Fix primary key constraint (#66)

Author: ryannedolan

.github/workflows/integration-tests.yml

@@ -59,11 +59,18 @@ jobs:
         kubectl describe kafkas -n kafka
         kubectl describe flinkdeployments
         kubectl describe subscriptions
+    - name: Capture Flink Job Logs
+      if: always()
+      run: |
+        kubectl logs $(kubectl get pods -l component=jobmanager -o name) --since=0s || echo "skipped."
+        kubectl logs $(kubectl get pods -l component=taskmanager -o name) --since=0s || echo "skipped."
     - name: Capture Hoptimator Operator Logs
       if: always()
       run: kubectl logs $(kubectl get pods -l app=hoptimator-operator -o name)
     - name: Capture Flink Operator Logs
       if: always()
       run: kubectl logs $(kubectl get pods -l app.kubernetes.io/name=flink-kubernetes-operator -o name)
+    - name: Capture Flink Job Logs
+      if: always()
+      run: kubectl logs $(kubectl get pods -l app.kubernetes.io/name=flink-kubernetes-operator -o name)
 
-

Makefile

@@ -25,8 +25,7 @@ quickstart: build deploy-dev-environment deploy
 deploy-dev-environment: 
 	kubectl create -f https://github.com/jetstack/cert-manager/releases/download/v1.8.2/cert-manager.yaml || echo "skipping"
 	kubectl create namespace kafka || echo "skipping"
-	kubectl create namespace mysql || echo "skipping"
-	helm repo add flink-operator-repo https://downloads.apache.org/flink/flink-kubernetes-operator-1.6.1/
+	helm repo add flink-operator-repo https://downloads.apache.org/flink/flink-kubernetes-operator-1.9.0/
 	helm upgrade --install --atomic --set webhook.create=false flink-kubernetes-operator flink-operator-repo/flink-kubernetes-operator
 	kubectl apply -f "https://strimzi.io/install/latest?namespace=kafka" -n kafka
 	kubectl wait --for=condition=Established=True crds/kafkas.kafka.strimzi.io

deploy/dev/mysql.yaml

@@ -2,9 +2,9 @@
 apiVersion: hoptimator.linkedin.com/v1alpha1
 kind: Subscription
 metadata:
-  name: products
+  name: names
 spec:
-  sql: SELECT "quantity", "product_id" AS KEY FROM INVENTORY."products_on_hand"
+  sql: SELECT NAME, NAME AS KEY FROM DATAGEN.PERSON
   database: RAWKAFKA
 
 

deploy/samples/subscriptions.yaml

@@ -15,12 +15,6 @@ SELECT * FROM DATAGEN.COMPANY;
 !insert into RAWKAFKA."test-sink" SELECT AGE AS PAYLOAD, NAME AS KEY FROM DATAGEN.PERSON
 SELECT * FROM RAWKAFKA."test-sink" LIMIT 5;
 
--- MySQL CDC tables
-SELECT * FROM INVENTORY."products_on_hand" LIMIT 1;
-
--- Test check command
-!check not empty SELECT * FROM INVENTORY."products_on_hand";
-
--- MySQL CDC -> Kafka (via sample subscription "products")
-SELECT * FROM RAWKAFKA."products" LIMIT 1;
+-- read from sample subscription "names"
+SELECT * FROM RAWKAFKA."names" LIMIT 1;
 

etc/integration-tests.sql

@@ -3,5 +3,4 @@
 
 SELECT * FROM DATAGEN.PERSON;
 SELECT * FROM DATAGEN.COMPANY;
-SELECT * FROM INVENTORY."products_on_hand" LIMIT 1;
 SELECT * FROM RAWKAFKA."test-sink" LIMIT 0;

etc/readiness-probe.sql

@@ -170,7 +170,7 @@ public SqlNode visit(SqlCall call) {
    * Implements a CREATE TABLE...WITH... DDL statement.
    *
    * N.B. the following magic:
-   *  - field 'KEY' is treated as a PRIMARY KEY
+   *  - field 'PRIMARY_KEY' is treated as a PRIMARY KEY
    */
   class ConnectorImplementor implements ScriptImplementor {
     private final String database;
@@ -192,9 +192,9 @@ public void implement(SqlWriter w) {
       (new CompoundIdentifierImplementor(database, name)).implement(w);
       SqlWriter.Frame frame1 = w.startList("(", ")");
       (new RowTypeSpecImplementor(rowType)).implement(w);
-      if (rowType.getField("KEY", true, false) != null) {
+      if (rowType.getField("PRIMARY_KEY", true, false) != null) {
         w.sep(",");
-        w.literal("PRIMARY KEY (KEY) NOT ENFORCED");
+        w.literal("PRIMARY KEY (PRIMARY_KEY) NOT ENFORCED");
       }
       w.endList(frame1);
       // TODO support PARTITIONED BY for Tables that support it

hoptimator-catalog/src/main/java/com/linkedin/hoptimator/catalog/ScriptImplementor.java

@@ -31,8 +31,9 @@ public Schema create(SchemaPlus parentSchema, String name, Map<String, Object> o
       .with("KEY", DataType.VARCHAR);
     ConfigProvider connectorConfigProvider = ConfigProvider.from(clientConfig)
       .withPrefix("properties.")
-      .with("connector", "upsert-kafka")
+      .with("connector", "kafka")
       .with("key.format", "csv")
+      .with("key.fields", "KEY")
       .with("value.format", "csv")
       .with("value.fields-include", "EXCEPT_KEY")
       .with("topic", x -> x);

hoptimator-kafka-adapter/src/main/java/com/linkedin/hoptimator/catalog/kafka/RawKafkaSchemaFactory.java

@@ -14,17 +14,3 @@ schemas:
       bootstrap.servers: one-kafka-bootstrap.kafka.svc:9092
       group.id: hoptimator-test
       auto.offset.reset: earliest
-
-- name: INVENTORY
-  type: custom
-  factory: com.linkedin.hoptimator.catalog.mysql.MySqlCdcSchemaFactory
-  operand:
-    username: root
-    password: debezium
-    hostname: mysql.mysql.svc.cluster.local
-    port: 3306
-    database: inventory
-    urlSuffix: "?useUnicode=true&characterEncoding=UTF-8&useSSL=false"
-    connectorConfig:
-      scan.incremental.snapshot.enabled: false
-