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

planner: make index-hash-join and index-merge-join consider collation #21201

Merged
merged 5 commits into from
Nov 23, 2020
Merged

planner: make index-hash-join and index-merge-join consider collation #21201

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
c1efdd4
done
wjhuang2016 Nov 23, 2020
1d0071a
fix
wjhuang2016 Nov 23, 2020
4f4466c
address comment
wjhuang2016 Nov 23, 2020
85cc3c1
clean
wjhuang2016 Nov 23, 2020
ec628b6
Merge branch 'master' into fix_join_coll
ti-srebot Nov 23, 2020
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
18 changes: 18 additions & 0 deletions expression/integration_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7737,3 +7737,21 @@ func (s *testIntegrationSerialSuite) TestIssue20608(c *C) {
tk := testkit.NewTestKit(c, s.store)
tk.MustQuery("select '䇇Հ' collate utf8mb4_bin like '___Հ';").Check(testkit.Rows("0"))
}

func (s *testIntegrationSerialSuite) TestCollationIndexJoin(c *C) {
collate.SetNewCollationEnabledForTest(true)
defer collate.SetNewCollationEnabledForTest(false)
tk := testkit.NewTestKit(c, s.store)
tk.MustExec("use test")
tk.MustExec("drop table if exists t1, t2")
tk.MustExec("create table t1(a int, b char(10), key(b)) collate utf8mb4_general_ci")
tk.MustExec("create table t2(a int, b char(10), key(b)) collate ascii_bin")
tk.MustExec("insert into t1 values (1, 'a')")
tk.MustExec("insert into t2 values (1, 'A')")

tk.MustQuery("select /*+ inl_join(t1) */ t1.b, t2.b from t1 join t2 where t1.b=t2.b").Check(testkit.Rows("a A"))
tk.MustQuery("select /*+ hash_join(t1) */ t1.b, t2.b from t1 join t2 where t1.b=t2.b").Check(testkit.Rows("a A"))
tk.MustQuery("select /*+ merge_join(t1) */ t1.b, t2.b from t1 join t2 where t1.b=t2.b").Check(testkit.Rows("a A"))
tk.MustQuery("select /*+ inl_hash_join(t1) */ t1.b, t2.b from t1 join t2 where t1.b=t2.b").Check(testkit.Rows("a A"))
tk.MustQuery("select /*+ inl_merge_join(t1) */ t1.b, t2.b from t1 join t2 where t1.b=t2.b").Check(testkit.Rows("a A"))
}
40 changes: 26 additions & 14 deletions planner/core/exhaust_physical_plans.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ import (
"github.com/pingcap/tidb/statistics"
"github.com/pingcap/tidb/types"
"github.com/pingcap/tidb/util/chunk"
"github.com/pingcap/tidb/util/collate"
"github.com/pingcap/tidb/util/logutil"
"github.com/pingcap/tidb/util/plancodec"
"github.com/pingcap/tidb/util/ranger"
Expand Down Expand Up @@ -479,6 +480,16 @@ func (p *LogicalJoin) constructIndexJoin(
IsNullEQ: newIsNullEQ,
DefaultValues: p.DefaultValues,
}
// Correct the collation used by hash.
for i := range outerHashKeys {
// Make compiler happy.
if len(innerHashKeys) == 0 {
return nil
}
chs, coll := expression.DeriveCollationFromExprs(nil, outerHashKeys[i], innerHashKeys[i])
outerHashKeys[i].GetType().Charset, outerHashKeys[i].GetType().Collate = chs, coll
innerHashKeys[i].GetType().Charset, innerHashKeys[i].GetType().Collate = chs, coll
}
join := PhysicalIndexJoin{
basePhysicalJoin: baseJoin,
innerTask: innerTask,
Expand Down Expand Up @@ -657,15 +668,14 @@ func (p *LogicalJoin) getIndexJoinByOuterIdx(prop *property.PhysicalProperty, ou
return p.buildIndexJoinInner2IndexScan(prop, ds, innerJoinKeys, outerJoinKeys, outerIdx, us, avgInnerRowCnt)
}

func (p *LogicalJoin) getIndexJoinBuildHelper(ds *DataSource, innerJoinKeys []*expression.Column,
checkPathValid func(path *util.AccessPath) bool) (*indexJoinBuildHelper, []int) {
func (p *LogicalJoin) getIndexJoinBuildHelper(ds *DataSource, innerJoinKeys []*expression.Column, checkPathValid func(path *util.AccessPath) bool, outerJoinKeys []*expression.Column) (*indexJoinBuildHelper, []int) {
helper := &indexJoinBuildHelper{
join: p,
innerPlan: ds,
}
for _, path := range ds.possibleAccessPaths {
if checkPathValid(path) {
emptyRange, err := helper.analyzeLookUpFilters(path, ds, innerJoinKeys)
emptyRange, err := helper.analyzeLookUpFilters(path, ds, innerJoinKeys, outerJoinKeys)
if emptyRange {
return nil, nil
}
Expand Down Expand Up @@ -713,7 +723,7 @@ func (p *LogicalJoin) buildIndexJoinInner2TableScan(
var innerTask, innerTask2 task
var helper *indexJoinBuildHelper
if ds.tableInfo.IsCommonHandle {
helper, keyOff2IdxOff = p.getIndexJoinBuildHelper(ds, innerJoinKeys, func(path *util.AccessPath) bool { return path.IsCommonHandlePath })
helper, keyOff2IdxOff = p.getIndexJoinBuildHelper(ds, innerJoinKeys, func(path *util.AccessPath) bool { return path.IsCommonHandlePath }, outerJoinKeys)
if helper == nil {
return nil
}
Expand Down Expand Up @@ -774,7 +784,7 @@ func (p *LogicalJoin) buildIndexJoinInner2TableScan(
func (p *LogicalJoin) buildIndexJoinInner2IndexScan(
prop *property.PhysicalProperty, ds *DataSource, innerJoinKeys, outerJoinKeys []*expression.Column,
outerIdx int, us *LogicalUnionScan, avgInnerRowCnt float64) (joins []PhysicalPlan) {
helper, keyOff2IdxOff := p.getIndexJoinBuildHelper(ds, innerJoinKeys, func(path *util.AccessPath) bool { return !path.IsTablePath() })
helper, keyOff2IdxOff := p.getIndexJoinBuildHelper(ds, innerJoinKeys, func(path *util.AccessPath) bool { return !path.IsTablePath() }, outerJoinKeys)
if helper == nil {
return nil
}
Expand Down Expand Up @@ -1250,16 +1260,18 @@ loopOtherConds:
// It's clearly that the column c cannot be used to access data. So we need to remove it and reset the IdxOff2KeyOff to
// [0 -1 -1].
// So that we can use t1.a=t2.a and t1.b > t2.b-10 and t1.b < t2.b+10 to build ranges then access data.
func (ijHelper *indexJoinBuildHelper) removeUselessEqAndInFunc(
idxCols []*expression.Column,
notKeyEqAndIn []expression.Expression) (
usefulEqAndIn, uselessOnes []expression.Expression,
) {
func (ijHelper *indexJoinBuildHelper) removeUselessEqAndInFunc(idxCols []*expression.Column, notKeyEqAndIn []expression.Expression, outerJoinKeys []*expression.Column) (usefulEqAndIn, uselessOnes []expression.Expression) {
ijHelper.curPossibleUsedKeys = make([]*expression.Column, 0, len(idxCols))
for idxColPos, notKeyColPos := 0, 0; idxColPos < len(idxCols); idxColPos++ {
if ijHelper.curIdxOff2KeyOff[idxColPos] != -1 {
ijHelper.curPossibleUsedKeys = append(ijHelper.curPossibleUsedKeys, idxCols[idxColPos])
continue
// Check collation is the new collation is enabled.
tmpType := make([]expression.Expression, 0)
expression.DeriveCollationFromExprs(nil, idxCols[idxColPos], outerJoinKeys[ijHelper.curIdxOff2KeyOff[idxColPos]])
_, coll := expression.DeriveCollationFromExprs(nil, tmpType...)
if !collate.NewCollationEnabled() || collate.CompatibleCollate(idxCols[idxColPos].GetType().Collate, coll) {
ijHelper.curPossibleUsedKeys = append(ijHelper.curPossibleUsedKeys, idxCols[idxColPos])
continue
}
}
if notKeyColPos < len(notKeyEqAndIn) && ijHelper.curNotUsedIndexCols[notKeyColPos].Equal(nil, idxCols[idxColPos]) {
notKeyColPos++
Expand All @@ -1276,15 +1288,15 @@ func (ijHelper *indexJoinBuildHelper) removeUselessEqAndInFunc(
return notKeyEqAndIn, nil
}

func (ijHelper *indexJoinBuildHelper) analyzeLookUpFilters(path *util.AccessPath, innerPlan *DataSource, innerJoinKeys []*expression.Column) (emptyRange bool, err error) {
func (ijHelper *indexJoinBuildHelper) analyzeLookUpFilters(path *util.AccessPath, innerPlan *DataSource, innerJoinKeys []*expression.Column, outerJoinKeys []*expression.Column) (emptyRange bool, err error) {
if len(path.IdxCols) == 0 {
return false, nil
}
accesses := make([]expression.Expression, 0, len(path.IdxCols))
ijHelper.resetContextForIndex(innerJoinKeys, path.IdxCols, path.IdxColLens)
notKeyEqAndIn, remained, rangeFilterCandidates := ijHelper.findUsefulEqAndInFilters(innerPlan)
var remainedEqAndIn []expression.Expression
notKeyEqAndIn, remainedEqAndIn = ijHelper.removeUselessEqAndInFunc(path.IdxCols, notKeyEqAndIn)
notKeyEqAndIn, remainedEqAndIn = ijHelper.removeUselessEqAndInFunc(path.IdxCols, notKeyEqAndIn, outerJoinKeys)
matchedKeyCnt := len(ijHelper.curPossibleUsedKeys)
// If no join key is matched while join keys actually are not empty. We don't choose index join for now.
if matchedKeyCnt <= 0 && len(innerJoinKeys) > 0 {
Expand Down
2 changes: 1 addition & 1 deletion planner/core/exhaust_physical_plans_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -248,7 +248,7 @@ func (s *testUnitTestSuit) TestIndexJoinAnalyzeLookUpFilters(c *C) {
c.Assert(err, IsNil)
joinNode.OtherConditions = others
helper := &indexJoinBuildHelper{join: joinNode, lastColManager: nil, innerPlan: dataSourceNode}
_, err = helper.analyzeLookUpFilters(path, dataSourceNode, tt.innerKeys)
_, err = helper.analyzeLookUpFilters(path, dataSourceNode, tt.innerKeys, tt.innerKeys)
c.Assert(err, IsNil)
c.Assert(fmt.Sprintf("%v", helper.chosenAccess), Equals, tt.accesses)
c.Assert(fmt.Sprintf("%v", helper.chosenRanges), Equals, tt.ranges, Commentf("test case: #%v", i))
Expand Down