# pyscn configuration file
# This file configures all analysis features of pyscn
# Place this file in your project root to customize analysis behavior
# =============================================================================
# OUTPUT CONFIGURATION
# =============================================================================
[output]
format = "text" # Default output format: text, json, yaml, csv, html
show_details = false # Show detailed breakdown by default
sort_by = "complexity" # Default sort: name, complexity, risk
min_complexity = 1 # Minimum complexity to report
directory = "" # Output directory for reports (empty = current directory)
# =============================================================================
# COMPLEXITY ANALYSIS
# =============================================================================
[complexity]
enabled = true # Enable complexity analysis
low_threshold = 15 # Functions with complexity ≤ 15 are low risk
medium_threshold = 25 # Functions with complexity 16-25 are medium risk
# Functions with complexity ≥ 26 are high risk
max_complexity = 0 # Maximum allowed complexity (0 = no limit)
report_unchanged = true # Report functions with complexity = 1
# =============================================================================
# DEAD CODE DETECTION
# =============================================================================
[dead_code]
enabled = true # Enable dead code detection
min_severity = "warning" # Minimum severity to report: critical, warning, info
show_context = false # Show surrounding code context
context_lines = 3 # Number of context lines to show
sort_by = "severity" # Sort by: severity, line, file, function
# Detection options - configure what types of dead code to detect
detect_after_return = true # Code after return statements
detect_after_break = true # Code after break statements
detect_after_continue = true # Code after continue statements
detect_after_raise = true # Code after raise statements
detect_unreachable_branches = true # Unreachable conditional branches
# Patterns to ignore (regex patterns)
ignore_patterns = []
# =============================================================================
# CLONE DETECTION
# =============================================================================
[clones]
# Analysis settings
min_lines = 5 # Minimum lines for clone candidates
min_nodes = 10 # Minimum AST nodes for clone candidates
max_edit_distance = 50.0 # Maximum edit distance allowed
ignore_literals = false # Ignore differences in literal values
ignore_identifiers = false # Ignore differences in identifier names
cost_model_type = "python" # Cost model: default, python, weighted
# Threshold settings for clone type classification (0.0 - 1.0)
type1_threshold = 0.98 # Type-1: Identical code (except whitespace/comments)
type2_threshold = 0.95 # Type-2: Syntactically identical (different identifiers)
type3_threshold = 0.85 # Type-3: Syntactically similar (small modifications)
type4_threshold = 0.70 # Type-4: Functionally similar (different syntax)
similarity_threshold = 0.8 # General minimum similarity threshold
# Advanced analysis settings
enable_dfa = true # Enable Data Flow Analysis for enhanced Type-4 detection
# Filtering settings
min_similarity = 0.0 # Minimum similarity to report
max_similarity = 1.0 # Maximum similarity to report
enabled_clone_types = ["type1", "type2", "type3", "type4"] # Clone types to detect
max_results = 10000 # Maximum results (0 = no limit)
# Grouping settings
grouping_mode = "connected" # Grouping mode: connected, star, complete_linkage, k_core
grouping_threshold = 0.85 # Threshold for grouping
k_core_k = 2 # K-core parameter
# LSH acceleration settings
lsh_enabled = "auto" # Enable LSH: true, false, auto (based on project size)
lsh_auto_threshold = 500 # Auto-enable LSH for projects with >500 fragments
lsh_similarity_threshold = 0.50 # LSH similarity threshold
lsh_bands = 32 # Number of LSH bands
lsh_rows = 4 # Number of rows per band
lsh_hashes = 128 # Number of hash functions
# Performance settings
max_memory_mb = 100 # Memory limit in MB (0 = no limit)
batch_size = 100 # Batch size for processing
enable_batching = true # Enable batching for large projects
max_goroutines = 4 # Maximum concurrent goroutines
timeout_seconds = 300 # Timeout for clone analysis
# Output settings
show_details = false # Show detailed clone information
show_content = false # Include source code content in output
sort_by = "similarity" # Sort by: similarity, size, location, type
group_clones = true # Group related clones together
# =============================================================================
# CBO (COUPLING BETWEEN OBJECTS) ANALYSIS
# =============================================================================
[cbo]
enabled = true # Enable CBO analysis
low_threshold = 5 # CBO values ≤ 5 are low risk
medium_threshold = 10 # CBO values 6-10 are medium risk
# CBO values > 10 are high risk
include_builtins = false # Include built-in type dependencies
include_imports = true # Include imported module dependencies
min_cbo = 0 # Minimum CBO to report
show_zeros = false # Include classes with CBO = 0
# =============================================================================
# ANALYSIS CONFIGURATION
# =============================================================================
[analysis]
recursive = true # Recursively analyze directories
follow_symlinks = false # Follow symbolic links
include_patterns = ["**/*.py"] # File patterns to include
exclude_patterns = [ # File patterns to exclude
"**/test_*.py",
"**/*_test.py",
"**/__pycache__/*",
"**/*.pyc",
"**/.pytest_cache/",
".tox/",
"venv/",
"env/",
".venv/",
".env/"
]
# =============================================================================
# ARCHITECTURE VALIDATION
# =============================================================================
# Define architectural layers and validate dependencies
[architecture]
enabled = true
validate_layers = true
[[architecture.layers]]
name = "presentation"
packages = ["router", "routers", "route", "routes", "endpoint", "endpoints", "handler", "handlers", "controller", "controllers", "view", "views", "api", "apis", "ui", "web", "rest", "graphql"]
[[architecture.layers]]
name = "application"
packages = ["service", "services", "usecase", "usecases", "use_case", "use_cases", "workflow", "workflows", "command", "commands", "query", "queries", "manager", "managers", "dependencies", "dependency"]
[[architecture.layers]]
name = "domain"
packages = ["model", "models", "entity", "entities", "schema", "schemas", "domain", "domains", "core", "business", "aggregate", "aggregates", "valueobject", "valueobjects"]
[[architecture.layers]]
name = "infrastructure"
packages = ["repository", "repositories", "repo", "repos", "db", "database", "adapter", "adapters", "persistence", "storage", "cache", "client", "clients", "external"]
[[architecture.rules]]
from = "presentation"
allow = ["presentation", "application", "domain", "infrastructure"]
[[architecture.rules]]
from = "application"
allow = ["application", "domain", "infrastructure"]
[[architecture.rules]]
from = "domain"
allow = ["domain", "infrastructure"]
deny = ["presentation", "application"]
[[architecture.rules]]
from = "infrastructure"
allow = ["infrastructure", "domain", "application"]
# =============================================================================
# EXAMPLE CONFIGURATIONS
# =============================================================================
# Uncomment and modify these sections for common use cases:
# # Strict mode - fail on any issues
# [complexity]
# max_complexity = 10
# [dead_code]
# min_severity = "critical"
#
# # Relaxed mode - only catch major issues
# [complexity]
# low_threshold = 15
# medium_threshold = 25
# [dead_code]
# min_severity = "warning"
#
# # Clone detection focused on exact matches
# [clones]
# similarity_threshold = 0.95
# enabled_clone_types = ["type1", "type2"]
#
# # Performance optimized for large codebases
# [clones]
# max_goroutines = 8
# max_memory_mb = 1024
# min_lines = 10
# min_nodes = 20
# lsh_enabled = true
