diff --git a/app/search/search_utils.py b/app/search/search_utils.py
index c648e7a9..64324657 100644
--- a/app/search/search_utils.py
+++ b/app/search/search_utils.py
@@ -233,36 +233,18 @@ def get_code_region_containing_code(
     pattern = re.compile(re.escape(code_str))
     # each occurrence is a tuple of (line_no, code_snippet) (1-based line number)
     occurrences: list[tuple[int, str]] = []
+    file_content_lines = file_content.splitlines()
     for match in pattern.finditer(file_content):
         matched_start_pos = match.start()
-        # first, find the line number of the matched start position (1-based)
-        matched_line_no = file_content.count("\n", 0, matched_start_pos) + 1
-        # next, get a few surrounding lines as context
-        search_start = match.start() - 1
-        search_end = match.end() + 1
-        # from the matched position, go left to find 5 new lines.
-        for _ in range(context_size):
-            # find the \n to the left
-            left_newline = file_content.rfind("\n", 0, search_start)
-            if left_newline == -1:
-                # no more new line to the left
-                search_start = 0
-                break
-            else:
-                search_start = left_newline
-        # go right to fine 5 new lines
-        for _ in range(context_size):
-            right_newline = file_content.find("\n", search_end + 1)
-            if right_newline == -1:
-                # no more new line to the right
-                search_end = len(file_content)
-                break
-            else:
-                search_end = right_newline
+        # first, find the line number of the matched start position (0-based)
+        matched_line_no = file_content.count("\n", 0, matched_start_pos)
+
+        window_start_index = max(0, matched_line_no - context_size)
+        window_end_index = min(
+            len(file_content_lines), matched_line_no + context_size + 1
+        )
 
-        start = max(0, search_start)
-        end = min(len(file_content), search_end)
-        context = file_content[start:end]
+        context = "\n".join(file_content_lines[window_start_index:window_end_index])
         occurrences.append((matched_line_no, context))
 
     return occurrences