Clean up the trash
TL;DR: Eliminate unused functions, constants, and "just-in-case" code.
Problems Addressed
Just-in-case code
Reduced maintainability
Cognitive Load
Related Code Smells
Steps
Ensure your code has good functional coverage.
Identify unused functions and constants by reviewing your code or using static analysis tools.
Analyze the added speculative code, just in case.
Remove anything unnecessary or unused.
Perform comprehensive regression testing on your code.
Sample Code
Before
from flask import Flask, jsonify, make_response
app = Flask(__name__)
HTTP_100_CONTINUE = 100
HTTP_202_ACCEPTED = 202 # Not used
HTTP_204_NO_CONTENT = 204 # Not Used
HTTP_302_FOUND = 302 # Not Used
HTTP_400_BAD_REQUEST = 400 # Not Used
HTTP_401_UNAUTHORIZED = 401 # Not Used
HTTP_403_FORBIDDEN = 403
HTTP_404_NOT_FOUND = 404
HTTP_410_GONE = 410
HTTP_500_INTERNAL_SERVER_ERROR = 500
HTTP_501_NOT_IMPLEMENTED = 501
probe_telemetry = {
"temperature": {"solar_panels": 150, "instrument_1": 50},
"position": {"x": 1000000, "y": 2000000, "z": 3000000,
"velocity": {"vx": 100, "vy": 200, "vz": 300}},
"status": {"power_level": 95, "communication_status": "OK"}
}
@app.route('/api/v1/probe/telemetry', methods=['GET'])
def get_telemetry():
return jsonify(probe_telemetry), HTTP_200_OK
# The following function is not invoked
# and not implemented
# It is a dead placeholder
@app.route('/api/v1/probe/send_command', methods=['POST'])
def send_command():
return jsonify({"message": "Command endpoint not implemented yet."}),
HTTP_501_NOT_IMPLEMENTED
@app.route('/api/v1/probe/data', methods=['GET'])
def get_data():
return jsonify({"message": "Data not found"}),
HTTP_404_NOT_FOUND
@app.route('/api/v1/probe/redirect', methods=['GET'])
def redirect_endpoint():
response = make_response(jsonify({"message": "Redirecting..."}),
HTTP_301_MOVED_PERMANENTLY)
response.headers['Location'] = '/api/v1/probe/telemetry'
return response
@app.route('/api/v1/probe/not_modified', methods=['GET'])
def not_modified_endpoint():
response = make_response(jsonify({"message": "Not Modified"}),
HTTP_304_NOT_MODIFIED)
response.headers['ETag'] = 'some_etag'
return response
@app.route('/api/v1/probe/gone', methods=['GET'])
def gone_endpoint():
return jsonify({"message": "Resource permanently gone"}),
HTTP_410_GONE
After
# 1. Ensure your code has good functional coverage.
from flask import Flask, jsonify, make_response
from http import HTTPStatus
app = Flask(__name__)
# 2. Identify unused functions and constants
# by reviewing your code or using static analysis tools.
HTTP_200_OK = HTTPStatus.OK
HTTP_301_MOVED_PERMANENTLY = HTTPStatus.MOVED_PERMANENTLY
HTTP_304_NOT_MODIFIED = HTTPStatus.NOT_MODIFIED
HTTP_404_NOT_FOUND = HTTPStatus.NOT_FOUND
HTTP_410_GONE = HTTPStatus.GONE
HTTP_501_NOT_IMPLEMENTED = HTTPStatus.NOT_IMPLEMENTED
probe_telemetry = {
"temperature": {"solar_panels": 150, "instrument_1": 50},
"position": {"x": 1000000, "y": 2000000, "z": 3000000,
"velocity": {"vx": 100, "vy": 200, "vz": 300}},
"status": {"power_level": 95, "communication_status": "OK"}
}
@app.route('/api/v1/probe/telemetry', methods=['GET'])
def get_telemetry():
return jsonify(probe_telemetry), HTTP_200_OK
# 3. Analyze the added speculative code, just in case.
@app.route('/api/v1/probe/send_command', methods=['POST'])
def send_command():
return jsonify({"message": "Command endpoint not implemented yet."}),
HTTP_501_NOT_IMPLEMENTED
@app.route('/api/v1/probe/data', methods=['GET'])
def get_data():
return jsonify({"message": "Data not found"}),
HTTP_404_NOT_FOUND
# 4. Remove anything unnecessary or unused.
# 5. Perform comprehensive regression testing on your code.
Type
[X] Semi-Automatic
Safety
This refactoring is safe if you thoroughly test your application after the changes. Static analysis tools can help ensure you don't remove anything still in use.
Why is the Code Better?
You improve clarity and reduce complexity by removing unused elements.
Your code becomes easier to understand and maintain.
Reducing speculative code also keeps your focus on current, actual requirements.
How Does it Improve the Bijection?
Dead code and speculative elements break Bijection between your software and the real-world model.
Removing these elements ensures your code accurately represents your
MAPPER, making it cleaner and closer to reality.
Limitations
Removing dead code requires confidence that it's truly unused.
This process relies on static analysis or thorough codebase knowledge, which can be error-prone without robust tools.
Refactor with AI
Without Proper Instructions | With Specific Instructions |
---|---|
ChatGPT | ChatGPT |
Claude | Claude |
Perplexity | Perplexity |
Copilot | Copilot |
Gemini | Gemini |
Tags
- Bloaters
Related Refactorings
Refactoring 003 - Extract Constant
Maxi Contieri ・ Jan 2 '22
Credits
This article is part of the Refactoring Series.
Top comments (0)