allowed_values: [ "value_1", "value_2" , … ]
AcceptsThe name of a user attribute with the
An access grant is a LookML structure that controls access to other LookML structures, specifically Explores, joins, views, and fields. The
access_grant parameter defines an access grant.
access_grant takes the name of a user attribute with the
user_attribute subparameter and a list of acceptable values for the user attribute with the
allowed_values subparameter. Only those users who are assigned one of the allowed values in the specified user attribute can access structures to which the access grant is required.
Once defined, you can use the
required_access_grants parameter at the Explore, join, view, or field level to require the access grant to access those structures.
For example, the LookML below creates an access grant called
can_view_financial_data, which is based on the
department user attribute. Only those users who are assigned the values
"executive" in the
department user attribute are given access to the
can_view_financial_data access grant:
You then associate the
can_view_financial_data access grant with a LookML structure using the
In the example above, only users who have the proper user attribute value for the
can_view_financial_data access grant will see the
You can define multiple access grants in a model, and you can assign multiple access grants to a LookML structure with the
required_access_grants parameter. In that case, a user must have access to all of the specified access grants to have access to the LookML structure.
For example, the LookML below defines two different access grants:
Then in the view file below, the
required_access_grants parameter specifies both access grants:
In this case, only users that have either the value
"finance" or the value
"executive" assigned to their
department user attribute and have the value
"yes" assigned to their
view_payroll user attribute can access the view.
Define an access grant that requires users to have either the value
"product_management" or the value
"engineering" in the
department user attribute to have access to the
engineering access grant:
You can also define access grants with user attributes that take numeric or date/time data. To do this, you must enclose the allowed values in double quotes, just as you would with a string. For example, the following access grant references the
id user attribute, which has a data type of number. Only users with the
id value of 1, 2, 3, 4, or 5 will be granted access:
The following example references the user attribute
start_date, which has the data type Date/Time. Only users who have the value
2020-01-01 in the user attribute will be granted access:
Things to Consider
Values Listed in
allowed_values Must Match User Attribute Values Exactly
access_grant will work with user attributes that have the String Filter (advanced), Number Filter (advanced), or Date/Time Filter (advanced) data type. But, in order to grant access, the values listed in the
allowed_values parameter must match the value in the user attribute exactly.
For example, if you created a user attribute called
numeric_range with the data type Number Filter (advanced), you could use a Looker filter expression to enter a range of numbers, such as
[1, 20]. In that example, the Looker filter expression in the user attribute will return a range of numbers between 1 and 20, inclusive. Since
access_grant requires an exact match, however, using the parameter
allowed_values: ["10"] would not grant access. To grant access, you would have to use
allowed_values: ["[1, 20]"].
access_grant requires an exact match with user attributes of data type String Filter (advanced). Unlike typical Looker filter expressions, using the parameter
allowed_values: [ "Ca%" ] does not not match a user attribute with the values
California. Only a user attribute value of exactly
Ca% would be matched and granted access.
Users Who Are Not Granted Access Experience Different Behavior Depending on LookML Structure
A user who does not have access to an access grant will experience different behavior depending on which LookML structure they are trying to access. See the
required_access_grants documentation pages at the Explore, join, view, or field level for information about how access to those structures is restricted.
Access Grants at Multiple Levels Are Added Together
If you nest access grants, the access grants are additive. For example, you can create
required_access_grants for a view and create
required_access_grants for a field inside the view. In order to see the field, a user must have access grants to both the field and the view. Likewise for joins: If you create
required_access_grants for the views in a join and also create
required_access_grants for the join of these two views, a user must have access grants to both views and the join in order to see the joined view.
Accessing Structures That Reference Restricted Structures
Users can have access to Looks or dashboards that contain LookML objects they don’t have access to. In these situations the Look or dashboard will display as if those LookML objects have been removed from the model.
Suppose we have an Explore A, which contains join A, view A, and field A. Next, we place an access restriction on Explore A. As expected, join A, view A, and field A will inherit that restriction, but only when users are interacting with Explore A. If join A, view A, or field A is used in a different Explore B, they will not necessarily have any access restrictions. Therefore, if you plan to re-use LookML elements, we suggest you apply access restrictions at the lowest level possible.