From 6af4e4b160dd5c04d7379caf2a16a3daeb9a3e56 Mon Sep 17 00:00:00 2001
From: Song Gao <disxiaofei@163.com>
Date: Tue, 15 Nov 2022 18:27:55 +0800
Subject: [PATCH] statistics: fix merge global stats panic if the partition
 hasn't been analyzed before (#39123)

close pingcap/tidb#39121
---
 executor/analyze_test.go    | 21 +++++++++++++++++++++
 statistics/handle/handle.go | 12 +++++++++++-
 statistics/table.go         | 10 +++++-----
 3 files changed, 37 insertions(+), 6 deletions(-)

diff --git a/executor/analyze_test.go b/executor/analyze_test.go
index bbe3f5b8d1b1e..a6cdea833df50 100644
--- a/executor/analyze_test.go
+++ b/executor/analyze_test.go
@@ -417,3 +417,24 @@ func TestAnalyzePartitionTableByConcurrencyInDynamic(t *testing.T) {
 		tk.MustQuery("show stats_topn where partition_name = 'global' and table_name = 't'").CheckAt([]int{5, 6}, expected)
 	}
 }
+
+func TestMergeGlobalStatsWithUnAnalyzedPartition(t *testing.T) {
+	store := testkit.CreateMockStore(t)
+	tk := testkit.NewTestKit(t, store)
+	tk.MustExec("use test")
+	tk.MustExec("set tidb_partition_prune_mode=dynamic;")
+	tk.MustExec("CREATE TABLE `t` (   `id` int(11) DEFAULT NULL,   `a` int(11) DEFAULT NULL, `b` int(11) DEFAULT NULL, `c` int(11) DEFAULT NULL ) PARTITION BY RANGE (`id`) (PARTITION `p0` VALUES LESS THAN (3),  PARTITION `p1` VALUES LESS THAN (7),  PARTITION `p2` VALUES LESS THAN (11));")
+	tk.MustExec("insert into t values (1,1,1,1),(2,2,2,2),(4,4,4,4),(5,5,5,5),(6,6,6,6),(8,8,8,8),(9,9,9,9);")
+	tk.MustExec("create index idxa on t (a);")
+	tk.MustExec("create index idxb on t (b);")
+	tk.MustExec("create index idxc on t (c);")
+	tk.MustExec("analyze table t partition p0 index idxa;")
+	tk.MustExec("analyze table t partition p1 index idxb;")
+	tk.MustExec("analyze table t partition p2 index idxc;")
+	tk.MustQuery("show warnings").Check(testkit.Rows(
+		"Warning 1105 The version 2 would collect all statistics not only the selected indexes",
+		"Note 1105 Analyze use auto adjusted sample rate 1.000000 for table test.t's partition p2"))
+	tk.MustExec("analyze table t partition p0;")
+	tk.MustQuery("show warnings").Check(testkit.Rows(
+		"Note 1105 Analyze use auto adjusted sample rate 1.000000 for table test.t's partition p0"))
+}
diff --git a/statistics/handle/handle.go b/statistics/handle/handle.go
index 7dfc3f47d2d62..344ac0f72c755 100644
--- a/statistics/handle/handle.go
+++ b/statistics/handle/handle.go
@@ -775,7 +775,17 @@ func (h *Handle) mergePartitionStats2GlobalStats(sc sessionctx.Context,
 			allPartitionStats[partitionID] = partitionStats
 		}
 		for i := 0; i < globalStats.Num; i++ {
-			_, hg, cms, topN, fms := partitionStats.GetStatsInfo(histIDs[i], isIndex == 1)
+			_, hg, cms, topN, fms, analyzed := partitionStats.GetStatsInfo(histIDs[i], isIndex == 1)
+			if !analyzed {
+				var errMsg string
+				if isIndex == 0 {
+					errMsg = fmt.Sprintf("`%s`", tableInfo.Name.L)
+				} else {
+					errMsg = fmt.Sprintf("`%s` index: `%s`", tableInfo.Name.L, tableInfo.FindIndexNameByID(histIDs[0]))
+				}
+				err = types.ErrPartitionStatsMissing.GenWithStackByArgs(errMsg)
+				return
+			}
 			// partition stats is not empty but column stats(hist, topn) is missing
 			if partitionStats.Count > 0 && (hg == nil || hg.TotalRowCount() <= 0) && (topN == nil || topN.TotalCount() <= 0) {
 				var errMsg string
diff --git a/statistics/table.go b/statistics/table.go
index 2e66e39ab8152..20bc5d38b0a5e 100644
--- a/statistics/table.go
+++ b/statistics/table.go
@@ -377,19 +377,19 @@ func (t *Table) ColumnByName(colName string) *Column {
 }
 
 // GetStatsInfo returns their statistics according to the ID of the column or index, including histogram, CMSketch, TopN and FMSketch.
-func (t *Table) GetStatsInfo(ID int64, isIndex bool) (int64, *Histogram, *CMSketch, *TopN, *FMSketch) {
+func (t *Table) GetStatsInfo(ID int64, isIndex bool) (int64, *Histogram, *CMSketch, *TopN, *FMSketch, bool) {
 	if isIndex {
 		if idxStatsInfo, ok := t.Indices[ID]; ok {
-			return int64(idxStatsInfo.TotalRowCount()), idxStatsInfo.Histogram.Copy(), idxStatsInfo.CMSketch.Copy(), idxStatsInfo.TopN.Copy(), idxStatsInfo.FMSketch.Copy()
+			return int64(idxStatsInfo.TotalRowCount()), idxStatsInfo.Histogram.Copy(), idxStatsInfo.CMSketch.Copy(), idxStatsInfo.TopN.Copy(), idxStatsInfo.FMSketch.Copy(), true
 		}
 		// newly added index which is not analyzed yet
-		return 0, nil, nil, nil, nil
+		return 0, nil, nil, nil, nil, false
 	}
 	if colStatsInfo, ok := t.Columns[ID]; ok {
-		return int64(colStatsInfo.TotalRowCount()), colStatsInfo.Histogram.Copy(), colStatsInfo.CMSketch.Copy(), colStatsInfo.TopN.Copy(), colStatsInfo.FMSketch.Copy()
+		return int64(colStatsInfo.TotalRowCount()), colStatsInfo.Histogram.Copy(), colStatsInfo.CMSketch.Copy(), colStatsInfo.TopN.Copy(), colStatsInfo.FMSketch.Copy(), true
 	}
 	// newly added column which is not analyzed yet
-	return 0, nil, nil, nil, nil
+	return 0, nil, nil, nil, nil, false
 }
 
 // GetColRowCount tries to get the row count of the a column if possible.