intermediate⏱60 minutesLección 8 de 9

Static Analysis & Code Scanning

Explore static analysis tools: mypy for type checking, bandit for security scanning, and SonarQube for comprehensive code quality analysis

Static Analysis & Code Scanning

Static analysis examines your code without running it to find bugs, security vulnerabilities, type errors, and code quality issues. Unlike linters (which enforce style), static analyzers find real potential bugs in your logic.

Types of Static Analysis

💡 Tip: Use Ctrl + Mouse Wheel to zoom, or click the fullscreen icon

Tool	Purpose	Finds	Speed
mypy	Type checking	Type mismatches, None safety, missing returns	Moderate
bandit	Security scanning	SQL injection, hardcoded passwords, unsafe eval	Fast
SonarQube	Comprehensive quality	Bugs, vulnerabilities, code smells, duplication	Slow

Type Checking with mypy

mypy is the standard static type checker for Python. It verifies that your type annotations are consistent with how the code actually uses values.

Installation

bash

# Install mypy
pip install mypy

# Install type stubs for libraries
pip install types-requests types-PyYAML types-python-dateutil

# Run mypy on a file
mypy src/main.py

# Run mypy on a module
mypy src

# Run mypy with strict mode
mypy --strict src/

# Generate a configuration file
mypy --generate-config

Basic mypy Configuration

toml

# pyproject.toml
[tool.mypy]
python_version = "3.12"
strict = true
ignore_missing_imports = true
disallow_untyped_defs = true
disallow_any_unimported = false
no_implicit_optional = true
warn_redundant_casts = true
warn_unused_ignores = true
warn_return_any = true
warn_unreachable = true

exclude = [
    "migrations/",
    "build/",
    ".venv/",
    "tests/",
]

[[tool.mypy.overrides]]
module = "tests.*"
disallow_untyped_defs = false
ignore_missing_imports = true

mypy in Action

python

# BEFORE: No type annotations (mypy cannot help)
def process_user(user_data):
    name = user_data["name"]
    age = user_data["age"]
    return f"{name} is {age} years old"


def calculate_discount(price, percentage):
    return price * percentage

python

# AFTER: With type annotations (mypy catches errors)
from typing import Protocol, Any


class UserData(Protocol):
    name: str
    age: int


def process_user(user_data: UserData) -> str:
    name: str = user_data.name
    age: int = user_data.age
    return f"{name} is {age} years old"


def calculate_discount(price: float, percentage: float) -> float:
    return price * percentage

Common mypy Errors

python

from typing import Optional, Union, Any, Callable


# Error 1: None safety
def get_user_name(user_id: int) -> Optional[str]:
    if user_id == 1:
        return "Alice"
    return None


def process_user(user_id: int) -> str:
    name = get_user_name(user_id)
    # ERROR: Argument 1 to "len" has incompatible type "Optional[str]"; expected "str"
    return f"User: {name.upper()}"  # mypy error!
    # Fix: name = get_user_name(user_id) or "Unknown"
    #       name.upper()


# Error 2: Type mismatch
def add(a: int, b: int) -> int:
    return a + b


result = add(1, "2")  # mypy error: Argument 2 to "add" has incompatible type "str"; expected "int"


# Error 3: Missing return
def validate_age(age: int) -> bool:
    if age < 0:
        return False
    # ERROR: Missing return statement
    # Fix: return age >= 0


# Error 4: Incompatible return type
def get_item(items: list[int], index: int) -> int:
    if index < len(items):
        return items[index]
    return None  # mypy error: Incompatible return value type "None"; expected "int"
    # Fix: return -1 or use Optional[int]


# Error 5: Any type leak
def process(data: Any) -> str:
    return data  # mypy warns about returning Any unless silenced

Gradual Typing Strategy

python

# Stage 1: Function signatures only (most value)
def create_user(name: str, email: str) -> User:
    ...

# Stage 2: Add return types everywhere
def find_user(user_id: int) -> Optional[User]:
    ...

# Stage 3: Full annotations (internal variables too)
def calculate_total(items: list[Item]) -> float:
    total: float = 0.0
    for item in items:
        total += item.price * item.quantity
    return total

Type Stubs for External Libraries

bash

# Types for popular libraries
pip install types-requests
pip install types-PyYAML
pip install types-python-dateutil
pip install types-beautifulsoup4
pip install types-redis
pip install types-pytz

python

# With types-requests installed, mypy checks this:
import requests


def fetch_data(url: str) -> dict:
    response = requests.get(url)
    response.raise_for_status()
    return response.json()  # Correctly typed as dict

mypy Strict Mode Flags

Flag	What It Does
`--strict`	Enables all strict flags below
`--disallow-untyped-defs`	All functions must have type annotations
`--disallow-incomplete-defs`	Partial annotations not allowed (e.g., `def f(x)` without typing `x`)
`--disallow-untyped-calls`	Cannot call functions without type annotations
`--no-implicit-optional`	`Optional[T]` must be explicit
`--warn-redundant-casts`	Warns about unnecessary `cast()` calls
`--warn-unused-ignores`	Warns about unused `# type: ignore` comments
`--warn-return-any`	Warns when function returns inferred `Any`
`--warn-unreachable`	Warns about unreachable code

Security Scanning with bandit

Bandit finds common security issues in Python code.

Installation

bash

# Install bandit
pip install bandit

# Run bandit on a file
bandit -r src/

# Run with specific severity/confidence
bandit -r src/ -ll  # Only medium+ severity
bandit -r src/ -iii # Only medium+ confidence

# Generate HTML report
bandit -r src/ -f html -o bandit_report.html

# Generate JSON report
bandit -r src/ -f json -o bandit_report.json

# Use a baseline (skip known issues)
bandit -r src/ -b .bandit_baseline.json

Bandit Configuration

ini

# .bandit
[bandit]
exclude: tests,.venv,migrations
tests: B101,B102,B301,B302,B303,B304,B305,B306,B307,B308,B309,B310,B311,B312,B313,B314,B315,B316,B317,B318,B319,B320,B321,B322,B323,B324,B325,B326,B327,B328,B329,B330,B331,B332,B333,B334,B335,B336,B337,B338,B339,B340,B341,B342,B343,B344,B345,B346,B347,B348,B349,B350,B351,B352,B353,B354,B355,B356,B357,B358,B359,B360,B361,B362,B363,B364,B365,B366,B367,B368,B369,B370,B371,B372,B373,B374,B375,B376,B377,B378,B379,B380,B381,B382,B383,B384,B385,B386,B387,B388,B389,B390,B391,B392,B393,B394,B395,B396,B397,B398,B399,B400,B401,B402,B403,B404,B405,B406,B407,B408,B409,B410,B411,B412,B413,B414,B415,B416,B417,B418,B419,B420,B421,B422,B423,B424,B425,B426,B427,B428,B429,B430,B431,B432,B433,B434,B435,B436,B437,B438,B439,B440,B441,B442,B443,B444,B445,B446,B447,B448,B449,B450,B451,B452,B453,B454,B455,B456,B457,B458,B459,B460,B461,B462,B463,B464,B465,B466,B467,B468,B469,B470,B471,B472,B473,B474,B475,B476,B477,B478,B479,B480,B481,B482,B483,B484,B485,B486,B487,B488,B489,B490,B491,B492,B493,B494,B495,B496,B497,B498,B499,B500
skips: B101,B311

Bandit Security Issues

python

import hashlib
import subprocess
import os

# B102: exec() usage (dangerous)
exec("print('hello')")  # BANDIT: B102

# B201: subprocess with shell=True
subprocess.call("ls -la", shell=True)  # BANDIT: B201

# B303: MD5 used for password
hashlib.md5(b"password")  # BANDIT: B303

# B105: Hardcoded password
password = "super_secret_123"  # BANDIT: B105

# B106: Hardcoded API key
API_KEY = "sk-abc123def456"  # BANDIT: B106

# B108: Temp file in insecure location
file_path = "/tmp/data.txt"  # BANDIT: B108

# B110: Bare except (catches all exceptions)
try:
    do_something()
except:  # BANDIT: B110
    pass

# B112: Try/except/pass
try:
    do_something()
except ValueError:
    pass  # BANDIT: B112

# B301: Pickle (dangerous deserialization)
import pickle
data = pickle.loads(unsafe_data)  # BANDIT: B301

# B320: SQL injection risk
cursor.execute("SELECT * FROM users WHERE id = " + user_input)  # BANDIT: B320

Writing Secure Code (Bandit-Safe)

python

import secrets
import hashlib
import subprocess
from typing import Optional


# ✅ Safe: Use environment variables for secrets
DATABASE_PASSWORD = os.environ.get("DATABASE_PASSWORD")
if not DATABASE_PASSWORD:
    raise ValueError("DATABASE_PASSWORD not set")


# ✅ Safe: Use parameterized queries
def get_user(user_id: int) -> Optional[dict]:
    cursor.execute("SELECT * FROM users WHERE id = ?", (user_id,))
    return cursor.fetchone()


# ✅ Safe: Use subprocess without shell
def list_directory(path: str) -> list[str]:
    result = subprocess.run(
        ["ls", "-la", path],
        capture_output=True,
        text=True,
        check=True,
    )
    return result.stdout.splitlines()


# ✅ Safe: Use secrets module for security-sensitive randomness
def generate_reset_token() -> str:
    return secrets.token_hex(32)


# ✅ Safe: Use hashlib securely
def hash_password(password: str) -> str:
    salt = secrets.token_hex(16)
    return hashlib.pbkdf2_hmac("sha256", password.encode(), salt.encode(), 100000).hex()


# ✅ Safe: Use context managers for files
def read_config(path: str) -> str:
    with open(path) as f:
        return f.read()

Bandit CI Integration

yaml

# .github/workflows/security.yml
name: Security Scan

on: [pull_request, push]

jobs:
  security:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Install bandit
        run: pip install bandit

      - name: Run bandit security scan
        run: |
          bandit -r src/ -ll -f json -o bandit_report.json || true

      - name: Upload report
        uses: actions/upload-artifact@v4
        with:
          name: bandit-report
          path: bandit_report.json

      - name: Fail on high severity issues
        run: |
          HIGH_ISSUES=$(python -c "
          import json
          with open('bandit_report.json') as f:
              data = json.load(f)
          high = [r for r in data.get('results', []) 
                  if r.get('issue_severity') == 'HIGH']
          print(f'High severity issues: {len(high)}')
          for r in high:
              print(f'  - {r[\"test_id\"]}: {r[\"issue_text\"]} at {r[\"filename\"]}:{r[\"line_number\"]}')
          exit(1 if high else 0)
          ")

SonarQube: Comprehensive Quality Analysis

SonarQube provides a dashboard for tracking code quality over time, including bugs, vulnerabilities, code smells, and technical debt.

Running SonarQube Locally

bash

# Run SonarQube with Docker
docker run -d --name sonarqube \
  -p 9000:9000 \
  sonarqube:community

# Install SonarScanner
pip install sonar-scanner

# Run analysis
sonar-scanner \
  -Dsonar.projectKey=my_project \
  -Dsonar.sources=src \
  -Dsonar.tests=tests \
  -Dsonar.python.coverage.reportPaths=coverage.xml \
  -Dsonar.host.url=http://localhost:9000 \
  -Dsonar.login=your_token

sonar-project.properties

properties

# sonar-project.properties
sonar.projectKey=my_python_project
sonar.projectName=My Python Project
sonar.projectVersion=1.0
sonar.sources=src
sonar.tests=tests
sonar.sourceEncoding=UTF-8
sonar.language=py
sonar.python.version=3.12

# Coverage
sonar.python.coverage.reportPaths=coverage.xml
sonar.python.xunit.reportPath=test-report.xml

# Exclusions
sonar.exclusions=**/migrations/**,**/__init__.py
sonar.test.exclusions=**/migrations/**

# Quality gates
sonar.qualitygate.wait=true
sonar.qualitygate.timeout=300

# Duplication
sonar.cpd.exclusions=**/migrations/**

SonarQube Quality Metrics

Metric	What It Measures	Target
Bugs	Runtime errors and incorrect behavior	0
Vulnerabilities	Security weaknesses	0
Code Smells	Maintainability issues	Low
Technical Debt	Time to fix all issues	< 5%
Duplications	Repeated code blocks	< 3%
Coverage	Test coverage percentage	>= 80%
Security Rating	A-E based on vulnerabilities	A
Reliability Rating	A-E based on bugs	A
Maintainability Rating	A-E based on code smells	A

💡 Tip: Use Ctrl + Mouse Wheel to zoom, or click the fullscreen icon

Comprehensive CI Workflow

yaml

# .github/workflows/static-analysis.yml
name: Static Analysis

on:
  pull_request:
  push:
    branches: [main]

jobs:
  type-check:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Install dependencies
        run: |
          pip install -r requirements.txt
          pip install mypy types-requests types-PyYAML

      - name: Run mypy
        run: mypy src/ --strict

  security:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Install bandit
        run: pip install bandit

      - name: Run bandit
        run: bandit -r src/ -ll

  sonarqube:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Install dependencies
        run: |
          pip install -r requirements.txt
          pip install pytest pytest-cov

      - name: Run tests with coverage
        run: |
          pytest --cov=src --cov-report=xml --junitxml=test-report.xml

      - name: SonarQube Scan
        uses: SonarSource/sonarqube-scan-action@v4
        env:
          SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
        with:
          args: >
            -Dsonar.python.coverage.reportPaths=coverage.xml
            -Dsonar.python.xunit.reportPath=test-report.xml

Practice Exercises

mypy Setup: Install mypy and configure it with --strict mode. Add type annotations to a Python file until mypy reports no errors. Start with a file that has no type annotations.
Fix mypy Errors: Given this code, fix all mypy errors:

python
def get_user(id): if id == 1: return {"name": "Alice", "age": 30} return None def process_user(user): return user["name"].upper()
Bandit Security Audit: Create a Python file with 5 common security vulnerabilities (hardcoded passwords, SQL injection, eval usage, shell injection, pickle). Run bandit and verify it catches all 5.
Fix Security Issues: Given the bandit findings from exercise 3, fix all security vulnerabilities following best practices. Verify bandit reports no issues after fixes.
SonarQube Setup: Run SonarQube locally using Docker. Configure a project and run SonarScanner on a Python project. Analyze the quality gate results.
Gradual Typing: Take a 200+ line Python module with no type annotations. Add type annotations gradually: first function signatures, then return types, then internal variables. Track how mypy catches issues at each stage.
CI Integration: Create a GitHub Actions workflow that runs mypy (strict mode) and bandit on every PR. The pipeline should fail if any type error exists or if bandit finds high-severity issues.
Quality Gate Setup: Configure SonarQube quality gates that require: 0 bugs, 0 vulnerabilities, at least 80% coverage, and less than 5% duplicated code. Then create a PR that violates one gate and verify it fails.

Summary

mypy catches type errors at compile time — the earlier the annotation, the more it helps
Strict mode (--strict) is the most valuable — it forces complete annotations
bandit finds security issues: hardcoded secrets, injection risks, unsafe APIs
SonarQube provides comprehensive quality tracking: bugs, vulnerabilities, technical debt
CI integration makes static analysis a mandatory part of the development process
Fix issues at the source — don't suppress warnings without understanding them
Static analysis complements testing — tests verify behavior, static analysis verifies code structure

✅Success

Static analysis tools are your automated code reviewers. mypy checks your types, bandit checks your security, and SonarQube tracks your quality over time. Together, they catch what tests miss.

Progreso89%

Code Formatting

Complete CI/CD Pipeline

Contenido del Curso

1. Introduction to TDD 2. Writing Your First Tests 3. Unit Testing Best Practices 4. Test Coverage & Quality 5. Pre-Commit Hooks 6. Linting & Code Quality 7. Code Formatting 8. Static Analysis & Code Scanning 9. Complete CI/CD Pipeline