Parse Quickstart

Parse converts a document into Markdown and structured elements while preserving reading order, layout, tables, images, and formulas. In this quickstart, you will:

Install Python dependencies.
Parse a sample document.
Read the returned Markdown and structured elements.

Before you start

You need:

TextIn API credentials. See Authentication.
Python 3.8 or later.
A PDF, image, Office file, or other supported document.

You can use your own file or download a sample PDF.

For supported formats, file size limits, page limits, and concurrency limits, see Supported Files & Limits.

Install dependencies

pip install requests

Set your credentials

Set your App ID and Secret Code as environment variables:

export TEXTIN_APP_ID="your-app-id"
export TEXTIN_SECRET_CODE="your-secret-code"

On Windows PowerShell:

$env:TEXTIN_APP_ID="your-app-id"
$env:TEXTIN_SECRET_CODE="your-secret-code"

Parse a document

import os
import requests

with open("document.pdf", "rb") as file:
    response = requests.post(
        "https://api.textin.ai/api/v1/xparse/parse/sync",
        headers={
            "x-ti-app-id": os.environ["TEXTIN_APP_ID"],
            "x-ti-secret-code": os.environ["TEXTIN_SECRET_CODE"],
        },
        files={"file": ("document.pdf", file)},
    )

response.raise_for_status()
result = response.json()["data"]

print(result["markdown"])

Inspect the result

Parse returns Markdown and structured document elements.

Markdown

# Annual Report 2024

## Executive Summary

This report presents...

Use Markdown for:

RAG pipelines and retrieval workflows
Full-text search and indexing
Summarization
Chunking and embedding
Document review applications

Structured elements

Each element represents part of the document, such as a heading, paragraph, table, image, or formula.

for element in result["elements"][:5]:
    print(element["type"], element["page_number"], element["text"][:80])

Common element fields include:

Field	Description
`element_id`	Unique identifier for the element
`type`	Element type, such as `Title`, `NarrativeText`, `Table`, or `Image`
`text`	Recognized text content
`page_number`	Page containing the element
`coordinates`	Normalized four-point bounding box

See Response Format for the complete response structure.

Save the output

from pathlib import Path

Path("output.md").write_text(result["markdown"], encoding="utf-8")

Use cURL

You can also call the synchronous Parse endpoint directly:

curl -X POST "https://api.textin.ai/api/v1/xparse/parse/sync" \
  -H "x-ti-app-id: $TEXTIN_APP_ID" \
  -H "x-ti-secret-code: $TEXTIN_SECRET_CODE" \
  -F "file=@document.pdf"

Use the Python SDK

If you are using the Python SDK version that exposes XParseClient, you can call Parse like this:

from xparse_client import XParseClient

client = XParseClient(
    app_id="your-app-id",
    secret_code="your-secret-code",
)

with open("document.pdf", "rb") as file:
    result = client.parse.run(
        file=file,
        filename="document.pdf",
    )

print(result.markdown)

The SDK interface may vary by package version. If your installed xparse-client package does not expose XParseClient, use the REST API examples above or the SDK version provided by TextIn.

Next steps

Configuration

Customize parsing behavior and returned data.

Response Format

Understand Markdown, elements, coordinates, pages, and optional fields.

Async Processing

Process large documents or batch jobs asynchronously.

Examples

Explore common implementation patterns.

Getting Started

Parse

Extract

Before you start

Install dependencies

Set your credentials

Parse a document

Inspect the result

Markdown

Structured elements

Save the output

Use cURL

Use the Python SDK

Next steps

Configuration

Response Format

Async Processing

Examples

​Before you start

​Install dependencies

​Set your credentials

​Parse a document

​Inspect the result

​Markdown

​Structured elements

​Save the output

​Use cURL

​Use the Python SDK

​Next steps

Configuration

Response Format

Async Processing

Examples

Before you start

Install dependencies

Set your credentials

Parse a document

Inspect the result

Markdown

Structured elements

Save the output

Use cURL

Use the Python SDK

Next steps