crawlo 1.3.3__py3-none-any.whl → 1.3.5__py3-none-any.whl

tests/performance_comparison.py

@@ -0,0 +1,246 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+Crawlo vs Scrapy 性能对比测试
+"""
+import asyncio
+import time
+import subprocess
+import sys
+import os
+import re
+
+# 添加项目根目录到Python路径
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))
+
+def run_crawlo_test():
+    """运行Crawlo性能测试"""
+    print("开始运行Crawlo性能测试...")
+    start_time = time.time()
+    
+    # 运行Crawlo测试
+    try:
+        # 使用较小的页数进行测试以节省时间
+        result = subprocess.run([
+            'python', '-c', '''
+import asyncio
+import sys
+import os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+
+from crawlo import Spider, Request
+from crawlo.crawler import CrawlerProcess
+from crawlo.items import Item, Field
+
+class NewsItem(Item):
+    title = Field()
+    publish_time = Field()
+    url = Field()
+    source = Field()
+    content = Field()
+
+class OfweekPerformanceSpider(Spider):
+    name = "ofweek_performance"
+    
+    def start_requests(self):
+        headers = {
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7",
+            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+            "Cache-Control": "no-cache",
+            "Connection": "keep-alive",
+            "Pragma": "no-cache",
+            "Referer": "https://ee.ofweek.com/CATList-2800-8100-ee-2.html",
+            "Sec-Fetch-Dest": "document",
+            "Sec-Fetch-Mode": "navigate",
+            "Sec-Fetch-Site": "same-origin",
+            "Sec-Fetch-User": "?1",
+            "Upgrade-Insecure-Requests": "1",
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/139.0.0.0 Safari/537.36",
+        }
+        cookies = {
+            "__utmz": "57425525.1730117117.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none)",
+            "Hm_lvt_abe9900db162c6d089cdbfd107db0f03": "1739244841",
+            "Hm_lvt_af50e2fc51af73da7720fb324b88a975": "1740100727",
+            "JSESSIONID": "FEA96D3B5FC31350B2285E711BF2A541",
+            "Hm_lvt_28a416fcfc17063eb9c4f9bb1a1f5cda": "1757477622",
+            "HMACCOUNT": "08DF0D235A291EAA",
+            "__utma": "57425525.2080994505.1730117117.1747970718.1757477622.50",
+            "__utmc": "57425525",
+            "__utmt": "1",
+            "__utmb": "57425525.2.10.1757477622",
+            "Hm_lpvt_28a416fcfc17063eb9c4f9bb1a1f5cda": "1757477628",
+            "index_burying_point": "c64d6c31e69d560efe319cc9f8be279f"
+        }
+
+        # 使用较少的页数进行测试
+        max_page = 10
+        for page in range(1, max_page + 1):
+            url = f"https://ee.ofweek.com/CATList-2800-8100-ee-{page}.html"
+            yield Request(url=url, callback=self.parse, headers=headers, cookies=cookies)
+
+    def parse(self, response):
+        from urllib.parse import urljoin
+        rows = response.xpath("//div[@class=\"main_left\"]/div[@class=\"list_model\"]/div[@class=\"model_right model_right2\"]")
+        
+        for row in rows:
+            try:
+                url = row.xpath("./h3/a/@href").extract_first()
+                title = row.xpath("./h3/a/text()").extract_first()
+
+                if not url or not title:
+                    continue
+
+                absolute_url = urljoin(response.url, url)
+                if not absolute_url.startswith(("http://", "https://")):
+                    continue
+
+                yield Request(
+                    url=absolute_url,
+                    meta={"title": title.strip() if title else "", "parent_url": response.url},
+                    callback=self.parse_detail
+                )
+            except Exception:
+                continue
+
+    def parse_detail(self, response):
+        title = response.meta.get("title", "")
+        content_elements = response.xpath("//div[@class=\"TRS_Editor\"]|//*[@id=\"articleC\"]")
+        if content_elements:
+            content = content_elements.xpath(".//text()").extract()
+            content = "\\n".join([text.strip() for text in content if text.strip()])
+        else:
+            content = ""
+
+        publish_time = response.xpath("//div[@class=\"time fl\"]/text()").extract_first()
+        if publish_time:
+            publish_time = publish_time.strip()
+
+        source = response.xpath("//div[@class=\"source-name\"]/text()").extract_first()
+
+        item = NewsItem()
+        item["title"] = title.strip() if title else ""
+        item["publish_time"] = publish_time if publish_time else ""
+        item["url"] = response.url
+        item["source"] = source if source else ""
+        item["content"] = content
+
+        yield item
+
+async def main():
+    process = CrawlerProcess(settings={
+        "CONCURRENCY": 8,
+        "DOWNLOAD_DELAY": 0.1,
+        "LOG_LEVEL": "ERROR",  # 减少日志输出以提高性能
+    })
+    await process.crawl(OfweekPerformanceSpider)
+
+if __name__ == "__main__":
+    asyncio.run(main())
+'''
+        ], capture_output=True, text=True, timeout=300)  # 5分钟超时
+        
+        end_time = time.time()
+        execution_time = end_time - start_time
+        
+        # 解析结果
+        output = result.stdout
+        error = result.stderr
+        
+        # 统计处理的页面数
+        pages_crawled = output.count("正在解析页面:") + output.count("正在解析详情页:")
+        
+        print(f"Crawlo执行时间: {execution_time:.2f}秒")
+        print(f"Crawlo处理页面数: {pages_crawled}")
+        print(f"Crawlo平均速度: {pages_crawled/execution_time:.2f} 页面/秒")
+        
+        return execution_time, pages_crawled
+        
+    except subprocess.TimeoutExpired:
+        print("Crawlo测试超时")
+        return None, 0
+    except Exception as e:
+        print(f"Crawlo测试出错: {e}")
+        return None, 0
+
+def run_scrapy_test():
+    """运行Scrapy性能测试"""
+    print("\n开始运行Scrapy性能测试...")
+    start_time = time.time()
+    
+    try:
+        # 运行Scrapy测试
+        result = subprocess.run([
+            'scrapy', 'runspider', 
+            '/Users/oscar/projects/Crawlo/tests/scrapy_comparison/ofweek_scrapy.py',
+            '-s', 'CONCURRENT_REQUESTS=8',
+            '-s', 'DOWNLOAD_DELAY=0.1',
+            '-s', 'LOG_LEVEL=ERROR'
+        ], capture_output=True, text=True, timeout=300, cwd='/Users/oscar/projects/Crawlo')
+        
+        end_time = time.time()
+        execution_time = end_time - start_time
+        
+        # 解析结果
+        output = result.stdout
+        error = result.stderr
+        
+        # 统计处理的页面数
+        pages_crawled = output.count("正在解析页面:") + output.count("正在解析详情页:")
+        
+        print(f"Scrapy执行时间: {execution_time:.2f}秒")
+        print(f"Scrapy处理页面数: {pages_crawled}")
+        print(f"Scrapy平均速度: {pages_crawled/execution_time:.2f} 页面/秒")
+        
+        return execution_time, pages_crawled
+        
+    except subprocess.TimeoutExpired:
+        print("Scrapy测试超时")
+        return None, 0
+    except Exception as e:
+        print(f"Scrapy测试出错: {e}")
+        return None, 0
+
+def main():
+    """主函数"""
+    print("=== Crawlo vs Scrapy 性能对比测试 ===")
+    
+    # 创建测试目录
+    os.makedirs('/Users/oscar/projects/Crawlo/tests/scrapy_comparison', exist_ok=True)
+    
+    # 运行测试
+    crawlo_time, crawlo_pages = run_crawlo_test()
+    scrapy_time, scrapy_pages = run_scrapy_test()
+    
+    # 输出对比结果
+    print("\n=== 性能对比结果 ===")
+    if crawlo_time and scrapy_time:
+        print(f"Crawlo执行时间: {crawlo_time:.2f}秒")
+        print(f"Scrapy执行时间: {scrapy_time:.2f}秒")
+        print(f"时间差异: {abs(crawlo_time - scrapy_time):.2f}秒")
+        
+        if crawlo_time < scrapy_time:
+            improvement = (scrapy_time - crawlo_time) / scrapy_time * 100
+            print(f"Crawlo比Scrapy快 {improvement:.2f}%")
+        else:
+            improvement = (crawlo_time - scrapy_time) / crawlo_time * 100
+            print(f"Scrapy比Crawlo快 {improvement:.2f}%")
+    
+    if crawlo_pages and scrapy_pages:
+        print(f"\nCrawlo处理页面数: {crawlo_pages}")
+        print(f"Scrapy处理页面数: {scrapy_pages}")
+        
+        crawlo_speed = crawlo_pages / crawlo_time if crawlo_time else 0
+        scrapy_speed = scrapy_pages / scrapy_time if scrapy_time else 0
+        
+        print(f"Crawlo平均速度: {crawlo_speed:.2f} 页面/秒")
+        print(f"Scrapy平均速度: {scrapy_speed:.2f} 页面/秒")
+        
+        if crawlo_speed > scrapy_speed:
+            speed_improvement = (crawlo_speed - scrapy_speed) / scrapy_speed * 100
+            print(f"Crawlo速度比Scrapy快 {speed_improvement:.2f}%")
+        elif scrapy_speed > crawlo_speed:
+            speed_improvement = (scrapy_speed - crawlo_speed) / crawlo_speed * 100
+            print(f"Scrapy速度比Crawlo快 {speed_improvement:.2f}%")
+
+if __name__ == '__main__':
+    main()

tests/queue_blocking_test.py

@@ -0,0 +1,114 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+队列阻塞行为测试脚本
+验证队列满时的阻塞行为
+"""
+
+import sys
+import os
+import asyncio
+import time
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))
+
+from crawlo.network.request import Request
+from crawlo.queue.queue_manager import QueueConfig, QueueManager
+
+
+async def test_queue_blocking_behavior():
+    """测试队列阻塞行为"""
+    print("开始测试队列阻塞行为...")
+    
+    # 初始化框架
+    from crawlo.initialization import initialize_framework
+    settings = initialize_framework()
+    
+    # 创建一个小型队列配置进行测试
+    settings.set('SCHEDULER_MAX_QUEUE_SIZE', 5)  # 设置非常小的队列大小
+    
+    # 创建队列配置
+    queue_config = QueueConfig.from_settings(settings)
+    queue_config.max_queue_size = 5  # 设置小队列大小进行测试
+    
+    # 创建队列管理器
+    queue_manager = QueueManager(queue_config)
+    
+    # 初始化队列
+    await queue_manager.initialize()
+    
+    print(f"队列类型: {queue_manager._queue_type}")
+    print(f"队列最大大小: {queue_config.max_queue_size}")
+    
+    # 创建生产者任务
+    async def producer(queue_manager, name, count):
+        """生产者：向队列添加请求"""
+        print(f"生产者 {name} 开始工作，将添加 {count} 个请求")
+        start_time = time.time()
+        
+        for i in range(count):
+            request = Request(url=f'https://example.com/page{name}_{i}.html')
+            try:
+                # 这里应该会阻塞当队列满时
+                await queue_manager.put(request)
+                print(f"生产者 {name} 成功添加请求 {i}")
+            except Exception as e:
+                print(f"生产者 {name} 添加请求 {i} 失败: {e}")
+        
+        end_time = time.time()
+        print(f"生产者 {name} 完成工作，耗时 {end_time - start_time:.2f} 秒")
+    
+    # 创建消费者任务
+    async def consumer(queue_manager, name, count):
+        """消费者：从队列获取请求"""
+        print(f"消费者 {name} 开始工作，将获取 {count} 个请求")
+        start_time = time.time()
+        
+        retrieved_count = 0
+        while retrieved_count < count:
+            try:
+                request = await queue_manager.get(timeout=2.0)
+                if request:
+                    print(f"消费者 {name} 成功获取请求: {request.url}")
+                    retrieved_count += 1
+                    # 模拟处理时间
+                    await asyncio.sleep(0.1)
+                else:
+                    print(f"消费者 {name} 超时，没有获取到请求")
+                    break
+            except Exception as e:
+                print(f"消费者 {name} 获取请求失败: {e}")
+                break
+        
+        end_time = time.time()
+        print(f"消费者 {name} 完成工作，获取了 {retrieved_count} 个请求，耗时 {end_time - start_time:.2f} 秒")
+    
+    # 同时运行生产者和消费者
+    print("开始并发测试...")
+    
+    # 生产者尝试添加10个请求到大小为5的队列
+    # 消费者会逐渐消费，生产者应该会被阻塞直到有空间
+    tasks = [
+        asyncio.create_task(producer(queue_manager, "P1", 10)),
+        asyncio.create_task(consumer(queue_manager, "C1", 10))
+    ]
+    
+    # 等待所有任务完成
+    await asyncio.gather(*tasks, return_exceptions=True)
+    
+    final_queue_size = await queue_manager.size()
+    print(f"队列最终大小: {final_queue_size}")
+    
+    # 关闭队列
+    await queue_manager.close()
+    
+    print("队列阻塞行为测试完成！")
+
+
+def main():
+    """主函数"""
+    print("开始队列阻塞行为测试...")
+    asyncio.run(test_queue_blocking_behavior())
+
+
+if __name__ == "__main__":
+    main()

tests/queue_test.py

@@ -0,0 +1,90 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+队列系统测试脚本
+"""
+
+import sys
+import os
+import asyncio
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))
+
+from crawlo.network.request import Request
+from crawlo.queue.queue_manager import QueueConfig, QueueManager
+
+
+async def test_queue_system():
+    """测试队列系统"""
+    print("开始测试队列系统...")
+    
+    # 初始化框架
+    from crawlo.initialization import initialize_framework
+    settings = initialize_framework()
+    
+    # 创建一个小型队列配置进行测试
+    settings.set('SCHEDULER_MAX_QUEUE_SIZE', 10)
+    
+    # 创建队列配置
+    queue_config = QueueConfig.from_settings(settings)
+    queue_config.max_queue_size = 10  # 设置小队列大小进行测试
+    
+    # 创建队列管理器
+    queue_manager = QueueManager(queue_config)
+    
+    # 初始化队列
+    await queue_manager.initialize()
+    
+    print(f"队列类型: {queue_manager._queue_type}")
+    print(f"队列最大大小: {queue_config.max_queue_size}")
+    
+    # 测试添加请求到队列
+    print("测试添加请求到队列...")
+    tasks = []
+    for i in range(15):  # 尝试添加15个请求，超过队列大小
+        request = Request(url=f'https://example.com/page{i}.html')
+        task = asyncio.create_task(queue_manager.put(request))
+        tasks.append(task)
+    
+    # 等待所有任务完成
+    results = await asyncio.gather(*tasks, return_exceptions=True)
+    
+    success_count = sum(1 for result in results if result is True)
+    print(f"成功添加 {success_count} 个请求到队列")
+    
+    queue_size = await queue_manager.size()
+    print(f"队列当前大小: {queue_size}")
+    
+    # 测试从队列获取请求
+    print("测试从队列获取请求...")
+    retrieved_count = 0
+    while retrieved_count < 15:
+        try:
+            request = await queue_manager.get(timeout=1.0)
+            if request:
+                print(f"成功获取请求: {request.url}")
+                retrieved_count += 1
+            else:
+                break
+        except Exception as e:
+            print(f"获取请求失败: {e}")
+            break
+    
+    print(f"总共获取了 {retrieved_count} 个请求")
+    
+    final_queue_size = await queue_manager.size()
+    print(f"队列最终大小: {final_queue_size}")
+    
+    # 关闭队列
+    await queue_manager.close()
+    
+    print("队列系统测试完成！")
+
+
+def main():
+    """主函数"""
+    print("开始队列系统测试...")
+    asyncio.run(test_queue_system())
+
+
+if __name__ == "__main__":
+    main()