crawlo 1.2.8__py3-none-any.whl → 1.2.9__py3-none-any.whl

crawlo/core/processor.py

@@ -1,40 +1,40 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from asyncio import Queue
-from typing import Union, Optional
-
-from crawlo import Request, Item
-from crawlo.pipelines.pipeline_manager import PipelineManager
-
-
-class Processor(object):
-
-    def __init__(self, crawler):
-        self.crawler = crawler
-        self.queue: Queue = Queue()
-        self.pipelines: Optional[PipelineManager] = None
-
-    def open(self):
-        self.pipelines = PipelineManager.from_crawler(self.crawler)
-
-    async def process(self):
-        while not self.idle():
-            result = await self.queue.get()
-            if isinstance(result, Request):
-                await self.crawler.engine.enqueue_request(result)
-            else:
-                assert isinstance(result, Item)
-                await self._process_item(result)
-
-    async def _process_item(self, item):
-        await self.pipelines.process_item(item=item)
-
-    async def enqueue(self, output: Union[Request, Item]):
-        await self.queue.put(output)
-        await self.process()
-
-    def idle(self) -> bool:
-        return len(self) == 0
-
-    def __len__(self):
-        return self.queue.qsize()
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from asyncio import Queue
+from typing import Union, Optional
+
+from crawlo import Request, Item
+from crawlo.pipelines.pipeline_manager import PipelineManager
+
+
+class Processor(object):
+
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.queue: Queue = Queue()
+        self.pipelines: Optional[PipelineManager] = None
+
+    def open(self):
+        self.pipelines = PipelineManager.from_crawler(self.crawler)
+
+    async def process(self):
+        while not self.idle():
+            result = await self.queue.get()
+            if isinstance(result, Request):
+                await self.crawler.engine.enqueue_request(result)
+            else:
+                assert isinstance(result, Item)
+                await self._process_item(result)
+
+    async def _process_item(self, item):
+        await self.pipelines.process_item(item=item)
+
+    async def enqueue(self, output: Union[Request, Item]):
+        await self.queue.put(output)
+        await self.process()
+
+    def idle(self) -> bool:
+        return len(self) == 0
+
+    def __len__(self):
+        return self.queue.qsize()

crawlo/core/scheduler.py

@@ -1,252 +1,257 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from typing import Optional, Callable
-import traceback
-
-from crawlo.utils.log import get_logger
-from crawlo.utils.request import set_request
-from crawlo.utils.request_serializer import RequestSerializer
-from crawlo.utils.error_handler import ErrorHandler
-from crawlo.queue.queue_manager import QueueManager, QueueConfig, QueueType
-from crawlo.project import load_class, common_call
-
-
-class Scheduler:
-    def __init__(self, crawler, dupe_filter, stats, log_level, priority):
-        self.crawler = crawler
-        self.queue_manager: Optional[QueueManager] = None
-        self.request_serializer = RequestSerializer()  # 专门处理序列化
-
-        self.logger = get_logger(name=self.__class__.__name__, level=log_level)
-        self.error_handler = ErrorHandler(self.__class__.__name__, log_level)
-        self.stats = stats
-        self.dupe_filter = dupe_filter
-        self.priority = priority
-
-    @classmethod
-    def create_instance(cls, crawler):
-        filter_cls = load_class(crawler.settings.get('FILTER_CLASS'))
-        o = cls(
-            crawler=crawler,
-            dupe_filter=filter_cls.create_instance(crawler),
-            stats=crawler.stats,
-            log_level=crawler.settings.get('LOG_LEVEL'),
-            priority=crawler.settings.get('DEPTH_PRIORITY')
-        )
-        return o
-
-    async def open(self):
-        """初始化调度器和队列"""
-        self.logger.debug("开始初始化调度器...")
-        try:
-            # 创建队列配置
-            queue_config = QueueConfig.from_settings(self.crawler.settings)
-            
-            # 创建队列管理器
-            self.queue_manager = QueueManager(queue_config)
-            
-            # 初始化队列
-            needs_config_update = await self.queue_manager.initialize()
-            
-            # 检查是否需要更新过滤器配置
-            updated_configs = []
-            if needs_config_update:
-                # 如果返回True，说明队列类型发生了变化，需要检查当前队列类型来决定更新方向
-                if self.queue_manager._queue_type == QueueType.REDIS:
-                    self._update_filter_config_for_redis()
-                    updated_configs.append("Redis")
-                else:
-                    self._update_filter_config_if_needed()
-                    updated_configs.append("内存")
-            else:
-                # 检查是否需要更新配置（即使队列管理器没有要求更新）
-                if self.queue_manager._queue_type == QueueType.REDIS:
-                    # 检查当前过滤器是否为内存过滤器
-                    current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
-                    if 'memory_filter' in current_filter_class:
-                        self._update_filter_config_for_redis()
-                        updated_configs.append("Redis")
-                elif self.queue_manager._queue_type == QueueType.MEMORY:
-                    # 检查当前过滤器是否为Redis过滤器
-                    current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
-                    if 'aioredis_filter' in current_filter_class or 'redis_filter' in current_filter_class:
-                        self._update_filter_config_if_needed()
-                        updated_configs.append("内存")
-            
-            # 只有在确实需要更新配置时才重新创建过滤器实例
-            # 检查是否真的进行了配置更新
-            filter_updated = (
-                (self.queue_manager._queue_type == QueueType.REDIS and 'memory_filter' in self.crawler.settings.get('FILTER_CLASS', '')) or
-                (self.queue_manager._queue_type == QueueType.MEMORY and ('aioredis_filter' in self.crawler.settings.get('FILTER_CLASS', '') or 'redis_filter' in self.crawler.settings.get('FILTER_CLASS', '')))
-            )
-            
-            if needs_config_update or filter_updated:
-                # 重新创建过滤器实例，确保使用更新后的配置
-                filter_cls = load_class(self.crawler.settings.get('FILTER_CLASS'))
-                self.dupe_filter = filter_cls.create_instance(self.crawler)
-                
-                # 如果配置被更新，记录警告信息
-                if filter_updated:
-                    original_mode = "分布式" if 'memory_filter' in self.crawler.settings.get('FILTER_CLASS', '') else "单机"
-                    new_mode = "单机" if self.queue_manager._queue_type == QueueType.MEMORY else "分布式"
-                    self.logger.warning(f"⚠️ 检测到运行模式不一致，已从{original_mode}模式自动切换到{new_mode}模式")
-            
-            # 输出关键的调度器初始化完成信息
-            status = self.queue_manager.get_status()
-            current_filter = self.crawler.settings.get('FILTER_CLASS')
-            
-            # 优化日志输出，将多条日志合并为1条关键信息
-            # 显示完整路径的过滤器类名
-            if self.crawler.settings.get('QUEUE_TYPE', 'memory') == 'auto' and updated_configs:
-                concurrency = self.crawler.settings.get('CONCURRENCY', 8)
-                delay = self.crawler.settings.get('DOWNLOAD_DELAY', 1.0)
-                self.logger.info(f"✅ 调度器初始化完成 [队列类型: {status['type']}, 状态: {status['health']}, 并发数: {concurrency}, 延迟: {delay}s, 过滤器: {current_filter}]")
-            else:
-                self.logger.info(f"✅ 调度器初始化完成 [队列类型: {status['type']}, 状态: {status['health']}, 过滤器: {current_filter}]")
-        except Exception as e:
-            self.logger.error(f"❌ 调度器初始化失败: {e}")
-            self.logger.debug(f"详细错误信息:\n{traceback.format_exc()}")
-            raise
-    
-    def _update_filter_config_if_needed(self):
-        """如果队列类型切换到内存模式，则更新过滤器配置"""
-        if self.queue_manager and self.queue_manager._queue_type == QueueType.MEMORY:
-            # 检查当前过滤器是否为Redis过滤器
-            current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
-            updated_configs = []
-            
-            if 'aioredis_filter' in current_filter_class or 'redis_filter' in current_filter_class:
-                # 更新为内存过滤器
-                self.crawler.settings.set('FILTER_CLASS', 'crawlo.filters.memory_filter.MemoryFilter')
-                updated_configs.append("过滤器")
-            
-            # 检查当前去重管道是否为Redis去重管道
-            current_dedup_pipeline = self.crawler.settings.get('DEFAULT_DEDUP_PIPELINE', '')
-            if 'redis_dedup_pipeline' in current_dedup_pipeline:
-                # 更新为内存去重管道
-                self.crawler.settings.set('DEFAULT_DEDUP_PIPELINE', 'crawlo.pipelines.memory_dedup_pipeline.MemoryDedupPipeline')
-                # 同时更新PIPELINES列表中的去重管道
-                pipelines = self.crawler.settings.get('PIPELINES', [])
-                if current_dedup_pipeline in pipelines:
-                    # 找到并替换Redis去重管道为内存去重管道
-                    index = pipelines.index(current_dedup_pipeline)
-                    pipelines[index] = 'crawlo.pipelines.memory_dedup_pipeline.MemoryDedupPipeline'
-                    self.crawler.settings.set('PIPELINES', pipelines)
-                updated_configs.append("去重管道")
-            
-            # 合并日志输出
-            if updated_configs:
-                self.logger.info(f"✅ 已更新{', '.join(updated_configs)}配置为内存模式")
-
-    def _update_filter_config_for_redis(self):
-        """如果队列类型是Redis，则更新过滤器配置为Redis实现"""
-        if self.queue_manager and self.queue_manager._queue_type == QueueType.REDIS:
-            # 检查当前过滤器是否为内存过滤器
-            current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
-            updated_configs = []
-            
-            if 'memory_filter' in current_filter_class:
-                # 更新为Redis过滤器
-                self.crawler.settings.set('FILTER_CLASS', 'crawlo.filters.aioredis_filter.AioRedisFilter')
-                updated_configs.append("过滤器")
-            
-            # 检查当前去重管道是否为内存去重管道
-            current_dedup_pipeline = self.crawler.settings.get('DEFAULT_DEDUP_PIPELINE', '')
-            if 'memory_dedup_pipeline' in current_dedup_pipeline:
-                # 更新为Redis去重管道
-                self.crawler.settings.set('DEFAULT_DEDUP_PIPELINE', 'crawlo.pipelines.redis_dedup_pipeline.RedisDedupPipeline')
-                # 同时更新PIPELINES列表中的去重管道
-                pipelines = self.crawler.settings.get('PIPELINES', [])
-                if current_dedup_pipeline in pipelines:
-                    # 找到并替换内存去重管道为Redis去重管道
-                    index = pipelines.index(current_dedup_pipeline)
-                    pipelines[index] = 'crawlo.pipelines.redis_dedup_pipeline.RedisDedupPipeline'
-                    self.crawler.settings.set('PIPELINES', pipelines)
-                updated_configs.append("去重管道")
-            
-            # 合并日志输出
-            if updated_configs:
-                self.logger.info(f"✅ 已更新{', '.join(updated_configs)}配置为Redis模式")
-
-    async def next_request(self):
-        """获取下一个请求"""
-        if not self.queue_manager:
-            return None
-            
-        try:
-            request = await self.queue_manager.get()
-            
-            # 恢复 callback（从 Redis 队列取出时）
-            if request:
-                spider = getattr(self.crawler, 'spider', None)
-                request = self.request_serializer.restore_after_deserialization(request, spider)
-            
-            return request
-        except Exception as e:
-            self.error_handler.handle_error(
-                e, 
-                context="获取下一个请求失败", 
-                raise_error=False
-            )
-            return None
-
-    async def enqueue_request(self, request):
-        """将请求加入队列"""
-        if not request.dont_filter and await common_call(self.dupe_filter.requested, request):
-            self.dupe_filter.log_stats(request)
-            return False
-
-        if not self.queue_manager:
-            self.logger.error("队列管理器未初始化")
-            return False
-
-        set_request(request, self.priority)
-        
-        try:
-            # 使用统一的队列接口
-            success = await self.queue_manager.put(request, priority=getattr(request, 'priority', 0))
-            
-            if success:
-                self.logger.debug(f"✅ 请求入队成功: {request.url}")
-            
-            return success
-        except Exception as e:
-            self.error_handler.handle_error(
-                e, 
-                context="请求入队失败", 
-                raise_error=False
-            )
-            return False
-
-    def idle(self) -> bool:
-        """检查队列是否为空"""
-        return len(self) == 0
-
-    async def async_idle(self) -> bool:
-        """异步检查队列是否为空（更精确）"""
-        if not self.queue_manager:
-            return True
-        # 使用队列管理器的异步empty方法
-        return await self.queue_manager.async_empty()
-
-    async def close(self):
-        """关闭调度器"""
-        try:
-            if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
-                await closed()
-            
-            if self.queue_manager:
-                await self.queue_manager.close()
-        except Exception as e:
-            self.error_handler.handle_error(
-                e, 
-                context="关闭调度器失败", 
-                raise_error=False
-            )
-
-    def __len__(self):
-        """获取队列大小"""
-        if not self.queue_manager:
-            return 0
-        # 返回同步的近似值，实际大小需要异步获取
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from typing import Optional, Callable
+import traceback
+
+from crawlo.utils.log import get_logger
+from crawlo.utils.request import set_request
+from crawlo.utils.request_serializer import RequestSerializer
+from crawlo.utils.error_handler import ErrorHandler
+from crawlo.queue.queue_manager import QueueManager, QueueConfig, QueueType
+from crawlo.project import load_class, common_call
+
+
+class Scheduler:
+    def __init__(self, crawler, dupe_filter, stats, log_level, priority):
+        self.crawler = crawler
+        self.queue_manager: Optional[QueueManager] = None
+        self.request_serializer = RequestSerializer()
+
+        self.logger = get_logger(name=self.__class__.__name__, level=log_level)
+        self.error_handler = ErrorHandler(self.__class__.__name__, log_level)
+        self.stats = stats
+        self.dupe_filter = dupe_filter
+        self.priority = priority
+
+    @classmethod
+    def create_instance(cls, crawler):
+        filter_cls = load_class(crawler.settings.get('FILTER_CLASS'))
+        o = cls(
+            crawler=crawler,
+            dupe_filter=filter_cls.create_instance(crawler),
+            stats=crawler.stats,
+            log_level=crawler.settings.get('LOG_LEVEL'),
+            priority=crawler.settings.get('DEPTH_PRIORITY')
+        )
+        return o
+
+    async def open(self):
+        """Initialize scheduler and queue"""
+        self.logger.debug("开始初始化调度器...")
+        try:
+            # 创建队列配置
+            queue_config = QueueConfig.from_settings(self.crawler.settings)
+            
+            # 创建队列管理器
+            self.queue_manager = QueueManager(queue_config)
+            
+            # 初始化队列
+            needs_config_update = await self.queue_manager.initialize()
+            
+            # 检查是否需要更新过滤器配置
+            updated_configs = []
+            if needs_config_update:
+                # 如果返回True，说明队列类型发生了变化，需要检查当前队列类型来决定更新方向
+                if self.queue_manager._queue_type == QueueType.REDIS:
+                    self._update_filter_config_for_redis()
+                    updated_configs.append("Redis")
+                else:
+                    self._update_filter_config_if_needed()
+                    updated_configs.append("内存")
+            else:
+                # 检查是否需要更新配置（即使队列管理器没有要求更新）
+                if self.queue_manager._queue_type == QueueType.REDIS:
+                    # 检查当前过滤器是否为内存过滤器
+                    current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
+                    if 'memory_filter' in current_filter_class:
+                        self._update_filter_config_for_redis()
+                        updated_configs.append("Redis")
+                elif self.queue_manager._queue_type == QueueType.MEMORY:
+                    # 检查当前过滤器是否为Redis过滤器
+                    current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
+                    if 'aioredis_filter' in current_filter_class or 'redis_filter' in current_filter_class:
+                        self._update_filter_config_if_needed()
+                        updated_configs.append("内存")
+            
+            # 只有在确实需要更新配置时才重新创建过滤器实例
+            # 检查是否真的进行了配置更新
+            filter_updated = (
+                (self.queue_manager._queue_type == QueueType.REDIS and 'memory_filter' in self.crawler.settings.get('FILTER_CLASS', '')) or
+                (self.queue_manager._queue_type == QueueType.MEMORY and ('aioredis_filter' in self.crawler.settings.get('FILTER_CLASS', '') or 'redis_filter' in self.crawler.settings.get('FILTER_CLASS', '')))
+            )
+            
+            if needs_config_update or filter_updated:
+                # 重新创建过滤器实例，确保使用更新后的配置
+                filter_cls = load_class(self.crawler.settings.get('FILTER_CLASS'))
+                self.dupe_filter = filter_cls.create_instance(self.crawler)
+                
+                # 如果配置被更新，记录警告信息
+                if filter_updated:
+                    original_mode = "distributed" if 'memory_filter' in self.crawler.settings.get('FILTER_CLASS', '') else "standalone"
+                    new_mode = "standalone" if self.queue_manager._queue_type == QueueType.MEMORY else "distributed"
+                    self.logger.warning(f"runtime mode inconsistency detected: switched from {original_mode} to {new_mode} mode")
+            
+            # 输出关键的调度器初始化完成信息
+            status = self.queue_manager.get_status()
+            current_filter = self.crawler.settings.get('FILTER_CLASS')
+            
+            # 输出启用的过滤器信息（类似Downloader的格式）
+            self.logger.info(f"enabled filters: \n  {current_filter}")  # 注释掉重复的日志
+            
+            # 优化日志输出，将多条日志合并为1条关键信息
+            # 显示完整路径的过滤器类名
+            if self.crawler.settings.get('QUEUE_TYPE', 'memory') == 'auto' and updated_configs:
+                concurrency = self.crawler.settings.get('CONCURRENCY', 8)
+                delay = self.crawler.settings.get('DOWNLOAD_DELAY', 1.0)
+                # Change INFO level log to DEBUG level to avoid redundant output
+                self.logger.debug(f"Scheduler initialized [Queue type: {status['type']}, Status: {status['health']}, Concurrency: {concurrency}, Delay: {delay}s]")
+            else:
+                # Change INFO level log to DEBUG level to avoid redundant output
+                self.logger.debug(f"Scheduler initialized [Queue type: {status['type']}, Status: {status['health']}]")
+        except Exception as e:
+            self.logger.error(f"Scheduler initialization failed: {e}")
+            self.logger.debug(f"Detailed error information:\n{traceback.format_exc()}")
+            raise
+    
+    def _update_filter_config_if_needed(self):
+        """Update filter configuration if queue type switches to memory mode"""
+        if self.queue_manager and self.queue_manager._queue_type == QueueType.MEMORY:
+            # 检查当前过滤器是否为Redis过滤器
+            current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
+            updated_configs = []
+            
+            if 'aioredis_filter' in current_filter_class or 'redis_filter' in current_filter_class:
+                # 更新为内存过滤器
+                self.crawler.settings.set('FILTER_CLASS', 'crawlo.filters.memory_filter.MemoryFilter')
+                updated_configs.append("filter")
+            
+            # 检查当前去重管道是否为Redis去重管道
+            current_dedup_pipeline = self.crawler.settings.get('DEFAULT_DEDUP_PIPELINE', '')
+            if 'redis_dedup_pipeline' in current_dedup_pipeline:
+                # 更新为内存去重管道
+                self.crawler.settings.set('DEFAULT_DEDUP_PIPELINE', 'crawlo.pipelines.memory_dedup_pipeline.MemoryDedupPipeline')
+                # 同时更新PIPELINES列表中的去重管道
+                pipelines = self.crawler.settings.get('PIPELINES', [])
+                if current_dedup_pipeline in pipelines:
+                    # 找到并替换Redis去重管道为内存去重管道
+                    index = pipelines.index(current_dedup_pipeline)
+                    pipelines[index] = 'crawlo.pipelines.memory_dedup_pipeline.MemoryDedupPipeline'
+                    self.crawler.settings.set('PIPELINES', pipelines)
+                updated_configs.append("dedup pipeline")
+            
+            # 合并日志输出
+            if updated_configs:
+                self.logger.debug(f"configuration updated: {', '.join(updated_configs)} -> memory mode")
+
+    def _update_filter_config_for_redis(self):
+        """Update filter configuration to Redis implementation if queue type is Redis"""
+        if self.queue_manager and self.queue_manager._queue_type == QueueType.REDIS:
+            # 检查当前过滤器是否为内存过滤器
+            current_filter_class = self.crawler.settings.get('FILTER_CLASS', '')
+            updated_configs = []
+            
+            if 'memory_filter' in current_filter_class:
+                # 更新为Redis过滤器
+                self.crawler.settings.set('FILTER_CLASS', 'crawlo.filters.aioredis_filter.AioRedisFilter')
+                updated_configs.append("filter")
+            
+            # 检查当前去重管道是否为内存去重管道
+            current_dedup_pipeline = self.crawler.settings.get('DEFAULT_DEDUP_PIPELINE', '')
+            if 'memory_dedup_pipeline' in current_dedup_pipeline:
+                # 更新为Redis去重管道
+                self.crawler.settings.set('DEFAULT_DEDUP_PIPELINE', 'crawlo.pipelines.redis_dedup_pipeline.RedisDedupPipeline')
+                # 同时更新PIPELINES列表中的去重管道
+                pipelines = self.crawler.settings.get('PIPELINES', [])
+                if current_dedup_pipeline in pipelines:
+                    # 找到并替换内存去重管道为Redis去重管道
+                    index = pipelines.index(current_dedup_pipeline)
+                    pipelines[index] = 'crawlo.pipelines.redis_dedup_pipeline.RedisDedupPipeline'
+                    self.crawler.settings.set('PIPELINES', pipelines)
+                updated_configs.append("dedup pipeline")
+            
+            # 合并日志输出
+            if updated_configs:
+                self.logger.info(f"configuration updated: {', '.join(updated_configs)} -> redis mode")
+
+    async def next_request(self):
+        """Get next request"""
+        if not self.queue_manager:
+            return None
+            
+        try:
+            request = await self.queue_manager.get()
+            
+            # 恢复 callback（从 Redis 队列取出时）
+            if request:
+                spider = getattr(self.crawler, 'spider', None)
+                request = self.request_serializer.restore_after_deserialization(request, spider)
+            
+            return request
+        except Exception as e:
+            self.error_handler.handle_error(
+                e, 
+                context="Failed to get next request", 
+                raise_error=False
+            )
+            return None
+
+    async def enqueue_request(self, request):
+        """Add request to queue"""
+        if not request.dont_filter and await common_call(self.dupe_filter.requested, request):
+            self.dupe_filter.log_stats(request)
+            return False
+
+        if not self.queue_manager:
+            self.logger.error("Queue manager not initialized")
+            return False
+
+        set_request(request, self.priority)
+        
+        try:
+            # 使用统一的队列接口
+            success = await self.queue_manager.put(request, priority=getattr(request, 'priority', 0))
+            
+            if success:
+                self.logger.debug(f"Request enqueued successfully: {request.url}")
+            
+            return success
+        except Exception as e:
+            self.error_handler.handle_error(
+                e, 
+                context="Failed to enqueue request", 
+                raise_error=False
+            )
+            return False
+
+    def idle(self) -> bool:
+        """Check if queue is empty"""
+        return len(self) == 0
+
+    async def async_idle(self) -> bool:
+        """Asynchronously check if queue is empty (more accurate)"""
+        if not self.queue_manager:
+            return True
+        # 使用队列管理器的异步empty方法
+        return await self.queue_manager.async_empty()
+
+    async def close(self):
+        """Close scheduler"""
+        try:
+            if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
+                await closed()
+            
+            if self.queue_manager:
+                await self.queue_manager.close()
+        except Exception as e:
+            self.error_handler.handle_error(
+                e, 
+                context="Failed to close scheduler", 
+                raise_error=False
+            )
+
+    def __len__(self):
+        """Get queue size"""
+        if not self.queue_manager:
+            return 0
+        # 返回同步的近似值，实际大小需要异步获取
         return 0 if self.queue_manager.empty() else 1