提炼：

无覆盖创建文档有两种方案：
1.不指定 id 的 post 提交： POST /website/blog/
2.指定 id 的 put 提交：①PUT /website/blog/123?op_type=create ②PUT /website/blog/123/_create
指定 id 的 put 提交存在两种情况，一是成功创建返回 201， 二是冲突，创建失败返回 409

原文：

How can we be sure, when we index a document, that we are creating an entirely new document and not overwriting an existing one?
Remember that the combination of _index, _type, and _id uniquely identifies a document. So the easiest way to ensure that our document is new is by letting Elasticsearch autogenerate a new unique _id, using the POST version of the index request:

POST /website/blog/ {…}

However, if we already have an _id that we want to use, then we have to tell Elasticsearch that it should accept our index request only if a document with the same _index, _type, and _id doesn’t exist already. There are two ways of doing this, both of which amount to the same thing. Use whichever method is more vonvenient for you.
The first method uses the op_type query-string parameter:

PUT /website/blog/123?op_type=create {…}

And the second uses the /_create endpoint in the URL:

PUT /website/blog/123/_create {…}

If the request succeeds in creating a new document, Elasticsearch will return the usual metadata and an HTTP response code of 201 Created.
On the other hand, if a document with the same _index, _type, and _id already exists, Elasticsearch will respond with a 409 Conflict response code, and an error message like the following:

{ “error”: { “root_cause”: [ { “type”: “document_already_exist_exception”, “reason”: “[blog][123]: document already exists”, “shard”: “0”, “index”: “website” } ], “type”: “document_already_exists_exception”, “reason”: “[blog][123]: document already exists”, “shard”: “0”, “index”: “website” } “status”:409 }