SQLite

Check-in [8c5b9fedfc]
Login

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

Overview
Comment:More optimization for the 'y' and 'b' matchinfo directives.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | fts3-matchinfo-y
Files: files | file ages | folders
SHA1: 8c5b9fedfcee3ac22a222819dceb981ad94a9903
User & Date: dan 2015-05-06 17:41:19.953
Context
2015-05-06
17:51
Remove some dead code from fts3_snippet.c. (check-in: 46b2d3cef5 user: dan tags: fts3-matchinfo-y)
17:41
More optimization for the 'y' and 'b' matchinfo directives. (check-in: 8c5b9fedfc user: dan tags: fts3-matchinfo-y)
08:43
Further optimizations for the 'y' and 'b' matchinfo operators. (check-in: fbd038bb57 user: dan tags: fts3-matchinfo-y)
Changes
Unified Diff Ignore Whitespace Patch
Changes to ext/fts3/fts3Int.h.
424
425
426
427
428
429
430


431
432
433
434
435
436
437
438

  /* The following are used by the fts3_eval.c module. */
  sqlite3_int64 iDocid;      /* Current docid */
  u8 bEof;                   /* True this expression is at EOF already */
  u8 bStart;                 /* True if iDocid is valid */
  u8 bDeferred;              /* True if this expression is entirely deferred */



  u32 *aMI;
};

/*
** Candidate values for Fts3Query.eType. Note that the order of the first
** four values is in order of precedence when parsing expressions. For 
** example, the following:
**







>
>
|







424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440

  /* The following are used by the fts3_eval.c module. */
  sqlite3_int64 iDocid;      /* Current docid */
  u8 bEof;                   /* True this expression is at EOF already */
  u8 bStart;                 /* True if iDocid is valid */
  u8 bDeferred;              /* True if this expression is entirely deferred */

  /* The following are used by the fts3_snippet.c module. */
  int iPhrase;               /* Index of this phrase in matchinfo() results */
  u32 *aMI;                  /* See above */
};

/*
** Candidate values for Fts3Query.eType. Note that the order of the first
** four values is in order of precedence when parsing expressions. For 
** example, the following:
**
Changes to ext/fts3/fts3_snippet.c.
283
284
285
286
287
288
289

290
291
292
293
294
295
296
  Fts3Expr *pExpr,                /* Expression to iterate phrases of */
  int (*x)(Fts3Expr*,int,void*),  /* Callback function to invoke for phrases */
  void *pCtx                      /* Second argument to pass to callback */
){
  int iPhrase = 0;                /* Variable used as the phrase counter */
  return fts3ExprIterate2(pExpr, &iPhrase, x, pCtx);
}


/*
** This is an fts3ExprIterate() callback used while loading the doclists
** for each phrase into Fts3Expr.aDoclist[]/nDoclist. See also
** fts3ExprLoadDoclists().
*/
static int fts3ExprLoadDoclistsCb(Fts3Expr *pExpr, int iPhrase, void *ctx){







>







283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
  Fts3Expr *pExpr,                /* Expression to iterate phrases of */
  int (*x)(Fts3Expr*,int,void*),  /* Callback function to invoke for phrases */
  void *pCtx                      /* Second argument to pass to callback */
){
  int iPhrase = 0;                /* Variable used as the phrase counter */
  return fts3ExprIterate2(pExpr, &iPhrase, x, pCtx);
}


/*
** This is an fts3ExprIterate() callback used while loading the doclists
** for each phrase into Fts3Expr.aDoclist[]/nDoclist. See also
** fts3ExprLoadDoclists().
*/
static int fts3ExprLoadDoclistsCb(Fts3Expr *pExpr, int iPhrase, void *ctx){
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
  if( pnPhrase ) *pnPhrase = sCtx.nPhrase;
  if( pnToken ) *pnToken = sCtx.nToken;
  return rc;
}

static int fts3ExprPhraseCountCb(Fts3Expr *pExpr, int iPhrase, void *ctx){
  (*(int *)ctx)++;
  UNUSED_PARAMETER(pExpr);
  UNUSED_PARAMETER(iPhrase);
  return SQLITE_OK;
}
static int fts3ExprPhraseCount(Fts3Expr *pExpr){
  int nPhrase = 0;
  (void)fts3ExprIterate(pExpr, fts3ExprPhraseCountCb, (void *)&nPhrase);
  return nPhrase;
}







<
|







329
330
331
332
333
334
335

336
337
338
339
340
341
342
343
  if( pnPhrase ) *pnPhrase = sCtx.nPhrase;
  if( pnToken ) *pnToken = sCtx.nToken;
  return rc;
}

static int fts3ExprPhraseCountCb(Fts3Expr *pExpr, int iPhrase, void *ctx){
  (*(int *)ctx)++;

  pExpr->iPhrase = iPhrase;
  return SQLITE_OK;
}
static int fts3ExprPhraseCount(Fts3Expr *pExpr){
  int nPhrase = 0;
  (void)fts3ExprIterate(pExpr, fts3ExprPhraseCountCb, (void *)&nPhrase);
  return nPhrase;
}
850
851
852
853
854
855
856






















































857
858
859
860
861
862
863
    c = *pEnd++ & 0x80;
    if( !c ) nEntry++;
  }

  *ppCollist = pEnd;
  return nEntry;
}























































/*
** fts3ExprIterate() callback used to collect the "global" matchinfo stats
** for a single query. 
**
** fts3ExprIterate() callback to load the 'global' elements of a
** FTS3_MATCHINFO_HITS matchinfo array. The global stats are those elements 







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>







850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
    c = *pEnd++ & 0x80;
    if( !c ) nEntry++;
  }

  *ppCollist = pEnd;
  return nEntry;
}

/*
** This function gathers 'y' or 'b' data for a single phrase.
*/
static void fts3ExprLHits(
  Fts3Expr *pExpr,                /* Phrase expression node */
  MatchInfo *p                    /* Matchinfo context */
){
  Fts3Table *pTab = (Fts3Table *)p->pCursor->base.pVtab;
  int iStart;
  Fts3Phrase *pPhrase = pExpr->pPhrase;
  char *pIter = pPhrase->doclist.pList;
  int iCol = 0;

  assert( p->flag==FTS3_MATCHINFO_LHITS_BM || p->flag==FTS3_MATCHINFO_LHITS );
  if( p->flag==FTS3_MATCHINFO_LHITS ){
    iStart = pExpr->iPhrase * p->nCol;
  }else{
    iStart = pExpr->iPhrase * ((p->nCol + 31) / 32);
  }

  while( 1 ){
    int nHit = fts3ColumnlistCount(&pIter);
    if( (pPhrase->iColumn>=pTab->nColumn || pPhrase->iColumn==iCol) ){
      if( p->flag==FTS3_MATCHINFO_LHITS ){
        p->aMatchinfo[iStart + iCol] = (u32)nHit;
      }else if( nHit ){
        p->aMatchinfo[iStart + (iCol+1)/32] |= (1 << (iCol&0x1F));
      }
    }
    assert( *pIter==0x00 || *pIter==0x01 );
    if( *pIter!=0x01 ) break;
    pIter++;
    pIter += fts3GetVarint32(pIter, &iCol);
  }
}

/*
** Gather the results for matchinfo directives 'y' and 'b'.
*/
static void fts3ExprLHitGather(
  Fts3Expr *pExpr,
  MatchInfo *p
){
  assert( (pExpr->pLeft==0)==(pExpr->pRight==0) );
  if( pExpr->bEof==0 && pExpr->iDocid==p->pCursor->iPrevId ){
    if( pExpr->pLeft ){
      fts3ExprLHitGather(pExpr->pLeft, p);
      fts3ExprLHitGather(pExpr->pRight, p);
    }else{
      fts3ExprLHits(pExpr, p);
    }
  }
}

/*
** fts3ExprIterate() callback used to collect the "global" matchinfo stats
** for a single query. 
**
** fts3ExprIterate() callback to load the 'global' elements of a
** FTS3_MATCHINFO_HITS matchinfo array. The global stats are those elements 
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
        }
        break;

      case FTS3_MATCHINFO_LHITS_BM:
      case FTS3_MATCHINFO_LHITS: {
        int nZero = fts3MatchinfoSize(pInfo, zArg[i]) * sizeof(u32);
        memset(pInfo->aMatchinfo, 0, nZero);
        (void)fts3ExprIterate(pCsr->pExpr, fts3ExprLHitsCb, (void*)pInfo);
        break;
      }

      default: {
        Fts3Expr *pExpr;
        assert( zArg[i]==FTS3_MATCHINFO_HITS );
        pExpr = pCsr->pExpr;







|







1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
        }
        break;

      case FTS3_MATCHINFO_LHITS_BM:
      case FTS3_MATCHINFO_LHITS: {
        int nZero = fts3MatchinfoSize(pInfo, zArg[i]) * sizeof(u32);
        memset(pInfo->aMatchinfo, 0, nZero);
        fts3ExprLHitGather(pCsr->pExpr, pInfo);
        break;
      }

      default: {
        Fts3Expr *pExpr;
        assert( zArg[i]==FTS3_MATCHINFO_HITS );
        pExpr = pCsr->pExpr;
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
}


/*
** Populate pCsr->aMatchinfo[] with data for the current row. The 
** 'matchinfo' data is an array of 32-bit unsigned integers (C type u32).
*/
static int fts3GetMatchinfo(
  sqlite3_context *pCtx,        /* Return results here */
  Fts3Cursor *pCsr,               /* FTS3 Cursor object */
  const char *zArg                /* Second argument to matchinfo() function */
){
  MatchInfo sInfo;
  Fts3Table *pTab = (Fts3Table *)pCsr->base.pVtab;
  int rc = SQLITE_OK;







|







1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
}


/*
** Populate pCsr->aMatchinfo[] with data for the current row. The 
** 'matchinfo' data is an array of 32-bit unsigned integers (C type u32).
*/
static void fts3GetMatchinfo(
  sqlite3_context *pCtx,        /* Return results here */
  Fts3Cursor *pCsr,               /* FTS3 Cursor object */
  const char *zArg                /* Second argument to matchinfo() function */
){
  MatchInfo sInfo;
  Fts3Table *pTab = (Fts3Table *)pCsr->base.pVtab;
  int rc = SQLITE_OK;
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
  /* If Fts3Cursor.pMIBuffer is NULL, then this is the first time the
  ** matchinfo function has been called for this query. In this case 
  ** allocate the array used to accumulate the matchinfo data and
  ** initialize those elements that are constant for every row.
  */
  if( pCsr->pMIBuffer==0 ){
    int nMatchinfo = 0;           /* Number of u32 elements in match-info */
    int nArg;                     /* Bytes in zArg */
    int i;                        /* Used to iterate through zArg */

    /* Determine the number of phrases in the query */
    pCsr->nPhrase = fts3ExprPhraseCount(pCsr->pExpr);
    sInfo.nPhrase = pCsr->nPhrase;

    /* Determine the number of integers in the buffer returned by this call. */







<







1389
1390
1391
1392
1393
1394
1395

1396
1397
1398
1399
1400
1401
1402
  /* If Fts3Cursor.pMIBuffer is NULL, then this is the first time the
  ** matchinfo function has been called for this query. In this case 
  ** allocate the array used to accumulate the matchinfo data and
  ** initialize those elements that are constant for every row.
  */
  if( pCsr->pMIBuffer==0 ){
    int nMatchinfo = 0;           /* Number of u32 elements in match-info */

    int i;                        /* Used to iterate through zArg */

    /* Determine the number of phrases in the query */
    pCsr->nPhrase = fts3ExprPhraseCount(pCsr->pExpr);
    sInfo.nPhrase = pCsr->nPhrase;

    /* Determine the number of integers in the buffer returned by this call. */
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
  if( rc!=SQLITE_OK ){
    sqlite3_result_error_code(pCtx, rc);
    if( xDestroyOut ) xDestroyOut(aOut);
  }else{
    int n = pCsr->pMIBuffer->nElem * sizeof(u32);
    sqlite3_result_blob(pCtx, aOut, n, xDestroyOut);
  }

  return rc;
}

/*
** Implementation of snippet() function.
*/
void sqlite3Fts3Snippet(
  sqlite3_context *pCtx,          /* SQLite function call context */







<
<







1437
1438
1439
1440
1441
1442
1443


1444
1445
1446
1447
1448
1449
1450
  if( rc!=SQLITE_OK ){
    sqlite3_result_error_code(pCtx, rc);
    if( xDestroyOut ) xDestroyOut(aOut);
  }else{
    int n = pCsr->pMIBuffer->nElem * sizeof(u32);
    sqlite3_result_blob(pCtx, aOut, n, xDestroyOut);
  }


}

/*
** Implementation of snippet() function.
*/
void sqlite3Fts3Snippet(
  sqlite3_context *pCtx,          /* SQLite function call context */
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
void sqlite3Fts3Matchinfo(
  sqlite3_context *pContext,      /* Function call context */
  Fts3Cursor *pCsr,               /* FTS3 table cursor */
  const char *zArg                /* Second arg to matchinfo() function */
){
  Fts3Table *pTab = (Fts3Table *)pCsr->base.pVtab;
  const char *zFormat;
  int rc;

  if( zArg ){
    zFormat = zArg;
  }else{
    zFormat = FTS3_MATCHINFO_DEFAULT;
  }

  if( !pCsr->pExpr ){
    sqlite3_result_blob(pContext, "", 0, SQLITE_STATIC);
    return;
  }else{
    /* Retrieve matchinfo() data. */
    rc = fts3GetMatchinfo(pContext, pCsr, zFormat);
    sqlite3Fts3SegmentsClose(pTab);
  }
}

#endif







<












|





1735
1736
1737
1738
1739
1740
1741

1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
void sqlite3Fts3Matchinfo(
  sqlite3_context *pContext,      /* Function call context */
  Fts3Cursor *pCsr,               /* FTS3 table cursor */
  const char *zArg                /* Second arg to matchinfo() function */
){
  Fts3Table *pTab = (Fts3Table *)pCsr->base.pVtab;
  const char *zFormat;


  if( zArg ){
    zFormat = zArg;
  }else{
    zFormat = FTS3_MATCHINFO_DEFAULT;
  }

  if( !pCsr->pExpr ){
    sqlite3_result_blob(pContext, "", 0, SQLITE_STATIC);
    return;
  }else{
    /* Retrieve matchinfo() data. */
    fts3GetMatchinfo(pContext, pCsr, zFormat);
    sqlite3Fts3SegmentsClose(pTab);
  }
}

#endif