backend/app/scripts/seed.py

"""
Seed database with sample data for local development.
Run: python -m app.scripts.seed
"""
import asyncio
import uuid
from datetime import datetime, timedelta, timezone

from sqlalchemy import select

from app.core.database import async_session_factory
from app.core.security import hash_password

# Import all ORM modules so SQLAlchemy string relationships resolve on Windows/local runs.
import app.models.application  # noqa: F401
import app.models.resume  # noqa: F401
import app.models.alerts  # noqa: F401
import app.models.llm  # noqa: F401
import app.api.routes.bookmarks  # noqa: F401
import app.api.routes.cover_letters  # noqa: F401
import app.api.routes.reminders  # noqa: F401
import app.api.routes.qa_bank  # noqa: F401
import app.api.routes.webhooks  # noqa: F401
import app.services.analytics  # noqa: F401

from app.models.job import Company, Job
from app.models.scraper import JobSource
from app.models.user import Profile, User


SAMPLE_COMPANIES = [
    {"name": "Google", "slug": "google", "domain": "google.com", "industry": "Technology", "size_range": "10000+", "headquarters": "Mountain View, CA", "company_type": "public", "tech_stack": ["C++", "Python", "Go", "Kubernetes", "TensorFlow"]},
    {"name": "Stripe", "slug": "stripe", "domain": "stripe.com", "industry": "Fintech", "size_range": "5001-10000", "headquarters": "San Francisco, CA", "company_type": "private", "tech_stack": ["Ruby", "Java", "Scala", "PostgreSQL"]},
    {"name": "Vercel", "slug": "vercel", "domain": "vercel.com", "industry": "Developer Tools", "size_range": "201-500", "headquarters": "San Francisco, CA", "company_type": "startup", "tech_stack": ["TypeScript", "Next.js", "React", "Rust"]},
    {"name": "Anthropic", "slug": "anthropic", "domain": "anthropic.com", "industry": "AI/ML", "size_range": "501-1000", "headquarters": "San Francisco, CA", "company_type": "startup", "tech_stack": ["Python", "PyTorch", "Kubernetes", "Rust"]},
    {"name": "Linear", "slug": "linear", "domain": "linear.app", "industry": "Developer Tools", "size_range": "51-200", "headquarters": "San Francisco, CA", "company_type": "startup", "tech_stack": ["TypeScript", "React", "GraphQL", "PostgreSQL"]},
    {"name": "Figma", "slug": "figma", "domain": "figma.com", "industry": "Design", "size_range": "1001-5000", "headquarters": "San Francisco, CA", "company_type": "private", "tech_stack": ["TypeScript", "React", "WebGL", "Rust"]},
    {"name": "Datadog", "slug": "datadog", "domain": "datadoghq.com", "industry": "DevOps", "size_range": "5001-10000", "headquarters": "New York, NY", "company_type": "public", "tech_stack": ["Go", "Python", "Kubernetes", "Kafka"]},
    {"name": "Notion", "slug": "notion", "domain": "notion.so", "industry": "Productivity", "size_range": "501-1000", "headquarters": "San Francisco, CA", "company_type": "private", "tech_stack": ["TypeScript", "React", "Node.js", "PostgreSQL"]},
]

SAMPLE_JOBS = [
    {"title": "Senior Frontend Engineer", "company": "Vercel", "remote_type": "remote", "employment_type": "full_time", "seniority_level": "senior", "salary_min": 180000, "salary_max": 250000, "skills_required": ["React", "Next.js", "TypeScript", "CSS"], "description_text": "Join our frontend team to build the future of web development. You'll work on Next.js, Turbopack, and our deployment platform used by millions of developers."},
    {"title": "Staff Software Engineer - AI Infrastructure", "company": "Anthropic", "remote_type": "hybrid", "employment_type": "full_time", "seniority_level": "senior", "salary_min": 300000, "salary_max": 450000, "skills_required": ["Python", "Distributed Systems", "ML Infrastructure", "Kubernetes"], "description_text": "Build the infrastructure powering Claude. Work on training pipelines, inference systems, and developer APIs at scale."},
    {"title": "Product Engineer", "company": "Linear", "remote_type": "remote", "employment_type": "full_time", "seniority_level": "mid", "salary_min": 150000, "salary_max": 200000, "skills_required": ["React", "TypeScript", "Node.js", "PostgreSQL"], "description_text": "Build features end-to-end for our project management tool used by the best product teams. Own features from conception to deployment."},
    {"title": "Backend Engineer - Payments", "company": "Stripe", "remote_type": "hybrid", "employment_type": "full_time", "seniority_level": "mid", "salary_min": 175000, "salary_max": 250000, "skills_required": ["Ruby", "Java", "Distributed Systems", "SQL"], "description_text": "Build the economic infrastructure for the internet. Work on payment processing, fraud detection, and financial services APIs."},
    {"title": "Senior Software Engineer - Search", "company": "Google", "remote_type": "onsite", "employment_type": "full_time", "seniority_level": "senior", "salary_min": 200000, "salary_max": 350000, "skills_required": ["C++", "Python", "Distributed Systems", "Information Retrieval"], "description_text": "Improve search quality for billions of users. Work on ranking algorithms, knowledge graphs, and AI-powered features."},
    {"title": "Full Stack Engineer", "company": "Notion", "remote_type": "hybrid", "employment_type": "full_time", "seniority_level": "mid", "salary_min": 160000, "salary_max": 220000, "skills_required": ["React", "TypeScript", "Node.js", "PostgreSQL"], "description_text": "Build collaborative features for Notion's workspace platform. Work across the stack from real-time sync to UI components."},
    {"title": "Design Engineer", "company": "Figma", "remote_type": "hybrid", "employment_type": "full_time", "seniority_level": "mid", "salary_min": 170000, "salary_max": 240000, "skills_required": ["React", "TypeScript", "WebGL", "Design Systems"], "description_text": "Bridge design and engineering at Figma. Build interactive UI components, design system primitives, and creative tools."},
    {"title": "Senior Site Reliability Engineer", "company": "Datadog", "remote_type": "remote", "employment_type": "full_time", "seniority_level": "senior", "salary_min": 190000, "salary_max": 280000, "skills_required": ["Kubernetes", "Terraform", "Go", "Observability"], "description_text": "Keep Datadog running at scale. Own service reliability, incident response, and infrastructure automation for our monitoring platform."},
    {"title": "Machine Learning Engineer - NLP", "company": "Google", "remote_type": "onsite", "employment_type": "full_time", "seniority_level": "senior", "salary_min": 220000, "salary_max": 380000, "skills_required": ["Python", "PyTorch", "Transformers", "NLP"], "description_text": "Work on large language models and NLP systems. Apply cutting-edge research to production systems serving billions of queries."},
    {"title": "Software Engineering Intern", "company": "Stripe", "remote_type": "hybrid", "employment_type": "internship", "seniority_level": "intern", "salary_min": 80000, "salary_max": 100000, "skills_required": ["Python", "JavaScript", "SQL"], "description_text": "12-week summer internship. Work on real projects alongside experienced engineers. Past interns have shipped features used by millions."},
]

SAMPLE_SOURCES = [
    {"name": "Greenhouse - Vercel", "source_type": "ats_api", "domain": "vercel.com", "base_url": "https://boards-api.greenhouse.io/v1/boards/vercel/jobs", "ats_vendor": "greenhouse", "ats_config": {"board_token": "vercel"}},
    {"name": "Greenhouse - Anthropic", "source_type": "ats_api", "domain": "anthropic.com", "base_url": "https://boards-api.greenhouse.io/v1/boards/anthropic/jobs", "ats_vendor": "greenhouse", "ats_config": {"board_token": "anthropic"}},
    {"name": "Lever - Linear", "source_type": "ats_api", "domain": "linear.app", "base_url": "https://api.lever.co/v0/postings/linear", "ats_vendor": "lever", "ats_config": {"company_slug": "linear"}},
    {"name": "Ashby - Notion", "source_type": "ats_api", "domain": "notion.so", "base_url": "https://api.ashbyhq.com/posting-api/job-board", "ats_vendor": "ashby", "ats_config": {"board_slug": "notion"}},
]


async def seed():
    """Seed the database with sample data."""
    async with async_session_factory() as db:
        print("🌱 Seeding database...")

        # Avoid duplicate seed runs.
        existing = await db.scalar(select(User).where(User.email == "demo@jobportal.dev"))
        if existing:
            print("✅ Seed data already exists. Skipping.")
            return

        # 1. Create admin user
        admin = User(
            email="admin@jobportal.dev",
            hashed_password=hash_password("admin123"),
            full_name="Admin User",
            role="admin",
            is_active=True,
            is_verified=True,
            is_superuser=True,
        )
        db.add(admin)

        # 2. Create demo user
        demo = User(
            email="demo@jobportal.dev",
            hashed_password=hash_password("demo123"),
            full_name="Demo User",
            role="candidate",
            is_active=True,
            is_verified=True,
        )
        db.add(demo)
        await db.flush()

        # 3. Create demo profile
        profile = Profile(
            user_id=demo.id,
            headline="Full Stack Engineer",
            location="San Francisco, CA",
            summary="Passionate full-stack engineer with 5 years of experience building web applications.",
            skills=[
                {"name": "React", "category": "Frameworks", "level": 5},
                {"name": "TypeScript", "category": "Languages", "level": 5},
                {"name": "Python", "category": "Languages", "level": 4},
                {"name": "Node.js", "category": "Frameworks", "level": 4},
                {"name": "PostgreSQL", "category": "Databases", "level": 4},
                {"name": "AWS", "category": "Cloud", "level": 3},
                {"name": "Docker", "category": "Tools", "level": 4},
                {"name": "Next.js", "category": "Frameworks", "level": 5},
            ],
            experience=[
                {
                    "company": "TechCorp",
                    "title": "Senior Software Engineer",
                    "location": "San Francisco, CA",
                    "start_date": "2022-01",
                    "end_date": "",
                    "is_current": True,
                    "bullets": [
                        "Led development of real-time collaboration features serving 100K+ daily active users",
                        "Reduced page load time by 40% through code splitting and performance optimization",
                        "Mentored 3 junior engineers and led weekly architecture reviews",
                    ],
                },
            ],
            remote_preference="remote",
            salary_min=180000,
            salary_max=250000,
            salary_currency="USD",
            open_to_work=True,
        )
        db.add(profile)

        # 4. Create companies
        company_map = {}
        for c in SAMPLE_COMPANIES:
            company = Company(**c)
            db.add(company)
            company_map[c["name"]] = company
        await db.flush()

        # 5. Create jobs
        now = datetime.now(timezone.utc)
        for i, j in enumerate(SAMPLE_JOBS):
            company = company_map.get(j["company"])
            job = Job(
                title=j["title"],
                company_id=company.id if company else None,
                company_name=j["company"],
                source_url=f"https://{company.domain}/careers/{uuid.uuid4().hex[:8]}" if company else "https://example.com/jobs/1",
                source_domain=company.domain if company else "example.com",
                source_type="seed",
                remote_type=j.get("remote_type"),
                employment_type=j.get("employment_type"),
                seniority_level=j.get("seniority_level"),
                salary_min=j.get("salary_min"),
                salary_max=j.get("salary_max"),
                salary_currency="USD",
                salary_interval="yearly",
                skills_required=j.get("skills_required"),
                description_text=j.get("description_text"),
                status="active",
                date_posted=now - timedelta(days=i * 2),
                locations=[{"raw": "San Francisco, CA", "city": "San Francisco", "state": "CA", "country": "US"}],
            )
            db.add(job)

        # 6. Create sources
        for s in SAMPLE_SOURCES:
            source = JobSource(**s)
            db.add(source)

        await db.commit()
        print("✅ Database seeded successfully!")
        print("   Admin: admin@jobportal.dev / admin123")
        print("   Demo:  demo@jobportal.dev / demo123")
        print(f"   Companies: {len(SAMPLE_COMPANIES)}")
        print(f"   Jobs: {len(SAMPLE_JOBS)}")
        print(f"   Sources: {len(SAMPLE_SOURCES)}")


if __name__ == "__main__":
    asyncio.run(seed())