From 09a3854ed82e3f40932a6a0b4a5506a29e9eaf9c Mon Sep 17 00:00:00 2001
From: Kevin Hu <kevinhu.sh@gmail.com>
Date: Mon, 8 Dec 2025 14:28:23 +0800
Subject: [PATCH] Fix: chunk method error. (#11807)

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
---
 deepdoc/parser/excel_parser.py |  4 ++--
 rag/app/laws.py                | 21 ++++++++++++++++-----
 2 files changed, 18 insertions(+), 7 deletions(-)

diff --git a/deepdoc/parser/excel_parser.py b/deepdoc/parser/excel_parser.py
index 868fc5f41..906de8135 100644
--- a/deepdoc/parser/excel_parser.py
+++ b/deepdoc/parser/excel_parser.py
@@ -41,7 +41,7 @@ class RAGFlowExcelParser:
 
             try:
                 file_like_object.seek(0)
-                df = pd.read_csv(file_like_object)
+                df = pd.read_csv(file_like_object, on_bad_lines='skip')
                 return RAGFlowExcelParser._dataframe_to_workbook(df)
 
             except Exception as e_csv:
@@ -164,7 +164,7 @@ class RAGFlowExcelParser:
         except Exception as e:
             logging.warning(f"Parse spreadsheet error: {e}, trying to interpret as CSV file")
             file_like_object.seek(0)
-            df = pd.read_csv(file_like_object)
+            df = pd.read_csv(file_like_object, on_bad_lines='skip')
         df = df.replace(r"^\s*$", "", regex=True)
         return df.to_markdown(index=False)
 
diff --git a/rag/app/laws.py b/rag/app/laws.py
index ba2592833..e09bb4d67 100644
--- a/rag/app/laws.py
+++ b/rag/app/laws.py
@@ -201,12 +201,23 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
 
     elif re.search(r"\.doc$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")
-        binary = BytesIO(binary)
-        doc_parsed = parser.from_buffer(binary)
-        sections = doc_parsed['content'].split('\n')
-        sections = [s for s in sections if s]
-        callback(0.8, "Finish parsing.")
+        try:
+            from tika import parser as tika_parser
+        except Exception as e:
+            callback(0.8, f"tika not available: {e}. Unsupported .doc parsing.")
+            logging.warning(f"tika not available: {e}. Unsupported .doc parsing for {filename}.")
+            return []
 
+        binary = BytesIO(binary)
+        doc_parsed = tika_parser.from_buffer(binary)
+        if doc_parsed.get('content', None) is not None:
+            sections = doc_parsed['content'].split('\n')
+            sections = [s for s in sections if s]
+            callback(0.8, "Finish parsing.")
+        else:
+            callback(0.8, f"tika.parser got empty content from {filename}.")
+            logging.warning(f"tika.parser got empty content from {filename}.")
+            return []
     else:
         raise NotImplementedError(
             "file type not supported yet(doc, docx, pdf, txt supported)")