Controlling access to tables and views

This document describes how to use BigQuery Table ACL to control access to tables and views. For an overview of BigQuery Table ACL, see Introduction to table access controls.

After you create a table or view, you can set its policy in the following ways:

• using the console
• using the bq set-iam-policy command
• calling the tables.setIamPolicy method
• using the GRANT or REVOKE data control language statements

You can use BigQuery Table ACL to set access on both views and dataset-level authorized views. A view can also reference other source tables and views that you shared using BigQuery Table ACL.

Before you begin

You can use BigQuery Table ACL on BigQuery resources such as tables and views. To learn how to create tables, see Create a table, and for views, see Create a view.

Required permissions

To create a table access policy, you need the bigquery.tables.setIamPolicy Identity and Access Management (IAM) permission.

Each of the following predefined IAM roles includes the permissions that you need in order to create access policies on a table or view:

• bigquery.dataOwner
• bigquery.admin

For more information about BigQuery IAM, see Access control with IAM.

Required permission to query tables and views

Users need the bigquery.tables.getData permission on all tables and views that their query references. In addition, when querying a view users need this permission on all underlying tables and views. However, if you are using authorized views or authorized datasets, you don't need to give users access to the underlying source data.

Each of the following predefined IAM roles includes the permission that users need on all tables and views that the query references:

• roles/bigquery.admin
• roles/bigquery.dataOwner
• roles/bigquery.dataEditor
• roles/bigquery.dataViewer

Create an access policy

To create an access policy on a table or view:

Console

1. Go to the BigQuery page.

   Go to BigQuery

2. In the Explorer panel, expand your project and select a dataset.

3. Expand the dataset and select a table or view.

4. Click Share.

5. On the Share page, to add a user (or principal), click Add principal.

6. On the Add principals page, do the following:

   • For New principals, enter a user. You can add individual users, groups, service accounts, and workspace domains.
   • From the Select a role drop-down list, select the role that you want to grant to the user.
   • Click Add condition to add conditional access for the user.

7. Click Save to save the changes for the new user.

8. Click Close to close the Share page.

SQL

To grant roles to users, use the GRANT DCL statement:

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

   Go to BigQuery

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

   GRANT `ROLE_LIST`
   ON TABLE DATASET.TABLE_OR_VIEW
   TO 'USER_LIST';
   

   Replace the following:

   • ROLE_LIST: a role or a comma-separated list of roles that contains the permissions that you want to give, such as roles/bigquery.dataViewer
   • DATASET: the name of the dataset that the resource is in
   • TABLE_OR_VIEW: the table or view that you are revoking access from
   • USER_LIST: a comma-separated list of users that the role is granted to, such as user:[e-mail]

3. Click Run.

For more information about how to run queries, see Running interactive queries.

bq

1. Retrieve the existing policy to a local file.

   bq get-iam-policy \
    project-id:dataset.table_or_view \
    > policy.json
   

   where:

   • project-id is your project ID.
   • dataset is the name of the dataset that contains the resource (table or view) that you are updating.
   • table_or_view is the name of the resource that you are updating.

   More examples of identifying a table or view, and redirecting the policy output to a file:

   • bq get-iam-policy dataset1.table1 > policy.json
   • bq get-iam-policy --project_id=project1 -t dataset1.table1 > policy.json
   • bq get-iam-policy project1:dataset1.table1 > policy.json

2. If you haven't yet added any members to the policy, the policy.json file will contain an etag value and no other fields. See For more information about how to format the policy.json file, see Understanding policies.

3. To add the first member, add a bindings field to the policy. For example, to grant the BigQuery Data Viewer (roles/bigquery.dataViewer) role to [e-mail]:

   "bindings": [
    {
      "members": [
        "user:[e-mail]"
      ],
      "role": "roles/bigquery.dataViewer"
    }
   ]
   

   If you need to add more members to an existing binding, just add the member. This example shows granting [e-mail] the BigQuery Data Viewer (roles/bigquery.dataViewer) role, for a binding that already exists.

   "members": [
          "user:[e-mail]",
          "user:[e-mail]"
        ],
        "role": "roles/bigquery.dataViewer"
      }
   

4. Update the policy.

   bq set-iam-policy \
    project-id:dataset.table_or_view \
    policy.json
   

   Caution: After you create a policy, subsequent retrieval of the policy will show that its version field is 1. Do not change this version number in future updates to the policy. This version number refers to the IAM policy schema version, not the version of your policy for the table or view. The etag value is the policy version number.

   For information about IAM policy schema versions, see Policy versions.

API

1. Call tables.getIamPolicy to retrieve the current policy.

2. Edit the policy to add members and/or bindings. See the bq example for the format of the policy.

3. Call tables.setIamPolicy to write the new policy.

Caution: After you create a policy, subsequent retrieval of the policy will show that its version field is 1. Do not change this version number in future updates to the policy. This version number refers to the IAM policy schema version, not the version of your policy for the table or view. The etag value is the policy version number.

For information about IAM policy schema versions, see Policy versions.

Java

Before trying this sample, follow the Java setup instructions in the BigQuery quickstart using client libraries. For more information, see the BigQuery Java API reference documentation.

View on GitHub Feedback

import com.google.cloud.Identity;
import com.google.cloud.Policy;
import com.google.cloud.Role;
import com.google.cloud.bigquery.BigQuery;
import com.google.cloud.bigquery.BigQueryException;
import com.google.cloud.bigquery.BigQueryOptions;
import com.google.cloud.bigquery.TableId;

// Sample to create iam policy for table
public class CreateIamPolicy {

  public static void main(String[] args) {
    // TODO(developer): Replace these variables before running the sample.
    String datasetName = "MY_DATASET_NAME";
    String tableName = "MY_TABLE_NAME";
    createIamPolicy(datasetName, tableName);
  }

  public static void createIamPolicy(String datasetName, String tableName) {
    try {
      // Initialize client that will be used to send requests. This client only needs to be created
      // once, and can be reused for multiple requests.
      BigQuery bigquery = BigQueryOptions.getDefaultInstance().getService();

      TableId tableId = TableId.of(datasetName, tableName);

      Policy policy = bigquery.getIamPolicy(tableId);
      policy
          .toBuilder()
          .addIdentity(Role.of("roles/bigquery.dataViewer"), Identity.allUsers())
          .build();
      bigquery.setIamPolicy(tableId, policy);
      System.out.println("Iam policy created successfully");
    } catch (BigQueryException e) {
      System.out.println("Iam policy was not created. \n" + e.toString());
    }
  }
}

Update an access policy

To update an access policy on a table or view:

Console

1. Go to the BigQuery page.

   Go to BigQuery

2. In the Explorer panel, expand your project and select a dataset.

3. Expand the dataset and select a table or view.

4. Click Share.

5. On the Share page, do the following:

   • To add new users, follow the steps as shown in Creating an access policy.

   • To remove access for a user, either search for the user by using the Search field or expand the role to which the user belongs. For the user you want to remove, click Delete .

   • To modify access for a user, click Edit .

6. Click Save when you have made the changes.

7. Click Close to close the Share page.

SQL

To remove roles from users on a table or view in your dataset, use the REVOKE DCL statement:

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

   Go to BigQuery

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

   REVOKE `ROLE_LIST`
   ON TABLE DATASET.TABLE_OR_VIEW
   FROM 'USER_LIST';
   

   Replace the following:

   • ROLE_LIST: a role or a comma-separated list of roles that contains the permissions that you want to give, such as roles/bigquery.dataViewer
   • DATASET: the name of the dataset that the resource is in
   • TABLE_OR_VIEW: the table or view that you are revoking access from
   • USER_LIST: a comma-separated list of users that the role is granted to, such as user:[e-mail]

3. Click Run.

For more information about how to run queries, see Running interactive queries.

bq

1. Retrieve the existing policy to a local file.

   bq get-iam-policy --format=prettyjson \
    project-id:dataset.table_or_view \
    > policy.json
   

   where:

   • project-id is your project ID.
   • dataset is the name of the dataset that contains the table that you are updating.
   • table_or_view is the name of the table or view you are updating.

   More examples of identifying a table or view, and redirecting the policy output to a file:

   • bq get-iam-policy dataset1.table1 > policy.json
   • bq get-iam-policy --project_id=project1 -t dataset1.table1 > policy.json
   • bq get-iam-policy project1:dataset1.table1 > policy.json
   Caution: When setting the policy, always retrieve the current policy as a first step, to obtain the current etag value. Your updated policy file must include the same etag value as the current policy you are replacing, or the update fails. This feature prevents concurrent updates from happening.
   
   The version number in the policy file remains 1. This version number refers to the IAM policy schema version, not the version of the policy. The etag value is the policy version number.

2. Modify policy.json as needed.

   For more information about how to format the policy.json file, see Understanding policies.

   For information about IAM policy schema versions, see Policy versions.

   Caution: Empty bindings with no members are not allowed and result in an error.

3. Update the policy.

   bq set-iam-policy \
    project-id:dataset.table_or_view \
    policy.json
   

API

1. Call tables.getIamPolicy to retrieve the current policy.

2. Edit the policy to add members and/or bindings.

   For the format required for the policy, see the Policy reference topic.

   Caution: When setting the policy, always retrieve the current policy as a first step, to obtain the current etag value. Your updated policy file must include the same etag value as the current policy you are replacing, or the update fails. This feature prevents concurrent updates from happening.
   
   The version number in the policy file remains 1. This version number refers to the IAM policy schema version, not the version of the policy. The etag value is the policy version number.

3. Call tables.setIamPolicy to write the updated policy.

   Caution: Empty bindings with no members are not allowed and result in an error.

Java

Before trying this sample, follow the Java setup instructions in the BigQuery quickstart using client libraries. For more information, see the BigQuery Java API reference documentation.

View on GitHub Feedback

import com.google.cloud.Identity;
import com.google.cloud.Policy;
import com.google.cloud.Role;
import com.google.cloud.bigquery.BigQuery;
import com.google.cloud.bigquery.BigQueryException;
import com.google.cloud.bigquery.BigQueryOptions;
import com.google.cloud.bigquery.TableId;
import java.util.HashMap;
import java.util.Map;
import java.util.Set;

// Sample to update iam policy in table
public class UpdateIamPolicy {

  public static void main(String[] args) {
    // TODO(developer): Replace these variables before running the sample.
    String datasetName = "MY_DATASET_NAME";
    String tableName = "MY_TABLE_NAME";
    updateIamPolicy(datasetName, tableName);
  }

  public static void updateIamPolicy(String datasetName, String tableName) {
    try {
      // Initialize client that will be used to send requests. This client only needs to be created
      // once, and can be reused for multiple requests.
      BigQuery bigquery = BigQueryOptions.getDefaultInstance().getService();

      TableId tableId = TableId.of(datasetName, tableName);

      Policy policy = bigquery.getIamPolicy(tableId);
      Map<Role, Set<Identity>> binding = new HashMap<>(policy.getBindings());
      binding.remove(Role.of("roles/bigquery.dataViewer"));

      policy.toBuilder().setBindings(binding).build();
      bigquery.setIamPolicy(tableId, policy);

      System.out.println("Iam policy updated successfully");
    } catch (BigQueryException e) {
      System.out.println("Iam policy was not updated. \n" + e.toString());
    }
  }
}

For more information about Identity and Access Management policies, see Understanding policies and the Policy reference topic.

What's next

• Read the FAQ.
• Learn about audit logging of BigQuery Table ACL admin activities at Audit logging.