SQLite
Check-in [9ce6f40dfb54b35cecba3cc9c1ec0d111f6e9f11]
Not logged in
Overview
SHA1 Hash:9ce6f40dfb54b35cecba3cc9c1ec0d111f6e9f11
Date: 2013-09-13 12:10:09
User: dan
Comment:Add tests for the fts4 unicode61 tokenchars and separators options.
Tags And Properties
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to test/fts4unicode.test

433
434
435
436
437
438
439
440






































































































441




}
do_execsql_test 8.2.2 {
  SELECT rowid FROM t4 WHERE t4 MATCH 'o';
} {1 3}
do_execsql_test 8.2.3 {
  SELECT rowid FROM t4 WHERE t4 MATCH 'a';
} {2 4}







































































































finish_test












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

>
>
>
>
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
}
do_execsql_test 8.2.2 {
  SELECT rowid FROM t4 WHERE t4 MATCH 'o';
} {1 3}
do_execsql_test 8.2.3 {
  SELECT rowid FROM t4 WHERE t4 MATCH 'a';
} {2 4}

#-------------------------------------------------------------------------
#
foreach {tn sql} {
  1 {
    CREATE VIRTUAL TABLE t5 USING fts4(tokenize=unicode61 [tokenchars= .]);
    CREATE VIRTUAL TABLE t6 USING fts4(
        tokenize=unicode61 [tokenchars=="] "tokenchars=[]");
    CREATE VIRTUAL TABLE t7 USING fts4(tokenize=unicode61 [separators=x\xC4]);
  }
  2 {
    CREATE VIRTUAL TABLE t5 USING fts4(tokenize=unicode61 "tokenchars= .");
    CREATE VIRTUAL TABLE t6 USING fts4(tokenize=unicode61 "tokenchars=[=""]");
    CREATE VIRTUAL TABLE t7 USING fts4(tokenize=unicode61 "separators=x\xC4");
  }
  3 {
    CREATE VIRTUAL TABLE t5 USING fts4(tokenize=unicode61 'tokenchars= .');
    CREATE VIRTUAL TABLE t6 USING fts4(tokenize=unicode61 'tokenchars=="[]');
    CREATE VIRTUAL TABLE t7 USING fts4(tokenize=unicode61 'separators=x\xC4');
  }
  4 {
    CREATE VIRTUAL TABLE t5 USING fts4(tokenize=unicode61 `tokenchars= .`);
    CREATE VIRTUAL TABLE t6 USING fts4(tokenize=unicode61 `tokenchars=[="]`);
    CREATE VIRTUAL TABLE t7 USING fts4(tokenize=unicode61 `separators=x\xC4`);
  }
} {
  do_execsql_test 9.$tn.0 { 
    DROP TABLE IF EXISTS t5;
    DROP TABLE IF EXISTS t5aux;
    DROP TABLE IF EXISTS t6;
    DROP TABLE IF EXISTS t6aux;
    DROP TABLE IF EXISTS t7;
    DROP TABLE IF EXISTS t7aux;
  }
  do_execsql_test 9.$tn.1 $sql

  do_execsql_test 9.$tn.2 {
    CREATE VIRTUAL TABLE t5aux USING fts4aux(t5);
    INSERT INTO t5 VALUES('one two three/four.five.six');
    SELECT * FROM t5aux;
  } {
    four.five.six   * 1 1 four.five.six   0 1 1 
    {one two three} * 1 1 {one two three} 0 1 1
  }

  do_execsql_test 9.$tn.3 {
    CREATE VIRTUAL TABLE t6aux USING fts4aux(t6);
    INSERT INTO t6 VALUES('alpha=beta"gamma/delta[epsilon]zeta');
    SELECT * FROM t6aux;
  } {
    {alpha=beta"gamma}   * 1 1 {alpha=beta"gamma} 0 1 1 
    {delta[epsilon]zeta} * 1 1 {delta[epsilon]zeta} 0 1 1
  }

  do_execsql_test 9.$tn.4 {
    CREATE VIRTUAL TABLE t7aux USING fts4aux(t7);
    INSERT INTO t7 VALUES('alephxbeth\xC4gimel');
    SELECT * FROM t7aux;
  } {
    aleph * 1 1 aleph 0 1 1 
    beth  * 1 1 beth  0 1 1 
    gimel * 1 1 gimel 0 1 1
  }
}

# Check that multiple options are handled correctly.
#
do_execsql_test 10.1 {
  DROP TABLE IF EXISTS t1;
  CREATE VIRTUAL TABLE t1 USING fts4(tokenize=unicode61
    "tokenchars=xyz" "tokenchars=.=" "separators=.=" "separators=xy"
    "separators=a" "separators=a" "tokenchars=a" "tokenchars=a"
  );

  INSERT INTO t1 VALUES('oneatwoxthreeyfour');
  INSERT INTO t1 VALUES('a.single=word');
  CREATE VIRTUAL TABLE t1aux USING fts4aux(t1);
  SELECT * FROM t1aux;
} {
  .single=word * 1 1 .single=word 0 1 1 
  four         * 1 1 four         0 1 1 
  one          * 1 1 one          0 1 1 
  three        * 1 1 three        0 1 1 
  two          * 1 1 two          0 1 1
}

# Test that case folding happens after tokenization, not before.
#
do_execsql_test 10.2 {
  DROP TABLE IF EXISTS t2;
  CREATE VIRTUAL TABLE t2 USING fts4(tokenize=unicode61 "separators=aB");
  INSERT INTO t2 VALUES('oneatwoBthree');
  INSERT INTO t2 VALUES('onebtwoAthree');
  CREATE VIRTUAL TABLE t2aux USING fts4aux(t2);
  SELECT * FROM t2aux;
} {
  one           * 1 1 one           0 1 1 
  onebtwoathree * 1 1 onebtwoathree 0 1 1 
  three         * 1 1 three         0 1 1 
  two           * 1 1 two           0 1 1
}


finish_test