feat(eap): Add a column to store arrays in EAP

.github/workflows/ci.yml

@@ -386,7 +386,6 @@ jobs:
       matrix:
         version:
           [
-            "24.8.11.51285.altinitystable",
             "25.3.6.10034.altinitystable",
           ]
 

docker-compose.gcb.yml

@@ -70,7 +70,7 @@ services:
       KAFKA_LOG4J_ROOT_LOGLEVEL: "WARN"
       KAFKA_TOOLS_LOG4J_LOGLEVEL: "WARN"
   clickhouse:
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse.local
     extra_hosts:
       - "clickhouse.local:127.0.0.1"  # Add entry to /etc/hosts file
@@ -92,7 +92,7 @@ services:
   clickhouse-query:
     depends_on:
       - zookeeper
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse-query.local
     extra_hosts:
       - "clickhouse-query.local:127.0.0.1"  # Add entry to /etc/hosts file
@@ -108,7 +108,7 @@ services:
   clickhouse-01:
     depends_on:
       - zookeeper
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse-01.local
     extra_hosts:
       - "clickhouse-01.local:127.0.0.1"  # Add entry to /etc/hosts file
@@ -125,7 +125,7 @@ services:
   clickhouse-02:
     depends_on:
       - zookeeper
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse-02.local
     extra_hosts:
       - "clickhouse-02.local:127.0.0.1"  # Add entry to /etc/hosts file
@@ -142,7 +142,7 @@ services:
   clickhouse-03:
     depends_on:
       - zookeeper
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse-03.local
     extra_hosts:
       - "clickhouse-03.local:127.0.0.1"  # Add entry to /etc/hosts file
@@ -159,7 +159,7 @@ services:
   clickhouse-04:
     depends_on:
       - zookeeper
-    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:23.8.11.29.altinitystable}"
+    image: "${CLICKHOUSE_IMAGE:-ghcr.io/getsentry/image-mirror-altinity-clickhouse-server:25.3.6.10034.altinitystable}"
     hostname: clickhouse-04.local
     extra_hosts:
       - "clickhouse-04.local:127.0.0.1"  # Add entry to /etc/hosts file

snuba/snuba_migrations/events_analytics_platform/0050_add_attributes_array_column.py

@@ -0,0 +1,101 @@
+from snuba.clickhouse.columns import JSON, Column
+from snuba.clusters.storage_sets import StorageSetKey
+from snuba.migrations import migration, operations
+from snuba.migrations.columns import MigrationModifiers as Modifiers
+from snuba.migrations.operations import OperationTarget
+
+storage_set = StorageSetKey.EVENTS_ANALYTICS_PLATFORM
+table_name_prefix = "eap_items_1"
+new_columns = [
+    Column(
+        "attributes_array",
+        JSON(
+            max_dynamic_paths=128,
+            modifiers=Modifiers(
+                codecs=[
+                    "ZSTD(1)",
+                ],
+            ),
+        ),
+    ),
+]
+after = "attributes_float_39"
+sampling_weights = [8, 8**2, 8**3]
+
+
+class Migration(migration.ClickhouseNodeMigration):
+    blocking = False
+
+    def forwards_ops(self) -> list[operations.SqlOperation]:
+        ops: list[operations.SqlOperation] = [
+            operations.AddColumn(
+                storage_set=storage_set,
+                table_name=f"{table_name_prefix}_{suffix}",
+                column=new_column,
+                after=after,
+                target=target,
+            )
+            for suffix, target in [
+                ("local", OperationTarget.LOCAL),
+                ("dist", OperationTarget.DISTRIBUTED),
+            ]
+            for new_column in new_columns
+        ]
+
+        for sampling_weight in sampling_weights:
+            downsampled_table_prefix = f"eap_items_1_downsample_{sampling_weight}"
+
+            ops.extend(
+                [
+                    operations.AddColumn(
+                        storage_set=storage_set,
+                        table_name=f"{downsampled_table_prefix}_{suffix}",
+                        column=new_column,
+                        after=after,
+                        target=target,
+                    )
+                    for suffix, target in [
+                        ("local", OperationTarget.LOCAL),
+                        ("dist", OperationTarget.DISTRIBUTED),
+                    ]
+                    for new_column in new_columns
+                ]
+            )
+
+        return ops
+
+    def backwards_ops(self) -> list[operations.SqlOperation]:
+        ops: list[operations.SqlOperation] = [
+            operations.DropColumn(
+                storage_set=storage_set,
+                table_name=f"{table_name_prefix}_{suffix}",
+                column_name=new_column.name,
+                target=target,
+            )
+            for suffix, target in [
+                ("dist", OperationTarget.DISTRIBUTED),
+                ("local", OperationTarget.LOCAL),
+            ]
+            for new_column in new_columns
+        ]
+
+        for sampling_weight in sampling_weights:
+            downsampled_table_prefix = f"eap_items_1_downsample_{sampling_weight}"
+
+            ops.extend(
+                [
+                    operations.DropColumn(
+                        storage_set=storage_set,
+                        table_name=f"{downsampled_table_prefix}_{suffix}",
+                        column_name=new_column.name,
+                        target=target,
+                    )
+                    for suffix, target in [
+                        ("dist", OperationTarget.DISTRIBUTED),
+                        ("local", OperationTarget.LOCAL),
+                    ]
+                    for new_column in new_columns
+                ]
+            )
+
+        return ops