From 077773a9675fe5467661e8ffa545bd8896b68e72 Mon Sep 17 00:00:00 2001
From: Luck <2115759892@qq.com>
Date: Fri, 13 Nov 2026 23:17:18 +0800
Subject: [PATCH] fix(sandbox): match local grep lines without their terminator
 (#8210)

* fix(sandbox): match local grep lines without their terminator

find_grep_matches ran the regex on each line with its trailing "\n", so
patterns that can consume the newline (\s$, [^a-z]$, \n) matched every
newline-ended line while the returned line was shown stripped. Strip the
LF or CRLF terminator before matching, as grep(1) and the remote sandbox
providers do.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* docs(sandbox): clarify local grep CRLF handling

* test(sandbox): cover literal grep line terminators

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
---
 .../harness/deerflow/sandbox/search.py        |  9 +++++---
 backend/tests/test_sandbox_line_endings.py    | 22 +++++++++++++++++++
 2 files changed, 28 insertions(+), 3 deletions(-)

diff --git a/backend/packages/harness/deerflow/sandbox/search.py b/backend/packages/harness/deerflow/sandbox/search.py
index 634d12aabe..bb7588d24c 100644
--- a/backend/packages/harness/deerflow/sandbox/search.py
+++ b/backend/packages/harness/deerflow/sandbox/search.py
@@ -311,12 +311,15 @@ def find_grep_matches(
                 continue
             # newline="\n" ends lines where LocalSandbox.read_file does, so a
             # hit's line_number is the start_line that reads it back; a bare
-            # "\r" stays content. A CRLF terminator still reads as "\n", so $
-            # anchors match as before.
+            # "\r" stays content. Like native grep on Unix, local matching
+            # excludes LF. LocalSandbox also strips the CR in CRLF, so $ anchors
+            # at the end of the line text for both line endings.
             with file_path.open(encoding="utf-8", errors="replace", newline="\n") as handle:
                 for line_number, line in enumerate(handle, start=1):
                     if line.endswith("\r\n"):
-                        line = line[:-2] + "\n"
+                        line = line[:-2]
+                    elif line.endswith("\n"):
+                        line = line[:-1]
                     if not literal and len(line) > _max_line_chars:
                         continue
                     if matches_line(line):
diff --git a/backend/tests/test_sandbox_line_endings.py b/backend/tests/test_sandbox_line_endings.py
index 70aa950517..79ff49f4fd 100644
--- a/backend/tests/test_sandbox_line_endings.py
+++ b/backend/tests/test_sandbox_line_endings.py
@@ -15,6 +15,8 @@ from pathlib import Path
 from types import SimpleNamespace
 from unittest.mock import MagicMock
 
+import pytest
+
 import deerflow.sandbox.local.local_sandbox as local_sandbox
 from deerflow.sandbox.local.local_sandbox import LocalSandbox
 from deerflow.sandbox.read_file_contract import count_file_lines
@@ -216,3 +218,23 @@ def test_read_before_write_content_changes_when_only_line_endings_change(tmp_pat
     after = read_current_file_content(runtime, "/mnt/user-data/outputs/file.txt")
 
     assert before != after
+
+
+@pytest.mark.parametrize(
+    ("pattern", "literal", "expected"),
+    [
+        (r"\s$", False, [(2, "trailing ")]),
+        (r"[^a-z]$", False, [(2, "trailing ")]),
+        (r"\n", False, []),
+        pytest.param("\n", True, [], id="literal-newline"),
+    ],
+)
+def test_grep_does_not_match_the_line_terminator(tmp_path, pattern, literal, expected) -> None:
+    # LocalSandbox excludes LF and CRLF terminators from matching, so
+    # a trailing-whitespace search only matches whitespace in the line text.
+    path = tmp_path / "ws.txt"
+    path.write_bytes(b"clean\ntrailing \r\nalso clean\r\nlast clean")
+
+    matches, _ = LocalSandbox("t").grep(str(path), pattern, literal=literal)
+
+    assert [(match.line_number, match.line) for match in matches] == expected