fix(artifacts): 修复文件名差异导致 Artifact not found

backend/app/gateway/routers/artifacts.py

@@ -1,5 +1,7 @@
 import logging
 import mimetypes
+import re
+import unicodedata
 import zipfile
 from pathlib import Path
 from urllib.parse import quote
@@ -19,6 +21,9 @@ ACTIVE_CONTENT_MIME_TYPES = {
     "image/svg+xml",
 }
 
+_DASH_VARIANTS_RE = re.compile(r"\s*[-\u2010\u2011\u2012\u2013\u2014\u2212]\s*")
+_WHITESPACE_RE = re.compile(r"\s+")
+
 
 def _build_content_disposition(disposition_type: str, filename: str) -> str:
     """Build an RFC 5987 encoded Content-Disposition header value."""
@@ -32,6 +37,31 @@ def _build_attachment_headers(filename: str, extra_headers: dict[str, str] | Non
     return headers
 
 
+def _canonicalize_filename_for_lookup(filename: str) -> str:
+    """Canonical form used for conservative compatibility lookup."""
+    normalized = unicodedata.normalize("NFKC", filename).strip()
+    normalized = _DASH_VARIANTS_RE.sub("-", normalized)
+    normalized = _WHITESPACE_RE.sub(" ", normalized)
+    return normalized
+
+
+def _find_compat_filename_match(missing_path: Path) -> Path | None:
+    """Find a same-directory file whose canonicalized name uniquely matches."""
+    parent = missing_path.parent
+    if not parent.is_dir():
+        return None
+
+    target_name = _canonicalize_filename_for_lookup(missing_path.name)
+    matches: list[Path] = []
+    for candidate in parent.iterdir():
+        if not candidate.is_file():
+            continue
+        if _canonicalize_filename_for_lookup(candidate.name) == target_name:
+            matches.append(candidate)
+
+    return matches[0] if len(matches) == 1 else None
+
+
 def is_text_file_by_content(path: Path, sample_size: int = 8192) -> bool:
     """Check if file is text by examining content for null bytes."""
     try:
@@ -157,7 +187,15 @@ async def get_artifact(thread_id: str, path: str, request: Request, download: bo
     logger.info(f"Resolving artifact path: thread_id={thread_id}, requested_path={path}, actual_path={actual_path}")
 
     if not actual_path.exists():
-        raise HTTPException(status_code=404, detail=f"Artifact not found: {path}")
+        compat_path = _find_compat_filename_match(actual_path)
+        if compat_path is None:
+            raise HTTPException(status_code=404, detail=f"Artifact not found: {path}")
+        logger.info(
+            "Artifact compatibility fallback applied: requested_path=%s, resolved_path=%s",
+            actual_path,
+            compat_path,
+        )
+        actual_path = compat_path
 
     if not actual_path.is_file():
         raise HTTPException(status_code=400, detail=f"Path is not a file: {path}")

backend/packages/harness/deerflow/tools/builtins/present_file_tool.py

@@ -56,6 +56,11 @@ def _normalize_presented_filepath(
     except ValueError as exc:
         raise ValueError(f"Only files in {OUTPUTS_VIRTUAL_PREFIX} can be presented: {filepath}") from exc
 
+    if not actual_path.exists():
+        raise ValueError(f"File does not exist: {filepath}")
+    if not actual_path.is_file():
+        raise ValueError(f"Path is not a file: {filepath}")
+
     return f"{OUTPUTS_VIRTUAL_PREFIX}/{relative_path.as_posix()}"
 
 

backend/tests/test_artifacts_router.py

@@ -117,3 +117,16 @@ def test_get_artifact_pdf_with_no_null_bytes_and_non_utf8_content_is_served_inli
     assert bytes(response.body) == binary_content
     assert response.media_type == "application/pdf"
     assert response.headers.get("content-disposition", "").startswith("inline;")
+
+
+def test_get_artifact_compat_fallback_for_dash_spacing(tmp_path, monkeypatch) -> None:
+    artifact_path = tmp_path / "xhs-note-唯-疲劳端茶.md"
+    artifact_path.write_text("ok", encoding="utf-8")
+    requested_path = tmp_path / "xhs-note-唯 - 疲劳端茶.md"
+
+    monkeypatch.setattr(artifacts_router, "resolve_thread_virtual_path", lambda _thread_id, _path: requested_path)
+
+    response = asyncio.run(artifacts_router.get_artifact("thread-1", "mnt/user-data/outputs/xhs-note-唯 - 疲劳端茶.md", _make_request()))
+
+    assert bytes(response.body).decode("utf-8") == "ok"
+    assert response.media_type == "text/markdown"

backend/tests/test_present_file_tool_core_logic.py

@@ -66,3 +66,18 @@ def test_present_files_rejects_paths_outside_outputs(tmp_path):
 
     assert "artifacts" not in result.update
     assert result.update["messages"][0].content == f"Error: Only files in /mnt/user-data/outputs can be presented: {leaked_path}"
+
+
+def test_present_files_rejects_nonexistent_file_in_outputs(tmp_path):
+    outputs_dir = tmp_path / "threads" / "thread-1" / "user-data" / "outputs"
+    outputs_dir.mkdir(parents=True)
+    missing_path = outputs_dir / "missing.md"
+
+    result = present_file_tool_module.present_file_tool.func(
+        runtime=_make_runtime(str(outputs_dir)),
+        filepaths=[str(missing_path)],
+        tool_call_id="tc-4",
+    )
+
+    assert "artifacts" not in result.update
+    assert result.update["messages"][0].content == f"Error: File does not exist: {missing_path}"