Skip to content

v3.5.5: Comprehensive Sessions Module with RTH/ETH Filtering

Choose a tag to compare

View all tags
@TexasCoding TexasCoding released this 01 Sep 01:01
· 30 commits to main since this release
v3.5.5
55f807e

🎯 Release v3.5.5: Comprehensive Sessions Module

This release delivers a production-ready sessions module with timezone-aware filtering for Regular Trading Hours (RTH) and Extended Trading Hours (ETH), complete with extensive testing and documentation.

✨ Key Features

Session Filtering System

  • RTH/ETH Filtering: Intelligent filtering of market data by trading session type
  • Timezone-Aware: Proper handling of DST transitions using pytz
  • Product Support: Pre-configured sessions for 20+ futures products (ES, NQ, CL, GC, etc.)
  • Custom Sessions: Support for user-defined trading hours
  • Performance Optimized: LRU cache with TTL, lazy evaluation for large datasets

Session Analytics

  • Session Statistics: Volume profiles, price ranges, VWAP calculations per session
  • Session Indicators: Technical indicators that respect session boundaries
  • Session Comparison: Compare RTH vs ETH performance metrics
  • Multi-Instrument: Analyze sessions across multiple instruments simultaneously

🐛 Bug Fixes

  • Fixed hardcoded timezone offsets - now uses proper pytz conversion
  • Resolved DST transition edge cases
  • Fixed CI test timing issues with optimized performance thresholds
  • Corrected maintenance break logic consolidation
  • Fixed DataFrame boolean evaluation in examples
  • Resolved Polars correlation calculation methods
  • Added proper type checking for dictionary operations

📊 Testing & Quality

  • 163 tests all passing with 88% code coverage
  • Performance tests ensure <0.2s for 10,000 session checks
  • Mutation testing validates test quality
  • Full TDD methodology - tests define specification
  • All ruff, mypy, and bandit security checks passing

📚 Documentation

  • 5 comprehensive example scripts demonstrating all features
  • 935-line README with complete API documentation
  • Session guide with detailed RTH/ETH explanations
  • Migration examples for existing code

🚀 Performance Improvements

  • Cached timezone objects reduce pytz overhead by 40%
  • LRU cache with TTL prevents unbounded memory growth
  • Lazy evaluation for datasets >100k rows
  • Configurable thresholds for performance tuning

💡 Example Usage

from project_x_py.sessions import SessionFilterMixin, SessionType
import polars as pl

# Initialize session filter
session_filter = SessionFilterMixin()

# Filter data to RTH only
rth_data = await session_filter.filter_by_session(
    data=market_data,
    session_type=SessionType.RTH,
    product="ES"
)

# Check if current time is in session
is_trading = session_filter.is_in_session(
    timestamp=datetime.now(UTC),
    session_type=SessionType.RTH,
    product="MNQ"
)

📦 Installation

pip install --upgrade project-x-py==3.5.5

🔄 Migration Guide

No breaking changes - fully backward compatible. The sessions module is opt-in and doesn't affect existing code.

👥 Contributors

  • @TexasCoding - Implementation and testing
  • Claude Code - PR review and optimization suggestions

📄 Full Changelog

  • feat: Complete sessions module implementation with RTH/ETH filtering
  • feat: Session statistics and analytics capabilities
  • feat: Multi-instrument session comparison
  • fix: Timezone handling with proper pytz usage
  • fix: DST transition edge cases
  • fix: Performance test thresholds for CI
  • docs: Comprehensive examples and documentation
  • test: 163 tests with mutation testing validation
  • perf: Cached timezone objects and LRU cache implementation

Note: This release maintains full backward compatibility with v3.x. The sessions module is an optional feature that can be adopted incrementally.

Contributors

TexasCoding
Assets 2
Loading