Update Docker files and other settings required to start the application with Docker/compose

Author: kinow

.github/workflows/ci-build.yml

@@ -18,9 +18,6 @@ jobs:
           ref: ${{ github.event.pull_request.head.ref }}
           repository: ${{ github.event.pull_request.head.repo.full_name }}
 
-      - name: MongoDB in GitHub Actions
-        uses: supercharge/mongodb-github-action@1.3.0
-
       - name: Restore dependency cache
         uses: actions/cache@v2.1.4
         with:

CONTRIBUTING.md

@@ -78,8 +78,8 @@ Java annotation based.
 Templates for the view use [Thymeleaf](http://www.thymeleaf.org/), 
 which allows them to be displayed in browsers as static prototypes.
 
-MongoDB is used to store information about `Workflow` and `QueuedWorkflow` 
-objects using [Spring Data JPA](https://docs.spring.io/spring-data/mongodb/docs/current/reference/html/).
+PostgreSQL is used to store information about `Workflow` and `QueuedWorkflow` 
+objects using [Spring Data JPA](https://docs.spring.io/spring-data/jpa/docs/current/reference/html/).
 
 The application also uses a triple store to keep the RDF representing 
 workflows (gathered from [cwltool](https://github.com/common-workflow-language/cwltool)'s 

Dockerfile

@@ -45,7 +45,7 @@ ADD pom.xml LICENSE.md NOTICE.md README.md /usr/src/app/
 
 # add src/ (which often change)
 ADD src /usr/src/app/src
-# Skip tests while building as that requires a local mongodb
+# Skip tests while building as that requires a local postgres
 RUN mvn clean package -DskipTests && cp target/cwlviewer-*.jar /usr/lib/cwlviewer.jar && rm -rf target
 
 # NOTE: ~/.m2/repository is a VOLUME and so will be deleted anyway

README.md

@@ -59,16 +59,16 @@ If you have modified the source code, then you may want to build the docker imag
 
     docker build -t commonworkflowlanguage/cwlviewer .
 
-## Running Spring Boot locally for development, with MongoDB and Jena Fuseki in Docker
+## Running Spring Boot locally for development, with PostgreSQL and Jena Fuseki in Docker
 
 Create `docker-compose.override.yml`:
 
 ```
 version: '3.2'
 services:
-  mongo:
+  postgres:
     ports:
-     - "27017:27017"
+     - "5432:5432"
   sparql:
     ports:
      - "3030:3030"
@@ -94,21 +94,21 @@ To completely reset the state, you must delete the data volumes:
 
 ```
 docker-compose down
-docker volume rm  cwlviewer_bundle cwlviewer_git cwlviewer_graphviz cwlviewer_mongo cwlviewer_sparql
+docker volume rm  cwlviewer_bundle cwlviewer_git cwlviewer_graphviz cwlviewer_postgres cwlviewer_sparql
 ```
 
 ## Running without Docker
 
 ### Requirements
 
-#### MongoDB
+#### PostgreSQL
 
-You will need to have [MongoDB](https://www.mongodb.com/) running,
-by default on `localhost:27017`
+You will need to have [PostgreSQL](https://www.postgresql.org/) running,
+by default on `localhost:5432`
 
 If you are running from the command line, you can override this by supplying
-system properties like `-Dspring.data.mongodb.host=mongo.example.org` and
-`-Dspring.data.mongodb.port=1337`
+system properties like `-Dspring.datasource.url=jdbc:postgresql://localhost:5432/cwlviewer` and
+`-Dspring.datasource.password=sa`
 
 #### Apache Jena Fuseki (or alternative SPARQL server)
 

docker-compose.yml

@@ -3,7 +3,7 @@ volumes:
   bundle:
   git:
   graphviz:
-  mongo:
+  postgres:
   sparql:
 services:
   spring:
@@ -15,7 +15,7 @@ services:
     ports:
      - "8080:8080"
     links:
-     - mongo
+     - postgres
      - sparql
     volumes:
      - type: volume
@@ -29,19 +29,26 @@ services:
        target: /data/graphviz
     environment:
      - SPARQL_ENDPOINT=http://sparql:3030/cwlviewer/
-     - SPRING_DATA_MONGODB_HOST=mongo
-     - SPRING_DATA_MONGODB_PORT=27017
+     - SPRING_DATASOURCE_URL=jdbc:postgresql://postgres:5432/cwlviewer
+     - SPRING_DATASOURCE_USERNAME=sa
+     - SPRING_DATASOURCE_PASSWORD=sa
      - BUNDLESTORAGE=/data/bundle
      - GITSTORAGE=/data/git
      - GRAPHVIZSTORAGE=/data/graphviz
-  mongo:
+  postgres:
     restart: always
-    image: mongo:3.4
-    command: mongod --setParameter="internalQueryExecMaxBlockingSortBytes=335544320"
+    image: postgres:14-alpine
+    command: postgres -c listen_addresses='*' -c log_statement=all
+    environment:
+      - POSTGRES_USER=sa
+      - POSTGRES_PASSWORD=sa
+      - POSTGRES_DB=cwlviewer
+      - PGDATA=/var/lib/postgresql/data/cwlviewer/
     volumes:
       - type: volume
-        source: mongo
-        target: /data/db
+        source: postgres
+        target: /var/lib/postgresql/data
+      - ./pg_hba.conf/:/var/lib/postgresql/data/pg_hba.conf
   sparql:
     restart: always
 ## For debugging, expose the Fuseki port by enabling:

pga_hba.conf

@@ -0,0 +1 @@
+host all all 0.0.0.0/0 trust

src/main/java/org/commonwl/view/workflow/QueuedWorkflowRepository.java

@@ -7,7 +7,10 @@
 import java.util.List;
 
 /**
- * Stores workflows in the queue waiting for cwltool
+ * Stores workflows in the queue waiting for cwltool.
+ *
+ * <p>Use only queries without objects and JSON here. For other methods use the Impl class to avoid
+ * issues with serialization.</p>
  */
 public interface QueuedWorkflowRepository extends JpaRepository<QueuedWorkflow, String>, QueuedWorkflowRepositoryCustom {
 
@@ -17,7 +20,7 @@ public interface QueuedWorkflowRepository extends JpaRepository<QueuedWorkflow,
      * @param retrievedOn Date of when the queued workflow was retrieved
      * @return The number of queued workflows deleted
      */
-    @Query(value = "delete from queued_workflow q where q.tempRepresentation->>retrieved_on <= ?1", nativeQuery = true)
+    @Query(value = "DELETE FROM queued_workflow q WHERE q.tempRepresentation ->> 'retrieved_on' <= ?1", nativeQuery = true)
     Long deleteByTempRepresentation_RetrievedOnLessThanEqual(Date retrievedOn);
 
     /**
@@ -26,7 +29,7 @@ public interface QueuedWorkflowRepository extends JpaRepository<QueuedWorkflow,
      * @param retrievedOn Details of where the queued workflow is from
      * @return A list of queued workflows
      */
-    @Query(value = "select q.* from queued_workflow q where q.tempRepresentation->>retrieved_on <= ?1", nativeQuery = true)
+    @Query(value = "SELECT q.* FROM queued_workflow q WHERE q.tempRepresentation ->> 'retrieved_on' <= ?1", nativeQuery = true)
     List<QueuedWorkflow> findByTempRepresentation_RetrievedOnLessThanEqual(Date retrievedOn);
 
 }

src/main/java/org/commonwl/view/workflow/WorkflowRepository.java

@@ -30,8 +30,8 @@
 /**
  * Stores and retrieved workflow objects from the database
  *
- * See Spring Data MongoDB docs:
- * https://docs.spring.io/spring-data/data-mongo/docs/current/reference/html/
+ * See Spring Data JPA docs:
+ * https://docs.spring.io/spring-data/jpa/docs/current/reference/html/
  */
 public interface WorkflowRepository extends JpaRepository<Workflow, String> {
 
@@ -48,7 +48,7 @@ public interface WorkflowRepository extends JpaRepository<Workflow, String> {
      * @param path The path to the workflow within the repository
      * @return The workflow model
      */
-    @Query(value = "select * from workflow w where w.lastCommit = ?1 and w.retrievedFrom->>path = ?2", nativeQuery = true)
+    @Query(value = "SELECT w.* FROM workflow w WHERE w.lastCommit = ?1 AND w.retrievedFrom ->> 'path' = ?2", nativeQuery = true)
     List<Workflow> findByCommitAndPath(String commitId, String path);
 
     /**
@@ -58,7 +58,7 @@ public interface WorkflowRepository extends JpaRepository<Workflow, String> {
      *            The latest commit ID of the workflow
      * @return The workflow model
      */
-    @Query("select w from Workflow w where w.lastCommit = ?1")
+    @Query("SELECT w FROM Workflow w WHERE w.lastCommit = ?1")
     List<Workflow> findByCommit(String commitId);
 
     /**