ai-lls-lib 1.5.0rc4

Landline Scrubber core library - phone verification and DNC checking

AI LLS Library

Core business logic library and CLI tools for Landline Scrubber - phone verification and DNC checking.

Version 2.1.0 - Streaming & Provider Architecture

New features:

• Streaming support for large CSV files to reduce memory usage
• Provider architecture for clean separation of verification logic
• Contract tests ensuring all providers behave consistently

Version 2.0.0 - Breaking Changes

This is a greenfield rewrite with no backwards compatibility:

• All file-based CSV processing replaced with text-based methods
• Removed _sync suffix from all methods (everything is sync)
• process_csv_sync(file_path) → process_csv(csv_text)
• generate_results_csv(...) now returns CSV string instead of writing to file

Features

• Phone number normalization (E.164 format)
• Line type detection (mobile/landline/voip)
• DNC (Do Not Call) list checking
• DynamoDB caching with 30-day TTL
• Bulk CSV processing
• Infrastructure-aware CLI for admin operations
• AWS Lambda PowerTools integration

Installation

# Install library with Poetry
poetry install

# Install CLI globally
pip install -e .

Library Usage

Single Phone Verification

from ai_lls_lib import PhoneVerifier, DynamoDBCache

cache = DynamoDBCache(table_name="phone-cache")
verifier = PhoneVerifier(cache)

result = verifier.verify("[phone]")
print(f"Line type: {result.line_type}")
print(f"DNC: {result.dnc}")
print(f"From cache: {result.cached}")

Bulk Processing

from ai_lls_lib import BulkProcessor, PhoneVerifier, DynamoDBCache

cache = DynamoDBCache(table_name="phone-cache")
verifier = PhoneVerifier(cache)
processor = BulkProcessor(verifier)

# Process CSV text content
csv_text = "name,phone\nJohn,[phone]\nJane,[phone]"
results = processor.process_csv(csv_text)

# Generate results CSV
results_csv = processor.generate_results_csv(csv_text, results)
print(results_csv)  # CSV string with added line_type, dnc, cached columns

Streaming Large Files

For memory-efficient processing of large CSV files:

from ai_lls_lib import BulkProcessor, PhoneVerifier, DynamoDBCache

cache = DynamoDBCache(table_name="phone-cache")
verifier = PhoneVerifier(cache)
processor = BulkProcessor(verifier)

# Process CSV as a stream, yielding batches
csv_lines = open('large_file.csv').readlines()
for batch in processor.process_csv_stream(csv_lines, batch_size=100):
    print(f"Processed batch of {len(batch)} phones")
    # Each batch is a list of PhoneVerification objects

Custom Verification Providers

Use different verification providers based on your needs:

from ai_lls_lib import PhoneVerifier, DynamoDBCache
from ai_lls_lib.providers import StubProvider

# Use stub provider for testing
cache = DynamoDBCache(table_name="phone-cache")
provider = StubProvider()  # Deterministic testing provider
verifier = PhoneVerifier(cache, provider=provider)

# When external APIs are ready, switch to:
# from ai_lls_lib.providers.external import ExternalAPIProvider
# provider = ExternalAPIProvider(phone_api_key="...", dnc_api_key="...")

CLI Usage

The ai-lls CLI provides infrastructure-aware administrative tools:

Verification Commands

# Verify single phone
ai-lls verify phone [phone] --stack landline-api

# Bulk verify CSV
ai-lls verify bulk input.csv -o output.csv --stack landline-api

Cache Management

# Show cache statistics
ai-lls cache stats --stack landline-api

# Get cached entry
ai-lls cache get [phone] --stack landline-api

# Invalidate cache entry
ai-lls cache invalidate [phone] --stack landline-api

# Clear old entries
ai-lls cache clear --older-than 20 --stack landline-api

Administrative Commands

# Manage user credits
ai-lls admin user-credits user123 --add 100
ai-lls admin user-credits user123 --set 500

# List API keys
ai-lls admin api-keys --user user123

# Check queue status
ai-lls admin queue-stats

# View secrets (masked)
ai-lls admin secrets --stack landline-api

Test Stack Management

# Deploy test stack
ai-lls test-stack deploy

# Check status
ai-lls test-stack status

# Run integration tests
ai-lls test-stack test

# Delete test stack
ai-lls test-stack delete

Project Structure

ai-lls-lib/
├── src/ai_lls_lib/
│   ├── core/           # Business logic (infrastructure-agnostic)
│   │   ├── models.py   # Pydantic models
│   │   ├── verifier.py # Phone verification
│   │   ├── processor.py # Bulk processing
│   │   └── cache.py    # DynamoDB cache
│   ├── cli/            # Infrastructure-aware CLI
│   │   ├── __main__.py # Entry point
│   │   ├── commands/   # Command modules
│   │   └── aws_client.py # AWS operations
│   └── testing/        # Test utilities
│       └── fixtures.py # Test data
├── tests/
│   ├── unit/          # Mocked tests
│   └── integration/   # AWS integration tests
└── test-stack.yaml    # Test infrastructure

Testing

# Run unit tests (mocked AWS)
poetry run pytest tests/unit -v

# Deploy test stack for integration tests
ai-lls test-stack deploy

# Run integration tests (requires test stack)
TEST_STACK_NAME=ai-lls-lib-test poetry run pytest tests/integration -v

# All tests with coverage
poetry run pytest --cov=src --cov-report=html

# Clean up
ai-lls test-stack delete

Development

Current Stub Implementation

For demo purposes, verification uses stub logic based on last digit:

• Ends in 3: mobile, not on DNC
• Ends in 2: landline, not on DNC
• Ends in 1: mobile, on DNC
• Ends in 0: landline, on DNC
• Otherwise: mobile, not on DNC

TODO markers indicate where real API integration will be added.

Code Quality

# Format code
poetry run black src/ tests/
poetry run isort src/ tests/

# Type checking
poetry run mypy src/

# Run pre-commit hooks
pre-commit run --all-files

Environment Variables

• DNC_API_KEY - DNC verification API key
• DNC_CHECK_API_KEY - Alternative DNC service
• PHONE_VERIFY_API_KEY - Line type verification
• AWS_REGION - AWS region (default: us-east-1)
• AWS_PROFILE - AWS profile for CLI operations

License

Proprietary - All rights reserved

Release Process

This library uses semantic versioning and publishes to:

• TestPyPI on dev branch pushes (pre-release versions)
• PyPI on main branch pushes (stable releases)