Installation

The Weaviate CLI is available on Pypi.org. The package can be easily installed using pip. The client is developed and tested for Python 3.7 and higher.

The Weaviate CLI can be installed with:

$ pip install weaviate-cli

To check if the cli is installed correctly, run:

$ weaviate version

which should return 2.1.2.

Functions

Configuration

You need to configure the CLI tool before you can interact with you Weaviate instance. This can be done manually or by adding flags to commands.

• Manually (interactive):

  $ weaviate config set
  

  or

  $ weaviate init
  

  After which you will be asked to enter the Weaviate URL and authentication mode.

• Flags: if you didn’t configure the CLI manually, you can add a configuration flag pointing to a configuration json file (--config-file myconfig.json) with every command you execute.

  $ weaviate --config-file myconfig.json
  

  in which myconfig.json should look like:

  {
    "url": "http://localhost:8080",
    "auth": null
  }   
  

  or

  {
    "url": "http://localhost:8080",
    "auth": {
        "secret": <your_client_secret>
    }
  }   
  

  or

  {
    "url": "http://localhost:8080",
    "auth": {
        "user": <user name>,
        "pass": <user password>
    }
  }   
  

You can view the configuration with the command:

$ weaviate config view

Ping

You can ping the Weaviate URL you’re connected to with:

$ weaviate ping

Which returns Weaviate is reachable! if the connection with the Weaviate server is set up correctly.

Schema

There are three operations available with regard to the schema: import, export and truncate.

Import

Adding a schema can be done via:

$ weaviate schema import my_schema.json

Where my_schema.json contains the schema as described here.

To overwrite your schema you can use the --force flag, this will clear the index and replace your schema:

$ weaviate schema import --force my_schema.json # using --force will delete your data

Export

You can export a schema to a json file that is present in the Weaviate instance by:

$ weaviate schema export my_schema.json

Where my_schema.json can be replaces by a json file and local location. Naturally this function only outputs the schema to the given location when a schema is present in Weaviate.

Truncate

With delete you can remove the entire schema and all the data that is associated with it. You will be asked for confirmation unless the --force flag is added.

$ weaviate schema delete

Data

Import

The import function enables data import from a json file. When the flag --fail-on-error is added, this command execution will fail if an error was thrown by Weaviate when loading the data object in.

$ weaviate data import my_data_objects.json

The json file and location is passed in the command. The file needs to be formatted according to the Weaviate data schema, for example:

{
    "classes": [
        {
            "class": "Publication",
            "id": "f81bfe5e-16ba-4615-a516-46c2ae2e5a80",
            "properties": {
                "name": "New York Times"
            }
        },
        {
            "class": "Author",
            "id": "36ddd591-2dee-4e7e-a3cc-eb86d30a4303",
            "properties": {
                "name": "Jodi Kantor",
                "writesFor": [{
                    "beacon": "weaviate://localhost/f81bfe5e-16ba-4615-a516-46c2ae2e5a80",
                    "href": "/v1/f81bfe5e-16ba-4615-a516-46c2ae2e5a80"
                }]
            }
        }
    ]
}

Empty

With delete you can remove all data objects in Weaviate. You will be asked for confirmation unless the --force flag is added.

$ weaviate data delete

Change logs

2.1.2

From 2.1.1 to 2.1.2: fixes compatibility for Weaviate client v3.0.0.

2.0.0

From 1.2.0 to 2.0.0.

New release of the weaviate CLI that is compatible with weaviate v1.x.x and uses weaviate-python-client v2.0.x.

Changes with this release:

1. The name of the CLI command is changed from weaviate-cli to weaviate.
2. The name of schema truncate is changed to schema delete, because delete is more appropriate for its functionality.
3. The name of data empty is changed to data delete, because empty might be confused with checking if it is empty.

More Resources

If you can’t find the answer to your question here, please look at the:

1. Frequently Asked Questions. Or,
2. Knowledge base of old issues. Or,
3. For questions: Stackoverflow. Or,
4. For issues: Github. Or,
5. Ask your question in the Slack channel: Slack.