Core: Include summary without 'operation' when comparing view versions

apache · Sep 29, 2023 · 8ae63c2 · 8ae63c2
1 parent 8062aef
commit 8ae63c2
Show file tree

Hide file tree

Showing 2 changed files with 48 additions and 2 deletions.
diff --git a/core/src/main/java/org/apache/iceberg/view/ViewMetadata.java b/core/src/main/java/org/apache/iceberg/view/ViewMetadata.java
@@ -321,14 +321,20 @@ private int reuseOrCreateNewViewVersionId(ViewVersion viewVersion) {
 
     /**
      * Checks whether the given view versions would behave the same while ignoring the view version
-     * id, the creation timestamp, and the summary.
+     * id, the creation timestamp, and the operation.
      *
      * @param one the view version to compare
      * @param two the view version to compare
      * @return true if the given view versions would behave the same
      */
     private boolean sameViewVersion(ViewVersion one, ViewVersion two) {
-      return Objects.equals(one.representations(), two.representations())
+      // ignore the "operation" contained in the summary for the comparison
+      Map<String, String> summaryOne = Maps.newHashMap(one.summary());
+      Map<String, String> summaryTwo = Maps.newHashMap(two.summary());
+      summaryOne.remove("operation");
+      summaryTwo.remove("operation");
+      return Objects.equals(summaryOne, summaryTwo)
+          && Objects.equals(one.representations(), two.representations())
           && Objects.equals(one.defaultCatalog(), two.defaultCatalog())
           && Objects.equals(one.defaultNamespace(), two.defaultNamespace())
           && one.schemaId() == two.schemaId();

diff --git a/core/src/test/java/org/apache/iceberg/view/TestViewMetadata.java b/core/src/test/java/org/apache/iceberg/view/TestViewMetadata.java
@@ -578,6 +578,46 @@ public void viewVersionDeduplication() {
             ImmutableViewVersion.builder().from(viewVersionThree).versionId(3).build());
   }
 
+  @Test
+  public void viewVersionDeduplicationWithCustomSummary() {
+    // all view versions have the same ID
+    // additionally, there are duplicate view versions that only differ in the summary
+    ViewVersion viewVersionOne = newViewVersion(1, "select * from ns.tbl");
+    ViewVersion viewVersionTwo = newViewVersion(1, "select count(*) from ns.tbl");
+    ViewVersion viewVersionOneUpdated =
+        ImmutableViewVersion.builder()
+            .from(viewVersionOne)
+            .summary(ImmutableMap.of("operation", "replace"))
+            .build();
+    ViewVersion viewVersionTwoUpdated =
+        ImmutableViewVersion.builder()
+            .from(viewVersionTwo)
+            .summary(ImmutableMap.of("operation", "replace", "key", "val"))
+            .build();
+
+    ViewMetadata viewMetadata =
+        ViewMetadata.builder()
+            .setLocation("custom-location")
+            .addSchema(new Schema(Types.NestedField.required(1, "x", Types.LongType.get())))
+            .addVersion(viewVersionOne)
+            .addVersion(viewVersionTwo)
+            .addVersion(viewVersionOneUpdated)
+            .addVersion(viewVersionTwoUpdated)
+            .setCurrentVersionId(3)
+            .build();
+
+    assertThat(viewMetadata.currentVersion())
+        .isEqualTo(ImmutableViewVersion.builder().from(viewVersionTwoUpdated).versionId(3).build());
+
+    // IDs of the view versions should be re-assigned and view versions should be de-duplicated
+    assertThat(viewMetadata.versions())
+        .hasSize(3)
+        .containsExactly(
+            viewVersionOne,
+            ImmutableViewVersion.builder().from(viewVersionTwo).versionId(2).build(),
+            ImmutableViewVersion.builder().from(viewVersionTwoUpdated).versionId(3).build());
+  }
+
   @Test
   public void schemaIDReassignment() {
     Schema schemaOne = new Schema(5, Types.NestedField.required(1, "x", Types.LongType.get()));