Files
ragflow/api/utils/file_response.py
monsterDavid 7da4f200e5 fix(agent): enable MCP file preview via doc_id (#15399)
## Summary
MCP-wrapped agents could only force-download files looked up by
`doc_id`. This adds an explicit preview path and inline response headers
for previewable file types.

- **New** `GET /api/v1/agents/attachments/{attachment_id}/preview` —
inline preview for PDFs, images, and other safe types (pass `ext` and/or
`mime_type`)
- **Improved** `GET /api/v1/documents/{doc_id}/preview` — sets inline
disposition using the document filename
- **Improved** attachment download routing — resolves `mime_type` /
`ext` query params (no default `markdown`), supports
`disposition=inline`
- **DocGenerator output** — includes URL-encoded `preview_url` for MCP
clients
- **Legacy `/document/download/...` aliases** — still use download
semantics; MCP clients should call `/preview` explicitly

Fixes #15398

## Test plan
- [x] `pytest test/unit_test/api/utils/test_file_response_headers.py`
(6/6)

---------

Co-authored-by: MkDev11 <mkdev11@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Ling Qin <qinling0210@163.com>
2026-07-03 19:56:01 +08:00

148 lines
4.6 KiB
Python

#
# Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
#
import re
from urllib.parse import urlencode
CONTENT_TYPE_MAP = {
"docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"doc": "application/msword",
"pdf": "application/pdf",
"csv": "text/csv",
"xls": "application/vnd.ms-excel",
"xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
"txt": "text/plain",
"py": "text/plain",
"js": "text/plain",
"java": "text/plain",
"c": "text/plain",
"cpp": "text/plain",
"h": "text/plain",
"php": "text/plain",
"go": "text/plain",
"ts": "text/plain",
"sh": "text/plain",
"cs": "text/plain",
"kt": "text/plain",
"sql": "text/plain",
"md": "text/markdown",
"markdown": "text/markdown",
"mdx": "text/markdown",
"htm": "text/html",
"html": "text/html",
"json": "application/json",
"png": "image/png",
"jpg": "image/jpeg",
"jpeg": "image/jpeg",
"gif": "image/gif",
"bmp": "image/bmp",
"tiff": "image/tiff",
"tif": "image/tiff",
"webp": "image/webp",
"svg": "image/svg+xml",
"ico": "image/x-icon",
"avif": "image/avif",
"heic": "image/heic",
"ppt": "application/vnd.ms-powerpoint",
"pptx": "application/vnd.openxmlformats-officedocument.presentationml.presentation",
}
FORCE_ATTACHMENT_EXTENSIONS = {
"htm",
"html",
"shtml",
"xht",
"xhtml",
"xml",
"mhtml",
"svg",
}
FORCE_ATTACHMENT_CONTENT_TYPES = {
"text/html",
"image/svg+xml",
"application/xhtml+xml",
"text/xml",
"application/xml",
"multipart/related",
}
def should_force_attachment(ext: str | None, content_type: str | None = None) -> bool:
normalized_ext = (ext or "").lower().strip(".")
if normalized_ext in FORCE_ATTACHMENT_EXTENSIONS:
return True
normalized_type = (content_type or "").lower().split(";")[0].strip()
return normalized_type in FORCE_ATTACHMENT_CONTENT_TYPES
def sanitize_content_disposition_filename(filename: str | None) -> str | None:
if not filename:
return None
base = re.sub(r"[^\w.\-]", "_", str(filename).split("/")[-1].split("\\")[-1])
return base or None
def resolve_attachment_content_type(ext: str | None = None, mime_type: str | None = None) -> tuple[str | None, str | None]:
if mime_type:
normalized_type = mime_type.lower().split(";")[0].strip()
for known_ext, known_type in CONTENT_TYPE_MAP.items():
if known_type == normalized_type:
return normalized_type, known_ext
return normalized_type, (ext or "").lower().strip(".") or None
if ext:
normalized_ext = ext.lower().strip(".")
return CONTENT_TYPE_MAP.get(normalized_ext, f"application/{normalized_ext}"), normalized_ext
return None, None
def apply_preview_file_response_headers(
response,
content_type: str | None,
ext: str | None = None,
filename: str | None = None,
):
if content_type:
response.headers.set("Content-Type", content_type)
if should_force_attachment(ext, content_type):
response.headers.set("X-Content-Type-Options", "nosniff")
response.headers.set("Content-Disposition", "attachment")
return response
safe_filename = sanitize_content_disposition_filename(filename)
if safe_filename:
response.headers.set("Content-Disposition", f'inline; filename="{safe_filename}"')
else:
response.headers.set("Content-Disposition", "inline")
return response
def apply_download_file_response_headers(
response,
content_type: str | None,
ext: str | None = None,
filename: str | None = None,
):
if content_type:
response.headers.set("Content-Type", content_type)
if should_force_attachment(ext, content_type):
response.headers.set("X-Content-Type-Options", "nosniff")
response.headers.set("Content-Disposition", "attachment")
return response
safe_filename = sanitize_content_disposition_filename(filename)
if safe_filename:
response.headers.set("Content-Disposition", f'attachment; filename="{safe_filename}"')
else:
response.headers.set("Content-Disposition", "attachment")
return response
def agent_attachment_preview_path(attachment_id: str, *, ext: str | None = None, mime_type: str | None = None) -> str:
query: dict[str, str] = {}
if ext:
query["ext"] = ext
if mime_type:
query["mime_type"] = mime_type
suffix = f"?{urlencode(query)}" if query else ""
return f"/api/v1/agents/attachments/{attachment_id}/preview{suffix}"