crawlo 1.0.1__py3-none-any.whl → 1.0.3__py3-none-any.whl

crawlo/extension/log_stats.py

@@ -1,44 +1,44 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from crawlo import event
-from crawlo.utils.date_tools import get_current_time, time_diff_seconds
-
-
-class LogStats(object):
-
-    def __init__(self, stats):
-        self._stats = stats
-
-    @classmethod
-    def create_instance(cls, crawler):
-        o = cls(crawler.stats)
-        crawler.subscriber.subscribe(o.spider_opened, event=event.spider_opened)
-        crawler.subscriber.subscribe(o.spider_closed, event=event.spider_closed)
-        crawler.subscriber.subscribe(o.item_successful, event=event.item_successful)
-        crawler.subscriber.subscribe(o.item_discard, event=event.item_discard)
-        crawler.subscriber.subscribe(o.response_received, event=event.response_received)
-        crawler.subscriber.subscribe(o.request_scheduled, event=event.request_scheduled)
-
-        return o
-
-    async def spider_opened(self):
-        self._stats['start_time'] = get_current_time(fmt='%Y-%m-%d %H:%M:%S')
-
-    async def spider_closed(self):
-        self._stats['end_time'] = get_current_time(fmt='%Y-%m-%d %H:%M:%S')
-        self._stats['cost_time(s)'] = time_diff_seconds(start_time=self._stats['start_time'], end_time=self._stats['end_time'])
-
-    async def item_successful(self, _item, _spider):
-        self._stats.inc_value('item_successful_count')
-
-    async def item_discard(self, _item, exc, _spider):
-        self._stats.inc_value('item_discard_count')
-        reason = exc.msg
-        if reason:
-            self._stats.inc_value(f"item_discard/{reason}")
-
-    async def response_received(self, _response, _spider):
-        self._stats.inc_value('response_received_count')
-
-    async def request_scheduled(self, _request, _spider):
-        self._stats.inc_value('request_scheduler_count')
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from crawlo import event
+from crawlo.utils.date_tools import get_current_time, time_diff_seconds
+
+
+class LogStats(object):
+
+    def __init__(self, stats):
+        self._stats = stats
+
+    @classmethod
+    def create_instance(cls, crawler):
+        o = cls(crawler.stats)
+        crawler.subscriber.subscribe(o.spider_opened, event=event.spider_opened)
+        crawler.subscriber.subscribe(o.spider_closed, event=event.spider_closed)
+        crawler.subscriber.subscribe(o.item_successful, event=event.item_successful)
+        crawler.subscriber.subscribe(o.item_discard, event=event.item_discard)
+        crawler.subscriber.subscribe(o.response_received, event=event.response_received)
+        crawler.subscriber.subscribe(o.request_scheduled, event=event.request_scheduled)
+
+        return o
+
+    async def spider_opened(self):
+        self._stats['start_time'] = get_current_time(fmt='%Y-%m-%d %H:%M:%S')
+
+    async def spider_closed(self):
+        self._stats['end_time'] = get_current_time(fmt='%Y-%m-%d %H:%M:%S')
+        self._stats['cost_time(s)'] = time_diff_seconds(start_time=self._stats['start_time'], end_time=self._stats['end_time'])
+
+    async def item_successful(self, _item, _spider):
+        self._stats.inc_value('item_successful_count')
+
+    async def item_discard(self, _item, exc, _spider):
+        self._stats.inc_value('item_discard_count')
+        reason = exc.msg
+        if reason:
+            self._stats.inc_value(f"item_discard/{reason}")
+
+    async def response_received(self, _response, _spider):
+        self._stats.inc_value('response_received_count')
+
+    async def request_scheduled(self, _request, _spider):
+        self._stats.inc_value('request_scheduler_count')

crawlo/filters/__init__.py

@@ -1,37 +1,37 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from abc import ABC, abstractmethod
-
-from crawlo import Request
-from crawlo.utils.request import request_fingerprint
-
-
-class BaseFilter(ABC):
-
-    def __init__(self, logger, stats, debug: bool):
-        self.logger = logger
-        self.stats = stats
-        self.debug = debug
-
-    @classmethod
-    def create_instance(cls, *args, **kwargs) -> 'BaseFilter':
-        return cls(*args, **kwargs)
-
-    def requested(self, request: Request):
-        fp = request_fingerprint(request)
-        if fp in self:
-            return True
-        self.add_fingerprint(fp)
-        return False
-
-    @abstractmethod
-    def add_fingerprint(self, fp) -> None:
-        pass
-
-    def log_stats(self, request: Request) -> None:
-        if self.debug:
-            self.logger.debug(f'Filtered duplicate request: {request}')
-        self.stats.inc_value(f'{self}/filtered_count')
-
-    def __str__(self) -> str:
-        return f'{self.__class__.__name__}'
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from abc import ABC, abstractmethod
+
+from crawlo import Request
+from crawlo.utils.request import request_fingerprint
+
+
+class BaseFilter(ABC):
+
+    def __init__(self, logger, stats, debug: bool):
+        self.logger = logger
+        self.stats = stats
+        self.debug = debug
+
+    @classmethod
+    def create_instance(cls, *args, **kwargs) -> 'BaseFilter':
+        return cls(*args, **kwargs)
+
+    def requested(self, request: Request):
+        fp = request_fingerprint(request)
+        if fp in self:
+            return True
+        self.add_fingerprint(fp)
+        return False
+
+    @abstractmethod
+    def add_fingerprint(self, fp) -> None:
+        pass
+
+    def log_stats(self, request: Request) -> None:
+        if self.debug:
+            self.logger.debug(f'Filtered duplicate request: {request}')
+        self.stats.inc_value(f'{self}/filtered_count')
+
+    def __str__(self) -> str:
+        return f'{self.__class__.__name__}'

crawlo/filters/aioredis_filter.py

@@ -1,130 +1,158 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-from typing import Optional
-
-import aioredis
-
-from crawlo import Request
-from crawlo.filters import BaseFilter
-from crawlo.utils.log import get_logger
-from crawlo.utils.request import request_fingerprint
-
-
-class AioRedisFilter(BaseFilter):
-    """使用Redis集合实现的异步请求去重过滤器（适用于分布式爬虫）"""
-
-    def __init__(
-            self,
-            redis_key: str,
-            client: aioredis.Redis,
-            stats: dict,
-            debug: bool,
-            log_level: str,
-            cleanup_fp: bool = False
-    ):
-        """
-        初始化过滤器
-
-        参数说明:
-            redis_key: Redis中存储指纹的键名
-            client: aioredis客户端实例
-            stats: 统计信息字典
-            debug: 是否启用调试模式
-            log_level: 日志级别
-            save_fp: 爬虫关闭时是否保留指纹数据
-        """
-        # 初始化日志记录器（使用类名作为日志标识）
-        self.logger = get_logger(self.__class__.__name__, log_level)
-        super().__init__(self.logger, stats, debug)
-
-        self.redis_key = redis_key  # Redis存储键（如："project:request_fingerprints"）
-        self.redis = client  # Redis异步客户端
-        self.cleanup_fp = cleanup_fp  # 是否持久化指纹数据
-
-    @classmethod
-    def create_instance(cls, crawler) -> 'BaseFilter':
-        """从爬虫配置创建过滤器实例（工厂方法）"""
-        # 从配置获取Redis连接参数（带默认值）
-        redis_url = crawler.settings.get('REDIS_URL', 'redis://localhost:6379')
-        decode_responses = crawler.settings.get_bool('DECODE_RESPONSES', True)
-
-        try:
-            # 创建Redis连接池（限制最大连接数20）
-            redis_client = aioredis.from_url(
-                redis_url,
-                decode_responses=decode_responses,
-                max_connections=20
-            )
-        except Exception as e:
-            raise RuntimeError(f"Redis连接失败 {redis_url}: {str(e)}")
-
-        # 使用项目名+配置键组合作为Redis键
-        return cls(
-            redis_key=f"{crawler.settings.get('PROJECT_NAME')}:{crawler.settings.get('REDIS_KEY', 'request_fingerprints')}",
-            client=redis_client,
-            stats=crawler.stats,
-            cleanup_fp=crawler.settings.get_bool('CLEANUP_FP', False),
-            debug=crawler.settings.get_bool('FILTER_DEBUG', False),
-            log_level=crawler.settings.get('LOG_LEVEL', 'INFO')
-        )
-
-    async def requested(self, request: Request) -> bool:
-        """
-        检查请求是否重复
-
-        参数:
-            request: 要检查的请求对象
-
-        返回:
-            bool: True表示重复请求，False表示新请求
-        """
-        fp = request_fingerprint(request)  # 生成请求指纹
-        try:
-            # 检查指纹是否已存在集合中
-            is_duplicate = await self.redis.sismember(self.redis_key, fp)
-            if is_duplicate:
-                # self.logger.debug(f"发现重复请求: {fp}")
-                return True
-
-            # 新请求则添加指纹
-            await self.add_fingerprint(fp)
-            return False
-        except aioredis.RedisError as e:
-            self.logger.error(f"Redis操作失败: {str(e)}")
-            raise  # 向上抛出异常
-
-    async def add_fingerprint(self, fp: str) -> None:
-        """向Redis集合添加新指纹"""
-        try:
-            await self.redis.sadd(self.redis_key, fp)
-            self.logger.debug(f"新增指纹: {fp}")
-        except aioredis.RedisError as e:
-            self.logger.error(f"指纹添加失败: {str(e)}")
-            raise
-
-    async def closed(self, reason: Optional[str] = None) -> None:
-        """
-        爬虫关闭时的处理（兼容Scrapy的关闭逻辑）
-
-        参数:
-            reason: 爬虫关闭原因（Scrapy标准参数）
-        """
-        if self.cleanup_fp:  # 仅在配置明确要求时清理
-            try:
-                deleted = await self.redis.delete(self.redis_key)
-                self.logger.info(
-                    f"Cleaned {deleted} fingerprints from {self.redis_key} "
-                    f"(reason: {reason or 'manual'})"
-                )
-            except aioredis.RedisError as e:
-                self.logger.warning(f"Cleanup failed: {e}")
-            finally:
-                await self._close_redis()
-
-    async def _close_redis(self) -> None:
-        """安全关闭Redis连接"""
-        try:
-            await self.redis.close()
-            await self.redis.connection_pool.disconnect()
-        except Exception as e:
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+from typing import Optional
+import aioredis
+from crawlo import Request
+from crawlo.filters import BaseFilter
+from crawlo.utils.log import get_logger
+from crawlo.utils.request import request_fingerprint
+
+
+class AioRedisFilter(BaseFilter):
+    """使用Redis集合实现的异步请求去重过滤器（适用于分布式爬虫）"""
+
+    def __init__(
+            self,
+            redis_key: str,
+            client: aioredis.Redis,
+            stats: dict,
+            debug: bool,
+            log_level: str,
+            cleanup_fp: bool = False
+    ):
+        """初始化过滤器"""
+        self.logger = get_logger(self.__class__.__name__, log_level)
+        super().__init__(self.logger, stats, debug)
+
+        self.redis_key = redis_key
+        self.redis = client
+        self.cleanup_fp = cleanup_fp
+
+    @classmethod
+    def create_instance(cls, crawler) -> 'BaseFilter':
+        """从爬虫配置创建过滤器实例"""
+        redis_url = crawler.settings.get('REDIS_URL', 'redis://localhost:6379')
+        decode_responses = crawler.settings.get_bool('DECODE_RESPONSES', False)  # 关键：改为False
+
+        try:
+            redis_client = aioredis.from_url(
+                redis_url,
+                decode_responses=decode_responses,
+                max_connections=20,
+                encoding='utf-8'
+            )
+        except Exception as e:
+            raise RuntimeError(f"Redis连接失败 {redis_url}: {str(e)}")
+
+        return cls(
+            redis_key=f"{crawler.settings.get('PROJECT_NAME', 'default')}:{crawler.settings.get('REDIS_KEY', 'request_fingerprints')}",
+            client=redis_client,
+            stats=crawler.stats,
+            cleanup_fp=crawler.settings.get_bool('CLEANUP_FP', False),
+            debug=crawler.settings.get_bool('FILTER_DEBUG', False),
+            log_level=crawler.settings.get('LOG_LEVEL', 'INFO')
+        )
+
+    async def requested(self, request: Request) -> bool:
+        """
+        检查请求是否重复
+        """
+        try:
+            fp = request_fingerprint(request)
+            self.logger.debug(f"Checking fingerprint: {fp}")
+
+            # 确保fp是字符串类型
+            if not isinstance(fp, str):
+                fp = str(fp)
+
+            # 检查Redis连接状态
+            if not self.redis:
+                raise RuntimeError("Redis client is not initialized")
+
+            # 检查指纹是否已存在
+            is_member = await self.redis.sismember(self.redis_key, fp)
+            self.logger.debug(f"Fingerprint {fp} exists: {is_member}")
+
+            if is_member:
+                if self.debug:
+                    self.logger.debug(f"Filtered duplicate request: {fp}")
+                return True
+
+            # 添加新指纹
+            result = await self.redis.sadd(self.redis_key, fp)
+
+            if self.debug:
+                if result == 1:
+                    self.logger.debug(f"Added new fingerprint: {fp}")
+                else:
+                    self.logger.warning(f"Failed to add fingerprint: {fp}")
+
+            return False
+
+        except Exception as e:
+            self.logger.error(f"Filter check failed for {getattr(request, 'url', 'unknown')}: {str(e)}")
+            # 可以选择抛出异常或返回False（不过滤）
+            raise
+
+    async def add_fingerprint(self, fp: str) -> bool:
+        """向Redis集合添加新指纹"""
+        try:
+            if not isinstance(fp, str):
+                fp = str(fp)
+
+            result = await self.redis.sadd(self.redis_key, fp)
+            if self.debug:
+                self.logger.debug(f"Added fingerprint {fp}, result: {result}")
+            return result == 1
+        except Exception as e:
+            self.logger.error(f"Failed to add fingerprint {fp}: {str(e)}")
+            raise
+
+    async def get_stats(self) -> dict:
+        """获取当前过滤器统计信息"""
+        try:
+            count = await self.redis.scard(self.redis_key)
+            return {
+                'total_fingerprints': count,
+                'redis_key': self.redis_key,
+                **self.stats
+            }
+        except Exception as e:
+            self.logger.error(f"Failed to get stats: {str(e)}")
+            return self.stats
+
+    async def clear_all(self) -> int:
+        """清空所有指纹数据"""
+        try:
+            deleted = await self.redis.delete(self.redis_key)
+            self.logger.info(f"Cleared {deleted} keys")
+            return deleted
+        except Exception as e:
+            self.logger.error(f"Failed to clear fingerprints: {str(e)}")
+            raise
+
+    async def closed(self, reason: Optional[str] = None) -> None:
+        """爬虫关闭时的处理"""
+        try:
+            if self.cleanup_fp:
+                deleted = await self.redis.delete(self.redis_key)
+                self.logger.info(
+                    f"Cleaned {deleted} fingerprints from {self.redis_key} "
+                    f"(reason: {reason or 'manual'})"
+                )
+            else:
+                # 显示统计信息
+                count = await self.redis.scard(self.redis_key)
+                self.logger.info(f"Total fingerprints preserved: {count}")
+        except Exception as e:
+            self.logger.warning(f"Close operation failed: {e}")
+        finally:
+            await self._close_redis()
+
+    async def _close_redis(self) -> None:
+        """安全关闭Redis连接"""
+        try:
+            if hasattr(self.redis, 'close'):
+                await self.redis.close()
+        except Exception as e:
             self.logger.warning(f"Redis close error: {e}")