Interacting with app via API

Once you have your application up and running, you can start loading data and querying the API. Here's a step-by-step guide on how to do it:

Ingest an entry

1. Data Ingestion: You can test data ingestion by making a POST request to the /api/v1/ingest/your_schema endpoint. Here's an example using curl:

   curl -X POST \
       'http://localhost:8080/api/v1/ingest/your_schema' \
       --header 'Accept: */*' \
       --header 'Content-Type: application/json' \
       --data-raw '{
           "id": "your_id",
           ...
       }'

   Note: The current example will not work, please change the request body as your schema requires it.

Query your system

1. Query the API: After ingesting data, you can query the API by making a POST request to the /api/v1/search/query endpoint. Here's an example using curl:

   curl -X POST \
       'http://localhost:8080/api/v1/search/query' \
       --header 'Accept: */*' \
       --header 'Content-Type: application/json' \
       --header 'x-include-metadata: true' \
       --data-raw '{
           "query_text": "your_search_text"
       }'

   Note: The x-include-metadata: true header will include additional debug information about your query, such as search vector, weights and evaluated NLQ outputs.

Load data from file(s)

See available data loaders

To see what data loaders are available, send a request to the endpoint below:

curl 'http://localhost:8080/data-loader/'

Successful response (200 OK):

{
    "result": [
        "<NAME_OF_YOUR_DATA_LOADER>": "DataLoaderConfig(path='https://path-to-your-file.csv', format=<DataFormat.CSV: 2>, name=None, pandas_read_kwargs='{sep: ;}')"
    ]
}

The keys are the available data loader names that you can use for the rest of the data loader endpoints below. To see how these names are constructed and can be altered, read the docs here.

Trigger the data load

To initiate the data load, invoke its endpoint. This will spawn an asynchronous task DataLoaderSource by its name as defined in your api.py. To trigger the endpoint, simply send a request with curl as shown below. The response should be 202 Accepted. If the name you provided is not found in the system, a 404 NOT FOUND will be returned. See the logs to check the result of that task.

curl -X POST 'http://localhost:8080/data-loader/<NAME_OF_YOUR_DATA_LOADER>/run'

Successful response (200 OK):

{
    "result": "Background task successfully started with name: <NAME_OF_YOUR_DATA_LOADER>",
}