aparavi-dtc-sdk 1.1.0

Python SDK for Aparavi Data Toolchain API

Aparavi Data Toolchain SDK

A Python SDK for interacting with the Aparavi Web Services API. This SDK provides a clean, type-safe interface for validating pipelines, executing tasks, monitoring status, and managing webhooks with the Aparavi platform.

---

Description

This SDK simplifies integration with Aparavi's data processing pipelines by providing:

• validate_pipeline: Validates a pipeline structure against Aparavi's backend to ensure it is correctly formed.
• execute_pipeline: Submits a pipeline for execution on the Aparavi platform.
• get_pipeline_status: Fetches the current execution status of a pipeline task.
• teardown_pipeline: Gracefully ends a running pipeline task.
• send_payload_to_webhook: Sends file(s) to a running webhook-based pipeline.
• execute_pipeline_workflow: Performs a full end-to-end execution lifecycle for a pipeline.
• get_version: Fetches the current version of the Aparavi API/backend.

Perfect for data engineers, analysts, and developers building automated data processing workflows.

---

Setup

1. Get your credentials:

• Obtain your API key from the Aparavi console
• Note your API base URL (e.g. https://eaas.aparavi.com/)

2. Install package:

   pip install aparavi-dtc-sdk
   

3. Create .env file:

   Linux/macOS:

   touch .env
   

   Windows:

   type nul > .env
   

Env file

APARAVI_API_KEY=aparavi-dtc-api-key
APARAVI_BASE_URL=https://eaas.aparavi.com/

---

Quick Start

The Aparavi SDK supports a simplified, one-step method for executing pipelines. This method performs validation, execution, monitoring, and teardown in a single call:

import os

from dotenv import load_dotenv
from aparavi_dtc_sdk import AparaviClient

load_dotenv()

client = AparaviClient(
    base_url=os.getenv("APARAVI_BASE_URL"),
    api_key=os.getenv("APARAVI_API_KEY")
)

result = client.execute_pipeline_workflow(
    pipeline="./pipeline_config.json",
    file_glob="./*.png"
)

print(result)

NOTE: While we continue to evolve the SDK to support more flexible and modular pipeline workflows, the single-function call approach (execute_pipeline_workflow) will remain fully supported for the foreseeable future. This ensures backward compatibility and allows existing integrations to continue working without any changes.

Pre Built Pipelines

Available pre-built pipeline configurations:

• AUDIO_AND_SUMMARY: Processes audio content and produces both a transcription and a concise summary.
• SIMPLE_AUDIO_TRANSCRIBE: Processes audio files and returns transcriptions of spoken content.
• SIMPLE_PARSER: Extracts and processes metadata and content from uploaded documents.

import os

from dotenv import load_dotenv
from aparavi_dtc_sdk import AparaviClient, PredefinedPipelines # Import PredefinedPipelines enum

load_dotenv()

client = AparaviClient(
    base_url=os.getenv("APARAVI_BASE_URL"),
    api_key=os.getenv("APARAVI_API_KEY")
)

result = client.execute_pipeline_workflow(
    pipeline=PredefinedPipelines.SIMPLE_AUDIO_TRANSCRIBE, # Specify PredefinedPipelines
    file_glob="./audio/*.mp3"
)

print(result)

Power user quick start

import json
import os

from dotenv import load_dotenv
from aparavi_dtc_sdk import AparaviClient

load_dotenv()

client = AparaviClient(
    base_url=os.getenv("APARAVI_BASE_URL"),
    api_key=os.getenv("APARAVI_API_KEY")
)

try:
    validation_result = client.validate_pipeline("pipeline_config.json")
except Exception as e:
    print(f"Validation failed: {e}")

try:
    start_result = client.execute_pipeline(pipeline_config, name="my-task")
    if start_result.status == "OK":
        token = start_result.data["token"]
        task_type = start_result.data["type"]

        status_result = client.get_pipeline_status(token=token, task_type=task_type)

        end_result = client.teardown_pipeline(token=token, task_type=task_type)

except Exception as e:
    print(f"Task operation failed: {e}")

NOTE: We will soon sunset the multi-step pipeline execution process in favor of a simplified, single-method call that improves usability and reduces boilerplate code. The above pattern—using separate calls to validate_pipeline, execute_pipeline, get_pipeline_status, and teardown_pipeline—will no longer be supported in future SDK versions.

---

Development

Setting up for development

# Install in development mode
pip install -e ".[dev]"

# Find package Install
pip list | grep aparavi

# Show package info
pip show aparavi-dtc-sdk

# Run linting
flake8 aparavi-dtc-sdk/
black aparavi-dtc-sdk/
mypy aparavi-dtc-sdk/

Running Tests

pytest tests/

Support

For problems and questions, please open an issue on the GitHub repository.

---

Authors

Joshua D. Phillips
github.com/joshuadarron

Contributions welcome — feel free to submit a PR or open an issue!