neuwo-api 0.3.0

Neuwo API SDK - Python SDK client for the Neuwo content classification API.

Neuwo API Python SDK

Python SDK client for the Neuwo content classification API.

Support

• Neuwo: https://neuwo.ai
• Documentation: https://docs.neuwo.ai
• Repository: GitHub
• Issues: GitHub Issues
• Email Support: [e-mail]

Installation

Install the package using pip:

pip install neuwo-api

Install with optional development dependencies:

# For development
pip install neuwo-api[dev]

# For testing
pip install neuwo-api[test]

Requirements

• Python 3.8 or higher
• requests library (automatically installed)

Quick Start

For more detailed examples and use cases, see the examples folder in the repository.

REST API Client

from neuwo_api import NeuwoRestClient

# Initialize client
client = NeuwoRestClient(
    token="your-rest-api-token",
    base_url="https://custom.api.com",
)

# Analyze text content
response = client.get_ai_topics(
    content="Cats make wonderful pets for modern households.",
    document_id="article-123",
    headline="Why Cats Make Great Pets"
)

# Access results
print(f"Tags: {len(response.tags)}")
for tag in response.tags:
    print(f"  - {tag.value} (score: {tag.score})")

print(f"Brand Safe: {response.brand_safety.is_safe}")
print(f"IAB Categories: {len(response.marketing_categories.iab_tier_1)}")

EDGE API Client

from neuwo_api import NeuwoEdgeClient

# Initialize client
client = NeuwoEdgeClient(
    token="your-edge-api-token",
    base_url="https://custom.api.com",
    default_origin="https://yourwebsite.com"  # Optional: default origin for requests
)

# Analyze article by URL
response = client.get_ai_topics(url="https://example.com/article")

# Or wait for analysis to complete (with automatic retry)
response = client.get_ai_topics_wait(
    url="https://example.com/new-article",
    max_retries=10,
    retry_interval=6
)

print(f"Found {len(response.tags)} tags for the article")

Configuration

REST Client Parameters

Parameter	Type	Default	Description
token	str	Required	REST API authentication token
base_url	str	Required	Base URL for the API
timeout	int	60	Request timeout in seconds

Example:

client = NeuwoRestClient(
    token="your-token",
    base_url="https://custom.api.com",
    timeout=120
)

EDGE Client Parameters

Parameter	Type	Default	Description
token	str	Required	EDGE API authentication token
base_url	str	Required	Base URL for the API
timeout	int	60	Request timeout in seconds
default_origin	str	None	Default Origin header for requests

Example:

client = NeuwoEdgeClient(
    token="your-token",
    base_url="https://custom.api.com",
    default_origin="https://yoursite.com",
    timeout=90
)

API Methods

REST API

Get AI Topics

response = client.get_ai_topics(
    content="Text to analyze",           # Required
    document_id="doc123",                 # Optional: save to database
    lang="en",                            # Optional: ISO 639-1 code
    publication_id="pub1",                # Optional
    headline="Article Headline",          # Optional
    tag_limit=15,                         # Optional: max tags (default: 15)
    tag_min_score=0.1,                    # Optional: min score (default: 0.1)
    marketing_limit=None,                 # Optional
    marketing_min_score=0.3,              # Optional (default: 0.3)
    include_in_sim=True,                  # Optional (default: True)
    article_url="https://example.com"     # Optional
)

Get Similar Articles

articles = client.get_similar(
    document_id="doc123",                 # Required
    max_rows=10,                          # Optional: limit results
    past_days=30,                         # Optional: limit by date
    publication_ids=["pub1", "pub2"]      # Optional: filter by publication
)

Update Article

from datetime import date

article = client.update_article(
    document_id="doc123",                 # Required
    published=date(2024, 1, 15),          # Optional
    headline="Updated Headline",          # Optional
    writer="Author Name",                 # Optional
    category="News",                      # Optional
    content="Updated content",            # Optional
    summary="Summary",                    # Optional
    publication_id="pub1",                # Optional
    article_url="https://example.com",    # Optional
    include_in_sim=True                   # Optional
)

Train AI Topics

training_tags = client.train_ai_topics(
    document_id="doc123",                 # Required
    tags=["tag1", "tag2", "tag3"],        # Required
)

EDGE API

Get AI Topics (Single URL)

response = client.get_ai_topics(
    url="https://example.com/article",    # Required
    origin="https://yoursite.com"         # Optional: override default origin
)

Get AI Topics with Auto-Retry

response = client.get_ai_topics_wait(
    url="https://example.com/article",    # Required
    origin="https://yoursite.com",        # Optional
    max_retries=10,                       # Optional (default: 10)
    retry_interval=6,                     # Optional (default: 6s)
    initial_delay=2                       # Optional (default: 2s)
)

Get AI Topics (Multiple URLs)

# From list
results = client.get_ai_topics_list(
    urls=["https://example.com/1", "https://example.com/2"],
    origin="https://yoursite.com"
)

# From file bytes
with open("urls.txt", "rb") as f:
    urls_as_bytes = f.read()
result = client.get_ai_topics_list(urls_as_bytes)

Get Similar Articles

articles = client.get_similar(
    document_url="https://example.com/article",  # Required
    max_rows=10,                                 # Optional
    past_days=30,                                # Optional
    publication_ids=["pub1", "pub2"],            # Optional
    origin="https://yoursite.com"                # Optional
)

Raw Response Methods

All methods have _raw variants that return the raw requests.Response object:

# REST
raw_response = client.get_ai_topics_raw(content="Text")
print(raw_response.status_code)
print(raw_response.text)

# EDGE
raw_response = client.get_ai_topics_raw(url="https://example.com")
print(raw_response.json())

Error Handling

The SDK provides specific exceptions for different error scenarios:

from neuwo_api import (
    NeuwoRestClient,
    ValidationError,
    AuthenticationError,
    NoDataAvailableError,
    ContentNotAvailableError,
    NetworkError
)

client = NeuwoRestClient(
    token="your-token"
    base_url="https://custom.api.com",
)

try:
    response = client.get_ai_topics(content="Your content here")
except ValidationError as e:
    print(f"Invalid input: {e}")
except AuthenticationError as e:
    print(f"Authentication failed: {e}")
except NoDataAvailableError as e:
    print(f"Data not yet available: {e}")
except ContentNotAvailableError as e:
    print(f"Content could not be analyzed: {e}")
except NetworkError as e:
    print(f"Network error: {e}")
except Exception as e:
    print(f"Unexpected error: {e}")

Exception Hierarchy

• NeuwoAPIError - Base exception for all API errors
  • AuthenticationError - Invalid or missing token (401)
  • ForbiddenError - Token lacks permissions (403)
  • NotFoundError - Resource not found (404)
    • NoDataAvailableError - URL not yet processed (404)
  • BadRequestError - Malformed request (400)
  • ValidationError - Request validation failed (422)
  • RateLimitError - Rate limit exceeded (429)
  • ServerError - Server error (5xx)
  • NetworkError - Network communication failed
  • ContentNotAvailableError - Content tagging failed

Logging

The SDK uses Python's standard logging module. Enable logging to see detailed API communication:

from neuwo_api import setup_logger
import logging

# Enable debug logging
setup_logger(level=logging.DEBUG)

# Or just warnings and errors (default)
setup_logger(level=logging.WARNING)

# Disable logging
from neuwo_api import disable_logger
disable_logger()

Custom logging configuration:

import logging
from neuwo_api import get_logger

# Get the SDK logger
logger = get_logger()

# Add custom handler
handler = logging.FileHandler('neuwo_api.log')
handler.setLevel(logging.DEBUG)
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
handler.setFormatter(formatter)
logger.addHandler(handler)

Development

Setup Development Environment

# Clone repository
git clone https://github.com/neuwoai/neuwo-api-sdk-python.git
cd neuwo-api-sdk-python

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

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

Running Tests

# Run all tests
pytest

# Run with coverage
pytest --cov=neuwo_api --cov-report=html

# Run specific test file
pytest tests/test_rest_client.py

# Run with verbose output
pytest -v

Building Distribution

# Install build tools
pip install build twine

# Build
python -m build

# Upload to PyPI
twine upload dist/*

CI/CD

Automated Testing

The Unit Tests Coverage workflow automatically runs on every push to main or dev branches and on pull requests:

• Python versions tested: 3.8, 3.9, 3.10, 3.11, 3.12
• Coverage reporting: Results uploaded to Codecov
• Test execution: Full test suite with coverage analysis

Publishing Pipeline

The Publish Python Package workflow enables manual deployment with the following options:

Workflow Features:

• Manual trigger via GitHub Actions UI
• Deploy to TestPyPI or PyPI
• Upload artifacts to GitHub Packages
• Auto-create GitHub releases with tags
• Automatic version extraction from pyproject.toml

Setup Requirements:

Add GitHub secrets for API tokens:

• PYPI_API_TOKEN - Production PyPI token
• TEST_PYPI_API_TOKEN - TestPyPI token

Workflow Inputs

Input	Type	Default	Description
target	choice	TestPyPI	Deploy to TestPyPI or PyPI
publish_to_github	boolean	false	Upload artifacts to GitHub
create_release	boolean	false	Create GitHub release with tag (only for PyPI)

Usage:

1. Testing release:

   • Go to Actions > Publish Python Package > Run workflow
   • Select: TestPyPI
   • Test: pip install -i https://test.pypi.org/simple/ neuwo-api

2. Production release:

   • Update version in pyproject.toml, setup.py and README.md
   • Commit changes to repository
   • Go to Actions > Publish Python Package > Run workflow
   • Select: PyPI + Publish to GitHub Packages + Create GitHub release
   • Creates tag (e.g., v0.2.0), GitHub release, and publishes to PyPI

3. Verify:

   • Check PyPI: https://pypi.org/project/neuwo-api/
   • Check GitHub Release: https://github.com/neuwoai/neuwo-api-sdk-python/releases
   • Check packages appear in repo

What gets created:

• PyPI package: https://pypi.org/project/neuwo-api/
• GitHub release with .whl and .tar.gz files
• Git tag following v{VERSION} format
• Package artifacts in GitHub Actions

Versioning:

Follow Semantic Versioning (MAJOR.MINOR.PATCH):

• MAJOR: Breaking changes
• MINOR: New features (backward compatible)
• PATCH: Bug fixes (backward compatible)

License

This project is licensed under the MIT License - see the LICENCE file for details.