OpenClaw Agent
f26d4d8f7f
- Multi-source content aggregation (Bilibili/YouTube/Zhihu) - Smart inbox management with AI quality scoring - Long video chunking analysis (30min+ auto-segmentation) - Topic-based organization - Git auto-sync to Obsidian - Template system with Jinja2 and fallback rendering - CLI commands: subscribe, inbox, process, skip, status, search - Comprehensive test suite with unit and integration tests - Documentation: README.md and SKILL.md Checkpoint 7-8 completed by Agent E
316 lines
11 KiB
Python
316 lines
11 KiB
Python
"""Obsidian KB - 测试验证
|
||
验证Checkpoint 4-5的所有功能
|
||
"""
|
||
import sys
|
||
import os
|
||
sys.path.insert(0, os.path.dirname(__file__))
|
||
|
||
from src.models import Content, ContentType, ContentStatus, TimelineEntry
|
||
from src.light_scanner import LightScanner
|
||
from src.deep_analyzer import DeepAnalyzer
|
||
from src.pipeline import ProcessingPipeline
|
||
from src.content_chunker import ContentChunker, get_chunking_strategy, chunk_content, Chunk
|
||
from src.ai_client import AIClient
|
||
|
||
|
||
def test_light_scanner():
|
||
"""测试C4.1: 轻量扫描"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C4.1: 轻量扫描")
|
||
print("="*60)
|
||
|
||
# 创建测试内容
|
||
test_content = Content(
|
||
id="test-001",
|
||
url="https://example.com/video1",
|
||
title="测试视频：人工智能的未来发展",
|
||
content_type=ContentType.VIDEO,
|
||
source="youtube",
|
||
duration=600 # 10分钟
|
||
)
|
||
|
||
# 测试文本
|
||
test_text = """
|
||
大家好，今天我们来讨论人工智能的未来发展。
|
||
人工智能正在改变我们的生活和工作方式。
|
||
机器学习是人工智能的核心技术之一。
|
||
深度学习让神经网络能够处理更复杂的任务。
|
||
自然语言处理技术让机器能够理解人类语言。
|
||
计算机视觉让机器能够看懂图像和视频。
|
||
强化学习让AI能够通过试错来学习。
|
||
未来，AI将在医疗、教育、交通等领域发挥重要作用。
|
||
我们需要关注AI伦理和安全问题。
|
||
让我们共同迎接AI时代的到来。
|
||
""" * 10 # 重复以增加文本长度
|
||
|
||
# 执行轻量扫描
|
||
scanner = LightScanner()
|
||
result = scanner.scan(test_content, test_text)
|
||
|
||
# 验证结果
|
||
assert len(result.summary) > 0, "❌ 摘要为空"
|
||
assert len(result.key_points) == 5, f"❌ 关键词数量错误: {len(result.key_points)}"
|
||
assert 0 <= result.quality_metrics.overall_score <= 1, f"❌ 质量评分越界: {result.quality_metrics.overall_score}"
|
||
assert result.status == ContentStatus.SCANNED, "❌ 状态未更新"
|
||
|
||
print(f"✅ 摘要: {result.summary}")
|
||
print(f"✅ 关键词: {result.key_points}")
|
||
print(f"✅ 质量评分: {result.quality_metrics.overall_score:.2f}/1.0")
|
||
print("✅ C4.1 轻量扫描测试通过!")
|
||
|
||
return result
|
||
|
||
|
||
def test_deep_analyzer():
|
||
"""测试C4.2: 深度分析"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C4.2: 深度分析")
|
||
print("="*60)
|
||
|
||
ai_client = AIClient()
|
||
analyzer = DeepAnalyzer(ai_client=ai_client)
|
||
|
||
# 测试短视频（<10分钟）
|
||
print("\n📱 测试短视频分析...")
|
||
short_content = Content(
|
||
id="test-short",
|
||
url="https://example.com/short",
|
||
title="短视频测试",
|
||
content_type=ContentType.VIDEO,
|
||
duration=300 # 5分钟
|
||
)
|
||
short_text = "这是一个短视频的内容文本。" * 100
|
||
|
||
result_short = analyzer.analyze(short_content, short_text)
|
||
assert len(result_short.timeline) > 0, "❌ 短视频时间轴为空"
|
||
print(f"✅ 短视频分析完成，时间轴段数: {len(result_short.timeline)}")
|
||
|
||
# 测试长视频（>=10分钟）
|
||
print("\n🎬 测试长视频分析...")
|
||
long_content = Content(
|
||
id="test-long",
|
||
url="https://example.com/long",
|
||
title="长视频测试",
|
||
content_type=ContentType.VIDEO,
|
||
duration=3600 # 60分钟
|
||
)
|
||
# 创建长视频SRT字幕（60分钟）
|
||
long_text = ""
|
||
for i in range(120): # 120段，每段30秒 = 60分钟
|
||
start_sec = i * 30
|
||
end_sec = (i + 1) * 30
|
||
start_time = f"00:{start_sec//60:02d}:{start_sec%60:02d},000"
|
||
end_time = f"00:{end_sec//60:02d}:{end_sec%60:02d},000"
|
||
long_text += f"{i+1}\n{start_time} --> {end_time}\n这是第{i+1}段讨论内容关于人工智能和机器学习的应用场景\n\n"
|
||
|
||
result_long = analyzer.analyze(long_content, long_text)
|
||
assert len(result_long.timeline) > 0, "❌ 长视频时间轴为空"
|
||
assert len(result_long.main_topics) > 0, "❌ 长视频主题为空"
|
||
print(f"✅ 长视频分析完成，时间轴段数: {len(result_long.timeline)}")
|
||
print(f"✅ 主要主题: {result_long.main_topics}")
|
||
|
||
print("✅ C4.2 深度分析测试通过!")
|
||
|
||
return result_short, result_long
|
||
|
||
|
||
def test_pipeline():
|
||
"""测试C4.3: 处理管道"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C4.3: 处理管道")
|
||
print("="*60)
|
||
|
||
ai_client = AIClient()
|
||
pipeline = ProcessingPipeline(ai_client=ai_client)
|
||
|
||
# 创建测试内容
|
||
content = Content(
|
||
id="test-pipeline",
|
||
url="https://example.com/pipeline",
|
||
title="管道测试视频",
|
||
content_type=ContentType.VIDEO,
|
||
source="bilibili",
|
||
duration=600,
|
||
author="测试作者"
|
||
)
|
||
|
||
test_text = """
|
||
这是用于测试处理管道的内容。
|
||
包含多个主题：编程、AI、数据分析。
|
||
让我们看看管道能否正常工作。
|
||
""" * 50
|
||
|
||
# 执行完整流程
|
||
result = pipeline.process_full(content, test_text)
|
||
|
||
# 验证结果
|
||
assert result.status == ContentStatus.GENERATED, f"❌ 最终状态错误: {result.status}"
|
||
assert len(result.summary) > 0, "❌ 摘要为空"
|
||
assert len(result.key_points) == 5, "❌ 关键词数量错误"
|
||
assert result.output_path and len(result.output_path) > 0, "❌ 笔记路径为空"
|
||
|
||
print(f"✅ 处理完成，最终状态: {result.status.value}")
|
||
print(f"✅ 笔记路径: {result.output_path}")
|
||
print("✅ C4.3 处理管道测试通过!")
|
||
|
||
return result
|
||
|
||
|
||
def test_content_chunker():
|
||
"""测试C5.1: 内容分片"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C5.1: 内容分片器")
|
||
print("="*60)
|
||
|
||
# 创建测试SRT字幕
|
||
srt_text = ""
|
||
for i in range(100):
|
||
start_min = i * 0.5
|
||
end_min = (i + 1) * 0.5
|
||
start_time = f"00:{int(start_min):02d}:00,000"
|
||
end_time = f"00:{int(end_min):02d}:00,000"
|
||
srt_text += f"{i+1}\n{start_time} --> {end_time}\n这是第{i+1}段字幕内容\n\n"
|
||
|
||
chunker = ContentChunker()
|
||
|
||
# 测试按时长分片（15分钟）
|
||
print("\n⏱️ 测试按时长分片（15分钟）...")
|
||
chunks_by_duration = chunker.chunk_by_duration(srt_text, target_duration=900)
|
||
assert len(chunks_by_duration) > 0, "❌ 按时长分片失败"
|
||
print(f"✅ 按时长分片: {len(chunks_by_duration)} 段")
|
||
for chunk in chunks_by_duration[:3]: # 只显示前3段
|
||
print(f" 段{chunk.index}: {chunk.time_range}, {len(chunk.text)}字符")
|
||
|
||
# 测试按段数分片
|
||
print("\n🔢 测试按段数分片（4段）...")
|
||
chunks_by_count = chunker.chunk_by_count(srt_text, num_chunks=4)
|
||
assert len(chunks_by_count) > 0, "❌ 按段数分片失败"
|
||
print(f"✅ 按段数分片: {len(chunks_by_count)} 段")
|
||
|
||
print("✅ C5.1 内容分片器测试通过!")
|
||
|
||
return chunks_by_duration, chunks_by_count
|
||
|
||
|
||
def test_chunking_strategy():
|
||
"""测试C5.2: 分段策略"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C5.2: 分段策略")
|
||
print("="*60)
|
||
|
||
test_cases = [
|
||
(300, 'direct', False), # 5分钟 - 直接处理
|
||
(900, 'truncate', False), # 15分钟 - 截断
|
||
(1800, 'chunk_2', True), # 30分钟 - 2段
|
||
(3600, 'chunk_n', True), # 60分钟 - 多段
|
||
(7200, 'chunk_n', True), # 120分钟 - 多段
|
||
]
|
||
|
||
for duration, expected_strategy, expected_chunking in test_cases:
|
||
strategy = get_chunking_strategy(duration)
|
||
|
||
mins = duration // 60
|
||
print(f"\n⏱️ 测试 {mins}分钟视频:")
|
||
print(f" 策略: {strategy['strategy']} (预期: {expected_strategy})")
|
||
print(f" 是否分片: {strategy['use_chunking']} (预期: {expected_chunking})")
|
||
|
||
assert strategy['strategy'] == expected_strategy, f"❌ 策略不匹配"
|
||
assert strategy['use_chunking'] == expected_chunking, f"❌ 分片标志不匹配"
|
||
|
||
print("\n✅ C5.2 分段策略测试通过!")
|
||
|
||
|
||
def test_ai_chunk_analysis():
|
||
"""测试C5.3: AI分段分析"""
|
||
print("\n" + "="*60)
|
||
print("🧪 测试C5.3: AI分段分析")
|
||
print("="*60)
|
||
|
||
ai_client = AIClient()
|
||
|
||
# 创建长视频SRT（45分钟）
|
||
srt_text = ""
|
||
for i in range(90): # 90段，每段30秒 = 45分钟
|
||
start_sec = i * 30
|
||
end_sec = (i + 1) * 30
|
||
start_time = f"00:{start_sec//60:02d}:{start_sec%60:02d},000"
|
||
end_time = f"00:{end_sec//60:02d}:{end_sec%60:02d},000"
|
||
srt_text += f"{i+1}\n{start_time} --> {end_time}\n第{i+1}段讨论内容关于人工智能和机器学习的应用场景\n\n"
|
||
|
||
duration = 2700 # 45分钟
|
||
|
||
print(f"\n🎬 测试长视频（{duration//60}分钟）分段分析...")
|
||
|
||
# 执行分段分析
|
||
result = ai_client._deep_analyze(
|
||
content_text=srt_text,
|
||
content_type="video",
|
||
duration=duration
|
||
)
|
||
|
||
# 验证结果
|
||
assert 'summary' in result, "❌ 缺少整体摘要"
|
||
assert 'topics' in result, "❌ 缺少主题列表"
|
||
assert 'timeline' in result, "❌ 缺少时间轴"
|
||
assert len(result['timeline']) > 0, "❌ 时间轴为空"
|
||
|
||
print(f"✅ 分析完成")
|
||
print(f"✅ 整体摘要: {result['summary'][:80]}...")
|
||
print(f"✅ 主题数: {len(result['topics'])}")
|
||
print(f"✅ 时间轴段数: {len(result['timeline'])}")
|
||
|
||
print("\n✅ C5.3 AI分段分析测试通过!")
|
||
|
||
return result
|
||
|
||
|
||
def run_all_tests():
|
||
"""运行所有测试"""
|
||
print("\n" + "🚀"*30)
|
||
print("开始执行 Obsidian KB Checkpoint 4-5 测试")
|
||
print("🚀"*30)
|
||
|
||
results = {}
|
||
|
||
try:
|
||
# C4.1 轻量扫描
|
||
results['light_scan'] = test_light_scanner()
|
||
|
||
# C4.2 深度分析
|
||
results['deep_analyze_short'], results['deep_analyze_long'] = test_deep_analyzer()
|
||
|
||
# C4.3 处理管道
|
||
results['pipeline'] = test_pipeline()
|
||
|
||
# C5.1 内容分片
|
||
results['chunks_duration'], results['chunks_count'] = test_content_chunker()
|
||
|
||
# C5.2 分段策略
|
||
test_chunking_strategy()
|
||
|
||
# C5.3 AI分段分析
|
||
results['ai_chunk_analysis'] = test_ai_chunk_analysis()
|
||
|
||
# 总结
|
||
print("\n" + "="*60)
|
||
print("✅ 所有测试通过!")
|
||
print("="*60)
|
||
print("\n📋 验收标准检查:")
|
||
print(" [✅] 轻量扫描能生成摘要和评分")
|
||
print(" [✅] 深度分析能处理长短视频")
|
||
print(" [✅] 长视频(30分钟+)自动分段不报错")
|
||
print(" [✅] content_chunker 能正确分割SRT字幕")
|
||
|
||
return True
|
||
|
||
except Exception as e:
|
||
print(f"\n❌ 测试失败: {e}")
|
||
import traceback
|
||
traceback.print_exc()
|
||
return False
|
||
|
||
|
||
if __name__ == "__main__":
|
||
success = run_all_tests()
|
||
sys.exit(0 if success else 1)
|