Name: Doc Extract API: Document Schema, Fields & Validation Rules
Brand: nicheapi-llc-1
SKU: api-nicheapi-llc-1-docextract
Availability: InStock
Author: API Market

True Scary Stories

# Docextract API

## Overview

The Docextract API is a metadata catalog for document processing pipelines. It maps out which document types exist, what fields can be extracted from each, and how those fields should be validated — giving developers a structured reference for the shape of extraction rather than performing the extraction itself. Think of it as the schema layer that sits upstream of your OCR engine, NLP extractor, or document AI model: it tells your system what to look for, where to find it, and whether the value it returns makes sense. The three endpoints cover document type definitions (what categories of documents exist and what they contain), extraction field specifications (field names, data types, locations, and extraction methods), and field validation rules (patterns, ranges, formats, and constraints that determine whether an extracted value is trustworthy).

## Features

*   **3 Dedicated Endpoints**: Access three structured metadata categories — document type definitions, extraction field specifications, and field validation rules.
*   **Full-Text Search**: Filter records by keyword across all fields using the `?q=` query parameter.
*   **Pagination Support**: Efficiently retrieve large result sets with `?limit=` and `?offset=` parameters.
*   **Detailed Lookup**: Fetch any individual record directly by its unique ID.
*   **Authoritative Data**: All records are curated from established document processing standards, IRS and government form specifications, ANSI/ISO document standards, and industry extraction schema conventions — not auto-generated.
*   **Unified API Key**: A single `x-api-market-key` authenticates across all 415+ APIs on the platform — no separate credential management per API.
*   **High Availability**: A rate limit of 60 requests per minute across all plans.

## Use Cases

*   **Document AI & OCR Pipeline Configuration**: Developers building document processing systems can query extraction fields by document type to dynamically configure what their OCR or NLP model should extract from a given document — rather than hardcoding field lists per document category in application code.
*   **Extraction Quality & Confidence Scoring**: Intelligent document processing (IDP) platforms can use field validation rules to score extracted values post-extraction — flagging low-confidence results where the extracted value fails its expected pattern, range, or format constraint before passing it downstream.
*   **Document Classification Systems**: Machine learning teams building document classifiers can use document type records to define and standardize the target label space — ensuring classification outputs map to a consistent, well-described taxonomy with known field inventories.
*   **No-Code & Low-Code Document Workflow Builders**: Platforms that let non-technical users configure document extraction workflows can use the API to power field selectors and validation configurators — surfacing the available fields for a chosen document type and their expected data types and constraints without requiring users to understand the underlying extraction logic.
*   **Data Schema Generation & API Contract Design**: Backend engineers building data pipelines that ingest extracted document data can use the field specifications and validation rules to auto-generate JSON schemas, database table definitions, or API request/response contracts — keeping extraction output schemas in sync with the reference catalog.


## API Documentation

### Base URL

All API requests for the Docextract API should be made to the following base URL:

```
https://prod.api.market/api/v1/nicheapi-llc-1/docextract
```

### Authentication

Authentication is handled via a unified API key shared across all 415+ APIs on the platform. Include your key in the `x-api-market-key` header of every request:

```
x-api-market-key: YOUR_API_MARKET_KEY
```

Replace `YOUR_API_MARKET_KEY` with your actual API key obtained from API.market.


### Endpoints

#### 1. List Document Types

Retrieve a paginated list of document type definitions — the categories of documents the extraction schema catalog covers, with metadata about structure, complexity, and typical field inventory.

**Method**: `GET`
**Endpoint**: `/docextract/document_types`

**Parameters**:

| Name     | In    | Description                                           | Type      | Required | Default | Max Value |
| :------- | :---- | :---------------------------------------------------- | :-------- | :------- | :------ | :-------- |
| `q`      | query | Search query to filter document types by keyword      | `string`  | No       |         |           |
| `limit`  | query | Maximum number of results per page                    | `integer` | No       | `50`    | `1000`    |
| `offset` | query | Number of results to skip for pagination              | `integer` | No       | `0`     |           |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/document_types?q=invoice&limit=1" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
[
  {
    "id": "doctype_invoice",
    "document_name": "Invoice",
    "document_category": "Financial",
    "description": "A commercial document issued by a seller to a buyer itemizing goods or services delivered, amounts owed, and payment terms.",
    "typical_sources": ["Accounts payable systems", "Vendor email attachments", "Scanned paper documents", "EDI feeds"],
    "structured": true,
    "extraction_complexity": "Medium",
    "common_formats": ["PDF", "Scanned image", "HTML", "EDI X12 810"],
    "typical_field_count": 24,
    "has_line_items": true,
    "notes": "One of the most commonly processed document types in enterprise automation. Field layouts vary significantly by vendor, making template-free extraction preferable over fixed-template approaches for scale."
  }
]
```

---

#### 2. Get Document Type by ID

Retrieve a single document type record by its unique identifier.

**Method**: `GET`
**Endpoint**: `/docextract/document_types/{id}`

**Parameters**:

| Name | In   | Description                              | Type     | Required |
| :--- | :--- | :--------------------------------------- | :------- | :------- |
| `id` | path | Unique identifier for the document type  | `string` | Yes      |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/document_types/doctype_invoice" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
{
  "id": "doctype_invoice",
  "document_name": "Invoice",
  "document_category": "Financial",
  "description": "A commercial document issued by a seller to a buyer itemizing goods or services delivered, amounts owed, and payment terms.",
  "typical_sources": ["Accounts payable systems", "Vendor email attachments", "Scanned paper documents", "EDI feeds"],
  "structured": true,
  "extraction_complexity": "Medium",
  "common_formats": ["PDF", "Scanned image", "HTML", "EDI X12 810"],
  "typical_field_count": 24,
  "has_line_items": true,
  "notes": "One of the most commonly processed document types in enterprise automation. Field layouts vary significantly by vendor, making template-free extraction preferable over fixed-template approaches for scale."
}
```

---

#### 3. List Extraction Fields

Retrieve a paginated list of extraction field specifications — the individual data points that can be pulled from a document type, with data type, typical location, extraction method, and linked validation rule.

**Method**: `GET`
**Endpoint**: `/docextract/extraction_fields`

**Parameters**:

| Name     | In    | Description                                              | Type      | Required | Default | Max Value |
| :------- | :---- | :------------------------------------------------------- | :-------- | :------- | :------ | :-------- |
| `q`      | query | Search query to filter extraction fields by keyword      | `string`  | No       |         |           |
| `limit`  | query | Maximum number of results per page                       | `integer` | No       | `50`    | `1000`    |
| `offset` | query | Number of results to skip for pagination                 | `integer` | No       | `0`     |           |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/extraction_fields?q=invoice+number&limit=1" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
[
  {
    "id": "field_invoice_number",
    "field_name": "invoice_number",
    "field_label": "Invoice Number",
    "data_type": "string",
    "description": "The unique identifier assigned to this invoice by the issuing vendor. Used for matching against purchase orders and payment records.",
    "applicable_document_type_ids": ["doctype_invoice", "doctype_credit_note", "doctype_proforma_invoice"],
    "required": true,
    "typical_location": "Header",
    "extraction_method": "OCR + keyword proximity (labels: 'Invoice No.', 'Invoice #', 'Inv No', 'Reference')",
    "validation_rule_id": "validation_invoice_number",
    "example_values": ["INV-2024-00412", "2024-10459", "SI-88823"],
    "notes": "Alphanumeric; format varies widely by vendor. No universal standard — regex validation should be permissive."
  }
]
```

---

#### 4. Get Extraction Field by ID

Retrieve a single extraction field specification by its unique identifier.

**Method**: `GET`
**Endpoint**: `/docextract/extraction_fields/{id}`

**Parameters**:

| Name | In   | Description                                | Type     | Required |
| :--- | :--- | :----------------------------------------- | :------- | :------- |
| `id` | path | Unique identifier for the extraction field | `string` | Yes      |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/extraction_fields/field_invoice_number" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
{
  "id": "field_invoice_number",
  "field_name": "invoice_number",
  "field_label": "Invoice Number",
  "data_type": "string",
  "description": "The unique identifier assigned to this invoice by the issuing vendor. Used for matching against purchase orders and payment records.",
  "applicable_document_type_ids": ["doctype_invoice", "doctype_credit_note", "doctype_proforma_invoice"],
  "required": true,
  "typical_location": "Header",
  "extraction_method": "OCR + keyword proximity (labels: 'Invoice No.', 'Invoice #', 'Inv No', 'Reference')",
  "validation_rule_id": "validation_invoice_number",
  "example_values": ["INV-2024-00412", "2024-10459", "SI-88823"],
  "notes": "Alphanumeric; format varies widely by vendor. No universal standard — regex validation should be permissive."
}
```

---

#### 5. List Field Validation Rules

Retrieve a paginated list of field validation rules — the patterns, ranges, formats, and constraints used to determine whether an extracted field value is well-formed and trustworthy.

**Method**: `GET`
**Endpoint**: `/docextract/field_validation`

**Parameters**:

| Name     | In    | Description                                                 | Type      | Required | Default | Max Value |
| :------- | :---- | :---------------------------------------------------------- | :-------- | :------- | :------ | :-------- |
| `q`      | query | Search query to filter field validation rules by keyword    | `string`  | No       |         |           |
| `limit`  | query | Maximum number of results per page                          | `integer` | No       | `50`    | `1000`    |
| `offset` | query | Number of results to skip for pagination                    | `integer` | No       | `0`     |           |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/field_validation?q=date&limit=1" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
[
  {
    "id": "validation_issue_date",
    "field_name": "issue_date",
    "applies_to_field_id": "field_issue_date",
    "data_type": "date",
    "validation_type": "format + range",
    "accepted_formats": ["YYYY-MM-DD", "DD/MM/YYYY", "MM/DD/YYYY", "DD MMM YYYY", "MMMM DD, YYYY"],
    "normalize_to": "ISO 8601 (YYYY-MM-DD)",
    "min_value": "1990-01-01",
    "max_value": "today + 90 days",
    "pattern": null,
    "error_message": "Extracted date is not a recognized format or falls outside the acceptable range.",
    "examples_valid": ["2024-03-15", "15/03/2024", "March 15, 2024"],
    "examples_invalid": ["99/99/9999", "2087-01-01", "N/A"],
    "notes": "Dates more than 90 days in the future are likely extraction errors on standard commercial documents. Normalize all formats to ISO 8601 before downstream processing."
  }
]
```

---

#### 6. Get Field Validation Rule by ID

Retrieve a single field validation rule by its unique identifier.

**Method**: `GET`
**Endpoint**: `/docextract/field_validation/{id}`

**Parameters**:

| Name | In   | Description                                    | Type     | Required |
| :--- | :--- | :--------------------------------------------- | :------- | :------- |
| `id` | path | Unique identifier for the field validation rule | `string` | Yes      |

**Example Request**:

```bash
curl -X GET "https://prod.api.market/api/v1/nicheapi-llc-1/docextract/docextract/field_validation/validation_issue_date" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Example Response (200 OK)**:

```json
{
  "id": "validation_issue_date",
  "field_name": "issue_date",
  "applies_to_field_id": "field_issue_date",
  "data_type": "date",
  "validation_type": "format + range",
  "accepted_formats": ["YYYY-MM-DD", "DD/MM/YYYY", "MM/DD/YYYY", "DD MMM YYYY", "MMMM DD, YYYY"],
  "normalize_to": "ISO 8601 (YYYY-MM-DD)",
  "min_value": "1990-01-01",
  "max_value": "today + 90 days",
  "pattern": null,
  "error_message": "Extracted date is not a recognized format or falls outside the acceptable range.",
  "examples_valid": ["2024-03-15", "15/03/2024", "March 15, 2024"],
  "examples_invalid": ["99/99/9999", "2087-01-01", "N/A"],
  "notes": "Dates more than 90 days in the future are likely extraction errors on standard commercial documents. Normalize all formats to ISO 8601 before downstream processing."
}
```

---

## Status Codes

| Code  | Status                | Description |
| :---- | :-------------------- | :---------- |
| `200` | OK                    | The request succeeded and data has been returned. |
| `400` | Bad Request           | The request contains invalid parameters or a malformed query. Review your request syntax. |
| `401` | Unauthorized          | The `x-api-market-key` header is missing or contains an invalid key. |
| `403` | Forbidden             | Your API key is valid but lacks permission to access this resource. |
| `404` | Not Found             | The requested endpoint or record ID does not exist. Verify the URL path and identifier. |
| `429` | Too Many Requests     | You have exceeded the rate limit of 60 requests per minute. See the Rate Limits section. |
| `500` | Internal Server Error | An unexpected server-side error occurred. Retry the request after a brief delay. |
| `503` | Service Unavailable   | The server is temporarily unavailable due to maintenance or high load. Retry after a short delay. |

---

## Rate Limits

The Docextract API enforces a limit of **60 requests per minute** across all plans. If exceeded, the API returns a `429 Too Many Requests` response. To handle this gracefully, implement client-side throttling and exponential backoff in your application. Where provided, the `Retry-After` header in the 429 response will indicate how long to wait before retrying.

---

## FAQ

*   **Does this API extract data from documents I upload?**
    No. The Docextract API is a metadata catalog — it describes the structure of document extraction rather than performing it. It tells you what fields exist on an invoice, what data type each field should be, and what a valid extracted value looks like. The actual extraction (OCR, NLP, document AI) happens in your own pipeline; Docextract provides the reference schema that pipeline should conform to.

*   **How are the three endpoints related?**
    They form a chain: document types define which categories of documents exist → extraction fields specify what data points can be pulled from each document type (with `applicable_document_type_ids` linking them) → field validation rules define what a correctly extracted value for each field looks like (with `validation_rule_id` on the extraction field linking to its corresponding validation record). You can traverse the full chain starting from any endpoint.

*   **What does `extraction_method` mean on a field record?**
    This field describes the recommended technique for locating and extracting the field value — for example, `OCR + keyword proximity` (find the value near a known label), `NLP entity extraction` (use named entity recognition), `table parsing` (the value lives in a structured table cell), or `regex pattern match` (the value has a distinctive format that can be matched directly). It's advisory — your extraction system may use different methods — but it reflects best-practice approaches for that field type.

*   **What does `normalize_to` mean on a validation rule?**
    After extraction, values often come in inconsistent formats — dates as `15/03/2024`, `March 15, 2024`, or `2024-03-15` depending on the source document. The `normalize_to` field specifies the canonical target format your pipeline should convert extracted values into before storage or downstream processing, ensuring consistent data regardless of source document formatting.

*   **What is the `validation_type` field and what values can it take?**
    `validation_type` indicates the category of constraint applied to the field. Common values include `regex` (the value must match a regular expression pattern), `format + range` (date or numeric value within bounds), `enum` (value must be one of a fixed set of allowed values), `checksum` (value contains a self-verifying digit, such as an IBAN or routing number), and `format` (the value must conform to a recognized structural format without range constraints).

*   **How do I authenticate?**
    Include your API key in the `x-api-market-key` request header on every call. The same key authenticates across all 415+ APIs on the platform — no per-API credentials required.

*   **What is the rate limit?**
    60 requests per minute across all plans. Exceeding this returns a `429` response.

---

## Contact

For questions or support, please reach out via the chat on the bottom right of the page or email **21truescarystories@gmail.com**.

Access the Doc Extract API for structured reference data on document types, extraction fields, and validation rules. Power your AI/OCR pipelines and ensure data

Doc Extract API: Document Schema, Fields & Validation Rules

Free