feat：LLM 分析文件并生成结构化建议

app/datafiles/aggregate.py

@@ -3,9 +3,18 @@
 列名、别名一律通过 quote_identifier 引用，请求里的字符串不会拼进 SQL 的其他位置
 """
 
+from pathlib import Path
+
+from app.config import settings
 from app.core.exceptions import BizException
-from app.datafiles.query import quote_identifier
-from app.schemas.data_file import AggregateRequest, DataColumn, GroupBy, Metric
+from app.datafiles.query import quote_identifier, run_query
+from app.schemas.data_file import (
+    AggregateRequest,
+    AggregateResult,
+    DataColumn,
+    GroupBy,
+    Metric,
+)
 
 # 只能用于数值列的聚合；其余聚合对列类型没有要求
 NUMBER_ONLY = {"sum", "avg"}
@@ -20,7 +29,28 @@ TEMPLATES = {
 TIME_TYPES = {"date", "datetime"}
 
 
-def build_aggregate_sql(
+def run_aggregate(
+    path: Path, request: AggregateRequest, columns: list[DataColumn]
+) -> AggregateResult:
+    """对一张表的 Parquet 执行汇总；接口和上传后的分析共用
+
+    请求不合法时抛 422；执行失败（包括文件已被删除）抛 QueryError
+    """
+    sql, names = _build_sql(request, columns)
+    result = run_query(
+        sql,
+        {"t": path},
+        timeout=settings.data_query_timeout,
+        max_rows=request.limit,
+    )
+    return AggregateResult(
+        columns=names,
+        rows=[dict(zip(names, row)) for row in result.rows],
+        truncated=result.truncated,
+    )
+
+
+def _build_sql(
     request: AggregateRequest, columns: list[DataColumn]
 ) -> tuple[str, list[str]]:
     """返回 SQL（表名固定为 t）和结果列名；请求不合法时抛 422