save file document in safe mode for GridFS, so we can raise an exception on duplicates

gridfs/__init__.py

@@ -63,6 +63,10 @@ class GridFS(object):
         which data can be written. Any keyword arguments will be
         passed through to :meth:`~gridfs.grid_file.GridIn`.
 
+        If the ``"_id"`` of the file is manually specified, it must
+        not already exist in GridFS. Otherwise
+        :class:`~gridfs.errors.FileExists` is raised.
+
         :Parameters:
           - `**kwargs` (optional): keyword arguments for file creation
 
@@ -87,6 +91,10 @@ class GridFS(object):
         :meth:`~gridfs.grid_file.GridIn` for possible
         arguments. Returns the ``"_id"`` of the created file.
 
+        If the ``"_id"`` of the file is manually specified, it must
+        not already exist in GridFS. Otherwise
+        :class:`~gridfs.errors.FileExists` is raised.
+
         :Parameters:
           - `data`: data to be written as a file.
           - `**kwargs` (optional): keyword arguments for file creation

gridfs/errors.py

@@ -35,6 +35,12 @@ class NoFile(GridFSError):
     .. versionadded:: 1.6
     """
 
+class FileExists(GridFSError):
+    """Raised when trying to create a file that already exists.
+
+    .. versionadded:: 1.6+
+    """
+
 class UnsupportedAPI(GridFSError):
     """Raised when trying to use the old GridFS API.
 

gridfs/grid_file.py

@@ -23,10 +23,12 @@ except ImportError:
     from StringIO import StringIO
 
 from gridfs.errors import (CorruptGridFile,
+                           FileExists,
                            NoFile,
                            UnsupportedAPI)
 from pymongo.binary import Binary
 from pymongo.collection import Collection
+from pymongo.errors import DuplicateKeyError
 from pymongo.objectid import ObjectId
 
 try:
@@ -93,7 +95,8 @@ class GridIn(object):
         arguments include:
 
           - ``"_id"``: unique ID for this file (default:
-            :class:`~pymongo.objectid.ObjectId`)
+            :class:`~pymongo.objectid.ObjectId`) - this ``"_id"`` must
+            not have already been used for another file
 
           - ``"filename"``: human name for the file
 
@@ -192,7 +195,12 @@ class GridIn(object):
         self._file["md5"] = md5
         self._file["length"] = self._position
         self._file["uploadDate"] = datetime.datetime.utcnow()
-        return self._coll.files.insert(self._file)
+
+        try:
+            return self._coll.files.insert(self._file, safe=True)
+        except DuplicateKeyError:
+            raise FileExists("file with _id %r already exists" % self._id)
+
 
     def close(self):
         """Flush the file and close it.

test/test_gridfs.py

@@ -27,7 +27,8 @@ import sys
 sys.path[0:0] = [""]
 
 import gridfs
-from gridfs.errors import NoFile
+from gridfs.errors import (FileExists,
+                           NoFile)
 from test_connection import get_connection
 
 
@@ -182,6 +183,10 @@ class TestGridfs(unittest.TestCase):
         self.assertEqual(11, self.db.fs.chunks.count())
         self.assertEqual("hello world", self.fs.get(oid).read())
 
+    def test_put_duplicate(self):
+        oid = self.fs.put("hello")
+        self.assertRaises(FileExists, self.fs.put, "world", _id=oid)
+
 
 if __name__ == "__main__":
     unittest.main()