tryprogram commited on
Commit
4c1b798
·
verified ·
1 Parent(s): ca90afd

Delete offer-puplayer-hf

Browse files
Files changed (46) hide show
  1. offer-puplayer-hf/.gitignore +0 -10
  2. offer-puplayer-hf/Dockerfile +0 -35
  3. offer-puplayer-hf/README.md +0 -67
  4. offer-puplayer-hf/backend/app/agents/__init__.py +0 -0
  5. offer-puplayer-hf/backend/app/agents/advisor.py +0 -111
  6. offer-puplayer-hf/backend/app/agents/base_agent.py +0 -50
  7. offer-puplayer-hf/backend/app/agents/jd_parser.py +0 -70
  8. offer-puplayer-hf/backend/app/agents/matcher.py +0 -99
  9. offer-puplayer-hf/backend/app/agents/resume_parser.py +0 -41
  10. offer-puplayer-hf/backend/app/database.py +0 -124
  11. offer-puplayer-hf/backend/app/main.py +0 -184
  12. offer-puplayer-hf/backend/app/models.py +0 -32
  13. offer-puplayer-hf/backend/app/orchestrator.py +0 -33
  14. offer-puplayer-hf/backend/app/services/embedding_service.py +0 -38
  15. offer-puplayer-hf/backend/app/services/job_service.py +0 -232
  16. offer-puplayer-hf/backend/app/services/resume_service.py +0 -53
  17. offer-puplayer-hf/backend/data/jobs.json +0 -402
  18. offer-puplayer-hf/backend/models/config.json +0 -32
  19. offer-puplayer-hf/backend/models/model_optimized.onnx +0 -3
  20. offer-puplayer-hf/backend/models/special_tokens_map.json +0 -7
  21. offer-puplayer-hf/backend/models/tokenizer.json +0 -0
  22. offer-puplayer-hf/backend/models/tokenizer_config.json +0 -15
  23. offer-puplayer-hf/backend/requirements.txt +0 -8
  24. offer-puplayer-hf/frontend/index.html +0 -13
  25. offer-puplayer-hf/frontend/package.json +0 -24
  26. offer-puplayer-hf/frontend/public/favicon.svg +0 -1
  27. offer-puplayer-hf/frontend/public/icons.svg +0 -24
  28. offer-puplayer-hf/frontend/src/App.css +0 -535
  29. offer-puplayer-hf/frontend/src/App.jsx +0 -122
  30. offer-puplayer-hf/frontend/src/api/index.js +0 -130
  31. offer-puplayer-hf/frontend/src/api/mock.js +0 -73
  32. offer-puplayer-hf/frontend/src/components/JobCard.jsx +0 -189
  33. offer-puplayer-hf/frontend/src/components/PipelineFlow.jsx +0 -39
  34. offer-puplayer-hf/frontend/src/components/RadarChart.jsx +0 -49
  35. offer-puplayer-hf/frontend/src/components/ReportCard.jsx +0 -149
  36. offer-puplayer-hf/frontend/src/main.jsx +0 -10
  37. offer-puplayer-hf/frontend/src/pages/DiagnosePage.jsx +0 -271
  38. offer-puplayer-hf/frontend/src/pages/HomePage.jsx +0 -548
  39. offer-puplayer-hf/frontend/src/pages/RecommendPage.jsx +0 -113
  40. offer-puplayer-hf/frontend/src/pages/SuggestPage.jsx +0 -150
  41. offer-puplayer-hf/frontend/vite.config.js +0 -12
  42. offer-puplayer-hf/models/config.json +0 -32
  43. offer-puplayer-hf/models/model_optimized.onnx +0 -3
  44. offer-puplayer-hf/models/special_tokens_map.json +0 -7
  45. offer-puplayer-hf/models/tokenizer.json +0 -0
  46. offer-puplayer-hf/models/tokenizer_config.json +0 -15
offer-puplayer-hf/.gitignore DELETED
@@ -1,10 +0,0 @@
1
- __pycache__/
2
- *.pyc
3
- .env
4
- *.db
5
- node_modules/
6
- dist/
7
- static/
8
- .DS_Store
9
- *.log
10
- package-lock.json
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/Dockerfile DELETED
@@ -1,35 +0,0 @@
1
- # Stage 1: Build frontend
2
- FROM node:20 AS frontend-build
3
- WORKDIR /app
4
- COPY frontend/package.json frontend/package-lock.json* ./
5
- RUN npm install
6
- COPY frontend/ .
7
- RUN npm run build
8
-
9
- # Stage 2: Backend
10
- FROM python:3.10-slim
11
- WORKDIR /app
12
-
13
- # Install system dependencies for pdfplumber
14
- RUN apt-get update && apt-get install -y --no-install-recommends \
15
- gcc \
16
- && rm -rf /var/lib/apt/lists/*
17
-
18
- # Copy and install Python dependencies
19
- COPY backend/requirements.txt .
20
- RUN pip install --no-cache-dir -r requirements.txt
21
-
22
- # Copy backend code
23
- COPY backend/ .
24
-
25
- # Copy pre-downloaded ONNX model
26
- COPY models/ ./models/
27
-
28
- # Copy frontend build
29
- COPY --from=frontend-build /app/dist ./static
30
-
31
- # HF Space port
32
- EXPOSE 7860
33
-
34
- # Start server
35
- CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/README.md DELETED
@@ -1,67 +0,0 @@
1
- # Offer 捕手 — HuggingFace Spaces 部署指南
2
-
3
- ## 前置
4
- - 一个 GitHub 账号(https://github.com)
5
- - 一个 HuggingFace 账号(https://huggingface.co,建议用 GitHub 登录)
6
-
7
- ## 步骤
8
-
9
- ### 1. 在 HuggingFace 创建 Space
10
-
11
- 1. 打开 https://huggingface.co/new-space
12
- 2. 填写:
13
- - **Space Name**: `offer-puplayer`
14
- - **License**: `mit`
15
- - **Space SDK**: **Docker**
16
- - **Space Hardware**: CPU basic(免费)
17
- 3. 点击 **Create Space**
18
-
19
- ### 2. 上传代码
20
-
21
- **方式一:Git 推送(推荐)**
22
- ```bash
23
- # 克隆你的 Space(替换成你的用户名)
24
- git clone https://huggingface.co/spaces/你的用户名/offer-puplayer
25
- cd offer-puplayer
26
-
27
- # 把部署文件复制进去(从本地的 offer-puplayer-hf/ 目录复制)
28
- # 然后把所有文件 add、commit、push
29
- git add .
30
- git commit -m "Initial deploy"
31
- git push
32
- ```
33
-
34
- **方式二:网页上传(更简单)**
35
- 1. 进入你的 Space 页面
36
- 2. 点击 **Files** → **Add file** → **Upload files**
37
- 3. 把 `D:\tencent_HR_aiwork\offer-puplayer-hf\` 目录下的全部文件拖进去
38
- 4. 点击 **Commit changes**
39
-
40
- ### 3. 设置环境变量
41
-
42
- 在 Space 页面的 **Settings** → **Repository Secrets** 添加:
43
-
44
- | Name | Value |
45
- |------|-------|
46
- | `DEEPSEEK_API_KEY` | `sk-07006f4e7e7945a7bef2ec29879719fd` |
47
-
48
- ### 4. 等待构建
49
-
50
- 1. 回到 Space 主页面 → 切换到 **Builder** 标签
51
- 2. 看到 Docker 构建日志(约 10-15 分钟,因为要安装 torch)
52
- 3. 构建成功后自动启动,出现 **Running** 状态
53
- 4. 访问 `https://你的用户名-offer-puplayer.hf.space`
54
-
55
- ### 5. 验证
56
-
57
- 打开浏览器访问上面的 URL:
58
- - [ ] 首页能加载
59
- - [ ] 推荐岗位列表能显示
60
- - [ ] 上传简历/手工填写 → 确认
61
- - [ ] 点击「查看匹配诊断」→ Agent 流式输出
62
-
63
- ## 常见问题
64
-
65
- - **构建失败**: 查看 Builder 日志,通常是网络问题,点 **Restart** 重试
66
- - **启动超时**: 模型加载需要 30 秒左右,HF Spaces 默认 60 秒超时,如果超时去 Settings 调大
67
- - **首次访问慢**: 首次请求要加载向量模型(30s),后续请求正常
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/agents/__init__.py DELETED
File without changes
offer-puplayer-hf/backend/app/agents/advisor.py DELETED
@@ -1,111 +0,0 @@
1
- import os
2
- from app.agents.base_agent import BaseAgent
3
-
4
- MOCK_RESPONSE = {
5
- "rewrite_suggestions": [
6
- {
7
- "dimension": "技能匹配",
8
- "issue": "JD要求Python后端开发经验,简历未体现",
9
- "original": "熟悉Java、C++,了解数据结构与算法",
10
- "suggestion": "补充Python后端项目经历,例如\"使用FastAPI开发RESTful API,实现用户认证与数据持久化\"",
11
- "reason": "JD核心技术栈为Python后端,仅列举Java/C++无法满足岗位核心要求,需明确展示Python实战经验"
12
- },
13
- {
14
- "dimension": "经验匹配",
15
- "issue": "缺少企业级后端实习经历",
16
- "original": "在校课程项目:图书管理系统",
17
- "suggestion": "突出项目难点与高并发场景,例如\"优化数据库查询减少50%响应时间,支撑日均10万请求\"",
18
- "reason": "JD要求有后端实习经验,课程项目需用量化成果弥补实习缺失,展示工程化思维"
19
- },
20
- {
21
- "dimension": "项目契合",
22
- "issue": "项目方向与JD业务场景不匹配",
23
- "original": "基于Java Swing的桌面应用开发",
24
- "suggestion": "将项目描述重构为\"基于Python的Web全栈项目\",体现完整开发流程",
25
- "reason": "JD业务涉及Web后端服务,桌面应用项目与岗位技术栈差异大,需重新包装体现相关能力"
26
- },
27
- ],
28
- "interview_tips": [
29
- {
30
- "topic": "Python异步编程与FastAPI",
31
- "reason": "JD明确要求asyncio和FastAPI,简历中无相关经验",
32
- "tip": "准备asyncio核心概念(事件循环、协程、Task),结合FastAPI源码理解异步请求处理流程,准备一个异步爬虫或WebSocket demo"
33
- },
34
- {
35
- "topic": "数据库设计与ORM",
36
- "reason": "JD要求熟悉MySQL和Redis,简历仅提到\"了解数据库\"",
37
- "tip": "复习SQLAlchemy/FastAPI集成、索引优化、事务隔离级别,准备缓存穿透/雪崩的解决方案"
38
- },
39
- {
40
- "topic": "系统设计",
41
- "reason": "岗位为校招,可能会考察基础设计能力",
42
- "tip": "准备短URL服务、秒杀系统等常见设计题,重点练习Python伪代码表达能力"
43
- },
44
- ],
45
- "encouragement": "你的基础扎实,Java和C++的背景说明你有良好的编程思维。建议集中2-3周时间做一个完整的Python后端项目(如简易电商API),就能在简历上大幅提升匹配度。你的学习能力完全胜任这个方向,加油!"
46
- }
47
-
48
-
49
- class Advisor(BaseAgent):
50
- name = "advisor"
51
-
52
- def build_prompt(self, match_report: dict, jd_text: str) -> tuple[str, dict]:
53
- system = (
54
- "你是一个求职建议专家。你的任务是:\n"
55
- "1. 基于匹配报告,给出精准的简历修改建议。\n"
56
- "2. 建议必须包含:哪个维度、具体问题、原文、改写建议、原因。\n"
57
- "3. 每条建议必须做到\"HR一眼能看出改变\"的程度,不要泛泛而谈。\n"
58
- "4. 如果匹配度低于50分,必须在鼓励中包含'从匹配度角度不推荐'的提示。\n"
59
- "5. 提供不超过3个面试准备话题(每个话题包含 topic、reason、tip)。\n"
60
- "6. 最后输出一段鼓励语,要求真诚具体,不敷衍。\n\n"
61
- "请严格按以下JSON格式输出,字段名必须完全匹配:\n"
62
- '{\n'
63
- ' "rewrite_suggestions": [\n'
64
- ' {"dimension": "技能匹配", "issue": "...", "original": "...", "suggestion": "...", "reason": "..."}\n'
65
- ' ],\n'
66
- ' "interview_tips": [\n'
67
- ' {"topic": "...", "reason": "...", "tip": "..."}\n'
68
- ' ],\n'
69
- ' "encouragement": "..."\n'
70
- '}\n\n'
71
- "输出格式为JSON,不要加markdown代码块标记。"
72
- )
73
- user = {
74
- "match_report": match_report,
75
- "jd_text": jd_text
76
- }
77
- return system, user
78
-
79
- async def generate_suggestions(self, match_report: dict, jd_text: str) -> dict:
80
- api_key = os.environ.get("DEEPSEEK_API_KEY")
81
- if not api_key or os.environ.get("MOCK_MODE", "").lower() == "true":
82
- return MOCK_RESPONSE
83
- system, user_input = self.build_prompt(match_report, jd_text)
84
- result = await self.call_llm(system, user_input)
85
- raw_suggestions = result.get("rewrite_suggestions") or result.get("suggestions") or result.get("modifications") or []
86
- normalized = []
87
- for s in raw_suggestions:
88
- if isinstance(s, dict):
89
- normalized.append({
90
- "dimension": s.get("dimension") or s.get("name") or "",
91
- "issue": s.get("issue") or s.get("problem") or "",
92
- "original": s.get("original") or s.get("original_text") or "",
93
- "suggestion": s.get("suggestion") or s.get("rewrite") or "",
94
- "reason": s.get("reason") or s.get("rationale") or "",
95
- })
96
- result["rewrite_suggestions"] = normalized
97
- raw_tips = result.get("interview_tips") or result.get("tips") or result.get("preparation") or []
98
- normalized_tips = []
99
- for t in raw_tips:
100
- if isinstance(t, dict):
101
- normalized_tips.append({
102
- "topic": t.get("topic") or t.get("title") or "",
103
- "reason": t.get("reason") or t.get("why") or "",
104
- "tip": t.get("tip") or t.get("advice") or "",
105
- })
106
- result["interview_tips"] = normalized_tips
107
- if not result.get("encouragement"):
108
- result["encouragement"] = ""
109
- if "error" in result:
110
- return MOCK_RESPONSE
111
- return result
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/agents/base_agent.py DELETED
@@ -1,50 +0,0 @@
1
- import json
2
- import logging
3
- import os
4
-
5
- import httpx
6
-
7
- logger = logging.getLogger(__name__)
8
-
9
-
10
- class BaseAgent:
11
- name = "base"
12
-
13
- async def call_llm(self, system_prompt: str, user_input: dict) -> dict:
14
- api_key = os.getenv("DEEPSEEK_API_KEY")
15
- if not api_key:
16
- logger.warning("DEEPSEEK_API_KEY not set, agent=%s cannot call LLM", self.name)
17
- return {"error": "no_api_key", "agent": self.name}
18
- try:
19
- async with httpx.AsyncClient(timeout=60.0) as client:
20
- resp = await client.post(
21
- "https://api.deepseek.com/chat/completions",
22
- headers={
23
- "Authorization": f"Bearer {api_key}",
24
- "Content-Type": "application/json",
25
- },
26
- json={
27
- "model": "deepseek-chat",
28
- "response_format": {"type": "json_object"},
29
- "messages": [
30
- {"role": "system", "content": system_prompt},
31
- {"role": "user", "content": json.dumps(user_input, ensure_ascii=False)},
32
- ],
33
- },
34
- )
35
- resp.raise_for_status()
36
- body = resp.json()
37
- content = body["choices"][0]["message"]["content"]
38
- return json.loads(content)
39
- except httpx.TimeoutException:
40
- logger.error("LLM call timeout for agent=%s", self.name)
41
- return {"error": "timeout", "agent": self.name}
42
- except json.JSONDecodeError:
43
- logger.error("LLM response parse error for agent=%s", self.name)
44
- return {"error": "format_error", "agent": self.name}
45
- except Exception:
46
- logger.exception("LLM call failed for agent=%s", self.name)
47
- return {"error": "unknown", "agent": self.name}
48
-
49
- def build_prompt(self, **kwargs) -> tuple[str, dict]:
50
- raise NotImplementedError
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/agents/jd_parser.py DELETED
@@ -1,70 +0,0 @@
1
- import os
2
-
3
- from app.agents.base_agent import BaseAgent
4
-
5
- MOCK_RESULT = {
6
- "must_have": [
7
- "Java", "SpringBoot", "MyBatis", "MySQL", "Redis",
8
- "分布式系统", "微服务架构", "本科及以上学历", "计算机相关专业",
9
- ],
10
- "nice_to_have": [
11
- "Kubernetes", "Docker", "消息队列", "Elasticsearch",
12
- "大流量高并发经验", "开源项目贡献",
13
- ],
14
- "buzzwords": [
15
- "具有较强的学习能力和团队协作精神",
16
- "工作积极主动,责任心强",
17
- "具备良好的沟通表达能力",
18
- "有较强的抗压能力",
19
- ],
20
- "expectation_signals": [
21
- {"tag": "加班文化", "evidence": "能承受较大工作压力", "confidence": "medium"},
22
- {"tag": "技术氛围", "evidence": "钻研前沿技术,追求卓越", "confidence": "high"},
23
- {"tag": "团队协作", "evidence": "具备良好的团队协作精神", "confidence": "medium"},
24
- {"tag": "结果导向", "evidence": "以结果为导向,注重交付质量", "confidence": "high"},
25
- ],
26
- "position_summary": "Java后端开发工程师,负责核心业务系统的设计与开发,要求具备扎实的Java基础和微服务实践经验",
27
- }
28
-
29
-
30
- class JdParser(BaseAgent):
31
- name = "jd_parser"
32
-
33
- def build_prompt(self, jd_text: str) -> tuple[str, dict]:
34
- system = (
35
- "你是一个JD解析专家。请对提供的JD文本执行以下四项任务:\n\n"
36
- "1. 去噪(Denoise):识别并过滤掉模板话术/通用套话(buzzwords),"
37
- "这些描述通常在任何岗位中都能出现,缺乏具体信息量。\n\n"
38
- "2. 分层(Layer):将JD中要求的能力/条件分为两类:\n"
39
- " - must_have:明确要求的必备技能或条件\n"
40
- " - nice_to_have:加分项、优先考虑项、期望但不强制\n\n"
41
- "3. 翻译(Translate):将模糊描述转换为具体的技术栈或技能标签。\n"
42
- " 例如:'熟悉主流框架' → SpringBoot, MyBatis\n"
43
- " 例如:'熟悉数据库技术' → MySQL, Redis\n"
44
- " 例如:'熟练掌握Linux操作' → Linux\n\n"
45
- "4. 期望信号提取(Expectation Signals):提取JD中隐含的关于加班文化、"
46
- "团队氛围、薪酬待遇、技术风格等信号。\n"
47
- " 每条信号必须包含:\n"
48
- " - tag:信号类别(如'加班文化''技术氛围''结果导向''团队协作'等)\n"
49
- " - evidence:原文引用(精确的JD原文段落)\n"
50
- " - confidence:置信度(high/medium/low)\n\n"
51
- "请严格按以下JSON格式输出:\n"
52
- "{\n"
53
- ' "must_have": ["技能1", "技能2", ...],\n'
54
- ' "nice_to_have": ["技能1", "技能2", ...],\n'
55
- ' "buzzwords": ["套话1", "套话2", ...],\n'
56
- ' "expectation_signals": [\n'
57
- ' {"tag": "信号类别", "evidence": "原文引用", "confidence": "high/medium/low"}\n'
58
- " ],\n"
59
- ' "position_summary": "一句话总结该岗位"\n'
60
- "}"
61
- )
62
- user = {"jd_text": jd_text}
63
- return system, user
64
-
65
- async def parse_jd(self, jd_text: str) -> dict:
66
- if not os.getenv("DEEPSEEK_API_KEY"):
67
- return dict(MOCK_RESULT)
68
- system, user_input = self.build_prompt(jd_text)
69
- result = await self.call_llm(system, user_input)
70
- return result
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/agents/matcher.py DELETED
@@ -1,99 +0,0 @@
1
- import os
2
- from copy import deepcopy
3
-
4
- from app.agents.base_agent import BaseAgent
5
-
6
-
7
- _MOCK_RESULT = {
8
- "dimensions": [
9
- {"name": "技能匹配", "score": 80, "evidence": {"jd_quote": "熟悉Python", "resume_quote": "熟练使用Python进行数据处理和模型开发"}},
10
- {"name": "经验匹配", "score": 40, "evidence": {"jd_quote": "3年以上相关经验", "resume_quote": "6个月实习经历"}},
11
- {"name": "学历匹配", "score": 100, "evidence": {"jd_quote": "计算机相关专业", "resume_quote": "计算机科学与技术专业"}},
12
- {"name": "项目契合", "score": 65, "evidence": {"jd_quote": "有推荐系统经验优先", "resume_quote": "课程项目实现了协同过滤推荐算法"}},
13
- {"name": "软性素质", "score": 75, "evidence": {"jd_quote": "具备团队协作能力", "resume_quote": "项目描述中提及与3名同学协作完成"}},
14
- ],
15
- "total_score": 72,
16
- "strength": "技能基础扎实,学历完全符合要求",
17
- "gap": "缺乏全职工作经验,项目深度有待提升",
18
- "hr_view": {
19
- "first_impression": "技能列表与 JD 匹配度高,但项目描述缺乏量化成果",
20
- "attention_points": [
21
- {"aspect": "技能栈匹配", "finding": "Python/Flask 完全覆盖 JD 要求", "concern": "无"},
22
- {"aspect": "项目深度", "finding": "仅课程项目,无企业级实践", "concern": "项目描述偏功能实现,未见性能优化或架构设计"},
23
- {"aspect": "学历背景", "finding": "985 计算机专业,符合门槛", "concern": "无"},
24
- ],
25
- "hr_concern_level": "medium",
26
- "screen_summary": "简历整体合格,项目深度是最大短板,建议补充量化成果",
27
- },
28
- }
29
-
30
-
31
- class Matcher(BaseAgent):
32
- name = "matcher"
33
-
34
- def build_prompt(self, jd_parsed: dict, resume_data: dict) -> tuple[str, dict]:
35
- system = (
36
- "你是一个简历匹配诊断专家。你的任务是将候选人的简历与岗位要求进行多维度匹配评估。\n\n"
37
- "匹配维度(固定5个):\n"
38
- "1. 技能匹配 — 候选人掌握的技术栈 vs JD要求的技术栈\n"
39
- "2. 经验匹配 — 工作/实习年限 vs JD要求\n"
40
- "3. 学历匹配 — 学历/专业 vs JD要求\n"
41
- "4. 项目契合 — 项目经历与岗位方向的相关度\n"
42
- "5. 软性素质 — 沟通/协作/自驱力(从项目描述中推断)\n\n"
43
- "评分规则:\n"
44
- "- 每个维度 0-100 分\n"
45
- "- 每个分数必须附 evidence 字段:\n"
46
- " - jd_quote: JD中的原文依据\n"
47
- " - resume_quote: 简历中的原文依据\n"
48
- "- 证据必须精确到原文引用,不能模糊概括\n\n"
49
- "额外要求:\n"
50
- "- total_score 为5个维度的加权总分(权重:技能0.25, 经验0.20, 学历0.15, 项目0.25, 软性0.15)\n"
51
- "- strength 总结候选人优势\n"
52
- "- gap 总结候选人差距\n"
53
- "- hr_view 字段提供HR快速筛查模拟视角,包含 first_impression / attention_points(aspect/finding/concern)/ hr_concern_level(low/medium/high)/ screen_summary\n\n"
54
- "请严格按以下JSON格式输出,字段名必须完全匹配:\n"
55
- '{\n'
56
- ' "dimensions": [\n'
57
- ' {"name": "技能匹配", "score": 80, "evidence": {"jd_quote": "JD原文", "resume_quote": "简历原文"}},\n'
58
- ' {"name": "经验匹配", "score": 40, "evidence": {"jd_quote": "...", "resume_quote": "..."}},\n'
59
- ' {"name": "学历匹配", "score": 100, "evidence": {"jd_quote": "...", "resume_quote": "..."}},\n'
60
- ' {"name": "项目契合", "score": 65, "evidence": {"jd_quote": "...", "resume_quote": "..."}},\n'
61
- ' {"name": "软性素质", "score": 75, "evidence": {"jd_quote": "...", "resume_quote": "..."}}\n'
62
- ' ],\n'
63
- ' "total_score": 72,\n'
64
- ' "strength": "优势总结",\n'
65
- ' "gap": "差距总结",\n'
66
- ' "hr_view": {\n'
67
- ' "first_impression": "...",\n'
68
- ' "attention_points": [\n'
69
- ' {"aspect": "...", "finding": "...", "concern": "..."}\n'
70
- ' ],\n'
71
- ' "hr_concern_level": "medium",\n'
72
- ' "screen_summary": "..."\n'
73
- ' }\n'
74
- '}\n\n'
75
- "确保 total_score 为 0-100 的整数(加权:技能0.25, 经验0.20, 学历0.15, 项目0.25, 软性0.15)。\n\n"
76
- "输出格式: JSON"
77
- )
78
- user = {"jd_parsed": jd_parsed, "resume_data": resume_data}
79
- return system, user
80
-
81
- async def match_resume(self, resume_data: dict, jd_parsed: dict) -> dict:
82
- api_key = os.environ.get("DEEPSEEK_API_KEY")
83
- if not api_key or os.environ.get("MOCK_MODE", "").lower() == "true":
84
- return deepcopy(_MOCK_RESULT)
85
- system, user = self.build_prompt(jd_parsed=jd_parsed, resume_data=resume_data)
86
- result = await self.call_llm(system, user)
87
- raw_dims = result.get("dimensions") or result.get("dimension_list") or result.get("matches") or []
88
- normalized = []
89
- for d in raw_dims:
90
- if isinstance(d, dict):
91
- normalized.append({
92
- "name": d.get("name") or d.get("dimension") or d.get("title") or "",
93
- "score": d.get("score") or d.get("value") or 0,
94
- "evidence": d.get("evidence") or d.get("detail") or {"jd_quote": "", "resume_quote": ""},
95
- })
96
- result["dimensions"] = normalized
97
- if "error" in result:
98
- return deepcopy(_MOCK_RESULT)
99
- return result
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/agents/resume_parser.py DELETED
@@ -1,41 +0,0 @@
1
- import os
2
-
3
- from app.agents.base_agent import BaseAgent
4
-
5
- MOCK_RESULT = {
6
- "name": "张三",
7
- "school": "华南理工大学",
8
- "major": "计算机科学与技术",
9
- "grade": "研二",
10
- "skills": ["Python", "Java", "SQL", "PyTorch", "Flask"],
11
- "projects": ["基于协同过滤的推荐系统", "在线商城后端开发"],
12
- "summary": "技术栈集中在Python/Java后端开发,有推荐系统和Web开发项目经验",
13
- }
14
-
15
-
16
- class ResumeParser(BaseAgent):
17
- name = "resume_parser"
18
-
19
- def build_prompt(self, pdf_text: str) -> tuple[str, dict]:
20
- system = (
21
- "你是一个简历解析专家。从提供的简历文本中提取以下结构化信息,字段名必须精确匹配:\n"
22
- "1. name: 姓名\n"
23
- "2. school: 学校\n"
24
- "3. major: 专业\n"
25
- "4. grade: 年级/学历(如'研二'、'本科四年级'、'硕士')\n"
26
- "5. skills: 技能列表(数组,提取所有技术关键词,不遗漏)\n"
27
- "6. projects: 项目经历列表(数组,每项包含 name 和 description 字段)\n"
28
- "7. summary: 一句话摘要\n\n"
29
- "输出JSON格式,不要markdown代码块。"
30
- )
31
- user = {"resume_text": pdf_text}
32
- return system, user
33
-
34
- async def parse_resume(self, pdf_text: str) -> dict:
35
- if not os.getenv("DEEPSEEK_API_KEY"):
36
- return dict(MOCK_RESULT)
37
- system, user_input = self.build_prompt(pdf_text)
38
- result = await self.call_llm(system, user_input)
39
- if "error" in result:
40
- return dict(MOCK_RESULT)
41
- return result
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/database.py DELETED
@@ -1,124 +0,0 @@
1
- import sqlite3
2
- import json
3
- import os
4
- import logging
5
-
6
- logger = logging.getLogger(__name__)
7
-
8
- DB_PATH = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), 'data', 'offer.db')
9
-
10
-
11
- def get_conn():
12
- conn = sqlite3.connect(DB_PATH)
13
- conn.row_factory = sqlite3.Row
14
- return conn
15
-
16
-
17
- def init_db():
18
- conn = get_conn()
19
- conn.executescript("""
20
- CREATE TABLE IF NOT EXISTS users (
21
- id INTEGER PRIMARY KEY AUTOINCREMENT,
22
- name TEXT DEFAULT '',
23
- school TEXT DEFAULT '',
24
- major TEXT DEFAULT '',
25
- grade TEXT DEFAULT '',
26
- skills TEXT DEFAULT '[]',
27
- projects TEXT DEFAULT '[]',
28
- summary TEXT DEFAULT '',
29
- experience TEXT DEFAULT '',
30
- education TEXT DEFAULT '',
31
- created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
32
- );
33
-
34
- CREATE TABLE IF NOT EXISTS jobs (
35
- id INTEGER PRIMARY KEY AUTOINCREMENT,
36
- job_id INTEGER UNIQUE,
37
- name TEXT NOT NULL,
38
- company TEXT DEFAULT '',
39
- city TEXT DEFAULT '',
40
- salary TEXT DEFAULT '',
41
- tags TEXT DEFAULT '[]',
42
- description TEXT DEFAULT '',
43
- requirements TEXT DEFAULT '',
44
- expectation_signals TEXT DEFAULT '[]',
45
- embedding TEXT DEFAULT '',
46
- created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
47
- );
48
-
49
- CREATE TABLE IF NOT EXISTS diagnoses (
50
- id INTEGER PRIMARY KEY AUTOINCREMENT,
51
- user_id INTEGER REFERENCES users(id),
52
- job_id INTEGER REFERENCES jobs(job_id),
53
- match_report TEXT DEFAULT '{}',
54
- advice TEXT DEFAULT '{}',
55
- created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
56
- );
57
- """)
58
-
59
- count = conn.execute("SELECT COUNT(*) FROM jobs").fetchone()[0]
60
- if count == 0:
61
- jobs_path = os.path.join(os.path.dirname(DB_PATH), 'jobs.json')
62
- jobs = json.load(open(jobs_path, encoding='utf-8'))
63
- for j in jobs:
64
- conn.execute(
65
- "INSERT INTO jobs (job_id, name, company, city, salary, tags, description, requirements, expectation_signals) VALUES (?,?,?,?,?,?,?,?,?)",
66
- (int(j['id'].split('_')[-1]) if isinstance(j['id'], str) else j['id'],
67
- j['name'], j.get('company',''), j.get('city',''), j.get('salary',''),
68
- json.dumps(j.get('tags',[]), ensure_ascii=False),
69
- j.get('description',''), j.get('requirements',''),
70
- json.dumps(j.get('expectation_signals',[]), ensure_ascii=False))
71
- )
72
- conn.commit()
73
- _compute_job_embeddings(conn)
74
-
75
-
76
- def _compute_job_embeddings(conn):
77
- try:
78
- from app.services.embedding_service import compute_embedding
79
- rows = conn.execute("SELECT job_id, name, description, requirements, tags FROM jobs").fetchall()
80
- for r in rows:
81
- text = " ".join([
82
- r["name"] or "",
83
- r["description"] or "",
84
- r["requirements"] or "",
85
- " ".join(json.loads(r["tags"] or "[]")),
86
- ])
87
- vec = compute_embedding(text)
88
- conn.execute(
89
- "UPDATE jobs SET embedding=? WHERE job_id=?",
90
- (json.dumps(vec, ensure_ascii=False), r["job_id"])
91
- )
92
- conn.commit()
93
- logger.info("Computed embeddings for %d jobs", len(rows))
94
- except Exception as e:
95
- logger.warning("Failed to compute job embeddings: %s", e)
96
-
97
- count = conn.execute("SELECT COUNT(*) FROM users").fetchone()[0]
98
- if count == 0:
99
- conn.execute(
100
- "INSERT INTO users (name, school, major, grade, skills, projects, summary) VALUES (?,?,?,?,?,?,?)",
101
- ('张三', '华南理工大学', '计算机科学与技术', '研二',
102
- json.dumps(['Python','Java','SQL','PyTorch','Flask'], ensure_ascii=False),
103
- json.dumps(['基于协同过滤的推荐系统','在线商城后端开发'], ensure_ascii=False),
104
- '技术栈集中在Python/Java后端开发,有推荐系统和Web开发项目经验')
105
- )
106
-
107
- conn.commit()
108
- conn.close()
109
-
110
-
111
- def migrate_db():
112
- conn = get_conn()
113
- try:
114
- conn.execute("SELECT embedding FROM jobs LIMIT 1")
115
- except:
116
- conn.execute("ALTER TABLE jobs ADD COLUMN embedding TEXT DEFAULT ''")
117
- conn.commit()
118
- empty = conn.execute("SELECT COUNT(*) FROM jobs WHERE embedding IS NULL OR embedding = ''").fetchone()[0]
119
- if empty > 0:
120
- try:
121
- _compute_job_embeddings(conn)
122
- except Exception as e:
123
- logger.warning("Embedding computation failed: %s", e)
124
- conn.close()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/main.py DELETED
@@ -1,184 +0,0 @@
1
- import json
2
- from contextlib import asynccontextmanager
3
- from pathlib import Path
4
-
5
- from dotenv import load_dotenv
6
- from fastapi import FastAPI, File, UploadFile, Query
7
- from fastapi.middleware.cors import CORSMiddleware
8
- from fastapi.responses import StreamingResponse
9
-
10
- from app import database
11
- from app.models import GenerateRequest, RerankRequest, UploadResponse, ResumeData
12
- from app.orchestrator import ReportOrchestrator
13
- from app.services import resume_service, job_service
14
-
15
- load_dotenv()
16
-
17
- orchestrator = ReportOrchestrator()
18
-
19
- _confirmed_resume = None
20
-
21
-
22
- @asynccontextmanager
23
- async def lifespan(app: FastAPI):
24
- database.init_db()
25
- database.migrate_db()
26
- print("Offer 捕手后端已启动")
27
- yield
28
-
29
-
30
- app = FastAPI(lifespan=lifespan)
31
-
32
- app.add_middleware(
33
- CORSMiddleware,
34
- allow_origins=["*"],
35
- allow_credentials=True,
36
- allow_methods=["*"],
37
- allow_headers=["*"],
38
- )
39
-
40
-
41
- @app.post("/api/resume/upload")
42
- async def upload_resume(file: UploadFile = File(...)):
43
- content = await file.read()
44
- result = await resume_service.parse_resume_pdf(content)
45
- if result is None:
46
- return UploadResponse(success=False, error="简历解析失败")
47
- return UploadResponse(success=True, data=result)
48
-
49
-
50
- @app.post("/api/resume/confirm")
51
- async def confirm_resume(req: ResumeData):
52
- global _confirmed_resume
53
- data = req.model_dump()
54
- edu = data.get("education", "")
55
- if isinstance(edu, dict):
56
- edu = edu.get("school", "") or edu.get("name", "") or json.dumps(edu, ensure_ascii=False)
57
- data["education"] = edu
58
- _confirmed_resume = data
59
- conn = database.get_conn()
60
- conn.execute(
61
- "UPDATE users SET name=?, school=?, major=?, grade=?, skills=?, projects=?, summary=?, experience=?, education=? WHERE id=1",
62
- (data.get("name",""), data.get("school",""), data.get("major",""), data.get("grade",""),
63
- json.dumps(data.get("skills",[]), ensure_ascii=False),
64
- json.dumps(data.get("projects",[]), ensure_ascii=False),
65
- data.get("summary",""), data.get("experience",""), data.get("education",""))
66
- )
67
- conn.commit()
68
- conn.close()
69
- return {"success": True, "data": _confirmed_resume}
70
-
71
-
72
- @app.get("/api/jobs/recommend")
73
- async def recommend_jobs(resume_id: int = Query(...)):
74
- conn = database.get_conn()
75
- user = conn.execute("SELECT * FROM users WHERE id = ?", (resume_id,)).fetchone()
76
- conn.close()
77
- if user:
78
- resume_data = {
79
- "name": user["name"] or "",
80
- "school": user["school"] or "",
81
- "major": user["major"] or "",
82
- "grade": user["grade"] or "",
83
- "skills": json.loads(user["skills"] or "[]"),
84
- "projects": json.loads(user["projects"] or "[]"),
85
- "summary": user["summary"] or "",
86
- "experience": user["experience"] or "",
87
- "education": user["education"] or "",
88
- }
89
- else:
90
- resume_data = resume_service.MOCK_RESULT
91
- jobs = job_service.recommend_jobs(resume_data)
92
- return {"jobs": jobs}
93
-
94
-
95
- def _load_job_by_id(job_id: int) -> dict | None:
96
- conn = database.get_conn()
97
- row = conn.execute("SELECT * FROM jobs WHERE job_id = ?", (job_id,)).fetchone()
98
- conn.close()
99
- if row is None:
100
- return None
101
- return dict(row)
102
-
103
-
104
- @app.post("/api/report/generate")
105
- async def generate_report(req: GenerateRequest):
106
- job = _load_job_by_id(req.job_id)
107
- jd_text = (job.get("description", "") + "\n" + job.get("requirements", "")) if job else ""
108
- resume_dict = req.resume_data.model_dump() if req.resume_data else resume_service.MOCK_RESULT
109
-
110
- async def event_stream():
111
- async for event in orchestrator.generate_report(resume_dict, jd_text):
112
- yield f"data: {event}\n\n"
113
-
114
- return StreamingResponse(
115
- event_stream(),
116
- media_type="text/event-stream",
117
- )
118
-
119
-
120
- @app.post("/api/report/rerank")
121
- async def rerank_report(req: RerankRequest):
122
- job = _load_job_by_id(req.job_id)
123
- jd_text = (job.get("description", "") + "\n" + job.get("requirements", "")) if job else ""
124
-
125
- from app.agents.jd_parser import JdParser
126
- from app.agents.matcher import Matcher
127
-
128
- parser = JdParser()
129
- matcher = Matcher()
130
-
131
- jd_parsed = await parser.parse_jd(jd_text)
132
-
133
- if req.edited_resume_data is not None:
134
- edited_resume_data = req.edited_resume_data
135
- else:
136
- try:
137
- parsed = json.loads(req.edited_resume)
138
- if isinstance(parsed, dict) and ("skills" in parsed or "projects" in parsed):
139
- edited_resume_data = parsed
140
- else:
141
- edited_resume_data = {
142
- "skills": [],
143
- "projects": [req.edited_resume],
144
- "summary": req.edited_resume,
145
- }
146
- except (json.JSONDecodeError, TypeError):
147
- edited_resume_data = {
148
- "skills": [],
149
- "projects": [req.edited_resume],
150
- "summary": req.edited_resume,
151
- }
152
-
153
- resume_for_previous = _confirmed_resume if _confirmed_resume else (req.resume_data.model_dump() if req.resume_data else resume_service.MOCK_RESULT)
154
-
155
- match_result = await matcher.match_resume(edited_resume_data, jd_parsed)
156
- current_rate = match_result.get("total_score", 50)
157
-
158
- previous_result = await matcher.match_resume(resume_for_previous, jd_parsed)
159
- previous_rate = previous_result.get("total_score", 50)
160
-
161
- return {"match_rate": current_rate, "previous_rate": previous_rate}
162
-
163
-
164
- # Serve frontend static files (for HuggingFace Spaces deployment)
165
- from fastapi.responses import FileResponse
166
-
167
-
168
- @app.get("/{full_path:path}")
169
- async def serve_frontend(full_path: str):
170
- static_dir = Path(__file__).parent.parent / "static"
171
- file_path = static_dir / full_path
172
- if file_path.exists() and file_path.is_file():
173
- return FileResponse(str(file_path))
174
- index_path = static_dir / "index.html"
175
- if index_path.exists():
176
- return FileResponse(str(index_path))
177
- from fastapi.responses import JSONResponse
178
- return JSONResponse({"detail": "Not Found"}, status_code=404)
179
-
180
-
181
- if __name__ == "__main__":
182
- import uvicorn
183
-
184
- uvicorn.run("app.main:app", host="0.0.0.0", port=7860)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/models.py DELETED
@@ -1,32 +0,0 @@
1
- from pydantic import BaseModel, Field
2
- from typing import Optional
3
-
4
-
5
- class ResumeData(BaseModel):
6
- name: str = ""
7
- school: str = ""
8
- major: str = ""
9
- grade: str = ""
10
- skills: list[str] = Field(default_factory=list)
11
- projects: list[str | dict] = Field(default_factory=list)
12
- summary: str = ""
13
- experience: str = ""
14
- education: str | dict = ""
15
-
16
-
17
- class UploadResponse(BaseModel):
18
- success: bool
19
- data: Optional[ResumeData] = None
20
- error: Optional[str] = None
21
-
22
-
23
- class GenerateRequest(BaseModel):
24
- job_id: int
25
- resume_data: Optional[ResumeData] = None
26
-
27
-
28
- class RerankRequest(BaseModel):
29
- job_id: int
30
- edited_resume: str
31
- resume_data: Optional[ResumeData] = None
32
- edited_resume_data: Optional[dict] = None
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/orchestrator.py DELETED
@@ -1,33 +0,0 @@
1
- import json
2
- from typing import AsyncGenerator
3
-
4
- from app.agents.jd_parser import JdParser
5
- from app.agents.matcher import Matcher
6
- from app.agents.advisor import Advisor
7
-
8
-
9
- class ReportOrchestrator:
10
- def __init__(self):
11
- self.jd_parser = JdParser()
12
- self.matcher = Matcher()
13
- self.advisor = Advisor()
14
-
15
- async def generate_report(self, resume_data: dict, jd_text: str) -> AsyncGenerator[str, None]:
16
- yield json.dumps({"type": "agent_start", "agent": "agent1"}, ensure_ascii=False)
17
- jd_parsed = await self.jd_parser.parse_jd(jd_text)
18
- yield json.dumps({"type": "agent_complete", "agent": "agent1", "data": jd_parsed}, ensure_ascii=False)
19
-
20
- yield json.dumps({"type": "agent_start", "agent": "agent2"}, ensure_ascii=False)
21
- match_report = await self.matcher.match_resume(resume_data, jd_parsed)
22
- yield json.dumps({"type": "agent_complete", "agent": "agent2", "data": match_report}, ensure_ascii=False)
23
-
24
- yield json.dumps({"type": "agent_start", "agent": "agent3"}, ensure_ascii=False)
25
- advice = await self.advisor.generate_suggestions(match_report, jd_text)
26
- yield json.dumps({"type": "agent_complete", "agent": "agent3", "data": advice}, ensure_ascii=False)
27
-
28
- full_report = {
29
- "jd_parsed": jd_parsed,
30
- "match_report": match_report,
31
- "advice": advice,
32
- }
33
- yield json.dumps({"type": "report_complete", "data": full_report}, ensure_ascii=False)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/services/embedding_service.py DELETED
@@ -1,38 +0,0 @@
1
- import logging
2
- import math
3
- import os
4
-
5
- logger = logging.getLogger(__name__)
6
-
7
- _model = None
8
- _LOCAL_MODEL_PATH = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "models")
9
-
10
-
11
- def _get_model():
12
- global _model
13
- if _model is None:
14
- logger.info("Loading embedding model ...")
15
- from fastembed import TextEmbedding
16
- local_path = _LOCAL_MODEL_PATH
17
- if os.path.isdir(local_path) and os.path.isfile(os.path.join(local_path, "model_optimized.onnx")):
18
- logger.info("Loading from local path: %s", local_path)
19
- _model = TextEmbedding(model_name=local_path, cache_dir=os.path.dirname(local_path))
20
- else:
21
- os.environ.setdefault("HF_ENDPOINT", "https://hf-mirror.com")
22
- _model = TextEmbedding("BAAI/bge-small-zh-v1.5")
23
- logger.info("Embedding model loaded")
24
- return _model
25
-
26
-
27
- def compute_embedding(text: str) -> list[float]:
28
- vec = list(_get_model().embed(text))[0]
29
- return vec.tolist()
30
-
31
-
32
- def cosine_similarity(a: list[float], b: list[float]) -> float:
33
- dot = sum(x * y for x, y in zip(a, b))
34
- na = math.sqrt(sum(x * x for x in a))
35
- nb = math.sqrt(sum(y * y for y in b))
36
- if na == 0 or nb == 0:
37
- return 0.0
38
- return dot / (na * nb)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/services/job_service.py DELETED
@@ -1,232 +0,0 @@
1
- import json
2
- import os
3
- import logging
4
-
5
- import httpx
6
-
7
- from app import database
8
-
9
- logger = logging.getLogger(__name__)
10
-
11
-
12
- _985_UNIVERSITIES = {
13
- "清华大学","北京大学","复旦大学","上海交通大学","浙江大学","南京大学","中国科学技术大学",
14
- "哈尔滨工业大学","西安交通大学","武汉大学","华中科技大学","中山大学","四川大学","南开大学",
15
- "天津大学","山东大学","东南大学","吉林大学","厦门大学","同济大学","北京师范大学",
16
- "国防科技大学","中国人民大学","北京航空航天大学","北京理工大学","中国农业大学",
17
- "中央民族大学","华东师范大学","兰州大学","西北工业大学","电子科技大学",
18
- "重庆大学","湖南大学","中南大学","大连理工大学","东北大学","华南理工大学"
19
- }
20
-
21
- _211_UNIVERSITIES = _985_UNIVERSITIES | {
22
- "北京科技大学","北京交通大学","北京邮电大学","北京林业大学","北京外国语大学",
23
- "中国传媒大学","中央财经大学","对外经济贸易大学","北京体育大学","中国政法大学",
24
- "华北电力大学","河北工业大学","太原理工大学","内蒙古大学","辽宁大学","大连海事大学",
25
- "延边大学","东北师范大学","哈尔滨工程大学","东北农业大学","华东理工大学","东华大学",
26
- "上海外国语大学","上海财经大学","上海大学","苏州大学","南京航空航天大学","南京理工大学",
27
- "中国矿业大学","河海大学","江南大学","南京农业大学","中国药科大学","南京师范大学",
28
- "安徽大学","合肥工业大学","福州大学","南昌大学","中国石油大学","郑州大学",
29
- "中国地质大学","武汉理工大学","华中农业大学","华中师范大学","中南财经政法大学",
30
- "湖南师范大学","暨南大学","华南师范大学","广西大学","海南大学","西南交通大学",
31
- "西南大学","西南财经大学","贵州大学","云南大学","西藏大学","西北大学",
32
- "西安电子科技大学","长安大学","陕西师范大学","青海大学","宁夏大学","石河子大学",
33
- "新疆大学","北京工业大学","天津医科大学","南京信息工程大学","南方科技大学",
34
- "上海科技大学","中国社科院大学","中国科学院大学"
35
- }
36
-
37
-
38
- def _get_school_tier(school: str) -> int:
39
- if not school:
40
- return 0
41
- if school in _985_UNIVERSITIES:
42
- return 2
43
- if school in _211_UNIVERSITIES:
44
- return 1
45
- return 0
46
-
47
-
48
- def _build_resume_text(resume_data: dict) -> str:
49
- skills = " ".join(s.lower().strip() for s in resume_data.get("skills", []))
50
- projects = " ".join(
51
- p if isinstance(p, str) else (p.get("name", "") + " " + p.get("description", ""))
52
- for p in resume_data.get("projects", [])
53
- )
54
- parts = [
55
- resume_data.get("summary") or "",
56
- projects,
57
- resume_data.get("experience") or "",
58
- resume_data.get("education") or "",
59
- resume_data.get("major") or "",
60
- skills,
61
- ]
62
- return " ".join(parts)
63
-
64
-
65
- def _calc_match_rate(resume_data: dict, job: dict, tag_score: float, edu_score: float, vec_score: float) -> float:
66
- final_score = vec_score * 0.35 + tag_score * 0.45 + edu_score * 0.20
67
- return round(min(final_score, 1.0), 4)
68
-
69
-
70
- def _calc_tag_score(resume_data: dict, job: dict) -> float:
71
- skills = [s.lower().strip() for s in resume_data.get("skills", [])]
72
- tags = [t.lower().strip() for t in job.get("tags", [])]
73
- if not tags:
74
- return 0.0
75
- matched = sum(1 for t in tags if t in skills)
76
- return matched / len(tags)
77
-
78
-
79
- def _calc_edu_score(resume_data: dict, job: dict) -> float:
80
- school = (resume_data.get("school") or "")
81
- grade = (resume_data.get("grade") or "").lower()
82
- jd_requirements = (job.get("requirements") or "").lower()
83
- score = 0.4
84
- tier = _get_school_tier(school)
85
- if tier >= 2:
86
- score = 1.0
87
- elif tier >= 1:
88
- score = 0.8
89
- elif school:
90
- score = 0.5
91
- if ("硕士" in grade or "博士" in grade or "研" in grade) and "硕士" in jd_requirements:
92
- score = min(score + 0.15, 1.0)
93
- return score
94
-
95
-
96
- def recommend_jobs(resume_data: dict) -> list[dict]:
97
- if not resume_data or not resume_data.get("skills"):
98
- return []
99
-
100
- conn = database.get_conn()
101
- rows = conn.execute("SELECT * FROM jobs ORDER BY job_id").fetchall()
102
- conn.close()
103
-
104
- jobs = []
105
- for r in rows:
106
- embedding_raw = r["embedding"] if "embedding" in r.keys() else ""
107
- jobs.append({
108
- "job_id": r["job_id"],
109
- "name": r["name"],
110
- "company": r["company"],
111
- "city": r["city"],
112
- "salary": r["salary"],
113
- "tags": json.loads(r["tags"] or "[]"),
114
- "description": r["description"],
115
- "requirements": r["requirements"],
116
- "expectation_signals": json.loads(r["expectation_signals"] or "[]"),
117
- "_embedding": json.loads(embedding_raw) if embedding_raw else None,
118
- })
119
-
120
- resume_text = _build_resume_text(resume_data)
121
- resume_vec = None
122
- try:
123
- from app.services.embedding_service import compute_embedding, cosine_similarity
124
- resume_vec = compute_embedding(resume_text)
125
- except Exception as e:
126
- logger.warning("Resume embedding failed: %s", e)
127
-
128
- for job in jobs:
129
- tag_score = _calc_tag_score(resume_data, job)
130
- edu_score = _calc_edu_score(resume_data, job)
131
-
132
- vec_raw = 0.0
133
- if resume_vec and job["_embedding"]:
134
- vec_raw = cosine_similarity(resume_vec, job["_embedding"])
135
-
136
- job["_tag_score"] = tag_score
137
- job["_edu_score"] = edu_score
138
- job["_vec_raw"] = vec_raw
139
-
140
- for job in jobs:
141
- job["_vec_score"] = max(0, min((job["_vec_raw"] - 0.30) / 0.40, 1.0))
142
-
143
- scored = []
144
- for job in jobs:
145
- match_rate = _calc_match_rate(
146
- resume_data, job,
147
- job["_tag_score"], job["_edu_score"], job["_vec_score"]
148
- )
149
- scored.append({
150
- "job_id": job["job_id"],
151
- "match_rate": match_rate,
152
- "job": job,
153
- })
154
-
155
- scored.sort(key=lambda x: x["match_rate"], reverse=True)
156
- top_n = scored
157
-
158
- api_key = os.getenv("DEEPSEEK_API_KEY")
159
- if api_key and top_n:
160
- try:
161
- job_summaries = []
162
- for item in top_n:
163
- j = item["job"]
164
- job_summaries.append(
165
- f' {j.get("job_id")}: {j.get("name")} @ {j.get("company")} | '
166
- f'tags={j.get("tags")} | match_rate={item["match_rate"]}'
167
- )
168
-
169
- system_prompt = (
170
- "你是一个岗位推荐排序专家。给定候选人的技能列表和全部候选岗位,"
171
- "请根据岗位与候选人的匹配程度重新排序,输出最匹配的岗位ID列表(从最匹配到最不匹配)。"
172
- "输出格式为JSON:{\"ranked_job_ids\": [1, 2, 3, ...]}"
173
- )
174
- user_input = {
175
- "skills": resume_data.get("skills", []),
176
- "candidates": "\n".join(job_summaries),
177
- }
178
-
179
- with httpx.Client(timeout=30.0) as client:
180
- resp = client.post(
181
- "https://api.deepseek.com/chat/completions",
182
- headers={
183
- "Authorization": f"Bearer {api_key}",
184
- "Content-Type": "application/json",
185
- },
186
- json={
187
- "model": "deepseek-chat",
188
- "response_format": {"type": "json_object"},
189
- "messages": [
190
- {"role": "system", "content": system_prompt},
191
- {"role": "user", "content": json.dumps(user_input, ensure_ascii=False)},
192
- ],
193
- },
194
- )
195
- resp.raise_for_status()
196
- body = resp.json()
197
- content = body["choices"][0]["message"]["content"]
198
- llm_result = json.loads(content)
199
-
200
- ranked_ids = llm_result.get("ranked_job_ids", [])
201
- id_to_item = {item["job"]["job_id"]: item for item in top_n}
202
- re_ranked = []
203
- for raw_id in ranked_ids:
204
- if raw_id in id_to_item:
205
- re_ranked.append(id_to_item[raw_id])
206
- for item in top_n:
207
- if item["job"]["job_id"] not in ranked_ids:
208
- re_ranked.append(item)
209
- top_n = re_ranked
210
- except Exception as e:
211
- logger.warning("LLM ranking failed: %s", e)
212
-
213
- return [
214
- {
215
- "job_id": item["job_id"],
216
- "match_rate": item["match_rate"],
217
- "match_detail": {
218
- "semantic": round(item["job"]["_vec_score"], 4),
219
- "tags": round(item["job"]["_tag_score"], 4),
220
- "education": round(item["job"]["_edu_score"], 4),
221
- },
222
- "name": item["job"].get("name", ""),
223
- "company": item["job"].get("company", ""),
224
- "city": item["job"].get("city", ""),
225
- "salary": item["job"].get("salary", ""),
226
- "tags": item["job"].get("tags", []),
227
- "description": item["job"].get("description", ""),
228
- "requirements": item["job"].get("requirements", ""),
229
- "expectation_signals": item["job"].get("expectation_signals", []),
230
- }
231
- for item in top_n
232
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/app/services/resume_service.py DELETED
@@ -1,53 +0,0 @@
1
- import asyncio
2
- import io
3
- import logging
4
-
5
- try:
6
- import pdfplumber
7
- PDFPLUMBER_AVAILABLE = True
8
- except ImportError:
9
- PDFPLUMBER_AVAILABLE = False
10
-
11
- logger = logging.getLogger(__name__)
12
-
13
- MOCK_RESULT = {
14
- "name": "张三",
15
- "school": "华南理工大学",
16
- "major": "计算机科学与技术",
17
- "grade": "研二",
18
- "skills": ["Python", "Java", "SQL", "PyTorch", "Flask"],
19
- "projects": [
20
- "基于协同过滤的推荐系统",
21
- "在线商城后端开发",
22
- ],
23
- "summary": "技术栈集中在Python/Java后端开发,有推荐系统和Web开发项目经验",
24
- }
25
-
26
-
27
- async def parse_resume_pdf(file_bytes: bytes) -> dict:
28
- if not PDFPLUMBER_AVAILABLE:
29
- return dict(MOCK_RESULT)
30
-
31
- try:
32
- with pdfplumber.open(io.BytesIO(file_bytes)) as pdf:
33
- text_parts = []
34
- for page in pdf.pages:
35
- page_text = page.extract_text()
36
- if page_text:
37
- text_parts.append(page_text)
38
- full_text = "\n".join(text_parts)
39
-
40
- if not full_text.strip():
41
- return dict(MOCK_RESULT)
42
-
43
- from app.agents.resume_parser import ResumeParser
44
- parser = ResumeParser()
45
- result = await parser.parse_resume(full_text)
46
- if isinstance(result, dict) and "error" not in result:
47
- return result
48
- logger.warning("Resume parser returned error: %s", result.get("error"))
49
- return dict(MOCK_RESULT)
50
-
51
- except Exception as e:
52
- logger.error("Resume parsing failed: %s", e)
53
- return dict(MOCK_RESULT)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/data/jobs.json DELETED
@@ -1,402 +0,0 @@
1
- [
2
- {
3
- "id": "job_001",
4
- "name": "后端开发实习生",
5
- "company": "腾讯",
6
- "city": "深圳",
7
- "salary": "200-250元/天",
8
- "tags": ["Go", "微服务", "MySQL", "Redis", "Kafka"],
9
- "description": "负责腾讯云核心服务的后端模块开发,参与高并发分布式系统的设计与优化,完成接口文档编写与单元测试覆盖。",
10
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Go或Java语言,了解微服务架构;具备扎实的数据结构与算法基础;有分布式系统项目经验者优先。",
11
- "expectation_signals": [
12
- {"tag": "加班强度大", "evidence": "腾讯核心部门普遍加班,尤其是云业务线交付压力大", "confidence": 0.75},
13
- {"tag": "技术氛围浓厚", "evidence": "腾讯以技术驱动著称,内部技术分享和开源文化丰富", "confidence": 0.85},
14
- {"tag": "待遇高于行业平均", "evidence": "200-250元/天在实习岗位中属于较高水平", "confidence": 0.80},
15
- {"tag": "核心业务参与", "evidence": "腾讯云核心服务直接服务于公司战略级业务", "confidence": 0.85}
16
- ]
17
- },
18
- {
19
- "id": "job_002",
20
- "name": "AI算法实习生",
21
- "company": "字节跳动",
22
- "city": "北京",
23
- "salary": "250-300元/天",
24
- "tags": ["LLM", "PyTorch", "NLP", "模型微调", "AIGC"],
25
- "description": "参与大语言模型的训练与推理优化,负责数据清洗、Prompt工程、模型评测以及AIGC相关能力的落地探索。",
26
- "requirements": "2026届硕士及以上学历,人工智能/计算机相关专业;熟悉Transformer架构和PyTorch框架;有NLP/CV方向顶会论文或开源项目贡献者优先。",
27
- "expectation_signals": [
28
- {"tag": "加班强度大", "evidence": "字节跳动以快速迭代和高强度工作节奏著称", "confidence": 0.80},
29
- {"tag": "AI前沿方向", "evidence": "LLM训练与推理优化处于AI技术最前沿", "confidence": 0.90},
30
- {"tag": "待遇优厚", "evidence": "250-300元/天在算法实习岗位中极具竞争力", "confidence": 0.85},
31
- {"tag": "技术驱动文化", "evidence": "字节跳动重视技术投入,AI团队资源充足", "confidence": 0.85}
32
- ]
33
- },
34
- {
35
- "id": "job_003",
36
- "name": "前端开发实习生",
37
- "company": "阿里巴巴",
38
- "city": "杭州",
39
- "salary": "200-250元/天",
40
- "tags": ["React", "TypeScript", "Node.js", "Webpack", "CSS"],
41
- "description": "负责淘宝/天猫商家端PC及移动端H5页面开发,参与组件库建设与性能优化,推动前端工程化与自动化测试落地。",
42
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉HTML/CSS/JavaScript及React框架;了解前端构建工具链;有跨端开发经验者优先。",
43
- "expectation_signals": [
44
- {"tag": "大促期间加班多", "evidence": "阿里大促期间前端需随时待命保障页面稳定", "confidence": 0.70},
45
- {"tag": "技术沉淀深厚", "evidence": "阿里前端技术生态成熟,具有业界领先的组件库体系", "confidence": 0.80},
46
- {"tag": "待遇中等偏上", "evidence": "200-250元/天在前端实习中属于中等偏上", "confidence": 0.75},
47
- {"tag": "高流量业务场景", "evidence": "淘宝/天猫商家端服务千万级商家,流量巨大", "confidence": 0.85}
48
- ]
49
- },
50
- {
51
- "id": "job_004",
52
- "name": "产品经理实习生",
53
- "company": "美团",
54
- "city": "北京",
55
- "salary": "180-220元/天",
56
- "tags": ["用户研究", "数据分析", "PRD", "竞品分析", "Axure"],
57
- "description": "负责到店餐饮业务线产品需求分析,输出PRD并跟进开发迭代,通过用户调研和数据埋点驱动产品体验优化。",
58
- "requirements": "2026届本科及以上学历,专业不限;具备出色的逻辑分析能力和用户同理心;有互联网产品实习经验或独立上线产品项目者优先。",
59
- "expectation_signals": [
60
- {"tag": "业务节奏快", "evidence": "美团到店餐饮业务竞争激烈,产品迭代速度快", "confidence": 0.70},
61
- {"tag": "数据驱动文化", "evidence": "美团以数据分析驱动决策著称,产品岗位需强数据能力", "confidence": 0.85},
62
- {"tag": "待遇行业中等", "evidence": "180-220元/天属于产品实习岗位中等水平", "confidence": 0.70},
63
- {"tag": "核心业务线", "evidence": "到店餐饮是美团核心业务之一,成长空间大", "confidence": 0.80}
64
- ]
65
- },
66
- {
67
- "id": "job_005",
68
- "name": "数据分析实习生",
69
- "company": "快手",
70
- "city": "北京",
71
- "salary": "200-250元/天",
72
- "tags": ["SQL", "Python", "Tableau", "AB实验", "数据建模"],
73
- "description": "负责快手主站DAU增长方向的数据监控与归因分析,搭建AB实验指标体系,产出业务洞察报告辅助运营决策。",
74
- "requirements": "2026届本科及以上学历,统计学/数学/计算机相关专业;精通SQL和Python数据分���库;有海量数据处理经验;熟悉因果推断方法者优先。",
75
- "expectation_signals": [
76
- {"tag": "工作节奏较快", "evidence": "快手增长团队工作节奏快,数据产出时效性要求高", "confidence": 0.65},
77
- {"tag": "数据导向文化", "evidence": "快手高度重视数据驱动业务决策,数据岗位话语权强", "confidence": 0.85},
78
- {"tag": "待遇中等偏上", "evidence": "200-250元/天在数据分析实习中属于较高水平", "confidence": 0.75},
79
- {"tag": "核心业务指标", "evidence": "DAU增长方向直接关系公司核心KPI", "confidence": 0.85}
80
- ]
81
- },
82
- {
83
- "id": "job_006",
84
- "name": "客户端开发实习生",
85
- "company": "百度",
86
- "city": "北京",
87
- "salary": "200-250元/天",
88
- "tags": ["Android", "Kotlin", "性能优化", "Jetpack", "Gradle"],
89
- "description": "参与百度App Android端的功能开发与性能优化,负责模块化架构下的需求迭代、启动速度优化及崩溃率治理。",
90
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Android开发及Kotlin/Java语言;了解常用性能分析工具;有开源Android项目者优先。",
91
- "expectation_signals": [
92
- {"tag": "核心产品维护任务重", "evidence": "百度App用户量巨大,性能优化工作持续不断", "confidence": 0.70},
93
- {"tag": "技术积累深厚", "evidence": "百度Android技术团队积累了大量的性能优化经验", "confidence": 0.80},
94
- {"tag": "待遇中等偏上", "evidence": "200-250元/天在客户端实习中较为优厚", "confidence": 0.75},
95
- {"tag": "大规模用户场景", "evidence": "百度App服务亿万用户,技术挑战大", "confidence": 0.85}
96
- ]
97
- },
98
- {
99
- "id": "job_007",
100
- "name": "测试开发实习生",
101
- "company": "网易",
102
- "city": "广州",
103
- "salary": "180-220元/天",
104
- "tags": ["自动化测试", "Python", "Selenium", "CI/CD", "接口测试"],
105
- "description": "负责网易游戏周边产品的质量保障体系建设,设计和实现接口自动化与UI自动化用例,推动流水线集成与测试左移。",
106
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Python及至少一种自动化测试框架;了解CI/CD流程;有游戏测试经验者优先。",
107
- "expectation_signals": [
108
- {"tag": "阶段性加班", "evidence": "游戏行业上线前测试密集,版本节点需加班", "confidence": 0.60},
109
- {"tag": "重视质量文化", "evidence": "网易游戏周边产品质量保障体系完善", "confidence": 0.80},
110
- {"tag": "待遇行业中等", "evidence": "180-220元/天属于测试岗位中等水平", "confidence": 0.70},
111
- {"tag": "测试体系建设", "evidence": "从零建设质量保障体系,锻炼系统化思维", "confidence": 0.80}
112
- ]
113
- },
114
- {
115
- "id": "job_008",
116
- "name": "运维开发实习生",
117
- "company": "小红书",
118
- "city": "上海",
119
- "salary": "200-250元/天",
120
- "tags": ["Kubernetes", "Docker", "Prometheus", "Terraform", "Shell"],
121
- "description": "负责小红书推荐系统SRE工作,参与容器化部署改造、监控告警规则配置及运维自动化脚本开发,保障线上服务高可用。",
122
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Linux操作系统和容器技术;了解监控与日志系统;有大规模集群运维经验者优先。",
123
- "expectation_signals": [
124
- {"tag": "需值班响应", "evidence": "SRE岗位需24小时值班,紧急情况及时响应", "confidence": 0.75},
125
- {"tag": "DevOps文化深厚", "evidence": "小红书技术团队DevOps实践成熟,自动化程度高", "confidence": 0.80},
126
- {"tag": "待遇中等偏上", "evidence": "200-250元/天在运维岗位中属于较高水平", "confidence": 0.75},
127
- {"tag": "核心基础设施", "evidence": "推荐系统SRE保障公司核心业务的高可用", "confidence": 0.85}
128
- ]
129
- },
130
- {
131
- "id": "job_009",
132
- "name": "运营实习生",
133
- "company": "B站",
134
- "city": "上海",
135
- "salary": "150-200元/天",
136
- "tags": ["内容运营", "社群运营", "数据分析", "活动策划", "PS"],
137
- "description": "负责B站知识区创作者运营,制定社群运营策略并落地创作者拉新与活跃活动,通过数据复盘优化运营SOP。",
138
- "requirements": "2026届本科及以上学历,专业不限;B站重度用户,熟悉站内生态和UP主文化;有较强的文案和活动策划能力;有社群运营实习经验者优先。",
139
- "expectation_signals": [
140
- {"tag": "活动节点较忙", "evidence": "活动运营期间工作节奏较快,需多线程推进", "confidence": 0.55},
141
- {"tag": "年轻化社区氛围", "evidence": "B站内容运营团队年轻有活力,氛围轻松", "confidence": 0.85},
142
- {"tag": "待遇偏低", "evidence": "150-200元/天在实习岗位中处于偏低水平", "confidence": 0.60},
143
- {"tag": "核心运营岗位", "evidence": "知识区创作者运营直接服务B站核心内容生态", "confidence": 0.80}
144
- ]
145
- },
146
- {
147
- "id": "job_010",
148
- "name": "游戏开发实习生",
149
- "company": "米哈游",
150
- "city": "上海",
151
- "salary": "250-300元/天",
152
- "tags": ["Unity", "C#", "渲染管线", "性能优化", "Shader"],
153
- "description": "参与米哈游在研项目的游戏客户端功能开发,负责战斗系统模块迭代、渲染效果调优及移动端性能优化。",
154
- "requirements": "2026届本科及以上学历,计算机/图形学相关专业;熟悉Unity引擎和C#开发;了解渲染管线与GPU调试;有Game Jam或独立游戏作品者优先。",
155
- "expectation_signals": [
156
- {"tag": "加班较为常见", "evidence": "米哈游项目研发周期紧张,游戏行业加班普遍", "confidence": 0.80},
157
- {"tag": "追求极致品质", "evidence": "米哈游以技术和内容品质著称,对交付质量要求极高", "confidence": 0.90},
158
- {"tag": "待遇行业较高", "evidence": "250-300元/天在游戏开发实习中属于较高水平", "confidence": 0.85},
159
- {"tag": "核心技术岗位", "evidence": "战斗系统与渲染性能是游戏核心体验的关键", "confidence": 0.90}
160
- ]
161
- },
162
- {
163
- "id": "job_011",
164
- "name": "后端开发实习生(Java)",
165
- "company": "字节跳动",
166
- "city": "北京",
167
- "salary": "250-300元/天",
168
- "tags": ["Java", "Spring Boot", "微服务", "MySQL", "Kafka", "Redis"],
169
- "description": "参与字节跳动抖音电商后端服务开发,负责订单、支付等核心链路的迭代优化,支撑高并发秒杀场景下的系统稳定性。",
170
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Java及Spring Boot框架;了解分布式系统基本原理;有高并发项目经验者优先。",
171
- "expectation_signals": [
172
- {"tag": "加班强度高", "evidence": "抖音电商业务快速扩张,研发节奏极快", "confidence": 0.85},
173
- {"tag": "核心业务线", "evidence": "电商是字节跳动战略级业务,资源倾斜明显", "confidence": 0.90},
174
- {"tag": "待遇优厚", "evidence": "250-300元/天处于后端实习的顶尖水平", "confidence": 0.85},
175
- {"tag": "海量并发场景", "evidence": "抖音电商服务亿万用户,秒杀场景极其挑战", "confidence": 0.90}
176
- ]
177
- },
178
- {
179
- "id": "job_012",
180
- "name": "数据平台开发实习生",
181
- "company": "蚂蚁集团",
182
- "city": "杭州",
183
- "salary": "200-280元/天",
184
- "tags": ["Flink", "Spark", "Hadoop", "Kafka", "SQL", "Java"],
185
- "description": "参与蚂蚁集团数据中台建设,负责实时计算管线的开发与优化,保障百亿级数据量的离线与实时处理链路稳定运行。",
186
- "requirements": "2026届本科及以上学历,计算机/数学相关专业;熟悉Java/Scala;了解大数据生态组件(Flink/Spark/Hive);有数据仓库建模经验者优先。",
187
- "expectation_signals": [
188
- {"tag": "业务稳定性要求高", "evidence": "金融级数据平台的稳定性直接关系到核心交易链路", "confidence": 0.85},
189
- {"tag": "大数据技术深度", "evidence": "蚂蚁数据中台具备业界顶级的大数据技术积累", "confidence": 0.90},
190
- {"tag": "待遇优厚", "evidence": "200-280元/天在大数据岗位中属于较高水平", "confidence": 0.80},
191
- {"tag": "核心基础设施", "evidence": "数据中台支撑全集团的数据业务决策", "confidence": 0.85}
192
- ]
193
- },
194
- {
195
- "id": "job_013",
196
- "name": "全栈开发实习生",
197
- "company": "华为",
198
- "city": "武汉",
199
- "salary": "200-250元/天",
200
- "tags": ["React", "TypeScript", "Node.js", "Python", "Docker", "PostgreSQL"],
201
- "description": "参与华为云DevOps平台的全栈开发,负责前端界面交互、后端API设计及云服务部署,推动内部工具的平台化升级。",
202
- "requirements": "2026届本科及以上学历,计算机相关专业;同时具备前端(React/TypeScript)和后端(Node.js/Python)开发能力;有全栈项目经验者优先。",
203
- "expectation_signals": [
204
- {"tag": "项目周期相对可控", "evidence": "内部工具平台开发节奏相比核心业务略为平稳", "confidence": 0.60},
205
- {"tag": "技术栈全面", "evidence": "可接触前后端+云服务全套技术栈", "confidence": 0.85},
206
- {"tag": "内部转正机会高", "evidence": "华为云实习生转正率较高,表现优秀可留用", "confidence": 0.80},
207
- {"tag": "武汉性价比高", "evidence": "武汉生活成本远低于北上深,实习工资购买力强", "confidence": 0.75}
208
- ]
209
- },
210
- {
211
- "id": "job_014",
212
- "name": "推荐系统开发实习生",
213
- "company": "快手",
214
- "city": "北京",
215
- "salary": "250-300元/天",
216
- "tags": ["C++", "Python", "推荐算法", "TensorFlow", "Spark", "特征工程"],
217
- "description": "负责快手主站推荐系统的特征工程与模型serving优化,参与召回/排序/重排全链路迭代,提升核心内容消费指标。",
218
- "requirements": "2026届硕士及以上学历,计算机/机器学习相关专业;熟悉C++或Python,了解常用推荐算法(FM/DeepFM/DIN);有推荐系统实习经验者优先。",
219
- "expectation_signals": [
220
- {"tag": "技术密集型岗位", "evidence": "推荐系统是快手最核心的技术方向,投入巨大", "confidence": 0.90},
221
- {"tag": "加班强度较大", "evidence": "推荐算法团队每周多次实验迭代,节奏快", "confidence": 0.75},
222
- {"tag": "待遇行业顶级", "evidence": "250-300元/天是算法岗位中的高薪水平", "confidence": 0.85},
223
- {"tag": "核心业务影响力", "evidence": "推荐效果直接决定快手DAU和使用时长", "confidence": 0.90}
224
- ]
225
- },
226
- {
227
- "id": "job_015",
228
- "name": "安全工程师实习生",
229
- "company": "奇安信",
230
- "city": "北京",
231
- "salary": "200-250元/天",
232
- "tags": ["渗透测试", "Web安全", "Python", "Burp Suite", "SQL注入"],
233
- "description": "参与奇安信安全服务项目,负责Web应用渗透测试、漏洞挖掘与复现、安全报告编写,支撑国家级安全保障活动。",
234
- "requirements": "2026届本科及以上学历,信息安全/计算机相关专业;熟悉OWASP Top 10漏洞原理;至少有1次CTF或漏洞挖掘实战经历;持有CISP-PTE认证者优先。",
235
- "expectation_signals": [
236
- {"tag": "项目节奏明显", "evidence": "安全服务项目按合同交付,节点性加班", "confidence": 0.60},
237
- {"tag": "国家项目经验", "evidence": "有机会参与国家级安全保障活动,履历含金量高", "confidence": 0.85},
238
- {"tag": "待遇中等偏上", "evidence": "200-250元/天在安全岗位中处于较好水平", "confidence": 0.75},
239
- {"tag": "实战能力锻炼", "evidence": "大量真实攻防场景,技术成长极快", "confidence": 0.90}
240
- ]
241
- },
242
- {
243
- "id": "job_016",
244
- "name": "AI产品经理实习生",
245
- "company": "百度",
246
- "city": "北京",
247
- "salary": "180-220元/天",
248
- "tags": ["AI产品", "数据分析", "用户调研", "PRD", "AIGC"],
249
- "description": "参与百度文心一言的AI产品能力建设,负责竞品分析、用户需求调研及产品功能定义,推动模型能力的场景化落地。",
250
- "requirements": "2026届本科及以上学历,专业不限;对大模型产品有深度使用经验和独到理解;有数据分析能力(SQL/Python);有AI相关产品实习经验者优先。",
251
- "expectation_signals": [
252
- {"tag": "AI前沿领域", "evidence": "文心一言是百度AI战略的核心产品", "confidence": 0.90},
253
- {"tag": "节奏紧凑", "evidence": "大模型行业竞争白热化,产品迭代频率极高", "confidence": 0.75},
254
- {"tag": "待遇中等", "evidence": "180-220元/天在产品实习中属于一般水平", "confidence": 0.65},
255
- {"tag": "技术理解要求高", "evidence": "AI产品经理需比普通PM更深入的技术理解力", "confidence": 0.80}
256
- ]
257
- },
258
- {
259
- "id": "job_017",
260
- "name": "DevOps开发实习生",
261
- "company": "腾讯",
262
- "city": "深圳",
263
- "salary": "200-250元/天",
264
- "tags": ["Docker", "Kubernetes", "Terraform", "CI/CD", "Python", "Linux"],
265
- "description": "负责腾讯云TKE容器服务的运维工具开发,参与云原生监控告警系统的建设,提升Kubernetes集群的运维自动化水平。",
266
- "requirements": "2026届本科及以上学历,计算机相关专业;熟悉Linux和容器技术;有Kubernetes集群操作经验;有Go或Python开发能力者优先。",
267
- "expectation_signals": [
268
- {"tag": "需on-call值班", "evidence": "云原生基础设施岗位需7x24小时响应机制", "confidence": 0.80},
269
- {"tag": "云原生前沿技术", "evidence": "Kubernetes/Terraform是当前基础设施方向的核心技术栈", "confidence": 0.90},
270
- {"tag": "待遇较高", "evidence": "200-250元/天在运维开发岗位中属于较高水平", "confidence": 0.75},
271
- {"tag": "大规模集群管理", "evidence": "腾讯云管理数百万核的Kubernetes集群", "confidence": 0.90}
272
- ]
273
- },
274
- {
275
- "id": "job_018",
276
- "name": "硬件开发实习生",
277
- "company": "大疆",
278
- "city": "深圳",
279
- "salary": "200-300元/天",
280
- "tags": ["嵌入式", "C", "RTOS", "I2C", "SPI", "PCB设计"],
281
- "description": "参与大疆无人机飞控系统的硬件驱动开发与调优,负责传感器数据采集、电机控制算法的嵌入式实现及硬件调试。",
282
- "requirements": "2026届本科及以上学历,电子/自动化/计算机相关专业;熟悉C语言和嵌入式开发(STM32/FreeRTOS);了解常用通信协议(I2C/SPI/UART);有RoboMaster或电子设计竞赛经历者优先。",
283
- "expectation_signals": [
284
- {"tag": "研发导向重技术", "evidence": "大疆以技术驱动产品创新,研发投入占比很高", "confidence": 0.85},
285
- {"tag": "硬件调试节奏波动大", "evidence": "硬件开发在流片和测试阶段需集中攻关", "confidence": 0.65},
286
- {"tag": "待遇行业较好", "evidence": "200-300元/天在硬件实习中属于上等水平", "confidence": 0.80},
287
- {"tag": "核心技术岗位", "evidence": "飞控系统是无人机最核心的技术壁垒", "confidence": 0.90}
288
- ]
289
- },
290
- {
291
- "id": "job_019",
292
- "name": "算法研究实习生(大模型方向)",
293
- "company": "DeepSeek",
294
- "city": "杭州",
295
- "salary": "300-400元/天",
296
- "tags": ["PyTorch", "Transformer", "CUDA", "分布式训练", "模型压缩", "RLHF"],
297
- "description": "参与大语言模型的预训练与对齐算法研究,负责模型训练流程优化、RLHF算法实现及推理加速,探索下一代模型架构。",
298
- "requirements": "2026届硕士及以上学历(博士优先),人工智能/计算机相关专业;有AI顶会论文(NeurIPS/ICML/ACL等);精通PyTorch及分布式训练框架;有LLM训练经验者优先。",
299
- "expectation_signals": [
300
- {"tag": "学术导向极强", "evidence": "该岗位要求顶会论文,以发表研究成果为主要目标", "confidence": 0.90},
301
- {"tag": "研究氛围浓厚", "evidence": "DeepSeek团队以年轻研究员为主,学术交流频繁", "confidence": 0.85},
302
- {"tag": "待遇行业顶尖", "evidence": "300-400元/天是所有实习岗位中的天花板水平", "confidence": 0.90},
303
- {"tag": "算力资源充足", "evidence": "拥有大规模GPU集群,支持前沿研究探索", "confidence": 0.95}
304
- ]
305
- },
306
- {
307
- "id": "job_020",
308
- "name": "技术产品运营实习生",
309
- "company": "小红书",
310
- "city": "上海",
311
- "salary": "180-220元/天",
312
- "tags": ["用户增长", "AB实验", "数据分析", "SQL", "内容运营"],
313
- "description": "负责小红书搜索产品的用户增长策略制定与落地,通过数据分析和AB实验优化搜索引导和推荐策略,提升搜索渗透率。",
314
- "requirements": "2026届本科及以上学历,专业不限;具备SQL数据提取和分析能力;有产品运营或用户增长实习经验者优先;小红书重度用户加分。",
315
- "expectation_signals": [
316
- {"tag": "数据驱动要求高", "evidence": "该岗位以AB实验和数据分析为日常核心工作", "confidence": 0.80},
317
- {"tag": "节奏较快", "evidence": "用户增长团队需持续追赶增长目标,节奏紧凑", "confidence": 0.70},
318
- {"tag": "待遇中等", "evidence": "180-220元/天在运营岗中属于一般水平", "confidence": 0.65},
319
- {"tag": "核心业务方向", "evidence": "搜索是小红书用户获取信息的主要入口,战略地位高", "confidence": 0.85}
320
- ]
321
- },
322
- {
323
- "id": "job_021",
324
- "name": "产品经理实习生",
325
- "company": "腾讯",
326
- "city": "深圳",
327
- "salary": "180-220元/天",
328
- "tags": ["用户调研", "数据分析", "PRD", "Axure", "需求分析"],
329
- "description": "负责腾讯云产品线的需求收集与PRD输出,协调前后端与设计团队推进需求落地,通过数据埋点验证产品迭代效果。",
330
- "requirements": "2026届本科及以上学历,专业不限;具备良好的逻辑能力和沟通协调能力;有互联网产品实习经验者优先;有技术背景或ToB产品经验者加分。",
331
- "expectation_signals": [
332
- {"tag": "985/211院校偏好", "evidence": "腾讯产品岗简历池竞争激烈,HR会优先筛选985/211院校学生", "confidence": 0.85},
333
- {"tag": "实习经历硬门槛", "evidence": "产品岗几乎不招收零实习经历的学生,至少有一段互联网公司实习", "confidence": 0.90},
334
- {"tag": "行业经验匹配度高", "evidence": "有云服务或ToB产品实习经验的候选人会优先进入面试", "confidence": 0.80},
335
- {"tag": "逻辑表达能力要求高", "evidence": "PRD质量直接反映候选人的逻辑思维和书面表达能力", "confidence": 0.85}
336
- ]
337
- },
338
- {
339
- "id": "job_022",
340
- "name": "内容运营实习生",
341
- "company": "字节跳动",
342
- "city": "北京",
343
- "salary": "150-200元/天",
344
- "tags": ["内容策划", "社群运营", "数据分析", "PS", "文案撰写"],
345
- "description": "负责抖音生活服务内容运营,策划话题活动与达人合作方案,通过内容数据分析和用户反馈持续优化运营策略。",
346
- "requirements": "2026届本科及以上学历,新闻传播/中文/市场营销相关专业优先;具备优秀的文案能力和内容敏感度;熟练使用PS/剪映等工具;有自媒体或内容运营经验者优先。",
347
- "expectation_signals": [
348
- {"tag": "专业背景偏好", "evidence": "HR会优先看新闻传播/中文/市场营销等文科专业简历", "confidence": 0.75},
349
- {"tag": "作品集是加分项", "evidence": "有个人公众号/小红书账号或短视频作品集的候选人更受青睐", "confidence": 0.85},
350
- {"tag": "工具技能基础门槛", "evidence": "PS/剪映等工具是内容运营的基础能力,HR会留意技能关键词", "confidence": 0.80},
351
- {"tag": "平台熟悉度隐性要求", "evidence": "重度抖音用户更容易理解平台生态和内容调性", "confidence": 0.75}
352
- ]
353
- },
354
- {
355
- "id": "job_023",
356
- "name": "品牌营销实习生",
357
- "company": "宝洁",
358
- "city": "广州",
359
- "salary": "180-220元/天",
360
- "tags": ["品牌策划", "市场调研", "活动执行", "PPT", "数据整理"],
361
- "description": "参与宝洁大中华区洗护品牌的营销战役策划,协助完成竞品分析、消费者洞察报告以及线下推广活动的落地执行与效果复盘。",
362
- "requirements": "2026届本科及以上学历,市场营销/广告/商科相关专业优先;出色的PPT制作与表达能力;有快消或广告公司实习经历者优先;英语读写流利。",
363
- "expectation_signals": [
364
- {"tag": "英语能力硬指标", "evidence": "外企HR会重点看四六级/雅思成绩,英语能力是简历初筛红线", "confidence": 0.90},
365
- {"tag": "商科专业背景偏好", "evidence": "市场营销和商科专业出身的简历通过率远高于其他专业", "confidence": 0.80},
366
- {"tag": "实习经历相关度", "evidence": "有快消或4A广告公司实习经历是简历亮点,HR重点关注", "confidence": 0.85},
367
- {"tag": "逻辑表达与数据分析并重", "evidence": "快消营销越来越依赖数据驱动,会数据分析的候选人更有竞争力", "confidence": 0.75}
368
- ]
369
- },
370
- {
371
- "id": "job_024",
372
- "name": "UI/UX设计实习生",
373
- "company": "网易",
374
- "city": "杭州",
375
- "salary": "180-220元/天",
376
- "tags": ["Figma", "Sketch", "设计系统", "用户研究", "交互设计"],
377
- "description": "负责网易云音乐产品线的界面设计与交互优化,参与设计系统组件维护,基于用户研究数据输出高保真原型与设计规范文档。",
378
- "requirements": "2026届本科及以上学历,设计/艺术/人机交互相关专业;精通Figma或Sketch等设计工具;了解设计系统和工作流规范;有完整上线作品者优先。",
379
- "expectation_signals": [
380
- {"tag": "作品集是必须项", "evidence": "设计岗HR必看作品集,无作品集直接淘汰,作品集质量决定初筛结果", "confidence": 0.95},
381
- {"tag": "设计相关专业要求", "evidence": "设计/艺术/人机交互专业为HR筛选时的偏好方向", "confidence": 0.80},
382
- {"tag": "设计系统理解力", "evidence": "有大厂设计系统参与经验的候选人会被高亮标记", "confidence": 0.80},
383
- {"tag": "工具掌握是基础分", "evidence": "Figma已是设计行业标配,HR会从技能关键词判断基础能力", "confidence": 0.85}
384
- ]
385
- },
386
- {
387
- "id": "job_025",
388
- "name": "商业分析实习生",
389
- "company": "美团",
390
- "city": "北京",
391
- "salary": "200-250元/天",
392
- "tags": ["SQL", "Python", "商业建模", "行业研究", "数据可视化"],
393
- "description": "负责美团到店事业群的商业分析工作,搭建业务指标体系并产出周报/月报,通过行业对标和用户分层分析驱动业务策略调整。",
394
- "requirements": "2026届本科及以上学历,统计学/经济学/金融/计算机相关专业;精通SQL和Excel;掌握Python或R进行数据处理和分析;有咨询公司或互联网商业分析实习经历者优先。",
395
- "expectation_signals": [
396
- {"tag": "量化背景是硬门槛", "evidence": "HR会优先筛选统计学/经济学等量化背景专业的学生", "confidence": 0.85},
397
- {"tag": "SQL能力是下限要求", "evidence": "商业分析岗SQL是必选项,无SQL能力的简历直接淘汰", "confidence": 0.90},
398
- {"tag": "咨询/行研实习加分", "evidence": "有MBB或四大咨询实习经历的候选人会被优先考虑", "confidence": 0.80},
399
- {"tag": "数据分析案例亮点", "evidence": "简历中如有独立完成的商业分析项目案例会是关键加分项", "confidence": 0.85}
400
- ]
401
- }
402
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/models/config.json DELETED
@@ -1,32 +0,0 @@
1
- {
2
- "_name_or_path": "BAAI/bge-small-zh-v1.5",
3
- "architectures": [
4
- "BertModel"
5
- ],
6
- "attention_probs_dropout_prob": 0.1,
7
- "classifier_dropout": null,
8
- "gradient_checkpointing": false,
9
- "hidden_act": "gelu",
10
- "hidden_dropout_prob": 0.1,
11
- "hidden_size": 512,
12
- "id2label": {
13
- "0": "LABEL_0"
14
- },
15
- "initializer_range": 0.02,
16
- "intermediate_size": 2048,
17
- "label2id": {
18
- "LABEL_0": 0
19
- },
20
- "layer_norm_eps": 1e-12,
21
- "max_position_embeddings": 512,
22
- "model_type": "bert",
23
- "num_attention_heads": 8,
24
- "num_hidden_layers": 4,
25
- "pad_token_id": 0,
26
- "position_embedding_type": "absolute",
27
- "torch_dtype": "float32",
28
- "transformers_version": "4.33.2",
29
- "type_vocab_size": 2,
30
- "use_cache": true,
31
- "vocab_size": 21128
32
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/models/model_optimized.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:1294ea4b6331115a353d81f96b85e8c8d7fdcc284453d5b2fab5b016230aad38
3
- size 94781076
 
 
 
 
offer-puplayer-hf/backend/models/special_tokens_map.json DELETED
@@ -1,7 +0,0 @@
1
- {
2
- "cls_token": "[CLS]",
3
- "mask_token": "[MASK]",
4
- "pad_token": "[PAD]",
5
- "sep_token": "[SEP]",
6
- "unk_token": "[UNK]"
7
- }
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/models/tokenizer.json DELETED
The diff for this file is too large to render. See raw diff
 
offer-puplayer-hf/backend/models/tokenizer_config.json DELETED
@@ -1,15 +0,0 @@
1
- {
2
- "clean_up_tokenization_spaces": true,
3
- "cls_token": "[CLS]",
4
- "do_basic_tokenize": true,
5
- "do_lower_case": false,
6
- "mask_token": "[MASK]",
7
- "model_max_length": 512,
8
- "never_split": null,
9
- "pad_token": "[PAD]",
10
- "sep_token": "[SEP]",
11
- "strip_accents": null,
12
- "tokenize_chinese_chars": true,
13
- "tokenizer_class": "BertTokenizer",
14
- "unk_token": "[UNK]"
15
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/backend/requirements.txt DELETED
@@ -1,8 +0,0 @@
1
- fastapi
2
- uvicorn
3
- pydantic
4
- python-multipart
5
- pdfplumber
6
- httpx
7
- python-dotenv
8
- fastembed
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/index.html DELETED
@@ -1,13 +0,0 @@
1
- <!doctype html>
2
- <html lang="zh-CN">
3
- <head>
4
- <meta charset="UTF-8" />
5
- <link rel="icon" type="image/svg+xml" href="/favicon.svg" />
6
- <meta name="viewport" content="width=device-width, initial-scale=1.0" />
7
- <title>Offer 捕手 — 求职匹配助手</title>
8
- </head>
9
- <body>
10
- <div id="app"></div>
11
- <script type="module" src="/src/main.jsx"></script>
12
- </body>
13
- </html>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/package.json DELETED
@@ -1,24 +0,0 @@
1
- {
2
- "name": "offer_puplayer_frontend",
3
- "private": true,
4
- "version": "0.0.0",
5
- "type": "module",
6
- "scripts": {
7
- "dev": "vite",
8
- "build": "vite build",
9
- "preview": "vite preview"
10
- },
11
- "devDependencies": {
12
- "@vitejs/plugin-react": "^6.0.2",
13
- "typescript": "~6.0.2",
14
- "vite": "^8.0.12"
15
- },
16
- "dependencies": {
17
- "axios": "^1.17.0",
18
- "echarts": "^6.1.0",
19
- "echarts-for-react": "^3.0.6",
20
- "react": "^19.2.7",
21
- "react-dom": "^19.2.7",
22
- "tslib": "^2.8.1"
23
- }
24
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/public/favicon.svg DELETED
offer-puplayer-hf/frontend/public/icons.svg DELETED
offer-puplayer-hf/frontend/src/App.css DELETED
@@ -1,535 +0,0 @@
1
- :root {
2
- --color-primary: #4F46E5;
3
- --color-primary-light: #6366F1;
4
- --color-primary-dark: #3730A3;
5
- --color-success: #10B981;
6
- --color-warning: #F59E0B;
7
- --color-error: #EF4444;
8
- --color-bg: #F3F4F6;
9
- --color-bg-white: #FFFFFF;
10
- --color-text: #1F2937;
11
- --color-text-secondary: #6B7280;
12
- --color-text-light: #9CA3AF;
13
- --color-border: #E5E7EB;
14
- --color-border-light: #F3F4F6;
15
- --shadow-sm: 0 1px 2px rgba(0, 0, 0, 0.05);
16
- --shadow-md: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -2px rgba(0, 0, 0, 0.1);
17
- --shadow-lg: 0 10px 15px -3px rgba(0, 0, 0, 0.1), 0 4px 6px -4px rgba(0, 0, 0, 0.1);
18
- --radius-sm: 6px;
19
- --radius-md: 8px;
20
- --radius-lg: 12px;
21
- --font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, 'Helvetica Neue', Arial, sans-serif;
22
- }
23
-
24
- *,
25
- *::before,
26
- *::after {
27
- margin: 0;
28
- padding: 0;
29
- box-sizing: border-box;
30
- }
31
-
32
- html {
33
- font-size: 16px;
34
- -webkit-font-smoothing: antialiased;
35
- -moz-osx-font-smoothing: grayscale;
36
- }
37
-
38
- body {
39
- font-family: var(--font-family);
40
- font-size: 14px;
41
- line-height: 1.6;
42
- color: var(--color-text);
43
- background-color: var(--color-bg);
44
- }
45
-
46
- #app {
47
- min-height: 100vh;
48
- }
49
-
50
- .app {
51
- min-height: 100vh;
52
- display: flex;
53
- flex-direction: column;
54
- }
55
-
56
- .nav-bar {
57
- background-color: var(--color-bg-white);
58
- border-bottom: 1px solid var(--color-border);
59
- box-shadow: var(--shadow-sm);
60
- position: sticky;
61
- top: 0;
62
- z-index: 100;
63
- }
64
-
65
- .nav-inner {
66
- max-width: 1200px;
67
- margin: 0 auto;
68
- padding: 0 24px;
69
- height: 56px;
70
- display: flex;
71
- align-items: center;
72
- justify-content: space-between;
73
- }
74
-
75
- .nav-logo {
76
- font-size: 18px;
77
- font-weight: 700;
78
- color: var(--color-primary);
79
- cursor: pointer;
80
- user-select: none;
81
- }
82
-
83
- .nav-links {
84
- display: flex;
85
- align-items: center;
86
- gap: 8px;
87
- }
88
-
89
- .nav-link {
90
- padding: 6px 16px;
91
- font-size: 14px;
92
- font-weight: 500;
93
- color: var(--color-text-secondary);
94
- cursor: pointer;
95
- border-radius: var(--radius-sm);
96
- transition: color 0.2s, background-color 0.2s;
97
- user-select: none;
98
- }
99
-
100
- .nav-link:hover {
101
- color: var(--color-primary);
102
- background-color: rgba(79, 70, 229, 0.06);
103
- }
104
-
105
- .nav-link.active {
106
- color: var(--color-primary);
107
- background-color: rgba(79, 70, 229, 0.1);
108
- font-weight: 600;
109
- }
110
-
111
- .page-container {
112
- flex: 1;
113
- max-width: 1200px;
114
- width: 100%;
115
- margin: 0 auto;
116
- padding: 32px 24px;
117
- }
118
-
119
- .flex {
120
- display: flex;
121
- }
122
-
123
- .flex-col {
124
- flex-direction: column;
125
- }
126
-
127
- .flex-center {
128
- display: flex;
129
- align-items: center;
130
- justify-content: center;
131
- }
132
-
133
- .flex-between {
134
- display: flex;
135
- align-items: center;
136
- justify-content: space-between;
137
- }
138
-
139
- .flex-wrap {
140
- flex-wrap: wrap;
141
- }
142
-
143
- .gap-8 {
144
- gap: 8px;
145
- }
146
-
147
- .gap-12 {
148
- gap: 12px;
149
- }
150
-
151
- .gap-16 {
152
- gap: 16px;
153
- }
154
-
155
- .gap-24 {
156
- gap: 24px;
157
- }
158
-
159
- .grid-2 {
160
- display: grid;
161
- grid-template-columns: repeat(2, 1fr);
162
- gap: 24px;
163
- }
164
-
165
- .grid-3 {
166
- display: grid;
167
- grid-template-columns: repeat(3, 1fr);
168
- gap: 24px;
169
- }
170
-
171
- .btn {
172
- display: inline-flex;
173
- align-items: center;
174
- justify-content: center;
175
- gap: 8px;
176
- padding: 10px 20px;
177
- font-size: 14px;
178
- font-weight: 500;
179
- line-height: 1;
180
- border: 1px solid transparent;
181
- border-radius: var(--radius-sm);
182
- cursor: pointer;
183
- transition: all 0.2s;
184
- user-select: none;
185
- white-space: nowrap;
186
- }
187
-
188
- .btn:disabled {
189
- opacity: 0.5;
190
- cursor: not-allowed;
191
- }
192
-
193
- .btn-primary {
194
- background-color: var(--color-primary);
195
- color: #FFFFFF;
196
- border-color: var(--color-primary);
197
- }
198
-
199
- .btn-primary:hover:not(:disabled) {
200
- background-color: var(--color-primary-dark);
201
- border-color: var(--color-primary-dark);
202
- }
203
-
204
- .btn-secondary {
205
- background-color: var(--color-bg-white);
206
- color: var(--color-text);
207
- border-color: var(--color-border);
208
- }
209
-
210
- .btn-secondary:hover:not(:disabled) {
211
- background-color: var(--color-border-light);
212
- border-color: var(--color-border);
213
- }
214
-
215
- .btn-outline {
216
- background-color: transparent;
217
- color: var(--color-primary);
218
- border-color: var(--color-primary);
219
- }
220
-
221
- .btn-outline:hover:not(:disabled) {
222
- background-color: rgba(79, 70, 229, 0.06);
223
- }
224
-
225
- .btn-sm {
226
- padding: 6px 12px;
227
- font-size: 13px;
228
- }
229
-
230
- .btn-lg {
231
- padding: 14px 28px;
232
- font-size: 16px;
233
- }
234
-
235
- .card {
236
- background-color: var(--color-bg-white);
237
- border: 1px solid var(--color-border);
238
- border-radius: var(--radius-lg);
239
- box-shadow: var(--shadow-sm);
240
- padding: 24px;
241
- transition: box-shadow 0.2s;
242
- }
243
-
244
- .card:hover {
245
- box-shadow: var(--shadow-md);
246
- }
247
-
248
- .text-primary {
249
- color: var(--color-primary);
250
- }
251
-
252
- .text-success {
253
- color: var(--color-success);
254
- }
255
-
256
- .text-warning {
257
- color: var(--color-warning);
258
- }
259
-
260
- .text-error {
261
- color: var(--color-error);
262
- }
263
-
264
- .text-secondary {
265
- color: var(--color-text-secondary);
266
- }
267
-
268
- .text-light {
269
- color: var(--color-text-light);
270
- }
271
-
272
- .text-center {
273
- text-align: center;
274
- }
275
-
276
- .text-sm {
277
- font-size: 13px;
278
- }
279
-
280
- .text-lg {
281
- font-size: 16px;
282
- }
283
-
284
- .text-xl {
285
- font-size: 20px;
286
- }
287
-
288
- .font-bold {
289
- font-weight: 600;
290
- }
291
-
292
- .mt-8 {
293
- margin-top: 8px;
294
- }
295
-
296
- .mt-16 {
297
- margin-top: 16px;
298
- }
299
-
300
- .mt-24 {
301
- margin-top: 24px;
302
- }
303
-
304
- .mb-8 {
305
- margin-bottom: 8px;
306
- }
307
-
308
- .mb-16 {
309
- margin-bottom: 16px;
310
- }
311
-
312
- .mb-24 {
313
- margin-bottom: 24px;
314
- }
315
-
316
- .empty-state {
317
- display: flex;
318
- flex-direction: column;
319
- align-items: center;
320
- justify-content: center;
321
- padding: 64px 24px;
322
- text-align: center;
323
- }
324
-
325
- .empty-state-icon {
326
- font-size: 48px;
327
- margin-bottom: 16px;
328
- color: var(--color-text-light);
329
- }
330
-
331
- .empty-state-text {
332
- font-size: 16px;
333
- color: var(--color-text-secondary);
334
- margin-bottom: 24px;
335
- }
336
-
337
- .error-state {
338
- display: flex;
339
- flex-direction: column;
340
- align-items: center;
341
- justify-content: center;
342
- padding: 48px 24px;
343
- text-align: center;
344
- }
345
-
346
- .error-state-text {
347
- font-size: 15px;
348
- color: var(--color-error);
349
- margin-bottom: 16px;
350
- }
351
-
352
- .loading-state {
353
- display: flex;
354
- flex-direction: column;
355
- align-items: center;
356
- justify-content: center;
357
- padding: 48px 24px;
358
- }
359
-
360
- @keyframes spin {
361
- from {
362
- transform: rotate(0deg);
363
- }
364
- to {
365
- transform: rotate(360deg);
366
- }
367
- }
368
-
369
- @keyframes pulse {
370
- 0%, 100% {
371
- opacity: 1;
372
- }
373
- 50% {
374
- opacity: 0.5;
375
- }
376
- }
377
-
378
- @keyframes shimmer {
379
- 0% {
380
- background-position: -200% 0;
381
- }
382
- 100% {
383
- background-position: 200% 0;
384
- }
385
- }
386
-
387
- .loading-spinner {
388
- width: 32px;
389
- height: 32px;
390
- border: 3px solid var(--color-border);
391
- border-top-color: var(--color-primary);
392
- border-radius: 50%;
393
- animation: spin 0.8s linear infinite;
394
- }
395
-
396
- .loading-spinner-sm {
397
- width: 20px;
398
- height: 20px;
399
- border-width: 2px;
400
- }
401
-
402
- .loading-spinner-lg {
403
- width: 48px;
404
- height: 48px;
405
- border-width: 4px;
406
- }
407
-
408
- .skeleton {
409
- background: linear-gradient(90deg, var(--color-border-light) 25%, var(--color-border) 50%, var(--color-border-light) 75%);
410
- background-size: 200% 100%;
411
- animation: shimmer 1.5s ease-in-out infinite;
412
- border-radius: var(--radius-sm);
413
- }
414
-
415
- .pipeline-step-running {
416
- animation: spin 1s linear infinite;
417
- }
418
-
419
- .pipeline-step-pulse {
420
- animation: pulse 1.5s ease-in-out infinite;
421
- }
422
-
423
- /* fix: 补充审查缺失的CSS类名 -- BEGIN */
424
- .tag {
425
- display: inline-block;
426
- padding: 2px 8px;
427
- font-size: 12px;
428
- border-radius: var(--radius-sm);
429
- background-color: rgba(79, 70, 229, 0.08);
430
- color: var(--color-primary);
431
- }
432
-
433
- .input {
434
- width: 100%;
435
- padding: 10px 12px;
436
- font-size: 14px;
437
- border: 1px solid var(--color-border);
438
- border-radius: var(--radius-sm);
439
- outline: none;
440
- transition: border-color 0.2s;
441
- font-family: inherit;
442
- }
443
-
444
- .input:focus {
445
- border-color: var(--color-primary);
446
- }
447
-
448
- .pipeline-node {
449
- width: 40px;
450
- height: 40px;
451
- border-radius: 50%;
452
- display: flex;
453
- align-items: center;
454
- justify-content: center;
455
- font-size: 16px;
456
- font-weight: 600;
457
- border: 2px solid var(--color-border);
458
- color: var(--color-text-light);
459
- background-color: var(--color-bg-white);
460
- transition: all 0.3s;
461
- }
462
-
463
- .pipeline-node-idle {
464
- border-color: var(--color-border);
465
- color: var(--color-text-light);
466
- }
467
-
468
- .pipeline-node-running {
469
- border-color: var(--color-primary);
470
- color: var(--color-primary);
471
- animation: spin 1s linear infinite;
472
- }
473
-
474
- .pipeline-node-completed {
475
- border-color: var(--color-success);
476
- color: var(--color-success);
477
- background-color: rgba(16, 185, 129, 0.08);
478
- }
479
-
480
- .pipeline-node-error {
481
- border-color: var(--color-error);
482
- color: var(--color-error);
483
- background-color: rgba(239, 68, 68, 0.08);
484
- }
485
-
486
- .pipeline-connector {
487
- width: 60px;
488
- height: 2px;
489
- transition: background-color 0.3s;
490
- }
491
-
492
- .pipeline-connector-completed {
493
- background-color: var(--color-success);
494
- }
495
-
496
- .pipeline-connector-idle {
497
- background-color: var(--color-border);
498
- }
499
-
500
- .pipeline-label-idle {
501
- color: var(--color-text-light);
502
- }
503
-
504
- .pipeline-label-running {
505
- color: var(--color-primary);
506
- }
507
-
508
- .pipeline-label-completed {
509
- color: var(--color-success);
510
- }
511
-
512
- .pipeline-label-error {
513
- color: var(--color-error);
514
- }
515
- /* fix: 补充审查缺失的CSS类名 -- END */
516
-
517
- @media (min-width: 768px) {
518
- .page-container {
519
- padding: 40px 32px;
520
- }
521
- }
522
-
523
- @media (min-width: 1024px) {
524
- .page-container {
525
- padding: 48px 32px;
526
- }
527
-
528
- .grid-2 {
529
- gap: 32px;
530
- }
531
-
532
- .grid-3 {
533
- gap: 32px;
534
- }
535
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/App.jsx DELETED
@@ -1,122 +0,0 @@
1
- import { useState, useEffect } from 'react'
2
- import { confirmResume } from './api'
3
- import HomePage from './pages/HomePage'
4
- import RecommendPage from './pages/RecommendPage'
5
- import DiagnosePage from './pages/DiagnosePage'
6
- import SuggestPage from './pages/SuggestPage'
7
-
8
- function App() {
9
- const [currentPage, setCurrentPage] = useState('recommend')
10
- const [resumeData, setResumeData] = useState(null)
11
- const [selectedJobId, setSelectedJobId] = useState(null)
12
- const [selectedJob, setSelectedJob] = useState(null)
13
- const [advice, setAdvice] = useState(null)
14
- const [pendingJobForDiagnose, setPendingJobForDiagnose] = useState(null)
15
-
16
- useEffect(() => {
17
- const saved = localStorage.getItem('offer_resume_data')
18
- if (saved) {
19
- try { setResumeData(JSON.parse(saved)) } catch {}
20
- }
21
- }, [])
22
-
23
- useEffect(() => {
24
- if (resumeData) {
25
- localStorage.setItem('offer_resume_data', JSON.stringify(resumeData))
26
- }
27
- }, [resumeData])
28
-
29
- const goToHome = () => {
30
- setCurrentPage('home')
31
- }
32
-
33
- const goToRecommend = () => {
34
- setCurrentPage('recommend')
35
- }
36
-
37
- const goToDiagnose = (jobId, jobData) => {
38
- if (!jobId) return
39
- setSelectedJobId(jobId)
40
- setSelectedJob(jobData || null)
41
- setCurrentPage('diagnose')
42
- }
43
-
44
- const goToSuggest = (payload) => {
45
- if (payload?.advice) setAdvice(payload.advice)
46
- setCurrentPage('suggest')
47
- }
48
-
49
- const handleNavDiagnose = () => {
50
- if (!selectedJobId) return
51
- setCurrentPage('diagnose')
52
- }
53
-
54
- const handleConfirmResume = async (data) => {
55
- let resolved = data
56
- try {
57
- const res = await confirmResume(data)
58
- if (res.data?.data) {
59
- resolved = res.data.data
60
- }
61
- } catch {}
62
- setResumeData(resolved)
63
- if (pendingJobForDiagnose) {
64
- const { jobId, jobData } = pendingJobForDiagnose
65
- setPendingJobForDiagnose(null)
66
- setSelectedJobId(jobId)
67
- setSelectedJob(jobData || null)
68
- setCurrentPage('diagnose')
69
- } else {
70
- setCurrentPage('recommend')
71
- }
72
- }
73
-
74
- const goHomeFromDiagnose = () => {
75
- if (selectedJobId && !resumeData) {
76
- setPendingJobForDiagnose({ jobId: selectedJobId, jobData: selectedJob })
77
- }
78
- setCurrentPage('home')
79
- }
80
-
81
- const renderNav = () => (
82
- <nav className="nav-bar">
83
- <div className="nav-inner">
84
- <span className="nav-logo" onClick={goToRecommend}>Offer 捕手</span>
85
- <div className="nav-links">
86
- <span className={`nav-link${currentPage === 'home' ? ' active' : ''}`} onClick={goToHome}>个人中心</span>
87
- <span className={`nav-link${currentPage === 'recommend' ? ' active' : ''}`} onClick={goToRecommend}>推荐岗位</span>
88
- <span className={`nav-link${currentPage === 'diagnose' ? ' active' : ''}`} onClick={handleNavDiagnose}>匹配诊断</span>
89
- <span className={`nav-link${currentPage === 'suggest' ? ' active' : ''}`} onClick={() => goToSuggest()}>改进建议</span>
90
- </div>
91
- </div>
92
- </nav>
93
- )
94
-
95
- const renderPage = () => {
96
- switch (currentPage) {
97
- case 'home':
98
- return <HomePage resumeData={resumeData} onConfirm={handleConfirmResume} onGoRecommend={goToRecommend} />
99
- case 'recommend':
100
- return <RecommendPage
101
- resumeData={resumeData}
102
- onSelectJob={(jobId, jobData) => goToDiagnose(jobId, jobData)}
103
- onGoHome={goToHome}
104
- />
105
- case 'diagnose':
106
- return <DiagnosePage jobId={selectedJobId} jobData={selectedJob} resumeData={resumeData} onComplete={(payload) => goToSuggest(payload)} onBack={() => setCurrentPage('recommend')} onGoHome={goHomeFromDiagnose} />
107
- case 'suggest':
108
- return <SuggestPage jobId={selectedJobId} resumeData={resumeData} advice={advice} onBack={() => setCurrentPage('diagnose')} />
109
- default:
110
- return <RecommendPage resumeData={resumeData} onSelectJob={(jobId, jobData) => goToDiagnose(jobId, jobData)} />
111
- }
112
- }
113
-
114
- return (
115
- <div className="app">
116
- {renderNav()}
117
- <main className="page-container">{renderPage()}</main>
118
- </div>
119
- )
120
- }
121
-
122
- export default App
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/api/index.js DELETED
@@ -1,130 +0,0 @@
1
- import axios from 'axios'
2
- import { getMockResumeData, getMockJobs, getMockReport, getMockSuggestions } from './mock'
3
-
4
- const USE_MOCK = false
5
-
6
- function delay(ms) {
7
- return new Promise(resolve => setTimeout(resolve, ms))
8
- }
9
-
10
- const api = axios.create({
11
- baseURL: '',
12
- timeout: 10000,
13
- })
14
-
15
- api.interceptors.response.use(
16
- (response) => response,
17
- (error) => {
18
- if (error.response) {
19
- const status = error.response.status
20
- if (status >= 500) {
21
- console.error('服务器错误,请稍后重试')
22
- } else if (status >= 400) {
23
- console.error('请求错误,请检查输入')
24
- }
25
- } else if (error.request) {
26
- console.error('网络连接失败,请检查网络')
27
- }
28
- return Promise.reject(error)
29
- }
30
- )
31
-
32
- export async function uploadResume(file) {
33
- if (USE_MOCK) {
34
- await delay(800)
35
- return getMockResumeData()
36
- }
37
- const formData = new FormData()
38
- formData.append('file', file)
39
- return api.post('/api/resume/upload', formData, {
40
- headers: { 'Content-Type': 'multipart/form-data' },
41
- })
42
- }
43
-
44
- export async function getRecommendJobs(resumeId) {
45
- if (USE_MOCK) {
46
- await delay(500)
47
- return getMockJobs()
48
- }
49
- const res = await api.get('/api/jobs/recommend', { params: { resume_id: resumeId } })
50
- const list = (res.data?.jobs || []).map(job => ({
51
- ...job,
52
- match_rate: Math.round((job.match_rate || 0) * 100)
53
- }))
54
- return { data: { jobs: list } }
55
- }
56
-
57
- export async function generateReport(jobId, resumeData) {
58
- if (USE_MOCK) {
59
- const encoder = new TextEncoder()
60
- const report = getMockReport()
61
- const suggestions = getMockSuggestions()
62
- const stream = new ReadableStream({
63
- async start(controller) {
64
- const events = [
65
- { type: 'agent_start', agent: 'agent1' },
66
- { type: 'agent_complete', agent: 'agent1', data: { must_have: ['Python', 'Redis', '分布式系统'], nice_to_have: ['Go', '微服务'], position_summary: '后端开发实习生 - 腾讯' } },
67
- { type: 'agent_start', agent: 'agent2' },
68
- { type: 'agent_complete', agent: 'agent2', data: { dimensions: report.dimensions, total_score: report.total_score, strength: report.strength, gap: report.gap, hr_view: report.hr_view } },
69
- { type: 'agent_start', agent: 'agent3' },
70
- { type: 'agent_complete', agent: 'agent3', data: suggestions },
71
- { type: 'report_complete', data: { jd_parsed: { must_have: ['Python', 'Redis', '分布式系统'], nice_to_have: ['Go', '微服务'], position_summary: '后端开发实习生 - 腾讯' }, match_report: { dimensions: report.dimensions, total_score: report.total_score, strength: report.strength, gap: report.gap, hr_view: report.hr_view }, advice: suggestions } },
72
- ]
73
- for (const event of events) {
74
- await delay(600)
75
- const line = JSON.stringify(event) + '\n'
76
- controller.enqueue(encoder.encode(line))
77
- }
78
- controller.close()
79
- }
80
- })
81
- return { ok: true, body: { getReader: () => stream.getReader() } }
82
- }
83
- const body = { job_id: jobId }
84
- if (resumeData) {
85
- const normalizeSkills = (s) => typeof s === 'string' ? s : (s.name || s)
86
- const normalizeProjects = (p) => typeof p === 'string' ? p : (p.name || JSON.stringify(p))
87
- const normalizeString = (v) => {
88
- if (typeof v === 'string') return v
89
- if (v && typeof v === 'object') return v.school || v.name || JSON.stringify(v)
90
- return ''
91
- }
92
- body.resume_data = {
93
- name: resumeData.name || '',
94
- school: normalizeString(resumeData.school),
95
- major: normalizeString(resumeData.major),
96
- grade: normalizeString(resumeData.grade),
97
- skills: (resumeData.skills || []).map(normalizeSkills),
98
- projects: (resumeData.projects || []).map(normalizeProjects),
99
- summary: resumeData.summary || '',
100
- experience: resumeData.experience || '',
101
- education: normalizeString(resumeData.education),
102
- }
103
- }
104
- return fetch('/api/report/generate', {
105
- method: 'POST',
106
- headers: { 'Content-Type': 'application/json' },
107
- body: JSON.stringify(body),
108
- })
109
- }
110
-
111
- export async function rerankReport(jobId, editedResume) {
112
- if (USE_MOCK) {
113
- await delay(800)
114
- return { data: { match_rate: Math.floor(Math.random() * 30) + 60, previous_rate: 62 } }
115
- }
116
- return api.post('/api/report/rerank', {
117
- job_id: jobId,
118
- edited_resume: editedResume,
119
- })
120
- }
121
-
122
- export async function confirmResume(resumeData) {
123
- if (USE_MOCK) {
124
- await delay(300)
125
- return { data: { success: true, data: resumeData } }
126
- }
127
- return api.post('/api/resume/confirm', resumeData)
128
- }
129
-
130
- export default api
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/api/mock.js DELETED
@@ -1,73 +0,0 @@
1
- const mockJobs = [
2
- { job_id: 1, name: '后端开发实习生', company: '腾讯', city: '深圳', salary: '200-250/天', match_rate: 85, tags: ['Python', 'Go', '后端'], description: '负责后端服务开发和维护,参与微服务架构设计', requirements: '熟悉Python/Go,了解分布式系统', expectation_signals: [{"tag":"加班强度大","evidence":"腾讯核心部门普遍加班,尤其是云业务线交付压力大","confidence":0.75},{"tag":"技术氛围浓厚","evidence":"腾讯以技术驱动著称,内部技术分享和开源文化丰富","confidence":0.85},{"tag":"待遇高于行业平均","evidence":"200-250元/天在实习岗位中属于较高水平","confidence":0.8},{"tag":"核心业务参与","evidence":"腾讯云核心服务直接服务于公司战略级业务","confidence":0.85}] },
3
- { job_id: 2, name: 'AI算法实习生', company: '字节跳动', city: '北京', salary: '250-300/天', match_rate: 72, tags: ['机器学习', 'Python', '深度学习'], description: '参与推荐系统算法优化,提升推荐准确率', requirements: '熟悉机器学习基础,有相关项目经验优先', expectation_signals: [{"tag":"加班强度大","evidence":"字节跳动以快速迭代和高强度工作节奏著称","confidence":0.8},{"tag":"AI前沿方向","evidence":"LLM训练与推理优化处于AI技术最前沿","confidence":0.9},{"tag":"待遇优厚","evidence":"250-300元/天在算法实习岗位中极具竞争力","confidence":0.85},{"tag":"技术驱动文化","evidence":"字节跳动重视技术投入,AI团队资源充足","confidence":0.85}] },
4
- { job_id: 3, name: '前端开发实习生', company: '阿里巴巴', city: '杭州', salary: '200-250/天', match_rate: 68, tags: ['React', 'TypeScript', '前端'], description: '参与中后台系统前端开发', requirements: '熟悉React/Vue,了解前端工程化', expectation_signals: [{"tag":"大促期间加班","evidence":"阿里大促期间前端需随时待命保障页面稳定","confidence":0.7},{"tag":"技术沉淀深厚","evidence":"阿里前端技术生态成熟,具有业界领先的组件库体系","confidence":0.8},{"tag":"待遇中等偏上","evidence":"200-250元/天在前端实习中属于中等偏上","confidence":0.75},{"tag":"高流量业务场景","evidence":"淘宝/天猫商家端服务千万级商家,流量巨大","confidence":0.85}] },
5
- { job_id: 4, name: '产品经理实习生', company: '美团', city: '北京', salary: '180-220/天', match_rate: 55, tags: ['产品设计', '数据分析', '用户研究'], description: '负责B端产品功能设计和迭代', requirements: '有产品实习经验优先,逻辑清晰', expectation_signals: [{"tag":"业务竞争激烈","evidence":"美团所处生活服务赛道竞争激烈,产品推进速度要快","confidence":0.7},{"tag":"数据驱动文化","evidence":"美团以数据驱动运营决策著称,产品岗需强分析能力","confidence":0.85},{"tag":"待遇行业中等","evidence":"180-220元/天在产品实习岗位中属中等水平","confidence":0.7},{"tag":"核心业务板块","evidence":"美团到店/外卖是公司核心业务之一,成长空间大","confidence":0.8}] },
6
- { job_id: 5, name: '数据分析实习生', company: '快手', city: '北京', salary: '200-250/天', match_rate: 78, tags: ['SQL', 'Python', '数据分析'], description: '负责业务数据分析和报表建设', requirements: '熟练使用SQL和Python,有数据分析项目经验', expectation_signals: [{"tag":"工作节奏较快","evidence":"快手增长团队工作节奏快,数据产出时效性要求高","confidence":0.65},{"tag":"数据导向文化","evidence":"快手高度重视数据驱动业务决策,数据岗位话语权强","confidence":0.85},{"tag":"待遇中等偏上","evidence":"200-250元/天在数据分析实习中属于较高水平","confidence":0.75},{"tag":"核心业务指标","evidence":"DAU增长方向直接关系公司核心KPI","confidence":0.85}] },
7
- ]
8
-
9
- const mockReport = {
10
- dimensions: [
11
- { name: '技能匹配', score: 80, evidence: { jd_quote: '熟悉Python', resume_quote: '熟练使用Python进行数据处理和模型开发' } },
12
- { name: '经验匹配', score: 40, evidence: { jd_quote: '3年以上相关经验', resume_quote: '6个月实习经历' } },
13
- { name: '学历匹配', score: 100, evidence: { jd_quote: '计算机相关专业', resume_quote: '计算机科学与技术专业' } },
14
- { name: '项目契合', score: 65, evidence: { jd_quote: '有推荐系统经验优先', resume_quote: '课程项目实现了协同过滤推荐算法' } },
15
- { name: '软性素质', score: 75, evidence: { jd_quote: '具备团队协作能力', resume_quote: '与3名同学协作完成课程项目' } },
16
- ],
17
- total_score: 72,
18
- strength: '技能基础扎实,学历完全符合要求',
19
- gap: '缺乏全职工作经验,项目深度有待提升',
20
- hr_view: {
21
- first_impression: '技能列表与 JD 匹配度高,但项目描述缺乏量化成果,简历整体合格',
22
- attention_points: [
23
- { aspect: '技能栈匹配', finding: 'Python/Flask 完全覆盖 JD 要求', concern: '无' },
24
- { aspect: '项目深度', finding: '仅课程项目,无企业级实践', concern: '项目描述偏功能实现,未见性能优化或架构设计' },
25
- { aspect: '学历背景', finding: '985 计算机专业,符合门槛', concern: '无' },
26
- ],
27
- hr_concern_level: 'medium',
28
- screen_summary: '简历整体合格,项目深度是最大短板,建议补充量化成果',
29
- },
30
- }
31
-
32
- const mockSuggestions = {
33
- rewrite_suggestions: [
34
- { dimension: '项目契合', issue: '项目描述侧重原理,未体现工程落地能力', original: '使用协同过滤算法实现推荐系统', suggestion: '独立设计并实现基于协同过滤的推荐系统,使用Python+Flask搭建Web服务,处理了10000+用户行为数据,推荐准确率达到85%', reason: 'JD偏向工程落地,HR更关注实际产出而非原理' },
35
- { dimension: '技能匹配', issue: 'JD要求Redis但简历未提及', original: '无', suggestion: '在技能列表或项目描述中补充Redis使用经验,如"使用Redis缓存热点数据,查询响应时间降低60%"', reason: 'Redis是JD明确要求的技能,不提及会被直接筛掉' },
36
- ],
37
- interview_tips: [
38
- { topic: 'Redis缓存', reason: 'JD要求熟悉Redis但简历未提及', tip: '准备Redis基础数据结构、缓存穿透/击穿/雪崩、在推荐系统中如何使用缓存' },
39
- { topic: '项目深挖', reason: '项目描述较浅', tip: '准备:项目的技术选型原因、遇到的难点和解决方案、如果重做会怎么改进' },
40
- ],
41
- encouragement: '你的专业背景和技能基础与岗位匹配度不错,只需在项目描述上做针对性调整,命中率会明显提升',
42
- }
43
-
44
- const mockResumeData = {
45
- name: '张三',
46
- school: '华南理工大学',
47
- major: '计算机科学与技术',
48
- grade: '研二',
49
- skills: ['Python', 'Java', 'SQL', 'PyTorch', 'Flask'],
50
- projects: [
51
- { name: '基于协同过滤的推荐系统', desc: '使用协同过滤算法实现电影推荐,处理了MovieLens数据集' },
52
- { name: '在线商城后端开发', desc: '使用Spring Boot实现RESTful API,包括用户管理、商品管理模块' },
53
- ],
54
- experience: '在字节跳动有过6个月后端开发实习经历,参与推荐系统数据管道维护',
55
- education: { school: '华南理工大学', major: '计算机科学与技术', degree: '硕士', startYear: '2023', endYear: '2025' },
56
- summary: '技术栈集中在Python/Java后端开发,有推荐系统和Web开发项目经验,适合投递后端开发/算法类岗位',
57
- }
58
-
59
- export function getMockResumeData() {
60
- return { data: mockResumeData }
61
- }
62
-
63
- export function getMockJobs() {
64
- return { data: { jobs: mockJobs } }
65
- }
66
-
67
- export function getMockReport() {
68
- return mockReport
69
- }
70
-
71
- export function getMockSuggestions() {
72
- return mockSuggestions
73
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/components/JobCard.jsx DELETED
@@ -1,189 +0,0 @@
1
- import { useState } from 'react'
2
-
3
- const RADIUS = 40
4
- const CIRCUMFERENCE = 2 * Math.PI * RADIUS
5
-
6
- function MatchRing({ rate }) {
7
- const color = rate >= 80 ? '#10B981' : rate >= 60 ? '#F59E0B' : '#EF4444'
8
- const offset = CIRCUMFERENCE * (1 - rate / 100)
9
-
10
- return (
11
- <svg width="96" height="96" viewBox="0 0 96 96">
12
- <circle cx="48" cy="48" r={RADIUS} fill="none" stroke="#E5E7EB" strokeWidth="8" />
13
- <circle
14
- cx="48" cy="48" r={RADIUS}
15
- fill="none"
16
- stroke={color}
17
- strokeWidth="8"
18
- strokeDasharray={CIRCUMFERENCE}
19
- strokeDashoffset={offset}
20
- strokeLinecap="round"
21
- transform="rotate(-90 48 48)"
22
- style={{ transition: 'stroke-dashoffset 0.6s ease' }}
23
- />
24
- <text
25
- x="48" y="48"
26
- textAnchor="middle"
27
- dominantBaseline="central"
28
- fontSize="20"
29
- fontWeight="700"
30
- fill={color}
31
- >
32
- {rate}%
33
- </text>
34
- </svg>
35
- )
36
- }
37
-
38
- function MatchBreakdown({ detail, rate }) {
39
- if (!detail) return null
40
- const bars = [
41
- { key: 'semantic', label: '语义匹配', weight: 40, color: '#6366F1' },
42
- { key: 'tags', label: '技能匹配', weight: 35, color: '#10B981' },
43
- { key: 'education', label: '学历匹配', weight: 25, color: '#F59E0B' },
44
- ]
45
- return (
46
- <div className="mt-16" style={{ padding: '12px 16px', backgroundColor: '#FFFFFF', borderRadius: 'var(--radius-sm)', border: '1px solid #E5E7EB' }}>
47
- <div className="flex-between mb-8">
48
- <h4 className="font-bold text-sm">匹配度计算</h4>
49
- <span className="text-xs text-light">总分 = 语义×40% + 技能×35% + 学历×25%</span>
50
- </div>
51
- {bars.map(b => {
52
- const val = detail[b.key] || 0
53
- const pct = Math.round(val * 100)
54
- return (
55
- <div key={b.key} className="flex items-center gap-8 mb-6">
56
- <span className="text-xs" style={{ width: 56, flexShrink: 0 }}>{b.label}</span>
57
- <div style={{ flex: 1, height: 8, backgroundColor: '#F3F4F6', borderRadius: 4, overflow: 'hidden' }}>
58
- <div style={{ width: pct + '%', height: '100%', backgroundColor: b.color, borderRadius: 4, transition: 'width 0.5s ease' }} />
59
- </div>
60
- <span className="text-xs font-bold" style={{ width: 32, textAlign: 'right', color: b.color }}>{pct}%</span>
61
- <span className="text-xs text-light" style={{ width: 28, textAlign: 'right' }}>×{b.weight}%</span>
62
- </div>
63
- )
64
- })}
65
- <div className="flex-between mt-8" style={{ borderTop: '1px solid #E5E7EB', paddingTop: 8 }}>
66
- <span className="text-sm font-bold">最终匹配度</span>
67
- <span className="text-sm font-bold" style={{ color: rate >= 80 ? '#10B981' : rate >= 60 ? '#F59E0B' : '#EF4444' }}>{rate}%</span>
68
- </div>
69
- </div>
70
- )
71
- }
72
- function ExpectationSignals({ signals }) {
73
- if (!signals || signals.length === 0) return null
74
- return (
75
- <div className="mt-12">
76
- <h4 className="font-bold text-sm mb-8">岗位画像</h4>
77
- <div className="flex flex-wrap gap-8">
78
- {signals.map((signal, i) => (
79
- <div key={i} className="tag" style={{
80
- display: 'flex', flexDirection: 'column', gap: 4,
81
- padding: '8px 12px', backgroundColor: '#EEF2FF',
82
- border: '1px solid #C7D2FE', borderRadius: 'var(--radius-sm)'
83
- }}>
84
- <span className="text-sm font-bold" style={{ color: '#4338CA' }}>{signal.name || signal.tag}</span>
85
- {signal.evidence && (
86
- <span className="text-xs text-secondary">{signal.evidence}</span>
87
- )}
88
- {signal.confidence !== undefined && (
89
- <div className="flex items-center gap-4">
90
- <span className="text-xs text-light">置信度</span>
91
- <div style={{
92
- flex: 1, height: 6, backgroundColor: '#E5E7EB',
93
- borderRadius: 3, overflow: 'hidden'
94
- }}>
95
- <div style={{
96
- width: `${Math.round(signal.confidence * 100)}%`,
97
- height: '100%', backgroundColor: '#6366F1',
98
- borderRadius: 3, transition: 'width 0.4s ease'
99
- }} />
100
- </div>
101
- <span className="text-xs font-bold" style={{ color: '#4338CA' }}>
102
- {Math.round(signal.confidence * 100)}%
103
- </span>
104
- </div>
105
- )}
106
- </div>
107
- ))}
108
- </div>
109
- </div>
110
- )
111
- }
112
-
113
- function JobCard({ job, onClick, lowMatchWarning, showMatch = true, onUploadClick }) {
114
- const { job_id, name: title, company, city, match_rate, match_detail, tags = [], description, requirements, expectation_signals } = job
115
- const [expanded, setExpanded] = useState(false)
116
-
117
- const visibleTags = tags.slice(0, 3)
118
- const extraCount = tags.length - 3
119
- const descSummary = description && description.length > 80
120
- ? description.slice(0, 80) + '...'
121
- : description
122
-
123
- return (
124
- <div className="card">
125
- <div style={{ cursor: 'pointer' }} onClick={() => setExpanded(!expanded)}>
126
- <div className="gap-16" style={{ display: 'flex', alignItems: 'center' }}>
127
- {showMatch && (
128
- <div style={{ flexShrink: 0 }}>
129
- <MatchRing rate={match_rate} />
130
- </div>
131
- )}
132
- <div style={{ flex: 1, minWidth: 0 }}>
133
- <h3 className="font-bold text-lg">{title}</h3>
134
- <p className="text-secondary mt-8">{company}</p>
135
- <div className="flex flex-wrap gap-8" style={{ marginTop: 12 }}>
136
- <span className="text-sm text-secondary">{city}</span>
137
- {visibleTags.map((tag, i) => (
138
- <span key={i} className="tag">{tag}</span>
139
- ))}
140
- {extraCount > 0 && (
141
- <span className="text-sm text-light">+{extraCount}</span>
142
- )}
143
- </div>
144
- </div>
145
- </div>
146
- {showMatch && lowMatchWarning && (
147
- <div className="text-sm text-warning" style={{ marginTop: 12 }}>
148
- {match_rate < 60 ? '匹配度较低,别灰心,看看建议提升一下吧' : '匹配度较低'}
149
- </div>
150
- )}
151
- {descSummary && (
152
- <p className="text-sm text-secondary mt-12" style={{ lineHeight: 1.5 }}>{descSummary}</p>
153
- )}
154
- </div>
155
-
156
- {expanded && (
157
- <div style={{ marginTop: 12, padding: 16, backgroundColor: '#F9FAFB', borderRadius: 'var(--radius-sm)' }}>
158
- {description && (
159
- <div className="mb-12">
160
- <h4 className="font-bold text-sm mb-4">职位描述</h4>
161
- <p className="text-sm text-secondary">{description}</p>
162
- </div>
163
- )}
164
- {requirements && (
165
- <div className="mb-12">
166
- <h4 className="font-bold text-sm mb-4">职位要求</h4>
167
- <p className="text-sm text-secondary">{requirements}</p>
168
- </div>
169
- )}
170
- {showMatch && match_detail && <MatchBreakdown detail={match_detail} rate={match_rate} />}
171
- <ExpectationSignals signals={expectation_signals} />
172
- <div className="mt-12 flex gap-8" style={{ justifyContent: 'center' }}>
173
- {showMatch ? (
174
- <button className="btn btn-primary btn-sm" onClick={(e) => { e.stopPropagation(); onClick && onClick(job_id, { job_id, name: title, company, city, match_rate, tags, description, requirements, expectation_signals }) }}>
175
- 查看匹配诊断
176
- </button>
177
- ) : (
178
- <button className="btn btn-primary btn-sm" onClick={(e) => { e.stopPropagation(); onUploadClick && onUploadClick() }}>
179
- 上传简历后使用匹配
180
- </button>
181
- )}
182
- </div>
183
- </div>
184
- )}
185
- </div>
186
- )
187
- }
188
-
189
- export default JobCard
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/components/PipelineFlow.jsx DELETED
@@ -1,39 +0,0 @@
1
- function PipelineFlow({ steps = [] }) {
2
- return (
3
- <div className="flex-center" style={{ gap: 0, padding: '24px 0' }}>
4
- {steps.map((step, index) => {
5
- const isLast = index === steps.length - 1
6
- return (
7
- <div key={step.id} className="flex-center" style={{ gap: 0 }}>
8
- <div className="flex flex-col flex-center" style={{ gap: 8 }}>
9
- <div
10
- className={`pipeline-node pipeline-node-${step.status}`}
11
- >
12
- {step.status === 'completed' && '✓'}
13
- {step.status === 'error' && '❗'}
14
- {(step.status === 'idle' || step.status === 'running') && (
15
- <span className={step.status === 'running' ? 'pipeline-step-running' : ''}>
16
- {step.icon || '○'}
17
- </span>
18
- )}
19
- </div>
20
- <span className={`text-sm pipeline-label-${step.status}`}>
21
- {step.label}
22
- </span>
23
- {step.status === 'error' && (
24
- <span className="text-sm text-error">暂不可用</span>
25
- )}
26
- </div>
27
- {!isLast && (
28
- <div
29
- className={`pipeline-connector pipeline-connector-${step.status === 'completed' ? 'completed' : 'idle'}`}
30
- />
31
- )}
32
- </div>
33
- )
34
- })}
35
- </div>
36
- )
37
- }
38
-
39
- export default PipelineFlow
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/components/RadarChart.jsx DELETED
@@ -1,49 +0,0 @@
1
- import ReactECharts from 'echarts-for-react'
2
-
3
- function RadarChart({ dimensions = [], scores = [], loading = false }) {
4
- if (loading) {
5
- return (
6
- <div className="loading-state" style={{ height: 320 }}>
7
- <div className="loading-spinner loading-spinner-lg" />
8
- <p className="text-secondary mt-16">加载中...</p>
9
- </div>
10
- )
11
- }
12
-
13
- const isEmpty = !dimensions.length || !scores.length || scores.every((s) => s === 0)
14
-
15
- if (isEmpty) {
16
- return (
17
- <div className="flex-center" style={{ height: 320 }}>
18
- <p className="text-secondary">图表数据暂不可用</p>
19
- </div>
20
- )
21
- }
22
-
23
- const option = {
24
- radar: {
25
- indicator: dimensions.map((name) => ({ name, max: 100 })),
26
- center: ['50%', '50%'],
27
- radius: '65%',
28
- },
29
- series: [
30
- {
31
- type: 'radar',
32
- data: [{ value: scores, name: '匹配得分' }],
33
- areaStyle: { color: 'rgba(79, 70, 229, 0.2)' },
34
- lineStyle: { color: '#4F46E5' },
35
- itemStyle: { color: '#4F46E5' },
36
- },
37
- ],
38
- animationDuration: 1000,
39
- animationEasing: 'cubicOut',
40
- }
41
-
42
- return (
43
- <div style={{ height: 320 }}>
44
- <ReactECharts option={option} style={{ height: '100%' }} />
45
- </div>
46
- )
47
- }
48
-
49
- export default RadarChart
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/components/ReportCard.jsx DELETED
@@ -1,149 +0,0 @@
1
- function ReportCard({ dimension, score, evidence = [], viewType, footnote, dataUnavailable, hrView }) {
2
- if (viewType === 'hr' && hrView && hrView.first_impression) {
3
- const levelColor = hrView.hr_concern_level === 'low' ? '#10B981' : hrView.hr_concern_level === 'medium' ? '#F59E0B' : '#EF4444'
4
- const levelLabel = hrView.hr_concern_level === 'low' ? '低' : hrView.hr_concern_level === 'medium' ? '中' : '高'
5
- return (
6
- <div className="card">
7
- <h4 className="font-bold mb-12">{dimension}</h4>
8
- <div style={{ padding: 12, backgroundColor: 'rgba(79,70,229,0.06)', borderRadius: 'var(--radius-sm)', borderLeft: '3px solid #4F46E5', marginBottom: 16 }}>
9
- <p className="text-sm" style={{ fontStyle: 'italic' }}>"{hrView.first_impression}"</p>
10
- </div>
11
- <div className="flex flex-col gap-12 mb-16">
12
- {hrView.attention_points && hrView.attention_points.map((point, i) => (
13
- <div key={i} style={{ padding: 12, backgroundColor: '#F9FAFB', borderRadius: 'var(--radius-sm)' }}>
14
- <p className="font-bold text-sm">{point.aspect}</p>
15
- <p className="text-sm text-secondary mt-4">{point.finding}</p>
16
- {point.concern && point.concern !== '无' && (
17
- <p className="text-sm" style={{ color: '#DC2626', marginTop: 4 }}>关注点: {point.concern}</p>
18
- )}
19
- </div>
20
- ))}
21
- </div>
22
- <div className="flex items-center gap-8 mb-16">
23
- <span className="text-sm text-secondary">HR 关注度:</span>
24
- <span className="tag" style={{ backgroundColor: `${levelColor}18`, color: levelColor, border: `1px solid ${levelColor}40`, fontWeight: 700 }}>{levelLabel}</span>
25
- </div>
26
- {hrView.screen_summary && (
27
- <p className="text-sm" style={{ padding: 12, backgroundColor: 'rgba(16,185,129,0.06)', borderRadius: 'var(--radius-sm)' }}>
28
- <span className="font-bold">筛选总结: </span>{hrView.screen_summary}
29
- </p>
30
- )}
31
- </div>
32
- )
33
- }
34
-
35
- if (dataUnavailable) {
36
- return (
37
- <div className="card" style={{ opacity: 0.5 }}>
38
- <h4 className="font-bold">{dimension}</h4>
39
- <p className="text-secondary mt-8">数据暂不可用</p>
40
- </div>
41
- )
42
- }
43
-
44
- if (viewType === 'hr') {
45
- const conclusion = score >= 70 ? 'OK ✓' : score >= 40 ? '⚠️' : '✗'
46
- const finding = evidence.length > 0 ? evidence[0].text : '暂无关键发现'
47
- return (
48
- <div className="card">
49
- <div className="flex-between">
50
- <h4 className="font-bold">{dimension}</h4>
51
- <span className="text-lg font-bold">{conclusion}</span>
52
- </div>
53
- <p className="text-secondary mt-8">{finding}</p>
54
- {footnote && (
55
- <p className="text-sm text-light mt-8" style={{ padding: '8px 12px', backgroundColor: 'rgba(79,70,229,0.06)', borderRadius: 'var(--radius-sm)' }}>
56
- {footnote}
57
- </p>
58
- )}
59
- </div>
60
- )
61
- }
62
-
63
- const evidenceObj = Array.isArray(evidence) ? (evidence.length > 0 ? evidence[0] : null) : evidence
64
- const hasJdQuote = evidenceObj && (evidenceObj.jd_quote || evidenceObj.resume_quote)
65
-
66
- if (hasJdQuote) {
67
- return (
68
- <div className="card">
69
- <div className="flex-between">
70
- <h4 className="font-bold">{dimension}</h4>
71
- <span className="text-lg font-bold">{score}%</span>
72
- </div>
73
- <div
74
- style={{
75
- height: 8,
76
- backgroundColor: '#E5E7EB',
77
- borderRadius: 4,
78
- marginTop: 12,
79
- overflow: 'hidden',
80
- }}
81
- >
82
- <div
83
- style={{
84
- width: `${score}%`,
85
- height: '100%',
86
- backgroundColor: '#4F46E5',
87
- borderRadius: 4,
88
- transition: 'width 0.5s ease',
89
- }}
90
- />
91
- </div>
92
- <div className="mt-16" style={{ display: 'grid', gridTemplateColumns: '1fr 1fr', gap: 12 }}>
93
- <div style={{ padding: 12, border: '1px solid var(--color-border)', borderRadius: 'var(--radius-sm)', backgroundColor: 'rgba(79,70,229,0.04)' }}>
94
- <p className="font-bold text-sm mb-4">📋 JD要求</p>
95
- <p className="text-sm text-secondary">{evidenceObj.jd_quote}</p>
96
- </div>
97
- <div style={{ padding: 12, border: '1px solid var(--color-border)', borderRadius: 'var(--radius-sm)', backgroundColor: 'rgba(16,185,129,0.04)' }}>
98
- <p className="font-bold text-sm mb-4">📝 简历匹配</p>
99
- <p className="text-sm text-secondary">{evidenceObj.resume_quote}</p>
100
- </div>
101
- </div>
102
- </div>
103
- )
104
- }
105
-
106
- return (
107
- <div className="card">
108
- <div className="flex-between">
109
- <h4 className="font-bold">{dimension}</h4>
110
- <span className="text-lg font-bold">{score}%</span>
111
- </div>
112
- <div
113
- style={{
114
- height: 8,
115
- backgroundColor: '#E5E7EB',
116
- borderRadius: 4,
117
- marginTop: 12,
118
- overflow: 'hidden',
119
- }}
120
- >
121
- <div
122
- style={{
123
- width: `${score}%`,
124
- height: '100%',
125
- backgroundColor: '#4F46E5',
126
- borderRadius: 4,
127
- transition: 'width 0.5s ease',
128
- }}
129
- />
130
- </div>
131
- {(Array.isArray(evidence) && evidence.length > 0) && (
132
- <ul className="mt-16" style={{ listStyle: 'none' }}>
133
- {evidence.map((item, i) => (
134
- <li key={i} className="text-sm text-secondary" style={{ marginTop: 4 }}>
135
- <span style={{ color: '#4F46E5' }}>{item.text}</span>
136
- {item.source && (
137
- <span className="text-light" style={{ marginLeft: 4 }}>
138
- — {item.source}
139
- </span>
140
- )}
141
- </li>
142
- ))}
143
- </ul>
144
- )}
145
- </div>
146
- )
147
- }
148
-
149
- export default ReportCard
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/main.jsx DELETED
@@ -1,10 +0,0 @@
1
- import React from 'react'
2
- import ReactDOM from 'react-dom/client'
3
- import App from './App'
4
- import './App.css'
5
-
6
- ReactDOM.createRoot(document.getElementById('app')).render(
7
- <React.StrictMode>
8
- <App />
9
- </React.StrictMode>
10
- )
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/pages/DiagnosePage.jsx DELETED
@@ -1,271 +0,0 @@
1
- import { useState, useRef, useCallback, useEffect } from 'react'
2
- import { generateReport } from '../api'
3
- import PipelineFlow from '../components/PipelineFlow'
4
- import RadarChart from '../components/RadarChart'
5
- import ReportCard from '../components/ReportCard'
6
-
7
- const INITIAL_STEPS = [
8
- { id: 'agent1', label: '简历解析', status: 'idle' },
9
- { id: 'agent2', label: '匹配诊断', status: 'idle' },
10
- { id: 'agent3', label: '报告生成', status: 'idle' },
11
- ]
12
-
13
- function DiagnosePage({ jobId, jobData, resumeData, onComplete, onBack, onGoHome }) {
14
- const [pageStatus, setPageStatus] = useState('idle')
15
- const [steps, setSteps] = useState(INITIAL_STEPS)
16
- const [viewType, setViewType] = useState('student')
17
- const [jdParsed, setJdParsed] = useState(null)
18
- const [matchReport, setMatchReport] = useState(null)
19
- const [advice, setAdvice] = useState(null)
20
- const [errorMsg, setErrorMsg] = useState('')
21
- const abortRef = useRef(null)
22
- const retryCountRef = useRef(0)
23
- const retryTimeoutRef = useRef(null)
24
-
25
- const jobInfo = jobData || {
26
- name: '岗位名称',
27
- company: '公司名称',
28
- city: '',
29
- salary: '',
30
- tags: [],
31
- description: '',
32
- requirements: '',
33
- expectation_signals: []
34
- }
35
-
36
- const updateStep = useCallback((agentId, newStatus) => {
37
- setSteps((prev) => prev.map((s) => (s.id === agentId ? { ...s, status: newStatus } : s)))
38
- }, [])
39
-
40
- const handleAnalyze = useCallback(async () => {
41
- setPageStatus('loading')
42
- setSteps(INITIAL_STEPS.map((s) => ({ ...s, status: 'idle' })))
43
- setJdParsed(null)
44
- setMatchReport(null)
45
- setAdvice(null)
46
- retryCountRef.current = 0
47
-
48
- const controller = new AbortController()
49
- abortRef.current = controller
50
-
51
- const doFetch = async () => {
52
- try {
53
- const response = await generateReport(jobId, resumeData)
54
- if (!response.ok) throw new Error('请求失败')
55
- const reader = response.body.getReader()
56
- const decoder = new TextDecoder()
57
- let buffer = ''
58
-
59
- while (true) {
60
- const { done, value } = await reader.read()
61
- if (done) break
62
-
63
- buffer += decoder.decode(value, { stream: true })
64
- const lines = buffer.split('\n')
65
- buffer = lines.pop() || ''
66
-
67
- for (const line of lines) {
68
- const trimmed = line.trim()
69
- if (!trimmed) continue
70
- const jsonStr = trimmed.startsWith('data: ') ? trimmed.slice(6).trim() : trimmed
71
- if (!jsonStr) continue
72
- try {
73
- const event = JSON.parse(jsonStr)
74
- const { type, agent, data } = event
75
-
76
- if (type === 'agent_start') {
77
- updateStep(agent, 'running')
78
- } else if (type === 'agent_complete') {
79
- updateStep(agent, 'completed')
80
- if (agent === 'agent1' && data) setJdParsed(data)
81
- if (agent === 'agent2' && data) setMatchReport(data)
82
- if (agent === 'agent3' && data) setAdvice(data)
83
- } else if (type === 'report_complete') {
84
- if (data) {
85
- if (data.jd_parsed) setJdParsed(data.jd_parsed)
86
- if (data.match_report) setMatchReport(data.match_report)
87
- if (data.advice) setAdvice(data.advice)
88
- }
89
- } else if (type === 'error') {
90
- updateStep(agent || 'agent3', 'error')
91
- }
92
- } catch {
93
- // skip invalid lines
94
- }
95
- }
96
- }
97
-
98
- setPageStatus('success')
99
- } catch (err) {
100
- if (err.name === 'AbortError') return
101
- retryCountRef.current += 1
102
- if (retryCountRef.current <= 3) {
103
- const delay = retryCountRef.current * 1000
104
- retryTimeoutRef.current = setTimeout(doFetch, delay)
105
- } else {
106
- setPageStatus('error')
107
- setErrorMsg('连接中断,请刷新页面')
108
- }
109
- }
110
- }
111
-
112
- doFetch()
113
- }, [jobId, updateStep])
114
-
115
- const abort = useCallback(() => {
116
- clearTimeout(retryTimeoutRef.current)
117
- abortRef.current?.abort()
118
- setPageStatus('idle')
119
- }, [])
120
-
121
- useEffect(() => {
122
- return () => {
123
- clearTimeout(retryTimeoutRef.current)
124
- abortRef.current?.abort()
125
- }
126
- }, [])
127
-
128
- const studentDims = matchReport?.dimensions?.map(d => d.name) || ['技能匹配', '经验匹配', '项目匹配', '学历匹配']
129
- const studentScores = matchReport?.dimensions?.map(d => d.score) || []
130
- const studentEvidence = matchReport?.dimensions?.map(d => d.evidence) || []
131
-
132
- const hrViewData = matchReport?.hr_view || {}
133
-
134
- const showResults = pageStatus === 'loading' || pageStatus === 'success' || pageStatus === 'error'
135
- const analyzeDone = pageStatus === 'success'
136
-
137
- if (!resumeData) {
138
- return (
139
- <div>
140
- <div className="card" style={{ textAlign: 'center', padding: 48 }}>
141
- <h2 className="font-bold text-xl mb-16">匹配诊断</h2>
142
- <p className="text-secondary mb-24">请先���传或填写简历,才能进行匹配诊断</p>
143
- <div className="flex gap-12" style={{ justifyContent: 'center' }}>
144
- <button className="btn btn-primary" onClick={onGoHome}>上传简历</button>
145
- <button className="btn btn-secondary" onClick={onGoHome}>手工填写</button>
146
- </div>
147
- <div className="mt-16">
148
- <button className="btn btn-secondary" onClick={onBack}>返回推荐</button>
149
- </div>
150
- </div>
151
- </div>
152
- )
153
- }
154
-
155
- return (
156
- <div>
157
- <div className="card mb-24">
158
- <div className="flex-between mb-8">
159
- <div>
160
- <h2 className="font-bold text-xl">{jobInfo.name}</h2>
161
- <p className="text-secondary">{jobInfo.company} · {jobInfo.city || ''} {jobInfo.salary ? '· ' + jobInfo.salary : ''}</p>
162
- </div>
163
- <div className="flex gap-8 flex-wrap" style={{ maxWidth: '50%', justifyContent: 'flex-end' }}>
164
- {(jobInfo.tags || []).map((tag, i) => (
165
- <span key={i} className="tag" style={{ fontSize: 12 }}>{tag}</span>
166
- ))}
167
- </div>
168
- </div>
169
-
170
- {jobInfo.description && (
171
- <div className="mb-12">
172
- <h4 className="font-bold text-sm mb-4">职位描述</h4>
173
- <p className="text-sm text-secondary" style={{ lineHeight: 1.6, whiteSpace: 'pre-wrap' }}>{jobInfo.description}</p>
174
- </div>
175
- )}
176
-
177
- {jobInfo.requirements && (
178
- <div className="mb-12">
179
- <h4 className="font-bold text-sm mb-4">职位要求</h4>
180
- <p className="text-sm text-secondary" style={{ lineHeight: 1.6, whiteSpace: 'pre-wrap' }}>{jobInfo.requirements}</p>
181
- </div>
182
- )}
183
-
184
- {jobInfo.expectation_signals && jobInfo.expectation_signals.length > 0 && (
185
- <div className="mb-12">
186
- <h4 className="font-bold text-sm mb-4">岗位画像</h4>
187
- <div className="flex flex-col gap-8">
188
- {jobInfo.expectation_signals.map((signal, i) => (
189
- <div key={i} style={{ padding: '8px 12px', backgroundColor: '#F9FAFB', borderRadius: 'var(--radius-sm)', border: '1px solid #E5E7EB' }}>
190
- <div className="flex-between mb-4">
191
- <span className="text-sm font-bold">{signal.tag || signal.name}</span>
192
- <span className="text-xs text-light">可信度 {Math.round((signal.confidence || 0) * 100)}%</span>
193
- </div>
194
- <p className="text-xs text-secondary mb-4">{signal.evidence || ''}</p>
195
- <div style={{ height: 4, backgroundColor: '#E5E7EB', borderRadius: 2, overflow: 'hidden' }}>
196
- <div style={{ width: `${Math.round((signal.confidence || 0) * 100)}%`, height: '100%', backgroundColor: (signal.confidence || 0) > 0.7 ? '#10B981' : (signal.confidence || 0) > 0.4 ? '#F59E0B' : '#EF4444', borderRadius: 2 }} />
197
- </div>
198
- </div>
199
- ))}
200
- </div>
201
- </div>
202
- )}
203
- </div>
204
-
205
- {pageStatus === 'idle' && (
206
- <div className="card mb-24" style={{ backgroundColor: '#FAFAFA' }}>
207
- <h4 className="font-bold mb-8">分析流程说明</h4>
208
- <p className="text-sm text-secondary mb-12">分析将由 3 个 AI Agent 依次完成:</p>
209
- <div className="flex flex-col gap-8">
210
- <div style={{ padding: '8px 12px', backgroundColor: '#FFFFFF', borderRadius: 'var(--radius-sm)', border: '1px solid #E5E7EB' }}>
211
- <span className="text-sm font-bold">① JD 解析器</span>
212
- <p className="text-xs text-secondary mt-4">提取岗位核心要求、技术栈、加分项、文化信号</p>
213
- </div>
214
- <div style={{ padding: '8px 12px', backgroundColor: '#FFFFFF', borderRadius: 'var(--radius-sm)', border: '1px solid #E5E7EB' }}>
215
- <span className="text-sm font-bold">② 匹配诊断器</span>
216
- <p className="text-xs text-secondary mt-4">多维度拆解匹配度,输出分数和引用证据</p>
217
- </div>
218
- <div style={{ padding: '8px 12px', backgroundColor: '#FFFFFF', borderRadius: 'var(--radius-sm)', border: '1px solid #E5E7EB' }}>
219
- <span className="text-sm font-bold">③ 建议生成器</span>
220
- <p className="text-xs text-secondary mt-4">生成简历修改建议和面试准备方向</p>
221
- </div>
222
- </div>
223
- <button className="btn btn-primary mt-16" onClick={handleAnalyze}>开始分析</button>
224
- </div>
225
- )}
226
-
227
- {showResults && (
228
- <>
229
- <div className="card mb-24"><PipelineFlow steps={steps} /></div>
230
- {pageStatus !== 'loading' && matchReport && (
231
- <>
232
- <div className="flex gap-12 mb-24">
233
- <button className={`btn ${viewType === 'student' ? 'btn-primary' : 'btn-secondary'}`} onClick={() => setViewType('student')}>学生视角</button>
234
- <button className={`btn ${viewType === 'hr' ? 'btn-primary' : 'btn-secondary'}`} onClick={() => setViewType('hr')}>HR 视角</button>
235
- </div>
236
- {viewType === 'student' ? (
237
- <>
238
- <div className="card mb-24"><RadarChart dimensions={studentDims} scores={studentScores} /></div>
239
- <div className="flex flex-col gap-16">
240
- {studentDims.map((dim, i) => (
241
- <ReportCard key={dim} dimension={dim} score={studentScores[i]} evidence={studentEvidence[i]} viewType={viewType} />
242
- ))}
243
- </div>
244
- </>
245
- ) : (
246
- <div className="flex flex-col gap-16">
247
- <ReportCard dimension="HR 快速筛查模拟" viewType="hr" hrView={hrViewData} />
248
- </div>
249
- )}
250
- </>
251
- )}
252
- {pageStatus === 'error' && (
253
- <div className="error-state">
254
- <p className="error-state-text">{errorMsg}</p>
255
- <button className="btn btn-primary" onClick={handleAnalyze}>重新分析</button>
256
- </div>
257
- )}
258
- </>
259
- )}
260
-
261
- {analyzeDone && (
262
- <div className="mt-24 flex gap-12" style={{ justifyContent: 'center' }}>
263
- <button className="btn btn-primary" onClick={() => onComplete({ advice })}>查看改进建议</button>
264
- <button className="btn btn-secondary" onClick={onBack}>返回推荐</button>
265
- </div>
266
- )}
267
- </div>
268
- )
269
- }
270
-
271
- export default DiagnosePage
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/pages/HomePage.jsx DELETED
@@ -1,548 +0,0 @@
1
- import { useState, useEffect, useRef, useCallback } from 'react'
2
- import { uploadResume, getRecommendJobs } from '../api'
3
- import RadarChart from '../components/RadarChart'
4
-
5
- const SKILL_LEVELS = ['精通', '熟悉', '了解']
6
-
7
- function HomePage({ resumeData, onConfirm, onGoRecommend }) {
8
- const [status, setStatus] = useState(resumeData ? 'dashboard' : 'idle')
9
- const [errorMsg, setErrorMsg] = useState('')
10
- const [localParsed, setLocalParsed] = useState(null)
11
- const [failCount, setFailCount] = useState(0)
12
- const [manualMode, setManualMode] = useState(false)
13
- const [autoTriggered, setAutoTriggered] = useState(false)
14
- const [previewJobs, setPreviewJobs] = useState([])
15
- const inputRef = useRef(null)
16
-
17
- const [formData, setFormData] = useState({
18
- skillInput: '',
19
- skills: [],
20
- projects: [{ name: '', description: '', techStack: '' }],
21
- experience: '',
22
- education: { school: '', major: '', degree: '', startYear: '', endYear: '' }
23
- })
24
-
25
- const [editValues, setEditValues] = useState({ name: '', school: '', major: '', grade: '', skills: '', projects: '', summary: '', experience: '', education: '' })
26
-
27
- useEffect(() => {
28
- if (localParsed) {
29
- const projToString = (p) => typeof p === 'string' ? p : (p.name || p.title || JSON.stringify(p))
30
- setEditValues({
31
- name: localParsed.name || '',
32
- school: localParsed.school || '',
33
- major: localParsed.major || '',
34
- grade: localParsed.grade || '',
35
- skills: (localParsed.skills || []).join(', '),
36
- projects: (localParsed.projects || []).map(projToString).join(', '),
37
- summary: localParsed.summary || '',
38
- experience: localParsed.experience || '',
39
- education: localParsed.education || '',
40
- })
41
- }
42
- }, [localParsed])
43
-
44
- useEffect(() => {
45
- if (manualMode && !localParsed && !resumeData) {
46
- const s = formData.skillInput || formData.skills.length > 0 || formData.projects.some(p => p.name)
47
- if (!s) handleReEdit()
48
- }
49
- }, [manualMode])
50
-
51
- useEffect(() => {
52
- if (localParsed || resumeData) {
53
- const fetchPreview = async () => {
54
- try {
55
- const resumeId = (localParsed || resumeData)?.id || 1
56
- const res = await getRecommendJobs(resumeId)
57
- const list = res.data?.jobs || []
58
- setPreviewJobs(list.slice(0, 3))
59
- } catch {
60
- setPreviewJobs([])
61
- }
62
- }
63
- fetchPreview()
64
- }
65
- }, [localParsed, resumeData])
66
-
67
- useEffect(() => {
68
- if (resumeData) {
69
- setStatus('dashboard')
70
- }
71
- }, [resumeData])
72
-
73
- const handleFile = useCallback(async (selectedFile) => {
74
- if (!selectedFile) return
75
- if (selectedFile.type !== 'application/pdf') {
76
- setFailCount(c => { const n = c + 1; if (n >= 3) { setManualMode(true); setAutoTriggered(true) }; return n })
77
- setStatus('error'); setErrorMsg('请上传 PDF 格式简历'); return
78
- }
79
- setStatus('uploading')
80
- try {
81
- setStatus('loading')
82
- const res = await uploadResume(selectedFile)
83
- const parsed = res.data?.data || res.data
84
- setLocalParsed(parsed)
85
- setStatus('preview')
86
- setFailCount(0)
87
- } catch {
88
- setFailCount(c => { const n = c + 1; if (n >= 3) { setManualMode(true); setAutoTriggered(true) }; return n })
89
- setStatus('error'); setErrorMsg('简历解析失败,请重新上传或手工填写')
90
- }
91
- }, [])
92
-
93
- const handleConfirm = (formData) => {
94
- const data = formData || localParsed
95
- if (data) {
96
- const confirmed = {
97
- ...data,
98
- name: data.name || localParsed?.name || '',
99
- school: data.school || localParsed?.school || '',
100
- major: data.major || localParsed?.major || '',
101
- grade: data.grade || localParsed?.grade || '',
102
- skills: typeof data.skills === 'string' ? data.skills.split(',').map(s => s.trim()).filter(Boolean) : (data.skills || localParsed?.skills || []),
103
- projects: typeof data.projects === 'string' ? data.projects.split(',').map(s => s.trim()).filter(Boolean) : (data.projects || localParsed?.projects || []),
104
- summary: data.summary || localParsed?.summary || '',
105
- experience: data.experience || localParsed?.experience || '',
106
- education: data.education || localParsed?.education || { school: '', major: '', degree: '', startYear: '', endYear: '' },
107
- }
108
- onConfirm(confirmed)
109
- }
110
- }
111
-
112
- const handleReEdit = () => {
113
- const source = localParsed || resumeData
114
- if (!source) {
115
- setFormData({
116
- skillInput: 'Python, Java, SQL, PyTorch, Flask',
117
- skills: [
118
- { name: 'Python', level: '熟悉' },
119
- { name: 'Java', level: '熟悉' },
120
- { name: 'SQL', level: '熟悉' },
121
- { name: 'PyTorch', level: '了解' },
122
- { name: 'Flask', level: '了解' },
123
- ],
124
- projects: [
125
- { name: '基于协同过滤的推荐系统', description: '使用协同过滤算法实现电影推荐,处理了MovieLens数据集', techStack: 'Python, Flask' },
126
- { name: '在线商城后端开发', description: '使用Spring Boot实现RESTful API,包括用户管理、商品管理模块', techStack: 'Java, Spring Boot' },
127
- ],
128
- experience: '6个月实习经历,参与推荐系统数据管道维护',
129
- education: { school: '华南理工大学', major: '计算机科学与技术', degree: '本科', startYear: '2021', endYear: '2025' }
130
- })
131
- } else {
132
- const mapProject = (p) => {
133
- const pName = typeof p === 'string' ? p : (p.name || '')
134
- const pDesc = typeof p === 'string' ? p : (p.desc || p.name || '')
135
- return { name: pName, description: pDesc, techStack: '' }
136
- }
137
- setFormData({
138
- skillInput: (source.skills || []).join(', '),
139
- skills: (source.skills || []).map(name => ({ name, level: '熟悉' })),
140
- projects: (source.projects || []).map(mapProject),
141
- experience: source.experience || '',
142
- education: {
143
- school: source.education?.school || '',
144
- major: source.education?.major || '',
145
- degree: source.education?.degree || '',
146
- startYear: source.education?.startYear || '',
147
- endYear: source.education?.endYear || ''
148
- }
149
- })
150
- }
151
- setManualMode(true)
152
- }
153
-
154
- const parseSkillInput = (text) => {
155
- const names = text.split(',').map(s => s.trim()).filter(Boolean)
156
- return names.map(name => {
157
- const existing = formData.skills.find(s => s.name === name)
158
- return { name, level: existing ? existing.level : '熟悉' }
159
- })
160
- }
161
-
162
- const handleSkillInputChange = (text) => {
163
- setFormData(p => ({ ...p, skillInput: text, skills: parseSkillInput(text) }))
164
- }
165
-
166
- const handleSkillLevelChange = (skillName, level) => {
167
- setFormData(p => ({
168
- ...p,
169
- skills: p.skills.map(s => s.name === skillName ? { ...s, level } : s)
170
- }))
171
- }
172
-
173
- const handleProjectAdd = () => {
174
- setFormData(p => ({ ...p, projects: [...p.projects, { name: '', description: '', techStack: '' }] }))
175
- }
176
-
177
- const handleProjectRemove = (index) => {
178
- setFormData(p => ({ ...p, projects: p.projects.filter((_, i) => i !== index) }))
179
- }
180
-
181
- const handleProjectChange = (index, field, value) => {
182
- setFormData(p => {
183
- const projects = [...p.projects]
184
- projects[index] = { ...projects[index], [field]: value }
185
- return { ...p, projects }
186
- })
187
- }
188
-
189
- const handleEducationChange = (field, value) => {
190
- setFormData(p => ({ ...p, education: { ...p.education, [field]: value } }))
191
- }
192
-
193
- const handleManualSubmit = useCallback(() => {
194
- const hasAny = formData.skillInput.trim() || formData.projects.some(p => p.name.trim()) || formData.experience.trim() || formData.education.school.trim()
195
- if (!hasAny) { setErrorMsg('请至少填写一个字段'); return }
196
- const data = {
197
- name: '',
198
- school: formData.education.school || '',
199
- major: formData.education.major || '',
200
- grade: formData.education.degree || '',
201
- skills: formData.skills.map(s => s.name),
202
- projects: formData.projects.filter(p => p.name.trim()).map(p => p.name),
203
- projectDetails: formData.projects.filter(p => p.name.trim()),
204
- summary: '',
205
- experience: formData.experience,
206
- education: formData.education.school
207
- ? `${formData.education.school} ${formData.education.major} ${formData.education.degree} ${formData.education.startYear}-${formData.education.endYear}`
208
- : ''
209
- }
210
- setLocalParsed(data)
211
- setStatus('preview')
212
- setManualMode(false)
213
- setAutoTriggered(false)
214
- }, [formData])
215
-
216
- if (manualMode) {
217
- return (
218
- <div className="page-container">
219
- <div className="card" style={{ maxWidth: 640, margin: '0 auto' }}>
220
- <h2 className="font-bold text-xl mb-24">手工填写简历</h2>
221
- {autoTriggered && (
222
- <div className="text-warning text-sm mb-16" style={{ padding: '8px 12px', backgroundColor: 'rgba(245,158,11,0.08)', borderRadius: 'var(--radius-sm)' }}>
223
- 已连续3次上传失败,已自动切换为手工填写模式
224
- </div>
225
- )}
226
- {errorMsg && status === 'error' && (
227
- <div className="text-error text-sm mb-16" style={{ padding: '8px 12px', backgroundColor: 'rgba(239,68,68,0.08)', borderRadius: 'var(--radius-sm)' }}>
228
- {errorMsg}
229
- </div>
230
- )}
231
- <div className="flex flex-col gap-16">
232
- <div>
233
- <label className="text-sm font-bold">技能(逗号分隔)</label>
234
- <input
235
- value={formData.skillInput}
236
- onChange={e => handleSkillInputChange(e.target.value)}
237
- className="input"
238
- placeholder="如: React, Python, SQL"
239
- />
240
- {formData.skills.length > 0 && (
241
- <div className="flex flex-col gap-8 mt-8">
242
- {formData.skills.map((skill, i) => (
243
- <div key={i} className="flex items-center gap-8">
244
- <span className="text-sm" style={{ minWidth: 80 }}>{skill.name}</span>
245
- <select
246
- value={skill.level}
247
- onChange={e => handleSkillLevelChange(skill.name, e.target.value)}
248
- className="input"
249
- style={{ width: 120 }}
250
- >
251
- {SKILL_LEVELS.map(level => (
252
- <option key={level} value={level}>{level}</option>
253
- ))}
254
- </select>
255
- </div>
256
- ))}
257
- </div>
258
- )}
259
- </div>
260
-
261
- <div>
262
- <div className="flex-between mb-8">
263
- <label className="text-sm font-bold">项目经历</label>
264
- <button className="btn btn-sm btn-secondary" onClick={handleProjectAdd}>+ 添加项目</button>
265
- </div>
266
- {formData.projects.map((project, i) => (
267
- <div key={i} className="card mb-8" style={{ padding: 12 }}>
268
- <div className="flex-between mb-8">
269
- <span className="text-sm font-bold">项目 #{i + 1}</span>
270
- {formData.projects.length > 1 && (
271
- <button className="btn btn-sm btn-secondary" onClick={() => handleProjectRemove(i)} style={{ color: '#EF4444' }}>删除</button>
272
- )}
273
- </div>
274
- <input
275
- className="input mb-8"
276
- placeholder="项目名称"
277
- value={project.name}
278
- onChange={e => handleProjectChange(i, 'name', e.target.value)}
279
- />
280
- <textarea
281
- className="input mb-8"
282
- rows={2}
283
- placeholder="项目描述"
284
- value={project.description}
285
- onChange={e => handleProjectChange(i, 'description', e.target.value)}
286
- />
287
- <input
288
- className="input"
289
- placeholder="技术栈(逗号分隔)"
290
- value={project.techStack}
291
- onChange={e => handleProjectChange(i, 'techStack', e.target.value)}
292
- />
293
- </div>
294
- ))}
295
- </div>
296
-
297
- <div>
298
- <label className="text-sm font-bold">实习经历</label>
299
- <textarea
300
- className="input"
301
- rows={3}
302
- value={formData.experience}
303
- onChange={e => setFormData(p => ({ ...p, experience: e.target.value }))}
304
- placeholder="简要描述实习经历"
305
- />
306
- </div>
307
-
308
- <div>
309
- <label className="text-sm font-bold">教育背景</label>
310
- <div className="grid" style={{ display: 'grid', gridTemplateColumns: '1fr 1fr', gap: 8, marginTop: 8 }}>
311
- <input className="input" placeholder="学校" value={formData.education.school} onChange={e => handleEducationChange('school', e.target.value)} />
312
- <input className="input" placeholder="专业" value={formData.education.major} onChange={e => handleEducationChange('major', e.target.value)} />
313
- <input className="input" placeholder="学位(本科/硕士/博士)" value={formData.education.degree} onChange={e => handleEducationChange('degree', e.target.value)} />
314
- <div className="flex gap-8">
315
- <input className="input" placeholder="开始年份" value={formData.education.startYear} onChange={e => handleEducationChange('startYear', e.target.value)} />
316
- <input className="input" placeholder="结束年份" value={formData.education.endYear} onChange={e => handleEducationChange('endYear', e.target.value)} />
317
- </div>
318
- </div>
319
- </div>
320
-
321
- <button className="btn btn-primary" onClick={handleManualSubmit}>提交预览</button>
322
- <button className="btn btn-secondary" onClick={() => { setManualMode(false); setStatus('idle'); setAutoTriggered(false) }}>返回</button>
323
- </div>
324
- </div>
325
- </div>
326
- )
327
- }
328
-
329
- if (status === 'uploading' || status === 'loading') {
330
- return (
331
- <div className="loading-state" style={{ height: 320 }}>
332
- <div className="loading-spinner loading-spinner-lg" />
333
- <p className="mt-16 text-secondary">{status === 'uploading' ? '上传中...' : '解析中...'}</p>
334
- </div>
335
- )
336
- }
337
-
338
- if (status === 'error') {
339
- return (
340
- <div className="error-state">
341
- <p className="error-state-text">{errorMsg}</p>
342
- <div className="flex gap-12">
343
- <button className="btn btn-primary" onClick={() => setStatus('idle')}>重新上传</button>
344
- <button className="btn btn-secondary" onClick={() => { setManualMode(true); setAutoTriggered(true) }}>手工填写</button>
345
- </div>
346
- </div>
347
- )
348
- }
349
-
350
- if (status === 'preview' && localParsed) {
351
- return (
352
- <div className="page-container">
353
- <div className="card" style={{ maxWidth: 640, margin: '0 auto' }}>
354
- <h2 className="font-bold text-xl mb-16">简历解析结果</h2>
355
- <p className="text-sm text-secondary mb-16">请核对并修改解析结果,确认后进入推荐</p>
356
- <div className="flex flex-col gap-12">
357
- <div>
358
- <label className="text-sm font-bold">姓名</label>
359
- <input className="input mt-4" value={editValues.name} onChange={e => setEditValues(f => ({ ...f, name: e.target.value }))} />
360
- </div>
361
- <div>
362
- <label className="text-sm font-bold">学校</label>
363
- <input className="input mt-4" value={editValues.school} onChange={e => setEditValues(f => ({ ...f, school: e.target.value }))} />
364
- </div>
365
- <div>
366
- <label className="text-sm font-bold">专业</label>
367
- <input className="input mt-4" value={editValues.major} onChange={e => setEditValues(f => ({ ...f, major: e.target.value }))} />
368
- </div>
369
- <div>
370
- <label className="text-sm font-bold">年级</label>
371
- <input className="input mt-4" value={editValues.grade} onChange={e => setEditValues(f => ({ ...f, grade: e.target.value }))} />
372
- </div>
373
- <div>
374
- <label className="text-sm font-bold">技能(逗号分隔)</label>
375
- <input className="input mt-4" value={editValues.skills} onChange={e => setEditValues(f => ({ ...f, skills: e.target.value }))} placeholder="如: Python, Java, SQL" />
376
- </div>
377
- <div>
378
- <label className="text-sm font-bold">项目经历(逗号分隔)</label>
379
- <input className="input mt-4" value={editValues.projects} onChange={e => setEditValues(f => ({ ...f, projects: e.target.value }))} placeholder="如: 推荐系统, 电商后端" />
380
- </div>
381
- <div>
382
- <label className="text-sm font-bold">实习经历</label>
383
- <textarea className="input mt-4" rows={3} value={editValues.experience} onChange={e => setEditValues(f => ({ ...f, experience: e.target.value }))} placeholder="描述实习经历" />
384
- </div>
385
- <div>
386
- <label className="text-sm font-bold">自我评价</label>
387
- <textarea className="input mt-4" rows={3} value={editValues.summary} onChange={e => setEditValues(f => ({ ...f, summary: e.target.value }))} />
388
- </div>
389
- </div>
390
- <div className="flex gap-12 mt-24">
391
- <button className="btn btn-primary" onClick={() => handleConfirm(editValues)} style={{ flex: 1 }}>确认并继续</button>
392
- <button className="btn btn-secondary" onClick={handleReEdit} style={{ flex: 1 }}>重新编辑</button>
393
- </div>
394
- </div>
395
- </div>
396
- )
397
- }
398
-
399
- if (status === 'dashboard' && resumeData) {
400
- const skillNames = (resumeData.skills || []).map(s => typeof s === 'string' ? s : s.name).join(', ')
401
- const metrics = [
402
- { label: '画像完成度', value: '85%', color: '#10B981' },
403
- { label: '已匹配岗位', value: `${previewJobs.length}`, color: '#3B82F6' },
404
- { label: '待改进维度', value: '2', color: '#F59E0B' },
405
- ]
406
- return (
407
- <div className="page-container">
408
- <div className="flex-between mb-24" style={{ alignItems: 'flex-start' }}>
409
- <div>
410
- <h2 className="font-bold text-xl">个人中心</h2>
411
- <p className="text-sm text-secondary mt-4">
412
- {resumeData.name || '求职者'} | {resumeData.school || '未设置'}
413
- </p>
414
- </div>
415
- <div className="flex gap-8">
416
- <button className="btn btn-sm btn-secondary" onClick={() => inputRef.current?.click()}>上传简历</button>
417
- <input ref={inputRef} type="file" accept=".pdf" style={{ display: 'none' }} onChange={e => { const f = e.target.files[0]; if (f) handleFile(f); e.target.value = '' }} />
418
- <button className="btn btn-sm btn-secondary" onClick={() => {
419
- setFormData({
420
- skillInput: skillNames,
421
- skills: (resumeData.skills || []).map(s => ({ name: typeof s === 'string' ? s : s.name, level: '熟悉' })),
422
- projects: [{ name: '', description: '', techStack: '' }],
423
- experience: resumeData.experience || '',
424
- education: { school: '', major: '', degree: '', startYear: '', endYear: '' }
425
- })
426
- setManualMode(true)
427
- }}>编辑简历</button>
428
- </div>
429
- </div>
430
-
431
- <div className="flex gap-12 mb-24">
432
- {metrics.map((m, i) => (
433
- <div key={i} className="card" style={{ flex: 1, textAlign: 'center' }}>
434
- <p className="text-2xl font-bold" style={{ color: m.color }}>{m.value}</p>
435
- <p className="text-sm text-secondary mt-4">{m.label}</p>
436
- </div>
437
- ))}
438
- </div>
439
-
440
- <div className="card mb-24" style={{ textAlign: 'center' }}>
441
- <h3 className="font-bold mb-16">能力画像</h3>
442
- <RadarChart dimensions={resumeData.dimensions || ['技能', '项目', '实习', '学历', '综合素质']} scores={resumeData.scores || [80, 75, 70, 85, 78]} />
443
- <p className="text-sm text-secondary mt-16">{resumeData.summary || '技术栈集中在Python/Java后端开发,有推荐系统和Web开发项目经验'}</p>
444
- </div>
445
-
446
- {previewJobs.length > 0 && (
447
- <div className="card mb-24">
448
- <div className="flex-between mb-16">
449
- <h3 className="font-bold">推荐岗位 Top 3</h3>
450
- </div>
451
- <div className="flex flex-col gap-12">
452
- {previewJobs.map(job => (
453
- <div key={job.job_id} className="flex-between" style={{ padding: '8px 12px', backgroundColor: '#F9FAFB', borderRadius: 'var(--radius-sm)' }}>
454
- <div>
455
- <span className="font-bold text-sm">{job.name}</span>
456
- <span className="text-sm text-secondary ml-8">{job.company}</span>
457
- </div>
458
- <span className="font-bold text-sm" style={{ color: job.match_rate >= 80 ? '#10B981' : job.match_rate >= 60 ? '#F59E0B' : '#EF4444' }}>
459
- {job.match_rate}%
460
- </span>
461
- </div>
462
- ))}
463
- </div>
464
- </div>
465
- )}
466
-
467
- {previewJobs.length > 0 && (
468
- <div className="card mb-24">
469
- <div className="flex-between mb-16">
470
- <h3 className="font-bold">💡 HR 视角速览</h3>
471
- </div>
472
- <p className="text-sm mb-16" style={{ lineHeight: 1.7, color: '#4B5563' }}>
473
- HR 初筛一份简历平均只花 <strong>10~15 秒</strong>⏱️,他们最先扫一眼你的学校、实习经历和技能关键词。下面是你 Top 3 岗位的 HR 关注点,帮你理解他们筛选时在想什么👇
474
- </p>
475
- <div className="flex flex-col gap-12">
476
- {previewJobs.map(job => (
477
- <div key={job.job_id} className="card" style={{ padding: '10px 14px', backgroundColor: '#F9FAFB' }}>
478
- <div className="flex-between mb-8">
479
- <span className="font-bold text-sm">{job.name}<span className="text-secondary ml-8" style={{ fontWeight: 400 }}>{job.company}</span></span>
480
- </div>
481
- <div className="flex flex-wrap gap-4">
482
- {(job.expectation_signals || []).map((signal, i) => (
483
- <span key={i} className="text-xs" style={{
484
- padding: '3px 10px',
485
- backgroundColor: '#EFF6FF',
486
- color: '#1D4ED8',
487
- borderRadius: 'var(--radius-sm)',
488
- display: 'inline-block',
489
- marginBottom: 4,
490
- lineHeight: 1.6
491
- }}>
492
- {signal.tag}
493
- <span className="ml-4" style={{ fontSize: 11, color: '#6B7280' }}>— {signal.evidence}</span>
494
- </span>
495
- ))}
496
- </div>
497
- </div>
498
- ))}
499
- </div>
500
- <p className="text-xs mt-16" style={{
501
- padding: '10px 14px',
502
- backgroundColor: 'rgba(245,158,11,0.08)',
503
- borderRadius: 'var(--radius-sm)',
504
- lineHeight: 1.6,
505
- color: '#92400E'
506
- }}>
507
- ✨ <strong>小提醒:</strong>针对每个岗位的 HR 关注点,在简历中有针对性地突出相关关键词和经历,能大幅提高初筛通过率!
508
- </p>
509
- </div>
510
- )}
511
-
512
- <div className="card mb-24">
513
- <h3 className="font-bold mb-16">开始匹配</h3>
514
- <div className="flex flex-row gap-12">
515
- <button className="btn btn-primary" onClick={() => onConfirm(resumeData)}>确认并继续</button>
516
- <button className="btn btn-secondary" onClick={handleReEdit}>编辑简历</button>
517
- <button className="btn btn-secondary" onClick={onGoRecommend}>浏览推荐岗位</button>
518
- </div>
519
- </div>
520
- </div>
521
- )
522
- }
523
-
524
- return (
525
- <div className="page-container">
526
- <div className="card" style={{ textAlign: 'center', padding: 48 }}>
527
- <h2 className="font-bold text-xl mb-8">我的简历</h2>
528
- <p className="text-secondary mb-24">上传简历,AI 帮你匹配合适岗位、诊断简历匹配度、提供改进建议</p>
529
- <div
530
- style={{ border: '2px dashed var(--color-border)', borderRadius: 'var(--radius-lg)', padding: 48, cursor: 'pointer' }}
531
- onClick={() => inputRef.current?.click()}
532
- onDrop={e => { e.preventDefault(); const f = e.dataTransfer.files[0]; if (f) handleFile(f) }}
533
- onDragOver={e => e.preventDefault()}
534
- >
535
- <p className="text-secondary">拖拽 PDF 到此处,或点击上传</p>
536
- <p className="text-sm text-light mt-8">支持 PDF 格式</p>
537
- </div>
538
- <input ref={inputRef} type="file" accept=".pdf" style={{ display: 'none' }} onChange={e => { const f = e.target.files[0]; if (f) handleFile(f); e.target.value = '' }} />
539
- <button className="btn btn-secondary mt-16" onClick={handleReEdit}>没有简历?手工填写</button>
540
- <div className="mt-16">
541
- <button className="btn btn-primary" onClick={onGoRecommend}>浏览推荐岗位</button>
542
- </div>
543
- </div>
544
- </div>
545
- )
546
- }
547
-
548
- export default HomePage
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/pages/RecommendPage.jsx DELETED
@@ -1,113 +0,0 @@
1
- import { useState, useEffect } from 'react'
2
- import { getRecommendJobs } from '../api'
3
- import JobCard from '../components/JobCard'
4
-
5
- const PAGE_SIZE = 10
6
-
7
- function RecommendPage({ resumeData, onSelectJob, onBack, onGoHome }) {
8
- const [status, setStatus] = useState('loading')
9
- const [jobs, setJobs] = useState([])
10
- const [page, setPage] = useState(0)
11
- const [errorMsg, setErrorMsg] = useState('')
12
-
13
- const isBrowsing = !resumeData
14
-
15
- const fetchJobs = async () => {
16
- setStatus('loading')
17
- try {
18
- const resumeId = resumeData?.id || 1
19
- const res = await getRecommendJobs(resumeId)
20
- const list = res.data?.jobs || []
21
- setJobs(list)
22
- setPage(0)
23
- setStatus(list.length === 0 ? 'empty' : 'success')
24
- } catch {
25
- setStatus('error')
26
- setErrorMsg('加载失败,请重试')
27
- }
28
- }
29
-
30
- useEffect(() => {
31
- fetchJobs()
32
- }, [])
33
-
34
- const totalPages = Math.ceil(jobs.length / PAGE_SIZE)
35
- const pageJobs = jobs.slice(page * PAGE_SIZE, (page + 1) * PAGE_SIZE)
36
-
37
- if (status === 'loading') {
38
- return (
39
- <div className="loading-state">
40
- <div className="loading-spinner loading-spinner-lg" />
41
- <p className="mt-16 text-secondary">加载推荐岗位中...</p>
42
- </div>
43
- )
44
- }
45
-
46
- if (status === 'error') {
47
- return (
48
- <div className="error-state">
49
- <p className="error-state-text">{errorMsg}</p>
50
- <div className="flex gap-12">
51
- <button className="btn btn-primary" onClick={fetchJobs}>重试</button>
52
- <button className="btn btn-secondary" onClick={onBack}>返回首页</button>
53
- </div>
54
- </div>
55
- )
56
- }
57
-
58
- if (status === 'empty') {
59
- return (
60
- <div className="empty-state">
61
- <p className="empty-state-text">暂无可推荐岗位,请完善简历后重试</p>
62
- <button className="btn btn-primary" onClick={onBack}>返回首页</button>
63
- </div>
64
- )
65
- }
66
-
67
- return (
68
- <div>
69
- <div className="flex-between mb-24">
70
- <h2 className="font-bold text-xl">推荐岗位</h2>
71
- <span className="text-sm text-light">共 {jobs.length} 个岗位</span>
72
- </div>
73
- <div className="flex flex-col gap-16">
74
- {pageJobs.map((job) => (
75
- <JobCard
76
- key={job.job_id}
77
- job={job}
78
- onClick={(jobId, jobData) => onSelectJob(jobId, jobData || job)}
79
- lowMatchWarning={(job.match_rate || 0) < 40}
80
- showMatch={!isBrowsing}
81
- onUploadClick={isBrowsing ? onGoHome : undefined}
82
- />
83
- ))}
84
- </div>
85
- {totalPages > 1 && (
86
- <div className="flex-between mt-24" style={{ justifyContent: 'center', gap: 12 }}>
87
- <button
88
- className="btn btn-sm btn-secondary"
89
- disabled={page === 0}
90
- onClick={() => setPage(p => Math.max(0, p - 1))}
91
- >
92
- 上一页
93
- </button>
94
- <span className="text-sm text-secondary">
95
- {page + 1} / {totalPages}
96
- </span>
97
- <button
98
- className="btn btn-sm btn-secondary"
99
- disabled={page >= totalPages - 1}
100
- onClick={() => setPage(p => Math.min(totalPages - 1, p + 1))}
101
- >
102
- 下一页
103
- </button>
104
- </div>
105
- )}
106
- <div className="mt-24 text-center">
107
- <button className="btn btn-secondary" onClick={onBack}>返回首页</button>
108
- </div>
109
- </div>
110
- )
111
- }
112
-
113
- export default RecommendPage
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/src/pages/SuggestPage.jsx DELETED
@@ -1,150 +0,0 @@
1
- import { useState, useCallback } from 'react'
2
- import { rerankReport } from '../api'
3
-
4
- function SuggestPage({ jobId, resumeData, advice, onBack }) {
5
- const defaultSkills = 'Python, Java, SQL, PyTorch, Flask'
6
- const defaultProjects = '推荐系统, 电商后端'
7
- const defaultExp = '6个月实习经历,参与推荐系统数据管道维护'
8
-
9
- const initData = resumeData || {}
10
- const projToString = (p) => typeof p === 'string' ? p : (p.name || p.title || '')
11
- const [editedResume, setEditedResume] = useState({
12
- skills: (initData.skills || []).join(', ') || defaultSkills,
13
- projects: (initData.projects || []).map(projToString).join(', ') || defaultProjects,
14
- experience: initData.experience || defaultExp
15
- })
16
- const [rerankResult, setRerankResult] = useState(null)
17
- const [reranking, setReranking] = useState(false)
18
-
19
- const handleRerank = useCallback(async () => {
20
- const hasContent = editedResume.skills.trim() || editedResume.projects.trim() || editedResume.experience.trim()
21
- if (!hasContent) return
22
- setReranking(true)
23
- try {
24
- const text = `技能: ${editedResume.skills}\n项目: ${editedResume.projects}\n经历: ${editedResume.experience}`
25
- const res = await rerankReport(jobId, text)
26
- setRerankResult(res.data)
27
- } catch {
28
- setRerankResult({ error: '重评估失败' })
29
- }
30
- setReranking(false)
31
- }, [jobId, editedResume])
32
-
33
- return (
34
- <div>
35
- <h2 className="font-bold text-xl mb-24">简历修改建议</h2>
36
-
37
- {advice && (
38
- <>
39
- {advice.rewrite_suggestions && advice.rewrite_suggestions.length > 0 && (
40
- <div className="mb-24">
41
- <h3 className="font-bold mb-16">📝 简历修改建议</h3>
42
- <div className="flex flex-col gap-16">
43
- {advice.rewrite_suggestions.map((item, i) => (
44
- <div key={i} className="card">
45
- <div className="flex-between mb-8">
46
- <span className="font-bold">{item.dimension}</span>
47
- <span className="text-sm text-warning">{item.issue}</span>
48
- </div>
49
- <div className="mb-8">
50
- <span className="text-sm text-secondary">原文: </span>
51
- <span className="text-sm" style={{ textDecoration: 'line-through', color: '#EF4444' }}>{item.original}</span>
52
- </div>
53
- <div className="mb-8">
54
- <span className="text-sm text-secondary">建议: </span>
55
- <span className="text-sm font-bold" style={{ color: '#10B981' }}>{item.suggestion}</span>
56
- </div>
57
- <div>
58
- <span className="text-sm text-light">{item.reason}</span>
59
- </div>
60
- </div>
61
- ))}
62
- </div>
63
- </div>
64
- )}
65
-
66
- {advice.interview_tips && advice.interview_tips.length > 0 && (
67
- <div className="mb-24">
68
- <h3 className="font-bold mb-16">🎯 面试准备方向</h3>
69
- <div className="flex flex-col gap-16">
70
- {advice.interview_tips.map((item, i) => (
71
- <div key={i} className="card">
72
- <div className="flex-between mb-8">
73
- <span className="font-bold">{item.topic}</span>
74
- <span className="text-sm text-secondary">{item.reason}</span>
75
- </div>
76
- <p className="text-sm" style={{ color: '#4F46E5' }}>{item.tip}</p>
77
- </div>
78
- ))}
79
- </div>
80
- </div>
81
- )}
82
-
83
- {advice.encouragement && (
84
- <div className="card mb-24" style={{ backgroundColor: 'rgba(16,185,129,0.06)', borderColor: '#10B981' }}>
85
- <h3 className="font-bold mb-8">💪 鼓励</h3>
86
- <p className="text-secondary">{advice.encouragement}</p>
87
- </div>
88
- )}
89
-
90
- <hr className="divider mb-24" />
91
- </>
92
- )}
93
-
94
- <div className="card mb-24">
95
- <h3 className="font-bold mb-16">编辑简历</h3>
96
- <div className="flex flex-col gap-12">
97
- <div>
98
- <label className="text-sm font-bold">技能(逗号分隔)</label>
99
- <input
100
- className="input"
101
- value={editedResume.skills}
102
- onChange={e => setEditedResume(p => ({ ...p, skills: e.target.value }))}
103
- placeholder="如: React, Python, SQL"
104
- />
105
- </div>
106
- <div>
107
- <label className="text-sm font-bold">项目经历</label>
108
- <textarea
109
- className="input"
110
- rows={3}
111
- value={editedResume.projects}
112
- onChange={e => setEditedResume(p => ({ ...p, projects: e.target.value }))}
113
- placeholder="描述你的项目经历"
114
- />
115
- </div>
116
- <div>
117
- <label className="text-sm font-bold">实习经历</label>
118
- <textarea
119
- className="input"
120
- rows={3}
121
- value={editedResume.experience}
122
- onChange={e => setEditedResume(p => ({ ...p, experience: e.target.value }))}
123
- placeholder="描述你的实习经历"
124
- />
125
- </div>
126
- </div>
127
- <button className="btn btn-primary mt-16" onClick={handleRerank} disabled={reranking || !(editedResume.skills.trim() || editedResume.projects.trim() || editedResume.experience.trim())}>
128
- {reranking ? '重新评估中...' : '重新评估匹配度'}
129
- </button>
130
- </div>
131
-
132
- {rerankResult && !rerankResult.error && (
133
- <div className="card mb-24">
134
- <h3 className="font-bold mb-16">修改后匹配度</h3>
135
- <p>匹配度:<span className="font-bold text-lg">{rerankResult.match_rate}%</span></p>
136
- <p className="text-secondary mt-8">修改前:{rerankResult.previous_rate}%</p>
137
- {rerankResult.match_rate > rerankResult.previous_rate && (
138
- <p className="text-success mt-8">修改有效,匹配度提升了 {rerankResult.match_rate - rerankResult.previous_rate}%</p>
139
- )}
140
- </div>
141
- )}
142
-
143
- <div className="text-center">
144
- <button className="btn btn-secondary" onClick={onBack}>返回诊断</button>
145
- </div>
146
- </div>
147
- )
148
- }
149
-
150
- export default SuggestPage
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/frontend/vite.config.js DELETED
@@ -1,12 +0,0 @@
1
- import { defineConfig } from 'vite'
2
- import react from '@vitejs/plugin-react'
3
-
4
- export default defineConfig({
5
- plugins: [react()],
6
- server: {
7
- port: 5173,
8
- proxy: {
9
- '/api': 'http://localhost:8000'
10
- }
11
- }
12
- })
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/models/config.json DELETED
@@ -1,32 +0,0 @@
1
- {
2
- "_name_or_path": "BAAI/bge-small-zh-v1.5",
3
- "architectures": [
4
- "BertModel"
5
- ],
6
- "attention_probs_dropout_prob": 0.1,
7
- "classifier_dropout": null,
8
- "gradient_checkpointing": false,
9
- "hidden_act": "gelu",
10
- "hidden_dropout_prob": 0.1,
11
- "hidden_size": 512,
12
- "id2label": {
13
- "0": "LABEL_0"
14
- },
15
- "initializer_range": 0.02,
16
- "intermediate_size": 2048,
17
- "label2id": {
18
- "LABEL_0": 0
19
- },
20
- "layer_norm_eps": 1e-12,
21
- "max_position_embeddings": 512,
22
- "model_type": "bert",
23
- "num_attention_heads": 8,
24
- "num_hidden_layers": 4,
25
- "pad_token_id": 0,
26
- "position_embedding_type": "absolute",
27
- "torch_dtype": "float32",
28
- "transformers_version": "4.33.2",
29
- "type_vocab_size": 2,
30
- "use_cache": true,
31
- "vocab_size": 21128
32
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
offer-puplayer-hf/models/model_optimized.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:1294ea4b6331115a353d81f96b85e8c8d7fdcc284453d5b2fab5b016230aad38
3
- size 94781076
 
 
 
 
offer-puplayer-hf/models/special_tokens_map.json DELETED
@@ -1,7 +0,0 @@
1
- {
2
- "cls_token": "[CLS]",
3
- "mask_token": "[MASK]",
4
- "pad_token": "[PAD]",
5
- "sep_token": "[SEP]",
6
- "unk_token": "[UNK]"
7
- }
 
 
 
 
 
 
 
 
offer-puplayer-hf/models/tokenizer.json DELETED
The diff for this file is too large to render. See raw diff
 
offer-puplayer-hf/models/tokenizer_config.json DELETED
@@ -1,15 +0,0 @@
1
- {
2
- "clean_up_tokenization_spaces": true,
3
- "cls_token": "[CLS]",
4
- "do_basic_tokenize": true,
5
- "do_lower_case": false,
6
- "mask_token": "[MASK]",
7
- "model_max_length": 512,
8
- "never_split": null,
9
- "pad_token": "[PAD]",
10
- "sep_token": "[SEP]",
11
- "strip_accents": null,
12
- "tokenize_chinese_chars": true,
13
- "tokenizer_class": "BertTokenizer",
14
- "unk_token": "[UNK]"
15
- }