crawlo 1.1.2__py3-none-any.whl → 1.1.3__py3-none-any.whl

crawlo/core/scheduler.py

@@ -1,162 +1,166 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from typing import Optional, Callable
-
-from crawlo.utils.log import get_logger
-from crawlo.utils.request import set_request
-from crawlo.utils.request_serializer import RequestSerializer
-from crawlo.queue.queue_manager import QueueManager, QueueConfig
-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.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"""
-        # 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
-        success = await self.queue_manager.initialize()
-        if not success:
-            raise RuntimeError("队列初始化失败")
-        
-        # 输出队列状态
-        status = self.queue_manager.get_status()
-        self.logger.info(f'队列类型: {status["type"]}, 状态: {status["health"]}')
-        self.logger.info(f'requesting filter: {self.dupe_filter}')
-
-    async def next_request(self):
-        """获取下一个请求"""
-        if not self.queue_manager:
-            return None
-            
-        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
-
-    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)
-        
-        # 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
-
-    def idle(self) -> bool:
-        """检查队列是否为空"""
-        return len(self) == 0
-
-    async def close(self):
-        """关闭调度器"""
-        if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
-            await closed()
-        
-        if self.queue_manager:
-            await self.queue_manager.close()
-
-    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
-        return 0 if self.queue_manager.empty() else 1
-
-# #!/usr/bin/python
-# # -*- coding:UTF-8 -*-
-# from typing import Optional, Callable
-#
-# from crawlo.utils.log import get_logger
-# from crawlo.utils.request import set_request
-# from crawlo.utils.pqueue import SpiderPriorityQueue
-# from crawlo.project import load_class, common_call
-#
-#
-# class Scheduler:
-#     def __init__(self, crawler, dupe_filter, stats, log_level, priority):
-#         self.crawler = crawler
-#         self.request_queue: Optional[SpiderPriorityQueue] = None
-#
-#         self.logger = get_logger(name=self.__class__.__name__, level=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
-#
-#     def open(self):
-#         self.request_queue = SpiderPriorityQueue()
-#         self.logger.info(f'requesting filter: {self.dupe_filter}')
-#
-#     async def next_request(self):
-#         request = await self.request_queue.get()
-#         return request
-#
-#     async def enqueue_request(self, request):
-#         if not request.dont_filter and await common_call(self.dupe_filter.requested, request):
-#             self.dupe_filter.log_stats(request)
-#             return False
-#         set_request(request, self.priority)
-#         await self.request_queue.put(request)
-#         return True
-#
-#     def idle(self) -> bool:
-#         return len(self) == 0
-#
-#     async def close(self):
-#         if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
-#             await closed()
-#
-#     def __len__(self):
-#         return self.request_queue.qsize()
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from typing import Optional, Callable
+
+from crawlo.utils.log import get_logger
+from crawlo.utils.request import set_request
+from crawlo.utils.request_serializer import RequestSerializer
+from crawlo.queue.queue_manager import QueueManager, QueueConfig
+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.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"""
+        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
+            success = await self.queue_manager.initialize()
+            if not success:
+                raise RuntimeError("队列初始化失败")
+            
+            # 输出队列状态
+            status = self.queue_manager.get_status()
+            self.logger.info(f'队列类型: {status["type"]}, 状态: {status["health"]}')
+            self.logger.info(f'requesting filter: {self.dupe_filter}')
+        except Exception as e:
+            self.logger.error(f"❌ 调度器初始化失败: {e}")
+            raise
+
+    async def next_request(self):
+        """获取下一个请求"""
+        if not self.queue_manager:
+            return None
+            
+        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
+
+    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)
+        
+        # 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
+
+    def idle(self) -> bool:
+        """检查队列是否为空"""
+        return len(self) == 0
+
+    async def close(self):
+        """关闭调度器"""
+        if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
+            await closed()
+        
+        if self.queue_manager:
+            await self.queue_manager.close()
+
+    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
+        return 0 if self.queue_manager.empty() else 1
+
+# #!/usr/bin/python
+# # -*- coding:UTF-8 -*-
+# from typing import Optional, Callable
+#
+# from crawlo.utils.log import get_logger
+# from crawlo.utils.request import set_request
+# from crawlo.utils.pqueue import SpiderPriorityQueue
+# from crawlo.project import load_class, common_call
+#
+#
+# class Scheduler:
+#     def __init__(self, crawler, dupe_filter, stats, log_level, priority):
+#         self.crawler = crawler
+#         self.request_queue: Optional[SpiderPriorityQueue] = None
+#
+#         self.logger = get_logger(name=self.__class__.__name__, level=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
+#
+#     def open(self):
+#         self.request_queue = SpiderPriorityQueue()
+#         self.logger.info(f'requesting filter: {self.dupe_filter}')
+#
+#     async def next_request(self):
+#         request = await self.request_queue.get()
+#         return request
+#
+#     async def enqueue_request(self, request):
+#         if not request.dont_filter and await common_call(self.dupe_filter.requested, request):
+#             self.dupe_filter.log_stats(request)
+#             return False
+#         set_request(request, self.priority)
+#         await self.request_queue.put(request)
+#         return True
+#
+#     def idle(self) -> bool:
+#         return len(self) == 0
+#
+#     async def close(self):
+#         if isinstance(closed := getattr(self.dupe_filter, 'closed', None), Callable):
+#             await closed()
+#
+#     def __len__(self):
+#         return self.request_queue.qsize()