task: add real estate story
All checks were successful
Gitea Actions Demo / deploy (push) Successful in 26s
All checks were successful
Gitea Actions Demo / deploy (push) Successful in 26s
This commit is contained in:
17
.env
Normal file
17
.env
Normal file
@ -0,0 +1,17 @@
|
|||||||
|
ENV=dev
|
||||||
|
|
||||||
|
DEBUG=true
|
||||||
|
|
||||||
|
# 日志配置
|
||||||
|
LOG_LEVEL=DEBUG
|
||||||
|
LOG_TYPE=console
|
||||||
|
|
||||||
|
# 数据库配置
|
||||||
|
DB_HOST= 47.119.128.161 # 192.168.1.200
|
||||||
|
DB_PORT=19732
|
||||||
|
DB_USER=postgres
|
||||||
|
DB_PASS=postgres
|
||||||
|
DB_NAME=peter
|
||||||
|
|
||||||
|
# LLM配置
|
||||||
|
LLM_API_KEY=sk-88d6437a6c224ccbb761ec7d994e3b34
|
||||||
1
config/__init__.py
Normal file
1
config/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
from config.settings import settings
|
||||||
18
config/database.py
Normal file
18
config/database.py
Normal file
@ -0,0 +1,18 @@
|
|||||||
|
from sqlalchemy import create_engine
|
||||||
|
from sqlalchemy.orm import sessionmaker, scoped_session
|
||||||
|
from config.settings import settings
|
||||||
|
|
||||||
|
SQLALCHEMY_SYNC_URL = (
|
||||||
|
f"postgresql+psycopg://{settings.DB_USER}:{settings.DB_PASS}"
|
||||||
|
f"@{settings.DB_HOST}:{settings.DB_PORT}/{settings.DB_NAME}"
|
||||||
|
)
|
||||||
|
|
||||||
|
engine = create_engine(
|
||||||
|
SQLALCHEMY_SYNC_URL,
|
||||||
|
echo=False, # 开发可改 True
|
||||||
|
future=True
|
||||||
|
)
|
||||||
|
|
||||||
|
SessionLocal = scoped_session(
|
||||||
|
sessionmaker(bind=engine, autoflush=False, autocommit=False)
|
||||||
|
)
|
||||||
22
config/env_loader.py
Normal file
22
config/env_loader.py
Normal file
@ -0,0 +1,22 @@
|
|||||||
|
import os
|
||||||
|
from dotenv import load_dotenv
|
||||||
|
|
||||||
|
def load_env():
|
||||||
|
"""
|
||||||
|
自动根据 ENV 加载对应的 .env 文件
|
||||||
|
"""
|
||||||
|
base_file = ".env"
|
||||||
|
prod_file = ".env.prod"
|
||||||
|
test_file = ".env.test"
|
||||||
|
|
||||||
|
# 先加载基础 .env
|
||||||
|
if os.path.exists(base_file):
|
||||||
|
load_dotenv(base_file)
|
||||||
|
|
||||||
|
# 根据参数 ENV 再加载其他环境
|
||||||
|
env = os.getenv("ENV", "dev")
|
||||||
|
|
||||||
|
if env == "prod" and os.path.exists(prod_file):
|
||||||
|
load_dotenv(prod_file, override=True)
|
||||||
|
elif env == "test" and os.path.exists(test_file):
|
||||||
|
load_dotenv(test_file, override=True)
|
||||||
34
config/settings.py
Normal file
34
config/settings.py
Normal file
@ -0,0 +1,34 @@
|
|||||||
|
from pydantic_settings import BaseSettings
|
||||||
|
from pydantic import Field
|
||||||
|
from config.env_loader import load_env
|
||||||
|
|
||||||
|
# 先加载 ENV & .env
|
||||||
|
load_env()
|
||||||
|
|
||||||
|
class Settings(BaseSettings):
|
||||||
|
# 环境
|
||||||
|
ENV: str = Field("dev")
|
||||||
|
DEBUG: bool = Field(True)
|
||||||
|
|
||||||
|
# 日志
|
||||||
|
LOG_LEVEL: str = Field("LOG_LEVEL")
|
||||||
|
LOG_FILE_PATH: str = Field("logs")
|
||||||
|
LOG_TYPE: str = Field("console")
|
||||||
|
|
||||||
|
# 数据库
|
||||||
|
DB_HOST: str
|
||||||
|
DB_PORT: int
|
||||||
|
DB_USER: str
|
||||||
|
DB_PASS: str
|
||||||
|
DB_NAME: str
|
||||||
|
|
||||||
|
# LLM配置
|
||||||
|
LLM_API_KEY: str = Field("LLM_API_KEY")
|
||||||
|
|
||||||
|
class Config:
|
||||||
|
env_file = ".env"
|
||||||
|
env_file_encoding = "utf-8"
|
||||||
|
|
||||||
|
|
||||||
|
# 全局唯一配置实例
|
||||||
|
settings = Settings()
|
||||||
@ -5,3 +5,5 @@ services:
|
|||||||
image: edward:latest
|
image: edward:latest
|
||||||
container_name: edward
|
container_name: edward
|
||||||
restart: always
|
restart: always
|
||||||
|
environment:
|
||||||
|
- TZ=Asia/Shanghai # 设置时区环境变量
|
||||||
|
|||||||
@ -5,8 +5,8 @@ from apscheduler.events import EVENT_JOB_ERROR
|
|||||||
from apscheduler.schedulers.blocking import BlockingScheduler
|
from apscheduler.schedulers.blocking import BlockingScheduler
|
||||||
|
|
||||||
from config import config
|
from config import config
|
||||||
from log.log_manager import log, logger
|
|
||||||
from task.manager_task import manager_task
|
from task.manager_task import manager_task
|
||||||
|
from utils import logger
|
||||||
|
|
||||||
|
|
||||||
def job_error_listener(event):
|
def job_error_listener(event):
|
||||||
@ -29,7 +29,7 @@ if __name__ == '__main__':
|
|||||||
scheduler.add_listener(job_error_listener, EVENT_JOB_ERROR)
|
scheduler.add_listener(job_error_listener, EVENT_JOB_ERROR)
|
||||||
|
|
||||||
try:
|
try:
|
||||||
log("started successfully.")
|
logger.info("Edward started successfully.")
|
||||||
scheduler.start() # 阻塞运行
|
scheduler.start() # 阻塞运行
|
||||||
except (KeyboardInterrupt, SystemExit):
|
except (KeyboardInterrupt, SystemExit):
|
||||||
log("Shutting down ...")
|
logger.info("Shutting down ...")
|
||||||
|
|||||||
1
llm/__init__.py
Normal file
1
llm/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
from llm.llm_thinking_engine import LLMThinkingEngine
|
||||||
144
llm/llm_thinking_engine.py
Normal file
144
llm/llm_thinking_engine.py
Normal file
@ -0,0 +1,144 @@
|
|||||||
|
from typing import Optional, Dict
|
||||||
|
from dataclasses import dataclass
|
||||||
|
import os
|
||||||
|
from config.settings import settings
|
||||||
|
from openai import OpenAI
|
||||||
|
from utils import logger
|
||||||
|
|
||||||
|
|
||||||
|
@dataclass
|
||||||
|
class LLMConfig:
|
||||||
|
"""LLM配置类"""
|
||||||
|
api_key: Optional[str] = None
|
||||||
|
base_url: str = "https://dashscope.aliyuncs.com/compatible-mode/v1"
|
||||||
|
model: str = "deepseek-v3.2"
|
||||||
|
enable_thinking: bool = True
|
||||||
|
temperature: float = 0.7
|
||||||
|
max_tokens: int = 2048
|
||||||
|
|
||||||
|
|
||||||
|
class LLMThinkingEngine:
|
||||||
|
"""LLM驱动的思考引擎实现"""
|
||||||
|
|
||||||
|
def __init__(self, system_prompt_file: str = "system_prompt.txt", config: Optional[LLMConfig] = None):
|
||||||
|
"""
|
||||||
|
初始化LLMThinkingEngine
|
||||||
|
|
||||||
|
Args:
|
||||||
|
config: LLM配置对象,如果为None则使用默认配置
|
||||||
|
"""
|
||||||
|
self.system_prompt_file = system_prompt_file
|
||||||
|
self.config = config or LLMConfig()
|
||||||
|
self._init_client()
|
||||||
|
|
||||||
|
def _init_client(self):
|
||||||
|
"""初始化OpenAI客户端"""
|
||||||
|
api_key = self.config.api_key or settings.LLM_API_KEY
|
||||||
|
self.client = OpenAI(
|
||||||
|
api_key=api_key,
|
||||||
|
base_url=self.config.base_url,
|
||||||
|
)
|
||||||
|
|
||||||
|
def think(self, user_input: str) -> str:
|
||||||
|
"""
|
||||||
|
基于LLM进行思考,返回下一步的行动
|
||||||
|
|
||||||
|
Args:
|
||||||
|
user_input: 用户输入内容
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
Thought: 思考结果,包含行动类型和内容
|
||||||
|
"""
|
||||||
|
# 构建适用于LLM的消息
|
||||||
|
messages = self._build_messages(user_input)
|
||||||
|
logger.info(f"LLM构建的消息: {messages}")
|
||||||
|
# 调用LLM进行思考
|
||||||
|
thinking_content, response_content = self._call_llm(messages)
|
||||||
|
|
||||||
|
# logger.info(f"LLM思考结果: thinking_content={thinking_content}, response_content={response_content}")
|
||||||
|
return response_content
|
||||||
|
|
||||||
|
def _build_messages(self, user_input: str) -> list[Dict[str, str]]:
|
||||||
|
"""
|
||||||
|
构建发送给LLM的消息
|
||||||
|
|
||||||
|
Args:
|
||||||
|
user_input: 用户输入内容
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
消息列表,包含系统提示、历史和当前输入
|
||||||
|
"""
|
||||||
|
messages = []
|
||||||
|
|
||||||
|
# 系统提示
|
||||||
|
system_prompt = self._get_system_prompt()
|
||||||
|
messages.append({"role": "system", "content": system_prompt})
|
||||||
|
|
||||||
|
# 用户输入
|
||||||
|
messages.append({
|
||||||
|
"role": "user",
|
||||||
|
"content": user_input
|
||||||
|
})
|
||||||
|
|
||||||
|
return messages
|
||||||
|
|
||||||
|
def _get_system_prompt(self) -> str:
|
||||||
|
"""
|
||||||
|
获取系统提示词
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
系统提示词
|
||||||
|
"""
|
||||||
|
prompt_path = os.path.join(
|
||||||
|
os.path.dirname(__file__),
|
||||||
|
"prompts",
|
||||||
|
self.system_prompt_file
|
||||||
|
)
|
||||||
|
with open(prompt_path, "r", encoding="utf-8") as f:
|
||||||
|
return f.read()
|
||||||
|
|
||||||
|
def _call_llm(self, messages: list[Dict[str, str]]) -> tuple[str, str]:
|
||||||
|
"""
|
||||||
|
调用LLM API
|
||||||
|
|
||||||
|
Args:
|
||||||
|
messages: 消息列表
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
(thinking_content, response_content): 思考内容和响应内容
|
||||||
|
"""
|
||||||
|
thinking_content = ""
|
||||||
|
response_content = ""
|
||||||
|
|
||||||
|
try:
|
||||||
|
completion = self.client.chat.completions.create(
|
||||||
|
model=self.config.model,
|
||||||
|
messages=messages,
|
||||||
|
temperature=self.config.temperature,
|
||||||
|
max_tokens=self.config.max_tokens,
|
||||||
|
extra_body={"enable_thinking": self.config.enable_thinking},
|
||||||
|
stream=True
|
||||||
|
)
|
||||||
|
|
||||||
|
# 流式处理响应
|
||||||
|
for chunk in completion:
|
||||||
|
delta = chunk.choices[0].delta
|
||||||
|
|
||||||
|
# 收集思考内容
|
||||||
|
if hasattr(delta, "reasoning_content") and delta.reasoning_content:
|
||||||
|
thinking_content += delta.reasoning_content
|
||||||
|
|
||||||
|
# 收集响应内容
|
||||||
|
if hasattr(delta, "content") and delta.content:
|
||||||
|
response_content += delta.content
|
||||||
|
|
||||||
|
except Exception as e:
|
||||||
|
# 错误处理
|
||||||
|
response_content = f"调用LLM时出错:{str(e)}"
|
||||||
|
|
||||||
|
return thinking_content, response_content
|
||||||
|
|
||||||
|
def set_config(self, config: LLMConfig):
|
||||||
|
"""更新LLM配置"""
|
||||||
|
self.config = config
|
||||||
|
self._init_client()
|
||||||
41
llm/prompts/real_estate_story_system_prompt.txt
Normal file
41
llm/prompts/real_estate_story_system_prompt.txt
Normal file
@ -0,0 +1,41 @@
|
|||||||
|
你是一个收集并解读楼市众生相的观察者。每天从全国各地的购房故事里,抽取出“当下楼市最真实的情绪信号”,分享给你的粉丝。你不唱多不唱空,只是让故事本身说话。
|
||||||
|
|
||||||
|
请根据用户提供的购房故事素材,按照以下步骤生成一篇微头条,并以JSON格式输出。
|
||||||
|
|
||||||
|
## 第一步:素材筛选
|
||||||
|
根据提供的素材,分析其是否符合发布标准:
|
||||||
|
- 是否有普遍共鸣?
|
||||||
|
- 是否有情绪张力?
|
||||||
|
- 是否有信息增量?
|
||||||
|
在输出JSON中,需包含“素材分析”字段,简要说明理由。
|
||||||
|
|
||||||
|
## 第二步:撰写标题
|
||||||
|
从以下三个标题模板中选择最合适的一个(也可微调),并说明选择理由:
|
||||||
|
1. “[情绪钩子] + [具体信息] + [留白/反问]” 示例:“买完房三天,同小区冒出套更便宜更好的”:这位女孩的遭遇,评论区炸了。
|
||||||
|
2. “刚买房就亏13万,是什么体验?这个广东女生的帖子,看得人又笑又想哭。”
|
||||||
|
3. “我好像被贝壳耍了”:一个深圳女孩的买房后悔日记。
|
||||||
|
在JSON中输出所选标题。
|
||||||
|
|
||||||
|
## 第三步:构建正文
|
||||||
|
按照以下四段式结构撰写正文,每段内容需贴合素材,语言生动真实。
|
||||||
|
- 第1段:设问/引入,建立“观察者”视角。
|
||||||
|
- 第2段:讲故事(保留原帖语气,适当精简)。
|
||||||
|
- 第3段:加入评论区的声音(制造互动感)。
|
||||||
|
- 第4段:你的观察(保持理性,不煽动)。
|
||||||
|
在JSON中输出正文,可分段列出。
|
||||||
|
|
||||||
|
## 第四步:人设检查
|
||||||
|
在生成内容后,检查是否符合以下人设要求,并在JSON中输出布尔值:
|
||||||
|
- 开头是否用了“观察者”口吻?
|
||||||
|
- 转述故事时,是否保留了原帖的真实感?
|
||||||
|
- 结尾是否有自己的理性洞察?
|
||||||
|
- 是否引导了互动?
|
||||||
|
|
||||||
|
## 输出格式要求
|
||||||
|
请将最终结果以JSON格式输出,包含以下字段:
|
||||||
|
- material_analysis(对象):包含universal_resonance(字符串)、emotional_tension(字符串)、info_increment(字符串)。
|
||||||
|
- title(字符串):所选标题。
|
||||||
|
- body(数组):正文的四个段落,每个段落为字符串。
|
||||||
|
- persona_check(对象):包含observer_perspective(布尔)、authenticity(布尔)、rational_insight(布尔)、interaction_guidance(布尔)。
|
||||||
|
|
||||||
|
确保JSON格式正确,无多余字符。
|
||||||
71
llm/prompts/system_prompt.txt
Normal file
71
llm/prompts/system_prompt.txt
Normal file
@ -0,0 +1,71 @@
|
|||||||
|
# 核心定位
|
||||||
|
你是一个收集并解读楼市众生相的观察者。每天从全国各地的购房故事里,抽取出“当下楼市最真实的情绪信号”,分享给你的粉丝。你不唱多不唱空,只是让故事本身说话。
|
||||||
|
|
||||||
|
# 第一步:素材筛选标准(什么故事值得发?)
|
||||||
|
不是所有帖子都值得写成微头条。选素材时,问自己三个问题:
|
||||||
|
是否有普遍共鸣? 这个故事是孤例,还是很多人正在经历的?(如:买完就降价、卖不掉房、谈价拉扯)
|
||||||
|
是否有情绪张力? 读者看完会“代入”吗?会想起自己或身边人吗?
|
||||||
|
是否有信息增量? 故事里有没有具体细节(价格、户型、城市、谈判过程)?
|
||||||
|
|
||||||
|
根据你给的素材:
|
||||||
|
✅ 普遍共鸣:买完就降价,几乎每个近年买房的人都怕遇到。
|
||||||
|
✅ 情绪张力:从“有根了”的喜悦,到“亏了13万”的心痛,再到“算了拉倒”的释然。
|
||||||
|
✅ 信息增量:87平、南北通透、三房两卫、降价13万、22万装修(应为22万首付或总价?原文可能有笔误)。
|
||||||
|
|
||||||
|
结论:这是一个值得发的好素材。
|
||||||
|
|
||||||
|
# 第二步:撰写标题(3选1)
|
||||||
|
标题公式: [情绪钩子] + [具体信息] + [留白/反问]
|
||||||
|
|
||||||
|
根据你的素材,以下标题供选用:
|
||||||
|
1. “买完房三天,同小区冒出套更便宜更好的”:这位女孩的遭遇,评论区炸了。
|
||||||
|
2. 刚买房就亏13万,是什么体验?这个广东女生的帖子,看得人又笑又想哭。
|
||||||
|
3. “我好像被贝壳耍了”:一个深圳女孩的买房后悔日记。
|
||||||
|
|
||||||
|
选择建议:
|
||||||
|
想引发共鸣 → 选1
|
||||||
|
想激发好奇心 → 选2
|
||||||
|
想突出真实感 → 选3
|
||||||
|
|
||||||
|
# 第三步:正文结构(4段式,可灵活调整)
|
||||||
|
第1段:设问/引入,建立“观察者”视角
|
||||||
|
你有没有想过,买房后最难受的时刻是什么?
|
||||||
|
不是还贷压力大,也不是房子降价了。
|
||||||
|
而是——你刚签完合同,同小区就挂出一套户型更好、价格更便宜的房子。
|
||||||
|
昨晚刷到一个帖子,看完心里挺不是滋味的。
|
||||||
|
|
||||||
|
第2段:讲故事(保留原帖语气,适当精简)
|
||||||
|
发帖的是一位广东女孩,网名叫momo。
|
||||||
|
她说,自己攒了很久的钱,终于买下一套房。签完约那几天,心里美滋滋的,觉得自己“有根了”,再也不用像蒲公英一样飘着。
|
||||||
|
结果三天后,她刷贝壳,发现同小区新挂出一套房——87平、南北通透、三房两卫,户型比她那个更好,总价还便宜13万。
|
||||||
|
她说:
|
||||||
|
“我买房的时候,怎么砍价都砍不下来,中介说业主不缺钱。我一买完,新房就出来了,好像是等着我似的……”
|
||||||
|
“我平时买菜都斤斤计较,这一下亏掉13万,够我装修了。”
|
||||||
|
|
||||||
|
第3段:加入评论区的声音(制造互动感)
|
||||||
|
评论区里,有人安慰她:
|
||||||
|
“你跟21、22年高位上车的人比,已经赚了几十万了。”
|
||||||
|
momo回:你还真会安慰人。
|
||||||
|
也有人苦笑:
|
||||||
|
“我23年底上车的,后面都麻木了。”
|
||||||
|
还有人怀疑:不会是贝壳算法在搞鬼吧?故意先放差房源,再放好房源?
|
||||||
|
|
||||||
|
第4段:你的观察(保持理性,不煽动)
|
||||||
|
其实,momo的遭遇,不是个例。
|
||||||
|
现在的楼市,处于一个微妙期:
|
||||||
|
|
||||||
|
卖家心态分化,有人急售降价,有人还在硬扛;
|
||||||
|
|
||||||
|
买家只能在“当下挂牌”里选,看不到两天后才会出现的那套。
|
||||||
|
这不是谁在耍谁,而是市场流动性恢复后的正常现象。
|
||||||
|
就像炒股买在阶段性高点——只要你不卖,浮亏就不是真亏。
|
||||||
|
|
||||||
|
momo最后说了一句话,还挺打动我的:
|
||||||
|
“虽然亏了,但我有自己的家了,再也不用搬家了。算了拉倒!”
|
||||||
|
也许,这才是房子最大的意义。
|
||||||
|
|
||||||
|
# 第四步:固定人设检查
|
||||||
|
- 开头是否用了“观察者”口吻?(“刷到一个帖子”“看到一个故事”)
|
||||||
|
- 转述故事时,是否保留了原帖的真实感?(尽量用原话)
|
||||||
|
- 结尾是否有自己的理性洞察?(不煽动焦虑,不唱多空)
|
||||||
|
- 是否引导了互动?(“你遇到过吗?”“评论区聊聊”)
|
||||||
2
models/__init__.py
Normal file
2
models/__init__.py
Normal file
@ -0,0 +1,2 @@
|
|||||||
|
from models.source_content import SourceContent
|
||||||
|
from models.article import Article
|
||||||
50
models/article.py
Normal file
50
models/article.py
Normal file
@ -0,0 +1,50 @@
|
|||||||
|
from datetime import datetime
|
||||||
|
from sqlalchemy.orm import Mapped, mapped_column
|
||||||
|
from sqlalchemy import String, Text, Integer, DateTime, func
|
||||||
|
from models.base import Base
|
||||||
|
|
||||||
|
|
||||||
|
class Article(Base):
|
||||||
|
__tablename__ = "t_article"
|
||||||
|
|
||||||
|
id: Mapped[int] = mapped_column(
|
||||||
|
Integer,
|
||||||
|
primary_key=True,
|
||||||
|
autoincrement=True,
|
||||||
|
comment="自动递增的唯一内容ID"
|
||||||
|
)
|
||||||
|
|
||||||
|
title: Mapped[str] = mapped_column(
|
||||||
|
String(256),
|
||||||
|
nullable=False,
|
||||||
|
index=True,
|
||||||
|
comment="标题"
|
||||||
|
)
|
||||||
|
|
||||||
|
keywords: Mapped[str | None] = mapped_column(
|
||||||
|
Text,
|
||||||
|
nullable=True,
|
||||||
|
comment="关键词"
|
||||||
|
)
|
||||||
|
|
||||||
|
content: Mapped[str | None] = mapped_column(
|
||||||
|
Text,
|
||||||
|
nullable=True,
|
||||||
|
comment="内容"
|
||||||
|
)
|
||||||
|
|
||||||
|
create_time: Mapped[datetime] = mapped_column(
|
||||||
|
DateTime(timezone=True),
|
||||||
|
server_default=func.now(),
|
||||||
|
nullable=False,
|
||||||
|
comment="创建时间"
|
||||||
|
)
|
||||||
|
|
||||||
|
used: Mapped[bool] = mapped_column(
|
||||||
|
default=False,
|
||||||
|
nullable=False,
|
||||||
|
comment="是否已被使用"
|
||||||
|
)
|
||||||
|
|
||||||
|
def __repr__(self):
|
||||||
|
return f"<Article id={self.id} title={self.title!r} used={self.used!r}>"
|
||||||
4
models/base.py
Normal file
4
models/base.py
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
from sqlalchemy.orm import DeclarativeBase
|
||||||
|
|
||||||
|
class Base(DeclarativeBase):
|
||||||
|
pass
|
||||||
57
models/source_content.py
Normal file
57
models/source_content.py
Normal file
@ -0,0 +1,57 @@
|
|||||||
|
from datetime import datetime
|
||||||
|
from sqlalchemy.orm import Mapped, mapped_column
|
||||||
|
from sqlalchemy import String, Text, Integer, DateTime, Index, func
|
||||||
|
from models.base import Base
|
||||||
|
|
||||||
|
|
||||||
|
class SourceContent(Base):
|
||||||
|
__tablename__ = "t_source_content"
|
||||||
|
|
||||||
|
id: Mapped[int] = mapped_column(
|
||||||
|
Integer,
|
||||||
|
primary_key=True,
|
||||||
|
autoincrement=True,
|
||||||
|
comment="自动递增的唯一内容ID"
|
||||||
|
)
|
||||||
|
|
||||||
|
link: Mapped[str] = mapped_column(
|
||||||
|
String(2048),
|
||||||
|
nullable=False,
|
||||||
|
index=True,
|
||||||
|
comment="链接"
|
||||||
|
)
|
||||||
|
|
||||||
|
platform: Mapped[str] = mapped_column(
|
||||||
|
String(32),
|
||||||
|
nullable=False,
|
||||||
|
comment="平台"
|
||||||
|
)
|
||||||
|
|
||||||
|
content: Mapped[str | None] = mapped_column(
|
||||||
|
Text,
|
||||||
|
nullable=True,
|
||||||
|
comment="内容"
|
||||||
|
)
|
||||||
|
|
||||||
|
create_time: Mapped[datetime] = mapped_column(
|
||||||
|
DateTime(timezone=True),
|
||||||
|
server_default=func.now(),
|
||||||
|
nullable=False,
|
||||||
|
comment="创建时间"
|
||||||
|
)
|
||||||
|
|
||||||
|
update_time: Mapped[datetime] = mapped_column(
|
||||||
|
DateTime(timezone=True),
|
||||||
|
server_default=func.now(),
|
||||||
|
onupdate=func.now(),
|
||||||
|
nullable=False,
|
||||||
|
comment="更新时间"
|
||||||
|
)
|
||||||
|
|
||||||
|
# ——可选优化:添加 项目 + 主题 联合唯一索引——
|
||||||
|
__table_args__ = (
|
||||||
|
Index("link", "link", unique=True),
|
||||||
|
)
|
||||||
|
|
||||||
|
def __repr__(self):
|
||||||
|
return f"<SourceContent id={self.id} link={self.link!r} platform={self.platform!r}>"
|
||||||
BIN
requirements.txt
BIN
requirements.txt
Binary file not shown.
42
task/hot_topic/real_estate_story.py
Normal file
42
task/hot_topic/real_estate_story.py
Normal file
@ -0,0 +1,42 @@
|
|||||||
|
from datetime import datetime
|
||||||
|
import json
|
||||||
|
from task.manager_task import execute_task
|
||||||
|
from config.database import SessionLocal
|
||||||
|
from models import SourceContent, Article
|
||||||
|
from utils import logger
|
||||||
|
from llm import LLMThinkingEngine
|
||||||
|
|
||||||
|
|
||||||
|
def story_edit_task():
|
||||||
|
with SessionLocal() as db:
|
||||||
|
# 获取今天的所有帖子信息
|
||||||
|
today_contents = db.query(SourceContent).filter(
|
||||||
|
SourceContent.create_time >= datetime.today().replace(hour=0, minute=0, second=0, microsecond=0)
|
||||||
|
).limit(10).all()
|
||||||
|
if len(today_contents) == 0:
|
||||||
|
logger.info("story_edit_task finish, content size 0")
|
||||||
|
return
|
||||||
|
logger.info(f"story_edit_task get {len(today_contents)} contents")
|
||||||
|
|
||||||
|
llm_engine = LLMThinkingEngine(system_prompt_file="real_estate_story_system_prompt.txt")
|
||||||
|
for content in today_contents:
|
||||||
|
logger.info(f"story_edit_task content id: {content.id}, title: {content.link}, platform: {content.platform}")
|
||||||
|
story = llm_engine.think(f"故事素材:{content.content}")
|
||||||
|
logger.info(f"story_edit_task content id: {content.id} story: {story}")
|
||||||
|
# 将生成的故事写入Article表
|
||||||
|
json_story = json.loads(story)
|
||||||
|
title = json_story.get("title", "无标题")
|
||||||
|
paragraphs = json_story.get("body", ["无内容"])
|
||||||
|
article_content = "\n".join(paragraphs)
|
||||||
|
article = Article(
|
||||||
|
title=title,
|
||||||
|
keywords=None,
|
||||||
|
content=article_content,
|
||||||
|
used=False
|
||||||
|
)
|
||||||
|
db.add(article)
|
||||||
|
db.commit()
|
||||||
|
# break # 目前先处理一条内容,后续再改成批量处理
|
||||||
|
|
||||||
|
if __name__ == "__main__":
|
||||||
|
execute_task(story_edit_task)
|
||||||
@ -7,7 +7,7 @@ from apscheduler.schedulers.blocking import BlockingScheduler
|
|||||||
from config import config
|
from config import config
|
||||||
from database.database import get_session
|
from database.database import get_session
|
||||||
from database.tscheduler.crud import get_tasks_by_executor
|
from database.tscheduler.crud import get_tasks_by_executor
|
||||||
from log.log_manager import log
|
from utils import logger
|
||||||
|
|
||||||
"""
|
"""
|
||||||
这是一个特殊的任务,负责管理任务,命名为管理者任务。
|
这是一个特殊的任务,负责管理任务,命名为管理者任务。
|
||||||
@ -26,10 +26,10 @@ def log_task_execution(task_name: str, start_time: float, end_time: float = None
|
|||||||
start_time_str = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(start_time))
|
start_time_str = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(start_time))
|
||||||
end_time_str = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(end_time))
|
end_time_str = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(end_time))
|
||||||
if end_time is None:
|
if end_time is None:
|
||||||
log(f"{task_name} start execute at {start_time_str}")
|
logger.info(f"{task_name} start execute at {start_time_str}")
|
||||||
else:
|
else:
|
||||||
elapsed_time = end_time - start_time
|
elapsed_time = end_time - start_time
|
||||||
log(f"{task_name} end execute at {end_time_str}, use time {elapsed_time:.2f} seconds")
|
logger.info(f"{task_name} end execute at {end_time_str}, use time {elapsed_time:.2f} seconds")
|
||||||
|
|
||||||
|
|
||||||
def execute_task(task: callable):
|
def execute_task(task: callable):
|
||||||
@ -65,7 +65,7 @@ def load_tasks(scheduler: BlockingScheduler):
|
|||||||
id=str(task_id),
|
id=str(task_id),
|
||||||
replace_existing=True
|
replace_existing=True
|
||||||
)
|
)
|
||||||
log(f"Task {task.task_name} added with interval {interval_seconds} seconds")
|
logger.info(f"Task {task.task_name} added with interval {interval_seconds} seconds")
|
||||||
elif trigger == "cron":
|
elif trigger == "cron":
|
||||||
# 解析 cron 表达式的字段
|
# 解析 cron 表达式的字段
|
||||||
fields = task.cron_expression.split()
|
fields = task.cron_expression.split()
|
||||||
@ -89,7 +89,7 @@ def load_tasks(scheduler: BlockingScheduler):
|
|||||||
id=str(task_id),
|
id=str(task_id),
|
||||||
replace_existing=True
|
replace_existing=True
|
||||||
)
|
)
|
||||||
log(f"Task {task.task_name} added with cron {task.cron_expression}")
|
logger.info(f"Task {task.task_name} added with cron {task.cron_expression}")
|
||||||
elif trigger == "date":
|
elif trigger == "date":
|
||||||
scheduler.add_job(
|
scheduler.add_job(
|
||||||
task_function,
|
task_function,
|
||||||
@ -98,9 +98,9 @@ def load_tasks(scheduler: BlockingScheduler):
|
|||||||
id=str(task_id),
|
id=str(task_id),
|
||||||
replace_existing=True
|
replace_existing=True
|
||||||
)
|
)
|
||||||
log(f"Task {task.task_name} added with date {task.execution_date}")
|
logger.info(f"Task {task.task_name} added with date {task.execution_date}")
|
||||||
else:
|
else:
|
||||||
log(f"Invalid trigger type: {trigger}")
|
logger.warning(f"Invalid trigger type: {trigger}")
|
||||||
|
|
||||||
|
|
||||||
# 管理者任务
|
# 管理者任务
|
||||||
|
|||||||
1
utils/__init__.py
Normal file
1
utils/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
from utils.logger import logger
|
||||||
37
utils/logger.py
Normal file
37
utils/logger.py
Normal file
@ -0,0 +1,37 @@
|
|||||||
|
import sys
|
||||||
|
import os
|
||||||
|
from loguru import logger
|
||||||
|
from config.settings import settings
|
||||||
|
|
||||||
|
# 移除默认的 handler(否则重复输出)
|
||||||
|
logger.remove()
|
||||||
|
|
||||||
|
if "console" in settings.LOG_TYPE:
|
||||||
|
# ======== 控制台输出 ========
|
||||||
|
logger.add(
|
||||||
|
sys.stdout,
|
||||||
|
level=settings.LOG_LEVEL,
|
||||||
|
format="<green>{time:YYYY-MM-DD HH:mm:ss}</green> "
|
||||||
|
"| <level>{level: <8}</level> "
|
||||||
|
"| <cyan>{name}</cyan>:<cyan>{function}</cyan>:<cyan>{line}</cyan> "
|
||||||
|
"- <level>{message}</level>",
|
||||||
|
)
|
||||||
|
|
||||||
|
if "file" in settings.LOG_TYPE:
|
||||||
|
# 日志目录
|
||||||
|
LOG_DIR = settings.LOG_FILE_PATH
|
||||||
|
if not os.path.exists(LOG_DIR):
|
||||||
|
os.makedirs(LOG_DIR)
|
||||||
|
|
||||||
|
# ======== 文件输出(按天切割)========
|
||||||
|
logger.add(
|
||||||
|
f"{LOG_DIR}/app_{{time:YYYY-MM-DD}}.log",
|
||||||
|
rotation="00:00", # 每天 0 点切割
|
||||||
|
retention="7 days", # 保存 7 天
|
||||||
|
encoding="utf-8",
|
||||||
|
level=settings.LOG_LEVEL,
|
||||||
|
enqueue=True, # 多线程安全
|
||||||
|
compression="zip", # 自动压缩旧日志
|
||||||
|
)
|
||||||
|
|
||||||
|
__all__ = ["logger"]
|
||||||
Reference in New Issue
Block a user