github.com/cockroachdb/cockroach@v20.2.0-alpha.1+incompatible/pkg/sql/stats/row_sampling_test.go (about)

     1  // Copyright 2017 The Cockroach Authors.
     2  //
     3  // Use of this software is governed by the Business Source License
     4  // included in the file licenses/BSL.txt.
     5  //
     6  // As of the Change Date specified in that file, in accordance with
     7  // the Business Source License, use of this software will be governed
     8  // by the Apache License, Version 2.0, included in the file
     9  // licenses/APL.txt.
    10  
    11  package stats
    12  
    13  import (
    14  	"context"
    15  	"fmt"
    16  	"reflect"
    17  	"sort"
    18  	"testing"
    19  
    20  	"github.com/cockroachdb/cockroach/pkg/settings/cluster"
    21  	"github.com/cockroachdb/cockroach/pkg/sql/sem/tree"
    22  	"github.com/cockroachdb/cockroach/pkg/sql/sqlbase"
    23  	"github.com/cockroachdb/cockroach/pkg/sql/types"
    24  	"github.com/cockroachdb/cockroach/pkg/util"
    25  	"github.com/cockroachdb/cockroach/pkg/util/randutil"
    26  )
    27  
    28  // runSampleTest feeds rows with the given ranks through a reservoir
    29  // of a given size and verifies the results are correct.
    30  func runSampleTest(t *testing.T, evalCtx *tree.EvalContext, numSamples int, ranks []int) {
    31  	ctx := context.Background()
    32  	var sr SampleReservoir
    33  	sr.Init(numSamples, []*types.T{types.Int}, nil /* memAcc */, util.MakeFastIntSet(0))
    34  	for _, r := range ranks {
    35  		d := sqlbase.DatumToEncDatum(types.Int, tree.NewDInt(tree.DInt(r)))
    36  		if err := sr.SampleRow(ctx, evalCtx, sqlbase.EncDatumRow{d}, uint64(r)); err != nil {
    37  			t.Errorf("%v", err)
    38  		}
    39  	}
    40  	samples := sr.Get()
    41  	sampledRanks := make([]int, len(samples))
    42  
    43  	// Verify that the row and the ranks weren't mishandled.
    44  	for i, s := range samples {
    45  		if *s.Row[0].Datum.(*tree.DInt) != tree.DInt(s.Rank) {
    46  			t.Fatalf(
    47  				"mismatch between row %s and rank %d",
    48  				s.Row.String([]*types.T{types.Int}), s.Rank,
    49  			)
    50  		}
    51  		sampledRanks[i] = int(s.Rank)
    52  	}
    53  
    54  	// Verify the top ranks made it.
    55  	sort.Ints(sampledRanks)
    56  	expected := append([]int(nil), ranks...)
    57  	sort.Ints(expected)
    58  	if len(expected) > numSamples {
    59  		expected = expected[:numSamples]
    60  	}
    61  	if !reflect.DeepEqual(expected, sampledRanks) {
    62  		t.Errorf("invalid ranks: %v vs %v", sampledRanks, expected)
    63  	}
    64  }
    65  
    66  func TestSampleReservoir(t *testing.T) {
    67  	evalCtx := tree.MakeTestingEvalContext(cluster.MakeTestingClusterSettings())
    68  	for _, n := range []int{10, 100, 1000, 10000} {
    69  		rng, _ := randutil.NewPseudoRand()
    70  		ranks := make([]int, n)
    71  		for i := range ranks {
    72  			ranks[i] = rng.Int()
    73  		}
    74  		for _, k := range []int{1, 5, 10, 100} {
    75  			t.Run(fmt.Sprintf("%d/%d", n, k), func(t *testing.T) {
    76  				runSampleTest(t, &evalCtx, k, ranks)
    77  			})
    78  		}
    79  	}
    80  }
    81  
    82  func TestTruncateDatum(t *testing.T) {
    83  	evalCtx := tree.MakeTestingEvalContext(cluster.MakeTestingClusterSettings())
    84  	runTest := func(d, expected tree.Datum) {
    85  		actual := truncateDatum(&evalCtx, d, 10 /* maxBytes */)
    86  		if actual.Compare(&evalCtx, expected) != 0 {
    87  			t.Fatalf("expected %s but found %s", expected.String(), actual.String())
    88  		}
    89  	}
    90  
    91  	original1, err := tree.ParseDBitArray("0110110101111100001100110110101111100001100110110101111" +
    92  		"10000110011011010111110000110011011010111110000110011011010111110000110")
    93  	if err != nil {
    94  		t.Fatal(err)
    95  	}
    96  	expected1, err := tree.ParseDBitArray("0110110101111100001100110110101111100001100110110101111" +
    97  		"1000011001101101011111000")
    98  	if err != nil {
    99  		t.Fatal(err)
   100  	}
   101  	runTest(original1, expected1)
   102  
   103  	original2 := tree.DBytes("deadbeef1234567890")
   104  	expected2 := tree.DBytes("deadbeef12")
   105  	runTest(&original2, &expected2)
   106  
   107  	original3 := tree.DString("Hello 世界")
   108  	expected3 := tree.DString("Hello 世")
   109  	runTest(&original3, &expected3)
   110  
   111  	original4, err := tree.NewDCollatedString(`IT was lovely summer weather in the country, and the golden
   112  corn, the green oats, and the haystacks piled up in the meadows looked beautiful`,
   113  		"en_US", &tree.CollationEnvironment{})
   114  	if err != nil {
   115  		t.Fatal(err)
   116  	}
   117  	expected4, err := tree.NewDCollatedString("IT was lov", "en_US", &tree.CollationEnvironment{})
   118  	if err != nil {
   119  		t.Fatal(err)
   120  	}
   121  	runTest(original4, expected4)
   122  }