Create table clones

This document describes how to copy a table to a table clone by using a CREATE TABLE CLONE SQL statement, a bq cp command, or a jobs.insert API call. This document is intended for users who are familiar with table clones.

Permissions and roles

This section describes the Identity and Access Management (IAM) permissions that you need to create a table clone, and the predefined IAM roles that grant those permissions.

Permissions

To create a table clone, you need the following permissions:

Permission Resource
All of the following:

bigquery.tables.get
bigquery.tables.getData
 The table that you want to make a clone of.
bigquery.tables.create
bigquery.tables.updateData 		The dataset that contains the table clone.

Roles

The predefined BigQuery roles that provide the required permissions are as follows:

Role Resource
Any of the following:

bigquery.dataViewer
bigquery.dataEditor
bigquery.dataOwner
bigquery.admin 		The table that you want to make a clone of.
Any of the following:

bigquery.dataEditor
bigquery.dataOwner
bigquery.admin 		The dataset that contains the new table clone.

Create a table clone

Use GoogleSQL, the bq command-line tool, or the BigQuery API to create a table clone.

SQL

To clone a table, use the CREATE TABLE CLONE statement.

  1. In the Google Cloud console, go to the BigQuery page.

    Go to BigQuery

  2. In the query editor, enter the following statement:

    CREATE TABLE
myproject.myDataset_backup.myTableClone
CLONE myproject.myDataset.myTable;

  3. Click Run.

For more information about how to run queries, see Run an interactive query.

Replace the following:

  • PROJECT is the project ID of the target project. This project must be in the same organization as the project containing the table you are cloning.
  • DATASET is the name of the target dataset. This dataset must be in the same region as the dataset containing the table you are cloning.
  • CLONE_NAME is name of the table clone that you are creating.

bq

Use a bq cp command with the --clone flag:

Go to Cloud Shell 

bq cp --clone --no_clobber project1:myDataset.myTable PROJECT:DATASET.CLONE_NAME

Replace the following:

  • PROJECT is the project ID of the target project. This project must be in the same organization as the project containing the table you are cloning.
  • DATASET is the name of the target dataset. This dataset must be in the same region as the dataset containing the table you are cloning. If the dataset is not in the same region as the dataset containing the table you are cloning then a full table is copied.
  • CLONE_NAME is name of the table clone that you are creating.

The --no_clobber flag is required.

If you are creating a clone in the same project as the base table, you can skip specifying a project, as shown following:

bq cp --clone --no_clobber myDataset.myTable DATASET.CLONE_NAME

API

Call the jobs.insert method with the operationType field set to CLONE:

Parameter Value
projectId The project ID of the project that runs the job.
Request body 
{
  "configuration": {
    "copy": {
      "sourceTables": [
        {
          "projectId": "myProject",
          "datasetId": "myDataset",
          "tableId": "myTable"
        }
      ],
      "destinationTable": {
        "projectId": "PROJECT",
        "datasetId": "DATASET",
        "tableId": "CLONE_NAME"
      },
      "operationType": "CLONE",
      "writeDisposition": "WRITE_EMPTY",
    }
  }
}

Replace the following:

  • PROJECT is the project ID of the target project. This project must be in the same organization as the project containing the table you are cloning.
  • DATASET is the name of the target dataset. This dataset must be in the same region as the dataset containing the table you are cloning. If the dataset is not in the same region as the dataset containing the table you are cloning a full table is copied.
  • CLONE_NAME is name of the table clone that you are creating.

Access control

When you create a table clone, access to the table clone is set as follows:

  • Row-level access policies are copied from the base table to the table clone.
  • Column-level access policies are copied from the base table to the table clone.

  • Table-level access is determined as follows:

    • If the table clone overwrites an existing table, then the table-level access for the existing table is maintained. Tags aren't copied from the base table.
    • If the table clone is a new resource, then the table-level access for the table clone is determined by the access policies of the dataset in which the table clone is created. Additionally, tags are copied from the base table to the table clone.

What's next

  • After you create a table clone, you can use it like you use standard tables. For more information, see Manage tables.