▣ ARTIFICIAL INTELLIGENCE · DATA EXTRACTION

AI Document Table Data Extraction API

Hosted on ApyHub

What it does

Table Extraction helps you pull tabular data out of a file or a URL. Send a binary file to POST /file or a document URL to POST /url, and choose whether you want ApyHub's structured table extraction output or Azure Document Intelligence output by setting requested_service.

Use it when you need to turn PDFs or other table-bearing documents into machine-readable data for downstream processing. The file-based endpoint accepts file uploads and optional Azure credentials such as azure_key, azure_region, azure_endpoint, azure_account_id, and azure_access_token. The URL-based endpoint accepts url plus optional Azure credentials in an azure object. Both endpoints return a data object with azure and apyhub results, so you can compare the raw Azure parsing output with the structured ApyHub extraction output.

That makes Table Extraction useful for invoice capture, report ingestion, and spreadsheet-like data collection from documents that are not already structured. Build workflows that move table rows into your database, validate extracted line items, or feed document data into analytics pipelines.

If you already have documents stored remotely, use the URL endpoint. If you're processing uploads directly in your app, use the file endpoint.

Provider (requested_service)	Atoms
Azure	500
ApyHub	2000

▣ ENDPOINT 01 / 02

POST

Extract table data (file upload)

http://localhost:8080/apyhub/extract-table-data/file

QUICKSTARTGUIDE

Quickstart

Upload a table image or document for extraction with the required file field.

curl -X POST "http://localhost:8080/apyhub/extract-table-data/file" \
  -H "apy-token: $APY_TOKEN" \
  -F "file=@/path/to/file" \
  -F "requested_service=apyhub"

What you'll get back

Returns a JSON object with a data object field. Inside data, the response may include azure for raw Azure Document Intelligence parsing output and/or apyhub for structured ApyHub table extraction output.

{
  "data": {
    "azure": {},
    "apyhub": {}
  }
}

TRY ITLIVE · 500 ATOMS

API key

Loading your default key…

The full key is used to call the gateway and stays in this tab — never sent to orbit or saved.

Max 100MB total per request (all files combined). Larger? Use this API's URL-based endpoint instead, if it has one.

About this endpoint

What it does

Extracts table data from a document at the provided URL and returns parsed results in a JSON object. You can optionally supply Azure credentials and choose whether to request Azure or ApyHub processing through requested_service.

Request Body

Parameter	Type	Mandatory	Description
url	String	Yes	Document URL to extract table data from. Must be a valid URI.
azure	Object	No	Custom credentials for Azure services. Contains optional fields: `key`, `region`, `endpoint`, `account_id`, `access_token`.
azure.key	String	No	Azure subscription key.
azure.region	String	No	Azure service region.
azure.endpoint	String	No	Azure endpoint URL.
azure.account_id	String	No	Azure account ID.
azure.access_token	String	No	OAuth access token for Azure service.
requested_service	ENUM	No	Service to use. Allowed values: `azure`, `apyhub`. Default: `apyhub`.

Response

Returns a JSON object with a data object field. The data object may contain azure and/or apyhub fields, each of which is an object holding the corresponding table extraction output.

Parameter	Type	Mandatory	Description
data	Object	No	Response wrapper containing extraction results. May include `azure` and `apyhub` objects.
data.azure	Object	No	Raw Azure Document Intelligence parsing output.
data.apyhub	Object	No	Structured ApyHub Table Extraction parsing output.

Body

Name

Type

Description

bodyREQUIRED

object

—

Max 100MB total per request (all files combined). Larger? Use this API's URL-based endpoint instead, if it has one.

▣ ENDPOINT 02 / 02

POST

Extract table data (URL input)

http://localhost:8080/apyhub/extract-table-data/url

QUICKSTARTGUIDE

Quickstart

Extract table data from a PDF or web page by sending its URL.

curl -X POST "http://localhost:8080/apyhub/extract-table-data/url" \
  -H "apy-token: $APY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://assets.apyhub.com/samples/sample.pdf", "requested_service":"apyhub"}'

What you'll get back

Returns a JSON object with a data object. Inside data, the response may include an apyhub object for structured table extraction output and/or an azure object for raw Azure Document Intelligence parsing output.

{
  "data": {
    "apyhub": {},
    "azure": {}
  }
}

TRY ITLIVE · 500 ATOMS

API key

Loading your default key…

The full key is used to call the gateway and stays in this tab — never sent to orbit or saved.

About this endpoint

What it does

Extracts table data from a document or webpage referenced by a URL and returns parsed output in a JSON object. You can request the result from either Azure or ApyHub by setting requested_service; if omitted, it defaults to apyhub.

Request Body

Parameter	Type	Mandatory	Description
url	String	Yes	Source URL to the table document. Must be a valid URI.
azure	Object	No	Custom credentials for Azure services. See schema for nested fields.
azure.key	String	No	Azure subscription key.
azure.region	String	No	Azure service region.
azure.endpoint	String	No	Azure endpoint URL.
azure.account_id	String	No	Azure account ID.
azure.access_token	String	No	OAuth access token for Azure service.
requested_service	ENUM	No	Service to use for extraction. Allowed values: `azure`, `apyhub`. Default: `apyhub`.

Response

Returns a JSON object with a data object field. Inside data, the response may include an azure object with raw Azure Document Intelligence parsing output and/or an apyhub object with structured ApyHub Table Extraction parsing output.

Parameter	Type	Mandatory	Description
data	Object	No	Response wrapper. May contain `azure` and/or `apyhub` objects.
data.azure	Object	No	Raw Azure Document Intelligence parsing output.
data.apyhub	Object	No	Structured ApyHub Table Extraction parsing output.

Body

Name

Type

Description

bodyREQUIRED

object

—

▣ COMMON ERRORS

Errors any endpoint can return

400bad_request

Required parameter missing or malformed body.

401unauthorized

API key missing, revoked, or not authorized for this service.

429rate_limited

Your plan's per-second rate exceeded. Retry with exponential backoff.

503upstream_busy

Backend temporarily unavailable. Try again in a few seconds.