util/ranger: don't merge consecutive ranges when building ranges for …

…cnf item (#41661) (#44356)

close #41572

util/ranger/detacher.go

@@ -204,7 +204,14 @@ func extractIndexPointRangesForCNF(sctx sessionctx.Context, conds []expression.E
 		if colSets.Len() == 0 {
 			continue
 		}
-		res, err := DetachCondAndBuildRangeForIndex(sctx, tmpConds, cols, lengths, rangeMaxSize)
+		// When we build ranges for the CNF item, we choose not to merge consecutive ranges because we hope to get point
+		// ranges here. See https://github.com/pingcap/tidb/issues/41572 for more details.
+		//
+		// Here is an example. Assume that the index is `idx(a,b,c)` and the condition is `((a,b) in ((1,1),(1,2)) and c = 1`.
+		// We build ranges for `(a,b) in ((1,1),(1,2))` and get `[1 1, 1 1] [1 2, 1 2]`, which are point ranges and we can
+		// append `c = 1` to the point ranges. However, if we choose to merge consecutive ranges here, we get `[1 1, 1 2]`,
+		// which are not point ranges, and we cannot append `c = 1` anymore.
+		res, err := detachCondAndBuildRangeWithoutMerging(sctx, tmpConds, cols, lengths, rangeMaxSize)
 		if err != nil {
 			return nil, -1, nil, err
 		}
@@ -821,11 +828,9 @@ func DetachCondAndBuildRangeForIndex(sctx sessionctx.Context, conditions []expre
 	return d.detachCondAndBuildRangeForCols()
 }
 
-// DetachCondAndBuildRangeForPartition will detach the index filters from table filters.
-// rangeMaxSize is the max memory limit for ranges. O indicates no memory limit. If you ask that all conditions must be used
-// for building ranges, set rangeMemQuota to 0 to avoid range fallback.
-// The returned values are encapsulated into a struct DetachRangeResult, see its comments for explanation.
-func DetachCondAndBuildRangeForPartition(sctx sessionctx.Context, conditions []expression.Expression, cols []*expression.Column,
+// detachCondAndBuildRangeWithoutMerging detaches the index filters from table filters and uses them to build ranges.
+// When building ranges, it doesn't merge consecutive ranges.
+func detachCondAndBuildRangeWithoutMerging(sctx sessionctx.Context, conditions []expression.Expression, cols []*expression.Column,
 	lengths []int, rangeMaxSize int64) (*DetachRangeResult, error) {
 	d := &rangeDetacher{
 		sctx:             sctx,
@@ -838,6 +843,15 @@ func DetachCondAndBuildRangeForPartition(sctx sessionctx.Context, conditions []e
 	return d.detachCondAndBuildRangeForCols()
 }
 
+// DetachCondAndBuildRangeForPartition will detach the index filters from table filters.
+// rangeMaxSize is the max memory limit for ranges. O indicates no memory limit. If you ask that all conditions must be used
+// for building ranges, set rangeMemQuota to 0 to avoid range fallback.
+// The returned values are encapsulated into a struct DetachRangeResult, see its comments for explanation.
+func DetachCondAndBuildRangeForPartition(sctx sessionctx.Context, conditions []expression.Expression, cols []*expression.Column,
+	lengths []int, rangeMaxSize int64) (*DetachRangeResult, error) {
+	return detachCondAndBuildRangeWithoutMerging(sctx, conditions, cols, lengths, rangeMaxSize)
+}
+
 type rangeDetacher struct {
 	sctx             sessionctx.Context
 	allConds         []expression.Expression

util/ranger/ranger_test.go

@@ -1000,6 +1000,33 @@ func TestCompIndexMultiColDNF2(t *testing.T) {
 	}
 }
 
+func TestIssue41572(t *testing.T) {
+	store := testkit.CreateMockStore(t)
+
+	testKit := testkit.NewTestKit(t, store)
+	testKit.MustExec("use test")
+	testKit.MustExec("drop table if exists t")
+	testKit.MustExec("create table t(a varchar(100), b int, c int, d int, index idx(a, b, c))")
+	testKit.MustExec("insert into t values ('t',1,1,1),('t',1,3,3),('t',2,1,3),('t',2,3,1),('w',0,3,3),('z',0,1,1)")
+
+	var input []string
+	var output []struct {
+		SQL    string
+		Plan   []string
+		Result []string
+	}
+	rangerSuiteData.LoadTestCases(t, &input, &output)
+	for i, tt := range input {
+		testdata.OnRecord(func() {
+			output[i].SQL = tt
+			output[i].Plan = testdata.ConvertRowsToStrings(testKit.MustQuery("explain " + tt).Rows())
+			output[i].Result = testdata.ConvertRowsToStrings(testKit.MustQuery(tt).Rows())
+		})
+		testKit.MustQuery("explain " + tt).Check(testkit.Rows(output[i].Plan...))
+		testKit.MustQuery(tt).Sort().Check(testkit.Rows(output[i].Result...))
+	}
+}
+
 func TestPrefixIndexMultiColDNF(t *testing.T) {
 	store := testkit.CreateMockStore(t)
 

util/ranger/testdata/ranger_suite_in.json

@@ -59,6 +59,13 @@
       "select * from t where ((a = 1 and b = 1) or (a = 2 and b = 2)) and c > 2;"
     ]
   },
+  {
+    "name": "TestIssue41572",
+    "cases": [
+      "select * from t use index (idx) where ((a = 't' and b = 1) or (a = 't' and b = 2) or (a = 'w' and b = 0)) and c > 2",
+      "select * from t use index (idx) where ((a = 't' and b = 1) or (a = 't' and b = 2) or (a = 'w' and b = 0)) and d > 2"
+    ]
+  },
   {
     "name": "TestPrefixIndexMultiColDNF",
     "cases": [

util/ranger/testdata/ranger_suite_out.json

@@ -333,6 +333,38 @@
       }
     ]
   },
+  {
+    "Name": "TestIssue41572",
+    "Cases": [
+      {
+        "SQL": "select * from t use index (idx) where ((a = 't' and b = 1) or (a = 't' and b = 2) or (a = 'w' and b = 0)) and c > 2",
+        "Plan": [
+          "IndexLookUp_7 1.00 root  ",
+          "├─IndexRangeScan_5(Build) 1.00 cop[tikv] table:t, index:idx(a, b, c) range:(\"t\" 1 2,\"t\" 1 +inf], (\"t\" 2 2,\"t\" 2 +inf], (\"w\" 0 2,\"w\" 0 +inf], keep order:false, stats:pseudo",
+          "└─TableRowIDScan_6(Probe) 1.00 cop[tikv] table:t keep order:false, stats:pseudo"
+        ],
+        "Result": [
+          "t 1 3 3",
+          "t 2 3 1",
+          "w 0 3 3"
+        ]
+      },
+      {
+        "SQL": "select * from t use index (idx) where ((a = 't' and b = 1) or (a = 't' and b = 2) or (a = 'w' and b = 0)) and d > 2",
+        "Plan": [
+          "IndexLookUp_8 0.10 root  ",
+          "├─IndexRangeScan_5(Build) 0.30 cop[tikv] table:t, index:idx(a, b, c) range:[\"t\" 1,\"t\" 1], [\"t\" 2,\"t\" 2], [\"w\" 0,\"w\" 0], keep order:false, stats:pseudo",
+          "└─Selection_7(Probe) 0.10 cop[tikv]  gt(test.t.d, 2)",
+          "  └─TableRowIDScan_6 0.30 cop[tikv] table:t keep order:false, stats:pseudo"
+        ],
+        "Result": [
+          "t 1 3 3",
+          "t 2 1 3",
+          "w 0 3 3"
+        ]
+      }
+    ]
+  },
   {
     "Name": "TestPrefixIndexMultiColDNF",
     "Cases": [