Skip to content

Overview

What is DataStore?

DataStorcode is an in-browser store of tabular data. Mainly, the DataStore contains the DataTable derived from the original data source along with some data operations. You can then perform data operations on this DataTable to transform the data as per your requirement. For any operation applied a new DataTabe is generated keeping the source data intact.

Note: DataStore accepts both arrays of JSON object or 2D array.

The DataStore contains the following:

  • DataTable to represent the data on which operations can be performed.
  • Number of operators to transform the data.
  • A propagation mechanism to communicate between different DataTable created from the same DataStore.

Operations can only be performed on the DataTable of DataStore and each operation generates a new DataTable instead of modifying the same DataTable.

You can create new columns for a DataTable but new rows can only be inserted in the DataStore. The root tables and all derived tables will automatically reflect the new row of data.

What is a DataTable

The tabular representation of the data is called the DataTable, which is stored in the DataStore. You can apply one or more operations to an existing DataTable to create a new table with the resulting data.

Note that you cannot directly change data in the DataTable. Instead, you need to apply one or more operations to an existing DataTable to create a new table with the resulting data.

To create a DataTable, you need to provide:

  • The schema, which defines the properties of the columns.
  • The actual values for each row and column as the data.

Schema

The schema contains an array of objects. Each object represents a column in the DataTable. The schema helps DataTable understand the columns in the data, the data type for each column and the input format in case of date. It is mandatory to specify the schema for each DataTable.

The schema of a DataTable is shown below:

let schema = [{
  name: "Country",
  type: "string"
}, {
  name: "Time",
  type: "date",
  format: "%-m/%-d/%Y"
}, {
  name: "Sales",
  type: "number"
}];

In the above code:

  • schema is the variable in which the array is saved.
  • Each column in the DataTable is represented by a JSON object within the schema array. The JSON object has the following attributes:
    • name - Specify the name of the column.
    • type - Specify the type of the column.
    • format - Specify the input format of the date as per you data. In this example, the format is %-m/%-d/%Y. To know more on date formats refer to timeseries attributes.

Data

To add values in each column of the DataTable you can provide the data in arrays of JSON object or 2D array. In this article, we will use the 2D array format.

The data format is shown below:

let data = [
  ["United States", "1/4/2011", 16.448],
  ["United States", "1/5/2011", 272.736],
  ["United States", "1/5/2011", 11.784],
  ["United States", "12/31/2014", 20.72],
  ["United States", "12/31/2014", 13.904],
  ["United States", "12/31/2014", 3.024]
];

In the above code:

  • data is the variable in which the data array is saved.
  • Each row in the data array corresponds to a row in the DataTable and each element in a row are represented by each object of the schema. The values in each row of the array represents the following:
    • The name of the Country.
    • Time according to the format * Total Sales amount