| 1 |
/* |
| 2 |
* $LynxId: HTVMS_WaisProt.c,v 1.10 2020/01/21 22:10:03 tom Exp $ |
| 3 |
* |
| 4 |
* HTVMS_WAISProt.c |
| 5 |
* |
| 6 |
* Adaptation for Lynx by F.Macrides ([email protected]) |
| 7 |
* |
| 8 |
* 31-May-1994 FM Initial version. |
| 9 |
* |
| 10 |
*----------------------------------------------------------------------*/ |
| 11 |
|
| 12 |
/* |
| 13 |
* Routines originally from WProt.c -- FM |
| 14 |
* |
| 15 |
*----------------------------------------------------------------------*/ |
| 16 |
/* WIDE AREA INFORMATION SERVER SOFTWARE: |
| 17 |
* No guarantees or restrictions. See the readme file for the full standard |
| 18 |
* disclaimer. |
| 19 |
|
| 20 |
* 3.26.90 Harry Morris, [email protected] |
| 21 |
* 3.30.90 Harry Morris |
| 22 |
* - removed chunk code from WAISSearchAPDU, |
| 23 |
* - added makeWAISQueryType1Query() and readWAISType1Query() which replace |
| 24 |
* makeWAISQueryTerms() and makeWAISQueryDocs(). |
| 25 |
* 4.11.90 HWM - generalized conditional includes (see c-dialect.h) |
| 26 |
* - renamed makeWAISType1Query() to makeWAISTextQuery() |
| 27 |
* renamed readWAISType1Query() to readWAISTextQuery() |
| 28 |
* 5.29.90 TS - fixed bug in makeWAISQueryDocs |
| 29 |
* added CSTFreeWAISFoo functions |
| 30 |
*/ |
| 31 |
|
| 32 |
#define _C_WAIS_protocol_ |
| 33 |
|
| 34 |
/* This file implements the Z39.50 extensions required for WAIS |
| 35 |
*/ |
| 36 |
|
| 37 |
#include <HTUtils.h> |
| 38 |
#include <HTVMS_WaisUI.h> |
| 39 |
#include <HTVMS_WaisProt.h> |
| 40 |
|
| 41 |
#include <LYLeaks.h> |
| 42 |
|
| 43 |
/* very rough estimates of the size of an object */ |
| 44 |
#define DefWAISInitResponseSize (size_t)200 |
| 45 |
#define DefWAISSearchSize (size_t)3000 |
| 46 |
#define DefWAISSearchResponseSize (size_t)6000 |
| 47 |
#define DefWAISPresentSize (size_t)1000 |
| 48 |
#define DefWAISPresentResponseSize (size_t)6000 |
| 49 |
#define DefWAISDocHeaderSize (size_t)500 |
| 50 |
#define DefWAISShortHeaderSize (size_t)200 |
| 51 |
#define DefWAISLongHeaderSize (size_t)800 |
| 52 |
#define DefWAISDocTextSize (size_t)6000 |
| 53 |
#define DefWAISDocHeadlineSize (size_t)500 |
| 54 |
#define DefWAISDocCodeSize (size_t)500 |
| 55 |
|
| 56 |
#define RESERVE_SPACE_FOR_WAIS_HEADER(len) \ |
| 57 |
if (*len > 0) \ |
| 58 |
*len -= header_len; |
| 59 |
|
| 60 |
#define S_MALLOC(type) (type*)s_malloc(sizeof(type)) |
| 61 |
#define S_MALLOC2(type) (type*)s_malloc(sizeof(type) * 2) |
| 62 |
|
| 63 |
#define S_REALLOC2(type, ptr, num) (type*)s_realloc((char*)ptr, (sizeof(type) * (num + 2))) |
| 64 |
|
| 65 |
/*----------------------------------------------------------------------*/ |
| 66 |
|
| 67 |
static unsigned long userInfoTagSize(data_tag tag, |
| 68 |
unsigned long length) |
| 69 |
/* return the number of bytes required to write the user info tag and |
| 70 |
length |
| 71 |
*/ |
| 72 |
{ |
| 73 |
unsigned long size; |
| 74 |
|
| 75 |
/* calculate bytes required to represent tag. max tag is 16K */ |
| 76 |
size = writtenCompressedIntSize(tag); |
| 77 |
size += writtenCompressedIntSize(length); |
| 78 |
|
| 79 |
return (size); |
| 80 |
} |
| 81 |
|
| 82 |
/*----------------------------------------------------------------------*/ |
| 83 |
|
| 84 |
static char *writeUserInfoHeader(data_tag tag, |
| 85 |
long infoSize, |
| 86 |
long estHeaderSize, |
| 87 |
char *buffer, |
| 88 |
long *len) |
| 89 |
/* write the tag and size, making sure the info fits. return the true end |
| 90 |
of the info (after adjustment) note that the argument infoSize includes |
| 91 |
estHeaderSize. Note that the argument len is the number of bytes remaining |
| 92 |
in the buffer. Since we write the tag and size at the beginning of the |
| 93 |
buffer (in space that we reserved) we don't want to pass len the calls which |
| 94 |
do that writing. |
| 95 |
*/ |
| 96 |
{ |
| 97 |
long dummyLen = 100; /* plenty of space for a tag and size */ |
| 98 |
char *buf = buffer; |
| 99 |
long realSize = infoSize - estHeaderSize; |
| 100 |
long realHeaderSize = userInfoTagSize(tag, realSize); |
| 101 |
|
| 102 |
if (buffer == NULL || *len == 0) |
| 103 |
return (NULL); |
| 104 |
|
| 105 |
/* write the tag */ |
| 106 |
buf = writeTag(tag, buf, &dummyLen); |
| 107 |
|
| 108 |
/* see if the if the header size was correct. if not, |
| 109 |
we have to shift the info to fit the real header size */ |
| 110 |
if (estHeaderSize != realHeaderSize) { /* make sure there is enough space */ |
| 111 |
CHECK_FOR_SPACE_LEFT(realHeaderSize - estHeaderSize, len); |
| 112 |
memmove(buffer + realHeaderSize, buffer + estHeaderSize, (size_t) (realSize)); |
| 113 |
} |
| 114 |
|
| 115 |
/* write the size */ |
| 116 |
writeCompressedInteger(realSize, buf, &dummyLen); |
| 117 |
|
| 118 |
/* return the true end of buffer */ |
| 119 |
return (buffer + realHeaderSize + realSize); |
| 120 |
} |
| 121 |
|
| 122 |
/*----------------------------------------------------------------------*/ |
| 123 |
|
| 124 |
static char *readUserInfoHeader(data_tag *tag, |
| 125 |
unsigned long *num, |
| 126 |
char *buffer) |
| 127 |
/* read the tag and size */ |
| 128 |
{ |
| 129 |
char *buf = buffer; |
| 130 |
|
| 131 |
buf = readTag(tag, buf); |
| 132 |
buf = readCompressedInteger(num, buf); |
| 133 |
return (buf); |
| 134 |
} |
| 135 |
|
| 136 |
/*----------------------------------------------------------------------*/ |
| 137 |
|
| 138 |
WAISInitResponse *makeWAISInitResponse(long chunkCode, |
| 139 |
long chunkIDLen, |
| 140 |
char *chunkMarker, |
| 141 |
char *highlightMarker, |
| 142 |
char *deHighlightMarker, |
| 143 |
char *newLineChars) |
| 144 |
/* create a WAIS init response object */ |
| 145 |
{ |
| 146 |
WAISInitResponse *init = S_MALLOC(WAISInitResponse); |
| 147 |
|
| 148 |
init->ChunkCode = chunkCode; /* note: none are copied! */ |
| 149 |
init->ChunkIDLength = chunkIDLen; |
| 150 |
init->ChunkMarker = chunkMarker; |
| 151 |
init->HighlightMarker = highlightMarker; |
| 152 |
init->DeHighlightMarker = deHighlightMarker; |
| 153 |
init->NewlineCharacters = newLineChars; |
| 154 |
|
| 155 |
return (init); |
| 156 |
} |
| 157 |
|
| 158 |
/*----------------------------------------------------------------------*/ |
| 159 |
|
| 160 |
void freeWAISInitResponse(WAISInitResponse *init) |
| 161 |
/* free an object made with makeWAISInitResponse */ |
| 162 |
{ |
| 163 |
s_free(init->ChunkMarker); |
| 164 |
s_free(init->HighlightMarker); |
| 165 |
s_free(init->DeHighlightMarker); |
| 166 |
s_free(init->NewlineCharacters); |
| 167 |
s_free(init); |
| 168 |
} |
| 169 |
|
| 170 |
/*----------------------------------------------------------------------*/ |
| 171 |
|
| 172 |
char *writeInitResponseInfo(InitResponseAPDU *init, |
| 173 |
char *buffer, |
| 174 |
long *len) |
| 175 |
/* write an init response object */ |
| 176 |
{ |
| 177 |
unsigned long header_len = userInfoTagSize(DT_UserInformationLength, |
| 178 |
DefWAISInitResponseSize); |
| 179 |
char *buf = buffer + header_len; |
| 180 |
WAISInitResponse *info = (WAISInitResponse *) init->UserInformationField; |
| 181 |
unsigned long size; |
| 182 |
|
| 183 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 184 |
|
| 185 |
buf = writeNum(info->ChunkCode, DT_ChunkCode, buf, len); |
| 186 |
buf = writeNum(info->ChunkIDLength, DT_ChunkIDLength, buf, len); |
| 187 |
buf = writeString(info->ChunkMarker, DT_ChunkMarker, buf, len); |
| 188 |
buf = writeString(info->HighlightMarker, DT_HighlightMarker, buf, len); |
| 189 |
buf = writeString(info->DeHighlightMarker, DT_DeHighlightMarker, buf, len); |
| 190 |
buf = writeString(info->NewlineCharacters, DT_NewlineCharacters, buf, len); |
| 191 |
|
| 192 |
/* now write the header and size */ |
| 193 |
size = buf - buffer; |
| 194 |
buf = writeUserInfoHeader(DT_UserInformationLength, |
| 195 |
size, |
| 196 |
header_len, |
| 197 |
buffer, |
| 198 |
len); |
| 199 |
|
| 200 |
return (buf); |
| 201 |
} |
| 202 |
|
| 203 |
/*----------------------------------------------------------------------*/ |
| 204 |
|
| 205 |
char *readInitResponseInfo(void **info, |
| 206 |
char *buffer) |
| 207 |
/* read an init response object */ |
| 208 |
{ |
| 209 |
char *buf = buffer; |
| 210 |
unsigned long size; |
| 211 |
unsigned long headerSize; |
| 212 |
long chunkCode, chunkIDLen; |
| 213 |
data_tag tag1; |
| 214 |
char *chunkMarker = NULL; |
| 215 |
char *highlightMarker = NULL; |
| 216 |
char *deHighlightMarker = NULL; |
| 217 |
char *newLineChars = NULL; |
| 218 |
|
| 219 |
chunkCode = chunkIDLen = UNUSED; |
| 220 |
|
| 221 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 222 |
headerSize = buf - buffer; |
| 223 |
|
| 224 |
while (buf < (buffer + size + headerSize)) { |
| 225 |
data_tag tag = peekTag(buf); |
| 226 |
|
| 227 |
switch (tag) { |
| 228 |
case DT_ChunkCode: |
| 229 |
buf = readNum(&chunkCode, buf); |
| 230 |
break; |
| 231 |
case DT_ChunkIDLength: |
| 232 |
buf = readNum(&chunkIDLen, buf); |
| 233 |
break; |
| 234 |
case DT_ChunkMarker: |
| 235 |
buf = readString(&chunkMarker, buf); |
| 236 |
break; |
| 237 |
case DT_HighlightMarker: |
| 238 |
buf = readString(&highlightMarker, buf); |
| 239 |
break; |
| 240 |
case DT_DeHighlightMarker: |
| 241 |
buf = readString(&deHighlightMarker, buf); |
| 242 |
break; |
| 243 |
case DT_NewlineCharacters: |
| 244 |
buf = readString(&newLineChars, buf); |
| 245 |
break; |
| 246 |
default: |
| 247 |
s_free(highlightMarker); |
| 248 |
s_free(deHighlightMarker); |
| 249 |
s_free(newLineChars); |
| 250 |
REPORT_READ_ERROR(buf); |
| 251 |
break; |
| 252 |
} |
| 253 |
} |
| 254 |
|
| 255 |
*info = (void *) makeWAISInitResponse(chunkCode, chunkIDLen, chunkMarker, |
| 256 |
highlightMarker, deHighlightMarker, |
| 257 |
newLineChars); |
| 258 |
return (buf); |
| 259 |
} |
| 260 |
|
| 261 |
/*----------------------------------------------------------------------*/ |
| 262 |
|
| 263 |
WAISSearch *makeWAISSearch(char *seedWords, |
| 264 |
DocObj **docs, |
| 265 |
char **textList, |
| 266 |
long dateFactor, |
| 267 |
char *beginDateRange, |
| 268 |
char *endDateRange, |
| 269 |
long maxDocsRetrieved) |
| 270 |
|
| 271 |
/* create a type 3 query object */ |
| 272 |
{ |
| 273 |
WAISSearch *query = S_MALLOC(WAISSearch); |
| 274 |
|
| 275 |
query->SeedWords = seedWords; /* not copied! */ |
| 276 |
query->Docs = docs; /* not copied! */ |
| 277 |
query->TextList = textList; /* not copied! */ |
| 278 |
query->DateFactor = dateFactor; |
| 279 |
query->BeginDateRange = beginDateRange; |
| 280 |
query->EndDateRange = endDateRange; |
| 281 |
query->MaxDocumentsRetrieved = maxDocsRetrieved; |
| 282 |
|
| 283 |
return (query); |
| 284 |
} |
| 285 |
|
| 286 |
/*----------------------------------------------------------------------*/ |
| 287 |
|
| 288 |
void freeWAISSearch(WAISSearch *query) |
| 289 |
|
| 290 |
/* destroy an object made with makeWAISSearch() */ |
| 291 |
{ |
| 292 |
void *ptr = NULL; |
| 293 |
long i; |
| 294 |
|
| 295 |
s_free(query->SeedWords); |
| 296 |
|
| 297 |
if (query->Docs != NULL) |
| 298 |
for (i = 0, ptr = (void *) query->Docs[i]; |
| 299 |
ptr != NULL; |
| 300 |
ptr = (void *) query->Docs[++i]) |
| 301 |
freeDocObj((DocObj *) ptr); |
| 302 |
s_free(query->Docs); |
| 303 |
|
| 304 |
if (query->TextList != NULL) /* XXX revisit when textlist is fully defined */ |
| 305 |
for (i = 0, ptr = (void *) query->TextList[i]; |
| 306 |
ptr != NULL; |
| 307 |
ptr = (void *) query->TextList[++i]) |
| 308 |
s_free(ptr); |
| 309 |
s_free(query->TextList); |
| 310 |
|
| 311 |
s_free(query->BeginDateRange); |
| 312 |
s_free(query->EndDateRange); |
| 313 |
s_free(query); |
| 314 |
} |
| 315 |
|
| 316 |
/*----------------------------------------------------------------------*/ |
| 317 |
|
| 318 |
DocObj *makeDocObjUsingWholeDocument(any *docID, |
| 319 |
char *type) |
| 320 |
|
| 321 |
/* construct a document object using byte chunks - only for use by |
| 322 |
servers */ |
| 323 |
{ |
| 324 |
DocObj *doc = S_MALLOC(DocObj); |
| 325 |
|
| 326 |
doc->DocumentID = docID; /* not copied! */ |
| 327 |
doc->Type = type; /* not copied! */ |
| 328 |
doc->ChunkCode = CT_document; |
| 329 |
return (doc); |
| 330 |
} |
| 331 |
|
| 332 |
/*----------------------------------------------------------------------*/ |
| 333 |
|
| 334 |
DocObj *makeDocObjUsingLines(any *docID, |
| 335 |
char *type, |
| 336 |
long start, |
| 337 |
long end) |
| 338 |
|
| 339 |
/* construct a document object using line chunks - only for use by |
| 340 |
servers */ |
| 341 |
{ |
| 342 |
DocObj *doc = S_MALLOC(DocObj); |
| 343 |
|
| 344 |
doc->ChunkCode = CT_line; |
| 345 |
doc->DocumentID = docID; /* not copied */ |
| 346 |
doc->Type = type; /* not copied! */ |
| 347 |
doc->ChunkStart.Pos = start; |
| 348 |
doc->ChunkEnd.Pos = end; |
| 349 |
return (doc); |
| 350 |
} |
| 351 |
|
| 352 |
/*----------------------------------------------------------------------*/ |
| 353 |
|
| 354 |
DocObj *makeDocObjUsingBytes(any *docID, |
| 355 |
char *type, |
| 356 |
long start, |
| 357 |
long end) |
| 358 |
|
| 359 |
/* construct a document object using byte chunks - only for use by |
| 360 |
servers */ |
| 361 |
{ |
| 362 |
DocObj *doc = S_MALLOC(DocObj); |
| 363 |
|
| 364 |
doc->ChunkCode = CT_byte; |
| 365 |
doc->DocumentID = docID; /* not copied */ |
| 366 |
doc->Type = type; /* not copied! */ |
| 367 |
doc->ChunkStart.Pos = start; |
| 368 |
doc->ChunkEnd.Pos = end; |
| 369 |
return (doc); |
| 370 |
} |
| 371 |
|
| 372 |
/*----------------------------------------------------------------------*/ |
| 373 |
|
| 374 |
DocObj *makeDocObjUsingParagraphs(any *docID, |
| 375 |
char *type, |
| 376 |
any *start, |
| 377 |
any *end) |
| 378 |
|
| 379 |
/* construct a document object using byte chunks - only for use by |
| 380 |
servers */ |
| 381 |
{ |
| 382 |
DocObj *doc = S_MALLOC(DocObj); |
| 383 |
|
| 384 |
doc->ChunkCode = CT_paragraph; |
| 385 |
doc->DocumentID = docID; /* not copied */ |
| 386 |
doc->Type = type; |
| 387 |
doc->ChunkStart.ID = start; |
| 388 |
doc->ChunkEnd.ID = end; |
| 389 |
return (doc); |
| 390 |
} |
| 391 |
|
| 392 |
/*----------------------------------------------------------------------*/ |
| 393 |
|
| 394 |
void freeDocObj(DocObj *doc) |
| 395 |
|
| 396 |
/* free a docObj */ |
| 397 |
{ |
| 398 |
freeAny(doc->DocumentID); |
| 399 |
s_free(doc->Type); |
| 400 |
if (doc->ChunkCode == CT_paragraph) { |
| 401 |
freeAny(doc->ChunkStart.ID); |
| 402 |
freeAny(doc->ChunkEnd.ID); |
| 403 |
} |
| 404 |
s_free(doc); |
| 405 |
} |
| 406 |
|
| 407 |
/*----------------------------------------------------------------------*/ |
| 408 |
|
| 409 |
static char *writeDocObj(DocObj *doc, |
| 410 |
char *buffer, |
| 411 |
long *len) |
| 412 |
|
| 413 |
/* write as little as we can about the doc obj */ |
| 414 |
{ |
| 415 |
char *buf = buffer; |
| 416 |
|
| 417 |
/* we always have to write the id, but its tag depends on if it's a chunk */ |
| 418 |
if (doc->ChunkCode == CT_document) |
| 419 |
buf = writeAny(doc->DocumentID, DT_DocumentID, buf, len); |
| 420 |
else |
| 421 |
buf = writeAny(doc->DocumentID, DT_DocumentIDChunk, buf, len); |
| 422 |
|
| 423 |
if (doc->Type != NULL) |
| 424 |
buf = writeString(doc->Type, DT_TYPE, buf, len); |
| 425 |
|
| 426 |
switch (doc->ChunkCode) { |
| 427 |
case CT_document: |
| 428 |
/* do nothing - there is no chunk data */ |
| 429 |
break; |
| 430 |
case CT_byte: |
| 431 |
case CT_line: |
| 432 |
buf = writeNum(doc->ChunkCode, DT_ChunkCode, buf, len); |
| 433 |
buf = writeNum(doc->ChunkStart.Pos, DT_ChunkStartID, buf, len); |
| 434 |
buf = writeNum(doc->ChunkEnd.Pos, DT_ChunkEndID, buf, len); |
| 435 |
break; |
| 436 |
case CT_paragraph: |
| 437 |
buf = writeNum(doc->ChunkCode, DT_ChunkCode, buf, len); |
| 438 |
buf = writeAny(doc->ChunkStart.ID, DT_ChunkStartID, buf, len); |
| 439 |
buf = writeAny(doc->ChunkEnd.ID, DT_ChunkEndID, buf, len); |
| 440 |
break; |
| 441 |
default: |
| 442 |
panic("Implementation error: unknown chuck type %ld", |
| 443 |
doc->ChunkCode); |
| 444 |
break; |
| 445 |
} |
| 446 |
|
| 447 |
return (buf); |
| 448 |
} |
| 449 |
|
| 450 |
/*----------------------------------------------------------------------*/ |
| 451 |
|
| 452 |
static char *readDocObj(DocObj **doc, |
| 453 |
char *buffer) |
| 454 |
|
| 455 |
/* read whatever we have about the new document */ |
| 456 |
{ |
| 457 |
char *buf = buffer; |
| 458 |
data_tag tag; |
| 459 |
|
| 460 |
*doc = S_MALLOC(DocObj); |
| 461 |
|
| 462 |
tag = peekTag(buf); |
| 463 |
buf = readAny(&((*doc)->DocumentID), buf); |
| 464 |
|
| 465 |
if (tag == DT_DocumentID) { |
| 466 |
(*doc)->ChunkCode = CT_document; |
| 467 |
tag = peekTag(buf); |
| 468 |
if (tag == DT_TYPE) /* XXX depends on DT_TYPE != what comes next */ |
| 469 |
buf = readString(&((*doc)->Type), buf); |
| 470 |
/* ChunkStart and ChunkEnd are undefined */ |
| 471 |
} else if (tag == DT_DocumentIDChunk) { |
| 472 |
boolean readParagraphs = false; /* for cleanup */ |
| 473 |
|
| 474 |
tag = peekTag(buf); |
| 475 |
if (tag == DT_TYPE) /* XXX depends on DT_TYPE != CT_FOO */ |
| 476 |
buf = readString(&((*doc)->Type), buf); |
| 477 |
buf = readNum(&((*doc)->ChunkCode), buf); |
| 478 |
switch ((*doc)->ChunkCode) { |
| 479 |
case CT_byte: |
| 480 |
case CT_line: |
| 481 |
buf = readNum(&((*doc)->ChunkStart.Pos), buf); |
| 482 |
buf = readNum(&((*doc)->ChunkEnd.Pos), buf); |
| 483 |
break; |
| 484 |
case CT_paragraph: |
| 485 |
buf = readAny(&((*doc)->ChunkStart.ID), buf); |
| 486 |
buf = readAny(&((*doc)->ChunkEnd.ID), buf); |
| 487 |
break; |
| 488 |
default: |
| 489 |
freeAny((*doc)->DocumentID); |
| 490 |
if (readParagraphs) { |
| 491 |
freeAny((*doc)->ChunkStart.ID); |
| 492 |
freeAny((*doc)->ChunkEnd.ID); |
| 493 |
} |
| 494 |
s_free(doc); |
| 495 |
REPORT_READ_ERROR(buf); |
| 496 |
break; |
| 497 |
} |
| 498 |
} else { |
| 499 |
freeAny((*doc)->DocumentID); |
| 500 |
s_free(*doc); |
| 501 |
REPORT_READ_ERROR(buf); |
| 502 |
} |
| 503 |
return (buf); |
| 504 |
} |
| 505 |
|
| 506 |
/*----------------------------------------------------------------------*/ |
| 507 |
|
| 508 |
char *writeSearchInfo(SearchAPDU *query, |
| 509 |
char *buffer, |
| 510 |
long *len) |
| 511 |
|
| 512 |
/* write out a WAIS query (type 1 or 3) */ |
| 513 |
{ |
| 514 |
if (strcmp(query->QueryType, QT_TextRetrievalQuery) == 0) { |
| 515 |
return (writeAny((any *) query->Query, DT_Query, buffer, len)); |
| 516 |
} else { |
| 517 |
unsigned long header_len = userInfoTagSize(DT_UserInformationLength, |
| 518 |
DefWAISSearchSize); |
| 519 |
char *buf = buffer + header_len; |
| 520 |
WAISSearch *info = (WAISSearch *) query->Query; |
| 521 |
unsigned long size; |
| 522 |
long i; |
| 523 |
|
| 524 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 525 |
|
| 526 |
buf = writeString(info->SeedWords, DT_SeedWords, buf, len); |
| 527 |
|
| 528 |
if (info->Docs != NULL) { |
| 529 |
for (i = 0; info->Docs[i] != NULL; i++) { |
| 530 |
buf = writeDocObj(info->Docs[i], buf, len); |
| 531 |
} |
| 532 |
} |
| 533 |
|
| 534 |
/* XXX text list */ |
| 535 |
|
| 536 |
buf = writeNum(info->DateFactor, |
| 537 |
DT_DateFactor, |
| 538 |
buf, |
| 539 |
len); |
| 540 |
buf = writeString(info->BeginDateRange, |
| 541 |
DT_BeginDateRange, |
| 542 |
buf, |
| 543 |
len); |
| 544 |
buf = writeString(info->EndDateRange, |
| 545 |
DT_EndDateRange, |
| 546 |
buf, |
| 547 |
len); |
| 548 |
buf = writeNum(info->MaxDocumentsRetrieved, |
| 549 |
DT_MaxDocumentsRetrieved, |
| 550 |
buf, |
| 551 |
len); |
| 552 |
|
| 553 |
/* now write the header and size */ |
| 554 |
size = buf - buffer; |
| 555 |
buf = writeUserInfoHeader(DT_UserInformationLength, |
| 556 |
size, |
| 557 |
header_len, |
| 558 |
buffer, |
| 559 |
len); |
| 560 |
|
| 561 |
return (buf); |
| 562 |
} |
| 563 |
} |
| 564 |
|
| 565 |
/*----------------------------------------------------------------------*/ |
| 566 |
|
| 567 |
char *readSearchInfo(void **info, |
| 568 |
char *buffer) |
| 569 |
|
| 570 |
/* read a WAIS query (type 1 or 3) */ |
| 571 |
{ |
| 572 |
data_tag type = peekTag(buffer); |
| 573 |
|
| 574 |
if (type == DT_Query) /* this is a type 1 query */ |
| 575 |
{ |
| 576 |
char *buf = buffer; |
| 577 |
any *query = NULL; |
| 578 |
|
| 579 |
buf = readAny(&query, buf); |
| 580 |
*info = (void *) query; |
| 581 |
return (buf); |
| 582 |
} else { /* a type 3 query */ |
| 583 |
char *buf = buffer; |
| 584 |
unsigned long size; |
| 585 |
unsigned long headerSize; |
| 586 |
data_tag tag1; |
| 587 |
char *seedWords = NULL; |
| 588 |
char *beginDateRange = NULL; |
| 589 |
char *endDateRange = NULL; |
| 590 |
long dateFactor, maxDocsRetrieved; |
| 591 |
char **textList = NULL; |
| 592 |
DocObj **docIDs = NULL; |
| 593 |
DocObj *doc = NULL; |
| 594 |
long docs = 0; |
| 595 |
long i; |
| 596 |
void *ptr = NULL; |
| 597 |
|
| 598 |
dateFactor = maxDocsRetrieved = UNUSED; |
| 599 |
|
| 600 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 601 |
headerSize = buf - buffer; |
| 602 |
|
| 603 |
while (buf < (buffer + size + headerSize)) { |
| 604 |
data_tag tag = peekTag(buf); |
| 605 |
|
| 606 |
switch (tag) { |
| 607 |
case DT_SeedWords: |
| 608 |
buf = readString(&seedWords, buf); |
| 609 |
break; |
| 610 |
case DT_DocumentID: |
| 611 |
case DT_DocumentIDChunk: |
| 612 |
if (docIDs == NULL) /* create a new doc list */ |
| 613 |
{ |
| 614 |
docIDs = S_MALLOC2(DocObj *); |
| 615 |
} else { /* grow the doc list */ |
| 616 |
docIDs = S_REALLOC2(DocObj *, docIDs, docs); |
| 617 |
} |
| 618 |
buf = readDocObj(&doc, buf); |
| 619 |
if (buf == NULL) { |
| 620 |
s_free(seedWords); |
| 621 |
s_free(beginDateRange); |
| 622 |
s_free(endDateRange); |
| 623 |
if (docIDs != NULL) |
| 624 |
for (i = 0, ptr = (void *) docIDs[i]; |
| 625 |
ptr != NULL; |
| 626 |
ptr = (void *) docIDs[++i]) |
| 627 |
freeDocObj((DocObj *) ptr); |
| 628 |
s_free(docIDs); |
| 629 |
/* XXX should also free textlist when it is fully defined */ |
| 630 |
} |
| 631 |
RETURN_ON_NULL(buf); |
| 632 |
docIDs[docs++] = doc; /* put it in the list */ |
| 633 |
docIDs[docs] = NULL; |
| 634 |
break; |
| 635 |
case DT_TextList: |
| 636 |
/* XXX */ |
| 637 |
break; |
| 638 |
case DT_DateFactor: |
| 639 |
buf = readNum(&dateFactor, buf); |
| 640 |
break; |
| 641 |
case DT_BeginDateRange: |
| 642 |
buf = readString(&beginDateRange, buf); |
| 643 |
break; |
| 644 |
case DT_EndDateRange: |
| 645 |
buf = readString(&endDateRange, buf); |
| 646 |
break; |
| 647 |
case DT_MaxDocumentsRetrieved: |
| 648 |
buf = readNum(&maxDocsRetrieved, buf); |
| 649 |
break; |
| 650 |
default: |
| 651 |
s_free(seedWords); |
| 652 |
s_free(beginDateRange); |
| 653 |
s_free(endDateRange); |
| 654 |
if (docIDs != NULL) |
| 655 |
for (i = 0, ptr = (void *) docIDs[i]; |
| 656 |
ptr != NULL; |
| 657 |
ptr = (void *) docIDs[++i]) |
| 658 |
freeDocObj((DocObj *) ptr); |
| 659 |
s_free(docIDs); |
| 660 |
/* XXX should also free textlist when it is fully defined */ |
| 661 |
REPORT_READ_ERROR(buf); |
| 662 |
break; |
| 663 |
} |
| 664 |
} |
| 665 |
|
| 666 |
*info = (void *) makeWAISSearch(seedWords, docIDs, textList, |
| 667 |
dateFactor, beginDateRange, endDateRange, |
| 668 |
maxDocsRetrieved); |
| 669 |
return (buf); |
| 670 |
} |
| 671 |
} |
| 672 |
|
| 673 |
/*----------------------------------------------------------------------*/ |
| 674 |
|
| 675 |
WAISDocumentHeader *makeWAISDocumentHeader(any *docID, |
| 676 |
long versionNumber, |
| 677 |
long score, |
| 678 |
long bestMatch, |
| 679 |
long docLen, |
| 680 |
long lines, |
| 681 |
char **types, |
| 682 |
char *source, |
| 683 |
char *date, |
| 684 |
char *headline, |
| 685 |
char *originCity) |
| 686 |
|
| 687 |
/* construct a standard document header, note that no fields are copied! |
| 688 |
if the application needs to save these fields, it should copy them, |
| 689 |
or set the field in this object to NULL before freeing it. |
| 690 |
*/ |
| 691 |
{ |
| 692 |
WAISDocumentHeader *header = S_MALLOC(WAISDocumentHeader); |
| 693 |
|
| 694 |
header->DocumentID = docID; |
| 695 |
header->VersionNumber = versionNumber; |
| 696 |
header->Score = score; |
| 697 |
header->BestMatch = bestMatch; |
| 698 |
header->DocumentLength = docLen; |
| 699 |
header->Lines = lines; |
| 700 |
header->Types = types; |
| 701 |
header->Source = source; |
| 702 |
header->Date = date; |
| 703 |
header->Headline = headline; |
| 704 |
header->OriginCity = originCity; |
| 705 |
|
| 706 |
return (header); |
| 707 |
} |
| 708 |
|
| 709 |
/*----------------------------------------------------------------------*/ |
| 710 |
|
| 711 |
void freeWAISDocumentHeader(WAISDocumentHeader *header) |
| 712 |
{ |
| 713 |
freeAny(header->DocumentID); |
| 714 |
doList((void **) header->Types, fs_free); /* can't use the macro here ! */ |
| 715 |
s_free(header->Types); |
| 716 |
s_free(header->Source); |
| 717 |
s_free(header->Date); |
| 718 |
s_free(header->Headline); |
| 719 |
s_free(header->OriginCity); |
| 720 |
s_free(header); |
| 721 |
} |
| 722 |
|
| 723 |
/*----------------------------------------------------------------------*/ |
| 724 |
|
| 725 |
char *writeWAISDocumentHeader(WAISDocumentHeader *header, char *buffer, |
| 726 |
long *len) |
| 727 |
{ |
| 728 |
unsigned long header_len = userInfoTagSize(DT_DocumentHeaderGroup, |
| 729 |
DefWAISDocHeaderSize); |
| 730 |
char *buf = buffer + header_len; |
| 731 |
unsigned long size1; |
| 732 |
|
| 733 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 734 |
|
| 735 |
buf = writeAny(header->DocumentID, DT_DocumentID, buf, len); |
| 736 |
buf = writeNum(header->VersionNumber, DT_VersionNumber, buf, len); |
| 737 |
buf = writeNum(header->Score, DT_Score, buf, len); |
| 738 |
buf = writeNum(header->BestMatch, DT_BestMatch, buf, len); |
| 739 |
buf = writeNum(header->DocumentLength, DT_DocumentLength, buf, len); |
| 740 |
buf = writeNum(header->Lines, DT_Lines, buf, len); |
| 741 |
if (header->Types != NULL) { |
| 742 |
long size; |
| 743 |
char *ptr = NULL; |
| 744 |
long i; |
| 745 |
|
| 746 |
buf = writeTag(DT_TYPE_BLOCK, buf, len); |
| 747 |
for (i = 0, size = 0, ptr = header->Types[i]; |
| 748 |
ptr != NULL; |
| 749 |
ptr = header->Types[++i]) { |
| 750 |
long typeSize = strlen(ptr); |
| 751 |
|
| 752 |
size += writtenTagSize(DT_TYPE); |
| 753 |
size += writtenCompressedIntSize(typeSize); |
| 754 |
size += typeSize; |
| 755 |
} |
| 756 |
buf = writeCompressedInteger((unsigned long) size, buf, len); |
| 757 |
for (i = 0, ptr = header->Types[i]; ptr != NULL; ptr = header->Types[++i]) |
| 758 |
buf = writeString(ptr, DT_TYPE, buf, len); |
| 759 |
} |
| 760 |
buf = writeString(header->Source, DT_Source, buf, len); |
| 761 |
buf = writeString(header->Date, DT_Date, buf, len); |
| 762 |
buf = writeString(header->Headline, DT_Headline, buf, len); |
| 763 |
buf = writeString(header->OriginCity, DT_OriginCity, buf, len); |
| 764 |
|
| 765 |
/* now write the header and size */ |
| 766 |
size1 = buf - buffer; |
| 767 |
buf = writeUserInfoHeader(DT_DocumentHeaderGroup, |
| 768 |
size1, |
| 769 |
header_len, |
| 770 |
buffer, |
| 771 |
len); |
| 772 |
|
| 773 |
return (buf); |
| 774 |
} |
| 775 |
|
| 776 |
/*----------------------------------------------------------------------*/ |
| 777 |
|
| 778 |
char *readWAISDocumentHeader(WAISDocumentHeader **header, char *buffer) |
| 779 |
{ |
| 780 |
char *buf = buffer; |
| 781 |
unsigned long size1; |
| 782 |
unsigned long headerSize; |
| 783 |
data_tag tag1; |
| 784 |
any *docID = NULL; |
| 785 |
long versionNumber, score, bestMatch, docLength, lines; |
| 786 |
char **types = NULL; |
| 787 |
char *source = NULL; |
| 788 |
char *date = NULL; |
| 789 |
char *headline = NULL; |
| 790 |
char *originCity = NULL; |
| 791 |
|
| 792 |
versionNumber = score = bestMatch = docLength = lines = UNUSED; |
| 793 |
|
| 794 |
buf = readUserInfoHeader(&tag1, &size1, buf); |
| 795 |
headerSize = buf - buffer; |
| 796 |
|
| 797 |
while (buf < (buffer + size1 + headerSize)) { |
| 798 |
data_tag tag = peekTag(buf); |
| 799 |
|
| 800 |
switch (tag) { |
| 801 |
case DT_DocumentID: |
| 802 |
buf = readAny(&docID, buf); |
| 803 |
break; |
| 804 |
case DT_VersionNumber: |
| 805 |
buf = readNum(&versionNumber, buf); |
| 806 |
break; |
| 807 |
case DT_Score: |
| 808 |
buf = readNum(&score, buf); |
| 809 |
break; |
| 810 |
case DT_BestMatch: |
| 811 |
buf = readNum(&bestMatch, buf); |
| 812 |
break; |
| 813 |
case DT_DocumentLength: |
| 814 |
buf = readNum(&docLength, buf); |
| 815 |
break; |
| 816 |
case DT_Lines: |
| 817 |
buf = readNum(&lines, buf); |
| 818 |
break; |
| 819 |
case DT_TYPE_BLOCK: |
| 820 |
{ |
| 821 |
unsigned long size = -1; |
| 822 |
long numTypes = 0; |
| 823 |
|
| 824 |
buf = readTag(&tag, buf); |
| 825 |
buf = readCompressedInteger(&size, buf); |
| 826 |
while (size > 0) { |
| 827 |
char *type = NULL; |
| 828 |
char *originalBuf = buf; |
| 829 |
|
| 830 |
buf = readString(&type, buf); |
| 831 |
types = S_REALLOC2(char *, types, numTypes); |
| 832 |
|
| 833 |
types[numTypes++] = type; |
| 834 |
types[numTypes] = NULL; |
| 835 |
size -= (buf - originalBuf); |
| 836 |
} |
| 837 |
} |
| 838 |
/* FALLTHRU */ |
| 839 |
case DT_Source: |
| 840 |
buf = readString(&source, buf); |
| 841 |
break; |
| 842 |
case DT_Date: |
| 843 |
buf = readString(&date, buf); |
| 844 |
break; |
| 845 |
case DT_Headline: |
| 846 |
buf = readString(&headline, buf); |
| 847 |
break; |
| 848 |
case DT_OriginCity: |
| 849 |
buf = readString(&originCity, buf); |
| 850 |
break; |
| 851 |
default: |
| 852 |
freeAny(docID); |
| 853 |
s_free(source); |
| 854 |
s_free(date); |
| 855 |
s_free(headline); |
| 856 |
s_free(originCity); |
| 857 |
REPORT_READ_ERROR(buf); |
| 858 |
break; |
| 859 |
} |
| 860 |
} |
| 861 |
|
| 862 |
*header = makeWAISDocumentHeader(docID, versionNumber, score, bestMatch, |
| 863 |
docLength, lines, types, source, date, headline, |
| 864 |
originCity); |
| 865 |
return (buf); |
| 866 |
} |
| 867 |
|
| 868 |
/*----------------------------------------------------------------------*/ |
| 869 |
|
| 870 |
WAISDocumentShortHeader *makeWAISDocumentShortHeader(any *docID, |
| 871 |
long versionNumber, |
| 872 |
long score, |
| 873 |
long bestMatch, |
| 874 |
long docLen, |
| 875 |
long lines) |
| 876 |
/* construct a short document header, note that no fields are copied! |
| 877 |
if the application needs to save these fields, it should copy them, |
| 878 |
or set the field in this object to NULL before freeing it. |
| 879 |
*/ |
| 880 |
{ |
| 881 |
WAISDocumentShortHeader *header = S_MALLOC(WAISDocumentShortHeader); |
| 882 |
|
| 883 |
header->DocumentID = docID; |
| 884 |
header->VersionNumber = versionNumber; |
| 885 |
header->Score = score; |
| 886 |
header->BestMatch = bestMatch; |
| 887 |
header->DocumentLength = docLen; |
| 888 |
header->Lines = lines; |
| 889 |
|
| 890 |
return (header); |
| 891 |
} |
| 892 |
|
| 893 |
/*----------------------------------------------------------------------*/ |
| 894 |
|
| 895 |
void freeWAISDocumentShortHeader(WAISDocumentShortHeader *header) |
| 896 |
{ |
| 897 |
freeAny(header->DocumentID); |
| 898 |
s_free(header); |
| 899 |
} |
| 900 |
|
| 901 |
/*----------------------------------------------------------------------*/ |
| 902 |
|
| 903 |
char *writeWAISDocumentShortHeader(WAISDocumentShortHeader *header, char *buffer, |
| 904 |
long *len) |
| 905 |
{ |
| 906 |
unsigned long header_len = userInfoTagSize(DT_DocumentShortHeaderGroup, |
| 907 |
DefWAISShortHeaderSize); |
| 908 |
char *buf = buffer + header_len; |
| 909 |
unsigned long size; |
| 910 |
|
| 911 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 912 |
|
| 913 |
buf = writeAny(header->DocumentID, DT_DocumentID, buf, len); |
| 914 |
buf = writeNum(header->VersionNumber, DT_VersionNumber, buf, len); |
| 915 |
buf = writeNum(header->Score, DT_Score, buf, len); |
| 916 |
buf = writeNum(header->BestMatch, DT_BestMatch, buf, len); |
| 917 |
buf = writeNum(header->DocumentLength, DT_DocumentLength, buf, len); |
| 918 |
buf = writeNum(header->Lines, DT_Lines, buf, len); |
| 919 |
|
| 920 |
/* now write the header and size */ |
| 921 |
size = buf - buffer; |
| 922 |
buf = writeUserInfoHeader(DT_DocumentShortHeaderGroup, |
| 923 |
size, |
| 924 |
header_len, |
| 925 |
buffer, |
| 926 |
len); |
| 927 |
|
| 928 |
return (buf); |
| 929 |
} |
| 930 |
|
| 931 |
/*----------------------------------------------------------------------*/ |
| 932 |
|
| 933 |
char *readWAISDocumentShortHeader(WAISDocumentShortHeader **header, char *buffer) |
| 934 |
{ |
| 935 |
char *buf = buffer; |
| 936 |
unsigned long size; |
| 937 |
unsigned long headerSize; |
| 938 |
data_tag tag1; |
| 939 |
any *docID = NULL; |
| 940 |
long versionNumber, score, bestMatch, docLength, lines; |
| 941 |
|
| 942 |
versionNumber = score = bestMatch = docLength = lines = UNUSED; |
| 943 |
|
| 944 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 945 |
headerSize = buf - buffer; |
| 946 |
|
| 947 |
while (buf < (buffer + size + headerSize)) { |
| 948 |
data_tag tag = peekTag(buf); |
| 949 |
|
| 950 |
switch (tag) { |
| 951 |
case DT_DocumentID: |
| 952 |
buf = readAny(&docID, buf); |
| 953 |
break; |
| 954 |
case DT_VersionNumber: |
| 955 |
buf = readNum(&versionNumber, buf); |
| 956 |
break; |
| 957 |
case DT_Score: |
| 958 |
buf = readNum(&score, buf); |
| 959 |
break; |
| 960 |
case DT_BestMatch: |
| 961 |
buf = readNum(&bestMatch, buf); |
| 962 |
break; |
| 963 |
case DT_DocumentLength: |
| 964 |
buf = readNum(&docLength, buf); |
| 965 |
break; |
| 966 |
case DT_Lines: |
| 967 |
buf = readNum(&lines, buf); |
| 968 |
break; |
| 969 |
default: |
| 970 |
freeAny(docID); |
| 971 |
REPORT_READ_ERROR(buf); |
| 972 |
break; |
| 973 |
} |
| 974 |
} |
| 975 |
|
| 976 |
*header = makeWAISDocumentShortHeader(docID, versionNumber, score, bestMatch, |
| 977 |
docLength, lines); |
| 978 |
return (buf); |
| 979 |
} |
| 980 |
|
| 981 |
/*----------------------------------------------------------------------*/ |
| 982 |
|
| 983 |
WAISDocumentLongHeader *makeWAISDocumentLongHeader(any *docID, |
| 984 |
long versionNumber, |
| 985 |
long score, |
| 986 |
long bestMatch, |
| 987 |
long docLen, |
| 988 |
long lines, |
| 989 |
char **types, |
| 990 |
char *source, |
| 991 |
char *date, |
| 992 |
char *headline, |
| 993 |
char *originCity, |
| 994 |
char *stockCodes, |
| 995 |
char *companyCodes, |
| 996 |
char *industryCodes) |
| 997 |
/* construct a long document header, note that no fields are copied! |
| 998 |
if the application needs to save these fields, it should copy them, |
| 999 |
or set the field in this object to NULL before freeing it. |
| 1000 |
*/ |
| 1001 |
{ |
| 1002 |
WAISDocumentLongHeader *header = S_MALLOC(WAISDocumentLongHeader); |
| 1003 |
|
| 1004 |
header->DocumentID = docID; |
| 1005 |
header->VersionNumber = versionNumber; |
| 1006 |
header->Score = score; |
| 1007 |
header->BestMatch = bestMatch; |
| 1008 |
header->DocumentLength = docLen; |
| 1009 |
header->Lines = lines; |
| 1010 |
header->Types = types; |
| 1011 |
header->Source = source; |
| 1012 |
header->Date = date; |
| 1013 |
header->Headline = headline; |
| 1014 |
header->OriginCity = originCity; |
| 1015 |
header->StockCodes = stockCodes; |
| 1016 |
header->CompanyCodes = companyCodes; |
| 1017 |
header->IndustryCodes = industryCodes; |
| 1018 |
|
| 1019 |
return (header); |
| 1020 |
} |
| 1021 |
|
| 1022 |
/*----------------------------------------------------------------------*/ |
| 1023 |
|
| 1024 |
void freeWAISDocumentLongHeader(WAISDocumentLongHeader *header) |
| 1025 |
{ |
| 1026 |
freeAny(header->DocumentID); |
| 1027 |
doList((void **) header->Types, fs_free); /* can't use the macro here! */ |
| 1028 |
s_free(header->Source); |
| 1029 |
s_free(header->Date); |
| 1030 |
s_free(header->Headline); |
| 1031 |
s_free(header->OriginCity); |
| 1032 |
s_free(header->StockCodes); |
| 1033 |
s_free(header->CompanyCodes); |
| 1034 |
s_free(header->IndustryCodes); |
| 1035 |
s_free(header); |
| 1036 |
} |
| 1037 |
|
| 1038 |
/*----------------------------------------------------------------------*/ |
| 1039 |
|
| 1040 |
char *writeWAISDocumentLongHeader(WAISDocumentLongHeader *header, char *buffer, |
| 1041 |
long *len) |
| 1042 |
{ |
| 1043 |
unsigned long header_len = userInfoTagSize(DT_DocumentLongHeaderGroup, |
| 1044 |
DefWAISLongHeaderSize); |
| 1045 |
char *buf = buffer + header_len; |
| 1046 |
unsigned long size1; |
| 1047 |
|
| 1048 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 1049 |
|
| 1050 |
buf = writeAny(header->DocumentID, DT_DocumentID, buf, len); |
| 1051 |
buf = writeNum(header->VersionNumber, DT_VersionNumber, buf, len); |
| 1052 |
buf = writeNum(header->Score, DT_Score, buf, len); |
| 1053 |
buf = writeNum(header->BestMatch, DT_BestMatch, buf, len); |
| 1054 |
buf = writeNum(header->DocumentLength, DT_DocumentLength, buf, len); |
| 1055 |
buf = writeNum(header->Lines, DT_Lines, buf, len); |
| 1056 |
if (header->Types != NULL) { |
| 1057 |
long size; |
| 1058 |
char *ptr = NULL; |
| 1059 |
long i; |
| 1060 |
|
| 1061 |
buf = writeTag(DT_TYPE_BLOCK, buf, len); |
| 1062 |
for (i = 0, size = 0, ptr = header->Types[i]; |
| 1063 |
ptr != NULL; |
| 1064 |
ptr = header->Types[++i]) { |
| 1065 |
long typeSize = strlen(ptr); |
| 1066 |
|
| 1067 |
size += writtenTagSize(DT_TYPE); |
| 1068 |
size += writtenCompressedIntSize(typeSize); |
| 1069 |
size += typeSize; |
| 1070 |
} |
| 1071 |
buf = writeCompressedInteger((unsigned long) size, buf, len); |
| 1072 |
for (i = 0, ptr = header->Types[i]; ptr != NULL; ptr = header->Types[++i]) |
| 1073 |
buf = writeString(ptr, DT_TYPE, buf, len); |
| 1074 |
} |
| 1075 |
buf = writeString(header->Source, DT_Source, buf, len); |
| 1076 |
buf = writeString(header->Date, DT_Date, buf, len); |
| 1077 |
buf = writeString(header->Headline, DT_Headline, buf, len); |
| 1078 |
buf = writeString(header->OriginCity, DT_OriginCity, buf, len); |
| 1079 |
buf = writeString(header->StockCodes, DT_StockCodes, buf, len); |
| 1080 |
buf = writeString(header->CompanyCodes, DT_CompanyCodes, buf, len); |
| 1081 |
buf = writeString(header->IndustryCodes, DT_IndustryCodes, buf, len); |
| 1082 |
|
| 1083 |
/* now write the header and size */ |
| 1084 |
size1 = buf - buffer; |
| 1085 |
buf = writeUserInfoHeader(DT_DocumentLongHeaderGroup, |
| 1086 |
size1, |
| 1087 |
header_len, |
| 1088 |
buffer, |
| 1089 |
len); |
| 1090 |
|
| 1091 |
return (buf); |
| 1092 |
} |
| 1093 |
|
| 1094 |
/*----------------------------------------------------------------------*/ |
| 1095 |
|
| 1096 |
char *readWAISDocumentLongHeader(WAISDocumentLongHeader **header, char *buffer) |
| 1097 |
{ |
| 1098 |
char *buf = buffer; |
| 1099 |
unsigned long size1; |
| 1100 |
unsigned long headerSize; |
| 1101 |
data_tag tag1; |
| 1102 |
any *docID; |
| 1103 |
long versionNumber, score, bestMatch, docLength, lines; |
| 1104 |
char **types; |
| 1105 |
char *source, *date, *headline, *originCity, *stockCodes, *companyCodes, *industryCodes; |
| 1106 |
|
| 1107 |
docID = NULL; |
| 1108 |
versionNumber = |
| 1109 |
score = |
| 1110 |
bestMatch = |
| 1111 |
docLength = |
| 1112 |
lines = UNUSED; |
| 1113 |
types = NULL; |
| 1114 |
source = |
| 1115 |
date = |
| 1116 |
headline = |
| 1117 |
originCity = |
| 1118 |
stockCodes = |
| 1119 |
companyCodes = |
| 1120 |
industryCodes = NULL; |
| 1121 |
|
| 1122 |
buf = readUserInfoHeader(&tag1, &size1, buf); |
| 1123 |
headerSize = buf - buffer; |
| 1124 |
|
| 1125 |
while (buf < (buffer + size1 + headerSize)) { |
| 1126 |
data_tag tag = peekTag(buf); |
| 1127 |
|
| 1128 |
switch (tag) { |
| 1129 |
case DT_DocumentID: |
| 1130 |
buf = readAny(&docID, buf); |
| 1131 |
break; |
| 1132 |
case DT_VersionNumber: |
| 1133 |
buf = readNum(&versionNumber, buf); |
| 1134 |
break; |
| 1135 |
case DT_Score: |
| 1136 |
buf = readNum(&score, buf); |
| 1137 |
break; |
| 1138 |
case DT_BestMatch: |
| 1139 |
buf = readNum(&bestMatch, buf); |
| 1140 |
break; |
| 1141 |
case DT_DocumentLength: |
| 1142 |
buf = readNum(&docLength, buf); |
| 1143 |
break; |
| 1144 |
case DT_Lines: |
| 1145 |
buf = readNum(&lines, buf); |
| 1146 |
break; |
| 1147 |
case DT_TYPE_BLOCK: |
| 1148 |
{ |
| 1149 |
unsigned long size = -1; |
| 1150 |
long numTypes = 0; |
| 1151 |
|
| 1152 |
buf = readTag(&tag, buf); |
| 1153 |
readCompressedInteger(&size, buf); |
| 1154 |
while (size > 0) { |
| 1155 |
char *type = NULL; |
| 1156 |
char *originalBuf = buf; |
| 1157 |
|
| 1158 |
buf = readString(&type, buf); |
| 1159 |
types = S_REALLOC2(char *, types, numTypes); |
| 1160 |
|
| 1161 |
types[numTypes++] = type; |
| 1162 |
types[numTypes] = NULL; |
| 1163 |
size -= (buf - originalBuf); |
| 1164 |
} |
| 1165 |
} |
| 1166 |
/* FALLTHRU */ |
| 1167 |
case DT_Source: |
| 1168 |
buf = readString(&source, buf); |
| 1169 |
break; |
| 1170 |
case DT_Date: |
| 1171 |
buf = readString(&date, buf); |
| 1172 |
break; |
| 1173 |
case DT_Headline: |
| 1174 |
buf = readString(&headline, buf); |
| 1175 |
break; |
| 1176 |
case DT_OriginCity: |
| 1177 |
buf = readString(&originCity, buf); |
| 1178 |
break; |
| 1179 |
case DT_StockCodes: |
| 1180 |
buf = readString(&stockCodes, buf); |
| 1181 |
break; |
| 1182 |
case DT_CompanyCodes: |
| 1183 |
buf = readString(&companyCodes, buf); |
| 1184 |
break; |
| 1185 |
case DT_IndustryCodes: |
| 1186 |
buf = readString(&industryCodes, buf); |
| 1187 |
break; |
| 1188 |
default: |
| 1189 |
freeAny(docID); |
| 1190 |
s_free(source); |
| 1191 |
s_free(date); |
| 1192 |
s_free(headline); |
| 1193 |
s_free(originCity); |
| 1194 |
s_free(stockCodes); |
| 1195 |
s_free(companyCodes); |
| 1196 |
s_free(industryCodes); |
| 1197 |
REPORT_READ_ERROR(buf); |
| 1198 |
break; |
| 1199 |
} |
| 1200 |
} |
| 1201 |
|
| 1202 |
*header = makeWAISDocumentLongHeader(docID, |
| 1203 |
versionNumber, |
| 1204 |
score, |
| 1205 |
bestMatch, |
| 1206 |
docLength, |
| 1207 |
lines, |
| 1208 |
types, |
| 1209 |
source, |
| 1210 |
date, |
| 1211 |
headline, |
| 1212 |
originCity, |
| 1213 |
stockCodes, |
| 1214 |
companyCodes, |
| 1215 |
industryCodes); |
| 1216 |
return (buf); |
| 1217 |
} |
| 1218 |
|
| 1219 |
/*----------------------------------------------------------------------*/ |
| 1220 |
|
| 1221 |
WAISSearchResponse * |
| 1222 |
makeWAISSearchResponse( |
| 1223 |
char *seedWordsUsed, |
| 1224 |
WAISDocumentHeader **docHeaders, |
| 1225 |
WAISDocumentShortHeader **shortHeaders, |
| 1226 |
WAISDocumentLongHeader **longHeaders, |
| 1227 |
WAISDocumentText **text, |
| 1228 |
WAISDocumentHeadlines **headlines, |
| 1229 |
WAISDocumentCodes **codes, |
| 1230 |
diagnosticRecord ** diagnostics) |
| 1231 |
{ |
| 1232 |
WAISSearchResponse *response = S_MALLOC(WAISSearchResponse); |
| 1233 |
|
| 1234 |
response->SeedWordsUsed = seedWordsUsed; |
| 1235 |
response->DocHeaders = docHeaders; |
| 1236 |
response->ShortHeaders = shortHeaders; |
| 1237 |
response->LongHeaders = longHeaders; |
| 1238 |
response->Text = text; |
| 1239 |
response->Headlines = headlines; |
| 1240 |
response->Codes = codes; |
| 1241 |
response->Diagnostics = diagnostics; |
| 1242 |
|
| 1243 |
return (response); |
| 1244 |
} |
| 1245 |
|
| 1246 |
/*----------------------------------------------------------------------*/ |
| 1247 |
|
| 1248 |
void freeWAISSearchResponse(WAISSearchResponse * response) |
| 1249 |
{ |
| 1250 |
void *ptr = NULL; |
| 1251 |
long i; |
| 1252 |
|
| 1253 |
s_free(response->SeedWordsUsed); |
| 1254 |
|
| 1255 |
if (response->DocHeaders != NULL) |
| 1256 |
for (i = 0, ptr = (void *) response->DocHeaders[i]; |
| 1257 |
ptr != NULL; |
| 1258 |
ptr = (void *) response->DocHeaders[++i]) |
| 1259 |
freeWAISDocumentHeader((WAISDocumentHeader *) ptr); |
| 1260 |
s_free(response->DocHeaders); |
| 1261 |
|
| 1262 |
if (response->ShortHeaders != NULL) |
| 1263 |
for (i = 0, ptr = (void *) response->ShortHeaders[i]; |
| 1264 |
ptr != NULL; |
| 1265 |
ptr = (void *) response->ShortHeaders[++i]) |
| 1266 |
freeWAISDocumentShortHeader((WAISDocumentShortHeader *) ptr); |
| 1267 |
s_free(response->ShortHeaders); |
| 1268 |
|
| 1269 |
if (response->LongHeaders != NULL) |
| 1270 |
for (i = 0, ptr = (void *) response->LongHeaders[i]; |
| 1271 |
ptr != NULL; |
| 1272 |
ptr = (void *) response->LongHeaders[++i]) |
| 1273 |
freeWAISDocumentLongHeader((WAISDocumentLongHeader *) ptr); |
| 1274 |
s_free(response->LongHeaders); |
| 1275 |
|
| 1276 |
if (response->Text != NULL) |
| 1277 |
for (i = 0, ptr = (void *) response->Text[i]; |
| 1278 |
ptr != NULL; |
| 1279 |
ptr = (void *) response->Text[++i]) |
| 1280 |
freeWAISDocumentText((WAISDocumentText *) ptr); |
| 1281 |
s_free(response->Text); |
| 1282 |
|
| 1283 |
if (response->Headlines != NULL) |
| 1284 |
for (i = 0, ptr = (void *) response->Headlines[i]; |
| 1285 |
ptr != NULL; |
| 1286 |
ptr = (void *) response->Headlines[++i]) |
| 1287 |
freeWAISDocumentHeadlines((WAISDocumentHeadlines *) ptr); |
| 1288 |
s_free(response->Headlines); |
| 1289 |
|
| 1290 |
if (response->Codes != NULL) |
| 1291 |
for (i = 0, ptr = (void *) response->Codes[i]; |
| 1292 |
ptr != NULL; |
| 1293 |
ptr = (void *) response->Codes[++i]) |
| 1294 |
freeWAISDocumentCodes((WAISDocumentCodes *) ptr); |
| 1295 |
s_free(response->Codes); |
| 1296 |
|
| 1297 |
if (response->Diagnostics != NULL) |
| 1298 |
for (i = 0, ptr = (void *) response->Diagnostics[i]; |
| 1299 |
ptr != NULL; |
| 1300 |
ptr = (void *) response->Diagnostics[++i]) |
| 1301 |
freeDiag((diagnosticRecord *) ptr); |
| 1302 |
s_free(response->Diagnostics); |
| 1303 |
|
| 1304 |
s_free(response); |
| 1305 |
} |
| 1306 |
|
| 1307 |
/*----------------------------------------------------------------------*/ |
| 1308 |
|
| 1309 |
char *writeSearchResponseInfo(SearchResponseAPDU *query, |
| 1310 |
char *buffer, |
| 1311 |
long *len) |
| 1312 |
{ |
| 1313 |
unsigned long header_len = userInfoTagSize(DT_UserInformationLength, |
| 1314 |
DefWAISSearchResponseSize); |
| 1315 |
char *buf = buffer + header_len; |
| 1316 |
WAISSearchResponse *info = (WAISSearchResponse *) query->DatabaseDiagnosticRecords; |
| 1317 |
unsigned long size; |
| 1318 |
void *header = NULL; |
| 1319 |
long i; |
| 1320 |
|
| 1321 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 1322 |
|
| 1323 |
buf = writeString(info->SeedWordsUsed, DT_SeedWordsUsed, buf, len); |
| 1324 |
|
| 1325 |
/* write out all the headers */ |
| 1326 |
if (info->DocHeaders != NULL) { |
| 1327 |
for (i = 0, header = (void *) info->DocHeaders[i]; |
| 1328 |
header != NULL; |
| 1329 |
header = (void *) info->DocHeaders[++i]) |
| 1330 |
buf = writeWAISDocumentHeader((WAISDocumentHeader *) header, buf, len); |
| 1331 |
} |
| 1332 |
|
| 1333 |
if (info->ShortHeaders != NULL) { |
| 1334 |
for (i = 0, header = (void *) info->ShortHeaders[i]; |
| 1335 |
header != NULL; |
| 1336 |
header = (void *) info->ShortHeaders[++i]) |
| 1337 |
buf = writeWAISDocumentShortHeader((WAISDocumentShortHeader *) header, |
| 1338 |
buf, |
| 1339 |
len); |
| 1340 |
} |
| 1341 |
|
| 1342 |
if (info->LongHeaders != NULL) { |
| 1343 |
for (i = 0, header = (void *) info->LongHeaders[i]; |
| 1344 |
header != NULL; |
| 1345 |
header = (void *) info->LongHeaders[++i]) |
| 1346 |
buf = writeWAISDocumentLongHeader((WAISDocumentLongHeader *) header, |
| 1347 |
buf, |
| 1348 |
len); |
| 1349 |
} |
| 1350 |
|
| 1351 |
if (info->Text != NULL) { |
| 1352 |
for (i = 0, header = (void *) info->Text[i]; |
| 1353 |
header != NULL; |
| 1354 |
header = (void *) info->Text[++i]) |
| 1355 |
buf = writeWAISDocumentText((WAISDocumentText *) header, buf, len); |
| 1356 |
} |
| 1357 |
|
| 1358 |
if (info->Headlines != NULL) { |
| 1359 |
for (i = 0, header = (void *) info->Headlines[i]; |
| 1360 |
header != NULL; |
| 1361 |
header = (void *) info->Headlines[++i]) |
| 1362 |
buf = writeWAISDocumentHeadlines((WAISDocumentHeadlines *) header, |
| 1363 |
buf, |
| 1364 |
len); |
| 1365 |
} |
| 1366 |
|
| 1367 |
if (info->Codes != NULL) { |
| 1368 |
for (i = 0, header = (void *) info->Codes[i]; |
| 1369 |
header != NULL; |
| 1370 |
header = (void *) info->Codes[++i]) |
| 1371 |
buf = writeWAISDocumentCodes((WAISDocumentCodes *) header, buf, len); |
| 1372 |
} |
| 1373 |
|
| 1374 |
if (info->Diagnostics != NULL) { |
| 1375 |
for (i = 0, header = (void *) info->Diagnostics[i]; |
| 1376 |
header != NULL; |
| 1377 |
header = (void *) info->Diagnostics[++i]) |
| 1378 |
buf = writeDiag((diagnosticRecord *) header, buf, len); |
| 1379 |
} |
| 1380 |
|
| 1381 |
/* now write the header and size */ |
| 1382 |
size = buf - buffer; |
| 1383 |
buf = writeUserInfoHeader(DT_UserInformationLength, |
| 1384 |
size, |
| 1385 |
header_len, |
| 1386 |
buffer, |
| 1387 |
len); |
| 1388 |
|
| 1389 |
return (buf); |
| 1390 |
} |
| 1391 |
|
| 1392 |
/*----------------------------------------------------------------------*/ |
| 1393 |
|
| 1394 |
static void cleanUpWaisSearchResponse(char *buf, |
| 1395 |
char *seedWordsUsed, |
| 1396 |
WAISDocumentHeader **docHeaders, |
| 1397 |
WAISDocumentShortHeader **shortHeaders, |
| 1398 |
WAISDocumentLongHeader **longHeaders, |
| 1399 |
WAISDocumentText **text, |
| 1400 |
WAISDocumentHeadlines **headlines, |
| 1401 |
WAISDocumentCodes **codes, |
| 1402 |
diagnosticRecord ** diags) |
| 1403 |
/* if buf is NULL, we have just gotten a read error, and need to clean up |
| 1404 |
any state we have built. If not, then everything is going fine, and |
| 1405 |
we should just hang loose |
| 1406 |
*/ |
| 1407 |
{ |
| 1408 |
void *ptr = NULL; |
| 1409 |
long i; |
| 1410 |
|
| 1411 |
if (buf == NULL) { |
| 1412 |
s_free(seedWordsUsed); |
| 1413 |
if (docHeaders != NULL) |
| 1414 |
for (i = 0, ptr = (void *) docHeaders[i]; ptr != NULL; |
| 1415 |
ptr = (void *) docHeaders[++i]) |
| 1416 |
freeWAISDocumentHeader((WAISDocumentHeader *) ptr); |
| 1417 |
s_free(docHeaders); |
| 1418 |
if (shortHeaders != NULL) |
| 1419 |
for (i = 0, ptr = (void *) shortHeaders[i]; ptr != NULL; |
| 1420 |
ptr = (void *) shortHeaders[++i]) |
| 1421 |
freeWAISDocumentShortHeader((WAISDocumentShortHeader *) ptr); |
| 1422 |
s_free(shortHeaders); |
| 1423 |
if (longHeaders != NULL) |
| 1424 |
for (i = 0, ptr = (void *) longHeaders[i]; ptr != NULL; |
| 1425 |
ptr = (void *) longHeaders[++i]) |
| 1426 |
freeWAISDocumentLongHeader((WAISDocumentLongHeader *) ptr); |
| 1427 |
s_free(longHeaders); |
| 1428 |
if (text != NULL) |
| 1429 |
for (i = 0, ptr = (void *) text[i]; |
| 1430 |
ptr != NULL; |
| 1431 |
ptr = (void *) text[++i]) |
| 1432 |
freeWAISDocumentText((WAISDocumentText *) ptr); |
| 1433 |
s_free(text); |
| 1434 |
if (headlines != NULL) |
| 1435 |
for (i = 0, ptr = (void *) headlines[i]; ptr != NULL; |
| 1436 |
ptr = (void *) headlines[++i]) |
| 1437 |
freeWAISDocumentHeadlines((WAISDocumentHeadlines *) ptr); |
| 1438 |
s_free(headlines); |
| 1439 |
if (codes != NULL) |
| 1440 |
for (i = 0, ptr = (void *) codes[i]; ptr != NULL; |
| 1441 |
ptr = (void *) codes[++i]) |
| 1442 |
freeWAISDocumentCodes((WAISDocumentCodes *) ptr); |
| 1443 |
s_free(codes); |
| 1444 |
if (diags != NULL) |
| 1445 |
for (i = 0, ptr = (void *) diags[i]; ptr != NULL; |
| 1446 |
ptr = (void *) diags[++i]) |
| 1447 |
freeDiag((diagnosticRecord *) ptr); |
| 1448 |
s_free(diags); |
| 1449 |
} |
| 1450 |
} |
| 1451 |
|
| 1452 |
/*----------------------------------------------------------------------*/ |
| 1453 |
|
| 1454 |
char *readSearchResponseInfo(void **info, |
| 1455 |
char *buffer) |
| 1456 |
{ |
| 1457 |
char *buf = buffer; |
| 1458 |
unsigned long size; |
| 1459 |
unsigned long headerSize; |
| 1460 |
data_tag tag1; |
| 1461 |
void *header = NULL; |
| 1462 |
WAISDocumentHeader **docHeaders = NULL; |
| 1463 |
WAISDocumentShortHeader **shortHeaders = NULL; |
| 1464 |
WAISDocumentLongHeader **longHeaders = NULL; |
| 1465 |
WAISDocumentText **text = NULL; |
| 1466 |
WAISDocumentHeadlines **headlines = NULL; |
| 1467 |
WAISDocumentCodes **codes = NULL; |
| 1468 |
long numDocHeaders, numLongHeaders, numShortHeaders, numText, numHeadlines; |
| 1469 |
long numCodes; |
| 1470 |
char *seedWordsUsed = NULL; |
| 1471 |
diagnosticRecord **diags = NULL; |
| 1472 |
diagnosticRecord *diag = NULL; |
| 1473 |
long numDiags = 0; |
| 1474 |
|
| 1475 |
numDocHeaders = |
| 1476 |
numLongHeaders = |
| 1477 |
numShortHeaders = |
| 1478 |
numText = |
| 1479 |
numHeadlines = |
| 1480 |
numCodes = 0; |
| 1481 |
|
| 1482 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 1483 |
headerSize = buf - buffer; |
| 1484 |
|
| 1485 |
while (buf < (buffer + size + headerSize)) { |
| 1486 |
data_tag tag = peekTag(buf); |
| 1487 |
|
| 1488 |
switch (tag) { |
| 1489 |
case DT_SeedWordsUsed: |
| 1490 |
buf = readString(&seedWordsUsed, buf); |
| 1491 |
break; |
| 1492 |
case DT_DatabaseDiagnosticRecords: |
| 1493 |
if (diags == NULL) /* create a new diag list */ |
| 1494 |
{ |
| 1495 |
diags = S_MALLOC2(diagnosticRecord *); |
| 1496 |
} else { /* grow the diag list */ |
| 1497 |
diags = S_REALLOC2(diagnosticRecord *, diags, numDiags); |
| 1498 |
} |
| 1499 |
buf = readDiag(&diag, buf); |
| 1500 |
diags[numDiags++] = diag; /* put it in the list */ |
| 1501 |
diags[numDiags] = NULL; |
| 1502 |
break; |
| 1503 |
case DT_DocumentHeaderGroup: |
| 1504 |
if (docHeaders == NULL) /* create a new header list */ |
| 1505 |
{ |
| 1506 |
docHeaders = S_MALLOC2(WAISDocumentHeader *); |
| 1507 |
} else { /* grow the doc list */ |
| 1508 |
docHeaders = S_REALLOC2(WAISDocumentHeader *, docHeaders, numDocHeaders); |
| 1509 |
} |
| 1510 |
buf = readWAISDocumentHeader((WAISDocumentHeader **) &header, buf); |
| 1511 |
cleanUpWaisSearchResponse(buf, |
| 1512 |
seedWordsUsed, |
| 1513 |
docHeaders, |
| 1514 |
shortHeaders, |
| 1515 |
longHeaders, |
| 1516 |
text, |
| 1517 |
headlines, |
| 1518 |
codes, |
| 1519 |
diags); |
| 1520 |
RETURN_ON_NULL(buf); |
| 1521 |
docHeaders[numDocHeaders++] = |
| 1522 |
(WAISDocumentHeader *) header; /* put it in the list */ |
| 1523 |
docHeaders[numDocHeaders] = NULL; |
| 1524 |
break; |
| 1525 |
case DT_DocumentShortHeaderGroup: |
| 1526 |
if (shortHeaders == NULL) /* create a new header list */ |
| 1527 |
{ |
| 1528 |
shortHeaders = S_MALLOC2(WAISDocumentShortHeader *); |
| 1529 |
} else { /* grow the doc list */ |
| 1530 |
shortHeaders = S_REALLOC2(WAISDocumentShortHeader *, |
| 1531 |
shortHeaders, |
| 1532 |
numShortHeaders); |
| 1533 |
} |
| 1534 |
buf = readWAISDocumentShortHeader((WAISDocumentShortHeader **) &header, |
| 1535 |
buf); |
| 1536 |
cleanUpWaisSearchResponse(buf, |
| 1537 |
seedWordsUsed, |
| 1538 |
docHeaders, |
| 1539 |
shortHeaders, |
| 1540 |
longHeaders, |
| 1541 |
text, |
| 1542 |
headlines, |
| 1543 |
codes, |
| 1544 |
diags); |
| 1545 |
RETURN_ON_NULL(buf); |
| 1546 |
shortHeaders[numShortHeaders++] = |
| 1547 |
(WAISDocumentShortHeader *) header; /* put it in the list */ |
| 1548 |
shortHeaders[numShortHeaders] = NULL; |
| 1549 |
break; |
| 1550 |
case DT_DocumentLongHeaderGroup: |
| 1551 |
if (longHeaders == NULL) /* create a new header list */ |
| 1552 |
{ |
| 1553 |
longHeaders = S_MALLOC2(WAISDocumentLongHeader *); |
| 1554 |
} else { /* grow the doc list */ |
| 1555 |
longHeaders = S_REALLOC2(WAISDocumentLongHeader *, |
| 1556 |
longHeaders, |
| 1557 |
numLongHeaders); |
| 1558 |
} |
| 1559 |
buf = readWAISDocumentLongHeader((WAISDocumentLongHeader **) &header, |
| 1560 |
buf); |
| 1561 |
cleanUpWaisSearchResponse(buf, |
| 1562 |
seedWordsUsed, |
| 1563 |
docHeaders, |
| 1564 |
shortHeaders, |
| 1565 |
longHeaders, |
| 1566 |
text, |
| 1567 |
headlines, |
| 1568 |
codes, |
| 1569 |
diags); |
| 1570 |
RETURN_ON_NULL(buf); |
| 1571 |
longHeaders[numLongHeaders++] = |
| 1572 |
(WAISDocumentLongHeader *) header; /* put it in the list */ |
| 1573 |
longHeaders[numLongHeaders] = NULL; |
| 1574 |
break; |
| 1575 |
case DT_DocumentTextGroup: |
| 1576 |
if (text == NULL) /* create a new list */ |
| 1577 |
{ |
| 1578 |
text = S_MALLOC2(WAISDocumentText *); |
| 1579 |
} else { /* grow the list */ |
| 1580 |
text = S_REALLOC2(WAISDocumentText *, text, numText); |
| 1581 |
} |
| 1582 |
buf = readWAISDocumentText((WAISDocumentText **) &header, buf); |
| 1583 |
cleanUpWaisSearchResponse(buf, |
| 1584 |
seedWordsUsed, |
| 1585 |
docHeaders, |
| 1586 |
shortHeaders, |
| 1587 |
longHeaders, |
| 1588 |
text, |
| 1589 |
headlines, |
| 1590 |
codes, |
| 1591 |
diags); |
| 1592 |
RETURN_ON_NULL(buf); |
| 1593 |
text[numText++] = |
| 1594 |
(WAISDocumentText *) header; /* put it in the list */ |
| 1595 |
text[numText] = NULL; |
| 1596 |
break; |
| 1597 |
case DT_DocumentHeadlineGroup: |
| 1598 |
if (headlines == NULL) /* create a new list */ |
| 1599 |
{ |
| 1600 |
headlines = S_MALLOC2(WAISDocumentHeadlines *); |
| 1601 |
} else { /* grow the list */ |
| 1602 |
headlines = S_REALLOC2(WAISDocumentHeadlines *, headlines, numHeadlines); |
| 1603 |
} |
| 1604 |
buf = readWAISDocumentHeadlines((WAISDocumentHeadlines **) &header, |
| 1605 |
buf); |
| 1606 |
cleanUpWaisSearchResponse(buf, |
| 1607 |
seedWordsUsed, |
| 1608 |
docHeaders, |
| 1609 |
shortHeaders, |
| 1610 |
longHeaders, |
| 1611 |
text, |
| 1612 |
headlines, |
| 1613 |
codes, |
| 1614 |
diags); |
| 1615 |
RETURN_ON_NULL(buf); |
| 1616 |
headlines[numHeadlines++] = |
| 1617 |
(WAISDocumentHeadlines *) header; /* put it in the list */ |
| 1618 |
headlines[numHeadlines] = NULL; |
| 1619 |
break; |
| 1620 |
case DT_DocumentCodeGroup: |
| 1621 |
if (codes == NULL) /* create a new list */ |
| 1622 |
{ |
| 1623 |
codes = S_MALLOC2(WAISDocumentCodes *); |
| 1624 |
} else { /* grow the list */ |
| 1625 |
codes = S_REALLOC2(WAISDocumentCodes *, codes, numCodes); |
| 1626 |
} |
| 1627 |
buf = readWAISDocumentCodes((WAISDocumentCodes **) &header, buf); |
| 1628 |
cleanUpWaisSearchResponse(buf, |
| 1629 |
seedWordsUsed, |
| 1630 |
docHeaders, |
| 1631 |
shortHeaders, |
| 1632 |
longHeaders, |
| 1633 |
text, |
| 1634 |
headlines, |
| 1635 |
codes, |
| 1636 |
diags); |
| 1637 |
RETURN_ON_NULL(buf); |
| 1638 |
codes[numCodes++] = |
| 1639 |
(WAISDocumentCodes *) header; /* put it in the list */ |
| 1640 |
codes[numCodes] = NULL; |
| 1641 |
break; |
| 1642 |
default: |
| 1643 |
cleanUpWaisSearchResponse(buf, |
| 1644 |
seedWordsUsed, |
| 1645 |
docHeaders, |
| 1646 |
shortHeaders, |
| 1647 |
longHeaders, |
| 1648 |
text, |
| 1649 |
headlines, |
| 1650 |
codes, |
| 1651 |
diags); |
| 1652 |
REPORT_READ_ERROR(buf); |
| 1653 |
break; |
| 1654 |
} |
| 1655 |
} |
| 1656 |
|
| 1657 |
*info = (void *) makeWAISSearchResponse(seedWordsUsed, |
| 1658 |
docHeaders, |
| 1659 |
shortHeaders, |
| 1660 |
longHeaders, |
| 1661 |
text, |
| 1662 |
headlines, |
| 1663 |
codes, |
| 1664 |
diags); |
| 1665 |
|
| 1666 |
return (buf); |
| 1667 |
} |
| 1668 |
|
| 1669 |
/*----------------------------------------------------------------------*/ |
| 1670 |
|
| 1671 |
WAISDocumentText *makeWAISDocumentText(any *docID, |
| 1672 |
long versionNumber, |
| 1673 |
any *documentText) |
| 1674 |
{ |
| 1675 |
WAISDocumentText *docText = S_MALLOC(WAISDocumentText); |
| 1676 |
|
| 1677 |
docText->DocumentID = docID; |
| 1678 |
docText->VersionNumber = versionNumber; |
| 1679 |
docText->DocumentText = documentText; |
| 1680 |
|
| 1681 |
return (docText); |
| 1682 |
} |
| 1683 |
|
| 1684 |
/*----------------------------------------------------------------------*/ |
| 1685 |
|
| 1686 |
void freeWAISDocumentText(WAISDocumentText *docText) |
| 1687 |
{ |
| 1688 |
freeAny(docText->DocumentID); |
| 1689 |
freeAny(docText->DocumentText); |
| 1690 |
s_free(docText); |
| 1691 |
} |
| 1692 |
|
| 1693 |
/*----------------------------------------------------------------------*/ |
| 1694 |
|
| 1695 |
char *writeWAISDocumentText(WAISDocumentText *docText, char *buffer, |
| 1696 |
long *len) |
| 1697 |
{ |
| 1698 |
unsigned long header_len = userInfoTagSize(DT_DocumentTextGroup, |
| 1699 |
DefWAISDocTextSize); |
| 1700 |
char *buf = buffer + header_len; |
| 1701 |
unsigned long size; |
| 1702 |
|
| 1703 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 1704 |
|
| 1705 |
buf = writeAny(docText->DocumentID, DT_DocumentID, buf, len); |
| 1706 |
buf = writeNum(docText->VersionNumber, DT_VersionNumber, buf, len); |
| 1707 |
buf = writeAny(docText->DocumentText, DT_DocumentText, buf, len); |
| 1708 |
|
| 1709 |
/* now write the header and size */ |
| 1710 |
size = buf - buffer; |
| 1711 |
buf = writeUserInfoHeader(DT_DocumentTextGroup, size, header_len, buffer, len); |
| 1712 |
|
| 1713 |
return (buf); |
| 1714 |
} |
| 1715 |
|
| 1716 |
/*----------------------------------------------------------------------*/ |
| 1717 |
|
| 1718 |
char *readWAISDocumentText(WAISDocumentText **docText, char *buffer) |
| 1719 |
{ |
| 1720 |
char *buf = buffer; |
| 1721 |
unsigned long size; |
| 1722 |
unsigned long headerSize; |
| 1723 |
data_tag tag1; |
| 1724 |
any *docID, *documentText; |
| 1725 |
long versionNumber; |
| 1726 |
|
| 1727 |
docID = documentText = NULL; |
| 1728 |
versionNumber = UNUSED; |
| 1729 |
|
| 1730 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 1731 |
headerSize = buf - buffer; |
| 1732 |
|
| 1733 |
while (buf < (buffer + size + headerSize)) { |
| 1734 |
data_tag tag = peekTag(buf); |
| 1735 |
|
| 1736 |
switch (tag) { |
| 1737 |
case DT_DocumentID: |
| 1738 |
buf = readAny(&docID, buf); |
| 1739 |
break; |
| 1740 |
case DT_VersionNumber: |
| 1741 |
buf = readNum(&versionNumber, buf); |
| 1742 |
break; |
| 1743 |
case DT_DocumentText: |
| 1744 |
buf = readAny(&documentText, buf); |
| 1745 |
break; |
| 1746 |
default: |
| 1747 |
freeAny(docID); |
| 1748 |
freeAny(documentText); |
| 1749 |
REPORT_READ_ERROR(buf); |
| 1750 |
break; |
| 1751 |
} |
| 1752 |
} |
| 1753 |
|
| 1754 |
*docText = makeWAISDocumentText(docID, versionNumber, documentText); |
| 1755 |
return (buf); |
| 1756 |
} |
| 1757 |
|
| 1758 |
/*----------------------------------------------------------------------*/ |
| 1759 |
|
| 1760 |
WAISDocumentHeadlines *makeWAISDocumentHeadlines(any *docID, |
| 1761 |
long versionNumber, |
| 1762 |
char *source, |
| 1763 |
char *date, |
| 1764 |
char *headline, |
| 1765 |
char *originCity) |
| 1766 |
{ |
| 1767 |
WAISDocumentHeadlines *docHeadline = S_MALLOC(WAISDocumentHeadlines); |
| 1768 |
|
| 1769 |
docHeadline->DocumentID = docID; |
| 1770 |
docHeadline->VersionNumber = versionNumber; |
| 1771 |
docHeadline->Source = source; |
| 1772 |
docHeadline->Date = date; |
| 1773 |
docHeadline->Headline = headline; |
| 1774 |
docHeadline->OriginCity = originCity; |
| 1775 |
|
| 1776 |
return (docHeadline); |
| 1777 |
} |
| 1778 |
|
| 1779 |
/*----------------------------------------------------------------------*/ |
| 1780 |
|
| 1781 |
void freeWAISDocumentHeadlines(WAISDocumentHeadlines *docHeadline) |
| 1782 |
{ |
| 1783 |
freeAny(docHeadline->DocumentID); |
| 1784 |
s_free(docHeadline->Source); |
| 1785 |
s_free(docHeadline->Date); |
| 1786 |
s_free(docHeadline->Headline); |
| 1787 |
s_free(docHeadline->OriginCity); |
| 1788 |
s_free(docHeadline); |
| 1789 |
} |
| 1790 |
|
| 1791 |
/*----------------------------------------------------------------------*/ |
| 1792 |
|
| 1793 |
char *writeWAISDocumentHeadlines(WAISDocumentHeadlines *docHeadline, char *buffer, |
| 1794 |
long *len) |
| 1795 |
{ |
| 1796 |
unsigned long header_len = userInfoTagSize(DT_DocumentHeadlineGroup, |
| 1797 |
DefWAISDocHeadlineSize); |
| 1798 |
char *buf = buffer + header_len; |
| 1799 |
unsigned long size; |
| 1800 |
|
| 1801 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 1802 |
|
| 1803 |
buf = writeAny(docHeadline->DocumentID, DT_DocumentID, buf, len); |
| 1804 |
buf = writeNum(docHeadline->VersionNumber, DT_VersionNumber, buf, len); |
| 1805 |
buf = writeString(docHeadline->Source, DT_Source, buf, len); |
| 1806 |
buf = writeString(docHeadline->Date, DT_Date, buf, len); |
| 1807 |
buf = writeString(docHeadline->Headline, DT_Headline, buf, len); |
| 1808 |
buf = writeString(docHeadline->OriginCity, DT_OriginCity, buf, len); |
| 1809 |
|
| 1810 |
/* now write the header and size */ |
| 1811 |
size = buf - buffer; |
| 1812 |
buf = writeUserInfoHeader(DT_DocumentHeadlineGroup, |
| 1813 |
size, |
| 1814 |
header_len, |
| 1815 |
buffer, |
| 1816 |
len); |
| 1817 |
|
| 1818 |
return (buf); |
| 1819 |
} |
| 1820 |
|
| 1821 |
/*----------------------------------------------------------------------*/ |
| 1822 |
|
| 1823 |
char *readWAISDocumentHeadlines(WAISDocumentHeadlines **docHeadline, char *buffer) |
| 1824 |
{ |
| 1825 |
char *buf = buffer; |
| 1826 |
unsigned long size; |
| 1827 |
unsigned long headerSize; |
| 1828 |
data_tag tag1; |
| 1829 |
any *docID; |
| 1830 |
long versionNumber; |
| 1831 |
char *source, *date, *headline, *originCity; |
| 1832 |
|
| 1833 |
docID = NULL; |
| 1834 |
versionNumber = UNUSED; |
| 1835 |
source = date = headline = originCity = NULL; |
| 1836 |
|
| 1837 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 1838 |
headerSize = buf - buffer; |
| 1839 |
|
| 1840 |
while (buf < (buffer + size + headerSize)) { |
| 1841 |
data_tag tag = peekTag(buf); |
| 1842 |
|
| 1843 |
switch (tag) { |
| 1844 |
case DT_DocumentID: |
| 1845 |
buf = readAny(&docID, buf); |
| 1846 |
break; |
| 1847 |
case DT_VersionNumber: |
| 1848 |
buf = readNum(&versionNumber, buf); |
| 1849 |
break; |
| 1850 |
case DT_Source: |
| 1851 |
buf = readString(&source, buf); |
| 1852 |
break; |
| 1853 |
case DT_Date: |
| 1854 |
buf = readString(&date, buf); |
| 1855 |
break; |
| 1856 |
case DT_Headline: |
| 1857 |
buf = readString(&headline, buf); |
| 1858 |
break; |
| 1859 |
case DT_OriginCity: |
| 1860 |
buf = readString(&originCity, buf); |
| 1861 |
break; |
| 1862 |
default: |
| 1863 |
freeAny(docID); |
| 1864 |
s_free(source); |
| 1865 |
s_free(date); |
| 1866 |
s_free(headline); |
| 1867 |
s_free(originCity); |
| 1868 |
REPORT_READ_ERROR(buf); |
| 1869 |
break; |
| 1870 |
} |
| 1871 |
} |
| 1872 |
|
| 1873 |
*docHeadline = makeWAISDocumentHeadlines(docID, versionNumber, source, date, |
| 1874 |
headline, originCity); |
| 1875 |
return (buf); |
| 1876 |
} |
| 1877 |
|
| 1878 |
/*----------------------------------------------------------------------*/ |
| 1879 |
|
| 1880 |
WAISDocumentCodes *makeWAISDocumentCodes(any *docID, |
| 1881 |
long versionNumber, |
| 1882 |
char *stockCodes, |
| 1883 |
char *companyCodes, |
| 1884 |
char *industryCodes) |
| 1885 |
{ |
| 1886 |
WAISDocumentCodes *docCodes = S_MALLOC(WAISDocumentCodes); |
| 1887 |
|
| 1888 |
docCodes->DocumentID = docID; |
| 1889 |
docCodes->VersionNumber = versionNumber; |
| 1890 |
docCodes->StockCodes = stockCodes; |
| 1891 |
docCodes->CompanyCodes = companyCodes; |
| 1892 |
docCodes->IndustryCodes = industryCodes; |
| 1893 |
|
| 1894 |
return (docCodes); |
| 1895 |
} |
| 1896 |
|
| 1897 |
/*----------------------------------------------------------------------*/ |
| 1898 |
|
| 1899 |
void freeWAISDocumentCodes(WAISDocumentCodes *docCodes) |
| 1900 |
{ |
| 1901 |
freeAny(docCodes->DocumentID); |
| 1902 |
s_free(docCodes->StockCodes); |
| 1903 |
s_free(docCodes->CompanyCodes); |
| 1904 |
s_free(docCodes->IndustryCodes); |
| 1905 |
s_free(docCodes); |
| 1906 |
} |
| 1907 |
|
| 1908 |
/*----------------------------------------------------------------------*/ |
| 1909 |
|
| 1910 |
char *writeWAISDocumentCodes(WAISDocumentCodes *docCodes, |
| 1911 |
char *buffer, |
| 1912 |
long *len) |
| 1913 |
{ |
| 1914 |
unsigned long header_len = userInfoTagSize(DT_DocumentCodeGroup, |
| 1915 |
DefWAISDocCodeSize); |
| 1916 |
char *buf = buffer + header_len; |
| 1917 |
unsigned long size; |
| 1918 |
|
| 1919 |
RESERVE_SPACE_FOR_WAIS_HEADER(len); |
| 1920 |
|
| 1921 |
buf = writeAny(docCodes->DocumentID, DT_DocumentID, buf, len); |
| 1922 |
buf = writeNum(docCodes->VersionNumber, DT_VersionNumber, buf, len); |
| 1923 |
buf = writeString(docCodes->StockCodes, DT_StockCodes, buf, len); |
| 1924 |
buf = writeString(docCodes->CompanyCodes, DT_CompanyCodes, buf, len); |
| 1925 |
buf = writeString(docCodes->IndustryCodes, DT_IndustryCodes, buf, len); |
| 1926 |
|
| 1927 |
/* now write the header and size */ |
| 1928 |
size = buf - buffer; |
| 1929 |
buf = writeUserInfoHeader(DT_DocumentCodeGroup, size, header_len, buffer, len); |
| 1930 |
|
| 1931 |
return (buf); |
| 1932 |
} |
| 1933 |
|
| 1934 |
/*----------------------------------------------------------------------*/ |
| 1935 |
|
| 1936 |
char *readWAISDocumentCodes(WAISDocumentCodes **docCodes, |
| 1937 |
char *buffer) |
| 1938 |
{ |
| 1939 |
char *buf = buffer; |
| 1940 |
unsigned long size; |
| 1941 |
unsigned long headerSize; |
| 1942 |
data_tag tag1; |
| 1943 |
any *docID; |
| 1944 |
long versionNumber; |
| 1945 |
char *stockCodes, *companyCodes, *industryCodes; |
| 1946 |
|
| 1947 |
docID = NULL; |
| 1948 |
versionNumber = UNUSED; |
| 1949 |
stockCodes = companyCodes = industryCodes = NULL; |
| 1950 |
|
| 1951 |
buf = readUserInfoHeader(&tag1, &size, buf); |
| 1952 |
headerSize = buf - buffer; |
| 1953 |
|
| 1954 |
while (buf < (buffer + size + headerSize)) { |
| 1955 |
data_tag tag = peekTag(buf); |
| 1956 |
|
| 1957 |
switch (tag) { |
| 1958 |
case DT_DocumentID: |
| 1959 |
buf = readAny(&docID, buf); |
| 1960 |
break; |
| 1961 |
case DT_VersionNumber: |
| 1962 |
buf = readNum(&versionNumber, buf); |
| 1963 |
break; |
| 1964 |
case DT_StockCodes: |
| 1965 |
buf = readString(&stockCodes, buf); |
| 1966 |
break; |
| 1967 |
case DT_CompanyCodes: |
| 1968 |
buf = readString(&companyCodes, buf); |
| 1969 |
break; |
| 1970 |
case DT_IndustryCodes: |
| 1971 |
buf = readString(&industryCodes, buf); |
| 1972 |
break; |
| 1973 |
default: |
| 1974 |
freeAny(docID); |
| 1975 |
s_free(stockCodes); |
| 1976 |
s_free(companyCodes); |
| 1977 |
s_free(industryCodes); |
| 1978 |
REPORT_READ_ERROR(buf); |
| 1979 |
break; |
| 1980 |
} |
| 1981 |
} |
| 1982 |
|
| 1983 |
*docCodes = makeWAISDocumentCodes(docID, versionNumber, stockCodes, |
| 1984 |
companyCodes, industryCodes); |
| 1985 |
return (buf); |
| 1986 |
} |
| 1987 |
|
| 1988 |
/*----------------------------------------------------------------------*/ |
| 1989 |
|
| 1990 |
char *writePresentInfo(PresentAPDU * present GCC_UNUSED, char *buffer, |
| 1991 |
long *len GCC_UNUSED) |
| 1992 |
{ |
| 1993 |
/* The WAIS protocol doesn't use present info */ |
| 1994 |
return (buffer); |
| 1995 |
} |
| 1996 |
|
| 1997 |
/*----------------------------------------------------------------------*/ |
| 1998 |
|
| 1999 |
char *readPresentInfo(void **info, |
| 2000 |
char *buffer) |
| 2001 |
{ |
| 2002 |
/* The WAIS protocol doesn't use present info */ |
| 2003 |
*info = NULL; |
| 2004 |
return (buffer); |
| 2005 |
} |
| 2006 |
|
| 2007 |
/*----------------------------------------------------------------------*/ |
| 2008 |
|
| 2009 |
char *writePresentResponseInfo(PresentResponseAPDU * response GCC_UNUSED, char *buffer, |
| 2010 |
long *len GCC_UNUSED) |
| 2011 |
{ |
| 2012 |
/* The WAIS protocol doesn't use presentResponse info */ |
| 2013 |
return (buffer); |
| 2014 |
} |
| 2015 |
|
| 2016 |
/*----------------------------------------------------------------------*/ |
| 2017 |
|
| 2018 |
char *readPresentResponseInfo(void **info, |
| 2019 |
char *buffer) |
| 2020 |
{ |
| 2021 |
/* The WAIS protocol doesn't use presentResponse info */ |
| 2022 |
*info = NULL; |
| 2023 |
return (buffer); |
| 2024 |
} |
| 2025 |
|
| 2026 |
/*----------------------------------------------------------------------*/ |
| 2027 |
|
| 2028 |
/* support for type 1 queries */ |
| 2029 |
|
| 2030 |
/* new use values (for the chunk types) */ |
| 2031 |
#define BYTE "wb" |
| 2032 |
#define LINE "wl" |
| 2033 |
#define PARAGRAPH "wp" |
| 2034 |
#define DATA_TYPE "wt" |
| 2035 |
|
| 2036 |
/* WAIS supports the following semantics for type 1 queries: |
| 2037 |
|
| 2038 |
1. retrieve the header/codes from a document: |
| 2039 |
|
| 2040 |
System_Control_Number = docID |
| 2041 |
Data Type = type (optional) |
| 2042 |
And |
| 2043 |
|
| 2044 |
2. retrieve a fragment of the text of a document: |
| 2045 |
|
| 2046 |
System_Control_Number = docID |
| 2047 |
Data Type = type (optional) |
| 2048 |
And |
| 2049 |
Chunk >= start |
| 2050 |
And |
| 2051 |
Chunk < end |
| 2052 |
And |
| 2053 |
|
| 2054 |
Information from multiple documents may be requested by using |
| 2055 |
groups of the above joined by: |
| 2056 |
|
| 2057 |
OR |
| 2058 |
|
| 2059 |
( XXX does an OR come after every group but the first, or do they |
| 2060 |
all come at the end? ) |
| 2061 |
|
| 2062 |
( XXX return type could be in the element set) |
| 2063 |
*/ |
| 2064 |
|
| 2065 |
static query_term **makeWAISQueryTerms(DocObj **docs) |
| 2066 |
/* given a null terminated list of docObjs, construct the appropriate |
| 2067 |
query of the form given above |
| 2068 |
*/ |
| 2069 |
{ |
| 2070 |
query_term **terms = NULL; |
| 2071 |
long numTerms = 0; |
| 2072 |
DocObj *doc = NULL; |
| 2073 |
long i; |
| 2074 |
|
| 2075 |
if (docs == NULL) |
| 2076 |
return ((query_term **) NULL); |
| 2077 |
|
| 2078 |
terms = (query_term **) s_malloc((size_t) (sizeof(query_term *) * 1)); |
| 2079 |
|
| 2080 |
terms[numTerms] = NULL; |
| 2081 |
|
| 2082 |
/* loop through the docs making terms for them all */ |
| 2083 |
for (i = 0, doc = docs[i]; doc != NULL; doc = docs[++i]) { |
| 2084 |
any *type = NULL; |
| 2085 |
|
| 2086 |
if (doc->Type != NULL) |
| 2087 |
type = stringToAny(doc->Type); |
| 2088 |
|
| 2089 |
if (doc->ChunkCode == CT_document) /* a whole document */ |
| 2090 |
{ |
| 2091 |
terms = S_REALLOC2(query_term *, terms, numTerms + 2); |
| 2092 |
|
| 2093 |
terms[numTerms++] = makeAttributeTerm(SYSTEM_CONTROL_NUMBER, |
| 2094 |
EQUAL, IGNORE, IGNORE, |
| 2095 |
IGNORE, IGNORE, doc->DocumentID); |
| 2096 |
if (type != NULL) { |
| 2097 |
terms[numTerms++] = makeAttributeTerm(DATA_TYPE, EQUAL, |
| 2098 |
IGNORE, IGNORE, IGNORE, |
| 2099 |
IGNORE, type); |
| 2100 |
terms[numTerms++] = makeOperatorTerm(AND); |
| 2101 |
} |
| 2102 |
terms[numTerms] = NULL; |
| 2103 |
} else { /* a document fragment */ |
| 2104 |
char chunk_att[ATTRIBUTE_SIZE]; |
| 2105 |
any *startChunk = NULL; |
| 2106 |
any *endChunk = NULL; |
| 2107 |
|
| 2108 |
terms = S_REALLOC2(query_term *, terms, numTerms + 6); |
| 2109 |
|
| 2110 |
switch (doc->ChunkCode) { |
| 2111 |
case CT_byte: |
| 2112 |
case CT_line: |
| 2113 |
{ |
| 2114 |
char start[20], end[20]; |
| 2115 |
|
| 2116 |
(doc->ChunkCode == CT_byte) ? |
| 2117 |
StrNCpy(chunk_att, BYTE, ATTRIBUTE_SIZE) : |
| 2118 |
StrNCpy(chunk_att, LINE, ATTRIBUTE_SIZE); |
| 2119 |
sprintf(start, "%ld", doc->ChunkStart.Pos); |
| 2120 |
startChunk = stringToAny(start); |
| 2121 |
sprintf(end, "%ld", doc->ChunkEnd.Pos); |
| 2122 |
endChunk = stringToAny(end); |
| 2123 |
} |
| 2124 |
break; |
| 2125 |
case CT_paragraph: |
| 2126 |
StrNCpy(chunk_att, PARAGRAPH, ATTRIBUTE_SIZE); |
| 2127 |
startChunk = doc->ChunkStart.ID; |
| 2128 |
endChunk = doc->ChunkEnd.ID; |
| 2129 |
break; |
| 2130 |
default: |
| 2131 |
/* error */ |
| 2132 |
break; |
| 2133 |
} |
| 2134 |
|
| 2135 |
terms[numTerms++] = makeAttributeTerm(SYSTEM_CONTROL_NUMBER, |
| 2136 |
EQUAL, IGNORE, IGNORE, |
| 2137 |
IGNORE, |
| 2138 |
IGNORE, doc->DocumentID); |
| 2139 |
if (type != NULL) { |
| 2140 |
terms[numTerms++] = makeAttributeTerm(DATA_TYPE, EQUAL, IGNORE, |
| 2141 |
IGNORE, IGNORE, IGNORE, |
| 2142 |
type); |
| 2143 |
terms[numTerms++] = makeOperatorTerm(AND); |
| 2144 |
} |
| 2145 |
terms[numTerms++] = makeAttributeTerm(chunk_att, |
| 2146 |
GREATER_THAN_OR_EQUAL, |
| 2147 |
IGNORE, IGNORE, IGNORE, |
| 2148 |
IGNORE, |
| 2149 |
startChunk); |
| 2150 |
terms[numTerms++] = makeOperatorTerm(AND); |
| 2151 |
terms[numTerms++] = makeAttributeTerm(chunk_att, LESS_THAN, |
| 2152 |
IGNORE, IGNORE, IGNORE, |
| 2153 |
IGNORE, |
| 2154 |
endChunk); |
| 2155 |
terms[numTerms++] = makeOperatorTerm(AND); |
| 2156 |
terms[numTerms] = NULL; |
| 2157 |
|
| 2158 |
if (doc->ChunkCode == CT_byte || doc->ChunkCode == CT_line) { |
| 2159 |
freeAny(startChunk); |
| 2160 |
freeAny(endChunk); |
| 2161 |
} |
| 2162 |
} |
| 2163 |
|
| 2164 |
freeAny(type); |
| 2165 |
|
| 2166 |
if (i != 0) /* multiple independent queries, need a disjunction */ |
| 2167 |
{ |
| 2168 |
terms = S_REALLOC2(query_term *, terms, numTerms); |
| 2169 |
|
| 2170 |
terms[numTerms++] = makeOperatorTerm(OR); |
| 2171 |
terms[numTerms] = NULL; |
| 2172 |
} |
| 2173 |
} |
| 2174 |
|
| 2175 |
return (terms); |
| 2176 |
} |
| 2177 |
|
| 2178 |
/*----------------------------------------------------------------------*/ |
| 2179 |
|
| 2180 |
static DocObj **makeWAISQueryDocs(query_term **terms) |
| 2181 |
/* given a list of terms in the form given above, convert them to |
| 2182 |
DocObjs. |
| 2183 |
*/ |
| 2184 |
{ |
| 2185 |
query_term *docTerm = NULL; |
| 2186 |
query_term *fragmentTerm = NULL; |
| 2187 |
DocObj **docs = NULL; |
| 2188 |
DocObj *doc = NULL; |
| 2189 |
long docNum, termNum; |
| 2190 |
|
| 2191 |
docNum = termNum = 0; |
| 2192 |
|
| 2193 |
docs = S_MALLOC(DocObj *); |
| 2194 |
|
| 2195 |
docs[docNum] = NULL; |
| 2196 |
|
| 2197 |
/* translate the terms into DocObjs */ |
| 2198 |
while (true) { |
| 2199 |
query_term *typeTerm = NULL; |
| 2200 |
char *type = NULL; |
| 2201 |
long startTermOffset; |
| 2202 |
|
| 2203 |
docTerm = terms[termNum]; |
| 2204 |
|
| 2205 |
if (docTerm == NULL) |
| 2206 |
break; /* we're done converting */ |
| 2207 |
|
| 2208 |
typeTerm = terms[termNum + 1]; /* get the lead Term if it exists */ |
| 2209 |
|
| 2210 |
if (strcmp(typeTerm->Use, DATA_TYPE) == 0) /* we do have a type */ |
| 2211 |
{ |
| 2212 |
startTermOffset = 3; |
| 2213 |
type = anyToString(typeTerm->Term); |
| 2214 |
} else { /* no type */ |
| 2215 |
startTermOffset = 1; |
| 2216 |
typeTerm = NULL; |
| 2217 |
type = NULL; |
| 2218 |
} |
| 2219 |
|
| 2220 |
/* grow the doc list */ |
| 2221 |
docs = S_REALLOC2(DocObj *, docs, docNum); |
| 2222 |
|
| 2223 |
/* figure out what kind of docObj to build - and build it */ |
| 2224 |
fragmentTerm = terms[termNum + startTermOffset]; |
| 2225 |
if (fragmentTerm != NULL && fragmentTerm->TermType == TT_Attribute) { /* build a document fragment */ |
| 2226 |
query_term *startTerm = fragmentTerm; |
| 2227 |
query_term *endTerm = terms[termNum + startTermOffset + 2]; |
| 2228 |
|
| 2229 |
if (strcmp(startTerm->Use, BYTE) == 0) { /* a byte chunk */ |
| 2230 |
doc = makeDocObjUsingBytes(duplicateAny(docTerm->Term), |
| 2231 |
type, |
| 2232 |
anyToLong(startTerm->Term), |
| 2233 |
anyToLong(endTerm->Term)); |
| 2234 |
log_write("byte"); |
| 2235 |
} else if (strcmp(startTerm->Use, LINE) == 0) { /* a line chunk */ |
| 2236 |
doc = makeDocObjUsingLines(duplicateAny(docTerm->Term), |
| 2237 |
type, |
| 2238 |
anyToLong(startTerm->Term), |
| 2239 |
anyToLong(endTerm->Term)); |
| 2240 |
log_write("line"); |
| 2241 |
} else { |
| 2242 |
log_write("chunk"); /* a paragraph chunk */ |
| 2243 |
doc = makeDocObjUsingParagraphs(duplicateAny(docTerm->Term), |
| 2244 |
type, |
| 2245 |
duplicateAny(startTerm->Term), |
| 2246 |
duplicateAny(endTerm->Term)); |
| 2247 |
} |
| 2248 |
termNum += (startTermOffset + 4); /* point to next term */ |
| 2249 |
} else { /* build a full document */ |
| 2250 |
doc = makeDocObjUsingWholeDocument(duplicateAny(docTerm->Term), |
| 2251 |
type); |
| 2252 |
log_write("whole doc"); |
| 2253 |
termNum += startTermOffset; /* point to next term */ |
| 2254 |
} |
| 2255 |
|
| 2256 |
docs[docNum++] = doc; /* insert the new document */ |
| 2257 |
|
| 2258 |
docs[docNum] = NULL; /* keep the doc list terminated */ |
| 2259 |
|
| 2260 |
if (terms[termNum] != NULL) |
| 2261 |
termNum++; /* skip the OR operator it necessary */ |
| 2262 |
else |
| 2263 |
break; /* we are done */ |
| 2264 |
} |
| 2265 |
|
| 2266 |
return (docs); |
| 2267 |
} |
| 2268 |
|
| 2269 |
/*----------------------------------------------------------------------*/ |
| 2270 |
|
| 2271 |
any *makeWAISTextQuery(DocObj **docs) |
| 2272 |
/* given a list of DocObjs, return an any whose contents is the corresponding |
| 2273 |
type 1 query |
| 2274 |
*/ |
| 2275 |
{ |
| 2276 |
any *buf = NULL; |
| 2277 |
query_term **terms = NULL; |
| 2278 |
|
| 2279 |
terms = makeWAISQueryTerms(docs); |
| 2280 |
buf = writeQuery(terms); |
| 2281 |
|
| 2282 |
doList((void **) terms, freeTerm); |
| 2283 |
s_free(terms); |
| 2284 |
|
| 2285 |
return (buf); |
| 2286 |
} |
| 2287 |
|
| 2288 |
/*----------------------------------------------------------------------*/ |
| 2289 |
|
| 2290 |
DocObj **readWAISTextQuery(any *buf) |
| 2291 |
/* given an any whose contents are type 1 queries of the WAIS sort, |
| 2292 |
construct a list of the corresponding DocObjs |
| 2293 |
*/ |
| 2294 |
{ |
| 2295 |
query_term **terms = NULL; |
| 2296 |
DocObj **docs = NULL; |
| 2297 |
|
| 2298 |
terms = readQuery(buf); |
| 2299 |
docs = makeWAISQueryDocs(terms); |
| 2300 |
|
| 2301 |
doList((void **) terms, freeTerm); |
| 2302 |
s_free(terms); |
| 2303 |
|
| 2304 |
return (docs); |
| 2305 |
} |
| 2306 |
|
| 2307 |
/*----------------------------------------------------------------------*/ |
| 2308 |
/* Customized free WAIS object routines: */ |
| 2309 |
/* */ |
| 2310 |
/* This set of procedures is for applications to free a WAIS object */ |
| 2311 |
/* which was made with makeWAISFOO. */ |
| 2312 |
/* Each procedure frees only the memory that was allocated in its */ |
| 2313 |
/* associated makeWAISFOO routine, thus it's not necessary for the */ |
| 2314 |
/* caller to assign nulls to the pointer fields of the WAIS object. */ |
| 2315 |
/*----------------------------------------------------------------------*/ |
| 2316 |
|
| 2317 |
void CSTFreeWAISInitResponse(WAISInitResponse *init) |
| 2318 |
/* free an object made with makeWAISInitResponse */ |
| 2319 |
{ |
| 2320 |
s_free(init); |
| 2321 |
} |
| 2322 |
|
| 2323 |
/*----------------------------------------------------------------------*/ |
| 2324 |
|
| 2325 |
void CSTFreeWAISSearch(WAISSearch *query) |
| 2326 |
/* destroy an object made with makeWAISSearch() */ |
| 2327 |
{ |
| 2328 |
s_free(query); |
| 2329 |
} |
| 2330 |
|
| 2331 |
/*----------------------------------------------------------------------*/ |
| 2332 |
|
| 2333 |
void CSTFreeDocObj(DocObj *doc) |
| 2334 |
/* free a docObj */ |
| 2335 |
{ |
| 2336 |
s_free(doc); |
| 2337 |
} |
| 2338 |
|
| 2339 |
/*----------------------------------------------------------------------*/ |
| 2340 |
|
| 2341 |
void CSTFreeWAISDocumentHeader(WAISDocumentHeader *header) |
| 2342 |
{ |
| 2343 |
s_free(header); |
| 2344 |
} |
| 2345 |
|
| 2346 |
/*----------------------------------------------------------------------*/ |
| 2347 |
|
| 2348 |
void CSTFreeWAISDocumentShortHeader(WAISDocumentShortHeader *header) |
| 2349 |
{ |
| 2350 |
s_free(header); |
| 2351 |
} |
| 2352 |
|
| 2353 |
/*----------------------------------------------------------------------*/ |
| 2354 |
|
| 2355 |
void CSTFreeWAISDocumentLongHeader(WAISDocumentLongHeader *header) |
| 2356 |
{ |
| 2357 |
s_free(header); |
| 2358 |
} |
| 2359 |
|
| 2360 |
/*----------------------------------------------------------------------*/ |
| 2361 |
|
| 2362 |
void CSTFreeWAISSearchResponse(WAISSearchResponse * response) |
| 2363 |
{ |
| 2364 |
s_free(response); |
| 2365 |
} |
| 2366 |
|
| 2367 |
/*----------------------------------------------------------------------*/ |
| 2368 |
|
| 2369 |
void CSTFreeWAISDocumentText(WAISDocumentText *docText) |
| 2370 |
{ |
| 2371 |
s_free(docText); |
| 2372 |
} |
| 2373 |
|
| 2374 |
/*----------------------------------------------------------------------*/ |
| 2375 |
|
| 2376 |
void CSTFreeWAISDocHeadlines(WAISDocumentHeadlines *docHeadline) |
| 2377 |
{ |
| 2378 |
s_free(docHeadline); |
| 2379 |
} |
| 2380 |
|
| 2381 |
/*----------------------------------------------------------------------*/ |
| 2382 |
|
| 2383 |
void CSTFreeWAISDocumentCodes(WAISDocumentCodes *docCodes) |
| 2384 |
{ |
| 2385 |
s_free(docCodes); |
| 2386 |
} |
| 2387 |
|
| 2388 |
/*----------------------------------------------------------------------*/ |
| 2389 |
|
| 2390 |
void CSTFreeWAISTextQuery(any *query) |
| 2391 |
{ |
| 2392 |
freeAny(query); |
| 2393 |
} |
| 2394 |
|
| 2395 |
/*----------------------------------------------------------------------*/ |
| 2396 |
|
| 2397 |
/* |
| 2398 |
* Routines originally from WMessage.c -- FM |
| 2399 |
* |
| 2400 |
*----------------------------------------------------------------------*/ |
| 2401 |
/* WIDE AREA INFORMATION SERVER SOFTWARE |
| 2402 |
* No guarantees or restrictions. See the readme file for the full standard |
| 2403 |
* disclaimer. |
| 2404 |
* 3.26.90 |
| 2405 |
*/ |
| 2406 |
|
| 2407 |
/* This file is for reading and writing the wais packet header. |
| 2408 |
* [email protected] |
| 2409 |
*/ |
| 2410 |
|
| 2411 |
/* to do: |
| 2412 |
* add check sum |
| 2413 |
* what do you do when checksum is wrong? |
| 2414 |
*/ |
| 2415 |
|
| 2416 |
/*---------------------------------------------------------------------*/ |
| 2417 |
|
| 2418 |
void readWAISPacketHeader(char *msgBuffer, |
| 2419 |
WAISMessage * header_struct) |
| 2420 |
{ |
| 2421 |
/* msgBuffer is a string containing at least HEADER_LENGTH bytes. */ |
| 2422 |
|
| 2423 |
memmove(header_struct->msg_len, msgBuffer, (size_t) 10); |
| 2424 |
header_struct->msg_type = char_downcase((unsigned long) msgBuffer[10]); |
| 2425 |
header_struct->hdr_vers = char_downcase((unsigned long) msgBuffer[11]); |
| 2426 |
memmove(header_struct->server, (void *) (msgBuffer + 12), (size_t) 10); |
| 2427 |
header_struct->compression = char_downcase((unsigned long) msgBuffer[22]); |
| 2428 |
header_struct->encoding = char_downcase((unsigned long) msgBuffer[23]); |
| 2429 |
header_struct->msg_checksum = char_downcase((unsigned long) msgBuffer[24]); |
| 2430 |
} |
| 2431 |
|
| 2432 |
/*---------------------------------------------------------------------*/ |
| 2433 |
|
| 2434 |
/* this modifies the header argument. See wais-message.h for the different |
| 2435 |
* options for the arguments. |
| 2436 |
*/ |
| 2437 |
|
| 2438 |
void writeWAISPacketHeader(char *header, |
| 2439 |
long dataLen, |
| 2440 |
long type, |
| 2441 |
char *server, |
| 2442 |
long compression, |
| 2443 |
long encoding, |
| 2444 |
long version) |
| 2445 |
/* Puts together the new wais before-the-z39-packet header. */ |
| 2446 |
{ |
| 2447 |
char lengthBuf[11]; |
| 2448 |
char serverBuf[11]; |
| 2449 |
|
| 2450 |
long serverLen = strlen(server); |
| 2451 |
|
| 2452 |
if (serverLen > 10) |
| 2453 |
serverLen = 10; |
| 2454 |
|
| 2455 |
sprintf(lengthBuf, "%010ld", dataLen); |
| 2456 |
StrNCpy(header, lengthBuf, 10); |
| 2457 |
|
| 2458 |
header[10] = type & 0xFF; |
| 2459 |
header[11] = version & 0xFF; |
| 2460 |
|
| 2461 |
StrNCpy(serverBuf, server, serverLen); |
| 2462 |
StrNCpy((char *) (header + 12), serverBuf, serverLen); |
| 2463 |
|
| 2464 |
header[22] = compression & 0xFF; |
| 2465 |
header[23] = encoding & 0xFF; |
| 2466 |
header[24] = '0'; /* checkSum(header + HEADER_LENGTH,dataLen); XXX the result must be ascii */ |
| 2467 |
} |
| 2468 |
|
| 2469 |
/*---------------------------------------------------------------------*/ |