Method identify_flaky_tests

testing/test_analyzer.py:406–433

Identify tests that have inconsistent results across runs.

(self)

            return f"Profile {bottleneck_type.lower()} phase to identify specific bottlenecks"
    
    def identify_flaky_tests(self) -> List[str]:
        """Identify tests that have inconsistent results across runs."""
        conn = sqlite3.connect(self.db_path)
        
        # Find tests with both pass and fail outcomes in recent history
        query = '''
            SELECT test_name, COUNT(DISTINCT status) as status_count,
                   COUNT(*) as total_runs,
                   SUM(CASE WHEN status = 'failed' THEN 1 ELSE 0 END) as failures
            FROM test_runs 
            WHERE timestamp > datetime('now', '-7 days')
            GROUP BY test_name
            HAVING status_count > 1 AND total_runs >= 3
        '''
        
        cursor = conn.execute(query)
        flaky_tests = []
        
        for row in cursor.fetchall():
            test_name, status_count, total_runs, failures = row
            failure_rate = failures / total_runs
            
            # Consider a test flaky if it fails 20-80% of the time
            if 0.2 <= failure_rate <= 0.8:
                flaky_tests.append(test_name)
        
        conn.close()
        return flaky_tests
    
    def calculate_test_health_score(self, analysis_report: TestAnalysisReport) -> float:
        """Calculate overall test suite health score (0-100)."""