Dataset Service

The Dataset service allows you to read, write and delete columns/rows to a collection of data (database table).

The following example configuration shows you how to add a dataset and how to configure the Dataset service to insert, update, delete and find data within the dataset.

Creating a dataset¶

Before starting to use the dataset service a dataset needs to be created in the Spojit platform. Select "Datasets" on the left side bar and click "+ Dataset" to add a new dataset. Add a name for the dataset and define the columns:

Column Type¶

The column type is the type of data that is stored in the column.

Data Type¶

The data type is the type of data used for statistical and training purposes.

Once the columns are defined make sure that they are active in order to be used. After a dataset is saved you will be able to use it in any workflow.

Dataset Row Object¶

When dealing with a row in the dataset an object with be used with the fields matching the columns specified in the dataset configuration.

Reserved columns (fields)¶

The following reserved names can not be used to name a column or be a field in the dataset row object as they are automatically generated for an object:

Example dataset row object¶

If you have the columns id and name from the creating a dataset example above the object would consist of the following (with the reserved fields added):

{
  "id": 1,
  "name": "delectus aut autem",
  "_id": "60792518693cf2625e7530b2",
  "_createdAt": "2021-04-16T05:48:08+00:00",
  "_updatedAt": "2021-04-16T05:48:08+00:00"
}

Dataset Querying Criteria¶

You can query a dataset with critier to find a record to update or to search for records within the dataset.

The following operators can be used:

These operators can be chained to produce more specific results.

Primary Keys¶

The internal primary key (id) will always be generated automatically but if you want to pair it with an external primary key (or other unique identifier) the "Upsert" method can be used with criteria to identify it.

For example if the "Upsert" method is setup in the following way:

The record will be created if the value for external_id doesn't exist in the id column or updated if it does. Two or more columns can also be used to have a composite primary key.

Service Methods¶

The dataset service has various methods add to and perform actions on data in a dataset.

Insert a row¶

Inserting a row will add a new record into the dataset.

{
  "data": {
    "userId": 1,
    "id": 1,
    "title": "delectus aut autem",
    "completed": false
  }
}

{
  "data": {
    "id": 1,
    "name": "delectus aut autem",
    "_id": "60792518693cf2625e7530b2",
    "_createdAt": "2021-04-16T05:48:08+00:00",
    "_updatedAt": "2021-04-16T05:48:08+00:00"
  },
  "metadata": []
}

Update a row¶

Updating a row will update a record in the dataset.

{
  "data": {
    "userId": 1,
    "id": 1,
    "title": "hello world",
    "completed": false
  }
}

{
  "data": {
    "id": 1,
    "name": "hello world",
    "_id": "60792518693cf2625e7530b2",
    "_createdAt": "2021-04-16T05:48:08+00:00",
    "_updatedAt": "2021-04-16T05:58:08+00:00"
  },
  "metadata": []
}

Upsert a row¶

Upserting a row will either create a record in the dataset if it doesn't exist or update a record if it already exists.

{
  "data": {
    "userId": 1,
    "id": 1,
    "title": "hello world",
    "completed": false
  }
}

{
  "data": {
    "id": 1,
    "name": "hello world",
    "_id": "60792518693cf2625e7530b2",
    "_createdAt": "2021-04-16T05:48:08+00:00",
    "_updatedAt": "2021-04-16T05:58:08+00:00"
  },
  "metadata": []
}

Delete a row¶

Deleting a row will remove a record from the dataset if it exists.

Find one row¶

Find one row will search for and find one record from the dataset if it exists. It will find and return the first matching record. If there are multiple matching rows then "find" should be used.

{
  "data": {
    "id": 1,
    "name": "hello world",
    "_id": "60792518693cf2625e7530b2",
    "_createdAt": "2021-04-16T05:48:08+00:00",
    "_updatedAt": "2021-04-16T05:58:08+00:00"
  },
  "metadata": []
}

Find rows¶

Find rows will search for and find records from the dataset if they exists. It will find and return all matching records in an array. If you only want one result "findOne" should be used.

{
  "data": [
    {
      "id": 1,
      "name": "pants",
      "colour": "blue",
      "_id": "60792518693cf2625e7530b2",
      "_createdAt": "2021-04-16T05: 48: 08+00: 00",
      "_updatedAt": "2021-04-16T05: 58: 08+00: 00"
    },
    {
      "id": 2,
      "name": "tshirt",
      "colour": "blue",
      "_id": "60792518693cf2625e7530b2",
      "_createdAt": "2021-04-16T05: 48: 08+00: 00",
      "_updatedAt": "2021-04-16T05: 58: 08+00: 00"
    }
  ],
  "metadata": [

  ]
}