add_file_resource()

Private Preview

Addedv3.0.x

Registers a file that has already been uploaded to the object store configured for the Milvus cluster as a named file resource. Once registered, the resource can be referenced from analyzer parameters that accept an external dictionary — such as extra_dict_file on the jieba tokenizer, stop_words_file on the stop filter, word_list_file on the decompounder filter, and synonyms_file on the synonym filter — using {"type": "remote", "resource_name": "<name>", "file_name": "<file_name>"}. The target file must exist in the object store at the time of this call; the server validates path synchronously and fails the request if it does not resolve.

Request syntax

add_file_resource(
    name: str,
    path: str,
    timeout: float | None = None,
    **kwargs
)

PARAMETERS:

• name (str) - The unique name to register the resource under. This value is what you later pass as resource_name in analyzer configurations that reference this resource.

• path (str) - The object key of the file in the object store configured for the Milvus cluster, including the rootPath prefix. For example, if the cluster's rootPath is file and you uploaded the file to s3://<bucket>/file/dict.txt, set path to "file/dict.txt". A path that does not resolve to an existing object causes the call to fail with MilvusException (code=65535, message="file resource path not exist").

• timeout (float | None) - The timeout duration (in seconds) for this operation. A value of None indicates that no timeout is applied.

RETURNS:

None

Examples

from pymilvus import MilvusClient

client = MilvusClient(
    uri="YOUR_CLUSTER_ENDPOINT",
    token="YOUR_CLUSTER_TOKEN",
)

# Upload the file to the cluster's object store out-of-band first
# (e.g., via mc, boto3, or the AWS CLI), then register it here.
client.add_file_resource(
    name="zh_terms",
    path="file/zh_terms.txt",
)

# The registered resource can now be referenced from analyzer configs.
analyzer_params = {
    "tokenizer": {
        "type": "jieba",
        "dict": ["_default_"],
        "extra_dict_file": {
            "type": "remote",
            "resource_name": "zh_terms",
            "file_name": "zh_terms.txt",
        },
    },
}