Browse the data

1import hashlib
2import secrets
3import json
4from typing import Optional, List
5from fastapi import FastAPI, HTTPException, Header, Body
6from pydantic import BaseModel
7import httpx
8import pdfplumber
9from io import BytesIO
10import re
11 
12app = FastAPI()
13 
14users = {}
15tokens = {}
16resumes = {}
17skills_counter = {}
18current_user_id = 1
19current_resume_id = 1
20 
21class SignupRequest(BaseModel):
22    username: str
23    password: str
24 
25class LoginRequest(BaseModel):
26    username: str
27    password: str
28 
29class UploadRequest(BaseModel):
30    pdf_url: str
31 
32def hash_password(password: str) -> str:
33    return hashlib.sha256(password.encode()).hexdigest()
34 
35def generate_token() -> str:
36    return secrets.token_hex(32)
37 
38def get_current_user(authorization: Optional[str] = Header(None)):
39    if not authorization:
40        raise HTTPException(status_code=401, detail="Missing Authorization header")
41    token = authorization.replace("Bearer ", "")
42    user_id = tokens.get(token)
43    if not user_id:
44        raise HTTPException(status_code=401, detail="Invalid token")
45    return user_id
46 
47def parse_resume_from_url(url: str) -> dict:
48    response = httpx.get(url, timeout=30)
49    response.raise_for_status()
50    
51    with pdfplumber.open(BytesIO(response.content)) as pdf:
52        text = ""
53        for page in pdf.pages:
54            text += page.extract_text() or ""
55    
56    name = ""
57    email = ""
58    skills = []
59    work_history = []
60    
61    email_match = re.search(r'[\w\.-]+@[\w\.-]+\.\w+', text)
62    if email_match:
63        email = email_match.group()
64    
65    lines = text.split('\n')
66    if lines:
67        name = lines[0].strip()
68    
69    skill_keywords = ['python', 'java', 'javascript', 'react', 'node', 'sql', 'aws', 'docker', 'kubernetes', 'git', 'html', 'css', 'typescript', 'c++', 'c#', 'ruby', 'php', 'swift', 'kotlin', 'go', 'rust', 'scala', 'perl', 'r', 'matlab', 'angular', 'vue', 'django', 'flask', 'spring', 'express', 'mongodb', 'postgresql', 'mysql', 'redis', 'kafka', 'spark', 'hadoop', 'tensorflow', 'pytorch', 'machine learning', 'deep learning', 'nlp', 'computer vision', 'data analysis', 'data science', 'devops', 'ci/cd', 'terraform', 'ansible', 'jenkins']
70    
71    text_lower = text.lower()
72    for skill in skill_keywords:
73        if skill in text_lower:
74            skills.append(skill)
75    
76    work_pattern = re.compile(r'(\d{4})\s*[-–to]+\s*(\d{4}|present)', re.IGNORECASE)
77    work_matches = work_pattern.findall(text)
78    
79    for match in work_matches:
80        company = ""
81        title = ""
82        
83        match_start = text.find(match[0])
84        if match_start > 0:
85            before_text = text[max(0, match_start-200):match_start]
86            lines_before = before_text.split('\n')
87            if len(lines_before) >= 2:
88                title = lines_before[-2].strip()
89                company = lines_before[-1].strip()
90        
91        work_history.append({
92            "company": company,
93            "title": title,
94            "start_year": match[0],
95            "end_year": match[1]
96        })
97    
98    return {
99        "name": name,
100        "email": email,
101        "skills": list(set(skills)),
102        "work_history": work_history
103    }
104 
105@app.post("/signup")
106def signup(req: SignupRequest):
107    global current_user_id
108    if req.username in users:
109        raise HTTPException(status_code=400, detail="Username already exists")
110    user_id = current_user_id
111    users[req.username] = {"id": user_id, "password_hash": hash_password(req.password)}
112    current_user_id += 1
113    token = generate_token()
114    tokens[token] = user_id
115    return {"id": user_id, "token": token}
116 
117@app.post("/login")
118def login(req: LoginRequest):
119    user = users.get(req.username)
120    if not user or user["password_hash"] != hash_password(req.password):
121        raise HTTPException(status_code=401, detail="Invalid credentials")
122    token = generate_token()
123    tokens[token] = user["id"]
124    return {"token": token}
125 
126@app.post("/resumes/upload")
127def upload_resume(req: UploadRequest, authorization: Optional[str] = Header(None)):
128    global current_resume_id
129    user_id = get_current_user(authorization)
130    
131    try:
132        parsed = parse_resume_from_url(req.pdf_url)
133    except Exception as e:
134        raise HTTPException(status_code=400, detail=f"Failed to parse PDF: {str(e)}")
135    
136    resume_id = current_resume_id
137    resumes[resume_id] = {
138        "id": resume_id,
139        "user_id": user_id,
140        "name": parsed["name"],
141        "email": parsed["email"],
142        "skills": parsed["skills"],
143        "work_history": parsed["work_history"]
144    }
145    current_resume_id += 1
146    
147    for skill in parsed["skills"]:
148        skills_counter[skill] = skills_counter.get(skill, 0) + 1
149    
150    return resumes[resume_id]
151 
152@app.get("/resumes/{resume_id}")
153def get_resume(resume_id: int, authorization: Optional[str] = Header(None)):
154    get_current_user(authorization)
155    resume = resumes.get(resume_id)
156    if not resume:
157        raise HTTPException(status_code=404, detail="Resume not found")
158    return resume
159 
160@app.get("/skills/trending")
161def get_trending_skills(authorization: Optional[str] = Header(None)):
162    get_current_user(authorization)
163    sorted_skills = sorted(skills_counter.items(), key=lambda x: x[1], reverse=True)
164    return {"trending_skills": [{"skill": s, "count": c} for s, c in sorted_skills]}

1fastapi
2uvicorn
3httpx
4pdfplumber