devrpa 0.9.3

The Developer-First Automation Framework for Python.

devrpa

The Developer-First Automation Framework for Python.

[!WARNING] 🚧 Project Status: Under Development (Beta) devrpa is currently in active development (v0.9.0). APIs and features may change. We welcome feedback and contributions!

devrpa makes building, running, and monitoring automation workflows as simple as writing Python code. It combines the reliability of enterprise RPA with the joy of modern developer tools.

Key Features (v0.9 - TUI Update)

• Visual Workflow Builder (New!): Interactively build workflows using a powerful Terminal User Interface (TUI).
• Fluent Chain API: Build complex code-first workflows with regular Python method chaining.
• Type-Safe Context: Use Pydantic models for strict validation and IDE autocomplete.
• Built-in Integrations: GitHub, Slack, Web Scraping, File I/O, Database, and more.
• Production Resilience: Automatic retries, circuit breakers, backoff strategies, and resources pooling.

---

Installation

pip install -e .
playwright install

(Note: Requires Python 3.10+)

---

🚦 Basic Usage: 3 Steps to Automation

1. Install

Install the package and browser drivers.

pip install -e .
playwright install

2. Build (The Visual Way)

Use the new interactive builder to design your workflow without writing code.

devrpa create --interactive

• Add Steps: Choose from API, Database, or Transform steps.
• Configure: Set URLs, queries, and retries in the right panel.
• Save: Click "Save JSON" to keep your work (workflow.json).

3. Run

You can run your workflow directly from the builder (Click "Run") or export it to Python code (my_workflow.py) to run in production:

python my_workflow.py

---

Visual Workflow Builder

New in v0.9: Build workflows without writing code first!

Run the interactive builder:

devrpa create --interactive

Key Capabilities

1. Interactive Design

• Rich UI: Use the TUI to add, reorder (Up/Down), and delete steps easily.
• Instant Feedback: See steps visualized with type icons ([API], [DB], [TF]) and status indicators.

2. Advanced Configuration

Configure every aspect of your steps directly in the panel:

• API: Set Method, Headers, Timeout, and Validation rules.
• Resilience: Enable Retries and Response Caching with a single click.
• Variables: specificy Output Variables to pass data between steps.

3. Real-Time Testing & Execution

Don't wait until the end to find bugs.

• Test Step: Click to actually execute the step in real-time.
  • API: Sends real HTTP requests (via httpx).
  • DB: Runs real SQL queries (via sqlalchemy).
  • Transform: Executes Python code snippets instantly.
• Live Logs: View execution status, duration, and errors immediately.

4. Smart Validation

The builder validates your workflow before you run it:

• Checks for missing required fields (URLs, Queries).
• Static Analysis: Detects references to undefined variables (e.g., using {user_id} before it's assigned).

5. Project Management

• Persistence: Save your workflow to JSON (workflow.json) and load it back later to continue working.
• Code Export: Automatically generates python code (my_workflow.py) that you can run in production.

---

Use Cases

1. API Integration & ETL

Fetch data from multiple sources, transform it, and load it into your database.

workflow = (
    flow("user_sync")
    .from_api("https://api.example.com/users", output="users")
    .transform(lambda users: [u for u in users if u['active']], output="active_users")
    .from_db("INSERT INTO users (id, email) VALUES ...")
)

2. Intelligent Automation

Combine web scraping with AI processing.

workflow = (
    flow("market_research")
    .browser_goto("https://competitor.com/pricing")
    .browser_scrape_text("#price-table", output="pricing_data")
    .llm_analyze("Extract price points", input="pricing_data", output="report")
    .slack_notify("#strategy", "{report}")
)

3. DevOps & Monitoring

Monitor systems and auto-remediate issues.

workflow = (
    flow("system_health")
    .github_list_issues("owner", "repo", output="issues")
    .where("label == 'critical'", input="issues", output="critical_issues")
    .if_condition("len(critical_issues) > 0")
        .pagerduty_trigger("High load detected")
    .end_if()
)

---

Quick Start (Code-First)

The Fluent Way

The fluent API helps you write readable, chainable workflows.

from devrpa import flow

workflow = (
    flow("github_tracker")
    # Fetch issues from GitHub (GitHub Integration)
    .github_list_issues("owner", "repo", output="issues")
    
    # Filter for bugs (Selector Syntax)
    .where("title.startswith('[Bug]')", input="issues_response", output="bugs")
    
    # Notify team (Slack Integration)
    .slack_notify("#alerts", "Found {bugs.count} bugs!")
)

# Run it
import asyncio
asyncio.run(workflow.run())

The Type-Safe Way

Leverage Pydantic to ensure your workflow data is always correct.

from pydantic import BaseModel
from devrpa import flow

# Define your context model
class UserContext(BaseModel):
    user_id: int
    email: str
    active: bool = False

# Build workflow
wf = (
    flow("user_sync", context_model=UserContext)
    .from_api("https://api.com/users/{user_id}", output="api_user")
    .transform(lambda d: d['is_active'], input="api_user", output="active")
)

# Run with validation
wf.run(initial_data=UserContext(user_id=123, email="[e-mail]"))

---

Core Concepts

Workflow & Steps

A Workflow is a sequence of Steps. Steps are atomic units of work (API calls, browser actions, data transformations).

Execution Context

Data is shared between steps via the ExecutionContext. You can access it using {variable_name} templates in step parameters or pass it safely via Pydantic models.

Integrations

devrpa comes with built-in integrations for popular services:

• Web: WebStep (Playwright)
• API: ApiStep (HTTPX)
• GitHub: GitHub.get_repo, GitHub.list_issues
• Slack: Slack.post_message
• Files: ParseCSVStep, GenerateCSVStep
• Database: DatabaseStep (SQLAlchemy async)

---

Contributing

We welcome contributions! Please see examples/ for more usage patterns.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request