crawlo 1.1.3__py3-none-any.whl → 1.1.4__py3-none-any.whl

crawlo/pipelines/memory_dedup_pipeline.py

@@ -1,116 +1,116 @@
-#!/usr/bin/python
-# -*- coding:UTF-8 -*-
-"""
-基于内存的数据项去重管道
-======================
-提供单节点环境下的数据项去重功能，防止保存重复的数据记录。
-
-特点:
-- 高性能: 使用内存集合进行快速查找
-- 简单易用: 无需外部依赖
-- 轻量级: 适用于小规模数据采集
-- 低延迟: 内存操作无网络开销
-"""
-
-import hashlib
-from typing import Dict, Any, Set
-
-from crawlo import Item
-from crawlo.spider import Spider
-from crawlo.utils.log import get_logger
-from crawlo.exceptions import DropItem
-
-
-class MemoryDedupPipeline:
-    """基于内存的数据项去重管道"""
-
-    def __init__(self, log_level: str = "INFO"):
-        """
-        初始化内存去重管道
-        
-        :param log_level: 日志级别
-        """
-        self.logger = get_logger(self.__class__.__name__, log_level)
-        
-        # 使用集合存储已见过的数据项指纹
-        self.seen_items: Set[str] = set()
-        self.dropped_count = 0
-        
-        self.logger.info("内存去重管道初始化完成")
-
-    @classmethod
-    def from_crawler(cls, crawler):
-        """从爬虫配置创建管道实例"""
-        settings = crawler.settings
-        
-        return cls(
-            log_level=settings.get('LOG_LEVEL', 'INFO')
-        )
-
-    def process_item(self, item: Item, spider: Spider) -> Item:
-        """
-        处理数据项，进行去重检查
-        
-        :param item: 要处理的数据项
-        :param spider: 爬虫实例
-        :return: 处理后的数据项或抛出 DropItem 异常
-        """
-        try:
-            # 生成数据项指纹
-            fingerprint = self._generate_item_fingerprint(item)
-            
-            # 检查指纹是否已存在
-            if fingerprint in self.seen_items:
-                # 如果已存在，丢弃这个数据项
-                self.dropped_count += 1
-                self.logger.debug(f"丢弃重复数据项: {fingerprint[:20]}...")
-                raise DropItem(f"重复的数据项: {fingerprint}")
-            else:
-                # 记录新数据项的指纹
-                self.seen_items.add(fingerprint)
-                self.logger.debug(f"处理新数据项: {fingerprint[:20]}...")
-                return item
-                
-        except Exception as e:
-            self.logger.error(f"处理数据项时出错: {e}")
-            # 在错误时继续处理，避免丢失数据
-            return item
-
-    def _generate_item_fingerprint(self, item: Item) -> str:
-        """
-        生成数据项指纹
-        
-        基于数据项的所有字段生成唯一指纹，用于去重判断。
-        
-        :param item: 数据项
-        :return: 指纹字符串
-        """
-        # 将数据项转换为可序列化的字典
-        try:
-            item_dict = item.to_dict()
-        except AttributeError:
-            # 兼容没有to_dict方法的Item实现
-            item_dict = dict(item)
-        
-        # 对字典进行排序以确保一致性
-        sorted_items = sorted(item_dict.items())
-        
-        # 生成指纹字符串
-        fingerprint_string = '|'.join([f"{k}={v}" for k, v in sorted_items if v is not None])
-        
-        # 使用 SHA256 生成固定长度的指纹
-        return hashlib.sha256(fingerprint_string.encode('utf-8')).hexdigest()
-
-    def close_spider(self, spider: Spider) -> None:
-        """
-        爬虫关闭时的清理工作
-        
-        :param spider: 爬虫实例
-        """
-        self.logger.info(f"爬虫 {spider.name} 关闭:")
-        self.logger.info(f"  - 丢弃的重复数据项: {self.dropped_count}")
-        self.logger.info(f"  - 内存中存储的指纹数: {len(self.seen_items)}")
-        
-        # 清理内存
-        self.seen_items.clear()
+#!/usr/bin/python
+# -*- coding:UTF-8 -*-
+"""
+基于内存的数据项去重管道
+======================
+提供单节点环境下的数据项去重功能，防止保存重复的数据记录。
+
+特点:
+- 高性能: 使用内存集合进行快速查找
+- 简单易用: 无需外部依赖
+- 轻量级: 适用于小规模数据采集
+- 低延迟: 内存操作无网络开销
+"""
+
+import hashlib
+from typing import Dict, Any, Set
+
+from crawlo import Item
+from crawlo.spider import Spider
+from crawlo.utils.log import get_logger
+from crawlo.exceptions import DropItem
+
+
+class MemoryDedupPipeline:
+    """基于内存的数据项去重管道"""
+
+    def __init__(self, log_level: str = "INFO"):
+        """
+        初始化内存去重管道
+        
+        :param log_level: 日志级别
+        """
+        self.logger = get_logger(self.__class__.__name__, log_level)
+        
+        # 使用集合存储已见过的数据项指纹
+        self.seen_items: Set[str] = set()
+        self.dropped_count = 0
+        
+        self.logger.info("内存去重管道初始化完成")
+
+    @classmethod
+    def from_crawler(cls, crawler):
+        """从爬虫配置创建管道实例"""
+        settings = crawler.settings
+        
+        return cls(
+            log_level=settings.get('LOG_LEVEL', 'INFO')
+        )
+
+    def process_item(self, item: Item, spider: Spider) -> Item:
+        """
+        处理数据项，进行去重检查
+        
+        :param item: 要处理的数据项
+        :param spider: 爬虫实例
+        :return: 处理后的数据项或抛出 DropItem 异常
+        """
+        try:
+            # 生成数据项指纹
+            fingerprint = self._generate_item_fingerprint(item)
+            
+            # 检查指纹是否已存在
+            if fingerprint in self.seen_items:
+                # 如果已存在，丢弃这个数据项
+                self.dropped_count += 1
+                self.logger.debug(f"丢弃重复数据项: {fingerprint[:20]}...")
+                raise DropItem(f"重复的数据项: {fingerprint}")
+            else:
+                # 记录新数据项的指纹
+                self.seen_items.add(fingerprint)
+                self.logger.debug(f"处理新数据项: {fingerprint[:20]}...")
+                return item
+                
+        except Exception as e:
+            self.logger.error(f"处理数据项时出错: {e}")
+            # 在错误时继续处理，避免丢失数据
+            return item
+
+    def _generate_item_fingerprint(self, item: Item) -> str:
+        """
+        生成数据项指纹
+        
+        基于数据项的所有字段生成唯一指纹，用于去重判断。
+        
+        :param item: 数据项
+        :return: 指纹字符串
+        """
+        # 将数据项转换为可序列化的字典
+        try:
+            item_dict = item.to_dict()
+        except AttributeError:
+            # 兼容没有to_dict方法的Item实现
+            item_dict = dict(item)
+        
+        # 对字典进行排序以确保一致性
+        sorted_items = sorted(item_dict.items())
+        
+        # 生成指纹字符串
+        fingerprint_string = '|'.join([f"{k}={v}" for k, v in sorted_items if v is not None])
+        
+        # 使用 SHA256 生成固定长度的指纹
+        return hashlib.sha256(fingerprint_string.encode('utf-8')).hexdigest()
+
+    def close_spider(self, spider: Spider) -> None:
+        """
+        爬虫关闭时的清理工作
+        
+        :param spider: 爬虫实例
+        """
+        self.logger.info(f"爬虫 {spider.name} 关闭:")
+        self.logger.info(f"  - 丢弃的重复数据项: {self.dropped_count}")
+        self.logger.info(f"  - 内存中存储的指纹数: {len(self.seen_items)}")
+        
+        # 清理内存
+        self.seen_items.clear()
         self.dropped_count = 0

crawlo/pipelines/mongo_pipeline.py

@@ -1,117 +1,132 @@
-# -*- coding: utf-8 -*-
-from typing import Optional
-from motor.motor_asyncio import AsyncIOMotorClient
-from pymongo.errors import PyMongoError
-from crawlo.utils.log import get_logger
-from crawlo.exceptions import ItemDiscard
-
-
-class MongoPipeline:
-    def __init__(self, crawler):
-        self.crawler = crawler
-        self.settings = crawler.settings
-        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
-
-        # 初始化连接参数
-        self.client = None
-        self.db = None
-        self.collection = None
-
-        # 配置默认值
-        self.mongo_uri = self.settings.get('MONGO_URI', 'mongodb://localhost:27017')
-        self.db_name = self.settings.get('MONGO_DATABASE', 'scrapy_db')
-        self.collection_name = self.settings.get('MONGO_COLLECTION', crawler.spider.name)
-
-        # 注册关闭事件
-        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
-
-    @classmethod
-    def from_crawler(cls, crawler):
-        return cls(crawler)
-
-    async def _ensure_connection(self):
-        """确保连接已建立"""
-        if self.client is None:
-            self.client = AsyncIOMotorClient(self.mongo_uri)
-            self.db = self.client[self.db_name]
-            self.collection = self.db[self.collection_name]
-            self.logger.info(f"MongoDB连接建立 (集合: {self.collection_name})")
-
-    async def process_item(self, item, spider) -> Optional[dict]:
-        """处理item的核心方法"""
-        try:
-            await self._ensure_connection()
-
-            item_dict = dict(item)
-            result = await self.collection.insert_one(item_dict)
-
-            # 统计计数
-            self.crawler.stats.inc_value('mongodb/inserted')
-            self.logger.debug(f"插入文档ID: {result.inserted_id}")
-
-            return item
-
-        except Exception as e:
-            self.crawler.stats.inc_value('mongodb/failed')
-            self.logger.error(f"MongoDB插入失败: {e}")
-            raise ItemDiscard(f"MongoDB操作失败: {e}")
-
-    async def spider_closed(self):
-        """关闭爬虫时清理资源"""
-        if self.client:
-            self.client.close()
-            self.logger.info("MongoDB连接已关闭")
-
-
-class MongoPoolPipeline:
-    def __init__(self, crawler):
-        self.crawler = crawler
-        self.settings = crawler.settings
-        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
-
-        # 连接池配置
-        self.client = AsyncIOMotorClient(
-            self.settings.get('MONGO_URI', 'mongodb://localhost:27017'),
-            maxPoolSize=self.settings.getint('MONGO_MAX_POOL_SIZE', 100),
-            minPoolSize=self.settings.getint('MONGO_MIN_POOL_SIZE', 10),
-            connectTimeoutMS=5000,
-            socketTimeoutMS=30000
-        )
-
-        self.db = self.client[self.settings.get('MONGO_DATABASE', 'scrapy_db')]
-        self.collection = self.db[self.settings.get('MONGO_COLLECTION', crawler.spider.name)]
-
-        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
-        self.logger.info(f"MongoDB连接池已初始化 (集合: {self.collection.name})")
-
-    @classmethod
-    def create_instance(cls, crawler):
-        return cls(crawler)
-
-    async def process_item(self, item, spider) -> Optional[dict]:
-        """处理item方法（带重试机制）"""
-        try:
-            item_dict = dict(item)
-
-            # 带重试的插入操作
-            for attempt in range(3):
-                try:
-                    result = await self.collection.insert_one(item_dict)
-                    self.crawler.stats.inc_value('mongodb/insert_success')
-                    self.logger.debug(f"插入成功 [attempt {attempt + 1}]: {result.inserted_id}")
-                    return item
-                except PyMongoError as e:
-                    if attempt == 2:  # 最后一次尝试仍失败
-                        raise
-                    self.logger.warning(f"插入重试中 [attempt {attempt + 1}]: {e}")
-
-        except Exception as e:
-            self.crawler.stats.inc_value('mongodb/insert_failed')
-            self.logger.error(f"MongoDB操作最终失败: {e}")
-            raise ItemDiscard(f"MongoDB操作失败: {e}")
-
-    async def spider_closed(self):
-        """资源清理"""
-        if hasattr(self, 'client'):
-            self.client.close()
-            self.logger.info("MongoDB连接池已释放")
+# -*- coding: utf-8 -*-
+from typing import Optional, List, Dict
+from motor.motor_asyncio import AsyncIOMotorClient
+from pymongo.errors import PyMongoError
+from crawlo.utils.log import get_logger
+from crawlo.exceptions import ItemDiscard
+
+
+class MongoPipeline:
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+
+        # 初始化连接参数
+        self.client = None
+        self.db = None
+        self.collection = None
+
+        # 配置默认值
+        self.mongo_uri = self.settings.get('MONGO_URI', 'mongodb://localhost:27017')
+        self.db_name = self.settings.get('MONGO_DATABASE', 'scrapy_db')
+        self.collection_name = self.settings.get('MONGO_COLLECTION', crawler.spider.name)
+        
+        # 连接池配置
+        self.max_pool_size = self.settings.getint('MONGO_MAX_POOL_SIZE', 100)
+        self.min_pool_size = self.settings.getint('MONGO_MIN_POOL_SIZE', 10)
+        self.connect_timeout_ms = self.settings.getint('MONGO_CONNECT_TIMEOUT_MS', 5000)
+        self.socket_timeout_ms = self.settings.getint('MONGO_SOCKET_TIMEOUT_MS', 30000)
+
+        # 批量插入配置
+        self.batch_size = self.settings.getint('MONGO_BATCH_SIZE', 100)
+        self.use_batch = self.settings.getbool('MONGO_USE_BATCH', False)
+        self.batch_buffer: List[Dict] = []  # 批量缓冲区
+
+        # 注册关闭事件
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+
+    async def _ensure_connection(self):
+        """确保连接已建立"""
+        if self.client is None:
+            # 使用连接池配置创建客户端
+            self.client = AsyncIOMotorClient(
+                self.mongo_uri,
+                maxPoolSize=self.max_pool_size,
+                minPoolSize=self.min_pool_size,
+                connectTimeoutMS=self.connect_timeout_ms,
+                socketTimeoutMS=self.socket_timeout_ms
+            )
+            self.db = self.client[self.db_name]
+            self.collection = self.db[self.collection_name]
+            self.logger.info(f"MongoDB连接建立 (集合: {self.collection_name})")
+
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item的核心方法（带重试机制）"""
+        # 如果启用批量插入，将item添加到缓冲区
+        if self.use_batch:
+            self.batch_buffer.append(dict(item))
+            
+            # 如果缓冲区达到批量大小，执行批量插入
+            if len(self.batch_buffer) >= self.batch_size:
+                await self._flush_batch(spider)
+                
+            return item
+        else:
+            # 单条插入逻辑
+            try:
+                await self._ensure_connection()
+
+                item_dict = dict(item)
+
+                # 带重试的插入操作
+                for attempt in range(3):
+                    try:
+                        result = await self.collection.insert_one(item_dict)
+                        # 统一使用insert_success统计键名
+                        self.crawler.stats.inc_value('mongodb/insert_success')
+                        self.logger.debug(f"插入成功 [attempt {attempt + 1}]: {result.inserted_id}")
+                        return item
+                    except PyMongoError as e:
+                        if attempt == 2:  # 最后一次尝试仍失败
+                            raise
+                        self.logger.warning(f"插入重试中 [attempt {attempt + 1}]: {e}")
+
+            except Exception as e:
+                # 统一使用insert_failed统计键名
+                self.crawler.stats.inc_value('mongodb/insert_failed')
+                self.logger.error(f"MongoDB操作最终失败: {e}")
+                raise ItemDiscard(f"MongoDB操作失败: {e}")
+
+    async def _flush_batch(self, spider):
+        """刷新批量缓冲区并执行批量插入"""
+        if not self.batch_buffer:
+            return
+
+        try:
+            await self._ensure_connection()
+
+            # 带重试的批量插入操作
+            for attempt in range(3):
+                try:
+                    result = await self.collection.insert_many(self.batch_buffer, ordered=False)
+                    # 统一使用insert_success统计键名
+                    inserted_count = len(result.inserted_ids)
+                    self.crawler.stats.inc_value('mongodb/insert_success', inserted_count)
+                    self.logger.debug(f"批量插入成功 [attempt {attempt + 1}]: {inserted_count} 条记录")
+                    self.batch_buffer.clear()
+                    return
+                except PyMongoError as e:
+                    if attempt == 2:  # 最后一次尝试仍失败
+                        raise
+                    self.logger.warning(f"批量插入重试中 [attempt {attempt + 1}]: {e}")
+        except Exception as e:
+            # 统一使用insert_failed统计键名
+            failed_count = len(self.batch_buffer)
+            self.crawler.stats.inc_value('mongodb/insert_failed', failed_count)
+            self.logger.error(f"MongoDB批量插入最终失败: {e}")
+            raise ItemDiscard(f"MongoDB批量插入失败: {e}")
+
+    async def spider_closed(self):
+        """关闭爬虫时清理资源"""
+        # 在关闭前刷新剩余的批量数据
+        if self.use_batch and self.batch_buffer:
+            await self._flush_batch(self.crawler.spider)
+            
+        if self.client:
+            self.client.close()
+            self.logger.info("MongoDB连接已关闭")