App Upload

Requirements before uploading an app

Docker software must be installed on your machine. Docker’s installation procedure can be found here.
Python3 and PiP3 are required to use the Python script to upload an app. The installation of Python3 can be found here.
Your application should be uploaded to Docker Hub. For this, you will need a Docker Hub account. You can sign up for a new account here. Optionally, your application can be created on your local machine.

Dockerizing your app

Your app must be in a Docker format, as this is currently the only accepted format. In the future, other formats will also be allowed. Instructions for creating a Docker file can be found at the following link: https://docs.docker.com/develop/develop-images/dockerfile_best-practices/
You need to have your Docker image on Docker Hub. To do this, you need to log in to docker site and create a new public repository with your chosen name. You will receive the push command docker push username/test:tagname given that username is your username.
Once both steps have been performed, the image can be build and pushed to Docker Hub using the following commands: docker build . -t {yourImage:tag, e.g., username/test:tagname} and docker push {yourImage:tag, e.g., username/test:tagname}. Docker Hub will ask for your credentials.

Uploading process

Currently, an app is uploaded using a Python script. To upload an app using the Python script, perform the following three steps:

Make sure that you have the Python packages docker and requests installed. You can install them using the following commands:
```
pip3 install docker
pip3 install requests  
```

Before running the Python script create_resources_and_upload_app_local.py, the following information must be updated according to your settings.

You may need to update the settings of the Connector API. The Connector API settings are set by default as shown below. Your Connector API settings can be found in the application.properties file of the Dataspace Connector.

##########################
# APPSTORE API SETTINGS #
##########################
apiUser = "admin"
apiPassword = "password"
host = "localhost"
port = 8080

Adjust the App Store registry settings, which are the settings of Harbor if you followed the usual installation procedure. As part of the App Store configuration, you have already set the corresponding information in the application.properties file of the App Store.

########################################
# APPSTORE REGISTRY SETTINGS (HARBOR)  #
########################################
registry_address = "app.registry.example.org"
registry_repo_name = "library"
registry_user = "admin"
registry_password = "password"

Update the Docker settings. If the given image is not found locally, it will be cloned from the Docker Hub.

###################
# DOCKER SETTINGS #
###################
resource_id_tag_version = "latest"
image_name = "ahemid:idsapp"
resource_version = 1

Lastly, metadata of the app to be uploaded must be specified. For this, please update the JSON objects in the methods given below (section APP METADATA METHODS in the Python file).

Individualize the create_resource() method. Specify the title of the resource and give a resource description. List keywords for the search mechanism. Provide information about the publisher and the owner of the data, the resources’s license, and the abbreviation of the language used. Please find information about the range of the object properties in the IDS Information Model documentation: https://international-data-spaces-association.github.io/InformationModel/docs/index.html. Information about classes can be found there as well.

def create_resource():
 json = {
     "title": "DataProcessingApp", 
     "description": "data app for processing data.",
     "keywords": [ 
         "data",
         "processing",
         "fit"
     ],
     "publisher": "https://fit.fraunhofer.de",
     "sovereign": "https://fit.fraunhofer.de",
     "language": "EN",
     "license": "https://www.apache.org/licenses/LICENSE-2.0",
     "paymentMethod": "free" 
 }
 loc = post_request_check_response(f"{combined_host}/api/resources", json)
 return loc

Update the create_representation() method.

def create_representation():
 json = {
     "title": "Docker Representation",
     "description": "This is the docker representation for the DataProcessingApp",
     "language": "EN",
     "runtimeEnvironment": "docker",
     "distributionService": "https://localhost:8080"
 }
 loc = post_request_check_response(f"{combined_host}/api/representations", json)
 return loc

Adjust the create_dataApp() method.

def create_dataApp():
 json = {
     "title": "DataApp Information",
     "description": "This is the dataApp information for the DataProcessingApp.",
     "docs": "App-related human-readable documentation.",
     "environmentVariables": "Env1=environmentvariable;Env2=environmentvariable2",
     "storageConfig": "/data/temp:/temp",
     "supportedUsagePolicies": [
         "PROVIDE_ACCESS"
     ]
 }
 loc = post_request_check_response(f"{combined_host}/api/apps", json)
 return loc

Endpoints can be created using the create_endpoints() method. Here, an example of an input endpoint is given.

def create_endpoints():
 json = {
     "title": "DataApp Input Endpoint",
     "description": "This is the input endpoint for the DataProcessingApp.",
     "location": "/input",
     "mediaType": "application/json",
     "port": 5000,
     "protocol": "HTTP/1.1",
     "type": "Input",
     "docs": "https://app.swaggerhub.com/apis/app/1337",
     "info": "More information about the endpoint",
     "path": "/input"
 }
 loc = post_request_check_response(f"{combined_host}/api/endpoints", json)
 return loc

The artifact is an instance of a representation materialized at a partiuclar version and point in time. Give the title and the description for the artifact.

def create_artifact():
 json = {
     "title": "DataApp Template",
     "description": "This is the template for the DataProcessingApp",
     "value": ""
 }
 loc = post_request_check_response(f"{combined_host}/api/artifacts", json)
 return loc

To upload your app, running the following command: python3 create_resources_and_upload_app.py