From 47f82a6da0807b53328adfcf91b4903b6fbbc2bc Mon Sep 17 00:00:00 2001
From: Sun Dapeng <dapeng.sdp@alibaba-inc.com>
Date: Fri, 21 Aug 2026 09:43:25 +0800
Subject: [PATCH] [spark] Cover the zero a truncation reports surviving a later
 ANALYZE

A truncation reports an exact zero for the partitions it emptied. ANALYZE measures a partition and
writes what it found. Nothing pinned what the second does to the first, and the failure would be
silent in the direction that matters: a zero degraded back to unknown reads as "nobody measured
this", so a planner that guards on the value being known starts skipping the partition instead of
knowing it is empty.

The case truncates a measured partition and then runs both forms of ANALYZE over it, asserting the
zero is still a zero. It needs `TRUNCATE TABLE` and `ANALYZE TABLE` together, which is why it could
not go in with either of them.
---
 .../CatalogManagedPartitionAnalyzeTest.scala  | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/paimon-spark/paimon-spark-ut/src/test/scala/org/apache/paimon/spark/sql/CatalogManagedPartitionAnalyzeTest.scala b/paimon-spark/paimon-spark-ut/src/test/scala/org/apache/paimon/spark/sql/CatalogManagedPartitionAnalyzeTest.scala
index 89d57134f89c..b8d6b9a32077 100644
--- a/paimon-spark/paimon-spark-ut/src/test/scala/org/apache/paimon/spark/sql/CatalogManagedPartitionAnalyzeTest.scala
+++ b/paimon-spark/paimon-spark-ut/src/test/scala/org/apache/paimon/spark/sql/CatalogManagedPartitionAnalyzeTest.scala
@@ -42,6 +42,33 @@ import scala.collection.JavaConverters._
  */
 class CatalogManagedPartitionAnalyzeTest extends PaimonSparkTestWithRestCatalogBase {
 
+  test("ANALYZE after a TRUNCATE keeps the exact zero the truncation reported") {
+    val tableName = "analyze_after_truncate"
+    withTable(tableName) {
+      createTable(tableName)
+      writeCsvPartition(tableName, "20260101", "00", 1)
+      repair(tableName)
+      sql(s"ANALYZE TABLE ${qualified(tableName)} COMPUTE STATISTICS NOSCAN").collect()
+      assert(statisticsOf(tableName, "20260101", "00").fileCount() == 1L)
+
+      sql(s"TRUNCATE TABLE ${qualified(tableName)} PARTITION (dt = '20260101', hour = '00')")
+      val truncated = statisticsOf(tableName, "20260101", "00")
+      assert(truncated.fileCount() == 0L, truncated.toString)
+      assert(truncated.recordCount() == 0L, truncated.toString)
+
+      // Measuring an empty partition must not turn "nothing is here" back into "nobody measured".
+      sql(s"ANALYZE TABLE ${qualified(tableName)} COMPUTE STATISTICS NOSCAN").collect()
+      val listed = statisticsOf(tableName, "20260101", "00")
+      assert(listed.fileCount() == 0L, listed.toString)
+      assert(listed.recordCount() == 0L, listed.toString)
+
+      sql(s"ANALYZE TABLE ${qualified(tableName)} COMPUTE STATISTICS").collect()
+      val scanned = statisticsOf(tableName, "20260101", "00")
+      assert(scanned.fileCount() == 0L, scanned.toString)
+      assert(scanned.recordCount() == 0L, scanned.toString)
+    }
+  }
+
   test("ANALYZE with a full partition spec measures only that partition") {
     val tableName = "analyze_full_spec"
     withTable(tableName) {