Refactor: centralize config, improve logging, and clean up code

[d-roho]

Summary

This PR significantly improves code maintainability and observability by extracting hardcoded configuration values into a central config module, adding comprehensive logging throughout the application, and refactoring several functions for clarity and robustness.

Key Changes

Configuration Management

• Created new config.py module to centralize all configuration values (GSI settings, prediction thresholds, map codes, weapon classifications)
• Moved hardcoded values from gsi_pinger.py, snapshot_parser.py, and MainApp.py into config
• Updated weapon classification lists to use frozenset for better performance and immutability
• Fixed de_ancient map code collision (now uses code 8 instead of 4)

Logging Infrastructure

• Added structured logging throughout the application with appropriate log levels (DEBUG, INFO, WARNING)
• Configured logging in MainApp.py with -v flag support for verbose output
• Replaced print statements with logger calls for better observability and control
• Added debug logging for transient errors and GSI server activity

Code Quality Improvements

• Refactored snapshot_parser.py:

  • Simplified player iteration using enumerate and f-strings
  • Consolidated weapon counting into a single pass over players
  • Improved error handling with try-except for missing/malformed data
  • Enhanced docstrings with detailed attribute ordering

• Refactored MainApp.py:

  • Made Windows-only imports conditional with graceful fallback
  • Simplified welcome message display with better file handling
  • Consolidated match-start checking logic
  • Improved pause detection and prediction output formatting
  • Cleaner exception handling with more informative messages

• Refactored gsi_pinger.py:

  • Improved HTTP server implementation with better authentication
  • Used mutable container pattern to avoid global state
  • Added proper exception handling and logging
  • Suppressed per-request HTTP noise in logs

Robustness

• Added defensive .get() calls with defaults throughout to handle missing GSI data
• Improved error recovery in main prediction loop
• Better handling of edge cases (empty servers, warmup detection, bomb timer)

Notable Implementation Details

• Freezetime prediction override now uses configurable constant instead of magic number
• Bomb near-expiry threshold is now configurable
• Weapon classification uses frozensets for O(1) lookup performance
• GSI authentication and server configuration centralized for easier deployment customization

https://claude.ai/code/session_01BpRrR55tsrkVo9W5rtwYMK