Complete AI Data Analysis Agent implementation with 95.7% test coverage

tests/test_report_generation.py

@@ -0,0 +1,523 @@
+"""报告生成引擎的单元测试。"""
+
+import pytest
+import tempfile
+import os
+
+from src.engines.report_generation import (
+    extract_key_findings,
+    organize_report_structure,
+    generate_report,
+    _categorize_insight,
+    _calculate_importance,
+    _generate_report_title,
+    _generate_default_sections
+)
+from src.models.analysis_result import AnalysisResult
+from src.models.requirement_spec import RequirementSpec, AnalysisObjective
+from src.models.data_profile import DataProfile, ColumnInfo
+
+
+@pytest.fixture
+def sample_results():
+    """创建示例分析结果。"""
+    return [
+        AnalysisResult(
+            task_id='task1',
+            task_name='状态分布分析',
+            success=True,
+            data={'open': 50, 'closed': 30, 'pending': 20},
+            visualizations=['chart1.png'],
+            insights=[
+                '待处理工单占比50%，异常高',
+                '已关闭工单占比30%'
+            ],
+            execution_time=2.5
+        ),
+        AnalysisResult(
+            task_id='task2',
+            task_name='趋势分析',
+            success=True,
+            data={'trend': 'increasing'},
+            visualizations=['chart2.png'],
+            insights=[
+                '工单数量呈上升趋势',
+                '增长率为15%'
+            ],
+            execution_time=3.2
+        ),
+        AnalysisResult(
+            task_id='task3',
+            task_name='类型分析',
+            success=False,
+            data={},
+            visualizations=[],
+            insights=[],
+            error='数据缺少类型字段',
+            execution_time=0.1
+        )
+    ]
+
+
+@pytest.fixture
+def sample_requirement():
+    """创建示例需求规格。"""
+    return RequirementSpec(
+        user_input='分析工单健康度',
+        objectives=[
+            AnalysisObjective(
+                name='健康度分析',
+                description='评估工单处理的健康状况',
+                metrics=['关闭率', '处理时长', '积压情况'],
+                priority=5
+            )
+        ]
+    )
+
+
+@pytest.fixture
+def sample_data_profile():
+    """创建示例数据画像。"""
+    return DataProfile(
+        file_path='test.csv',
+        row_count=1000,
+        column_count=5,
+        columns=[
+            ColumnInfo(
+                name='status',
+                dtype='categorical',
+                missing_rate=0.0,
+                unique_count=3,
+                sample_values=['open', 'closed', 'pending']
+            ),
+            ColumnInfo(
+                name='created_at',
+                dtype='datetime',
+                missing_rate=0.0,
+                unique_count=1000
+            )
+        ],
+        inferred_type='ticket',
+        key_fields={'status': '状态', 'created_at': '创建时间'},
+        quality_score=85.0,
+        summary='工单数据，包含1000条记录'
+    )
+
+
+class TestExtractKeyFindings:
+    """测试关键发现提炼。"""
+    
+    def test_basic_functionality(self, sample_results):
+        """测试基本功能。"""
+        key_findings = extract_key_findings(sample_results)
+        
+        # 验证：返回列表
+        assert isinstance(key_findings, list)
+        
+        # 验证：只包含成功的结果
+        assert len(key_findings) == 4  # 2个任务，每个2个洞察
+        
+        # 验证：每个发现都有必需的字段
+        for finding in key_findings:
+            assert 'finding' in finding
+            assert 'importance' in finding
+            assert 'source_task' in finding
+            assert 'category' in finding
+    
+    def test_importance_sorting(self, sample_results):
+        """测试按重要性排序。"""
+        key_findings = extract_key_findings(sample_results)
+        
+        # 验证：按重要性降序排列
+        for i in range(len(key_findings) - 1):
+            assert key_findings[i]['importance'] >= key_findings[i + 1]['importance']
+    
+    def test_empty_results(self):
+        """测试空结果列表。"""
+        key_findings = extract_key_findings([])
+        
+        assert isinstance(key_findings, list)
+        assert len(key_findings) == 0
+    
+    def test_only_failed_results(self):
+        """测试只有失败的结果。"""
+        results = [
+            AnalysisResult(
+                task_id='task1',
+                task_name='失败任务',
+                success=False,
+                error='测试错误'
+            )
+        ]
+        
+        key_findings = extract_key_findings(results)
+        
+        # 失败的任务不应该产生发现
+        assert len(key_findings) == 0
+
+
+class TestCategorizeInsight:
+    """测试洞察分类。"""
+    
+    def test_anomaly_detection(self):
+        """测试异常检测。"""
+        insight = '待处理工单占比50%，异常高'
+        category = _categorize_insight(insight)
+        assert category == 'anomaly'
+    
+    def test_trend_detection(self):
+        """测试趋势检测。"""
+        insight = '工单数量呈上升趋势'
+        category = _categorize_insight(insight)
+        assert category == 'trend'
+    
+    def test_general_insight(self):
+        """测试一般洞察。"""
+        insight = '数据质量良好'
+        category = _categorize_insight(insight)
+        assert category == 'insight'
+    
+    def test_english_keywords(self):
+        """测试英文关键词。"""
+        assert _categorize_insight('This is an anomaly') == 'anomaly'
+        assert _categorize_insight('Showing growth trend') == 'trend'
+
+
+class TestCalculateImportance:
+    """测试重要性计算。"""
+    
+    def test_anomaly_importance(self):
+        """测试异常的重要性。"""
+        insight = '严重异常：系统故障'
+        importance = _calculate_importance(insight, {})
+        
+        # 异常 + 严重 = 高重要性
+        assert importance >= 4
+    
+    def test_percentage_importance(self):
+        """测试包含百分比的重要性。"""
+        insight = '占比达到80%'
+        importance = _calculate_importance(insight, {})
+        
+        # 包含百分比 = 较高重要性
+        assert importance >= 4
+    
+    def test_normal_importance(self):
+        """测试普通洞察的重要性。"""
+        insight = '数据正常'
+        importance = _calculate_importance(insight, {})
+        
+        # 默认中等重要性
+        assert importance == 3
+    
+    def test_importance_range(self):
+        """测试重要性范围。"""
+        # 测试多个洞察，确保重要性在1-5范围内
+        insights = [
+            '严重异常问题',
+            '占比80%',
+            '正常数据',
+            '轻微变化'
+        ]
+        
+        for insight in insights:
+            importance = _calculate_importance(insight, {})
+            assert 1 <= importance <= 5
+
+
+class TestOrganizeReportStructure:
+    """测试报告结构组织。"""
+    
+    def test_basic_structure(self, sample_results, sample_requirement, sample_data_profile):
+        """测试基本结构。"""
+        key_findings = extract_key_findings(sample_results)
+        structure = organize_report_structure(key_findings, sample_requirement, sample_data_profile)
+        
+        # 验证：包含必需的字段
+        assert 'title' in structure
+        assert 'sections' in structure
+        assert 'executive_summary' in structure
+        assert 'detailed_analysis' in structure
+        assert 'conclusions' in structure
+    
+    def test_with_template(self, sample_results, sample_data_profile):
+        """测试使用模板的结构。"""
+        # 创建带模板的需求
+        requirement = RequirementSpec(
+            user_input='按模板分析',
+            objectives=[
+                AnalysisObjective(
+                    name='分析',
+                    description='按模板分析',
+                    metrics=['指标1'],
+                    priority=5
+                )
+            ],
+            template_path='template.md',
+            template_requirements={
+                'sections': ['第一章', '第二章', '第三章'],
+                'required_metrics': ['指标1', '指标2'],
+                'required_charts': ['图表1']
+            }
+        )
+        
+        key_findings = extract_key_findings(sample_results)
+        structure = organize_report_structure(key_findings, requirement, sample_data_profile)
+        
+        # 验证：使用模板结构
+        assert structure['use_template'] is True
+        assert structure['sections'] == ['第一章', '第二章', '第三章']
+    
+    def test_without_template(self, sample_results, sample_requirement, sample_data_profile):
+        """测试不使用模板的结构。"""
+        key_findings = extract_key_findings(sample_results)
+        structure = organize_report_structure(key_findings, sample_requirement, sample_data_profile)
+        
+        # 验证：生成默认结构
+        assert structure['use_template'] is False
+        assert len(structure['sections']) > 0
+        assert '执行摘要' in structure['sections']
+    
+    def test_executive_summary(self, sample_results, sample_requirement, sample_data_profile):
+        """测试执行摘要组织。"""
+        key_findings = extract_key_findings(sample_results)
+        structure = organize_report_structure(key_findings, sample_requirement, sample_data_profile)
+        
+        exec_summary = structure['executive_summary']
+        
+        # 验证：包含关键发现
+        assert 'key_findings' in exec_summary
+        assert isinstance(exec_summary['key_findings'], list)
+        
+        # 验证：包含统计信息
+        assert 'anomaly_count' in exec_summary
+        assert 'trend_count' in exec_summary
+    
+    def test_detailed_analysis(self, sample_results, sample_requirement, sample_data_profile):
+        """测试详细分析组织。"""
+        key_findings = extract_key_findings(sample_results)
+        structure = organize_report_structure(key_findings, sample_requirement, sample_data_profile)
+        
+        detailed = structure['detailed_analysis']
+        
+        # 验证：包含分类
+        assert 'anomaly' in detailed
+        assert 'trend' in detailed
+        assert 'insight' in detailed
+        
+        # 验证：每个分类都是列表
+        assert isinstance(detailed['anomaly'], list)
+        assert isinstance(detailed['trend'], list)
+        assert isinstance(detailed['insight'], list)
+
+
+class TestGenerateReportTitle:
+    """测试报告标题生成。"""
+    
+    def test_health_analysis_title(self, sample_data_profile):
+        """测试健康度分析标题。"""
+        requirement = RequirementSpec(
+            user_input='分析工单健康度',
+            objectives=[]
+        )
+        
+        title = _generate_report_title(requirement, sample_data_profile)
+        
+        assert '工单' in title
+        assert '健康度' in title
+    
+    def test_trend_analysis_title(self, sample_data_profile):
+        """测试趋势分析标题。"""
+        requirement = RequirementSpec(
+            user_input='分析趋势',
+            objectives=[]
+        )
+        
+        title = _generate_report_title(requirement, sample_data_profile)
+        
+        assert '工单' in title
+        assert '趋势' in title
+    
+    def test_generic_title(self, sample_data_profile):
+        """测试通用标题。"""
+        requirement = RequirementSpec(
+            user_input='分析数据',
+            objectives=[]
+        )
+        
+        title = _generate_report_title(requirement, sample_data_profile)
+        
+        assert '工单' in title
+        assert '分析报告' in title
+
+
+class TestGenerateDefaultSections:
+    """测试默认章节生成。"""
+    
+    def test_with_anomalies(self):
+        """测试包含异常的章节。"""
+        key_findings = [
+            {
+                'finding': '异常情况',
+                'category': 'anomaly',
+                'importance': 5
+            }
+        ]
+        
+        data_profile = DataProfile(
+            file_path='test.csv',
+            row_count=100,
+            column_count=3,
+            columns=[],
+            inferred_type='ticket'
+        )
+        
+        sections = _generate_default_sections(key_findings, data_profile)
+        
+        # 验证：包含异常分析章节
+        assert '异常分析' in sections
+    
+    def test_with_trends(self):
+        """测试包含趋势的章节。"""
+        key_findings = [
+            {
+                'finding': '上升趋势',
+                'category': 'trend',
+                'importance': 4
+            }
+        ]
+        
+        data_profile = DataProfile(
+            file_path='test.csv',
+            row_count=100,
+            column_count=3,
+            columns=[],
+            inferred_type='sales'
+        )
+        
+        sections = _generate_default_sections(key_findings, data_profile)
+        
+        # 验证：包含趋势分析章节
+        assert '趋势分析' in sections
+    
+    def test_ticket_data_sections(self):
+        """测试工单数据的章节。"""
+        data_profile = DataProfile(
+            file_path='test.csv',
+            row_count=100,
+            column_count=3,
+            columns=[],
+            inferred_type='ticket'
+        )
+        
+        sections = _generate_default_sections([], data_profile)
+        
+        # 验证：包含工单相关章节
+        assert '状态分析' in sections or '类型分析' in sections
+
+
+class TestGenerateReport:
+    """测试完整报告生成。"""
+    
+    def test_basic_report_generation(self, sample_results, sample_requirement, sample_data_profile):
+        """测试基本报告生成。"""
+        report = generate_report(sample_results, sample_requirement, sample_data_profile)
+        
+        # 验证：返回字符串
+        assert isinstance(report, str)
+        
+        # 验证：报告不为空
+        assert len(report) > 0
+        
+        # 验证：包含标题
+        assert '#' in report
+        
+        # 验证：包含执行摘要
+        assert '执行摘要' in report or '摘要' in report
+    
+    def test_report_with_skipped_tasks(self, sample_results, sample_requirement, sample_data_profile):
+        """测试包含跳过任务的报告。"""
+        report = generate_report(sample_results, sample_requirement, sample_data_profile)
+        
+        # 验证：提到跳过的任务
+        assert '跳过' in report or '失败' in report
+        
+        # 验证：提到失败的任务名称
+        assert '类型分析' in report
+    
+    def test_report_with_visualizations(self, sample_results, sample_requirement, sample_data_profile):
+        """测试包含可视化的报告。"""
+        report = generate_report(sample_results, sample_requirement, sample_data_profile)
+        
+        # 验证：包含图表引用
+        assert 'chart1.png' in report or 'chart2.png' in report or '![' in report
+    
+    def test_report_with_insights(self, sample_results, sample_requirement, sample_data_profile):
+        """测试包含洞察的报告。"""
+        report = generate_report(sample_results, sample_requirement, sample_data_profile)
+        
+        # 验证：包含洞察内容
+        assert '待处理工单' in report or '趋势' in report
+    
+    def test_report_save_to_file(self, sample_results, sample_requirement, sample_data_profile):
+        """测试报告保存到文件。"""
+        with tempfile.NamedTemporaryFile(mode='w', suffix='.md', delete=False, encoding='utf-8') as f:
+            output_path = f.name
+        
+        try:
+            report = generate_report(
+                sample_results,
+                sample_requirement,
+                sample_data_profile,
+                output_path=output_path
+            )
+            
+            # 验证：文件已创建
+            assert os.path.exists(output_path)
+            
+            # 验证：文件内容与返回内容一致
+            with open(output_path, 'r', encoding='utf-8') as f:
+                saved_content = f.read()
+            
+            assert saved_content == report
+            
+        finally:
+            if os.path.exists(output_path):
+                os.unlink(output_path)
+    
+    def test_empty_results(self, sample_requirement, sample_data_profile):
+        """测试空结果列表。"""
+        report = generate_report([], sample_requirement, sample_data_profile)
+        
+        # 验证：仍然生成报告
+        assert isinstance(report, str)
+        assert len(report) > 0
+        
+        # 验证：包含基本结构
+        assert '执行摘要' in report or '摘要' in report
+    
+    def test_all_failed_results(self, sample_requirement, sample_data_profile):
+        """测试所有任务都失败的情况。"""
+        results = [
+            AnalysisResult(
+                task_id='task1',
+                task_name='失败任务1',
+                success=False,
+                error='错误1'
+            ),
+            AnalysisResult(
+                task_id='task2',
+                task_name='失败任务2',
+                success=False,
+                error='错误2'
+            )
+        ]
+        
+        report = generate_report(results, sample_requirement, sample_data_profile)
+        
+        # 验证：报告生成成功
+        assert isinstance(report, str)
+        assert len(report) > 0
+        
+        # 验证：提到失败
+        assert '失败' in report or '跳过' in report