API

Parse PDF

Create Synchronization Job

"Synchronization" mean that you can receive results immediately after a request is completed

Replace {{API_TOKEN}} with your actual token.

Example Request:

curl -X POST 'https://api.netmind.ai/inference-api/agent/v1/parse-pdf' \
--header 'Authorization: Bearer {{API_TOKEN}}' \
--header 'Content-Type: application/json' \
--data-raw '{
    "url": "https://netmind-public-files.s3.us-west-2.amazonaws.com/test.pdf",
    "format": "json"
}'

Example Response:

[
    {
        "id": "0eKSDqCl",
        "version": "1.0.1",
        "page": 1,
        "seq_no": 1,
        "sentence": "FABSA: An aspect-based sentiment analysis dataset of user reviews",
        "type": "title",
        "text_location": {
            "location": [
                [
                    37,
                    629,
                    427,
                    613
                ]
            ],
            "location_raw": [
                [
                    37,
                    164,
                    427,
                    180
                ]
            ]
        },
        "info": {}
    },
    ...
]

Create Asynchronous Job

"Asynchronous" mean that you need to obtain results through another request (Get Job)

Replace {{API_TOKEN}} with your actual token.

Example Request:

curl -X POST 'https://api.netmind.ai/inference-api/agent/v1/parse-pdf/async' \
--header 'Authorization: Bearer {{API_TOKEN}}' \
--header 'Content-Type: application/json' \
--data-raw '{
    "url": "https://netmind-public-files.s3.us-west-2.amazonaws.com/test.pdf",
    "format": "json"
}'

Example Response:

{
    "task_id":"1a9e46e...2a9fb",
    "status":"PENDING"
}

Get Job

Replace {{API_TOKEN}} with your actual token.

Replace {{TASK_ID}} with "task_id" you got from previous step.

Example Request:

curl -X GET 'https://api.netmind.ai/inference-api/agent/v1/parse-pdf/async/{{TASK_ID}}' \
--header 'Authorization: Bearer {{API_TOKEN}}'

Example Response:

[
    {
        "id": "0eKSDqCl",
        "version": "1.0.1",
        "page": 1,
        "seq_no": 1,
        "sentence": "FABSA: An aspect-based sentiment analysis dataset of user reviews",
        "type": "title",
        "text_location": {
            "location": [
                [
                    37,
                    629,
                    427,
                    613
                ]
            ],
            "location_raw": [
                [
                    37,
                    164,
                    427,
                    180
                ]
            ]
        },
        "info": {}
    },
    ...
]

Last updated

Was this helpful?