fix: 修复爬虫问题

scripts/verify-pipeline.sh

@@ -0,0 +1,124 @@
+#!/usr/bin/env bash
+# 验证爬虫 → 数据库 → API → 前端 全链路
+# 用法: ./scripts/verify-pipeline.sh [--start-crawler]
+set -e
+
+API_URL="${API_URL:-http://localhost:3001}"
+CRAWLER_URL="${CRAWLER_URL:-http://localhost:8000}"
+START_CRAWLER=false
+[[ "${1:-}" = "--start-crawler" ]] && START_CRAWLER=true
+
+PROJECT_ROOT="$(cd "$(dirname "$0")/.." && pwd)"
+
+echo "=========================================="
+echo "US-Iran 态势面板 链路验证"
+echo "API: $API_URL | Crawler: $CRAWLER_URL"
+echo "=========================================="
+echo ""
+
+# 可选：启动爬虫
+if $START_CRAWLER; then
+  echo "[0/6] 启动爬虫..."
+  if curl -sf "$CRAWLER_URL/crawler/status" >/dev/null 2>&1; then
+    echo "  ✓ 爬虫已在运行"
+  else
+    cd "$PROJECT_ROOT/crawler"
+    python3 -c "import uvicorn" 2>/dev/null || { echo "  需安装: pip install uvicorn"; exit 1; }
+    uvicorn realtime_conflict_service:app --host 127.0.0.1 --port 8000 &
+    echo "  等待爬虫就绪..."
+    for i in $(seq 1 15); do
+      sleep 2
+      if curl -sf "$CRAWLER_URL/crawler/status" >/dev/null 2>&1; then
+        echo "  ✓ 爬虫已启动"
+        echo "  等待首次 RSS 抓取（约 70 秒）..."
+        sleep 70
+        break
+      fi
+    done
+    if ! curl -sf "$CRAWLER_URL/crawler/status" >/dev/null 2>&1; then
+      echo "  ✗ 爬虫启动超时"
+      exit 1
+    fi
+  fi
+  echo ""
+fi
+
+# 1. API 健康检查
+echo "[1/6] API 健康检查..."
+if curl -sf "$API_URL/api/health" > /dev/null; then
+  echo "  ✓ API 正常"
+else
+  echo "  ✗ API 无响应，请先运行: npm run api"
+  exit 1
+fi
+
+# 2. 态势数据
+echo "[2/6] 态势数据..."
+SIT=$(curl -sf "$API_URL/api/situation" 2>/dev/null || echo "{}")
+if echo "$SIT" | grep -q "lastUpdated"; then
+  echo "  ✓ 态势数据可读"
+  LAST=$(echo "$SIT" | grep -o '"lastUpdated":"[^"]*"' | head -1)
+  echo "    $LAST"
+else
+  echo "  ✗ 态势数据异常"
+  exit 1
+fi
+
+# 3. 爬虫状态
+echo "[3/6] 爬虫状态..."
+CRAWLER=$(curl -sf "$CRAWLER_URL/crawler/status" 2>/dev/null || echo "{}")
+if echo "$CRAWLER" | grep -q "db_path\|db_exists"; then
+  echo "  ✓ 爬虫服务可访问"
+  if command -v jq &>/dev/null; then
+    CNT=$(echo "$CRAWLER" | jq -r '.situation_update_count // "?"')
+    echo "    situation_update 条数: $CNT"
+  fi
+else
+  echo "  ⚠ 爬虫未启动或不可达（可选，需单独运行爬虫）"
+fi
+
+# 4. 资讯表
+echo "[4/6] 资讯表 news_content..."
+NEWS=$(curl -sf "$API_URL/api/news?limit=3" 2>/dev/null || echo '{"items":[]}')
+if echo "$NEWS" | grep -q '"items"'; then
+  if command -v jq &>/dev/null; then
+    N=$(echo "$NEWS" | jq '.items | length')
+    echo "  ✓ 最近 $N 条资讯"
+  else
+    echo "  ✓ 资讯接口可读"
+  fi
+else
+  echo "  ⚠ news_content 可能为空（爬虫未跑或刚启动）"
+fi
+
+# 5. 战损数据
+echo "[5/6] 战损数据 combat_losses..."
+if echo "$SIT" | grep -q "personnelCasualties"; then
+  echo "  ✓ 战损字段存在"
+  if command -v jq &>/dev/null; then
+    US_K=$(echo "$SIT" | jq -r '.usForces.combatLosses.personnelCasualties.killed // "?"')
+    IR_K=$(echo "$SIT" | jq -r '.iranForces.combatLosses.personnelCasualties.killed // "?"')
+    echo "    美军阵亡: $US_K | 伊朗阵亡: $IR_K"
+  fi
+else
+  echo "  ✗ 战损结构异常"
+fi
+
+# 6. 通知接口（仅验证可调用）
+echo "[6/6] 通知接口 POST /api/crawler/notify..."
+NOTIFY=$(curl -sf -X POST "$API_URL/api/crawler/notify" 2>/dev/null || echo "{}")
+if echo "$NOTIFY" | grep -q '"ok"'; then
+  echo "  ✓ 通知接口正常"
+else
+  echo "  ⚠ 通知接口可能异常"
+fi
+
+echo ""
+echo "=========================================="
+echo "验证完成。"
+echo ""
+echo "建议："
+echo "  - 访问 $API_URL/db 查看各表数据"
+echo "  - 爬虫未启动时: ./scripts/verify-pipeline.sh --start-crawler"
+echo "  - 或手动启动: cd crawler && uvicorn realtime_conflict_service:app --port 8000"
+echo "=========================================="