planet/backend/app/api/v1/dashboard.py

"""Dashboard API with caching and optimizations"""

from datetime import datetime, timedelta
from fastapi import APIRouter, Depends
from sqlalchemy import select, func, text
from sqlalchemy.ext.asyncio import AsyncSession

from app.db.session import get_db
from app.models.user import User
from app.models.datasource import DataSource
from app.models.datasource_config import DataSourceConfig
from app.models.alert import Alert, AlertSeverity
from app.models.task import CollectionTask
from app.core.security import get_current_user
from app.core.cache import cache


# Built-in collectors info (mirrored from datasources.py)
COLLECTOR_INFO = {
    "top500": {
        "id": 1,
        "name": "TOP500 Supercomputers",
        "module": "L1",
        "priority": "P0",
        "frequency_hours": 4,
    },
    "epoch_ai_gpu": {
        "id": 2,
        "name": "Epoch AI GPU Clusters",
        "module": "L1",
        "priority": "P0",
        "frequency_hours": 6,
    },
    "huggingface_models": {
        "id": 3,
        "name": "HuggingFace Models",
        "module": "L2",
        "priority": "P1",
        "frequency_hours": 12,
    },
    "huggingface_datasets": {
        "id": 4,
        "name": "HuggingFace Datasets",
        "module": "L2",
        "priority": "P1",
        "frequency_hours": 12,
    },
    "huggingface_spaces": {
        "id": 5,
        "name": "HuggingFace Spaces",
        "module": "L2",
        "priority": "P2",
        "frequency_hours": 24,
    },
    "peeringdb_ixp": {
        "id": 6,
        "name": "PeeringDB IXP",
        "module": "L2",
        "priority": "P1",
        "frequency_hours": 24,
    },
    "peeringdb_network": {
        "id": 7,
        "name": "PeeringDB Networks",
        "module": "L2",
        "priority": "P2",
        "frequency_hours": 48,
    },
    "peeringdb_facility": {
        "id": 8,
        "name": "PeeringDB Facilities",
        "module": "L2",
        "priority": "P2",
        "frequency_hours": 48,
    },
    "telegeography_cables": {
        "id": 9,
        "name": "Submarine Cables",
        "module": "L2",
        "priority": "P1",
        "frequency_hours": 168,
    },
    "telegeography_landing": {
        "id": 10,
        "name": "Cable Landing Points",
        "module": "L2",
        "priority": "P2",
        "frequency_hours": 168,
    },
    "telegeography_systems": {
        "id": 11,
        "name": "Cable Systems",
        "module": "L2",
        "priority": "P2",
        "frequency_hours": 168,
    },
}

router = APIRouter()


@router.get("/stats")
async def get_stats(
    current_user: User = Depends(get_current_user),
    db: AsyncSession = Depends(get_db),
):
    """Get dashboard statistics with caching"""
    cache_key = "dashboard:stats"

    cached_result = cache.get(cache_key)
    if cached_result:
        return cached_result

    today_start = datetime.utcnow().replace(hour=0, minute=0, second=0, microsecond=0)

    # Count built-in collectors
    built_in_count = len(COLLECTOR_INFO)
    built_in_active = built_in_count  # Built-in are always "active" for counting purposes

    # Count custom configs from database
    result = await db.execute(select(func.count(DataSourceConfig.id)))
    custom_count = result.scalar() or 0

    result = await db.execute(
        select(func.count(DataSourceConfig.id)).where(DataSourceConfig.is_active == True)
    )
    custom_active = result.scalar() or 0

    # Total datasources
    total_datasources = built_in_count + custom_count
    active_datasources = built_in_active + custom_active

    # Tasks today (from database)
    result = await db.execute(
        select(func.count(CollectionTask.id)).where(CollectionTask.started_at >= today_start)
    )
    tasks_today = result.scalar() or 0

    result = await db.execute(
        select(func.count(CollectionTask.id)).where(
            CollectionTask.status == "success",
            CollectionTask.started_at >= today_start,
        )
    )
    success_tasks = result.scalar() or 0
    success_rate = (success_tasks / tasks_today * 100) if tasks_today > 0 else 0

    # Alerts
    result = await db.execute(
        select(func.count(Alert.id)).where(
            Alert.severity == AlertSeverity.CRITICAL,
            Alert.status == "active",
        )
    )
    critical_alerts = result.scalar() or 0

    result = await db.execute(
        select(func.count(Alert.id)).where(
            Alert.severity == AlertSeverity.WARNING,
            Alert.status == "active",
        )
    )
    warning_alerts = result.scalar() or 0

    result = await db.execute(
        select(func.count(Alert.id)).where(
            Alert.severity == AlertSeverity.INFO,
            Alert.status == "active",
        )
    )
    info_alerts = result.scalar() or 0

    response = {
        "total_datasources": total_datasources,
        "active_datasources": active_datasources,
        "tasks_today": tasks_today,
        "success_rate": round(success_rate, 1),
        "last_updated": datetime.utcnow().isoformat(),
        "alerts": {
            "critical": critical_alerts,
            "warning": warning_alerts,
            "info": info_alerts,
        },
    }

    cache.set(cache_key, response, expire_seconds=60)

    return response


@router.get("/summary")
async def get_summary(
    current_user: User = Depends(get_current_user),
    db: AsyncSession = Depends(get_db),
):
    """Get dashboard summary by module with caching"""
    cache_key = "dashboard:summary"

    cached_result = cache.get(cache_key)
    if cached_result:
        return cached_result

    # Count by module for built-in collectors
    builtin_by_module = {}
    for name, info in COLLECTOR_INFO.items():
        module = info["module"]
        if module not in builtin_by_module:
            builtin_by_module[module] = {"datasources": 0, "sources": []}
        builtin_by_module[module]["datasources"] += 1
        builtin_by_module[module]["sources"].append(info["name"])

    # Count custom configs by module (default to L3 for custom)
    result = await db.execute(
        select(DataSourceConfig.source_type, func.count(DataSourceConfig.id).label("count"))
        .where(DataSourceConfig.is_active == True)
        .group_by(DataSourceConfig.source_type)
    )
    custom_rows = result.fetchall()

    for row in custom_rows:
        source_type = row.source_type
        module = "L3"  # Custom configs default to L3
        if module not in builtin_by_module:
            builtin_by_module[module] = {"datasources": 0, "sources": []}
        builtin_by_module[module]["datasources"] += row.count
        builtin_by_module[module]["sources"].append(f"自定义 ({source_type})")

    summary = {}
    for module, data in builtin_by_module.items():
        summary[module] = {
            "datasources": data["datasources"],
            "total_records": 0,  # Built-in don't track this in dashboard stats
            "last_updated": datetime.utcnow().isoformat(),
        }

    response = {"modules": summary, "last_updated": datetime.utcnow().isoformat()}

    cache.set(cache_key, response, expire_seconds=300)

    return response