similarity

name	similarity
description	Detect semantic code similarities using similarity-rs. Creates refactoring plans for duplicate code.

Similarity - Code Similarity Analysis

Execution Steps

1. Run similarity-rs to detect code similarities
2. Analyze detected duplicates (95%+: immediate, 85-95%: review)
3. Create refactoring plan

Commands

# Basic scan
similarity-rs ./src

# With threshold
similarity-rs ./src --threshold 0.85

# Skip test functions
similarity-rs ./src --skip-test

# Show code in output
similarity-rs ./src --print

# Strict detection (95%+)
similarity-rs ./src --threshold 0.95 --skip-test

# Include type similarity
similarity-rs ./src --experimental-types

# CI integration
similarity-rs ./src --threshold 0.95 --skip-test --fail-on-duplicates

Options

Option	Description	Default
-t, --threshold	Similarity threshold (0.0-1.0)	0.85
-m, --min-lines	Minimum lines	3
--min-tokens	Minimum tokens	30
-p, --print	Show code in output	-
--skip-test	Skip test functions	-
--exclude	Exclude directories	-
--experimental-types	Check type similarity	-
--fail-on-duplicates	Exit 1 if duplicates found	-

Interpretation

Similarity	Category	Action
95%+	Near-exact duplicate	Extract to common function
85-95%	Structural similarity	Consider generics/traits
< 85%	Possible false positive	Review manually

Workflow with cargo-coupling

# 1. Detect similar code
similarity-rs ./src --threshold 0.85 --skip-test

# 2. Check coupling impact
cargo run -- coupling ./src

# 3. Visualize in Web UI
cargo run -- coupling --web ./src

Recommended Workflow

1. Initial scan: similarity-rs . --threshold 0.8 --skip-test
2. Detailed analysis: similarity-rs ./src --threshold 0.95 --print
3. Type check: similarity-rs ./src --experimental-types --threshold 0.85