/ Check-in [fbd038bb]
Login

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

Overview
Comment:Further optimizations for the 'y' and 'b' matchinfo operators.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | fts3-matchinfo-y
Files: files | file ages | folders
SHA1:fbd038bb57322e1ed2e1ee52f3d134594b6bfcc0
User & Date: dan 2015-05-06 08:43:26
Context
2015-05-06
17:41
More optimization for the 'y' and 'b' matchinfo directives. check-in: 8c5b9fed user: dan tags: fts3-matchinfo-y
08:43
Further optimizations for the 'y' and 'b' matchinfo operators. check-in: fbd038bb user: dan tags: fts3-matchinfo-y
2015-05-05
20:39
Add the fts3 matchinfo 'b' flag. check-in: b9b77972 user: dan tags: fts3-matchinfo-y
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to ext/fts3/fts3_snippet.c.

234
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
...
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
...
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
...
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
...
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
....
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
....
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
....
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
....
1344
1345
1346
1347
1348
1349
1350






1351
1352
1353
1354
1355
1356
1357
....
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
....
1677
1678
1679
1680
1681
1682
1683
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

1710
1711
1712
*/
static void fts3GetDeltaPosition(char **pp, int *piPos){
  int iVal;
  *pp += fts3GetVarint32(*pp, &iVal);
  *piPos += (iVal-2);
}



/*
** Helper function for fts3ExprIterate() (see below).
*/
static int fts3ExprIterate2(
  Fts3Expr *pExpr,                /* Expression to iterate phrases of */
  int bExcludeEof,
  int *piPhrase,                  /* Pointer to phrase counter */
  int (*x)(Fts3Expr*,int,void*),  /* Callback function to invoke for phrases */
  void *pCtx                      /* Second argument to pass to callback */
){
  int rc;                         /* Return code */

  if( bExcludeEof && pExpr->bEof ){
    rc = SQLITE_OK;
  }else{
    int eType = pExpr->eType;     /* Type of expression node pExpr */
    if( eType!=FTSQUERY_PHRASE ){
      assert( pExpr->pLeft && pExpr->pRight );
      rc = fts3ExprIterate2(pExpr->pLeft, bExcludeEof, piPhrase, x, pCtx);
      if( rc==SQLITE_OK && eType!=FTSQUERY_NOT ){
        rc = fts3ExprIterate2(pExpr->pRight, bExcludeEof, piPhrase, x, pCtx);
      }
    }else{
      rc = x(pExpr, *piPhrase, pCtx);
      (*piPhrase)++;
    }
  }
  return rc;
................................................................................
** If the callback function returns anything other than SQLITE_OK, 
** the iteration is abandoned and the error code returned immediately.
** Otherwise, SQLITE_OK is returned after a callback has been made for
** all eligible phrase nodes.
*/
static int fts3ExprIterate(
  Fts3Expr *pExpr,                /* Expression to iterate phrases of */
  int bExcludeEof,                /* Include nodes already at EOF */
  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, bExcludeEof, &iPhrase, x, pCtx);
}

/*
** This is an fts3ExprIterate() callback used while loading the doclists
** for each phrase into Fts3Expr.aDoclist[]/nDoclist. See also
** fts3ExprLoadDoclists().
*/
................................................................................
  Fts3Cursor *pCsr,               /* Fts3 cursor for current query */
  int *pnPhrase,                  /* OUT: Number of phrases in query */
  int *pnToken                    /* OUT: Number of tokens in query */
){
  int rc;                         /* Return Code */
  LoadDoclistCtx sCtx = {0,0,0};  /* Context for fts3ExprIterate() */
  sCtx.pCsr = pCsr;
  rc = fts3ExprIterate(pCsr->pExpr, 0, fts3ExprLoadDoclistsCb, (void *)&sCtx);
  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, 0, fts3ExprPhraseCountCb, (void *)&nPhrase);
  return nPhrase;
}

/*
** Advance the position list iterator specified by the first two 
** arguments so that it points to the first element with a value greater
** than or equal to parameter iNext.
................................................................................
  ** the set of phrases in the expression to populate the aPhrase[] array.
  */
  sIter.pCsr = pCsr;
  sIter.iCol = iCol;
  sIter.nSnippet = nSnippet;
  sIter.nPhrase = nList;
  sIter.iCurrent = -1;
  rc = fts3ExprIterate(pCsr->pExpr, 0, fts3SnippetFindPositions, (void*)&sIter);
  if( rc==SQLITE_OK ){

    /* Set the *pmSeen output variable. */
    for(i=0; i<nList; i++){
      if( sIter.aPhrase[i].pHead ){
        *pmSeen |= (u64)1 << i;
      }
................................................................................
  }

  return rc;
}

/*
** fts3ExprIterate() callback used to gather information for the matchinfo
** directive 'y'.
*/
static int fts3ExprLHitsCb(
  Fts3Expr *pExpr,                /* Phrase expression node */
  int iPhrase,                    /* Phrase number */
  void *pCtx                      /* Pointer to MatchInfo structure */
){
  int rc = SQLITE_OK;
................................................................................

  /* Allocate and populate the array of LcsIterator objects. The array
  ** contains one element for each matchable phrase in the query.
  **/
  aIter = sqlite3_malloc(sizeof(LcsIterator) * pCsr->nPhrase);
  if( !aIter ) return SQLITE_NOMEM;
  memset(aIter, 0, sizeof(LcsIterator) * pCsr->nPhrase);
  (void)fts3ExprIterate(pCsr->pExpr, 0, fts3MatchinfoLcsCb, (void*)aIter);

  for(i=0; i<pInfo->nPhrase; i++){
    LcsIterator *pIter = &aIter[i];
    nToken -= pIter->pExpr->pPhrase->nToken;
    pIter->iPosOffset = nToken;
  }

................................................................................
        }
        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, 1, fts3ExprLHitsCb, (void*)pInfo);
        break;
      }

      default: {
        Fts3Expr *pExpr;
        assert( zArg[i]==FTS3_MATCHINFO_HITS );
        pExpr = pCsr->pExpr;
................................................................................
        rc = fts3ExprLoadDoclists(pCsr, 0, 0);
        if( rc!=SQLITE_OK ) break;
        if( bGlobal ){
          if( pCsr->pDeferred ){
            rc = fts3MatchinfoSelectDoctotal(pTab, &pSelect, &pInfo->nDoc, 0);
            if( rc!=SQLITE_OK ) break;
          }
          rc = fts3ExprIterate(pExpr, 0, fts3ExprGlobalHitsCb,(void*)pInfo);
          if( rc!=SQLITE_OK ) break;
        }
        (void)fts3ExprIterate(pExpr, 0, fts3ExprLocalHitsCb,(void*)pInfo);
        break;
      }
    }

    pInfo->aMatchinfo += fts3MatchinfoSize(pInfo, zArg[i]);
  }

................................................................................

    /* 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. */
    for(i=0; zArg[i]; i++){






      nMatchinfo += fts3MatchinfoSize(&sInfo, zArg[i]);
    }

    /* Allocate space for Fts3Cursor.aMatchinfo[] and Fts3Cursor.zMatchinfo. */
    pCsr->pMIBuffer = fts3MIBufferNew(nMatchinfo, zArg);
    if( !pCsr->pMIBuffer ) rc = SQLITE_NOMEM;

................................................................................

    /* Initialize the contents of sCtx.aTerm[] for column iCol. There is 
    ** no way that this operation can fail, so the return code from
    ** fts3ExprIterate() can be discarded.
    */
    sCtx.iCol = iCol;
    sCtx.iTerm = 0;
    (void)fts3ExprIterate(pCsr->pExpr, 0, fts3ExprTermOffsetInit, (void*)&sCtx);

    /* Retreive the text stored in column iCol. If an SQL NULL is stored 
    ** in column iCol, jump immediately to the next iteration of the loop.
    ** If an OOM occurs while retrieving the data (this can happen if SQLite
    ** needs to transform the data from utf-16 to utf-8), return SQLITE_NOMEM 
    ** to the caller. 
    */
................................................................................
*/
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;
  int rc;
  int i;
  const char *zFormat;


  if( zArg ){
    for(i=0; zArg[i]; i++){
      char *zErr = 0;
      if( fts3MatchinfoCheck(pTab, zArg[i], &zErr) ){
        sqlite3_result_error(pContext, zErr, -1);
        sqlite3_free(zErr);
        return;
      }
    }
    zFormat = zArg;
  }else{
    zFormat = FTS3_MATCHINFO_DEFAULT;
  }

  if( !pCsr->pExpr ){
    sqlite3_result_blob(pContext, "", 0, SQLITE_STATIC);
    return;
  }

  /* Retrieve matchinfo() data. */
  rc = fts3GetMatchinfo(pContext, pCsr, zFormat);
  sqlite3Fts3SegmentsClose(pTab);

}

#endif







>
>





<






|





|

|







 







<




|







 







|













|







 







|







 







|







 







|







 







|







 







|


|







 







>
>
>
>
>
>







 







|







 







<
<

>


<
<
<
<
<
<
<
<








|
<
|
|
|
>



234
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
...
277
278
279
280
281
282
283

284
285
286
287
288
289
290
291
292
293
294
295
...
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
...
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
...
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
....
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
....
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
....
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
....
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
....
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
....
1683
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
*/
static void fts3GetDeltaPosition(char **pp, int *piPos){
  int iVal;
  *pp += fts3GetVarint32(*pp, &iVal);
  *piPos += (iVal-2);
}

static int fts3ExprLHitsCb(Fts3Expr*, int, void*);

/*
** Helper function for fts3ExprIterate() (see below).
*/
static int fts3ExprIterate2(
  Fts3Expr *pExpr,                /* Expression to iterate phrases of */

  int *piPhrase,                  /* Pointer to phrase counter */
  int (*x)(Fts3Expr*,int,void*),  /* Callback function to invoke for phrases */
  void *pCtx                      /* Second argument to pass to callback */
){
  int rc;                         /* Return code */

  if( x==fts3ExprLHitsCb && pExpr->bEof ){
    rc = SQLITE_OK;
  }else{
    int eType = pExpr->eType;     /* Type of expression node pExpr */
    if( eType!=FTSQUERY_PHRASE ){
      assert( pExpr->pLeft && pExpr->pRight );
      rc = fts3ExprIterate2(pExpr->pLeft, piPhrase, x, pCtx);
      if( rc==SQLITE_OK && eType!=FTSQUERY_NOT ){
        rc = fts3ExprIterate2(pExpr->pRight, piPhrase, x, pCtx);
      }
    }else{
      rc = x(pExpr, *piPhrase, pCtx);
      (*piPhrase)++;
    }
  }
  return rc;
................................................................................
** If the callback function returns anything other than SQLITE_OK, 
** the iteration is abandoned and the error code returned immediately.
** Otherwise, SQLITE_OK is returned after a callback has been made for
** all eligible phrase nodes.
*/
static int fts3ExprIterate(
  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().
*/
................................................................................
  Fts3Cursor *pCsr,               /* Fts3 cursor for current query */
  int *pnPhrase,                  /* OUT: Number of phrases in query */
  int *pnToken                    /* OUT: Number of tokens in query */
){
  int rc;                         /* Return Code */
  LoadDoclistCtx sCtx = {0,0,0};  /* Context for fts3ExprIterate() */
  sCtx.pCsr = pCsr;
  rc = fts3ExprIterate(pCsr->pExpr, fts3ExprLoadDoclistsCb, (void *)&sCtx);
  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;
}

/*
** Advance the position list iterator specified by the first two 
** arguments so that it points to the first element with a value greater
** than or equal to parameter iNext.
................................................................................
  ** the set of phrases in the expression to populate the aPhrase[] array.
  */
  sIter.pCsr = pCsr;
  sIter.iCol = iCol;
  sIter.nSnippet = nSnippet;
  sIter.nPhrase = nList;
  sIter.iCurrent = -1;
  rc = fts3ExprIterate(pCsr->pExpr, fts3SnippetFindPositions, (void*)&sIter);
  if( rc==SQLITE_OK ){

    /* Set the *pmSeen output variable. */
    for(i=0; i<nList; i++){
      if( sIter.aPhrase[i].pHead ){
        *pmSeen |= (u64)1 << i;
      }
................................................................................
  }

  return rc;
}

/*
** fts3ExprIterate() callback used to gather information for the matchinfo
** directives 'y' and 'b'.
*/
static int fts3ExprLHitsCb(
  Fts3Expr *pExpr,                /* Phrase expression node */
  int iPhrase,                    /* Phrase number */
  void *pCtx                      /* Pointer to MatchInfo structure */
){
  int rc = SQLITE_OK;
................................................................................

  /* Allocate and populate the array of LcsIterator objects. The array
  ** contains one element for each matchable phrase in the query.
  **/
  aIter = sqlite3_malloc(sizeof(LcsIterator) * pCsr->nPhrase);
  if( !aIter ) return SQLITE_NOMEM;
  memset(aIter, 0, sizeof(LcsIterator) * pCsr->nPhrase);
  (void)fts3ExprIterate(pCsr->pExpr, fts3MatchinfoLcsCb, (void*)aIter);

  for(i=0; i<pInfo->nPhrase; i++){
    LcsIterator *pIter = &aIter[i];
    nToken -= pIter->pExpr->pPhrase->nToken;
    pIter->iPosOffset = nToken;
  }

................................................................................
        }
        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;
................................................................................
        rc = fts3ExprLoadDoclists(pCsr, 0, 0);
        if( rc!=SQLITE_OK ) break;
        if( bGlobal ){
          if( pCsr->pDeferred ){
            rc = fts3MatchinfoSelectDoctotal(pTab, &pSelect, &pInfo->nDoc, 0);
            if( rc!=SQLITE_OK ) break;
          }
          rc = fts3ExprIterate(pExpr, fts3ExprGlobalHitsCb,(void*)pInfo);
          if( rc!=SQLITE_OK ) break;
        }
        (void)fts3ExprIterate(pExpr, fts3ExprLocalHitsCb,(void*)pInfo);
        break;
      }
    }

    pInfo->aMatchinfo += fts3MatchinfoSize(pInfo, zArg[i]);
  }

................................................................................

    /* 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. */
    for(i=0; zArg[i]; i++){
      char *zErr = 0;
      if( fts3MatchinfoCheck(pTab, zArg[i], &zErr) ){
        sqlite3_result_error(pCtx, zErr, -1);
        sqlite3_free(zErr);
        return;
      }
      nMatchinfo += fts3MatchinfoSize(&sInfo, zArg[i]);
    }

    /* Allocate space for Fts3Cursor.aMatchinfo[] and Fts3Cursor.zMatchinfo. */
    pCsr->pMIBuffer = fts3MIBufferNew(nMatchinfo, zArg);
    if( !pCsr->pMIBuffer ) rc = SQLITE_NOMEM;

................................................................................

    /* Initialize the contents of sCtx.aTerm[] for column iCol. There is 
    ** no way that this operation can fail, so the return code from
    ** fts3ExprIterate() can be discarded.
    */
    sCtx.iCol = iCol;
    sCtx.iTerm = 0;
    (void)fts3ExprIterate(pCsr->pExpr, fts3ExprTermOffsetInit, (void*)&sCtx);

    /* Retreive the text stored in column iCol. If an SQL NULL is stored 
    ** in column iCol, jump immediately to the next iteration of the loop.
    ** If an OOM occurs while retrieving the data (this can happen if SQLite
    ** needs to transform the data from utf-16 to utf-8), return SQLITE_NOMEM 
    ** to the caller. 
    */
................................................................................
*/
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