ibis-project · cpcloud · Jan 28, 2024 · Dec 28, 2023 · Jan 28, 2024
diff --git a/.github/workflows/ibis-backends.yml b/.github/workflows/ibis-backends.yml
@@ -121,6 +121,12 @@ jobs:
               - postgres
             sys-deps:
               - libgeos-dev
+          - name: risingwave
+            title: Risingwave
+            services:
+              - risingwave
+            extras:
+              - risingwave
           - name: impala
             title: Impala
             extras:
@@ -209,6 +215,14 @@ jobs:
                 - postgres
               sys-deps:
                 - libgeos-dev
+          - os: windows-latest
+            backend:
+              name: risingwave
+              title: Risingwave
+              services:
+                - risingwave
+              extras:
+                - risingwave
           - os: windows-latest
             backend:
               name: postgres
@@ -677,7 +691,7 @@ jobs:
           python-version: ${{ matrix.python-version }}
 
       - name: remove deps that are not compatible with sqlalchemy 2
-        run: poetry remove snowflake-sqlalchemy sqlalchemy-exasol
+        run: poetry remove snowflake-sqlalchemy sqlalchemy-exasol sqlalchemy-risingwave
 
       - name: add sqlalchemy 2
         run: poetry update sqlalchemy

diff --git a/ci/schema/risingwave.sql b/ci/schema/risingwave.sql
@@ -0,0 +1,177 @@
+SET RW_IMPLICIT_FLUSH=true;
+
+DROP TABLE IF EXISTS diamonds CASCADE;
+
+CREATE TABLE diamonds (
+    carat FLOAT,
+    cut TEXT,
+    color TEXT,
+    clarity TEXT,
+    depth FLOAT,
+    "table" FLOAT,
+    price BIGINT,
+    x FLOAT,
+    y FLOAT,
+    z FLOAT
+) WITH (
+  connector = 'posix_fs',
+  match_pattern = 'diamonds.csv',
+  posix_fs.root = '/data',
+) FORMAT PLAIN ENCODE CSV ( without_header = 'false', delimiter = ',' );
+
+DROP TABLE IF EXISTS astronauts CASCADE;
+
+CREATE TABLE astronauts (
+    "id" BIGINT,
+    "number" BIGINT,
+    "nationwide_number" BIGINT,
+    "name" VARCHAR,
+    "original_name" VARCHAR,
+    "sex" VARCHAR,
+    "year_of_birth" BIGINT,
+    "nationality" VARCHAR,
+    "military_civilian" VARCHAR,
+    "selection" VARCHAR,
+    "year_of_selection" BIGINT,
+    "mission_number" BIGINT,
+    "total_number_of_missions" BIGINT,
+    "occupation" VARCHAR,
+    "year_of_mission" BIGINT,
+    "mission_title" VARCHAR,
+    "ascend_shuttle" VARCHAR,
+    "in_orbit" VARCHAR,
+    "descend_shuttle" VARCHAR,
+    "hours_mission" DOUBLE PRECISION,
+    "total_hrs_sum" DOUBLE PRECISION,
+    "field21" BIGINT,
+    "eva_hrs_mission" DOUBLE PRECISION,
+    "total_eva_hrs" DOUBLE PRECISION
+) WITH (
+  connector = 'posix_fs',
+  match_pattern = 'astronauts.csv',
+  posix_fs.root = '/data',
+) FORMAT PLAIN ENCODE CSV ( without_header = 'false', delimiter = ',' );
+
+DROP TABLE IF EXISTS batting CASCADE;
+
+CREATE TABLE batting (
+    "playerID" TEXT,
+    "yearID" BIGINT,
+    stint BIGINT,
+    "teamID" TEXT,
+    "lgID" TEXT,
+    "G" BIGINT,
+    "AB" BIGINT,
+    "R" BIGINT,
+    "H" BIGINT,
+    "X2B" BIGINT,
+    "X3B" BIGINT,
+    "HR" BIGINT,
+    "RBI" BIGINT,
+    "SB" BIGINT,
+    "CS" BIGINT,
+    "BB" BIGINT,
+    "SO" BIGINT,
+    "IBB" BIGINT,
+    "HBP" BIGINT,
+    "SH" BIGINT,
+    "SF" BIGINT,
+    "GIDP" BIGINT
+) WITH (
+  connector = 'posix_fs',
+  match_pattern = 'batting.csv',
+  posix_fs.root = '/data',
+) FORMAT PLAIN ENCODE CSV ( without_header = 'false', delimiter = ',' );
+
+DROP TABLE IF EXISTS awards_players CASCADE;
+
+CREATE TABLE awards_players (
+    "playerID" TEXT,
+    "awardID" TEXT,
+    "yearID" BIGINT,
+    "lgID" TEXT,
+    tie TEXT,
+    notes TEXT
+) WITH (
+  connector = 'posix_fs',
+  match_pattern = 'awards_players.csv',
+  posix_fs.root = '/data',
+) FORMAT PLAIN ENCODE CSV ( without_header = 'false', delimiter = ',' );
+
+DROP TABLE IF EXISTS functional_alltypes CASCADE;
+
+CREATE TABLE functional_alltypes (
+    id INTEGER,
+    bool_col BOOLEAN,
+    tinyint_col SMALLINT,
+    smallint_col SMALLINT,
+    int_col INTEGER,
+    bigint_col BIGINT,
+    float_col REAL,
+    double_col DOUBLE PRECISION,
+    date_string_col TEXT,
+    string_col TEXT,
+    timestamp_col TIMESTAMP WITHOUT TIME ZONE,
+    year INTEGER,
+    month INTEGER
+) WITH (
+  connector = 'posix_fs',
+  match_pattern = 'functional_alltypes.csv',
+  posix_fs.root = '/data',
+) FORMAT PLAIN ENCODE CSV ( without_header = 'false', delimiter = ',' );
+
+DROP TABLE IF EXISTS tzone CASCADE;
+
+CREATE TABLE tzone (
+    ts TIMESTAMP WITH TIME ZONE,
+    key TEXT,
+    value DOUBLE PRECISION
+);
+
+INSERT INTO tzone
+    SELECT
+	CAST('2017-05-28 11:01:31.000400' AS TIMESTAMP WITH TIME ZONE) +
+	    t * INTERVAL '1 day 1 second' AS ts,
+	CHR(97 + t) AS key,
+	t + t / 10.0 AS value
+    FROM generate_series(0, 9) AS t;
+
+DROP TABLE IF EXISTS array_types CASCADE;
+
+CREATE TABLE IF NOT EXISTS array_types (
+    x BIGINT[],
+    y TEXT[],
+    z DOUBLE PRECISION[],
+    grouper TEXT,
+    scalar_column DOUBLE PRECISION,
+    multi_dim BIGINT[][]
+);
+
+INSERT INTO array_types VALUES
+    (ARRAY[1, 2, 3], ARRAY['a', 'b', 'c'], ARRAY[1.0, 2.0, 3.0], 'a', 1.0, ARRAY[ARRAY[NULL::BIGINT, NULL, NULL], ARRAY[1, 2, 3]]),
+    (ARRAY[4, 5], ARRAY['d', 'e'], ARRAY[4.0, 5.0], 'a', 2.0, ARRAY[]::BIGINT[][]),
+    (ARRAY[6, NULL], ARRAY['f', NULL], ARRAY[6.0, NULL], 'a', 3.0, ARRAY[NULL, ARRAY[]::BIGINT[], NULL]),
+    (ARRAY[NULL, 1, NULL], ARRAY[NULL, 'a', NULL], ARRAY[]::DOUBLE PRECISION[], 'b', 4.0, ARRAY[ARRAY[1], ARRAY[2], ARRAY[NULL::BIGINT], ARRAY[3]]),
+    (ARRAY[2, NULL, 3], ARRAY['b', NULL, 'c'], NULL, 'b', 5.0, NULL),
+    (ARRAY[4, NULL, NULL, 5], ARRAY['d', NULL, NULL, 'e'], ARRAY[4.0, NULL, NULL, 5.0], 'c', 6.0, ARRAY[ARRAY[1, 2, 3]]);
+
+DROP TABLE IF EXISTS json_t CASCADE;
+
+CREATE TABLE IF NOT EXISTS json_t (js JSONB);
+
+INSERT INTO json_t VALUES
+    ('{"a": [1,2,3,4], "b": 1}'),
+    ('{"a":null,"b":2}'),
+    ('{"a":"foo", "c":null}'),
+    ('null'),
+    ('[42,47,55]'),
+    ('[]');
+
+DROP TABLE IF EXISTS win CASCADE;
+CREATE TABLE win (g TEXT, x BIGINT, y BIGINT);
+INSERT INTO win VALUES
+    ('a', 0, 3),
+    ('a', 1, 2),
+    ('a', 2, 0),
+    ('a', 3, 1),
+    ('a', 4, 1);
diff --git a/compose.yaml b/compose.yaml
@@ -537,6 +537,88 @@ services:
     networks:
       - impala
 
+  risingwave-minio:
+    image: "quay.io/minio/minio:latest"
+    command:
+      - server
+      - "--address"
+      - "0.0.0.0:9301"
+      - "--console-address"
+      - "0.0.0.0:9400"
+      - /data
+    expose:
+      - "9301"
+      - "9400"
+    ports:
+      - "9301:9301"
+      - "9400:9400"
+    depends_on: []
+    volumes:
+      - "risingwave-minio:/data"
+    entrypoint: /bin/sh -c "set -e; mkdir -p \"/data/hummock001\"; /usr/bin/docker-entrypoint.sh \"$$0\" \"$$@\" "
+    environment:
+      MINIO_CI_CD: "1"
+      MINIO_ROOT_PASSWORD: hummockadmin
+      MINIO_ROOT_USER: hummockadmin
+      MINIO_DOMAIN: "risingwave-minio"
+    container_name: risingwave-minio
+    healthcheck:
+      test:
+        - CMD-SHELL
+        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/9301; exit $$?;'
+      interval: 5s
+      timeout: 5s
+      retries: 20
+    restart: always
+    networks:
+      - risingwave
+
+  risingwave:
+    image: ghcr.io/risingwavelabs/risingwave:nightly-20240122
+    command: "standalone --meta-opts=\" \
+      --advertise-addr 0.0.0.0:5690 \
+      --backend mem \
+      --state-store hummock+minio://hummockadmin:hummockadmin@risingwave-minio:9301/hummock001 \
+      --data-directory hummock_001 \
+      --config-path /risingwave.toml\" \
+      --compute-opts=\" \
+      --config-path /risingwave.toml \
+      --advertise-addr 0.0.0.0:5688 \
+      --role both \" \
+      --frontend-opts=\" \
+      --config-path /risingwave.toml \
+      --listen-addr 0.0.0.0:4566 \
+      --advertise-addr 0.0.0.0:4566 \" \
+      --compactor-opts=\" \
+      --advertise-addr 0.0.0.0:6660 \""
+    expose:
+      - "4566"
+    ports:
+      - "4566:4566"
+    depends_on:
+      - risingwave-minio
+    volumes:
+      - "./docker/risingwave/risingwave.toml:/risingwave.toml"
+      - risingwave:/data
+    environment:
+      RUST_BACKTRACE: "1"
+      # If ENABLE_TELEMETRY is not set, telemetry will start by default
+      ENABLE_TELEMETRY: ${ENABLE_TELEMETRY:-true}
+    container_name: risingwave
+    healthcheck:
+      test:
+        - CMD-SHELL
+        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/6660; exit $$?;'
+        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/5688; exit $$?;'
+        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/4566; exit $$?;'
+        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/5690; exit $$?;'
+      interval: 5s
+      timeout: 5s
+      retries: 20
+    restart: always
+    networks:
+      - risingwave
+
 networks:
   impala:
     # docker defaults to naming networks "$PROJECT_$NETWORK" but the Java Hive
@@ -553,6 +635,7 @@ networks:
   oracle:
   exasol:
   flink:
+  risingwave:
 
 volumes:
   clickhouse:
@@ -563,3 +646,5 @@ volumes:
   postgres:
   exasol:
   impala:
+  risingwave-minio:
+  risingwave:
diff --git a/docker/risingwave/risingwave.toml b/docker/risingwave/risingwave.toml
@@ -0,0 +1,2 @@
+# RisingWave config file to be mounted into the Docker containers.
+# See https://github.com/risingwavelabs/risingwave/blob/main/src/config/example.toml for example
diff --git a/ibis/backends/conftest.py b/ibis/backends/conftest.py
@@ -539,6 +539,7 @@ def ddl_con(ddl_backend):
             "mysql",
             "oracle",
             "postgres",
+            "risingwave",
             "snowflake",
             "sqlite",
             "trino",
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		# RisingWave config file to be mounted into the Docker containers.
		# See https://github.com/risingwavelabs/risingwave/blob/main/src/config/example.toml for example
cpcloud marked this conversation as resolved. Show resolved Hide resolved