Merge pull request #1 from TobikoData/main

chore: Sync upstream `main`.

Author: walanguzzi

.circleci/continue_config.yml

@@ -258,6 +258,10 @@ jobs:
             echo "export REDSHIFT_DATABASE='$TEST_DB_NAME'" >> "$BASH_ENV"
             echo "export GCP_POSTGRES_DATABASE='$TEST_DB_NAME'" >> "$BASH_ENV"
             echo "export FABRIC_DATABASE='$TEST_DB_NAME'" >> "$BASH_ENV"
+
+            # Make snowflake private key available            
+            echo $SNOWFLAKE_PRIVATE_KEY_RAW | base64 -d > /tmp/snowflake-keyfile.p8
+            echo "export SNOWFLAKE_PRIVATE_KEY_FILE='/tmp/snowflake-keyfile.p8'" >> "$BASH_ENV"
       - run:
           name: Create test database
           command: ./.circleci/manage-test-db.sh << parameters.engine >> "$TEST_DB_NAME" up
@@ -309,7 +313,7 @@ workflows:
           matrix:
             parameters:
               engine:
-                #- snowflake
+                - snowflake
                 - databricks
                 - redshift
                 - bigquery

.circleci/manage-test-db.sh

@@ -25,7 +25,7 @@ function_exists() {
 # Snowflake
 snowflake_init() {
     echo "Installing Snowflake CLI"
-    pip install "snowflake-cli-labs<3.8.0"
+    pip install "snowflake-cli"
 }
 
 snowflake_up() {
@@ -40,20 +40,6 @@ snowflake_down() {
 databricks_init() {
     echo "Installing Databricks CLI"
     curl -fsSL https://raw.githubusercontent.com/databricks/setup-cli/main/install.sh | sudo sh || true
-
-    echo "Writing out Databricks CLI config file"
-    echo -e "[DEFAULT]\nhost = $DATABRICKS_SERVER_HOSTNAME\ntoken = $DATABRICKS_ACCESS_TOKEN" > ~/.databrickscfg
-
-    # this takes a path like 'sql/protocolv1/o/2934659247569/0723-005339-foobar' and extracts '0723-005339-foobar' from it
-    CLUSTER_ID=${DATABRICKS_HTTP_PATH##*/}
-
-    echo "Extracted cluster id: $CLUSTER_ID from '$DATABRICKS_HTTP_PATH'"
-
-    # Note: the cluster doesnt need to be running to create / drop catalogs, but it does need to be running to run the integration tests
-    echo "Ensuring cluster is running"
-    # the || true is to prevent the following error from causing an abort:
-    # > Error: is in unexpected state Running.
-    databricks clusters start $CLUSTER_ID || true
 }
 
 databricks_up() {

.github/workflows/pr.yaml

@@ -30,6 +30,8 @@ jobs:
   test-vscode-e2e:
     runs-on:
       labels: [ubuntu-2204-8]
+    # As at 2026-01-12 this job flakes 100% of the time. It needs investigation
+    if: false
     steps:
       - uses: actions/checkout@v5
       - uses: actions/setup-node@v6

Makefile

@@ -212,14 +212,14 @@ risingwave-test: engine-risingwave-up
 # Cloud Engines #
 #################
 
-snowflake-test: guard-SNOWFLAKE_ACCOUNT guard-SNOWFLAKE_WAREHOUSE guard-SNOWFLAKE_DATABASE guard-SNOWFLAKE_USER guard-SNOWFLAKE_PASSWORD engine-snowflake-install
+snowflake-test: guard-SNOWFLAKE_ACCOUNT guard-SNOWFLAKE_WAREHOUSE guard-SNOWFLAKE_DATABASE guard-SNOWFLAKE_USER engine-snowflake-install
 	pytest -n auto -m "snowflake" --reruns 3 --junitxml=test-results/junit-snowflake.xml
 
 bigquery-test: guard-BIGQUERY_KEYFILE engine-bigquery-install
 	$(PIP) install -e ".[bigframes]"
 	pytest -n auto -m "bigquery" --reruns 3 --junitxml=test-results/junit-bigquery.xml
 
-databricks-test: guard-DATABRICKS_CATALOG guard-DATABRICKS_SERVER_HOSTNAME guard-DATABRICKS_HTTP_PATH guard-DATABRICKS_ACCESS_TOKEN guard-DATABRICKS_CONNECT_VERSION engine-databricks-install
+databricks-test: guard-DATABRICKS_CATALOG guard-DATABRICKS_SERVER_HOSTNAME guard-DATABRICKS_HTTP_PATH guard-DATABRICKS_CONNECT_VERSION engine-databricks-install
 	$(PIP) install 'databricks-connect==${DATABRICKS_CONNECT_VERSION}'
 	pytest -n auto -m "databricks" --reruns 3 --junitxml=test-results/junit-databricks.xml
 

docs/integrations/engines/bigquery.md

@@ -193,6 +193,23 @@ If the `impersonated_service_account` argument is set, SQLMesh will:
 
 The user account must have [sufficient permissions to impersonate the service account](https://cloud.google.com/docs/authentication/use-service-account-impersonation).
 
+## Query Label
+
+BigQuery supports a `query_label` session variable which is attached to query jobs and can be used for auditing / attribution.
+
+SQLMesh supports setting it via `session_properties.query_label` on a model, as an array (or tuple) of key/value tuples.
+
+Example:
+```sql
+MODEL (
+  name my_project.my_dataset.my_model,
+  dialect 'bigquery',
+  session_properties (
+    query_label = [('team', 'data_platform'), ('env', 'prod')]
+  )
+);
+```
+
 ## Permissions Required
 With any of the above connection methods, ensure these BigQuery permissions are enabled to allow SQLMesh to work correctly.
 

docs/integrations/engines/trino.md

@@ -90,6 +90,7 @@ hive.metastore.glue.default-warehouse-dir=s3://my-bucket/
 | `http_scheme`             | The HTTP scheme to use when connecting to your cluster. By default, it's `https` and can only be `http` for no-auth or basic auth.                                                      | string |    N     |
 | `port`                    | The port to connect to your cluster. By default, it's `443` for `https` scheme and `80` for `http`                                                                                      |  int   |    N     |
 | `roles`                   | Mapping of catalog name to a role                                                                                                                                                       |  dict  |    N     |
+| `source`                  | Value to send as Trino's `source` field for query attribution / auditing. Default: `sqlmesh`.                                                                                           | string |    N     |
 | `http_headers`            | Additional HTTP headers to send with each request.                                                                                                                                      |  dict  |    N     |
 | `session_properties`      | Trino session properties. Run `SHOW SESSION` to see all options.                                                                                                                        |  dict  |    N     |
 | `retries`                 | Number of retries to attempt when a request fails. Default: `3`                                                                                                                         |  int   |    N     |