| 132 |
} |
| 133 |
printf("\nNumber of File Entries: %d\n", indexf->header.totalfiles); |
| 134 |
fflush(stdout); |
| 135 |
} |
| 136 |
|
| 137 |
|
| 138 |
|
| 139 |
|
| 140 |
/* Prints out the data in an index DB */ |
| 141 |
void DB_decompress(SWISH * sw, IndexFILE * indexf) |
| 142 |
{ |
| 143 |
int i, |
| 144 |
j, |
| 145 |
c, |
| 146 |
fieldnum, |
| 147 |
frequency, |
| 148 |
metaname, |
| 149 |
tmpval, |
| 150 |
filenum, |
| 151 |
*posdata; |
| 152 |
unsigned long nextposmetaname; |
| 153 |
char word[2]; |
| 154 |
char *resultword; |
| 155 |
unsigned char *worddata, *s, flag; |
| 156 |
int sz_worddata; |
| 157 |
long wordID; |
| 158 |
|
| 159 |
|
| 160 |
|
| 161 |
indexf->DB = DB_Open(sw, indexf->line,DB_READ); |
| 162 |
|
| 163 |
metaname = 0; |
| 164 |
|
| 165 |
nextposmetaname = 0L; |
| 166 |
|
| 167 |
c = 0; |
| 168 |
|
| 169 |
frequency = 0; |
| 170 |
|
| 171 |
/* Read header */ |
| 172 |
read_header(sw, &indexf->header, indexf->DB); |
| 173 |
|
| 174 |
|
| 175 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_HEADER) ) |
| 176 |
resultPrintHeader(sw, 0, &indexf->header, indexf->line, 0); |
| 177 |
|
| 178 |
fieldnum = 0; |
| 179 |
|
| 180 |
|
| 181 |
/* Do metanames first as that will be helpful for decoding next */ |
| 182 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_METANAMES) ) |
| 183 |
dump_metanames( sw, indexf, 1 ); |
| 184 |
|
| 185 |
if (DEBUG_MASK & DEBUG_INDEX_WORDS_ONLY) |
| 186 |
{ |
| 187 |
DB_InitReadWords(sw, indexf->DB); |
| 188 |
|
| 189 |
for( j = 0; j < 256; j++ ) |
| 190 |
{ |
| 191 |
word[0] = (unsigned char) j; |
| 192 |
word[1] = '\0'; |
| 193 |
DB_ReadFirstWordInvertedIndex(sw, word,&resultword,&wordID,indexf->DB); |
| 194 |
|
| 195 |
while(wordID) |
| 196 |
{ |
| 197 |
printf("%s\n",resultword); |
| 198 |
|
| 199 |
|
| 200 |
efree(resultword); |
| 201 |
DB_ReadNextWordInvertedIndex(sw, word,&resultword,&wordID,indexf->DB); |
| 202 |
|
| 203 |
} |
| 204 |
} |
| 205 |
DB_EndReadWords(sw, indexf->DB); |
| 206 |
} |
| 207 |
|
| 208 |
|
| 209 |
else if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_WORDS | DEBUG_INDEX_WORDS_FULL | DEBUG_INDEX_WORDS_META) ) |
| 210 |
{ |
| 211 |
int *meta_used; |
| 212 |
int end_meta = 0; |
| 213 |
|
| 214 |
printf("\n-----> WORD INFO in index %s <-----\n", indexf->line); |
| 215 |
|
| 216 |
for(i = 0; i < indexf->header.metaCounter; i++) |
| 217 |
if ( indexf->header.metaEntryArray[i]->metaID > end_meta ) |
| 218 |
end_meta = indexf->header.metaEntryArray[i]->metaID; |
| 219 |
|
| 220 |
meta_used = emalloc( sizeof(int) * ( end_meta + 1) ); |
| 221 |
|
| 222 |
/* _META only reports which tags the words are found in */ |
| 223 |
for(i = 0; i <= end_meta; i++) |
| 224 |
meta_used[i] = 0; |
| 225 |
|
| 226 |
|
| 227 |
DB_InitReadWords(sw, indexf->DB); |
| 228 |
|
| 229 |
for(j=1;j<256;j++) |
| 230 |
{ |
| 231 |
word[0] = (unsigned char) j; word[1] = '\0'; |
| 232 |
DB_ReadFirstWordInvertedIndex(sw, word,&resultword,&wordID,indexf->DB); |
| 233 |
|
| 234 |
while(wordID && (((int)((unsigned char)resultword[0]))== j)) |
| 235 |
{ |
| 236 |
printf("\n%s",resultword); |
| 237 |
|
| 238 |
/* Read Word's data */ |
| 239 |
DB_ReadWordData(sw, wordID, &worddata, &sz_worddata, indexf->DB); |
| 240 |
|
| 241 |
/* parse and print word's data */ |
| 242 |
s = worddata; |
| 243 |
|
| 244 |
tmpval = uncompress2(&s); /* tfrequency */ |
| 245 |
metaname = uncompress2(&s); /* metaname */ |
| 246 |
if (metaname) |
| 247 |
{ |
| 248 |
nextposmetaname = UNPACKLONG2(s); |
| 249 |
s += sizeof(long); |
| 250 |
} |
| 251 |
|
| 252 |
filenum = 0; |
| 253 |
while(1) |
| 254 |
{ /* Read on all items */ |
| 255 |
uncompress_location_values(&s,&flag,&tmpval,&frequency); |
| 256 |
filenum += tmpval; |
| 257 |
posdata = (int *) emalloc(frequency * sizeof(int)); |
| 258 |
uncompress_location_positions(&s,flag,frequency,posdata); |
| 259 |
|
| 260 |
|
| 261 |
// if (sw->verbose >= 4) |
| 262 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL|DEBUG_INDEX_WORDS_FULL)) |
| 263 |
{ |
| 264 |
struct metaEntry *m; |
| 265 |
|
| 266 |
printf("\n Meta:%d", metaname); |
| 267 |
|
| 268 |
|
| 269 |
/* Get path from property list */ |
| 270 |
if ( (m = getPropNameByName( &sw->indexlist->header, AUTOPROPERTY_DOCPATH )) ) |
| 271 |
{ |
| 272 |
RESULT r; |
| 273 |
char *s; |
| 274 |
|
| 275 |
memset( &r, 0, sizeof( RESULT ) ); |
| 276 |
|
| 277 |
r.indexf = indexf; |
| 278 |
r.filenum = filenum; |
| 279 |
r.fi.filenum = filenum; |
| 280 |
|
| 281 |
s = getResultPropAsString( sw, &r, m->metaID); |
| 282 |
|
| 283 |
printf(" %s", s ); |
| 284 |
efree( s ); |
| 285 |
|
| 286 |
} |
| 287 |
else |
| 288 |
printf(" Failed to lookup meta entry"); |
| 289 |
|
| 290 |
|
| 291 |
printf(" Freq:%d", frequency); |
| 292 |
printf(" Pos/Struct:"); |
| 293 |
} |
| 294 |
else if ( DEBUG_MASK & DEBUG_INDEX_WORDS_META) |
| 295 |
meta_used[ metaname ]++; |
| 296 |
else |
| 297 |
{ |
| 298 |
printf(" [%d", metaname); |
| 299 |
printf(" %d", filenum); |
| 300 |
printf(" %d (", frequency); |
| 301 |
} |
| 302 |
|
| 303 |
|
| 304 |
for (i = 0; i < frequency; i++) |
| 305 |
{ |
| 306 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_WORDS_FULL)) |
| 307 |
//if (sw->verbose >= 4) |
| 308 |
{ |
| 309 |
if (i) |
| 310 |
printf(",%d/%x", GET_POSITION(posdata[i]),GET_STRUCTURE(posdata[i])); |
| 311 |
else |
| 312 |
printf("%d/%x", GET_POSITION(posdata[i]), GET_STRUCTURE(posdata[i])); |
| 313 |
} |
| 314 |
else if ( DEBUG_MASK & DEBUG_INDEX_WORDS) |
| 315 |
{ |
| 316 |
if (i) |
| 317 |
printf(" %d/%x", GET_POSITION(posdata[i]),GET_STRUCTURE(posdata[i])); |
| 318 |
else |
| 319 |
printf("%d/%x", GET_POSITION(posdata[i]),GET_STRUCTURE(posdata[i])); |
| 320 |
} |
| 321 |
} |
| 322 |
|
| 323 |
efree(posdata); |
| 324 |
|
| 325 |
if ( DEBUG_MASK & DEBUG_INDEX_WORDS ) |
| 326 |
printf(")]"); |
| 327 |
|
| 328 |
if ((s - worddata) == sz_worddata) |
| 329 |
break; /* End of worddata */ |
| 330 |
|
| 331 |
if ((unsigned long)(s - worddata) == nextposmetaname) |
| 332 |
{ |
| 333 |
filenum = 0; |
| 334 |
metaname = uncompress2(&s); |
| 335 |
if (metaname) |
| 336 |
{ |
| 337 |
nextposmetaname = UNPACKLONG2(s); |
| 338 |
s += sizeof(long); |
| 339 |
} |
| 340 |
else |
| 341 |
nextposmetaname = 0L; |
| 342 |
} |
| 343 |
} |
| 344 |
|
| 345 |
if ( DEBUG_MASK & DEBUG_INDEX_WORDS_META) |
| 346 |
{ |
| 347 |
for(i = 0; i <= end_meta; i++) |
| 348 |
{ |
| 349 |
if ( meta_used[i] ) |
| 350 |
printf( "\t%d", i ); |
| 351 |
meta_used[i] = 0; |
| 352 |
} |
| 353 |
} |
| 354 |
|
| 355 |
|
| 356 |
if ( !( DEBUG_MASK & DEBUG_INDEX_WORDS_META )) |
| 357 |
printf("\n"); |
| 358 |
|
| 359 |
efree(worddata); |
| 360 |
efree(resultword); |
| 361 |
DB_ReadNextWordInvertedIndex(sw, word,&resultword,&wordID,indexf->DB); |
| 362 |
} |
| 363 |
} |
| 364 |
DB_EndReadWords(sw, indexf->DB); |
| 365 |
|
| 366 |
efree( meta_used ); |
| 367 |
} |
| 368 |
|
| 369 |
|
| 370 |
|
| 371 |
/* Decode Stop Words: All them are in just one line */ |
| 372 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_STOPWORDS) ) |
| 373 |
{ |
| 374 |
printf("\n\n-----> STOP WORDS in %s <-----\n" , indexf->line); |
| 375 |
for(i=0;i<indexf->header.stopPos;i++) |
| 376 |
printf("%s ",indexf->header.stopList[i]); |
| 377 |
printf("\n"); |
| 378 |
} |
| 379 |
|
| 380 |
|
| 381 |
|
| 382 |
/* Decode File Info */ |
| 383 |
if (DEBUG_MASK & (DEBUG_INDEX_ALL | DEBUG_INDEX_FILES) ) |
| 384 |
dump_index_file_list( sw, indexf ); |
| 385 |
|
| 386 |
|
| 387 |
DB_Close(sw, indexf->DB); |
| 388 |
|
| 389 |
} |
| 390 |
|
| 391 |
|
| 392 |
int check_sorted_index( SWISH *sw, IndexFILE *indexf, struct metaEntry *m ) |
| 393 |
{ |
| 394 |
unsigned char *buffer; |
| 395 |
int sz_buffer; |
| 396 |
|
| 397 |
DB_InitReadSortedIndex(sw, indexf->DB); |
| 398 |
|
| 399 |
/* Get the sorted index of the property */ |
| 400 |
DB_ReadSortedIndex(sw, m->metaID, &buffer, &sz_buffer, indexf->DB); |
| 401 |
|
| 402 |
if ( sz_buffer ) |
| 403 |
efree( buffer ); |
| 404 |
|
| 405 |
/* Table doesn't exist */ |
| 406 |
return sz_buffer; |
| 407 |
} |
| 408 |
|
| 409 |
|
| 410 |
void dump_metanames( SWISH *sw, IndexFILE *indexf, int check_presorted ) |
| 411 |
{ |
| 412 |
struct metaEntry *meta_entry; |
| 413 |
int i; |
| 414 |
|
| 415 |
printf("\n\n-----> METANAMES for %s <-----\n", indexf->line ); |
| 416 |
for(i = 0; i < indexf->header.metaCounter; i++) |
| 417 |
{ |
| 418 |
meta_entry = indexf->header.metaEntryArray[i]; |
| 419 |
|
| 420 |
printf("%20s : id=%2d type=%2d ",meta_entry->metaName, meta_entry->metaID, meta_entry->metaType); |
| 421 |
|
| 422 |
if ( is_meta_index( meta_entry ) ) |
| 423 |
printf(" META_INDEX Rank Bias=%3d", meta_entry->rank_bias ); |
| 424 |
|
| 425 |
|
| 426 |
|
| 427 |
if ( is_meta_internal( meta_entry ) ) |
| 428 |
printf(" META_INTERNAL"); |
| 429 |
|
| 430 |
|
| 431 |
if ( is_meta_property( meta_entry ) ) |
| 432 |
{ |
| 433 |
printf(" META_PROP:"); |
| 434 |
|
| 435 |
if ( is_meta_string(meta_entry) ) |
| 436 |
printf("STRING(case:%s)", is_meta_ignore_case(meta_entry)? "ignore" : "compare"); |
| 437 |
|
| 438 |
else if ( is_meta_date(meta_entry) ) |
| 439 |
printf("DATE"); |
| 440 |
|
| 441 |
else if ( is_meta_number(meta_entry) ) |
| 442 |
printf("NUMBER"); |
| 443 |
|
| 444 |
else |
| 445 |
printf("unknown!"); |
| 446 |
} |
| 447 |
|
| 448 |
|
| 449 |
if ( check_presorted && check_sorted_index( sw, indexf, meta_entry) ) |
| 450 |
printf(" *presorted*"); |
| 451 |
|
| 452 |
|
| 453 |
if ( meta_entry->alias ) |
| 454 |
{ |
| 455 |
struct metaEntry *m = is_meta_index( meta_entry ) |
| 456 |
? getMetaNameByID( &indexf->header, meta_entry->alias ) |
| 457 |
: getPropNameByID( &indexf->header, meta_entry->alias ); |
| 458 |
|
| 459 |
printf(" [Alias for %s (%d)]", m->metaName, m->metaID ); |
| 460 |
} |
| 461 |
|
| 462 |
|
| 463 |
printf("\n"); |
| 464 |
|
| 465 |
} |
| 466 |
printf("\n"); |
| 467 |
} |
| 468 |
|
| 469 |
|
Parent Directory
| 
Revision Log
| 
Revision Graph