Mapping

Mapping is the process of defining how fields contained in the data source are treated and presented within the component. For mapping in Flexmonster, you can use the Mapping object, which is a property of the Data Source.

The Mapping object is available for all data sources but with some differences.

For JSON, CSV, and the custom data source API, it’s possible to define field data types and captions, group fields under separate dimensions, create multilevel hierarchies, and more. For SSAS and Mondrian data sources, it’s possible to set captions for dimensions and measures. For data from Elasticsearch, it’s possible to customize hierarchy captions, formats, time zones, control field visibility, and more.

We recommend using mapping instead of defining a meta-object for JSON or adding prefixes for CSV data since the former presents a powerful way to neatly separate a data source from its representation. Moreover, mapping provides more options than using prefixes for CSV data.

This guide contains the following sections:

• Mapping properties
• Ways to define the mapping
• Configuring multilevel hierarchies
• Creating several fields from one field
• Defining available aggregations for all fields of a certain type
• Other ways to customize field presentation
• Examples

Mapping properties

The Mapping object has the following properties:

• (uniqueName) – Object | Array of objects. Allows setting the mapping for a specific field from the dataset. (uniqueName)is the field’s unique name.
  Specify the mapping as an object to set it for a certain field. See an example on JSFiddle.
  For "json" and "csv" data source types, it is possible to create several fields from a certain field. It can be done by specifying the mapping as an array of objects, where each object is a new field. Learn more in our guide.
  Each object can have the following properties:
  • caption (optional) – String. The hierarchy’s caption.
  • type (optional) – String. The field’s data type. Only for "json", "csv", and "api" data source types. type can be:
    • "string" – The field stores string data.
    • "number" – The field stores numerical data. It can be aggregated with all available aggregations.
    • "month" – The field stores months. Note that if the field stores month names only (in either short or full form), the field will be recognized by Flexmonster as a field of the "month" type automatically. If the field contains custom month names, specify its type as "month" explicitly.
    • "weekday" – The field stores days of the week.
    • "date" – The field stores a date. Fields of this type are split into 3 different fields: Year, Month, Day. Only for "json" and "csv" data source types.
    • "year/month/day" – The field stores a date. It’s displayed as a multilevel hierarchy with the following levels: Year > Month > Day. Only for "json" and "csv" data source types.
    • "year/quarter/month/day" – The field is a date. It’s displayed as a multilevel hierarchy with the following levels: Year > Quarter > Month > Day. Only for "json" and "csv" data source types.
    • "date string" – The field stores a date. Fields of this type are represented as strings and can be used in rows, columns, or report filters. The component sorts members of such a field as dates.
      Note that by default, "date string" fields are rounded down with the "1d" interval  (i.e., "04/25/2021T21:30:05" is rounded down to "04/25/2021T00:00:00"). If needed, you can set a custom interval for a field. Learn more about intervals.
      Fields of the "date string" type can be formatted using the datePattern option (the default pattern is "dd/MM/yyyy").
    • "datetime" – The field stores a date. You can select fields of this type for values.
      Fields of the "datetime" type can be formatted using the dateTimePattern option (the default pattern is "dd/MM/yyyy HH:mm:ss").
    • "time" – The field stores time. The "time" type is used for fields that store a time interval (e.g., duration or elapsed time), which is not related to a specific date.
      Fields of this type can be formatted using the timePattern option (the default pattern is "HH:mm:ss").
    • "id" – The field is an id. This field is not shown in the Field List. Fields of this type can be used for editing data.
    • "property" – The field for setting member properties. This field is not shown in the Field List. For example, it can be used to associate a productID with a product. See an example here.
  • hierarchy (optional) – String. The hierarchy’s name. When configuring hierarchies, specify this property to mark the field as a level of a hierarchy or as a member property of a hierarchy (in this case, the type parameter should be set to "property"). Only for "json", "csv", and "api" data source types.
    See how to configure multilevel hierarchies.
  • parent (optional) – String. The unique name of the parent level. This property is necessary if the field is a level of a hierarchy and has a parent level. Only for "json", "csv", and "api" data source types.
    See how to configure multilevel hierarchies.
  • folder (optional) – String. The field’s folder. Folders are used to group several fields in the Field List. folder supports nesting via / (e.g., "Folder/Subfolder/"). Only for "json", "csv", and "api" data source types.
  • aggregations (optional) — Array of strings. This property represents the list of aggregation functions that can be applied to the current measure.
  • filters (optional) – Boolean. This property allows enabling and disabling the UI filters for a specific hierarchy. When set to false, the UI filters are disabled. Default value: true.
  • visible (optional) – Boolean. When set to false, hides the field from the Field List. Only for "elasticsearch", "json", "csv", and "api" data source types.
  • interval (optional) – String. Allows rounding down dates by the given interval. For example, if the interval is "1d", 2021-05-25T21:30:00 will be rounded to 2021-05-25T00:00:00. See an example on JSFiddle.
    The interval property can be used in the following ways:
    • For the date histogram in Elasticsearch. Check out supported time units. Only for the "elasticsearch" data source type.
    • For the "date string" and "datetime" field types. Supported date intervals are the following:
      • "y" for one year (e.g., "y").
      • "q" for one quarter (e.g., "q").
      • "M" for one month (e.g., "M").
      • "w" for one week (e.g., "w").
      • "d" for days (e.g., "1d"). Note that rounding by days starts from 1 January 1970.
      • "h" for hours (e.g., "7h").
      • "m" for minutes (e.g., "20m").
      • "s" for seconds (e.g., "30s").
        
      Note that "y", "q", "M", and "w" intervals should be used without numbers. Only for "csv" and "json" data source types.
  • isMeasure (optional) – Boolean. When set to true, the field can be selected only as a measure. The isMeasure property should be used only with the strictDataTypes option (see an example on JSFiddle). Only for the "json" data source type. Default value: false.
  • time_zone (optional) – String. Used for the date histogram. You can specify time zones as either an ISO 8601 UTC offset (e.g., +01:00 or -08:00) or as a time zone ID as specified in the IANA time zone database, such as America/Los_Angeles. Only for the "elasticsearch" data source type. Check out an example here.
  • format (optional) – String. Used to format different types of date fields. format can be used in the following ways:
    • For the date histogram in Elasticsearch. The date format/pattern is described in the Elasticsearch documentation. Try a live sample on JSFiddle.
      If the datePattern option is defined, format will override its value for the field.
      Only for the "elasticsearch" data source type.
    • For a field of the "date string" type, it allows overriding the datePattern set in the Options Object. The pattern string for the format is the same as in the datePattern option (i.e., "dd/MM/yyyy"). Only for "json", "csv", and "api" data source types. Learn more about date and time formatting.
    • For a field of the "datetime" type, it allows overriding the dateTimePattern set in the Options Object. The pattern string for the format is the same as in the dateTimePattern option (i.e., "dd/MM/yyyy HH:mm:ss"). Only for "json", "csv", and "api" data source types. Learn more about date and time formatting.
    • For a field of the "time" type, it allows overriding the timePattern set in the Options Object. The pattern string for the format is the same as in the timePattern option (i.e., "HH:mm:ss"). Only for "json", "csv", and "api" data source types. Learn more about date and time formatting.
  • min_doc_count (optional) – Number. Only for the "elasticsearch" data source type. Used for the date histogram. Can be used to show intervals with empty values (min_doc_count: 0). Default value: 1 (empty intervals are hidden).
• aggregations (optional) – Object. Available aggregation functions for all fields of a certain type. See an example on JSFiddle. Only for "json" and "csv" data source types.
  Available aggregations for a certain field can be defined in the [field_name].aggregations property.
  The aggregations object can have the following properties:
  • date (optional) – Array of strings. Available aggregation functions for the "date string" field type.
  • number (optional) – Array of strings. Available aggregation functions for the "number" field type.
  • string (optional) – Array of strings. Available aggregation functions for the "string" field type.
  • time (optional) – Array of strings. Available aggregation functions for the "time" field type.
    
  See a list of supported aggregations in Flexmonster’s technical specifications.
  To learn more about defining aggregataions for fields of a certain type, refer to our guide.

Ways to define the mapping

In Flexmonster, there are two ways to define the mapping: 

1. As an inline Mapping Object.
2. As a URL to a JSON file with the mapping.

As an inline Mapping Object

The inline mapping can be defined as follows:

{
    dataSource: {
        filename: "data.csv",
        mapping: {
            "Month": {
                "type": "month"
            },
            "Company Name": {
                "type": "string"
            },
            "Region": {
                "type": "string",
                "hierarchy": "Geography"
            },
            "State": {
                "type": "string",
                "parent": "region",
                "hierarchy": "Geography"
            },
            "Price": {
                "type": "number",
                "caption": "Unit Price"
            }
        }
    }
}

See the full code on JSFiddle.

As a URL to a JSON file with the mapping

To define the mapping as a URL to a file, complete the steps below.

Step 1. Create a new JSON file with the mapping (e.g., mapping.json). Its contents should look similar to the following:

{
    "Month": {
        "type": "month"
    },
    "Company Name": {
        "type": "string"
    },
    "Region": {
        "type": "string",
        "hierarchy": "Geography"
    },
    "State": {
        "type": "string",
        "parent": "region",
        "hierarchy": "Geography"
    },
    "Price": {
        "type": "number",
        "caption": "Unit Price"
    }
}

Step 2. Put your mapping file where Flexmonster can access it.

Step 3. In Flexmonster, define the mapping as a URL to your file:

{
    dataSource: {
        filename: "data.csv",
        mapping: "<URL_to_your_mapping_file>"
    }
}

See a live sample on JSFiddle.

Configuring multilevel hierarchies

Using the mapping, you can create multilevel hierarchies from non-hierarchical data. This feature is supported for JSON, CSV, Flexmonster Data Server, MongoDB, and custom data source API.

In the custom data source API, multilevel hierarchies should be supported on the server side as well. Learn more in this guide: Supporting multilevel hierarchies.

To configure multilevel hierarchies in the component, specify the hierarchy and parent properties in the mapping.

The example below illustrates how to configure multilevel hierarchies. We create the "Geography" hierarchy with the "Country" field as a first level, "State" field as a second level, and "City" field as a third level:

report: {
    dataSource: {
        filename: "data.json",
        mapping: {
            "Country": {
                hierarchy: "Geography",
            },
            "State": {
                hierarchy: "Geography",
                parent: "Country"
            },
            "City": {
                hierarchy: "Geography",
                parent: "State"
            }
        }
    }
}

See a live example on JSFiddle.

Creating several fields from one field

Starting from version 2.9, you can create multiple fields from one field and use them separately in the slice. This feature is available for JSON and CSV data sources.

To create several fields, specify the mapping as follows:

{
    "mapping": { 
        "Price": [
            {
                "type": "number",
                "uniqueName": "PriceNumber",
                "caption": "Price as Number"
            },
            {
                "type": "string",
                "uniqueName": "PriceString",
                "caption": "Price as String"
            }
        ],
        ...
    }
}

Notice that instead of an object, we used an array of objects as a mapping value. Each object describes a separate field that will be created.

Every new field should have its own uniqueName property. As for other mapping properties, you can use any property available for your data source.

After you have defined the needed fields, you can include them in the slice as ordinary fields:

{
...
    "slice": {
        "rows": [
            {
                "uniqueName": "Category"
            }
        ],
        "columns": [
            {
                "uniqueName": "Color"
            },
            {
                "uniqueName": "[Measures]"
            }
        ],
        "measures": [
            {
                "uniqueName": "PriceNumber"
            }
        ]
    }, 
   ...
}

See a live sample on JSFiddle.

Defining available aggregations for all fields of a certain type

"mapping": {
    "aggregations": {
        "number": [ "sum", "average" ],
        "date": [ "max", "min" ],
        "string": [ "count", "distinctcount" ]
    }
}

Other ways to customize field presentation

Another way to define how the fields are displayed in the report is by configuring this right in the data source. Please note that this approach is available only for CSV and JSON data sources. For more details, please refer to the Data types in CSV and Data types in JSON articles. It also should be noted that there are certain limitations for the CSV data source – not all the field properties can be customized using prefixes.

For this reason, we strongly recommend using the Mapping Object to customize fields.

Examples

1) Creating multilevel hierarchies in JSON:

mapping: {
    "Color": { 
        type: "string" 
    },
    "Country": {
        type: "string",
        hierarchy: "Geography"
    },
    "State": {
        type: "string",
        hierarchy: "Geography",
        parent: "Country"
    },
    "City": {
        type: "string",
        hierarchy: "Geography",
        parent: "State"
    },
    "Price": {
        type: "number"
    }
}

Try the live demo on JSFiddle.

2) Setting custom captions and field data types as well as creating multilevel hierarchies in a CSV data source:

mapping: {
    "Order ID": { "type": "string" },
    "Month": { "type": "month" },
    "Company Name": { "type": "string" },
    "Customer": { "type": "string" },
    "region": {
         "caption": "Region",
         "type": "string",
         "hierarchy": "Geography"
     },
     "State": {
         "type": "string",
         "parent": "region",
         "hierarchy": "Geography"
     },
     "City": {
         "type": "string",
         "parent": "State",
         "hierarchy": "Geography"
     },
     "Salesperson": { "type": "string" },
     "Payment Method": { "type": "string" },
     "Category": { "type": "string" },
     "Name": { "type": "string", "caption": "Product Name" },
     "price": { "type": "number", "caption": "Unit Price" },
     "Quantity": { "type": "number" },
     "Revenue": { "type": "number" },
     "Shipping Cost": { "type": "number" }
}

Try it on JSFiddle.

3) Setting custom captions and grouping fields in separate folders in a JSON data source:

mapping: {
    "Color": {
        caption: "color"
    },
    "Country": {
        caption: "MyCountry",
        folder: "Place"
    },
    "State": {
        caption: "MyState",
        folder: "Place"
    },
    "City": {
        caption: "MyCity",
        folder: "Place"
    },
    "Price": {
        type: "number",
        caption: "MyPrice"
    },
    "Quantity": {
        type: "number",
        caption: "MyQuantity"
    }
}

See the live demo on JSFiddle.

4) Specifying custom captions for hierarchies and measures for SSAS:

mapping: {
    "[Geography].[Geography]": {
        caption: "My Geography"
    },
    "[Product].[Category]": {
        caption: "My Category"
    }
}

See the live demo on JSFiddle.