use app_name to set job.name-format

Author: razvan

docs/modules/trino/pages/usage-guide/openlineage.adoc

@@ -40,13 +40,13 @@ spec:
           #       caCert:
           #         secretClass: marquez-ca
       # namespace: trino-lineage # <3>
+      # appName: "$QUERY_ID" # <4>
   ...
 ----
 <1> Adding the `openLineage` block enables OpenLineage event emission. Omit it entirely to leave OpenLineage off (the default).
 <2> The OpenLineage backend connection. Use `inline` (shown here) to define the connection directly, or `reference: <name>` to point at an `OpenLineageConnection` resource in the same namespace. Required.
 <3> Optional. The OpenLineage namespace events are reported under. Defaults to the cluster's Kubernetes namespace.
-
-NOTE: The shared connection type also accepts an `appName` field. It is ignored for Trino, which is a long-running service rather than a single job and derives an OpenLineage job name per query.
+<4> Optional. Sets the OpenLineage job name format (`openlineage-event-listener.job.name-format`). Accepts an arbitrary string with optional `$QUERY_ID`, `$USER`, `$SOURCE` and `$CLIENT_IP` substitution variables. When unset, Trino defaults to `$QUERY_ID`. See the link:https://trino.io/docs/current/admin/event-listeners-openlineage.html[Trino OpenLineage event listener documentation {external-link-icon}^].
 
 The `OpenLineageConnection` resource (`openlineage.stackable.tech`) is a reusable definition of a connection to an OpenLineage backend, provided by the Stackable operator library.
 Reference it to share one backend definition across multiple clusters:

extra/crds.yaml

@@ -896,8 +896,9 @@ spec:
                       Emit [OpenLineage](https://openlineage.io/) lineage events for the queries run on this
                       Trino cluster. The OpenLineage event listener runs on the coordinator only.
                       The backend connection is either inlined or references an `OpenLineageConnection` resource.
-                      The shared `appName` field is not used by Trino, which derives an OpenLineage job name per
-                      query. See the [OpenLineage usage guide](https://docs.stackable.tech/home/nightly/trino/usage-guide/openlineage).
+                      The shared `appName` field sets the OpenLineage job name format
+                      (`openlineage-event-listener.job.name-format`); when unset Trino defaults to `$QUERY_ID`.
+                      See the [OpenLineage usage guide](https://docs.stackable.tech/home/nightly/trino/usage-guide/openlineage).
                     nullable: true
                     properties:
                       appName:

rust/operator-binary/src/config/open_lineage.rs

@@ -61,6 +61,10 @@ pub const OPENLINEAGE_TRANSPORT_URL_KEY: &str = "openlineage-event-listener.tran
 pub const OPENLINEAGE_TRANSPORT_API_KEY_KEY: &str = "openlineage-event-listener.transport.api-key";
 /// The OpenLineage namespace lineage is reported under.
 pub const OPENLINEAGE_NAMESPACE_KEY: &str = "openlineage-event-listener.namespace";
+/// Format for the emitted OpenLineage job name. Set from `spec.clusterConfig.openLineage.appName`.
+/// Accepts an arbitrary string with optional `$QUERY_ID`, `$USER`, `$SOURCE` and `$CLIENT_IP`
+/// substitution variables (Trino defaults to `$QUERY_ID` when unset).
+pub const OPENLINEAGE_JOB_NAME_FORMAT_KEY: &str = "openlineage-event-listener.job.name-format";
 /// The URI identifying this Trino cluster in emitted lineage. Computed from the coordinator service
 /// at ConfigMap-build time (see [`crate::controller::build::properties::event_listener_properties`]).
 pub const OPENLINEAGE_TRINO_URI_KEY: &str = "openlineage-event-listener.trino.uri";
@@ -148,6 +152,14 @@ impl ResolvedOpenLineageConfig {
                 .unwrap_or_else(|| namespace.to_string()),
         );
 
+        // The stable OpenLineage job name format. When unset, Trino defaults to `$QUERY_ID`.
+        if let Some(app_name) = &open_lineage.app_name {
+            properties.insert(
+                OPENLINEAGE_JOB_NAME_FORMAT_KEY.to_string(),
+                app_name.clone(),
+            );
+        }
+
         // Backend TLS: mount and import a `SecretClass` CA into the client truststore. WebPKI and
         // no verification need nothing (WebPKI is trusted via the system bundle already seeded into
         // the truststore; without server verification the URL is plain `http`).

rust/operator-binary/src/crd/mod.rs

@@ -296,8 +296,9 @@ pub mod versioned {
         /// Emit [OpenLineage](https://openlineage.io/) lineage events for the queries run on this
         /// Trino cluster. The OpenLineage event listener runs on the coordinator only.
         /// The backend connection is either inlined or references an `OpenLineageConnection` resource.
-        /// The shared `appName` field is not used by Trino, which derives an OpenLineage job name per
-        /// query. See the [OpenLineage usage guide](DOCS_BASE_URL_PLACEHOLDER/trino/usage-guide/openlineage).
+        /// The shared `appName` field sets the OpenLineage job name format
+        /// (`openlineage-event-listener.job.name-format`); when unset Trino defaults to `$QUERY_ID`.
+        /// See the [OpenLineage usage guide](DOCS_BASE_URL_PLACEHOLDER/trino/usage-guide/openlineage).
         #[serde(default, skip_serializing_if = "Option::is_none")]
         pub open_lineage: Option<openlineage::v1alpha1::OpenLineageJob>,
 

tests/templates/kuttl/openlineage/02_trino.yaml.j2

@@ -55,6 +55,8 @@ spec:
 {% if test_scenario['values']['openlineage-use-auth'] == 'true' %}
           authenticationClassRef: openlineage-auth-$NAMESPACE
 {% endif %}
+      # Sets openlineage-event-listener.job.name-format (asserted in 03-assert.yaml).
+      appName: "trino-$QUERY_ID"
 {% if lookup('env', 'VECTOR_AGGREGATOR') %}
     vectorAggregatorConfigMapName: vector-aggregator-discovery
 {% endif %}

tests/templates/kuttl/openlineage/03-assert.yaml

@@ -12,6 +12,8 @@ commands:
       echo "$CM" | grep -q "event-listener.name=openlineage"
       echo "$CM" | grep -q "openlineage-event-listener.transport.url"
       echo "$CM" | grep -q "event-listener.config-files=/stackable/rwconfig/event-listener.properties"
+      # appName maps to the OpenLineage job name format.
+      echo "$CM" | grep -q "openlineage-event-listener.job.name-format=trino-"
   # Workers must NOT get an event listener configured.
   - script: |
       ! kubectl get -n "$NAMESPACE" configmap trino-openlineage-worker-default -o yaml | grep -q "event-listener.properties"