requirements.txt

# Compliant.one Platform Requirements
# RegTech Platform for FATF-aligned AML/KYC Compliance Solutions

# Web Framework
streamlit>=1.28.0
fastapi>=0.104.1
uvicorn>=0.24.0

# Data Processing
pandas>=2.1.0
numpy>=1.25.0
sqlalchemy>=2.0.0
psycopg2-binary>=2.9.0
openpyxl>=3.1.0
python-docx>=0.8.11

# PDF Processing Libraries
PyPDF2>=3.0.0
pdfplumber>=0.10.0
pdfminer.six>=20221105

# Visualization
plotly>=5.17.0
altair>=5.1.0

# AI/ML Libraries
scikit-learn>=1.3.0
spacy>=3.7.0
transformers>=4.35.0
torch>=2.1.0
textblob>=0.17.0
nltk>=3.8.0

# OSINT & Data Collection
requests>=2.31.0
beautifulsoup4>=4.12.0
feedparser>=6.0.0
selenium>=4.15.0
scrapy>=2.11.0
crawl4ai>=0.2.0

# Identity Verification & OCR
opencv-python>=4.8.0
face-recognition>=1.3.0
pillow>=10.1.0
pytesseract>=0.3.10
pdf2image>=1.16.3

# Security & Encryption
cryptography>=41.0.0
bcrypt>=4.1.0
python-jose>=3.3.0
passlib>=1.7.0
streamlit-authenticator>=0.2.3

# Advanced NLP & ML (from sigmanaut merge)
fuzzywuzzy>=0.18.0
python-Levenshtein>=0.21.0
# python-mitre-attack>=1.0.0  # Package not available for Python 3.12
stix2>=3.0.1
yara-python>=4.3.1

# Task Queue & Scheduling
celery>=5.3.0
# celery-beat>=2.5.0  # Package not available for Python 3.12
schedule>=1.2.0
redis>=5.0.0

# Enhanced Monitoring & Alerting
python-telegram-bot>=20.7
sendgrid>=6.10.0
loguru>=0.7.0
prometheus-client>=0.19.0

# Advanced Data Collection
newspaper3k>=0.2.8
aiohttp>=3.9.0
httpx>=0.25.0

# Database & Caching
redis>=5.0.0
pymongo>=4.6.0
motor>=3.3.0
mongoengine>=0.27.0
alembic>=1.12.0

# API & HTTP
httpx>=0.25.0
aiohttp>=3.9.0
websockets>=11.0.0

# Async Processing
asyncio-mqtt>=0.15.0
celery>=5.3.0
kombu>=5.3.0

# Monitoring & Logging
prometheus-client>=0.19.0
structlog>=23.2.0
sentry-sdk>=1.38.0

# Testing
pytest>=7.4.0
pytest-asyncio>=0.21.0
pytest-cov>=4.1.0
faker>=20.1.0

# Documentation
mkdocs>=1.5.0
mkdocs-material>=9.4.0

# Development Tools
black>=23.10.0
flake8>=6.1.0
mypy>=1.7.0
pre-commit>=3.5.0

# Configuration & Environment
python-dotenv>=1.0.0
pydantic>=2.5.0
pydantic-settings>=2.1.0

# Date & Time
python-dateutil>=2.8.0
pytz>=2023.3

# Scraping & Data Collection
schedule>=1.2.0
croniter>=1.4.0
aiofiles>=23.2.0
lxml>=4.9.0

# File Processing
openpyxl>=3.1.0
python-docx>=1.1.0
PyPDF2>=3.0.0
reportlab>=4.0.0
fpdf>=3.0.0
pdfplumber>=0.9.0

# Text Processing & Analysis
fuzzywuzzy>=0.18.0
python-Levenshtein>=0.21.0
langchain>=0.0.300
textblob>=0.17.0
nltk>=3.8.0
cohere>=4.32.0

# RAGFlow Integration Dependencies
python-multipart>=0.0.6
filetype>=1.2.0
python-magic>=0.4.27
pypdf>=3.17.0  # Enhanced PDF processing for RAGFlow
faiss-cpu>=1.7.4  # Vector similarity search
chromadb>=0.4.0  # Vector database
sentence-transformers>=2.2.0  # Embedding models
openai>=1.3.0  # OpenAI API integration
anthropic>=0.7.0  # Claude API integration
tiktoken>=0.5.0  # Token counting for LLMs

# Geographic & Country Data
pycountry>=23.12.0
geopy>=2.4.0

# Compliance Specific
python-ldap>=3.4.0  # For enterprise integration
xmltodict>=0.13.0  # For XML data processing

# Performance & Monitoring
memory-profiler>=0.61.0
psutil>=5.9.0

# Email & Communication
sendgrid>=6.10.0
twilio>=8.11.0

# Enterprise Features
ldap3>=2.9.0
pysaml2>=7.4.0
authlib>=1.2.0

# Threat Intelligence Dependencies
aiohttp>=3.9.0
asyncio>=3.4.3
python-dotenv>=1.0.0
sqlite3  # Built-in Python module

# Additional Security Libraries
cryptography>=41.0.0
bcrypt>=4.1.0
passlib>=1.7.4

# Data Validation
pydantic>=2.5.0
validators>=0.22.0

# Background Tasks
celery>=5.3.0
redis>=5.0.0