rssKeeper/backend/routers/external_api.py

"""对外 API（供 AI/外部系统调用）"""
from typing import Optional, List
from datetime import datetime, timedelta
from fastapi import APIRouter, Depends, Query
from sqlalchemy.orm import Session
from sqlalchemy import desc, or_
from database import get_db
from models import Article, Feed
from fulltext_search import search_articles

router = APIRouter(tags=["external"])


@router.get("/recent")
def get_recent_articles(
    hours: int = 24,
    limit: int = 50,
    feed_id: Optional[int] = None,
    category: Optional[str] = None,
    search: Optional[str] = None,
    unread_only: bool = False,
    db: Session = Depends(get_db),
):
    """获取最近 N 小时的文章
    供 AI 分析的主要接口，支持多条件组合筛选
    """
    since = datetime.utcnow() - timedelta(hours=hours)

    query = db.query(Article, Feed.title.label("feed_title"), Feed.category.label("category")).join(Feed)
    query = query.filter(Article.created_at >= since)

    if feed_id:
        query = query.filter(Article.feed_id == feed_id)
    if category:
        query = query.filter(Feed.category == category)
    if search:
        query = query.filter(
            Article.title.contains(search) | Article.summary.contains(search)
        )
    if unread_only:
        query = query.filter(Article.is_read == False)

    rows = query.order_by(desc(Article.published_at)).limit(limit).all()

    return {
        "query": {
            "hours": hours,
            "limit": limit,
            "feed_id": feed_id,
            "category": category,
            "search": search,
            "unread_only": unread_only,
        },
        "count": len(rows),
        "articles": [
            {
                "id": article.id,
                "title": article.title or "",
                "link": article.link,
                "author": article.author or "",
                "summary": article.summary or "",
                "content": article.content or "" if len(article.content or "") < 10000 else article.summary or "",
                "published_at": article.published_at.isoformat() if article.published_at else None,
                "created_at": article.created_at.isoformat(),
                "feed_title": feed_title or "",
                "category": category or "",
            }
            for article, feed_title, category in rows
        ],
    }


@router.get("/search")
def fulltext_search(
    q: str = Query(..., description="搜索关键词"),
    limit: int = Query(50, ge=1, le=200),
    offset: int = Query(0, ge=0),
    category: Optional[str] = Query(None, description="按分类筛选"),
    feed_id: Optional[int] = Query(None, description="按源筛选"),
    db: Session = Depends(get_db),
):
    """全文搜索文章（FTS5）
    供 AI 按关键词检索文章内容
    """
    results, total = search_articles(q, limit, offset)

    # 二次过滤分类和源
    if category or feed_id:
        filtered = []
        for r in results:
            if category and r["category"] != category:
                continue
            if feed_id and r["feed_id"] != feed_id:
                continue
            filtered.append(r)
        results = filtered
        total = len(filtered)

    return {
        "query": q,
        "total": total,
        "offset": offset,
        "limit": limit,
        "articles": results,
    }


@router.get("/feeds")
def get_active_feeds(
    health_status: Optional[str] = Query(None, description="按健康度筛选: healthy/warning/unhealthy/unknown"),
    category: Optional[str] = Query(None, description="按分类筛选"),
    error_type: Optional[str] = Query(None, description="按错误类型筛选"),
    is_active: Optional[bool] = Query(None, description="按启用状态筛选"),
    db: Session = Depends(get_db),
):
    """获取 RSS 源列表（支持多条件筛选）"""
    query = db.query(Feed)

    if is_active is not None:
        query = query.filter(Feed.is_active == is_active)
    else:
        query = query.filter(Feed.is_active == True)

    if category:
        query = query.filter(Feed.category == category)

    feeds = query.all()

    results = []
    for feed in feeds:
        status = feed.health_status()
        if health_status and status != health_status:
            continue
        if error_type and feed.error_type != error_type:
            continue

        results.append({
            "id": feed.id,
            "title": feed.title or feed.url,
            "url": feed.url,
            "category": feed.category or "",
            "is_active": feed.is_active,
            "health_status": status,
            "error_type": feed.error_type,
            "article_count": feed.article_count,
            "last_fetch_at": feed.last_fetch_at.isoformat() if feed.last_fetch_at else None,
            "last_error": feed.last_error or "",
        })

    return {
        "count": len(results),
        "feeds": results,
    }


@router.get("/feeds/{feed_id}/articles")
def get_feed_articles(
    feed_id: int,
    limit: int = 100,
    since: Optional[str] = None,
    search: Optional[str] = None,
    unread_only: bool = False,
    db: Session = Depends(get_db),
):
    """获取指定 RSS 源的文章"""
    feed = db.query(Feed).filter(Feed.id == feed_id).first()
    if not feed:
        return {"error": "Feed not found"}

    query = db.query(Article).filter(Article.feed_id == feed_id)

    if since:
        query = query.filter(Article.published_at >= since)
    if search:
        query = query.filter(
            Article.title.contains(search) | Article.summary.contains(search)
        )
    if unread_only:
        query = query.filter(Article.is_read == False)

    articles = query.order_by(desc(Article.published_at)).limit(limit).all()

    return {
        "feed": {
            "id": feed.id,
            "title": feed.title or feed.url,
            "url": feed.url,
        },
        "count": len(articles),
        "articles": [
            {
                "id": article.id,
                "title": article.title or "",
                "link": article.link,
                "author": article.author or "",
                "summary": article.summary or "",
                "published_at": article.published_at.isoformat() if article.published_at else None,
            }
            for article in articles
        ],
    }


@router.get("/summary")
def get_daily_summary(
    date: Optional[str] = None,
    category: Optional[str] = None,
    db: Session = Depends(get_db),
):
    """获取指定日期的文章摘要统计
    供 AI 快速了解某天的 RSS 内容概况
    """
    if date:
        try:
            day = datetime.strptime(date, "%Y-%m-%d")
            next_day = day + timedelta(days=1)
        except ValueError:
            return {"error": "Invalid date format, use YYYY-MM-DD"}
    else:
        day = datetime.utcnow().replace(hour=0, minute=0, second=0, microsecond=0)
        next_day = day + timedelta(days=1)

    query = db.query(Article, Feed.title.label("feed_title"), Feed.category.label("category")).join(Feed)
    query = query.filter(Article.created_at >= day, Article.created_at < next_day)
    if category:
        query = query.filter(Feed.category == category)

    rows = query.order_by(desc(Article.published_at)).all()

    by_category = {}
    for article, feed_title, cat in rows:
        c = cat or "未分类"
        if category and c != category:
            continue
        if c not in by_category:
            by_category[c] = []
        by_category[c].append({
            "title": article.title or "",
            "link": article.link,
            "feed": feed_title or "",
            "summary": article.summary or "",
        })

    return {
        "date": day.strftime("%Y-%m-%d"),
        "total_articles": len(rows),
        "by_category": by_category,
    }