Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

fix(agent): sanitize Excel output sheet names (#18505)

Tyyyy02912f7d1bc4
diff --git a/agent/component/excel_processor.py b/agent/component/excel_processor.py
index df1bfaaf2c..f049acd9b4 100644
--- a/agent/component/excel_processor.py
+++ b/agent/component/excel_processor.py
@@ -23,6 +23,7 @@ Supports multiple Excel file inputs, data transformation, and Excel output gener
import logging
import os
+import re
from abc import ABC
from io import BytesIO
@@ -341,10 +342,32 @@ class ExcelProcessor(ComponentBase, ABC):
# Excel output
excel_io = BytesIO()
with pd.ExcelWriter(excel_io, engine="openpyxl") as writer:
+ used_sheet_names = set()
+ normalized_sheet_count = 0
+ deduplicated_sheet_count = 0
for sheet_name, df in dfs.items():
- # Sanitize sheet name (max 31 chars, no special chars)
- safe_name = sheet_name[:31].replace("/", "_").replace("\\", "_")
+ # Sanitize sheet name (max 31 chars, no special or reserved names)
+ original_name = str(sheet_name)
+ normalized_name = re.sub(r"[\x00-\x08\x0B\x0C\x0E-\x1F\\/*?:\[\]]", "_", original_name).strip("'")
+ if normalized_name.casefold() == "history":
+ normalized_name = f"{normalized_name}_"
+ base_name = normalized_name[:31] or "Sheet"
+ normalized_sheet_count += base_name != original_name
+ safe_name = base_name
+ counter = 2
+ if safe_name.casefold() in used_sheet_names:
+ deduplicated_sheet_count += 1
+ while safe_name.casefold() in used_sheet_names:
+ suffix = f"_{counter}"
+ safe_name = f"{base_name[: 31 - len(suffix)]}{suffix}"
+ counter += 1
+ used_sheet_names.add(safe_name.casefold())
df.to_excel(writer, sheet_name=safe_name, index=False)
+ logging.info(
+ "ExcelProcessor: Normalized %d sheet name(s) and deduplicated %d",
+ normalized_sheet_count,
+ deduplicated_sheet_count,
+ )
excel_io.seek(0)
binary_content = excel_io.read()
filename = f"{self._param.output_filename}.xlsx"
diff --git a/test/unit_test/agent/component/test_excel_processor_output.py b/test/unit_test/agent/component/test_excel_processor_output.py
new file mode 100644
index 0000000000..53b9c5741b
--- /dev/null
+++ b/test/unit_test/agent/component/test_excel_processor_output.py
@@ -0,0 +1,115 @@
+#
+# Copyright 2026 The InfiniFlow Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+import importlib.util
+import sys
+from io import BytesIO
+from pathlib import Path
+from types import ModuleType, SimpleNamespace
+
+import openpyxl
+
+
+def _load_excel_processor(monkeypatch):
+ repo_root = Path(__file__).resolve().parents[4]
+
+ class ComponentBase:
+ pass
+
+ class ComponentParamBase:
+ def __init__(self):
+ self.outputs = {}
+
+ base = ModuleType("agent.component.base")
+ base.ComponentBase = ComponentBase
+ base.ComponentParamBase = ComponentParamBase
+ monkeypatch.setitem(sys.modules, "agent.component.base", base)
+
+ file_service = ModuleType("api.db.services.file_service")
+ file_service.FileService = SimpleNamespace()
+ monkeypatch.setitem(sys.modules, "api.db.services.file_service", file_service)
+
+ api_utils = ModuleType("api.utils.api_utils")
+ api_utils.timeout = lambda _seconds: lambda func: func
+ monkeypatch.setitem(sys.modules, "api.utils.api_utils", api_utils)
+
+ storage = SimpleNamespace(put=lambda *_args, **_kwargs: None)
+ common = ModuleType("common")
+ common.settings = SimpleNamespace(STORAGE_IMPL=storage)
+ monkeypatch.setitem(sys.modules, "common", common)
+
+ misc_utils = ModuleType("common.misc_utils")
+ misc_utils.get_uuid = lambda: "attachment-id"
+ monkeypatch.setitem(sys.modules, "common.misc_utils", misc_utils)
+
+ module_path = repo_root / "agent" / "component" / "excel_processor.py"
+ spec = importlib.util.spec_from_file_location("test_excel_processor_module", module_path)
+ module = importlib.util.module_from_spec(spec)
+ monkeypatch.setitem(sys.modules, "test_excel_processor_module", module)
+ spec.loader.exec_module(module)
+ return module
+
+
+def test_output_excel_sanitizes_and_deduplicates_sheet_names(monkeypatch):
+ module = _load_excel_processor(monkeypatch)
+ stored = {}
+ module.settings.STORAGE_IMPL.put = lambda tenant_id, doc_id, data: stored.update(tenant_id=tenant_id, doc_id=doc_id, data=data)
+
+ long_sheet_name = "A" * 40
+ data = {
+ "Q1: Sales": [{"amount": 10}],
+ "Q1/ Sales": [{"amount": 20}],
+ "'Sales'": [{"amount": 30}],
+ "History": [{"amount": 40}],
+ "": [{"amount": 50}],
+ long_sheet_name: [{"amount": 60}],
+ "Nul\x00Name": [{"amount": 70}],
+ }
+ component = module.ExcelProcessor.__new__(module.ExcelProcessor)
+ component._canvas = SimpleNamespace(
+ _tenant_id="tenant-1",
+ get_variable_value=lambda _ref: data,
+ )
+ component._param = SimpleNamespace(
+ transform_data="source@data",
+ output_format="xlsx",
+ output_filename="quarterly",
+ )
+ component.set_input_value = lambda *_args, **_kwargs: None
+ outputs = {}
+ component.set_output = outputs.__setitem__
+
+ component._output_excel()
+
+ assert outputs["attachment"] == {
+ "doc_id": "attachment-id",
+ "format": "xlsx",
+ "file_name": "quarterly.xlsx",
+ }
+ assert stored["tenant_id"] == "tenant-1"
+ workbook = openpyxl.load_workbook(BytesIO(stored["data"]), read_only=True)
+ try:
+ assert workbook.sheetnames == [
+ "Q1_ Sales",
+ "Q1_ Sales_2",
+ "Sales",
+ "History_",
+ "Sheet",
+ "A" * 31,
+ "Nul_Name",
+ ]
+ finally:
+ workbook.close()