crawlo 1.1.0__py3-none-any.whl → 1.1.2__py3-none-any.whl

crawlo/pipelines/csv_pipeline.py

@@ -0,0 +1,317 @@
+# -*- coding: utf-8 -*-
+import csv
+import asyncio
+from pathlib import Path
+from typing import Optional, List
+from datetime import datetime
+from crawlo.utils.log import get_logger
+from crawlo.exceptions import ItemDiscard
+
+
+class CsvPipeline:
+    """CSV文件输出管道"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        # 配置文件路径
+        self.file_path = self._get_file_path()
+        self.file_handle = None
+        self.csv_writer = None
+        self.headers_written = False
+        self.lock = asyncio.Lock()  # 异步锁保证线程安全
+        
+        # CSV配置
+        self.delimiter = self.settings.get('CSV_DELIMITER', ',')
+        self.quotechar = self.settings.get('CSV_QUOTECHAR', '"')
+        self.include_headers = self.settings.get_bool('CSV_INCLUDE_HEADERS', True)
+        
+        # 注册关闭事件
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        file_path = (
+            self.settings.get('CSV_FILE') or
+            getattr(self.crawler.spider, 'csv_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.csv"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    async def _ensure_file_open(self):
+        """确保文件已打开"""
+        if self.file_handle is None:
+            self.file_handle = open(self.file_path, 'w', newline='', encoding='utf-8')
+            self.csv_writer = csv.writer(
+                self.file_handle,
+                delimiter=self.delimiter,
+                quotechar=self.quotechar,
+                quoting=csv.QUOTE_MINIMAL
+            )
+            self.logger.info(f"CSV文件已创建: {self.file_path}")
+    
+    async def _write_headers(self, item_dict: dict):
+        """写入CSV表头"""
+        if not self.headers_written and self.include_headers:
+            headers = list(item_dict.keys())
+            self.csv_writer.writerow(headers)
+            self.headers_written = True
+            self.logger.debug(f"CSV表头已写入: {headers}")
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item的核心方法"""
+        try:
+            async with self.lock:
+                await self._ensure_file_open()
+                
+                # 转换为字典
+                item_dict = dict(item)
+                
+                # 写入表头（仅第一次）
+                await self._write_headers(item_dict)
+                
+                # 写入数据行
+                values = [str(v) if v is not None else '' for v in item_dict.values()]
+                self.csv_writer.writerow(values)
+                self.file_handle.flush()  # 立即刷新到磁盘
+                
+                # 统计
+                self.crawler.stats.inc_value('csv_pipeline/items_written')
+                self.logger.debug(f"写入CSV行: {len(item_dict)} 字段")
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('csv_pipeline/items_failed')
+            self.logger.error(f"CSV写入失败: {e}")
+            raise ItemDiscard(f"CSV Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """关闭爬虫时清理资源"""
+        if self.file_handle:
+            self.file_handle.close()
+            self.logger.info(f"CSV文件已关闭: {self.file_path}")
+
+
+class CsvDictPipeline:
+    """CSV字典写入器管道（使用DictWriter，支持字段映射）"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        self.file_path = self._get_file_path()
+        self.file_handle = None
+        self.csv_writer = None
+        self.fieldnames = None
+        self.headers_written = False
+        self.lock = asyncio.Lock()
+        
+        # 配置选项
+        self.delimiter = self.settings.get('CSV_DELIMITER', ',')
+        self.quotechar = self.settings.get('CSV_QUOTECHAR', '"')
+        self.include_headers = self.settings.get_bool('CSV_INCLUDE_HEADERS', True)
+        self.extrasaction = self.settings.get('CSV_EXTRASACTION', 'ignore')  # ignore, raise
+        
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        file_path = (
+            self.settings.get('CSV_DICT_FILE') or
+            getattr(self.crawler.spider, 'csv_dict_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}_dict.csv"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    def _get_fieldnames(self, item_dict: dict) -> List[str]:
+        """获取字段名列表"""
+        # 优先使用配置的字段名
+        configured_fields = self.settings.get('CSV_FIELDNAMES')
+        if configured_fields:
+            return configured_fields if isinstance(configured_fields, list) else configured_fields.split(',')
+        
+        # 使用爬虫定义的字段名
+        spider_fields = getattr(self.crawler.spider, 'csv_fieldnames', None)
+        if spider_fields:
+            return spider_fields if isinstance(spider_fields, list) else spider_fields.split(',')
+        
+        # 使用item的字段名
+        return list(item_dict.keys())
+    
+    async def _ensure_file_open(self, item_dict: dict):
+        """确保文件已打开"""
+        if self.file_handle is None:
+            self.fieldnames = self._get_fieldnames(item_dict)
+            
+            self.file_handle = open(self.file_path, 'w', newline='', encoding='utf-8')
+            self.csv_writer = csv.DictWriter(
+                self.file_handle,
+                fieldnames=self.fieldnames,
+                delimiter=self.delimiter,
+                quotechar=self.quotechar,
+                quoting=csv.QUOTE_MINIMAL,
+                extrasaction=self.extrasaction
+            )
+            
+            # 写入表头
+            if self.include_headers:
+                self.csv_writer.writeheader()
+                self.headers_written = True
+            
+            self.logger.info(f"CSV字典文件已创建: {self.file_path}，字段: {self.fieldnames}")
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item方法"""
+        try:
+            async with self.lock:
+                item_dict = dict(item)
+                await self._ensure_file_open(item_dict)
+                
+                # 写入数据行
+                self.csv_writer.writerow(item_dict)
+                self.file_handle.flush()
+                
+                self.crawler.stats.inc_value('csv_dict_pipeline/items_written')
+                self.logger.debug(f"写入CSV字典行，字段数: {len(item_dict)}")
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('csv_dict_pipeline/items_failed')
+            self.logger.error(f"CSV字典写入失败: {e}")
+            raise ItemDiscard(f"CSV Dict Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """资源清理"""
+        if self.file_handle:
+            self.file_handle.close()
+            self.logger.info(f"CSV字典文件已关闭: {self.file_path}")
+
+
+class CsvBatchPipeline:
+    """CSV批量写入管道（内存缓存，批量写入，提高性能）"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        self.file_path = self._get_file_path()
+        self.file_handle = None
+        self.csv_writer = None
+        self.batch_buffer = []
+        self.headers_written = False
+        self.lock = asyncio.Lock()
+        
+        # 批量配置
+        self.batch_size = self.settings.get_int('CSV_BATCH_SIZE', 100)
+        self.delimiter = self.settings.get('CSV_DELIMITER', ',')
+        self.quotechar = self.settings.get('CSV_QUOTECHAR', '"')
+        self.include_headers = self.settings.get_bool('CSV_INCLUDE_HEADERS', True)
+        
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        file_path = (
+            self.settings.get('CSV_BATCH_FILE') or
+            getattr(self.crawler.spider, 'csv_batch_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}_batch.csv"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    async def _ensure_file_open(self):
+        """确保文件已打开"""
+        if self.file_handle is None:
+            self.file_handle = open(self.file_path, 'w', newline='', encoding='utf-8')
+            self.csv_writer = csv.writer(
+                self.file_handle,
+                delimiter=self.delimiter,
+                quotechar=self.quotechar,
+                quoting=csv.QUOTE_MINIMAL
+            )
+            self.logger.info(f"CSV批量文件已创建: {self.file_path}")
+    
+    async def _flush_batch(self):
+        """刷新批量缓存到文件"""
+        if not self.batch_buffer:
+            return
+        
+        await self._ensure_file_open()
+        
+        for row in self.batch_buffer:
+            self.csv_writer.writerow(row)
+        
+        self.file_handle.flush()
+        items_count = len(self.batch_buffer)
+        self.batch_buffer.clear()
+        
+        self.crawler.stats.inc_value('csv_batch_pipeline/batches_written')
+        self.crawler.stats.inc_value('csv_batch_pipeline/items_written', count=items_count)
+        self.logger.info(f"批量写入 {items_count} 行到CSV文件")
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item方法"""
+        try:
+            async with self.lock:
+                item_dict = dict(item)
+                
+                # 写入表头（仅第一次）
+                if not self.headers_written and self.include_headers:
+                    headers = list(item_dict.keys())
+                    self.batch_buffer.append(headers)
+                    self.headers_written = True
+                
+                # 添加数据到缓存
+                values = [str(v) if v is not None else '' for v in item_dict.values()]
+                self.batch_buffer.append(values)
+                
+                # 检查是否需要刷新批量缓存
+                if len(self.batch_buffer) >= self.batch_size:
+                    await self._flush_batch()
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('csv_batch_pipeline/items_failed')
+            self.logger.error(f"CSV批量处理失败: {e}")
+            raise ItemDiscard(f"CSV Batch Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """关闭时刷新剩余缓存"""
+        try:
+            # 刷新剩余的批量数据
+            async with self.lock:
+                await self._flush_batch()
+            
+            if self.file_handle:
+                self.file_handle.close()
+                self.logger.info(f"CSV批量文件已关闭: {self.file_path}")
+                
+        except Exception as e:
+            self.logger.error(f"关闭CSV批量管道时出错: {e}")

crawlo/pipelines/json_pipeline.py

@@ -0,0 +1,219 @@
+# -*- coding: utf-8 -*-
+import json
+import asyncio
+from pathlib import Path
+from typing import Optional
+from datetime import datetime
+from crawlo.utils.log import get_logger
+from crawlo.exceptions import ItemDiscard
+
+
+class JsonPipeline:
+    """JSON文件输出管道"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        # 配置文件路径
+        self.file_path = self._get_file_path()
+        self.file_handle = None
+        self.lock = asyncio.Lock()  # 异步锁保证线程安全
+        
+        # 注册关闭事件
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        # 优先级：设置 > 爬虫属性 > 默认路径
+        file_path = (
+            self.settings.get('JSON_FILE') or
+            getattr(self.crawler.spider, 'json_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    async def _ensure_file_open(self):
+        """确保文件已打开"""
+        if self.file_handle is None:
+            self.file_handle = open(self.file_path, 'w', encoding='utf-8')
+            self.logger.info(f"JSON文件已创建: {self.file_path}")
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item的核心方法"""
+        try:
+            async with self.lock:
+                await self._ensure_file_open()
+                
+                # 转换为字典并序列化
+                item_dict = dict(item)
+                json_line = json.dumps(item_dict, ensure_ascii=False, indent=None)
+                
+                # 写入文件（每行一个JSON对象）
+                self.file_handle.write(json_line + '\n')
+                self.file_handle.flush()  # 立即刷新到磁盘
+                
+                # 统计
+                self.crawler.stats.inc_value('json_pipeline/items_written')
+                self.logger.debug(f"写入JSON项: {len(item_dict)} 字段")
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('json_pipeline/items_failed')
+            self.logger.error(f"JSON写入失败: {e}")
+            raise ItemDiscard(f"JSON Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """关闭爬虫时清理资源"""
+        if self.file_handle:
+            self.file_handle.close()
+            self.logger.info(f"JSON文件已关闭: {self.file_path}")
+
+
+class JsonLinesPipeline:
+    """JSON Lines格式输出管道（每行一个JSON对象）"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        self.file_path = self._get_file_path()
+        self.file_handle = None
+        self.items_count = 0
+        self.lock = asyncio.Lock()
+        
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        file_path = (
+            self.settings.get('JSONLINES_FILE') or
+            getattr(self.crawler.spider, 'jsonlines_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.jsonl"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    async def _ensure_file_open(self):
+        """确保文件已打开"""
+        if self.file_handle is None:
+            self.file_handle = open(self.file_path, 'w', encoding='utf-8')
+            self.logger.info(f"JSONL文件已创建: {self.file_path}")
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item方法"""
+        try:
+            async with self.lock:
+                await self._ensure_file_open()
+                
+                item_dict = dict(item)
+                
+                # 添加元数据
+                if self.settings.get_bool('JSON_ADD_METADATA', False):
+                    item_dict['_crawl_time'] = datetime.now().isoformat()
+                    item_dict['_spider_name'] = spider.name
+                
+                # 写入JSONL格式
+                json_line = json.dumps(item_dict, ensure_ascii=False, separators=(',', ':'))
+                self.file_handle.write(json_line + '\n')
+                self.file_handle.flush()
+                
+                self.items_count += 1
+                
+                # 定期日志输出
+                if self.items_count % 100 == 0:
+                    self.logger.info(f"已写入 {self.items_count} 个JSON对象")
+                
+                self.crawler.stats.inc_value('jsonlines_pipeline/items_written')
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('jsonlines_pipeline/items_failed')
+            self.logger.error(f"JSONL写入失败: {e}")
+            raise ItemDiscard(f"JSON Lines Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """资源清理"""
+        if self.file_handle:
+            self.file_handle.close()
+            self.logger.info(f"JSONL文件已关闭，共写入 {self.items_count} 个项目: {self.file_path}")
+
+
+class JsonArrayPipeline:
+    """JSON数组格式输出管道（所有item组成一个JSON数组）"""
+    
+    def __init__(self, crawler):
+        self.crawler = crawler
+        self.settings = crawler.settings
+        self.logger = get_logger(self.__class__.__name__, self.settings.get('LOG_LEVEL'))
+        
+        self.file_path = self._get_file_path()
+        self.items = []  # 内存中暂存所有items
+        self.lock = asyncio.Lock()
+        
+        crawler.subscriber.subscribe(self.spider_closed, event='spider_closed')
+        
+    @classmethod
+    def from_crawler(cls, crawler):
+        return cls(crawler)
+    
+    def _get_file_path(self) -> Path:
+        """获取输出文件路径"""
+        file_path = (
+            self.settings.get('JSON_ARRAY_FILE') or
+            getattr(self.crawler.spider, 'json_array_file', None) or
+            f"output/{self.crawler.spider.name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}_array.json"
+        )
+        
+        path = Path(file_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        return path
+    
+    async def process_item(self, item, spider) -> Optional[dict]:
+        """处理item方法"""
+        try:
+            async with self.lock:
+                item_dict = dict(item)
+                self.items.append(item_dict)
+                
+                self.crawler.stats.inc_value('json_array_pipeline/items_collected')
+                self.logger.debug(f"收集item，当前总数: {len(self.items)}")
+                
+            return item
+            
+        except Exception as e:
+            self.crawler.stats.inc_value('json_array_pipeline/items_failed')
+            self.logger.error(f"JSON Array收集失败: {e}")
+            raise ItemDiscard(f"JSON Array Pipeline处理失败: {e}")
+    
+    async def spider_closed(self):
+        """关闭时写入所有items到JSON数组文件"""
+        try:
+            if self.items:
+                with open(self.file_path, 'w', encoding='utf-8') as f:
+                    json.dump(self.items, f, ensure_ascii=False, indent=2)
+                
+                self.logger.info(f"JSON数组文件已保存，包含 {len(self.items)} 个项目: {self.file_path}")
+                self.crawler.stats.set_value('json_array_pipeline/total_items', len(self.items))
+            else:
+                self.logger.warning("没有items需要保存")
+                
+        except Exception as e:
+            self.logger.error(f"保存JSON数组文件失败: {e}")