补充公司log

.env

@@ -39,3 +39,10 @@ JOB_EXPIRE_DAYS=7
 
 # 日志
 LOGGING_LEVEL=INFO
+
+# 阿里云 OSS
+OSS_ACCESS_KEY_ID=LTAI5tEdLKKQUKhTyUpfH5Mk
+OSS_ACCESS_KEY_SECRET=RjUdTrq0V5qA4b3BUElNhXqs3ZLp5k
+OSS_ENDPOINT=oss-cn-guangzhou.aliyuncs.com
+OSS_BUCKET=offerpie
+OSS_DOMAIN=https://offerpie.oss-cn-guangzhou.aliyuncs.com

.env.prod

@@ -33,3 +33,10 @@ JOB_EXPIRE_DAYS=7
 
 # 日志
 LOGGING_LEVEL=INFO
+
+# 阿里云 OSS
+OSS_ACCESS_KEY_ID=LTAI5tEdLKKQUKhTyUpfH5Mk
+OSS_ACCESS_KEY_SECRET=RjUdTrq0V5qA4b3BUElNhXqs3ZLp5k
+OSS_ENDPOINT=oss-cn-guangzhou.aliyuncs.com
+OSS_BUCKET=offerpie
+OSS_DOMAIN=https://offerpie.oss-cn-guangzhou.aliyuncs.com

.env.test

@@ -34,3 +34,10 @@ JOB_EXPIRE_DAYS=7
 
 # 日志
 LOGGING_LEVEL=INFO
+
+# 阿里云 OSS
+OSS_ACCESS_KEY_ID=LTAI5tEdLKKQUKhTyUpfH5Mk
+OSS_ACCESS_KEY_SECRET=RjUdTrq0V5qA4b3BUElNhXqs3ZLp5k
+OSS_ENDPOINT=oss-cn-guangzhou.aliyuncs.com
+OSS_BUCKET=offerpie
+OSS_DOMAIN=https://offerpie.oss-cn-guangzhou.aliyuncs.com

app/config/settings.py

@@ -48,6 +48,16 @@ class Settings(BaseSettings):
     # ──────────── 岗位下架参数 ────────────
     job_expire_days: int = 7
 
+    # ──────────── 阿里云 OSS ────────────
+    oss_access_key_id: str = ""
+    oss_access_key_secret: str = ""
+    oss_endpoint: str = "oss-cn-guangzhou.aliyuncs.com"
+    oss_bucket: str = "offerpie"
+    # 固定上传目录（对象 key 前缀）
+    oss_upload_dir: str = "company/logo"
+    # 下载访问域名
+    oss_domain: str = "https://offerpie.oss-cn-guangzhou.aliyuncs.com"
+
     # ──────────── 日志 ────────────
     logging_level: str = "INFO"
     log_file_name: str = "cleaner.log"

app/core/oss.py

@@ -0,0 +1,42 @@
+"""阿里云 OSS 上传：往固定目录上传，返回下载地址"""
+
+import asyncio
+import base64
+import os
+from uuid import uuid4
+
+import oss2
+
+from app.config import settings
+from app.core.logger import log
+
+_auth = oss2.Auth(settings.oss_access_key_id, settings.oss_access_key_secret)
+_bucket = oss2.Bucket(_auth, f"https://{settings.oss_endpoint}", settings.oss_bucket)
+
+
+async def upload(data: bytes, file_name: str) -> str:
+    """上传字节内容到固定目录，返回下载地址"""
+    key = f"{settings.oss_upload_dir}/{uuid4().hex[:18]}{os.path.splitext(file_name)[1]}"
+    await asyncio.to_thread(_bucket.put_object, key, data)
+    url = f"{settings.oss_domain}/{key}"
+    log.info("OSS 上传成功 [{}] -> {}", file_name, url)
+    return url
+
+
+async def upload_base64(b64: str, file_name: str = "image.png") -> str | None:
+    """上传 base64 图片到固定目录，返回下载地址；空值或解码失败返回 None
+
+    :param b64: base64 字符串，兼容 data URI 前缀（data:image/png;base64,xxx）
+    :param file_name: 用于取后缀，默认 png
+    """
+    if not b64 or not b64.strip():
+        return None
+    # 去掉 data URI 前缀
+    if "," in b64 and b64.strip().startswith("data:"):
+        b64 = b64.split(",", 1)[1]
+    try:
+        data = base64.b64decode(b64)
+    except Exception as e:
+        log.warning("base64 解码失败: {}", e)
+        return None
+    return await upload(data, file_name)

app/models/pg/app_url_list.py

@@ -0,0 +1,35 @@
+"""PostgreSQL: app_url_list 表模型"""
+
+from datetime import datetime
+from typing import Optional
+
+from sqlalchemy import BigInteger, DateTime, Integer, String, Text
+from sqlalchemy.orm import Mapped, mapped_column
+
+from app.core.database import PgBase
+
+
+class AppUrlList(PgBase):
+    """爬虫 URL 任务列表"""
+
+    __tablename__ = "app_url_list"
+
+    id: Mapped[int] = mapped_column(BigInteger, primary_key=True, autoincrement=True, comment="自增主键")
+    status: Mapped[str] = mapped_column(String(32), default="pending", nullable=False, comment="任务状态")
+    input_company_name: Mapped[str] = mapped_column(String(255), nullable=False, comment="输入公司名")
+    input_url: Mapped[Optional[str]] = mapped_column(String(1024), comment="输入URL")
+    output_url: Mapped[Optional[str]] = mapped_column(String(1024), comment="输出URL")
+    post_login: Mapped[Optional[int]] = mapped_column(Integer, comment="发布登录")
+    crawl_login: Mapped[Optional[int]] = mapped_column(Integer, comment="爬取登录")
+    type_sz: Mapped[Optional[int]] = mapped_column(Integer, comment="类型sz")
+    type_sx: Mapped[Optional[int]] = mapped_column(Integer, comment="类型sx")
+    crawlok_at: Mapped[Optional[str]] = mapped_column(String(32), comment="-2没有数据或者数据全部过期或者数据缺失 -3 过滤后没数据， -4无法修复，")
+    pyname: Mapped[Optional[str]] = mapped_column(String(255), comment="脚本名字")
+    type_xz: Mapped[Optional[int]] = mapped_column(Integer, comment="类型xz")
+    error_message: Mapped[Optional[str]] = mapped_column(Text, comment="错误信息")
+    logo: Mapped[Optional[str]] = mapped_column(Text, comment="公司logo (base64编码的64x64 PNG图片)")
+    started_at: Mapped[Optional[datetime]] = mapped_column(DateTime, comment="开始时间")
+    finished_at: Mapped[Optional[datetime]] = mapped_column(DateTime, comment="完成时间")
+    created_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, comment="创建时间")
+    updated_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, comment="更新时间")
+    output_company_name: Mapped[Optional[str]] = mapped_column(String(255), comment="输出公司名")

app/services/job_clean_service.py

@@ -10,6 +10,7 @@ from sqlalchemy import text, insert
 from app.config import settings
 from app.core.database import PgSession, MysqlSession
 from app.core.logger import log
+from app.core.oss import upload_base64
 from app.ai.model_config import JobCleanModel
 from app.ai.prompts import JOB_STRUCTURE_SYSTEM, MAJOR_MATCH_SYSTEM, SKILL_EXTRACT_SYSTEM
 from app.models.mysql.job import Job
@@ -134,7 +135,7 @@ async def _do_clean(data: dict) -> None:
 
     # 公司处理
     company_short_name = result.get("companyShortName") or data.get("company") or ""
-    company_id = await _find_or_create_company(company_short_name)
+    company_id = await _find_or_create_company(company_short_name, data.get("urllistid"))
 
     # 地区处理
     region_codes = []
@@ -284,8 +285,8 @@ async def _find_or_create_skill_tag(name: str) -> int | None:
         return row.scalar()
 
 
-async def _find_or_create_company(short_name: str) -> int:
+async def _find_or_create_company(short_name: str, urllistid: int | None = None) -> int:
-    """查找或创建公司（加锁防并发重复）"""
+    """查找或创建公司（加锁防并发重复）；新建公司时若有 logo 则上传 OSS 并回填地址"""
     async with _company_lock:
         async with MysqlSession() as mysql:
             row = await mysql.execute(
@@ -309,9 +310,37 @@ async def _find_or_create_company(short_name: str) -> int:
                 )
             )
             await mysql.commit()
+
+    # 锁外处理 logo：仅新建公司时执行，上传是网络IO，不阻塞其他协程；失败不影响主流程
+    try:
+        logo_b64 = await _get_logo_base64(urllistid)
+        if logo_b64:
+            logo_url = await upload_base64(logo_b64, "logo.png")
+            if logo_url:
+                async with MysqlSession() as mysql:
+                    await mysql.execute(
+                        text("UPDATE bg_company SET logo_url = :url, update_time = :t WHERE id = :id"),
+                        {"url": logo_url, "t": datetime.now(), "id": company_id},
+                    )
+                    await mysql.commit()
+    except Exception as e:
+        log.warning("[company={}] logo 上传失败: {}", company_id, e)
+
     return company_id
 
 
+async def _get_logo_base64(urllistid: int | None) -> str | None:
+    """从 PG app_url_list 按 urllistid 读取 logo（base64）"""
+    if not urllistid:
+        return None
+    async with PgSession() as pg:
+        row = await pg.execute(
+            text("SELECT logo FROM app_url_list WHERE id = :id"),
+            {"id": urllistid},
+        )
+        return row.scalar()
+
+
 async def _update_pg_status(data_id: int, status: str) -> None:
     """更新 PG 清洗状态"""
     async with PgSession() as pg:

requirements.txt

@@ -14,3 +14,4 @@ langchain-core>=0.3
 loguru>=0.7
 snowflake-id>=1.0
 json-repair>=0.30
+oss2>=2.18