Spektr API Integration Guide

To integrate with Spektr's APIs, follow these high-level steps to manage datasets, execute processes, and export results within the platform:

Import Overview

Purpose: Define the structure of your dataset by specifying the fields and their data types.
Action: Use the Import API to create a new dataset, outlining the schema that includes all necessary fields such as text, numbers, dates, or files.
Outcome: A dataset is created and becomes available on the platform, ready to accept data entries that conform to the defined schema.

Purpose: Add data entries to your dataset based on the schema you've defined.
Action: Use the Import API to submit records to the dataset. Each record should include values for the fields specified in your schema.
File fields handling: If your dataset includes file fields (e.g., documents, images), provide placeholders or filenames for these fields in your records.
Outcome: Records are added to your dataset, and for any file fields, the API responds with presigned URLs for file uploads along with unique spektrIds for each record.

Purpose: Upload the actual files associated with the file fields in your dataset records.
Action: Use the presigned URLs provided by the API to upload files directly to the storage service (AWS S3). This involves making a POST request with the file data to the specified URL.
Considerations: Presigned URLs are time-sensitive, so it's important to upload the files promptly. The presigned URL expires in 1 hour from the time it is generated.
Outcome: Files are successfully uploaded and associated with the corresponding records in your dataset.

Purpose: Execute specific processes defined in the platform using your imported data.
Action: Use the Execution API to initiate processes by providing the processId (obtained from the platform) and the list of spektrIds corresponding to your dataset records.
Outcome: The specified processes are executed for the provided records. Results can be obtained in our platform, through Fetch customer records Endpoint or through Webhooks .

Purpose: Retrieve the results of your executed processes.
Action: Use the Records API to query customer records using the spektrIds of your processed records.
Outcome: Access processed data, analyse results, and any generated outputs.