gensolution 0.0.8

Generates solution package when locally developing SAP DI operators

genjson

Generates the SAP Data Intelligence (SAP DI) solution files that enables to locally code new custom python operators and prepares for uploading to an SAP DI instance as a solution.

A specific folder structure is required in order use the gensolution that complies to a general github repository setup. This can be setup with gensolution --project.

In addition the operator script needs to provide additional information that is used to create the operator description files operator.json and configSchema.json.

Required Folder Structure

• project
  • src: containing the source code of the operator packages
    • package1
      • [operator] - operator is placeholder for the name of the operator
        • [operator].py
      • package2
        • ...
  • solution: containing the solution folders that corresponds to the SAP DI python subengine folder structure. The structure is generated if not existing.
    • operators
      • package_version.zip (solution for uploading to a SAP DI instance
      • package_version
        • manifest.json
        • content
          • files
            • ...

Usage

The genjson-script needs to be exectuted at the root of the project folder (Exception when using ```gensolution --project``).

usage: gensolution.py [-h] [--project PROJECT] [--version VERSION] [--debug]
                      [--force]

Generate SAP Data Intelligence solution for local operator code

optional arguments:
  -h, --help         show this help message and exit
  --project PROJECT  Creates new project with folder structure for locally
                     programming operators
  --version VERSION  version format <num.num.num>
  --debug            for debug-level information
  --force            removes subdirectories from <solution/operators>

Operator Template

def process(msg):

    # core process 
    # start custom code
    filename = api.config.filename
    with open(filename, 'r') as fin :
        data = fin.read()
        fin.close()
    # end custom code

    # return message
    return  api.Message(attributes = {'filename':api.config.filename}, body=data), log_stream.getvalue()

### definitions for locally coding

try:
    api
except NameError:
    class api:
        # default input - only for testing
        def get_default_input():
            body = '0'
            attributes = {'format': 'csv', "storage.filename": "test_file.txt"}
            return api.Message(attributes=attributes, body=body)

        # config parameter are required for testing and generating configSchema.json
        class config:
            # operator infos necessary for solution import
            tags = {'python36': ''}  # tags that helps to select the appropriate container
            operator_description = 'Read File from Container'
            version = "0.0.2"  # for creating the manifest.json

            # operator parameter for config the operator
            config_params=dict()
            filename = './data/test_file.txt'
            #info for necessary for solution import
            config_params['filename'] = {'title': 'Filename', 'description':'Filename (path)', 'type':'string'}

        class Message:
            def __init__(self,body = None,attributes = ""):
                self.body = body
                self.attributes = attributes

        def send(port, msg):
            if isinstance(msg,api.Message) :
                print('--- Attributes: --- ')
                print(msg.attributes)
                print('--- Body: --- ')
                print(msg.body)

        def set_port_callback(port, callback):
            print("Port: ",port)
            msg = api.get_default_input()
            callback(msg)

        def call(msg,config):
            api.config = config
            msg, info = process(msg)
            return msg, info

# list input and output ports with specified types for creating operator.json
inports = [{"name":"input","type":"message"}]
outports = [{"name":"output","type":"message"}]

def interface(msg):
    msg, info = process(msg)
    api.send(outports[0]["name"], msg)

# Triggers the request for every message (the message provides the stock_symbol)
api.set_port_callback(inports[0]["name"], interface)