Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

statistics, planner: use the correct IndexInfo in GenerateHistCollFromColumnInfo() | tidb-test=pr/2218 (#44441) #46730

Merged
merged 5 commits into from
Sep 28, 2023
Merged

statistics, planner: use the correct IndexInfo in GenerateHistCollFromColumnInfo() | tidb-test=pr/2218 (#44441) #46730

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
f4677ee
fix
time-and-fate Jun 6, 2023
4d1315c
fix fmt
time-and-fate Jun 6, 2023
d644cd6
change
time-and-fate Jun 6, 2023
56bd91c
remove unneeded parameter
time-and-fate Jun 6, 2023
45ec7da
make it more clear for review
time-and-fate Jun 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions planner/core/stats.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,7 @@ func (p *LogicalMemTable) DeriveStats(_ []*property.StatsInfo, selfSchema *expre
stats := &property.StatsInfo{
RowCount: float64(statsTable.RealtimeCount),
ColNDVs: make(map[int64]float64, len(p.TableInfo.Columns)),
HistColl: statsTable.GenerateHistCollFromColumnInfo(p.TableInfo.Columns, p.schema.Columns),
HistColl: statsTable.GenerateHistCollFromColumnInfo(p.TableInfo, p.schema.Columns),
StatsVersion: statistics.PseudoVersion,
}
for _, col := range selfSchema.Columns {
Expand Down Expand Up @@ -289,7 +289,7 @@ func (ds *DataSource) initStats(colGroups [][]*expression.Column) {
tableStats := &property.StatsInfo{
RowCount: float64(ds.statisticTable.RealtimeCount),
ColNDVs: make(map[int64]float64, ds.schema.Len()),
HistColl: ds.statisticTable.GenerateHistCollFromColumnInfo(ds.Columns, ds.schema.Columns),
HistColl: ds.statisticTable.GenerateHistCollFromColumnInfo(ds.tableInfo, ds.schema.Columns),
StatsVersion: ds.statisticTable.Version,
}
if ds.statisticTable.Pseudo {
Expand Down
16 changes: 16 additions & 0 deletions statistics/integration_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -891,3 +891,19 @@ func TestOrderingIdxSelectivityThreshold(t *testing.T) {
testKit.MustQuery(input[i]).Check(testkit.Rows(output[i].Result...))
}
}

func TestIssue44369(t *testing.T) {
store, dom := testkit.CreateMockStoreAndDomain(t)
h := dom.StatsHandle()
tk := testkit.NewTestKit(t, store)
tk.MustExec("use test")
tk.MustExec("create table t(a int, b int, index iab(a,b));")
require.NoError(t, h.HandleDDLEvent(<-h.DDLEventCh()))
tk.MustExec("insert into t value(1,1);")
require.NoError(t, h.DumpStatsDeltaToKV(handle.DumpAll))
tk.MustExec("analyze table t;")
is := dom.InfoSchema()
require.NoError(t, h.Update(is))
tk.MustExec("alter table t rename column b to bb;")
tk.MustExec("select * from t where a = 10 and bb > 20;")
}
4 changes: 2 additions & 2 deletions statistics/selectivity_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -574,7 +574,7 @@ func TestSelectivity(t *testing.T) {
sel := p.(plannercore.LogicalPlan).Children()[0].(*plannercore.LogicalSelection)
ds := sel.Children()[0].(*plannercore.DataSource)

histColl := statsTbl.GenerateHistCollFromColumnInfo(ds.Columns, ds.Schema().Columns)
histColl := statsTbl.GenerateHistCollFromColumnInfo(ds.TableInfo(), ds.Schema().Columns)

ratio, _, err := histColl.Selectivity(sctx, sel.Conditions, nil)
require.NoErrorf(t, err, "for %s", tt.exprs)
Expand Down Expand Up @@ -687,7 +687,7 @@ func TestDNFCondSelectivity(t *testing.T) {
sel := p.(plannercore.LogicalPlan).Children()[0].(*plannercore.LogicalSelection)
ds := sel.Children()[0].(*plannercore.DataSource)

histColl := statsTbl.GenerateHistCollFromColumnInfo(ds.Columns, ds.Schema().Columns)
histColl := statsTbl.GenerateHistCollFromColumnInfo(ds.TableInfo(), ds.Schema().Columns)

ratio, _, err := histColl.Selectivity(sctx, sel.Conditions, nil)
require.NoErrorf(t, err, "error %v, for expr %s", err, tt)
Expand Down
25 changes: 13 additions & 12 deletions statistics/table.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -939,33 +939,34 @@ func (coll *HistColl) ID2UniqueID(columns []*expression.Column) *HistColl {
}

// GenerateHistCollFromColumnInfo generates a new HistColl whose ColID2IdxIDs and IdxID2ColIDs is built from the given parameter.
func (coll *HistColl) GenerateHistCollFromColumnInfo(infos []*model.ColumnInfo, columns []*expression.Column) *HistColl {
func (coll *HistColl) GenerateHistCollFromColumnInfo(tblInfo *model.TableInfo, columns []*expression.Column) *HistColl {
newColHistMap := make(map[int64]*Column)
colInfoID2UniqueID := make(map[int64]int64, len(columns))
colNames2UniqueID := make(map[string]int64)
idxID2idxInfo := make(map[int64]*model.IndexInfo)
for _, col := range columns {
colInfoID2UniqueID[col.ID] = col.UniqueID
}
for _, colInfo := range infos {
uniqueID, ok := colInfoID2UniqueID[colInfo.ID]
if ok {
colNames2UniqueID[colInfo.Name.L] = uniqueID
}
}
for id, colHist := range coll.Columns {
uniqueID, ok := colInfoID2UniqueID[id]
// Collect the statistics by the given columns.
if ok {
newColHistMap[uniqueID] = colHist
}
}
for _, idxInfo := range tblInfo.Indices {
idxID2idxInfo[idxInfo.ID] = idxInfo
}
newIdxHistMap := make(map[int64]*Index)
idx2Columns := make(map[int64][]int64)
colID2IdxIDs := make(map[int64][]int64)
for _, idxHist := range coll.Indices {
ids := make([]int64, 0, len(idxHist.Info.Columns))
for _, idxCol := range idxHist.Info.Columns {
uniqueID, ok := colNames2UniqueID[idxCol.Name.L]
for id, idxHist := range coll.Indices {
idxInfo := idxID2idxInfo[id]
if idxInfo == nil {
continue
}
ids := make([]int64, 0, len(idxInfo.Columns))
for _, idxCol := range idxInfo.Columns {
uniqueID, ok := colInfoID2UniqueID[tblInfo.Columns[idxCol.Offset].ID]
if !ok {
break
}
Expand Down
8 changes: 4 additions & 4 deletions statistics/trace_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -221,9 +221,9 @@ func TestTraceDebugSelectivity(t *testing.T) {

// Collect common information for the following tests.
p := parser.New()
dsColInfos := make([][]*model.ColumnInfo, 0, len(in))
dsSchemaCols := make([][]*expression.Column, 0, len(in))
selConditions := make([][]expression.Expression, 0, len(in))
tblInfos := make([]*model.TableInfo, 0, len(in))
for _, sql := range in {
stmt, err := p.ParseOneStmt(sql, "", "")
require.NoError(t, err)
Expand All @@ -236,9 +236,9 @@ func TestTraceDebugSelectivity(t *testing.T) {
sel := p.(plannercore.LogicalPlan).Children()[0].(*plannercore.LogicalSelection)
ds := sel.Children()[0].(*plannercore.DataSource)

dsColInfos = append(dsColInfos, ds.Columns)
dsSchemaCols = append(dsSchemaCols, ds.Schema().Columns)
selConditions = append(selConditions, sel.Conditions)
tblInfos = append(tblInfos, ds.TableInfo())
}
var buf bytes.Buffer
encoder := json.NewEncoder(&buf)
Expand All @@ -247,7 +247,7 @@ func TestTraceDebugSelectivity(t *testing.T) {
// Test using ver2 stats.
for i, sql := range in {
stmtCtx.OptimizerDebugTrace = nil
histColl := statsTbl.GenerateHistCollFromColumnInfo(dsColInfos[i], dsSchemaCols[i])
histColl := statsTbl.GenerateHistCollFromColumnInfo(tblInfos[i], dsSchemaCols[i])
_, _, err = histColl.Selectivity(sctx, selConditions[i], nil)
require.NoError(t, err, sql, "For ver2")
traceInfo := stmtCtx.OptimizerDebugTrace
Expand All @@ -271,7 +271,7 @@ func TestTraceDebugSelectivity(t *testing.T) {
stmtCtx.EnableOptimizerDebugTrace = true
for i, sql := range in {
stmtCtx.OptimizerDebugTrace = nil
histColl := statsTbl.GenerateHistCollFromColumnInfo(dsColInfos[i], dsSchemaCols[i])
histColl := statsTbl.GenerateHistCollFromColumnInfo(tblInfos[i], dsSchemaCols[i])
_, _, err = histColl.Selectivity(sctx, selConditions[i], nil)
require.NoError(t, err, sql, "For ver1")
traceInfo := stmtCtx.OptimizerDebugTrace
Expand Down