👟Euno Query Language (EQL)

Euno Query Language (EQL) is a specialized query language designed for deep interaction with Euno’s data model. EQL allows users to define precise conditions to filter, retrieve, and analyze resources within Euno. Whether you're exploring relationships, tracking dependencies, or analyzing data, EQL provides the flexibility needed to extract actionable insights.

With EQL, users can:

• Filter resources using specific property values.

• Analyze relationships such as dependencies and hierarchies.

• Query metrics, usage patterns, and other actionable data insights.

EQL’s strength lies in its ability to handle complex queries by leveraging the properties and relationships of resources within Euno’s common model. The properties of resources are derived from multiple discovery sources, including:

• dbt discovered resources

• Tableau discovered resources

• Looker discovered resources

• Snowflake discovered resources

• Fivetran discovered resources

EQL is an essential tool for anyone seeking to unlock deep insights and harness the full potential of their data within Euno.

EQL Expressions & Operators

EQL expressions can be categorized based on the type of operation they perform. Each category allows you to query and manipulate data in different ways, using the properties defined in Euno's common model.

Logical Operations

Logical operations combine multiple conditions, allowing you to refine your query results based on the combined truth value of these conditions.

• AND Operation

  • Expression: (x AND y)

  • Meaning: Both conditions x and y must be satisfied.

  • Example: type = 'dbt_model' AND database_schema = 'salesforce'

• OR Operation

  • Expression: (x OR y)

  • Meaning: Either condition x or y must be satisfied.

  • Example: type = 'dbt_model' OR type = 'dbt_source'

• NOT Operation

  • Expression: NOT (x)

  • Meaning: The condition x must not be satisfied.

  • Example: NOT (type = 'looker_look')

• Boolean True

  • Expression: A is true

  • Meaning: The property A (of type boolean) is set to true.

  • Example: is_active is true

• Boolean False

  • Expression: A is false

  • Meaning: The property A (of type boolean) is set to false.

  • Example: is_active is false

• Boolean Not True

  • Expression: A is not true

  • Meaning: The property A (of type boolean) is not true.

  • Example: is_active is not true

• Boolean Not False

  • Expression: A is not false

  • Meaning: The property A (of type boolean) is not false.

  • Example: is_active is not false

• Boolean Not False

  • Expression: A is not false

  • Meaning: The property A (of type boolean) is not false.

  • Example: is_active is not false

• Null Check

  • Expression: A is null

  • Meaning: The property A is not set.

  • Example: last_run_time is null

Comparison Operations

Comparison operations are used to compare a property’s value with a specified value. All these operators can be used with both datetime and numeric EQL statements, with the exception of the In List Match, which cannot be used in datetime statements.

• Equality

  • Expression: A = 'foobar'

  • Meaning: The property A (of type text) equals 'foobar'.

  • Example: name = 'sales_dashboard'

• Inequality

  • Expression: A != 'foobar'

  • Meaning: The property A (of type text) is set and is not equal to 'foobar'.

  • Example: status != 'archived'

• Greater Than

  • Expression: A > 6

  • Meaning: The property A (of type numeric) is set and greater than 6.

  • Example: record_count > 1000

• Greater Than or Equal To

  • Expression: A >= 6

  • Meaning: The property A (of type numeric or datetime) is set and greater than or equal to 6.

  • Example: record_count >= 1000

• Less Than

  • Expression: A < 6

  • Meaning: The property A (of type numeric) is set and less than 6.

  • Example: record_count < 1000

• Less Than or Equal To

  • Expression: A <= 6

  • Meaning: The property A (of type numeric or datetime) is set and less than or equal to 6.

  • Example: record_count <= 1000

• In List Match

  • Expression: A IN [value1, value2, ...]

  • Meaning: The property A (of type numeric or string) is set and its value is one of the specified list of values.

  • Example: record_count IN [1000, 2000, 3000] or name IN ['sales_dashboard', 'marketing_dashboard']

Date and Time Operations

These operations compare timestamp properties against specific dates or relative time periods.

• Specific Date Comparison

  • Expression: A > '2023-01-01'

  • Meaning: The property A (of type timestamp) is set and the value is after 2023-01-01 00:00:00.

  • Example: created_at > '2023-01-01'

• Relative Time Comparison

  • Expression: A > -3d/h

  • Meaning: The property A (of type timestamp) is set and the value is within the last 3 days.

  • Example: updated_at > -7d/h (for the last 7 days/hours)

String Operations

String operations allow for querying text-based properties using conditions related to their contents.

• Substring Match

  • Expression: A ~ 'asdf'

  • Meaning: The property A (of type text) contains 'asdf' as a substring.

  • Example: description ~ 'critical'

• Regular Expression Match

  • Expression: A =~ 'asdf.*asdf'

  • Meaning: The property A (of type text) matches the regular expression 'asdf.*asdf'.

  • Example: log_entry =~ 'error.*timeout'

Map Operations

Map operations allow for checking specific keys or values within a map property.

• Key Existence

  • Expression: 'foo' in A

  • Meaning: The property A (of type string map) contains the key 'foo'.

  • Example: 'priority' in metadata

• Key-Value Match

  • Expression: A.foo = 'bar'

  • Meaning: The property A is set, it has the key 'foo', and the value of 'foo' is 'bar'.

  • Example: metadata.priority = 'high'

Boolean Literals

Boolean literals are static values representing true or false, often used in combination with other expressions.

• True Literal

  • Expression: True

  • Meaning: Always true.

• False Literal

  • Expression: False

  • Meaning: Always false.

Relationship Operations

One of the most powerful features of EQL is its ability to explore and analyze relationships between different resources. Relationships define how resources are connected within Euno’s data model, allowing users to traverse these connections and query related resources.

• Upstream relationships

  • Expression: has upstream(x)

  • Meaning: Represents resources that provide input or dependencies for another resource. For example, a dbt source might be upstream of a dbt model that uses it.

  • Example: type="dbt_model" AND has upstream(type="dbt_source" AND name="foobar")

• Downstream relationships

  • Expression: has downstream(x)

  • Meaning: Represents resources that are outputs or dependents of another resource. For example, a dbt model might be downstream of a specific tableau dashboard.

  • Example: type="dbt_model" and downstream(type="tableau_dashboard" and name="foobar")

• Parent relationship

  • Expression: has parent(x)

  • Meaning: Represents a hierarchical relationship where one resource contains or is the parent of another. For example, a folder containing multiple dashboards would have a parent-child relationship.

  • Example: type="looker_dashboard" AND has parent(type="looker_folder and name="foobar")

• Child relationship

  • Expression: has child(x)

  • Meaning: Represents a hierarchical relationship where one resource is contained within or is a child of another. This is the inverse of the parent relationship.

  • Example: type="looker_look" AND has child(type="looker_measure" AND name="foobar")

The has RELATION Operator

The has RELATION operator in EQL allows users to filter resources based on their relationships with other resources. This operator is key to exploring complex dependencies and associations within Euno’s data model.

Expression: has RELATION((x), n)

• RELATION: The type of relationship (e.g., upstream, downstream, parent, child).

• x: An EQL expression that defines the condition for the related resource.

• n (optional): The maximum length of the path. If not specified, paths of any length are considered.

Example:

type = 'dbt_model' AND has upstream((type='dbt_source' AND database='test_lab_jaffle_shop'), 1)

Leveraging Relationship Operators with Boolean Logic

One of the strengths of EQL is the ability to combine relationship operators with Boolean logic to construct powerful queries. For example:

Query: type = 'looker_view' AND NOT has upstream(TRUE)

Meaning: Return all Looker views that have no upstream connections. This allows users to pinpoint resources that are isolated or not dependent on any other resource in the model.