Skip to content

Analysis Lens (Math Constants & Schemas)

The Universal Schemas

The analysis_lens.py file operates as the strict mathematical and schema registry for GitGalaxy. While language_standards.py defines how to parse code, the Analysis Lens defines what that parsed data actually means.

It enforces the exact array structures, string translations, and mathematical thresholds required to turn raw regex counts into meaningful 3D visuals and forensic audits without breaking downstream Python dictionaries or GPU shaders.

The Triad of Schemas (RECORDING_SCHEMAS)

Because GitGalaxy eventually pivots data from object-oriented Python into flattened C-style arrays for WebGL, array order is critical. The Analysis Lens hardcodes the exact order of elements for the entire pipeline:

1. SIGNAL_SCHEMA (The 60-Point Heuristic Matrix)

This defines the absolute layout of the raw telemetry extracted by the LogicSplicer. Every language's regex hits are aggregated into this exact 60-element array. It represents the foundational signal data of the file, including active logic (e.g., branch, memory_alloc), passive data (e.g., doc, graveyard), and raw security triggers (e.g., sec_danger, sec_io).

2. RISK_SCHEMA (The 18-Point Exposure Vector)

This schema defines the final, processed risk metrics calculated by the SignalProcessor. It transforms the raw 60-point signal array into 18 standardized exposures (e.g., cognitive_load, tech_debt, logic_bomb, memory_corruption). Every file in the visible galaxy receives a score from 0.0 to 100.0 for every element in this array.

3. SAT_SCHEMA (Satellite Data)

Defines the strict 10-element array used to flatten individual internal functions (Satellites) into the GPU manifest. It enforces the order of metrics like LOC, Control Flow Ratio, and Big-O Depth.

The Translation Dictionaries

LLMs and human auditors cannot easily read raw backend keys like sec_bitwise_hits or cog_raw. The Analysis Lens contains the master translation maps utilized by the AuditRecorder and LLMRecorder:

  • FRIENDLY_MAP: Translates raw heuristic keys into descriptive text (e.g., converting sec_bitwise_hits into "Sub-Atomic Decryption (Custom XOR)").
  • EXPOSURE_LABELS: Formats the 18-point risk vector for the JSON audit (e.g., formatting secrets_risk as "Secrets Risk Exposure").
  • GPU_TEXTURE_LOOKUPS: A fixed string-interning dictionary that maps functional archetypes (like io, mutation, event, logic) into strict integers so the GPU knows exactly which 3D material to apply to a satellite node.

Mathematical Constants & Thresholds

The file houses the global physics constants used to calculate mass, momentum, and risk degradation:

  • Trust Constants (The Fog of War): Defines the "Opacity Tax" for dynamically typed or implicit languages (like Shell or SQL). Languages with higher opacity receive an artificial boost to their baseline risk scores to account for runtime unpredictability.
  • Network Load-Bearing Multipliers: Determines exactly how heavily PageRank and Betweenness scores are weighted when calculating a file's Systemic Threat Vector.
  • Security Tripewires: Houses the exact floating-point thresholds (e.g., > 60.0%) that trigger the ELEVATED_SURFACE_RISK and CRITICAL_THREATS_DETECTED statuses in the final SHBOM audit.

Architectural Stability

By centralizing these schemas, GitGalaxy achieves extreme stability. If a developer needs to add a new security metric to the engine, they do not have to update 15 different Python modules. They simply add the key to the RISK_SCHEMA array in the Analysis Lens, and the Splicer, Signal Processor, GPU Recorder, and Audit Recorder will automatically inherit, process, and serialize the new metric perfectly.



🌌 Powered by the blAST Engine

This documentation is part of the GitGalaxy Ecosystem, an AST-free, LLM-free heuristic knowledge graph engine.