From 67601e64945535859a5a1e7fccf1f429afd99cdf Mon Sep 17 00:00:00 2001
From: kris tan <145995558+KrisWuli006@users.noreply.github.com>
Date: Mon, 14 Sep 2026 11:35:48 +0800
Subject: [PATCH] fix(datasource): filter Hive partition metadata from table
 fields

Ignore blank names and DESCRIBE metadata headers. Retain partition columns once in their original order, preserve valid hash-prefixed column names, and keep support for extra driver columns.

Refs dataease/SQLBot#1250.
---
 backend/apps/db/db.py | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/backend/apps/db/db.py b/backend/apps/db/db.py
index 0b88f15a..b8ca4dba 100644
--- a/backend/apps/db/db.py
+++ b/backend/apps/db/db.py
@@ -598,7 +598,19 @@ def get_fields(ds: CoreDatasource, table_name: str = None):
             with get_driver_pool(ds).connection() as conn, conn.cursor() as cursor:
                 cursor.execute(sql)
                 res = cursor.fetchall()
-                res_list = [ColumnSchema(*item[:3]) for item in res]
+                res_list = []
+                seen_fields = set()
+                for item in res:
+                    field_name = item[0]
+                    if not field_name or not field_name.strip():
+                        continue
+                    # DESCRIBE includes section headings and repeats partition columns.
+                    if field_name.lstrip().startswith('#') and (item[1] or '').strip() in ('', 'data_type'):
+                        continue
+                    if field_name in seen_fields:
+                        continue
+                    seen_fields.add(field_name)
+                    res_list.append(ColumnSchema(*item[:3]))
                 return res_list