Mapping Rules Configuration

In MAS Data Lake, the device data for different device types are stored in their own table. The tables are named as IOT_<deviceType>. The configuration items that controls the transformation of extracted data from SCADA historian for a specific device type, are defined in mapping configuration file.
The location of this configuration file is <InstallRoot>/volume/config/mapping.json.

Configuration Object

Provide data configuration object in a json formated file:

{
    "serviceName": "String",
    "serviceType": "String",
    "description": "String",
    "csvFileName": "String",
    "deviceTypes": [
        {
            "type": "String",
            "tagpathFilters": [
                "String"
            ]
        }
    ],
    "metrics": {
        "name": "String",
        "value": "String",
        "unit": "String",
        "type": "String",
        "decimalAccuracy": "String",
        "label": "String"
    },
    "dimensions": {
       "tagpath": "String",
       "tagid": "String",
       "site": "String",
       "categories": "String"
    }
}

Where:

• serviceName: Defines the name of the service for example PITest1
• csvFileName: Name of the CSV file.
• deviceTypes: Specifies device type and corresponding tagpath patterns.
  • type: Specifies device type id.
  • tagpathFilters: Specifies list of tagpath patterns.
• metrics: This object defines the mapping rule for device metrics data items.
  • name: Column name in CSV file used to map metric "Name"
  • value: Metric value is extracted from PI Archive database at runtime. Set this to an empty string in this configuration object.
  • unit: Column name in CSV file used to map metric "Unit"
  • type: Column name in CSV file used to map metric "Type"
  • label: Column name in CSV file used to map metric "Label"
  • decimalAccuracy: Column name in CSV file used to map metric "Decimal Accuracy"
• dimensions: This object defines the mapping rule for device dimension data. Required items:
  • tagpath: One or more column name(s) in the CSV file used to create "tag path"
  • tagid: Column name in the CSV file used to map "tag id" Add more dimension data mapping
  • site: Column name in the CSV file used to map "site name".

Sample CSV file (pidemo.csv) used to define mapping.json configuration file

Parent,Name,ObjectType,Error,UniqueID,ParentUniqueID,Description,Categories,AttributeDefaultUOM,AttributeType,AttributeValue,AttributeDataReference,AttributeDisplayDigits
DistillExample\4820Column,Accumulator Level,Attribute,,5ef1040f-b6af-59f7-352c-0442281fb1f9,9f8a3f5d-2d1b-11ec-98b3-06abac9454c4,Liquid level in reflux accumulator,Process Parameters;,%,Double,,PI Point,-5,,,
...

Sample mapping.json Configuration File for

{
    "serviceName": "Service1",
    "csvFileName": "pidemo.csv",
    "deviceTypes": [
        {
            "type": "PIDemoType",
            "tagpathFilters": [
                "DistillExample.*"
            ]
        }
    ],
    "metrics": {
        "name": "${Name}",
        "value": "",
        "unit": "${AttributeDefaultUOM}",
        "type": "${AttributeType}",
        "decimalAccuracy": "${AttributeDisplayDigits}",
        "label": "${Description}"
    },
    "dimensions": {
       "tagpath": "${Parent},${Name}",
       "tagid": "${UniqueID}",
       "site": "IBMAustin",
       "categories": "${Categories}"
    }
}

NOTES:

• To map a column name specify the column name within curly brackets {}, For example "unit": "${AttributeDefaultUOM}. Column names are case sensitive. The "unit" in pidemo.csv example will map to "lb/h".
• You can use comma-separated list of column name(s) to map a metric or dimension data. For example "${Parent},${Name}" mapping rule is used to specify tagpath. The tagpath in this example will map to "DistillExample\4820Column\Bottoms Flow". A back-slash \ will be used as a field-separator.
• DO not use curly brackets {} to specify a fixed value. For example "site": "IBMAustin".