Add ClickHouse module

release-notes/2.2.0.md

@@ -15,7 +15,7 @@
 
 ## Library Changes and Additions
 
-- Added `mysql-event-listener`, `session-property-manager`, and
+- Added `clickhouse`, `mysql-event-listener`, `session-property-manager`, and
   `resource-groups` modules.
 - Added named volume to `insights` module to persist Postgres data.
 - Changed all `/etc/starburst/` references to `etc/` in properties files.

src/lib/minitrino.env

@@ -1,5 +1,6 @@
 COMPOSE_PROJECT_NAME=minitrino
 
+CLICKHOUSE_VER=23.10-alpine
 CURL_VER=8.4.0
 DB2_VER=11.5.8.0
 ELASTICSEARCH_VER=8.11.0

src/lib/modules/catalog/clickhouse/clickhouse.yaml

@@ -0,0 +1,19 @@
+---
+services:
+
+  trino:
+    volumes:
+      - ./modules/catalog/clickhouse/resources/trino/clickhouse.properties:${ETC}/catalog/clickhouse.properties
+
+  clickhouse:
+    image: clickhouse/clickhouse-server:${CLICKHOUSE_VER}
+    container_name: clickhouse
+    env_file:
+      - ./modules/catalog/clickhouse/resources/clickhouse/clickhouse.env
+    ports:
+      - :8123
+    volumes:
+      - ./modules/catalog/clickhouse/resources/clickhouse/init.sh:/docker-entrypoint-initdb.d/init.sh
+    labels:
+      - com.starburst.tests=minitrino
+      - com.starburst.tests.module.clickhouse=catalog-clickhouse

src/lib/modules/catalog/clickhouse/metadata.json

@@ -0,0 +1,5 @@
+{
+    "description": "ClickHouse catalog module",
+    "incompatibleModules": [],
+    "enterprise": false
+}

src/lib/modules/catalog/clickhouse/readme.md

@@ -0,0 +1,24 @@
+# ClickHouse Catalog Module
+
+This module provisions a ClickHouse server with some preloaded data. The
+preloaded tables are stored in the `minitrino` ClickHouse database, which is
+exposed as the `clickhouse.minitrino` schema in Trino.
+
+## Loading Data
+
+Data is loaded by mounting a shell script to the ClickHouse container's
+`docker-entrypoint-initdb.d/` directory. Scripts in this directory are executed
+whenever the container boots up, and is the ClickHouse-supported method for
+bootstrapping ClickHouse containers.
+
+The init script can be edited in the library to load different tables and/or
+additional data:
+
+    lib/modules/catalog/clickhouse/resources/clickhouse/init.sh
+
+## Usage
+
+    minitrino --env STARBURST_VER=<ver> provision --module clickhouse
+    docker exec -it trino bash 
+    trino-cli
+    trino> show tables in clickhouse.minitrino;

src/lib/modules/catalog/clickhouse/resources/clickhouse/clickhouse.env

@@ -0,0 +1,4 @@
+CLICKHOUSE_DB=minitrino
+CLICKHOUSE_USER=admin
+CLICKHOUSE_PASSWORD=trinoRocks15
+CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT=1

src/lib/modules/catalog/clickhouse/resources/clickhouse/init.sh

@@ -0,0 +1,52 @@
+#!/bin/bash
+
+set -e
+
+echo "Creating three sample tables..."
+
+clickhouse client -n <<-EOSQL
+    CREATE TABLE IF NOT EXISTS minitrino.table1 (
+        id UInt32,
+        name String,
+        value Float64
+    ) ENGINE = MergeTree()
+    ORDER BY id;
+
+    CREATE TABLE IF NOT EXISTS minitrino.table2 (
+        id UInt32,
+        category String,
+        amount Decimal(10, 2)
+    ) ENGINE = MergeTree()
+    ORDER BY id;
+
+    CREATE TABLE IF NOT EXISTS minitrino.table3 (
+        id UInt32,
+        timestamp DateTime,
+        is_active UInt8
+    ) ENGINE = MergeTree()
+    ORDER BY id;
+
+    -- Insert random data into table1
+    INSERT INTO minitrino.table1
+    SELECT
+        number AS id,
+        concat('Name_', toString(number % 100)) AS name,
+        rand() % 10000 / 100.0 AS value
+    FROM numbers(1000);
+
+    -- Insert random data into table2
+    INSERT INTO minitrino.table2
+    SELECT
+        number AS id,
+        concat('Category_', toString(rand() % 10)) AS category,
+        rand() % 5000 / 100.0 AS amount
+    FROM numbers(1000);
+
+    -- Insert random data into table3
+    INSERT INTO minitrino.table3
+    SELECT
+        number AS id,
+        now() - number * 60 AS timestamp,
+        rand() % 2 AS is_active
+    FROM numbers(1000);
+EOSQL

src/lib/modules/catalog/clickhouse/resources/trino/clickhouse.properties

@@ -0,0 +1,5 @@
+connector.name=clickhouse
+connection-url=jdbc:clickhouse://clickhouse:8123
+connection-user=admin
+connection-password=trinoRocks15
+case-insensitive-name-matching=true

src/test/src/lib/json/clickhouse.json

@@ -0,0 +1,70 @@
+{
+  "tests": [
+    {
+      "type": "query",
+      "name": "Test create schema",
+      "sql": "CREATE SCHEMA clickhouse.test_schema",
+      "trinoCliArgs": [
+        "--server 'http://trino:8080'",
+        "--user admin"
+      ],
+      "contains": [
+        "CREATE SCHEMA"
+      ]
+    },
+    {
+      "type": "query",
+      "name": "Test schema list",
+      "sql": "SHOW SCHEMAS IN clickhouse",
+      "trinoCliArgs": [
+        "--server 'http://trino:8080'",
+        "--user admin"
+      ],
+      "contains": [
+        "information_schema",
+        "test_schema",
+        "minitrino"
+      ]
+    },
+    {
+      "type": "query",
+      "name": "Test table list",
+      "sql": "SHOW TABLES IN clickhouse.minitrino",
+      "trinoCliArgs": [
+        "--server 'http://trino:8080'",
+        "--user admin"
+      ],
+      "contains": [
+        "table1",
+        "table2",
+        "table3"
+      ]
+    },
+    {
+      "type": "query",
+      "name": "Test CTAS",
+      "sql": "CREATE TABLE clickhouse.test_schema.test_table AS SELECT * FROM tpch.tiny.customer LIMIT 10",
+      "trinoCliArgs": [
+        "--server 'http://trino:8080'",
+        "--user admin"
+      ],
+      "contains": [
+        "CREATE TABLE: 10 rows"
+      ]
+    },
+    {
+      "type": "query",
+      "name": "Test table select",
+      "sql": "SELECT * FROM clickhouse.test_schema.test_table LIMIT 10",
+      "trinoCliArgs": [
+        "--server 'http://trino:8080'",
+        "--user admin"
+      ],
+      "contains": [
+        "custkey",
+        "name",
+        "address"
+      ]
+    }
+  ]
+}