jettask 0.2.5__tar.gz → 0.2.7__tar.gz

jettask-0.2.7/MANIFEST.in

@@ -0,0 +1,29 @@
+# 包含必要的文件
+include README.md
+include LICENSE
+include pyproject.toml
+
+# 包含包数据
+recursive-include jettask *.py
+recursive-include jettask/webui/static *
+recursive-include jettask/webui/frontend/dist *
+recursive-include jettask/webui *.sql
+recursive-include jettask/webui *.html
+recursive-include jettask/scheduler *.sql
+
+# 排除不必要的文件
+global-exclude __pycache__
+global-exclude *.py[co]
+global-exclude .git*
+global-exclude .DS_Store
+global-exclude *.swp
+global-exclude *~
+
+# 排除不必要的目录
+prune tests
+prune docs
+prune examples
+prune lib_reference
+prune docker
+prune jettask/webui/frontend/node_modules
+prune jettask/webui/frontend/src

{jettask-0.2.5 → jettask-0.2.7}/PKG-INFO

@@ -1,10 +1,9 @@
 Metadata-Version: 2.4
 Name: jettask
-Version: 0.2.5
+Version: 0.2.7
 Summary: A high-performance distributed task queue system with web monitoring
-Home-page: https://github.com/yourusername/jettask
-Author: JetTask Team
 Author-email: JetTask Team <support@jettask.io>
+License-Expression: MIT
 Project-URL: Homepage, https://github.com/yourusername/jettask
 Project-URL: Bug Tracker, https://github.com/yourusername/jettask/issues
 Project-URL: Documentation, https://jettask.readthedocs.io
@@ -12,7 +11,6 @@ Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Topic :: System :: Distributed Computing
-Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
@@ -23,17 +21,24 @@ Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: redis>=4.5.0
+Requires-Dist: aioredis>=2.0.0
+Requires-Dist: msgpack>=1.0.0
 Requires-Dist: watchdog>=3.0.0
 Requires-Dist: uvloop>=0.17.0
 Requires-Dist: ujson>=5.6.0
 Requires-Dist: fastapi>=0.100.0
-Requires-Dist: uvicorn>=0.23.0
+Requires-Dist: uvicorn[standard]>=0.23.0
 Requires-Dist: websockets>=11.0
-Requires-Dist: sqlalchemy[asyncio]>=2.0.0
-Requires-Dist: psycopg[binary,pool]>=3.1.0
 Requires-Dist: pydantic>=2.0.0
 Requires-Dist: python-multipart>=0.0.6
+Requires-Dist: aiohttp>=3.8.0
+Requires-Dist: sqlalchemy[asyncio]>=2.0.0
+Requires-Dist: psycopg[binary,pool]>=3.1.0
+Requires-Dist: asyncpg>=0.28.0
+Requires-Dist: croniter>=1.4.0
 Requires-Dist: click>=8.1.0
+Requires-Dist: python-dotenv>=1.0.0
+Requires-Dist: psutil>=5.9.0
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-asyncio>=0.21.0; extra == "dev"
@@ -41,7 +46,72 @@ Requires-Dist: black>=23.0; extra == "dev"
 Requires-Dist: flake8>=6.0; extra == "dev"
 Requires-Dist: mypy>=1.4.0; extra == "dev"
 Requires-Dist: coverage>=7.0; extra == "dev"
-Dynamic: author
-Dynamic: home-page
 Dynamic: license-file
-Dynamic: requires-python
+
+# JetTask
+
+一个高性能的分布式任务队列系统，支持Web监控界面。
+
+## 特性
+
+- 🚀 高性能异步任务执行
+- 📊 实时Web监控界面
+- ⏰ 支持定时任务和延迟任务
+- 🔄 任务重试和错误处理
+- 🎯 多队列和优先级支持
+- 🌍 多命名空间隔离
+- 📈 任务统计和性能监控
+- 🔧 简单易用的API
+
+## 安装
+
+```bash
+pip install jettask
+```
+
+## 快速开始
+
+### 1. 创建任务
+
+```python
+from jettask import JetTask
+
+app = JetTask()
+
+@app.task(queue="default")
+async def hello_task(name):
+    return f"Hello, {name}!"
+```
+
+### 2. 启动Worker
+
+```bash
+jettask worker -a app:app --queues default
+```
+
+### 3. 发送任务
+
+```python
+result = await hello_task.send("World")
+print(result)  # Hello, World!
+```
+
+### 4. 启动Web监控界面
+
+```bash
+# 启动API服务
+jettask api
+
+# 启动前端界面
+jettask frontend
+```
+
+然后访问 http://localhost:3000 查看监控界面。
+
+## 文档
+
+详细文档请参见 [docs/](docs/) 目录。
+
+## 许可证
+
+MIT License

jettask-0.2.7/README.md

@@ -0,0 +1,67 @@
+# JetTask
+
+一个高性能的分布式任务队列系统，支持Web监控界面。
+
+## 特性
+
+- 🚀 高性能异步任务执行
+- 📊 实时Web监控界面
+- ⏰ 支持定时任务和延迟任务
+- 🔄 任务重试和错误处理
+- 🎯 多队列和优先级支持
+- 🌍 多命名空间隔离
+- 📈 任务统计和性能监控
+- 🔧 简单易用的API
+
+## 安装
+
+```bash
+pip install jettask
+```
+
+## 快速开始
+
+### 1. 创建任务
+
+```python
+from jettask import JetTask
+
+app = JetTask()
+
+@app.task(queue="default")
+async def hello_task(name):
+    return f"Hello, {name}!"
+```
+
+### 2. 启动Worker
+
+```bash
+jettask worker -a app:app --queues default
+```
+
+### 3. 发送任务
+
+```python
+result = await hello_task.send("World")
+print(result)  # Hello, World!
+```
+
+### 4. 启动Web监控界面
+
+```bash
+# 启动API服务
+jettask api
+
+# 启动前端界面
+jettask frontend
+```
+
+然后访问 http://localhost:3000 查看监控界面。
+
+## 文档
+
+详细文档请参见 [docs/](docs/) 目录。
+
+## 许可证
+
+MIT License

jettask-0.2.7/jettask/monitor/run_backlog_collector.py

@@ -0,0 +1,96 @@
+#!/usr/bin/env python3
+"""
+Stream积压监控采集服务
+可以作为独立服务运行，定期采集Redis Stream的积压情况
+"""
+
+import asyncio
+import argparse
+import logging
+import signal
+import sys
+from stream_backlog_monitor import StreamBacklogMonitor
+
+# 配置日志
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)
+logger = logging.getLogger(__name__)
+
+
+async def main(args):
+    """主函数"""
+    # 创建监控器实例
+    monitor = StreamBacklogMonitor(
+        redis_url=args.redis_url,
+        pg_url=args.pg_url,
+        redis_prefix=args.redis_prefix
+    )
+    
+    # 设置信号处理
+    def signal_handler(sig, frame):
+        logger.info("Received stop signal, shutting down...")
+        sys.exit(0)
+    
+    signal.signal(signal.SIGINT, signal_handler)
+    signal.signal(signal.SIGTERM, signal_handler)
+    
+    # 运行采集器
+    logger.info(f"Starting backlog collector service...")
+    logger.info(f"  Redis URL: {args.redis_url}")
+    logger.info(f"  PostgreSQL URL: {args.pg_url}")
+    logger.info(f"  Redis Prefix: {args.redis_prefix}")
+    logger.info(f"  Collection Interval: {args.interval} seconds")
+    
+    try:
+        await monitor.run_collector(interval=args.interval)
+    except KeyboardInterrupt:
+        logger.info("Collector stopped by user")
+    except Exception as e:
+        logger.error(f"Collector failed: {e}")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Redis Stream Backlog Monitor Collector")
+    
+    parser.add_argument(
+        "--redis-url",
+        default="redis://localhost:6379/0",
+        help="Redis connection URL (default: redis://localhost:6379/0)"
+    )
+    
+    parser.add_argument(
+        "--pg-url",
+        default="postgresql+asyncpg://jettask:123456@localhost:5432/jettask",
+        help="PostgreSQL connection URL"
+    )
+    
+    parser.add_argument(
+        "--redis-prefix",
+        default="JETTASK",
+        help="Redis key prefix (default: JETTASK)"
+    )
+    
+    parser.add_argument(
+        "--interval",
+        type=int,
+        default=60,
+        help="Collection interval in seconds (default: 60)"
+    )
+    
+    parser.add_argument(
+        "--debug",
+        action="store_true",
+        help="Enable debug logging"
+    )
+    
+    args = parser.parse_args()
+    
+    # 设置日志级别
+    if args.debug:
+        logging.getLogger().setLevel(logging.DEBUG)
+    
+    # 运行主函数
+    asyncio.run(main(args))

jettask-0.2.7/jettask/monitor/stream_backlog_monitor.py

@@ -0,0 +1,362 @@
+"""
+Redis Stream积压监控模块
+用于监控任务队列的积压情况
+"""
+
+import asyncio
+import redis.asyncio as redis
+from datetime import datetime, timezone
+from typing import Dict, List, Optional, Tuple
+import logging
+from sqlalchemy import text
+from sqlalchemy.ext.asyncio import AsyncSession, create_async_engine
+from sqlalchemy.orm import sessionmaker
+import os
+
+logger = logging.getLogger(__name__)
+
+
+class StreamBacklogMonitor:
+    """Stream积压监控器"""
+    
+    def __init__(self, redis_url: str = None, pg_url: str = None, redis_prefix: str = "JETTASK"):
+        """
+        初始化监控器
+        
+        Args:
+            redis_url: Redis连接URL
+            pg_url: PostgreSQL连接URL  
+            redis_prefix: Redis键前缀
+        """
+        self.redis_url = redis_url or os.getenv('JETTASK_REDIS_URL', 'redis://localhost:6379/0')
+        self.pg_url = pg_url or os.getenv('JETTASK_PG_URL', 'postgresql+asyncpg://jettask:123456@localhost:5432/jettask')
+        self.redis_prefix = redis_prefix
+        
+        self.redis_client = None
+        self.engine = None
+        self.AsyncSessionLocal = None
+        
+    async def initialize(self):
+        """初始化连接"""
+        # 初始化Redis连接
+        self.redis_client = await redis.from_url(self.redis_url, decode_responses=True)
+        
+        # 初始化PostgreSQL连接
+        self.engine = create_async_engine(self.pg_url, echo=False)
+        self.AsyncSessionLocal = sessionmaker(self.engine, class_=AsyncSession, expire_on_commit=False)
+        
+    async def close(self):
+        """关闭连接"""
+        if self.redis_client:
+            await self.redis_client.close()
+        if self.engine:
+            await self.engine.dispose()
+    
+    async def update_delivered_offset(self, stream_name: str, group_name: str, messages: List[Tuple]):
+        """
+        更新消费组的已投递offset
+        
+        Args:
+            stream_name: Stream名称（队列名）
+            group_name: 消费者组名
+            messages: 消息列表
+        """
+        if not messages:
+            return
+            
+        try:
+            # 从消息中提取最大的offset
+            max_offset = 0
+            for _, msg_list in messages:
+                for msg_id, msg_data in msg_list:
+                    if b'offset' in msg_data:
+                        offset = int(msg_data[b'offset'])
+                        max_offset = max(max_offset, offset)
+            
+            if max_offset > 0:
+                # 更新Redis中的last_delivered_offset
+                key = f"{self.redis_prefix}:GROUP:{stream_name}:{group_name}:last_delivered_offset"
+                
+                # 使用Lua脚本确保只更新更大的值
+                lua_script = """
+                local current = redis.call('GET', KEYS[1])
+                if not current or tonumber(ARGV[1]) > tonumber(current) then
+                    redis.call('SET', KEYS[1], ARGV[1])
+                end
+                return redis.call('GET', KEYS[1])
+                """
+                
+                await self.redis_client.eval(lua_script, 1, key, str(max_offset))
+                logger.debug(f"Updated delivered offset for {stream_name}:{group_name} to {max_offset}")
+                
+        except Exception as e:
+            logger.error(f"Failed to update delivered offset: {e}")
+    
+    async def update_acked_offset(self, stream_name: str, group_name: str, acked_messages: List):
+        """
+        更新消费组的已确认offset
+        
+        Args:
+            stream_name: Stream名称
+            group_name: 消费者组名
+            acked_messages: 已确认的消息列表
+        """
+        if not acked_messages:
+            return
+            
+        try:
+            # 提取最大的已确认offset
+            max_offset = 0
+            for msg in acked_messages:
+                if 'offset' in msg:
+                    offset = int(msg['offset'])
+                    max_offset = max(max_offset, offset)
+            
+            if max_offset > 0:
+                # 更新Redis中的last_acked_offset
+                key = f"{self.redis_prefix}:GROUP:{stream_name}:{group_name}:last_acked_offset"
+                
+                # 使用Lua脚本确保只更新更大的值
+                lua_script = """
+                local current = redis.call('GET', KEYS[1])
+                if not current or tonumber(ARGV[1]) > tonumber(current) then
+                    redis.call('SET', KEYS[1], ARGV[1])
+                end
+                return redis.call('GET', KEYS[1])
+                """
+                
+                await self.redis_client.eval(lua_script, 1, key, str(max_offset))
+                logger.debug(f"Updated acked offset for {stream_name}:{group_name} to {max_offset}")
+                
+        except Exception as e:
+            logger.error(f"Failed to update acked offset: {e}")
+    
+    async def collect_metrics(self, namespace: str = "default", stream_names: List[str] = None) -> Dict:
+        """
+        采集指定stream的积压指标
+        使用 TASK_OFFSETS 和 QUEUE_OFFSETS 进行精确计算
+        
+        Args:
+            namespace: 命名空间
+            stream_names: 要监控的stream列表，None表示监控所有
+            
+        Returns:
+            采集的指标数据
+        """
+        metrics = {}
+        
+        try:
+            # 获取所有队列的最新offset (QUEUE_OFFSETS)
+            queue_offsets_key = f"{namespace}:QUEUE_OFFSETS"
+            queue_offsets = await self.redis_client.hgetall(queue_offsets_key)
+            
+            # 获取所有任务组的消费offset (TASK_OFFSETS)
+            task_offsets_key = f"{namespace}:TASK_OFFSETS"
+            task_offsets = await self.redis_client.hgetall(task_offsets_key)
+            
+            # 如果没有指定stream，从QUEUE_OFFSETS中获取所有队列
+            if not stream_names:
+                stream_names = list(queue_offsets.keys())
+            
+            # 对每个stream采集指标
+            for stream_name in stream_names:
+                # 使用实际的Stream键格式
+                stream_key = f"{self.redis_prefix.lower()}:QUEUE:{stream_name}"
+                
+                # 获取队列的最新offset
+                last_published_offset = int(queue_offsets.get(stream_name, 0))
+                
+                # 获取stream信息
+                try:
+                    stream_info = await self.redis_client.xinfo_stream(stream_key)
+                except:
+                    # Stream可能不存在
+                    continue
+                
+                # 获取所有消费者组信息
+                try:
+                    groups = await self.redis_client.xinfo_groups(stream_key)
+                except:
+                    groups = []
+                
+                stream_metrics = {
+                    'namespace': namespace,
+                    'stream_name': stream_name,
+                    'last_published_offset': last_published_offset,
+                    'groups': {}
+                }
+                
+                # 对每个消费者组采集指标
+                for group in groups:
+                    group_name = group['name']
+                    pending_count = group['pending']  # Redis Stream中的pending数量（已投递未ACK）
+                    
+                    # 从TASK_OFFSETS获取该组的消费offset
+                    # key格式: f"{queue}:{group_name}"
+                    task_offset_key = f"{stream_name}:{group_name}"
+                    last_acked_offset = int(task_offsets.get(task_offset_key, 0))
+                    print(f'{task_offset_key=} {last_acked_offset=}')
+                    # 计算各种积压指标
+                    # 1. 总积压 = 队列最新offset - 消费组已确认的offset
+                    total_backlog = max(0, last_published_offset - last_acked_offset)
+                    
+                    # 2. 未投递的积压 = 总积压 - pending数量
+                    #    pending_count 是已经投递给消费者但还未ACK的消息数量
+                    backlog_undelivered = max(0, total_backlog - pending_count)
+                    
+                    # 3. 已投递未确认 = pending数量（这是Redis Stream统计的）
+                    backlog_delivered_unacked = pending_count
+                    
+                    # 4. 已投递的offset = 已确认offset + pending数量
+                    last_delivered_offset = last_acked_offset + pending_count
+                    
+                    stream_metrics['groups'][group_name] = {
+                        'last_delivered_offset': last_delivered_offset,  # 已投递的最新offset
+                        'last_acked_offset': last_acked_offset,          # 已确认的最新offset
+                        'pending_count': pending_count,                  # 已投递未确认的数量
+                        'backlog_undelivered': backlog_undelivered,      # 未投递的积压
+                        'backlog_delivered_unacked': backlog_delivered_unacked,  # 已投递未确认的积压
+                        'backlog_unprocessed': total_backlog            # 总积压（未投递+已投递未确认）
+                    }
+                
+                # 如果没有消费组但有队列offset，也记录
+                if not stream_metrics['groups'] and last_published_offset > 0:
+                    stream_metrics['groups']['_total'] = {
+                        'last_delivered_offset': 0,
+                        'last_acked_offset': 0,
+                        'pending_count': 0,
+                        'backlog_undelivered': last_published_offset,
+                        'backlog_unprocessed': last_published_offset
+                    }
+                
+                metrics[stream_name] = stream_metrics
+                
+        except Exception as e:
+            logger.error(f"Failed to collect metrics: {e}")
+            import traceback
+            traceback.print_exc()
+            
+        return metrics
+    
+    async def save_metrics(self, metrics: Dict):
+        """
+        将采集的指标保存到数据库
+        
+        Args:
+            metrics: 采集的指标数据
+        """
+        if not metrics:
+            return
+            
+        try:
+            async with self.AsyncSessionLocal() as session:
+                # 准备插入数据
+                records = []
+                timestamp = datetime.now(timezone.utc)
+                
+                for stream_name, stream_data in metrics.items():
+                    # 保存每个消费组的数据
+                    for group_name, group_data in stream_data.get('groups', {}).items():
+                        record = {
+                            'namespace': stream_data['namespace'],
+                            'stream_name': stream_name,
+                            'consumer_group': group_name,
+                            'last_published_offset': stream_data['last_published_offset'],
+                            'last_delivered_offset': group_data['last_delivered_offset'],
+                            'last_acked_offset': group_data['last_acked_offset'],
+                            'pending_count': group_data['pending_count'],
+                            'backlog_undelivered': group_data['backlog_undelivered'],
+                            'backlog_unprocessed': group_data['backlog_unprocessed'],
+                            'backlog_delivered_unacked': group_data.get('backlog_delivered_unacked', group_data['pending_count']),
+                            'created_at': timestamp
+                        }
+                        records.append(record)
+                    
+                    # 如果没有消费组，也保存stream级别的数据
+                    if not stream_data.get('groups'):
+                        record = {
+                            'namespace': stream_data['namespace'],
+                            'stream_name': stream_name,
+                            'consumer_group': None,
+                            'last_published_offset': stream_data['last_published_offset'],
+                            'last_delivered_offset': 0,
+                            'last_acked_offset': 0,
+                            'pending_count': 0,
+                            'backlog_undelivered': stream_data['last_published_offset'],
+                            'backlog_unprocessed': stream_data['last_published_offset'],
+                            'created_at': timestamp
+                        }
+                        records.append(record)
+                
+                # 批量插入
+                if records:
+                    insert_sql = text("""
+                        INSERT INTO stream_backlog_monitor 
+                        (namespace, stream_name, consumer_group, last_published_offset, 
+                         last_delivered_offset, last_acked_offset, pending_count,
+                         backlog_undelivered, backlog_unprocessed, created_at)
+                        VALUES 
+                        (:namespace, :stream_name, :consumer_group, :last_published_offset,
+                         :last_delivered_offset, :last_acked_offset, :pending_count,
+                         :backlog_undelivered, :backlog_unprocessed, :created_at)
+                    """)
+                    # 注意：backlog_delivered_unacked 可以从 pending_count 推导，所以不单独存储
+                    
+                    await session.execute(insert_sql, records)
+                    await session.commit()
+                    logger.info(f"Saved {len(records)} monitoring records")
+                    
+        except Exception as e:
+            logger.error(f"Failed to save metrics: {e}")
+    
+    async def run_collector(self, interval: int = 60):
+        """
+        运行采集器
+        
+        Args:
+            interval: 采集间隔（秒）
+        """
+        await self.initialize()
+        
+        logger.info(f"Starting backlog monitor collector with {interval}s interval")
+        
+        try:
+            while True:
+                try:
+                    # 采集指标
+                    metrics = await self.collect_metrics()
+                    
+                    # 保存到数据库
+                    await self.save_metrics(metrics)
+                    
+                    # 等待下一次采集
+                    await asyncio.sleep(interval)
+                    
+                except Exception as e:
+                    logger.error(f"Collector error: {e}")
+                    await asyncio.sleep(interval)
+                    
+        except KeyboardInterrupt:
+            logger.info("Stopping collector...")
+        finally:
+            await self.close()
+
+# 辅助函数 - 供其他模块调用
+async def report_delivered_offset(redis_client, redis_prefix: str, queue: str, group_name: str, messages: List):
+    """
+    上报已投递的offset（供event_pool调用）
+    这个函数已弃用，改为直接更新TASK_OFFSETS
+    """
+    pass  # 现在offset更新在executor中完成
+
+async def report_queue_offset(redis_client, redis_prefix: str, queue: str, offset: int):
+    """
+    上报队列的最新offset（供发送消息时调用）
+    这个功能已经在发送时通过Lua脚本自动完成
+    """
+    pass  # 现在offset更新在发送时通过Lua脚本完成
+
+if __name__ == "__main__":
+    # 测试运行采集器
+    monitor = StreamBacklogMonitor()
+    asyncio.run(monitor.run_collector(interval=30))