Using the data import API | ioTORQ EMIS Help Center

In addition to manual uploads, ioTORQ EMIS allows data to be imported programmatically using the API. API data import is useful when you need to:

Automate data ingestion from external systems
Backfill large volumes of historical data
Update variables on a scheduled basis
Integrate ioTORQ EMIS with third-party platforms
Import data when a live connector is not available

API imports create or update data points in existing variables (auto-variables) within ioTORQ EMIS.

Before you begin

To use the API Data Import feature, you must have:

An active ioTORQ EMIS user account
An API token associated with that user
Existing variables in ioTORQ EMIS to receive the data

Each API token is linked to a specific user account, and all imported data is attributed to that user.

API endpoint

Data is imported using an HTTPS POST request to:

/ems/api/import/

All API URLs are relative to your company’s ioTORQ EMIS domain.
For example, if your account URL is:

https://mycompany.iotorq.com

The full API endpoint is:

https://mycompany.iotorq.com/ems/api/import/

Authentication and headers

Every request must include the following HTTP headers:

Authorization: Token <your_api_token>
Content-Type: application/json

Requests without valid authentication will be rejected.

Request format

Data must be sent as JSON.
The request body is a list of objects, where each object represents a variable (meter) and its time-series data.

Each object includes:

Variable identification

You can identify a variable using one of the following methods:

Meter Code (recommended for most users)
Variable ID (primary key)

Field	Description
`id_type`	How the variable is identified (`meter_code` or `pk`)
`id_value`	The meter code or variable ID that exists in ioTORQ EMIS

Timeseries data

Each variable includes a series object containing timestamps and values.

Field	Description
`datetime`	A list of UNIX timestamps
`val`	A list of numeric values (must match the datetime list)
`time_unit`	Optional. Use `"ms"` if timestamps are in milliseconds

Timestamps are assumed to be seconds since Jan 1, 1970 (UTC) unless otherwise specified.
The number of timestamps must match the number of values.

Example: Import data using meter codes

[
  {
    "id_type": "meter_code",
    "id_value": "my_meter_code1",
    "series": {
      "datetime": [1494880000, 1494880060, 1494880120],
      "val": [1, 2, 3]
    }
  },
  {
    "id_type": "meter_code",
    "id_value": "my_meter_code2",
    "series": {
      "datetime": [1494880000, 1494880060, 1494880120],
      "val": [2.0, 5.5, 4.1]
    }
  }
]

This example imports data for two variables, each with multiple timestamps and values.

Handling duplicate timestamps

If data is submitted for a timestamp that already exists:

The new value overwrites the existing value
If multiple values for the same timestamp are included in a single request, only one value will be stored, and the result is not guaranteed

To avoid unintended overwrites, ensure timestamps are unique per variable.