SQLite

Check-in [1ac7a8d0af]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Fix the fts5 bm25() function so that it matches the documentation.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | fts5
Files: files | file ages | folders
SHA1: 1ac7a8d0af9a71ddf6a1421033dcb9fa67c6120c
User & Date: dan 2014-12-23 19:18:34.426
Context
2014-12-29
11:24
Fixes to built-in tokenizers. (check-in: b33fe0dd89 user: dan tags: fts5)
2014-12-23
19:18
Fix the fts5 bm25() function so that it matches the documentation. (check-in: 1ac7a8d0af user: dan tags: fts5)
2014-12-22
21:01
Fixes and simplifications for the snippet() and highlight() functions. (check-in: ca5d44042a user: dan tags: fts5)
Changes
Unified Diff Ignore Whitespace Patch
Changes to ext/fts5/fts5.c.
︙︙
1258
1259
1260
1261
1262
1263
1264






1265
1266
1267
1268

1269
1270
1271
1272
1273
1274
1275

  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  int rc = SQLITE_OK;

  if( CsrFlagTest(pCsr, FTS5CSR_REQUIRE_DOCSIZE) ){
    i64 iRowid = fts5CursorRowid(pCsr);
    rc = sqlite3Fts5StorageDocsize(pTab->pStorage, iRowid, pCsr->aColumnSize);
  }






  if( iCol>=0 && iCol<pTab->pConfig->nCol ){
    *pnToken = pCsr->aColumnSize[iCol];
  }else{
    *pnToken = 0;

  }
  return rc;
}

static int fts5ApiPoslist(
  Fts5Context *pCtx, 
  int iPhrase, 








>
>
>
>
>
>
|



>








1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282

  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  int rc = SQLITE_OK;

  if( CsrFlagTest(pCsr, FTS5CSR_REQUIRE_DOCSIZE) ){
    i64 iRowid = fts5CursorRowid(pCsr);
    rc = sqlite3Fts5StorageDocsize(pTab->pStorage, iRowid, pCsr->aColumnSize);
  }
  if( iCol<0 ){
    int i;
    *pnToken = 0;
    for(i=0; i<pTab->pConfig->nCol; i++){
      *pnToken += pCsr->aColumnSize[i];
    }
  }else if( iCol<pTab->pConfig->nCol ){
    *pnToken = pCsr->aColumnSize[iCol];
  }else{
    *pnToken = 0;
    rc = SQLITE_RANGE;
  }
  return rc;
}

static int fts5ApiPoslist(
  Fts5Context *pCtx, 
  int iPhrase,
︙︙
Changes to ext/fts5/fts5.h.
︙︙
44
45
46
47
48
49
50
51
52



53
54
55




56
57
58
59
60
61
62

/*
** EXTENSION API FUNCTIONS
**
** xUserData(pFts):
**   Return a copy of the context pointer the extension function was 
**   registered with.
**
**
** xColumnTotalSize(pFts, iCol, pnToken):



**   Returns the total number of tokens in column iCol, considering all
**   rows in the FTS5 table.
**




**
** xColumnCount:
**   Returns the number of columns in the FTS5 table.
**
** xColumnSize:
**   Reports the size in tokens of a column value from the current row.
**








<

>
>
>
|
|

>
>
>
>








44
45
46
47
48
49
50

51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68

/*
** EXTENSION API FUNCTIONS
**
** xUserData(pFts):
**   Return a copy of the context pointer the extension function was 
**   registered with.
**

** xColumnTotalSize(pFts, iCol, pnToken):
**   If parameter iCol is less than zero, set output variable *pnToken
**   to the total number of tokens in the FTS5 table. Or, if iCol is
**   non-negative but less than the number of columns in the table, return
**   the total number of tokens in column iCol, considering all rows in 
**   the FTS5 table.
**
**   If parameter iCol is greater than or equal to the number of columns
**   in the table, SQLITE_RANGE is returned. Or, if an error occurs (e.g.
**   an OOM condition or IO error), an appropriate SQLite error code is 
**   returned.
**
** xColumnCount:
**   Returns the number of columns in the FTS5 table.
**
** xColumnSize:
**   Reports the size in tokens of a column value from the current row.
**
︙︙
Changes to ext/fts5/fts5_aux.c.
︙︙
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128


/*************************************************************************
** Start of highlight() implementation.
*/
typedef struct HighlightContext HighlightContext;
struct HighlightContext {
  CInstIter iter;                 /* Coalesced Instance Iterator */
  int iRangeStart;
  int iRangeEnd;
  const char *zOpen;              /* Opening highlight */
  const char *zClose;             /* Closing highlight */
  const char *zIn;                /* Input text */
  int nIn;                        /* Size of input text in bytes */
  int iOff;                       /* Current offset within zIn[] */
  char *zOut;                     /* Output value */
};








|
|








113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128


/*************************************************************************
** Start of highlight() implementation.
*/
typedef struct HighlightContext HighlightContext;
struct HighlightContext {
  CInstIter iter;                 /* Coalesced Instance Iterator */
  int iRangeStart;                /* First token to include */
  int iRangeEnd;                  /* If non-zero, last token to include */
  const char *zOpen;              /* Opening highlight */
  const char *zClose;             /* Closing highlight */
  const char *zIn;                /* Input text */
  int nIn;                        /* Size of input text in bytes */
  int iOff;                       /* Current offset within zIn[] */
  char *zOut;                     /* Output value */
};
︙︙
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174

  int iPos                        /* Position offset of token */
){
  HighlightContext *p = (HighlightContext*)pContext;
  int rc = SQLITE_OK;

  if( p->iRangeEnd>0 ){
    if( iPos<p->iRangeStart || iPos>p->iRangeEnd ) return SQLITE_OK;
    if( iPos==p->iRangeStart ) p->iOff = iStartOff;
  }

  if( iPos==p->iter.iStart ){
    fts5HighlightAppend(&rc, p, &p->zIn[p->iOff], iStartOff - p->iOff);
    fts5HighlightAppend(&rc, p, p->zOpen, -1);
    p->iOff = iStartOff;
  }








|








160
161
162
163
164
165
166
167
168
169
170
171
172
173
174

  int iPos                        /* Position offset of token */
){
  HighlightContext *p = (HighlightContext*)pContext;
  int rc = SQLITE_OK;

  if( p->iRangeEnd>0 ){
    if( iPos<p->iRangeStart || iPos>p->iRangeEnd ) return SQLITE_OK;
    if( p->iRangeStart && iPos==p->iRangeStart ) p->iOff = iStartOff;
  }

  if( iPos==p->iter.iStart ){
    fts5HighlightAppend(&rc, p, &p->zIn[p->iOff], iStartOff - p->iOff);
    fts5HighlightAppend(&rc, p, p->zOpen, -1);
    p->iOff = iStartOff;
  }
︙︙
235
236
237
238
239
240
241

242
243
244


245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276

277
278
279
280

281
282
283
284
285
286
287

    sqlite3_result_text(pCtx, (const char*)ctx.zOut, -1, SQLITE_TRANSIENT);
  }else{
    sqlite3_result_error_code(pCtx, rc);
  }
  sqlite3_free(ctx.zOut);
}
/*

**************************************************************************/




static void fts5SnippetFunction(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  sqlite3_context *pCtx,          /* Context for returning result/error */
  int nVal,                       /* Number of values in apVal[] array */
  sqlite3_value **apVal           /* Array of trailing arguments */
){
  HighlightContext ctx;
  int rc = SQLITE_OK;             /* Return code */
  int iCol;                       /* 1st argument to snippet() */
  const char *zEllips;            /* 4th argument to snippet() */
  int nToken;                     /* 5th argument to snippet() */
  int nInst;                      /* Number of instance matches this row */
  int i;                          /* Used to iterate through instances */
  int nPhrase;                    /* Number of phrases in query */
  unsigned char *aSeen;           /* Array of "seen instance" flags */
  int iBestCol;                   /* Column containing best snippet */
  int iBestStart = 0;             /* First token of best snippet */
  int iBestLast = nToken;         /* Last token of best snippet */
  int nBestScore = 0;             /* Score of best snippet */
  int nColSize;                   /* Total size of iBestCol in tokens */

  if( nVal!=5 ){
    const char *zErr = "wrong number of arguments to function snippet()";
    sqlite3_result_error(pCtx, zErr, -1);
    return;
  }

  memset(&ctx, 0, sizeof(HighlightContext));
  rc = pApi->xColumnText(pFts, iCol, &ctx.zIn, &ctx.nIn);

  iCol = sqlite3_value_int(apVal[0]);

  ctx.zOpen = (const char*)sqlite3_value_text(apVal[1]);
  ctx.zClose = (const char*)sqlite3_value_text(apVal[2]);
  zEllips = (const char*)sqlite3_value_text(apVal[3]);
  nToken = sqlite3_value_int(apVal[4]);


  iBestCol = (iCol>=0 ? iCol : 0);
  nPhrase = pApi->xPhraseCount(pFts);
  aSeen = sqlite3_malloc(nPhrase);
  if( aSeen==0 ){
    rc = SQLITE_NOMEM;
  }








>


|
>
>


















|










<
<

>




>








235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276


277
278
279
280
281
282
283
284
285
286
287
288
289
290

    sqlite3_result_text(pCtx, (const char*)ctx.zOut, -1, SQLITE_TRANSIENT);
  }else{
    sqlite3_result_error_code(pCtx, rc);
  }
  sqlite3_free(ctx.zOut);
}
/*
** End of highlight() implementation.
**************************************************************************/

/*
** Implementation of snippet() function.
*/
static void fts5SnippetFunction(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  sqlite3_context *pCtx,          /* Context for returning result/error */
  int nVal,                       /* Number of values in apVal[] array */
  sqlite3_value **apVal           /* Array of trailing arguments */
){
  HighlightContext ctx;
  int rc = SQLITE_OK;             /* Return code */
  int iCol;                       /* 1st argument to snippet() */
  const char *zEllips;            /* 4th argument to snippet() */
  int nToken;                     /* 5th argument to snippet() */
  int nInst;                      /* Number of instance matches this row */
  int i;                          /* Used to iterate through instances */
  int nPhrase;                    /* Number of phrases in query */
  unsigned char *aSeen;           /* Array of "seen instance" flags */
  int iBestCol;                   /* Column containing best snippet */
  int iBestStart = 0;             /* First token of best snippet */
  int iBestLast;                  /* Last token of best snippet */
  int nBestScore = 0;             /* Score of best snippet */
  int nColSize;                   /* Total size of iBestCol in tokens */

  if( nVal!=5 ){
    const char *zErr = "wrong number of arguments to function snippet()";
    sqlite3_result_error(pCtx, zErr, -1);
    return;
  }

  memset(&ctx, 0, sizeof(HighlightContext));


  iCol = sqlite3_value_int(apVal[0]);
  rc = pApi->xColumnText(pFts, iCol, &ctx.zIn, &ctx.nIn);
  ctx.zOpen = (const char*)sqlite3_value_text(apVal[1]);
  ctx.zClose = (const char*)sqlite3_value_text(apVal[2]);
  zEllips = (const char*)sqlite3_value_text(apVal[3]);
  nToken = sqlite3_value_int(apVal[4]);
  iBestLast = nToken-1;

  iBestCol = (iCol>=0 ? iCol : 0);
  nPhrase = pApi->xPhraseCount(pFts);
  aSeen = sqlite3_malloc(nPhrase);
  if( aSeen==0 ){
    rc = SQLITE_NOMEM;
  }
︙︙
359
360
361
362
363
364
365
366
367

368
369
370

371
372
373
374

375
376
377
378

379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466

467
468
469
470
471


472

473

474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490

491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530

531
532

533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597

598




599


600
601
602
603
604
605
606
607

608

609
610
611
612
613
614


615
616
617
618
619
620
621
622
623
624
625

626
627
628
629
630
631
632
633
634
635


636

637
638
639
640
641
642
643
644

645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679

  }
  sqlite3_free(ctx.zOut);
  sqlite3_free(aSeen);
}

/************************************************************************/


/*

** Context object passed by fts5GatherTotals() to xQueryPhrase callback
** fts5GatherCallback().
*/

struct Fts5GatherCtx {
  int nCol;                       /* Number of columns in FTS table */
  int iPhrase;                    /* Phrase currently under investigation */
  int *anVal;                     /* Array to populate */

};

/*
** Callback used by fts5GatherTotals() with the xQueryPhrase() API.

*/
static int fts5GatherCallback(
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts,
  void *pUserData                 /* Pointer to Fts5GatherCtx object */
){
  struct Fts5GatherCtx *p = (struct Fts5GatherCtx*)pUserData;
  int i = 0;
  int iPrev = -1;
  i64 iPos = 0;

  while( 0==pApi->xPoslist(pFts, 0, &i, &iPos) ){
    int iCol = FTS5_POS2COLUMN(iPos);
    if( iCol!=iPrev ){
      p->anVal[p->iPhrase * p->nCol + iCol]++;
      iPrev = iCol;
    }
  }

  return SQLITE_OK;
}

/*
** This function returns a pointer to an array of integers containing entries
** indicating the number of rows in the table for which each phrase features 
** at least once in each column.
**
** If nCol is the number of matchable columns in the table, and nPhrase is
** the number of phrases in the query, the array contains a total of
** (nPhrase*nCol) entries.
**
** For phrase iPhrase and column iCol:
**
**   anVal[iPhrase * nCol + iCol]
**
** is set to the number of rows in the table for which column iCol contains 
** at least one instance of phrase iPhrase.
*/
static int fts5GatherTotals(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  int **panVal
){
  int rc = SQLITE_OK;
  int *anVal = 0;
  int i;                          /* For iterating through expression phrases */
  int nPhrase = pApi->xPhraseCount(pFts);
  int nCol = pApi->xColumnCount(pFts);
  int nByte = nCol * nPhrase * sizeof(int);
  struct Fts5GatherCtx sCtx;

  sCtx.nCol = nCol;
  anVal = sCtx.anVal = (int*)sqlite3_malloc(nByte);
  if( anVal==0 ){
    rc = SQLITE_NOMEM;
  }else{
    memset(anVal, 0, nByte);
  }

  for(i=0; i<nPhrase && rc==SQLITE_OK; i++){
    sCtx.iPhrase = i;
    rc = pApi->xQueryPhrase(pFts, i, (void*)&sCtx, fts5GatherCallback);
  }

  if( rc!=SQLITE_OK ){
    sqlite3_free(anVal);
    anVal = 0;
  }

  *panVal = anVal;
  return rc;
}

typedef struct Fts5Bm25Context Fts5Bm25Context;
struct Fts5Bm25Context {
  int nPhrase;                    /* Number of phrases in query */
  int nCol;                       /* Number of columns in FTS table */
  double *aIDF;                   /* Array of IDF values */
  double *aAvg;                   /* Average size of each column in tokens */
};

static int fts5Bm25GetContext(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  Fts5Bm25Context **pp            /* OUT: Context object */
){
  Fts5Bm25Context *p;
  int rc = SQLITE_OK;


  p = pApi->xGetAuxdata(pFts, 0);
  if( p==0 ){
    int *anVal = 0;
    int ic;                       /* For iterating through columns */


    int ip;                       /* For iterating through phrases */

    i64 nRow;                     /* Total number of rows in table */

    int nPhrase = pApi->xPhraseCount(pFts);
    int nCol = pApi->xColumnCount(pFts);
    int nByte = sizeof(Fts5Bm25Context) 
              + sizeof(double) * nPhrase * nCol       /* aIDF[] */
              + sizeof(double) * nCol;                /* aAvg[] */

    p = (Fts5Bm25Context*)sqlite3_malloc(nByte);
    if( p==0 ){
      rc = SQLITE_NOMEM;
    }else{
      memset(p, 0, nByte);
      p->aAvg = (double*)&p[1];
      p->aIDF = (double*)&p->aAvg[nCol];
      p->nCol = nCol;
      p->nPhrase = nPhrase;
    }


    if( rc==SQLITE_OK ){
      rc = pApi->xRowCount(pFts, &nRow); 
      assert( nRow>0 || rc!=SQLITE_OK );
      if( nRow<2 ) nRow = 2;
    }

    for(ic=0; rc==SQLITE_OK && ic<nCol; ic++){
      i64 nToken = 0;
      rc = pApi->xColumnTotalSize(pFts, ic, &nToken);
      p->aAvg[ic] = (double)nToken / (double)nRow;
    }

    if( rc==SQLITE_OK ){
      rc = fts5GatherTotals(pApi, pFts, &anVal);
    }
    for(ic=0; ic<nCol; ic++){
      for(ip=0; rc==SQLITE_OK && ip<nPhrase; ip++){
        /* Calculate the IDF (Inverse Document Frequency) for phrase ip
        ** in column ic. This is done using the standard BM25 formula as
        ** found on wikipedia:
        **
        **   IDF = log( (N - nHit + 0.5) / (nHit + 0.5) )
        **
        ** where "N" is the total number of documents in the set and nHit
        ** is the number that contain at least one instance of the phrase
        ** under consideration.
        **
        ** The problem with this is that if (N < 2*nHit), the IDF is 
        ** negative. Which is undesirable. So the mimimum allowable IDF is
        ** (1e-6) - roughly the same as a term that appears in just over
        ** half of set of 5,000,000 documents.  */
        int idx = ip * nCol + ic; /* Index in aIDF[] and anVal[] arrays */
        int nHit = anVal[idx];    /* Number of docs matching "ic: ip" */

        p->aIDF[idx] = log( (0.5 + nRow - nHit) / (0.5 + nHit) );
        if( p->aIDF[idx]<=0.0 ) p->aIDF[idx] = 1e-6;
        assert( p->aIDF[idx]>=0.0 );
      }
    }


    sqlite3_free(anVal);
    if( rc==SQLITE_OK ){

      rc = pApi->xSetAuxdata(pFts, p, sqlite3_free);
    }
    if( rc!=SQLITE_OK ){
      sqlite3_free(p);
      p = 0;
    }
  }

  *pp = p;
  return rc;
}

static void fts5Bm25DebugContext(
  int *pRc,                       /* IN/OUT: Return code */
  Fts5Buffer *pBuf,               /* Buffer to populate */
  Fts5Bm25Context *p              /* Context object to decode */
){
  int ip;
  int ic;

  sqlite3Fts5BufferAppendString(pRc, pBuf, "idf ");
  if( p->nPhrase>1 || p->nCol>1 ){
    sqlite3Fts5BufferAppendString(pRc, pBuf, "{");
  }
  for(ip=0; ip<p->nPhrase; ip++){
    if( ip>0 ) sqlite3Fts5BufferAppendString(pRc, pBuf, " ");
    if( p->nCol>1 ) sqlite3Fts5BufferAppendString(pRc, pBuf, "{");
    for(ic=0; ic<p->nCol; ic++){
      if( ic>0 ) sqlite3Fts5BufferAppendString(pRc, pBuf, " ");
      sqlite3Fts5BufferAppendPrintf(pRc, pBuf, "%f", p->aIDF[ip*p->nCol+ic]);
    }
    if( p->nCol>1 ) sqlite3Fts5BufferAppendString(pRc, pBuf, "}");
  }
  if( p->nPhrase>1 || p->nCol>1 ){
    sqlite3Fts5BufferAppendString(pRc, pBuf, "}");
  }

  sqlite3Fts5BufferAppendString(pRc, pBuf, " avgdl ");
  if( p->nCol>1 ) sqlite3Fts5BufferAppendString(pRc, pBuf, "{");
  for(ic=0; ic<p->nCol; ic++){
    if( ic>0 ) sqlite3Fts5BufferAppendString(pRc, pBuf, " ");
    sqlite3Fts5BufferAppendPrintf(pRc, pBuf, "%f", p->aAvg[ic]);
  }
  if( p->nCol>1 ) sqlite3Fts5BufferAppendString(pRc, pBuf, "}");
}

static void fts5Bm25DebugRow(
  int *pRc, 
  Fts5Buffer *pBuf, 
  Fts5Bm25Context *p, 
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts
){
}

static void fts5Bm25Function(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  sqlite3_context *pCtx,          /* Context for returning result/error */
  int nVal,                       /* Number of values in apVal[] array */
  sqlite3_value **apVal           /* Array of trailing arguments */
){
  const double k1 = 1.2;
  const double B = 0.75;
  int rc = SQLITE_OK;

  Fts5Bm25Context *p;







  rc = fts5Bm25GetContext(pApi, pFts, &p);

  if( rc==SQLITE_OK ){
    /* If the bDebug flag is set, instead of returning a numeric rank, this
    ** function returns a text value showing how the rank is calculated. */
    Fts5Buffer debug;
    int bDebug = (pApi->xUserData(pFts)!=0);
    memset(&debug, 0, sizeof(Fts5Buffer));



    int ip;
    double score = 0.0;

    if( bDebug ){
      fts5Bm25DebugContext(&rc, &debug, p);
      fts5Bm25DebugRow(&rc, &debug, p, pApi, pFts);


    }

    for(ip=0; rc==SQLITE_OK && ip<p->nPhrase; ip++){
      int iPrev = 0;
      int nHit = 0;
      int i = 0;
      i64 iPos = 0;

      while( rc==SQLITE_OK ){
        int bDone = pApi->xPoslist(pFts, ip, &i, &iPos);
        int iCol = FTS5_POS2COLUMN(iPos);

        if( (iCol!=iPrev || bDone) && nHit>0 ){
          int sz = 0;
          int idx = ip * p->nCol + iPrev;
          double bm25;
          rc = pApi->xColumnSize(pFts, iPrev, &sz);

          bm25 = (p->aIDF[idx] * nHit * (k1+1.0)) /
            (nHit + k1 * (1.0 - B + B * sz / p->aAvg[iPrev]));




          score = score + bm25;

          nHit = 0;
        }
        if( bDone ) break;
        nHit++;
        iPrev = iCol;
      }
    }


    if( rc==SQLITE_OK ){
      if( bDebug ){
        sqlite3_result_text(pCtx, (const char*)debug.p, -1, SQLITE_TRANSIENT);
      }else{
        sqlite3_result_double(pCtx, score);
      }
    }
    sqlite3_free(debug.p);
  }

  if( rc!=SQLITE_OK ){
    sqlite3_result_error_code(pCtx, rc);
  }
}

int sqlite3Fts5AuxInit(fts5_api *pApi){
  struct Builtin {
    const char *zFunc;            /* Function name (nul-terminated) */
    void *pUserData;              /* User-data pointer */
    fts5_extension_function xFunc;/* Callback function */
    void (*xDestroy)(void*);      /* Destructor function */
  } aBuiltin [] = {
    { "bm25debug", (void*)1, fts5Bm25Function,    0 },
    { "snippet",   0, fts5SnippetFunction, 0 },
    { "highlight", 0, fts5HighlightFunction, 0 },
    { "bm25",      0, fts5Bm25Function,    0 },
  };

  int rc = SQLITE_OK;             /* Return code */
  int i;                          /* To iterate through builtin functions */

  for(i=0; rc==SQLITE_OK && i<sizeof(aBuiltin)/sizeof(aBuiltin[0]); i++){
    rc = pApi->xCreateFunction(pApi,
        aBuiltin[i].zFunc,
        aBuiltin[i].pUserData,








<

>
|
<

>
|
|
|
|
>



|
>

|


|

<
<
<
<
|
<
<
<
<
<
<
<
|




|
|
<
|
<
<
<
<
<
<
<
<
<
<

|
|
|
<
<
<
<
<
<
<
<
<
|
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<

<
|
>



<
|
>
>
|
>
|
>
|
<
|
<
<
<
|




|
|
<
|


>
|
|
|
<
|
|
|
|
<
<
<
|
|
<
<
<
<
|
|
<











<
<
|
<
|
|



>
|
<
>


|
<
<
|
<
<
|



<
<
<
<
<
<
<
|
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
|
<
<
<
<
<
<
<
<
|
<
<
<
<
<
<
<
<
<







|
|
|
>
|
>
>
>
>

>
>
|
<

<
<
<
|
|
>
|
>
|
<
|
|
<
<
>
>

|
<
<
<
<
<
<
<
<
|
>
|
|
<
<
|
|
<
<
|

>
>
|
>
|
<
|
<
<
|
|
|
>
|
<
<
<
|
|
<
<
<
<
<











<




<








362
363
364
365
366
367
368

369
370
371

372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389




390







391
392
393
394
395
396
397

398










399
400
401
402









403


































404

405
406
407
408
409

410
411
412
413
414
415
416
417

418



419
420
421
422
423
424
425

426
427
428
429
430
431
432

433
434
435
436



437
438




439
440

441
442
443
444
445
446
447
448
449
450
451


452

453
454
455
456
457
458
459

460
461
462
463


464


465
466
467
468







469
















470








471









472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491

492



493
494
495
496
497
498

499
500


501
502
503
504








505
506
507
508


509
510


511
512
513
514
515
516
517

518


519
520
521
522
523



524
525





526
527
528
529
530
531
532
533
534
535
536

537
538
539
540

541
542
543
544
545
546
547

  }
  sqlite3_free(ctx.zOut);
  sqlite3_free(aSeen);
}

/************************************************************************/


/*
** The first time the bm25() function is called for a query, an instance
** of the following structure is allocated and populated.

*/
typedef struct Fts5Bm25Data Fts5Bm25Data;
struct Fts5Bm25Data {
  int nPhrase;                    /* Number of phrases in query */
  double avgdl;                   /* Average number of tokens in each row */
  double *aIDF;                   /* IDF for each phrase */
  double *aFreq;                  /* Array used to calculate phrase freq. */
};

/*
** Callback used by fts5Bm25GetData() to count the number of rows in the
** table matched by each individual phrase within the query.
*/
static int fts5CountCb(
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts,
  void *pUserData                 /* Pointer to sqlite3_int64 variable */
){




  sqlite3_int64 *pn = (sqlite3_int64*)pUserData;







  (*pn)++;
  return SQLITE_OK;
}

/*
** Set *ppData to point to the Fts5Bm25Data object for the current query. 
** If the object has not already been allocated, allocate and populate it

** now.










*/
static int fts5Bm25GetData(
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts,









  Fts5Bm25Data **ppData           /* OUT: bm25-data object for this query */


































){

  int rc = SQLITE_OK;             /* Return code */
  Fts5Bm25Data *p;                /* Object to return */

  p = pApi->xGetAuxdata(pFts, 0);
  if( p==0 ){

    int nPhrase;                  /* Number of phrases in query */
    sqlite3_int64 nRow;           /* Number of rows in table */
    sqlite3_int64 nToken;         /* Number of tokens in table */
    int nByte;                    /* Bytes of space to allocate */
    int i;

    /* Allocate the Fts5Bm25Data object */
    nPhrase = pApi->xPhraseCount(pFts);

    nByte = sizeof(Fts5Bm25Data) + nPhrase*2*sizeof(double);



    p = (Fts5Bm25Data*)sqlite3_malloc(nByte);
    if( p==0 ){
      rc = SQLITE_NOMEM;
    }else{
      memset(p, 0, nByte);
      p->nPhrase = nPhrase;
      p->aIDF = (double*)&p[1];

      p->aFreq = &p->aIDF[nPhrase];
    }

    /* Calculate the average document length for this FTS5 table */
    if( rc==SQLITE_OK ) rc = pApi->xRowCount(pFts, &nRow);
    if( rc==SQLITE_OK ) rc = pApi->xColumnTotalSize(pFts, -1, &nToken);
    if( rc==SQLITE_OK ) p->avgdl = (double)nToken  / (double)nRow;


    /* Calculate an IDF for each phrase in the query */
    for(i=0; rc==SQLITE_OK && i<nPhrase; i++){
      sqlite3_int64 nHit = 0;



      rc = pApi->xQueryPhrase(pFts, i, (void*)&nHit, fts5CountCb);
      if( rc==SQLITE_OK ){




        /* Calculate the IDF (Inverse Document Frequency) for phrase i.
        ** This is done using the standard BM25 formula as found on wikipedia:

        **
        **   IDF = log( (N - nHit + 0.5) / (nHit + 0.5) )
        **
        ** where "N" is the total number of documents in the set and nHit
        ** is the number that contain at least one instance of the phrase
        ** under consideration.
        **
        ** The problem with this is that if (N < 2*nHit), the IDF is 
        ** negative. Which is undesirable. So the mimimum allowable IDF is
        ** (1e-6) - roughly the same as a term that appears in just over
        ** half of set of 5,000,000 documents.  */


        double idf = log( (nRow - nHit + 0.5) / (nHit + 0.5) );

        if( idf<=0.0 ) idf = 1e-6;
        p->aIDF[i] = idf;
      }
    }

    if( rc!=SQLITE_OK ){
      sqlite3_free(p);

    }else{
      rc = pApi->xSetAuxdata(pFts, p, sqlite3_free);
    }
    if( rc!=SQLITE_OK ) p = 0;


  }


  *ppData = p;
  return rc;
}








/*
















** Implementation of bm25() function.








*/









static void fts5Bm25Function(
  const Fts5ExtensionApi *pApi,   /* API offered by current FTS version */
  Fts5Context *pFts,              /* First arg to pass to pApi functions */
  sqlite3_context *pCtx,          /* Context for returning result/error */
  int nVal,                       /* Number of values in apVal[] array */
  sqlite3_value **apVal           /* Array of trailing arguments */
){
  const double k1 = 1.2;          /* Constant "k1" from BM25 formula */
  const double b = 0.75;          /* Constant "b" from BM25 formula */
  int rc = SQLITE_OK;             /* Error code */
  double score = 0.0;             /* SQL function return value */
  Fts5Bm25Data *pData;            /* Values allocated/calculated once only */
  int i;                          /* Iterator variable */
  int nInst;                      /* Value returned by xInstCount() */
  double D;                       /* Total number of tokens in row */
  double *aFreq;                  /* Array of phrase freq. for current row */

  /* Calculate the phrase frequency (symbol "f(qi,D)" in the documentation)
  ** for each phrase in the query for the current row. */
  rc = fts5Bm25GetData(pApi, pFts, &pData);

  if( rc==SQLITE_OK ){



    aFreq = pData->aFreq;
    memset(aFreq, 0, sizeof(double) * pData->nPhrase);
    rc = pApi->xInstCount(pFts, &nInst);
  }
  for(i=0; rc==SQLITE_OK && i<nInst; i++){
    int ip; int ic; int io;

    rc = pApi->xInst(pFts, i, &ip, &ic, &io);
    if( rc==SQLITE_OK ){


      double w = (nVal > ic) ? sqlite3_value_double(apVal[ic]) : 1.0;
      aFreq[ip] += w;
    }
  }









  /* Figure out the total size of the current row in tokens. */
  if( rc==SQLITE_OK ){
    int nTok;


    rc = pApi->xColumnSize(pFts, -1, &nTok);
    D = (double)nTok;


  }

  /* Determine the BM25 score for the current row. */
  for(i=0; rc==SQLITE_OK && i<pData->nPhrase; i++){
    score += pData->aIDF[i] * (
      ( aFreq[i] * (k1 + 1.0) ) / 
      ( aFreq[i] + k1 * (1 - b + b * D / pData->avgdl) )

    );


  }
  
  /* If no error has occurred, return the calculated score. Otherwise,
  ** throw an SQL exception.  */
  if( rc==SQLITE_OK ){



    sqlite3_result_double(pCtx, score);
  }else{





    sqlite3_result_error_code(pCtx, rc);
  }
}

int sqlite3Fts5AuxInit(fts5_api *pApi){
  struct Builtin {
    const char *zFunc;            /* Function name (nul-terminated) */
    void *pUserData;              /* User-data pointer */
    fts5_extension_function xFunc;/* Callback function */
    void (*xDestroy)(void*);      /* Destructor function */
  } aBuiltin [] = {

    { "snippet",   0, fts5SnippetFunction, 0 },
    { "highlight", 0, fts5HighlightFunction, 0 },
    { "bm25",      0, fts5Bm25Function,    0 },
  };

  int rc = SQLITE_OK;             /* Return code */
  int i;                          /* To iterate through builtin functions */

  for(i=0; rc==SQLITE_OK && i<sizeof(aBuiltin)/sizeof(aBuiltin[0]); i++){
    rc = pApi->xCreateFunction(pApi,
        aBuiltin[i].zFunc,
        aBuiltin[i].pUserData,
︙︙
Changes to ext/fts5/fts5_storage.c.
︙︙
725
726
727
728
729
730
731







732



733
734
735
736
737
738
739

  }
  return rc;
}

int sqlite3Fts5StorageSize(Fts5Storage *p, int iCol, i64 *pnToken){
  int rc = fts5StorageLoadTotals(p, 0);
  if( rc==SQLITE_OK ){







    *pnToken = p->aTotalSize[iCol];



  }
  return rc;
}

int sqlite3Fts5StorageRowCount(Fts5Storage *p, i64 *pnRow){
  int rc = fts5StorageLoadTotals(p, 0);
  if( rc==SQLITE_OK ){








>
>
>
>
>
>
>
|
>
>
>








725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749

  }
  return rc;
}

int sqlite3Fts5StorageSize(Fts5Storage *p, int iCol, i64 *pnToken){
  int rc = fts5StorageLoadTotals(p, 0);
  if( rc==SQLITE_OK ){
    *pnToken = 0;
    if( iCol<0 ){
      int i;
      for(i=0; i<p->pConfig->nCol; i++){
        *pnToken += p->aTotalSize[i];
      }
    }else if( iCol<p->pConfig->nCol ){
      *pnToken = p->aTotalSize[iCol];
    }else{
      rc = SQLITE_RANGE;
    }
  }
  return rc;
}

int sqlite3Fts5StorageRowCount(Fts5Storage *p, i64 *pnRow){
  int rc = fts5StorageLoadTotals(p, 0);
  if( rc==SQLITE_OK ){
︙︙