Merge branch 'main' into dependabot/go_modules/golang.org/x/mod-0.21.0

I plan to use this in https://github.com/databricks/cli/pull/1780, to
set the line and column numbers as well for the locations.

gopatch file used:
```
@@
var x expression
var y expression
var z expression
@@
-bundletest.SetLocation(x, y, z)
+bundletest.SetLocation(x, y, []dyn.Location{{File: z}})
```

.codegen.json

@@ -11,10 +11,10 @@
     "toolchain": {
         "required": ["go"],
         "post_generate": [
-            "go run ./bundle/internal/bundle/schema/main.go ./bundle/schema/docs/bundle_descriptions.json",
+            "go run ./bundle/internal/schema/*.go ./bundle/schema/jsonschema.json",
             "echo 'bundle/internal/tf/schema/\\*.go linguist-generated=true' >> ./.gitattributes",
             "echo 'go.sum linguist-generated=true' >> ./.gitattributes",
-            "echo 'bundle/schema/docs/bundle_descriptions.json linguist-generated=true' >> ./.gitattributes"
+            "echo 'bundle/schema/jsonschema.json linguist-generated=true' >> ./.gitattributes"
         ]
     }
 }

.gitattributes

@@ -120,4 +120,4 @@ cmd/workspace/workspace-conf/workspace-conf.go linguist-generated=true
 cmd/workspace/workspace/workspace.go linguist-generated=true
 bundle/internal/tf/schema/\*.go linguist-generated=true
 go.sum linguist-generated=true
-bundle/schema/docs/bundle_descriptions.json linguist-generated=true
+bundle/schema/jsonschema.json linguist-generated=true

.github/workflows/push.yml

@@ -33,7 +33,7 @@ jobs:
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22.x
+          go-version: 1.22.7
 
       - name: Setup Python
         uses: actions/setup-python@v5
@@ -68,7 +68,7 @@ jobs:
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22.x
+          go-version: 1.22.7
 
           # No need to download cached dependencies when running gofmt.
           cache: false
@@ -100,18 +100,25 @@ jobs:
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22.x
+          go-version: 1.22.7
 
       # Github repo: https://github.com/ajv-validator/ajv-cli
       - name: Install ajv-cli
         run: npm install -g ajv-cli@5.0.0
 
       # Assert that the generated bundle schema is a valid JSON schema by using
-      # ajv-cli to validate it against a sample configuration file.
+      # ajv-cli to validate it against bundle configuration files.
       # By default the ajv-cli runs in strict mode which will fail if the schema
       # itself is not valid. Strict mode is more strict than the JSON schema
       # specification. See for details: https://ajv.js.org/options.html#strict-mode-options
       - name: Validate bundle schema
         run: |
           go run main.go bundle schema > schema.json
-          ajv -s schema.json -d ./bundle/tests/basic/databricks.yml
+
+          for file in ./bundle/internal/schema/testdata/pass/*.yml; do
+            ajv test -s schema.json -d $file --valid
+          done
+
+          for file in ./bundle/internal/schema/testdata/fail/*.yml; do
+            ajv test -s schema.json -d $file --invalid
+          done

.github/workflows/release-snapshot.yml

@@ -21,7 +21,7 @@ jobs:
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22.x
+          go-version: 1.22.7
 
           # The default cache key for this action considers only the `go.sum` file.
           # We include .goreleaser.yaml here to differentiate from the cache used by the push action

.github/workflows/release.yml

@@ -22,7 +22,7 @@ jobs:
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22.x
+          go-version: 1.22.7
 
           # The default cache key for this action considers only the `go.sum` file.
           # We include .goreleaser.yaml here to differentiate from the cache used by the push action

CHANGELOG.md

@@ -1,5 +1,32 @@
 # Version changelog
 
+## [Release] Release v0.228.1
+
+Bundles:
+ * Added listing cluster filtering for cluster lookups ([#1754](https://github.com/databricks/cli/pull/1754)).
+ * Expand library globs relative to the sync root ([#1756](https://github.com/databricks/cli/pull/1756)).
+ * Fixed generated YAML missing 'default' for empty values ([#1765](https://github.com/databricks/cli/pull/1765)).
+ * Use periodic triggers in all templates ([#1739](https://github.com/databricks/cli/pull/1739)).
+ * Use the friendly name of service principals when shortening their name ([#1770](https://github.com/databricks/cli/pull/1770)).
+ * Fixed detecting full syntax variable override which includes type field ([#1775](https://github.com/databricks/cli/pull/1775)).
+
+Internal:
+ * Pass copy of `dyn.Path` to callback function ([#1747](https://github.com/databricks/cli/pull/1747)).
+ * Make bundle JSON schema modular with `` ([#1700](https://github.com/databricks/cli/pull/1700)).
+ * Alias variables block in the `Target` struct ([#1748](https://github.com/databricks/cli/pull/1748)).
+ * Add end to end integration tests for bundle JSON schema ([#1726](https://github.com/databricks/cli/pull/1726)).
+ * Fix artifact upload integration tests ([#1767](https://github.com/databricks/cli/pull/1767)).
+
+API Changes:
+ * Added `databricks quality-monitors regenerate-dashboard` command.
+
+OpenAPI commit d05898328669a3f8ab0c2ecee37db2673d3ea3f7 (2024-09-04)
+Dependency updates:
+ * Bump golang.org/x/term from 0.23.0 to 0.24.0 ([#1757](https://github.com/databricks/cli/pull/1757)).
+ * Bump golang.org/x/oauth2 from 0.22.0 to 0.23.0 ([#1761](https://github.com/databricks/cli/pull/1761)).
+ * Bump golang.org/x/text from 0.17.0 to 0.18.0 ([#1759](https://github.com/databricks/cli/pull/1759)).
+ * Bump github.com/databricks/databricks-sdk-go from 0.45.0 to 0.46.0 ([#1760](https://github.com/databricks/cli/pull/1760)).
+
 ## [Release] Release v0.228.0
 
 CLI:

bundle/artifacts/expand_globs_test.go

@@ -10,6 +10,7 @@ import (
 	"github.com/databricks/cli/bundle/config"
 	"github.com/databricks/cli/bundle/internal/bundletest"
 	"github.com/databricks/cli/internal/testutil"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -36,7 +37,7 @@ func TestExpandGlobs_Nominal(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "artifacts", filepath.Join(tmpDir, "databricks.yml"))
+	bundletest.SetLocation(b, "artifacts", []dyn.Location{{File: filepath.Join(tmpDir, "databricks.yml")}})
 
 	ctx := context.Background()
 	diags := bundle.Apply(ctx, b, bundle.Seq(
@@ -77,7 +78,7 @@ func TestExpandGlobs_InvalidPattern(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "artifacts", filepath.Join(tmpDir, "databricks.yml"))
+	bundletest.SetLocation(b, "artifacts", []dyn.Location{{File: filepath.Join(tmpDir, "databricks.yml")}})
 
 	ctx := context.Background()
 	diags := bundle.Apply(ctx, b, bundle.Seq(
@@ -125,7 +126,7 @@ func TestExpandGlobs_NoMatches(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "artifacts", filepath.Join(tmpDir, "databricks.yml"))
+	bundletest.SetLocation(b, "artifacts", []dyn.Location{{File: filepath.Join(tmpDir, "databricks.yml")}})
 
 	ctx := context.Background()
 	diags := bundle.Apply(ctx, b, bundle.Seq(

bundle/config/bundle.go

@@ -38,8 +38,11 @@ type Bundle struct {
 	// Annotated readonly as this should be set at the target level.
 	Mode Mode `json:"mode,omitempty" bundle:"readonly"`
 
-	// Overrides the compute used for jobs and other supported assets.
-	ComputeID string `json:"compute_id,omitempty"`
+	// DEPRECATED: Overrides the compute used for jobs and other supported assets.
+	ComputeId string `json:"compute_id,omitempty"`
+
+	// Overrides the cluster used for jobs and other supported assets.
+	ClusterId string `json:"cluster_id,omitempty"`
 
 	// Deployment section specifies deployment related configuration for bundle
 	Deployment Deployment `json:"deployment,omitempty"`

bundle/config/generate/job.go

@@ -25,6 +25,20 @@ func ConvertJobToValue(job *jobs.Job) (dyn.Value, error) {
 		value["tasks"] = dyn.NewValue(tasks, []dyn.Location{{Line: jobOrder.Get("tasks")}})
 	}
 
+	// We're processing job.Settings.Parameters separately to retain empty default values.
+	if len(job.Settings.Parameters) > 0 {
+		params := make([]dyn.Value, 0)
+		for _, parameter := range job.Settings.Parameters {
+			p := map[string]dyn.Value{
+				"name":    dyn.NewValue(parameter.Name, []dyn.Location{{Line: 0}}), // We use Line: 0 to ensure that the name goes first.
+				"default": dyn.NewValue(parameter.Default, []dyn.Location{{Line: 1}}),
+			}
+			params = append(params, dyn.NewValue(p, []dyn.Location{}))
+		}
+
+		value["parameters"] = dyn.NewValue(params, []dyn.Location{{Line: jobOrder.Get("parameters")}})
+	}
+
 	return yamlsaver.ConvertToMapValue(job.Settings, jobOrder, []string{"format", "new_cluster", "existing_cluster_id"}, value)
 }
 

bundle/config/mutator/apply_presets.go

@@ -160,6 +160,21 @@ func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnos
 		// the Databricks UI and via the SQL API.
 	}
 
+	// Clusters: Prefix, Tags
+	for _, c := range r.Clusters {
+		c.ClusterName = prefix + c.ClusterName
+		if c.CustomTags == nil {
+			c.CustomTags = make(map[string]string)
+		}
+		for _, tag := range tags {
+			normalisedKey := b.Tagging.NormalizeKey(tag.Key)
+			normalisedValue := b.Tagging.NormalizeValue(tag.Value)
+			if _, ok := c.CustomTags[normalisedKey]; !ok {
+				c.CustomTags[normalisedKey] = normalisedValue
+			}
+		}
+	}
+
 	return nil
 }
 

bundle/config/mutator/compute_id_compat.go

@@ -0,0 +1,87 @@
+package mutator
+
+import (
+	"context"
+
+	"github.com/databricks/cli/bundle"
+	"github.com/databricks/cli/libs/diag"
+	"github.com/databricks/cli/libs/dyn"
+)
+
+type computeIdToClusterId struct{}
+
+func ComputeIdToClusterId() bundle.Mutator {
+	return &computeIdToClusterId{}
+}
+
+func (m *computeIdToClusterId) Name() string {
+	return "ComputeIdToClusterId"
+}
+
+func (m *computeIdToClusterId) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics {
+	var diags diag.Diagnostics
+
+	// The "compute_id" key is set; rewrite it to "cluster_id".
+	err := b.Config.Mutate(func(v dyn.Value) (dyn.Value, error) {
+		v, d := rewriteComputeIdToClusterId(v, dyn.NewPath(dyn.Key("bundle")))
+		diags = diags.Extend(d)
+
+		// Check if the "compute_id" key is set in any target overrides.
+		return dyn.MapByPattern(v, dyn.NewPattern(dyn.Key("targets"), dyn.AnyKey()), func(p dyn.Path, v dyn.Value) (dyn.Value, error) {
+			v, d := rewriteComputeIdToClusterId(v, dyn.Path{})
+			diags = diags.Extend(d)
+			return v, nil
+		})
+	})
+
+	diags = diags.Extend(diag.FromErr(err))
+	return diags
+}
+
+func rewriteComputeIdToClusterId(v dyn.Value, p dyn.Path) (dyn.Value, diag.Diagnostics) {
+	var diags diag.Diagnostics
+	computeIdPath := p.Append(dyn.Key("compute_id"))
+	computeId, err := dyn.GetByPath(v, computeIdPath)
+
+	// If the "compute_id" key is not set, we don't need to do anything.
+	if err != nil {
+		return v, nil
+	}
+
+	if computeId.Kind() == dyn.KindInvalid {
+		return v, nil
+	}
+
+	diags = diags.Append(diag.Diagnostic{
+		Severity:  diag.Warning,
+		Summary:   "compute_id is deprecated, please use cluster_id instead",
+		Locations: computeId.Locations(),
+		Paths:     []dyn.Path{computeIdPath},
+	})
+
+	clusterIdPath := p.Append(dyn.Key("cluster_id"))
+	nv, err := dyn.SetByPath(v, clusterIdPath, computeId)
+	if err != nil {
+		return dyn.InvalidValue, diag.FromErr(err)
+	}
+	// Drop the "compute_id" key.
+	vout, err := dyn.Walk(nv, func(p dyn.Path, v dyn.Value) (dyn.Value, error) {
+		switch len(p) {
+		case 0:
+			return v, nil
+		case 1:
+			if p[0] == dyn.Key("compute_id") {
+				return v, dyn.ErrDrop
+			}
+			return v, nil
+		case 2:
+			if p[1] == dyn.Key("compute_id") {
+				return v, dyn.ErrDrop
+			}
+		}
+		return v, dyn.ErrSkip
+	})
+
+	diags = diags.Extend(diag.FromErr(err))
+	return vout, diags
+}

bundle/config/mutator/compute_id_compate_test.go

@@ -0,0 +1,57 @@
+package mutator_test
+
+import (
+	"context"
+	"testing"
+
+	"github.com/databricks/cli/bundle"
+	"github.com/databricks/cli/bundle/config"
+	"github.com/databricks/cli/bundle/config/mutator"
+	"github.com/databricks/cli/libs/diag"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestComputeIdToClusterId(t *testing.T) {
+	b := &bundle.Bundle{
+		Config: config.Root{
+			Bundle: config.Bundle{
+				ComputeId: "compute-id",
+			},
+		},
+	}
+
+	diags := bundle.Apply(context.Background(), b, mutator.ComputeIdToClusterId())
+	assert.NoError(t, diags.Error())
+	assert.Equal(t, "compute-id", b.Config.Bundle.ClusterId)
+	assert.Empty(t, b.Config.Bundle.ComputeId)
+
+	assert.Len(t, diags, 1)
+	assert.Equal(t, "compute_id is deprecated, please use cluster_id instead", diags[0].Summary)
+	assert.Equal(t, diag.Warning, diags[0].Severity)
+}
+
+func TestComputeIdToClusterIdInTargetOverride(t *testing.T) {
+	b := &bundle.Bundle{
+		Config: config.Root{
+			Targets: map[string]*config.Target{
+				"dev": {
+					ComputeId: "compute-id-dev",
+				},
+			},
+		},
+	}
+
+	diags := bundle.Apply(context.Background(), b, mutator.ComputeIdToClusterId())
+	assert.NoError(t, diags.Error())
+	assert.Empty(t, b.Config.Targets["dev"].ComputeId)
+
+	diags = diags.Extend(bundle.Apply(context.Background(), b, mutator.SelectTarget("dev")))
+	assert.NoError(t, diags.Error())
+
+	assert.Equal(t, "compute-id-dev", b.Config.Bundle.ClusterId)
+	assert.Empty(t, b.Config.Bundle.ComputeId)
+
+	assert.Len(t, diags, 1)
+	assert.Equal(t, "compute_id is deprecated, please use cluster_id instead", diags[0].Summary)
+	assert.Equal(t, diag.Warning, diags[0].Severity)
+}

bundle/config/mutator/expand_pipeline_glob_paths_test.go

@@ -10,6 +10,7 @@ import (
 	"github.com/databricks/cli/bundle/config"
 	"github.com/databricks/cli/bundle/config/resources"
 	"github.com/databricks/cli/bundle/internal/bundletest"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/databricks/databricks-sdk-go/service/compute"
 	"github.com/databricks/databricks-sdk-go/service/pipelines"
 	"github.com/stretchr/testify/require"
@@ -105,8 +106,8 @@ func TestExpandGlobPathsInPipelines(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
-	bundletest.SetLocation(b, "resources.pipelines.pipeline.libraries[3]", filepath.Join(dir, "relative", "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
+	bundletest.SetLocation(b, "resources.pipelines.pipeline.libraries[3]", []dyn.Location{{File: filepath.Join(dir, "relative", "resource.yml")}})
 
 	m := ExpandPipelineGlobPaths()
 	diags := bundle.Apply(context.Background(), b, m)

bundle/config/mutator/mutator.go

@@ -23,6 +23,7 @@ func DefaultMutators() []bundle.Mutator {
 		VerifyCliVersion(),
 
 		EnvironmentsToTargets(),
+		ComputeIdToClusterId(),
 		InitializeVariables(),
 		DefineDefaultTarget(),
 		LoadGitDetails(),

bundle/config/mutator/override_compute.go

@@ -39,22 +39,22 @@ func overrideJobCompute(j *resources.Job, compute string) {
 
 func (m *overrideCompute) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics {
 	if b.Config.Bundle.Mode != config.Development {
-		if b.Config.Bundle.ComputeID != "" {
+		if b.Config.Bundle.ClusterId != "" {
 			return diag.Errorf("cannot override compute for an target that does not use 'mode: development'")
 		}
 		return nil
 	}
 	if v := env.Get(ctx, "DATABRICKS_CLUSTER_ID"); v != "" {
-		b.Config.Bundle.ComputeID = v
+		b.Config.Bundle.ClusterId = v
 	}
 
-	if b.Config.Bundle.ComputeID == "" {
+	if b.Config.Bundle.ClusterId == "" {
 		return nil
 	}
 
 	r := b.Config.Resources
 	for i := range r.Jobs {
-		overrideJobCompute(r.Jobs[i], b.Config.Bundle.ComputeID)
+		overrideJobCompute(r.Jobs[i], b.Config.Bundle.ClusterId)
 	}
 
 	return nil

bundle/config/mutator/override_compute_test.go

@@ -20,7 +20,7 @@ func TestOverrideDevelopment(t *testing.T) {
 		Config: config.Root{
 			Bundle: config.Bundle{
 				Mode:      config.Development,
-				ComputeID: "newClusterID",
+				ClusterId: "newClusterID",
 			},
 			Resources: config.Resources{
 				Jobs: map[string]*resources.Job{
@@ -144,7 +144,7 @@ func TestOverrideProduction(t *testing.T) {
 	b := &bundle.Bundle{
 		Config: config.Root{
 			Bundle: config.Bundle{
-				ComputeID: "newClusterID",
+				ClusterId: "newClusterID",
 			},
 			Resources: config.Resources{
 				Jobs: map[string]*resources.Job{

bundle/config/mutator/paths/job_paths_visitor.go

@@ -0,0 +1,115 @@
+package paths
+
+import (
+	"github.com/databricks/cli/bundle/libraries"
+	"github.com/databricks/cli/libs/dyn"
+)
+
+type jobRewritePattern struct {
+	pattern     dyn.Pattern
+	kind        PathKind
+	skipRewrite func(string) bool
+}
+
+func noSkipRewrite(string) bool {
+	return false
+}
+
+func jobTaskRewritePatterns(base dyn.Pattern) []jobRewritePattern {
+	return []jobRewritePattern{
+		{
+			base.Append(dyn.Key("notebook_task"), dyn.Key("notebook_path")),
+			PathKindNotebook,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("spark_python_task"), dyn.Key("python_file")),
+			PathKindWorkspaceFile,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("dbt_task"), dyn.Key("project_directory")),
+			PathKindDirectory,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("sql_task"), dyn.Key("file"), dyn.Key("path")),
+			PathKindWorkspaceFile,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("whl")),
+			PathKindLibrary,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("jar")),
+			PathKindLibrary,
+			noSkipRewrite,
+		},
+		{
+			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("requirements")),
+			PathKindWorkspaceFile,
+			noSkipRewrite,
+		},
+	}
+}
+
+func jobRewritePatterns() []jobRewritePattern {
+	// Base pattern to match all tasks in all jobs.
+	base := dyn.NewPattern(
+		dyn.Key("resources"),
+		dyn.Key("jobs"),
+		dyn.AnyKey(),
+		dyn.Key("tasks"),
+		dyn.AnyIndex(),
+	)
+
+	// Compile list of patterns and their respective rewrite functions.
+	jobEnvironmentsPatterns := []jobRewritePattern{
+		{
+			dyn.NewPattern(
+				dyn.Key("resources"),
+				dyn.Key("jobs"),
+				dyn.AnyKey(),
+				dyn.Key("environments"),
+				dyn.AnyIndex(),
+				dyn.Key("spec"),
+				dyn.Key("dependencies"),
+				dyn.AnyIndex(),
+			),
+			PathKindWithPrefix,
+			func(s string) bool {
+				return !libraries.IsLibraryLocal(s)
+			},
+		},
+	}
+
+	taskPatterns := jobTaskRewritePatterns(base)
+	forEachPatterns := jobTaskRewritePatterns(base.Append(dyn.Key("for_each_task"), dyn.Key("task")))
+	allPatterns := append(taskPatterns, jobEnvironmentsPatterns...)
+	allPatterns = append(allPatterns, forEachPatterns...)
+	return allPatterns
+}
+
+// VisitJobPaths visits all paths in job resources and applies a function to each path.
+func VisitJobPaths(value dyn.Value, fn VisitFunc) (dyn.Value, error) {
+	var err error
+	var newValue = value
+
+	for _, rewritePattern := range jobRewritePatterns() {
+		newValue, err = dyn.MapByPattern(newValue, rewritePattern.pattern, func(p dyn.Path, v dyn.Value) (dyn.Value, error) {
+			if rewritePattern.skipRewrite(v.MustString()) {
+				return v, nil
+			}
+
+			return fn(p, rewritePattern.kind, v)
+		})
+
+		if err != nil {
+			return dyn.InvalidValue, err
+		}
+	}
+
+	return newValue, nil
+}

bundle/config/mutator/paths/job_paths_visitor_test.go

@@ -0,0 +1,168 @@
+package paths
+
+import (
+	"testing"
+
+	"github.com/databricks/cli/bundle/config"
+	"github.com/databricks/cli/bundle/config/resources"
+	"github.com/databricks/cli/libs/dyn"
+	assert "github.com/databricks/cli/libs/dyn/dynassert"
+	"github.com/databricks/databricks-sdk-go/service/compute"
+	"github.com/databricks/databricks-sdk-go/service/jobs"
+	"github.com/stretchr/testify/require"
+)
+
+func TestVisitJobPaths(t *testing.T) {
+	task0 := jobs.Task{
+		NotebookTask: &jobs.NotebookTask{
+			NotebookPath: "abc",
+		},
+	}
+	task1 := jobs.Task{
+		SparkPythonTask: &jobs.SparkPythonTask{
+			PythonFile: "abc",
+		},
+	}
+	task2 := jobs.Task{
+		DbtTask: &jobs.DbtTask{
+			ProjectDirectory: "abc",
+		},
+	}
+	task3 := jobs.Task{
+		SqlTask: &jobs.SqlTask{
+			File: &jobs.SqlTaskFile{
+				Path: "abc",
+			},
+		},
+	}
+	task4 := jobs.Task{
+		Libraries: []compute.Library{
+			{Whl: "dist/foo.whl"},
+		},
+	}
+	task5 := jobs.Task{
+		Libraries: []compute.Library{
+			{Jar: "dist/foo.jar"},
+		},
+	}
+	task6 := jobs.Task{
+		Libraries: []compute.Library{
+			{Requirements: "requirements.txt"},
+		},
+	}
+
+	job0 := &resources.Job{
+		JobSettings: &jobs.JobSettings{
+			Tasks: []jobs.Task{
+				task0,
+				task1,
+				task2,
+				task3,
+				task4,
+				task5,
+				task6,
+			},
+		},
+	}
+
+	root := config.Root{
+		Resources: config.Resources{
+			Jobs: map[string]*resources.Job{
+				"job0": job0,
+			},
+		},
+	}
+
+	actual := visitJobPaths(t, root)
+	expected := []dyn.Path{
+		dyn.MustPathFromString("resources.jobs.job0.tasks[0].notebook_task.notebook_path"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[1].spark_python_task.python_file"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[2].dbt_task.project_directory"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[3].sql_task.file.path"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[4].libraries[0].whl"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[5].libraries[0].jar"),
+		dyn.MustPathFromString("resources.jobs.job0.tasks[6].libraries[0].requirements"),
+	}
+
+	assert.ElementsMatch(t, expected, actual)
+}
+
+func TestVisitJobPaths_environments(t *testing.T) {
+	environment0 := jobs.JobEnvironment{
+		Spec: &compute.Environment{
+			Dependencies: []string{
+				"dist_0/*.whl",
+				"dist_1/*.whl",
+			},
+		},
+	}
+	job0 := &resources.Job{
+		JobSettings: &jobs.JobSettings{
+			Environments: []jobs.JobEnvironment{
+				environment0,
+			},
+		},
+	}
+
+	root := config.Root{
+		Resources: config.Resources{
+			Jobs: map[string]*resources.Job{
+				"job0": job0,
+			},
+		},
+	}
+
+	actual := visitJobPaths(t, root)
+	expected := []dyn.Path{
+		dyn.MustPathFromString("resources.jobs.job0.environments[0].spec.dependencies[0]"),
+		dyn.MustPathFromString("resources.jobs.job0.environments[0].spec.dependencies[1]"),
+	}
+
+	assert.ElementsMatch(t, expected, actual)
+}
+
+func TestVisitJobPaths_foreach(t *testing.T) {
+	task0 := jobs.Task{
+		ForEachTask: &jobs.ForEachTask{
+			Task: jobs.Task{
+				NotebookTask: &jobs.NotebookTask{
+					NotebookPath: "abc",
+				},
+			},
+		},
+	}
+	job0 := &resources.Job{
+		JobSettings: &jobs.JobSettings{
+			Tasks: []jobs.Task{
+				task0,
+			},
+		},
+	}
+
+	root := config.Root{
+		Resources: config.Resources{
+			Jobs: map[string]*resources.Job{
+				"job0": job0,
+			},
+		},
+	}
+
+	actual := visitJobPaths(t, root)
+	expected := []dyn.Path{
+		dyn.MustPathFromString("resources.jobs.job0.tasks[0].for_each_task.task.notebook_task.notebook_path"),
+	}
+
+	assert.ElementsMatch(t, expected, actual)
+}
+
+func visitJobPaths(t *testing.T, root config.Root) []dyn.Path {
+	var actual []dyn.Path
+	err := root.Mutate(func(value dyn.Value) (dyn.Value, error) {
+		return VisitJobPaths(value, func(p dyn.Path, kind PathKind, v dyn.Value) (dyn.Value, error) {
+			actual = append(actual, p)
+			return v, nil
+		})
+	})
+	require.NoError(t, err)
+	return actual
+}

bundle/config/mutator/paths/visitor.go

@@ -0,0 +1,26 @@
+package paths
+
+import "github.com/databricks/cli/libs/dyn"
+
+type PathKind int
+
+const (
+	// PathKindLibrary is a path to a library file
+	PathKindLibrary = iota
+
+	// PathKindNotebook is a path to a notebook file
+	PathKindNotebook
+
+	// PathKindWorkspaceFile is a path to a regular workspace file,
+	// notebooks are not allowed because they are uploaded a special
+	// kind of workspace object.
+	PathKindWorkspaceFile
+
+	// PathKindWithPrefix is a path that starts with './'
+	PathKindWithPrefix
+
+	// PathKindDirectory is a path to directory
+	PathKindDirectory
+)
+
+type VisitFunc func(path dyn.Path, kind PathKind, value dyn.Value) (dyn.Value, error)

bundle/config/mutator/populate_current_user.go

@@ -33,7 +33,7 @@ func (m *populateCurrentUser) Apply(ctx context.Context, b *bundle.Bundle) diag.
 	}
 
 	b.Config.Workspace.CurrentUser = &config.User{
-		ShortName: auth.GetShortUserName(me.UserName),
+		ShortName: auth.GetShortUserName(me),
 		User:      me,
 	}
 

bundle/config/mutator/process_target_mode_test.go

@@ -13,6 +13,7 @@ import (
 	"github.com/databricks/cli/libs/tags"
 	sdkconfig "github.com/databricks/databricks-sdk-go/config"
 	"github.com/databricks/databricks-sdk-go/service/catalog"
+	"github.com/databricks/databricks-sdk-go/service/compute"
 	"github.com/databricks/databricks-sdk-go/service/iam"
 	"github.com/databricks/databricks-sdk-go/service/jobs"
 	"github.com/databricks/databricks-sdk-go/service/ml"
@@ -119,6 +120,9 @@ func mockBundle(mode config.Mode) *bundle.Bundle {
 				Schemas: map[string]*resources.Schema{
 					"schema1": {CreateSchema: &catalog.CreateSchema{Name: "schema1"}},
 				},
+				Clusters: map[string]*resources.Cluster{
+					"cluster1": {ClusterSpec: &compute.ClusterSpec{ClusterName: "cluster1", SparkVersion: "13.2.x", NumWorkers: 1}},
+				},
 			},
 		},
 		// Use AWS implementation for testing.
@@ -177,6 +181,9 @@ func TestProcessTargetModeDevelopment(t *testing.T) {
 
 	// Schema 1
 	assert.Equal(t, "dev_lennart_schema1", b.Config.Resources.Schemas["schema1"].Name)
+
+	// Clusters
+	assert.Equal(t, "[dev lennart] cluster1", b.Config.Resources.Clusters["cluster1"].ClusterName)
 }
 
 func TestProcessTargetModeDevelopmentTagNormalizationForAws(t *testing.T) {
@@ -281,6 +288,7 @@ func TestProcessTargetModeDefault(t *testing.T) {
 	assert.Equal(t, "servingendpoint1", b.Config.Resources.ModelServingEndpoints["servingendpoint1"].Name)
 	assert.Equal(t, "registeredmodel1", b.Config.Resources.RegisteredModels["registeredmodel1"].Name)
 	assert.Equal(t, "qualityMonitor1", b.Config.Resources.QualityMonitors["qualityMonitor1"].TableName)
+	assert.Equal(t, "cluster1", b.Config.Resources.Clusters["cluster1"].ClusterName)
 }
 
 func TestProcessTargetModeProduction(t *testing.T) {
@@ -312,6 +320,7 @@ func TestProcessTargetModeProduction(t *testing.T) {
 	b.Config.Resources.Experiments["experiment2"].Permissions = permissions
 	b.Config.Resources.Models["model1"].Permissions = permissions
 	b.Config.Resources.ModelServingEndpoints["servingendpoint1"].Permissions = permissions
+	b.Config.Resources.Clusters["cluster1"].Permissions = permissions
 
 	diags = validateProductionMode(context.Background(), b, false)
 	require.NoError(t, diags.Error())
@@ -322,6 +331,7 @@ func TestProcessTargetModeProduction(t *testing.T) {
 	assert.Equal(t, "servingendpoint1", b.Config.Resources.ModelServingEndpoints["servingendpoint1"].Name)
 	assert.Equal(t, "registeredmodel1", b.Config.Resources.RegisteredModels["registeredmodel1"].Name)
 	assert.Equal(t, "qualityMonitor1", b.Config.Resources.QualityMonitors["qualityMonitor1"].TableName)
+	assert.Equal(t, "cluster1", b.Config.Resources.Clusters["cluster1"].ClusterName)
 }
 
 func TestProcessTargetModeProductionOkForPrincipal(t *testing.T) {

bundle/config/mutator/rewrite_sync_paths_test.go

@@ -9,6 +9,7 @@ import (
 	"github.com/databricks/cli/bundle/config"
 	"github.com/databricks/cli/bundle/config/mutator"
 	"github.com/databricks/cli/bundle/internal/bundletest"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/stretchr/testify/assert"
 )
 
@@ -33,12 +34,12 @@ func TestRewriteSyncPathsRelative(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "sync.paths[0]", "./databricks.yml")
-	bundletest.SetLocation(b, "sync.paths[1]", "./databricks.yml")
-	bundletest.SetLocation(b, "sync.include[0]", "./file.yml")
-	bundletest.SetLocation(b, "sync.include[1]", "./a/file.yml")
-	bundletest.SetLocation(b, "sync.exclude[0]", "./a/b/file.yml")
-	bundletest.SetLocation(b, "sync.exclude[1]", "./a/b/c/file.yml")
+	bundletest.SetLocation(b, "sync.paths[0]", []dyn.Location{{File: "./databricks.yml"}})
+	bundletest.SetLocation(b, "sync.paths[1]", []dyn.Location{{File: "./databricks.yml"}})
+	bundletest.SetLocation(b, "sync.include[0]", []dyn.Location{{File: "./file.yml"}})
+	bundletest.SetLocation(b, "sync.include[1]", []dyn.Location{{File: "./a/file.yml"}})
+	bundletest.SetLocation(b, "sync.exclude[0]", []dyn.Location{{File: "./a/b/file.yml"}})
+	bundletest.SetLocation(b, "sync.exclude[1]", []dyn.Location{{File: "./a/b/c/file.yml"}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.RewriteSyncPaths())
 	assert.NoError(t, diags.Error())
@@ -72,12 +73,12 @@ func TestRewriteSyncPathsAbsolute(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "sync.paths[0]", "/tmp/dir/databricks.yml")
-	bundletest.SetLocation(b, "sync.paths[1]", "/tmp/dir/databricks.yml")
-	bundletest.SetLocation(b, "sync.include[0]", "/tmp/dir/file.yml")
-	bundletest.SetLocation(b, "sync.include[1]", "/tmp/dir/a/file.yml")
-	bundletest.SetLocation(b, "sync.exclude[0]", "/tmp/dir/a/b/file.yml")
-	bundletest.SetLocation(b, "sync.exclude[1]", "/tmp/dir/a/b/c/file.yml")
+	bundletest.SetLocation(b, "sync.paths[0]", []dyn.Location{{File: "/tmp/dir/databricks.yml"}})
+	bundletest.SetLocation(b, "sync.paths[1]", []dyn.Location{{File: "/tmp/dir/databricks.yml"}})
+	bundletest.SetLocation(b, "sync.include[0]", []dyn.Location{{File: "/tmp/dir/file.yml"}})
+	bundletest.SetLocation(b, "sync.include[1]", []dyn.Location{{File: "/tmp/dir/a/file.yml"}})
+	bundletest.SetLocation(b, "sync.exclude[0]", []dyn.Location{{File: "/tmp/dir/a/b/file.yml"}})
+	bundletest.SetLocation(b, "sync.exclude[1]", []dyn.Location{{File: "/tmp/dir/a/b/c/file.yml"}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.RewriteSyncPaths())
 	assert.NoError(t, diags.Error())

bundle/config/mutator/run_as_test.go

@@ -32,6 +32,7 @@ func allResourceTypes(t *testing.T) []string {
 	// the dyn library gives us the correct list of all resources supported. Please
 	// also update this check when adding a new resource
 	require.Equal(t, []string{
+		"clusters",
 		"experiments",
 		"jobs",
 		"model_serving_endpoints",
@@ -133,6 +134,7 @@ func TestRunAsErrorForUnsupportedResources(t *testing.T) {
 	// some point in the future. These resources are (implicitly) on the deny list, since
 	// they are not on the allow list below.
 	allowList := []string{
+		"clusters",
 		"jobs",
 		"models",
 		"registered_models",

bundle/config/mutator/sync_infer_root_test.go

@@ -9,6 +9,7 @@ import (
 	"github.com/databricks/cli/bundle/config"
 	"github.com/databricks/cli/bundle/config/mutator"
 	"github.com/databricks/cli/bundle/internal/bundletest"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -184,7 +185,7 @@ func TestSyncInferRoot_Error(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "sync.paths", "databricks.yml")
+	bundletest.SetLocation(b, "sync.paths", []dyn.Location{{File: "databricks.yml"}})
 
 	ctx := context.Background()
 	diags := bundle.Apply(ctx, b, mutator.SyncInferRoot())

bundle/config/mutator/translate_paths_jobs.go

@@ -4,97 +4,11 @@ import (
 	"fmt"
 	"slices"
 
-	"github.com/databricks/cli/bundle/libraries"
+	"github.com/databricks/cli/bundle/config/mutator/paths"
+
 	"github.com/databricks/cli/libs/dyn"
 )
 
-type jobRewritePattern struct {
-	pattern     dyn.Pattern
-	fn          rewriteFunc
-	skipRewrite func(string) bool
-}
-
-func noSkipRewrite(string) bool {
-	return false
-}
-
-func rewritePatterns(t *translateContext, base dyn.Pattern) []jobRewritePattern {
-	return []jobRewritePattern{
-		{
-			base.Append(dyn.Key("notebook_task"), dyn.Key("notebook_path")),
-			t.translateNotebookPath,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("spark_python_task"), dyn.Key("python_file")),
-			t.translateFilePath,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("dbt_task"), dyn.Key("project_directory")),
-			t.translateDirectoryPath,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("sql_task"), dyn.Key("file"), dyn.Key("path")),
-			t.translateFilePath,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("whl")),
-			t.translateNoOp,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("jar")),
-			t.translateNoOp,
-			noSkipRewrite,
-		},
-		{
-			base.Append(dyn.Key("libraries"), dyn.AnyIndex(), dyn.Key("requirements")),
-			t.translateFilePath,
-			noSkipRewrite,
-		},
-	}
-}
-
-func (t *translateContext) jobRewritePatterns() []jobRewritePattern {
-	// Base pattern to match all tasks in all jobs.
-	base := dyn.NewPattern(
-		dyn.Key("resources"),
-		dyn.Key("jobs"),
-		dyn.AnyKey(),
-		dyn.Key("tasks"),
-		dyn.AnyIndex(),
-	)
-
-	// Compile list of patterns and their respective rewrite functions.
-	jobEnvironmentsPatterns := []jobRewritePattern{
-		{
-			dyn.NewPattern(
-				dyn.Key("resources"),
-				dyn.Key("jobs"),
-				dyn.AnyKey(),
-				dyn.Key("environments"),
-				dyn.AnyIndex(),
-				dyn.Key("spec"),
-				dyn.Key("dependencies"),
-				dyn.AnyIndex(),
-			),
-			t.translateNoOpWithPrefix,
-			func(s string) bool {
-				return !libraries.IsLibraryLocal(s)
-			},
-		},
-	}
-
-	taskPatterns := rewritePatterns(t, base)
-	forEachPatterns := rewritePatterns(t, base.Append(dyn.Key("for_each_task"), dyn.Key("task")))
-	allPatterns := append(taskPatterns, jobEnvironmentsPatterns...)
-	allPatterns = append(allPatterns, forEachPatterns...)
-	return allPatterns
-}
-
 func (t *translateContext) applyJobTranslations(v dyn.Value) (dyn.Value, error) {
 	var err error
 
@@ -111,30 +25,41 @@ func (t *translateContext) applyJobTranslations(v dyn.Value) (dyn.Value, error)
 		}
 	}
 
-	for _, rewritePattern := range t.jobRewritePatterns() {
-		v, err = dyn.MapByPattern(v, rewritePattern.pattern, func(p dyn.Path, v dyn.Value) (dyn.Value, error) {
-			key := p[2].Key()
+	return paths.VisitJobPaths(v, func(p dyn.Path, kind paths.PathKind, v dyn.Value) (dyn.Value, error) {
+		key := p[2].Key()
 
-			// Skip path translation if the job is using git source.
-			if slices.Contains(ignore, key) {
-				return v, nil
-			}
+		// Skip path translation if the job is using git source.
+		if slices.Contains(ignore, key) {
+			return v, nil
+		}
 
-			dir, err := v.Location().Directory()
-			if err != nil {
-				return dyn.InvalidValue, fmt.Errorf("unable to determine directory for job %s: %w", key, err)
-			}
+		dir, err := v.Location().Directory()
+		if err != nil {
+			return dyn.InvalidValue, fmt.Errorf("unable to determine directory for job %s: %w", key, err)
+		}
 
-			sv := v.MustString()
-			if rewritePattern.skipRewrite(sv) {
-				return v, nil
-			}
-			return t.rewriteRelativeTo(p, v, rewritePattern.fn, dir, fallback[key])
-		})
+		rewritePatternFn, err := t.getRewritePatternFn(kind)
 		if err != nil {
 			return dyn.InvalidValue, err
 		}
+
+		return t.rewriteRelativeTo(p, v, rewritePatternFn, dir, fallback[key])
+	})
+}
+
+func (t *translateContext) getRewritePatternFn(kind paths.PathKind) (rewriteFunc, error) {
+	switch kind {
+	case paths.PathKindLibrary:
+		return t.translateNoOp, nil
+	case paths.PathKindNotebook:
+		return t.translateNotebookPath, nil
+	case paths.PathKindWorkspaceFile:
+		return t.translateFilePath, nil
+	case paths.PathKindDirectory:
+		return t.translateDirectoryPath, nil
+	case paths.PathKindWithPrefix:
+		return t.translateNoOpWithPrefix, nil
 	}
 
-	return v, nil
+	return nil, fmt.Errorf("unsupported path kind: %d", kind)
 }

bundle/config/mutator/translate_paths_test.go

@@ -82,7 +82,7 @@ func TestTranslatePathsSkippedWithGitSource(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	require.NoError(t, diags.Error())
@@ -210,7 +210,7 @@ func TestTranslatePaths(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	require.NoError(t, diags.Error())
@@ -346,8 +346,8 @@ func TestTranslatePathsInSubdirectories(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "resources.jobs", filepath.Join(dir, "job/resource.yml"))
-	bundletest.SetLocation(b, "resources.pipelines", filepath.Join(dir, "pipeline/resource.yml"))
+	bundletest.SetLocation(b, "resources.jobs", []dyn.Location{{File: filepath.Join(dir, "job/resource.yml")}})
+	bundletest.SetLocation(b, "resources.pipelines", []dyn.Location{{File: filepath.Join(dir, "pipeline/resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	require.NoError(t, diags.Error())
@@ -408,7 +408,7 @@ func TestTranslatePathsOutsideSyncRoot(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "../resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "../resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.ErrorContains(t, diags.Error(), "is not contained in sync root path")
@@ -439,7 +439,7 @@ func TestJobNotebookDoesNotExistError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "fake.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "fake.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.EqualError(t, diags.Error(), "notebook ./doesnt_exist.py not found")
@@ -470,7 +470,7 @@ func TestJobFileDoesNotExistError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "fake.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "fake.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.EqualError(t, diags.Error(), "file ./doesnt_exist.py not found")
@@ -501,7 +501,7 @@ func TestPipelineNotebookDoesNotExistError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "fake.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "fake.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.EqualError(t, diags.Error(), "notebook ./doesnt_exist.py not found")
@@ -532,7 +532,7 @@ func TestPipelineFileDoesNotExistError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "fake.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "fake.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.EqualError(t, diags.Error(), "file ./doesnt_exist.py not found")
@@ -567,7 +567,7 @@ func TestJobSparkPythonTaskWithNotebookSourceError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.ErrorContains(t, diags.Error(), `expected a file for "resources.jobs.job.tasks[0].spark_python_task.python_file" but got a notebook`)
@@ -602,7 +602,7 @@ func TestJobNotebookTaskWithFileSourceError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.ErrorContains(t, diags.Error(), `expected a notebook for "resources.jobs.job.tasks[0].notebook_task.notebook_path" but got a file`)
@@ -637,7 +637,7 @@ func TestPipelineNotebookLibraryWithFileSourceError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.ErrorContains(t, diags.Error(), `expected a notebook for "resources.pipelines.pipeline.libraries[0].notebook.path" but got a file`)
@@ -672,7 +672,7 @@ func TestPipelineFileLibraryWithNotebookSourceError(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	assert.ErrorContains(t, diags.Error(), `expected a file for "resources.pipelines.pipeline.libraries[0].file.path" but got a notebook`)
@@ -710,7 +710,7 @@ func TestTranslatePathJobEnvironments(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "resources.jobs", filepath.Join(dir, "job/resource.yml"))
+	bundletest.SetLocation(b, "resources.jobs", []dyn.Location{{File: filepath.Join(dir, "job/resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, mutator.TranslatePaths())
 	require.NoError(t, diags.Error())
@@ -753,8 +753,8 @@ func TestTranslatePathWithComplexVariables(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "variables", filepath.Join(dir, "variables/variables.yml"))
-	bundletest.SetLocation(b, "resources.jobs", filepath.Join(dir, "job/resource.yml"))
+	bundletest.SetLocation(b, "variables", []dyn.Location{{File: filepath.Join(dir, "variables/variables.yml")}})
+	bundletest.SetLocation(b, "resources.jobs", []dyn.Location{{File: filepath.Join(dir, "job/resource.yml")}})
 
 	ctx := context.Background()
 	// Assign the variables to the dynamic configuration.

bundle/config/resources.go

@@ -19,6 +19,7 @@ type Resources struct {
 	RegisteredModels      map[string]*resources.RegisteredModel      `json:"registered_models,omitempty"`
 	QualityMonitors       map[string]*resources.QualityMonitor       `json:"quality_monitors,omitempty"`
 	Schemas               map[string]*resources.Schema               `json:"schemas,omitempty"`
+	Clusters              map[string]*resources.Cluster              `json:"clusters,omitempty"`
 }
 
 type ConfigResource interface {

bundle/config/resources/clusters.go

@@ -0,0 +1,39 @@
+package resources
+
+import (
+	"context"
+
+	"github.com/databricks/cli/libs/log"
+	"github.com/databricks/databricks-sdk-go"
+	"github.com/databricks/databricks-sdk-go/marshal"
+	"github.com/databricks/databricks-sdk-go/service/compute"
+)
+
+type Cluster struct {
+	ID             string         `json:"id,omitempty" bundle:"readonly"`
+	Permissions    []Permission   `json:"permissions,omitempty"`
+	ModifiedStatus ModifiedStatus `json:"modified_status,omitempty" bundle:"internal"`
+
+	*compute.ClusterSpec
+}
+
+func (s *Cluster) UnmarshalJSON(b []byte) error {
+	return marshal.Unmarshal(b, s)
+}
+
+func (s Cluster) MarshalJSON() ([]byte, error) {
+	return marshal.Marshal(s)
+}
+
+func (s *Cluster) Exists(ctx context.Context, w *databricks.WorkspaceClient, id string) (bool, error) {
+	_, err := w.Clusters.GetByClusterId(ctx, id)
+	if err != nil {
+		log.Debugf(ctx, "cluster %s does not exist", id)
+		return false, err
+	}
+	return true, nil
+}
+
+func (s *Cluster) TerraformResourceName() string {
+	return "databricks_cluster"
+}

bundle/config/root.go

@@ -366,9 +366,9 @@ func (r *Root) MergeTargetOverrides(name string) error {
 		}
 	}
 
-	// Merge `compute_id`. This field must be overwritten if set, not merged.
-	if v := target.Get("compute_id"); v.Kind() != dyn.KindInvalid {
-		root, err = dyn.SetByPath(root, dyn.NewPath(dyn.Key("bundle"), dyn.Key("compute_id")), v)
+	// Merge `cluster_id`. This field must be overwritten if set, not merged.
+	if v := target.Get("cluster_id"); v.Kind() != dyn.KindInvalid {
+		root, err = dyn.SetByPath(root, dyn.NewPath(dyn.Key("bundle"), dyn.Key("cluster_id")), v)
 		if err != nil {
 			return err
 		}
@@ -409,15 +409,51 @@ func (r *Root) MergeTargetOverrides(name string) error {
 var variableKeywords = []string{"default", "lookup"}
 
 // isFullVariableOverrideDef checks if the given value is a full syntax varaible override.
-// A full syntax variable override is a map with only one of the following
-// keys: "default", "lookup".
+// A full syntax variable override is a map with either 1 of 2 keys.
+// If it's 2 keys, the keys should be "default" and "type".
+// If it's 1 key, the key should be one of the following keys: "default", "lookup".
 func isFullVariableOverrideDef(v dyn.Value) bool {
 	mv, ok := v.AsMap()
 	if !ok {
 		return false
 	}
 
-	if mv.Len() != 1 {
+	// If the map has more than 3 keys, it is not a full variable override.
+	if mv.Len() > 3 {
+		return false
+	}
+
+	// If the map has 3 keys, they should be "description", "type" and "default" or "lookup"
+	if mv.Len() == 3 {
+		if _, ok := mv.GetByString("type"); ok {
+			if _, ok := mv.GetByString("description"); ok {
+				if _, ok := mv.GetByString("default"); ok {
+					return true
+				}
+			}
+		}
+
+		return false
+	}
+
+	// If the map has 2 keys, one of them should be "default" or "lookup" and the other is "type" or "description"
+	if mv.Len() == 2 {
+		if _, ok := mv.GetByString("type"); ok {
+			if _, ok := mv.GetByString("default"); ok {
+				return true
+			}
+		}
+
+		if _, ok := mv.GetByString("description"); ok {
+			if _, ok := mv.GetByString("default"); ok {
+				return true
+			}
+
+			if _, ok := mv.GetByString("lookup"); ok {
+				return true
+			}
+		}
+
 		return false
 	}
 

bundle/config/root_test.go

@@ -6,6 +6,7 @@ import (
 	"testing"
 
 	"github.com/databricks/cli/bundle/config/variable"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -139,7 +140,7 @@ func TestRootMergeTargetOverridesWithVariables(t *testing.T) {
 		},
 		Targets: map[string]*Target{
 			"development": {
-				Variables: map[string]*variable.Variable{
+				Variables: map[string]*variable.TargetVariable{
 					"foo": {
 						Default:     "bar",
 						Description: "wrong",
@@ -169,3 +170,87 @@ func TestRootMergeTargetOverridesWithVariables(t *testing.T) {
 	assert.Equal(t, "complex var", root.Variables["complex"].Description)
 
 }
+
+func TestIsFullVariableOverrideDef(t *testing.T) {
+	testCases := []struct {
+		value    dyn.Value
+		expected bool
+	}{
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type":        dyn.V("string"),
+				"default":     dyn.V("foo"),
+				"description": dyn.V("foo var"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type":        dyn.V("string"),
+				"lookup":      dyn.V("foo"),
+				"description": dyn.V("foo var"),
+			}),
+			expected: false,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type":    dyn.V("string"),
+				"default": dyn.V("foo"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type":   dyn.V("string"),
+				"lookup": dyn.V("foo"),
+			}),
+			expected: false,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"description": dyn.V("string"),
+				"default":     dyn.V("foo"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"description": dyn.V("string"),
+				"lookup":      dyn.V("foo"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"default": dyn.V("foo"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"lookup": dyn.V("foo"),
+			}),
+			expected: true,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type": dyn.V("string"),
+			}),
+			expected: false,
+		},
+		{
+			value: dyn.V(map[string]dyn.Value{
+				"type":        dyn.V("string"),
+				"default":     dyn.V("foo"),
+				"description": dyn.V("foo var"),
+				"lookup":      dyn.V("foo"),
+			}),
+			expected: false,
+		},
+	}
+
+	for i, tc := range testCases {
+		assert.Equal(t, tc.expected, isFullVariableOverrideDef(tc.value), "test case %d", i)
+	}
+
+}

bundle/config/target.go

@@ -24,8 +24,11 @@ type Target struct {
 	// name prefix of deployed resources.
 	Presets Presets `json:"presets,omitempty"`
 
-	// Overrides the compute used for jobs and other supported assets.
-	ComputeID string `json:"compute_id,omitempty"`
+	// DEPRECATED: Overrides the compute used for jobs and other supported assets.
+	ComputeId string `json:"compute_id,omitempty"`
+
+	// Overrides the cluster used for jobs and other supported assets.
+	ClusterId string `json:"cluster_id,omitempty"`
 
 	Bundle *Bundle `json:"bundle,omitempty"`
 
@@ -38,7 +41,26 @@ type Target struct {
 	// Override default values or lookup name for defined variables
 	// Does not permit defining new variables or redefining existing ones
 	// in the scope of an target
-	Variables map[string]*variable.Variable `json:"variables,omitempty"`
+	//
+	// There are two valid ways to define a variable override in a target:
+	// 1. Direct value override. We normalize this to the variable.Variable
+	//    struct format when loading the configuration YAML:
+	//
+	//   variables:
+	//     foo: "value"
+	//
+	// 2. Override matching the variable.Variable struct.
+	//
+	//   variables:
+	//     foo:
+	//       default: "value"
+	//
+	// OR
+	//
+	//   variables:
+	//     foo:
+	//       lookup: "resource_name"
+	Variables map[string]*variable.TargetVariable `json:"variables,omitempty"`
 
 	Git Git `json:"git,omitempty"`
 

bundle/config/validate/job_task_cluster_spec.go

@@ -0,0 +1,161 @@
+package validate
+
+import (
+	"context"
+	"fmt"
+	"strings"
+
+	"github.com/databricks/cli/bundle"
+	"github.com/databricks/cli/libs/diag"
+	"github.com/databricks/cli/libs/dyn"
+	"github.com/databricks/databricks-sdk-go/service/jobs"
+)
+
+// JobTaskClusterSpec validates that job tasks have cluster spec defined
+// if task requires a cluster
+func JobTaskClusterSpec() bundle.ReadOnlyMutator {
+	return &jobTaskClusterSpec{}
+}
+
+type jobTaskClusterSpec struct {
+}
+
+func (v *jobTaskClusterSpec) Name() string {
+	return "validate:job_task_cluster_spec"
+}
+
+func (v *jobTaskClusterSpec) Apply(ctx context.Context, rb bundle.ReadOnlyBundle) diag.Diagnostics {
+	diags := diag.Diagnostics{}
+
+	jobsPath := dyn.NewPath(dyn.Key("resources"), dyn.Key("jobs"))
+
+	for resourceName, job := range rb.Config().Resources.Jobs {
+		resourcePath := jobsPath.Append(dyn.Key(resourceName))
+
+		for taskIndex, task := range job.Tasks {
+			taskPath := resourcePath.Append(dyn.Key("tasks"), dyn.Index(taskIndex))
+
+			diags = diags.Extend(validateJobTask(rb, task, taskPath))
+		}
+	}
+
+	return diags
+}
+
+func validateJobTask(rb bundle.ReadOnlyBundle, task jobs.Task, taskPath dyn.Path) diag.Diagnostics {
+	diags := diag.Diagnostics{}
+
+	var specified []string
+	var unspecified []string
+
+	if task.JobClusterKey != "" {
+		specified = append(specified, "job_cluster_key")
+	} else {
+		unspecified = append(unspecified, "job_cluster_key")
+	}
+
+	if task.EnvironmentKey != "" {
+		specified = append(specified, "environment_key")
+	} else {
+		unspecified = append(unspecified, "environment_key")
+	}
+
+	if task.ExistingClusterId != "" {
+		specified = append(specified, "existing_cluster_id")
+	} else {
+		unspecified = append(unspecified, "existing_cluster_id")
+	}
+
+	if task.NewCluster != nil {
+		specified = append(specified, "new_cluster")
+	} else {
+		unspecified = append(unspecified, "new_cluster")
+	}
+
+	if task.ForEachTask != nil {
+		forEachTaskPath := taskPath.Append(dyn.Key("for_each_task"), dyn.Key("task"))
+
+		diags = diags.Extend(validateJobTask(rb, task.ForEachTask.Task, forEachTaskPath))
+	}
+
+	if isComputeTask(task) && len(specified) == 0 {
+		if task.NotebookTask != nil {
+			// notebook tasks without cluster spec will use notebook environment
+		} else {
+			// path might be not very helpful, adding user-specified task key clarifies the context
+			detail := fmt.Sprintf(
+				"Task %q requires a cluster or an environment to run.\nSpecify one of the following fields: %s.",
+				task.TaskKey,
+				strings.Join(unspecified, ", "),
+			)
+
+			diags = diags.Append(diag.Diagnostic{
+				Severity:  diag.Error,
+				Summary:   "Missing required cluster or environment settings",
+				Detail:    detail,
+				Locations: rb.Config().GetLocations(taskPath.String()),
+				Paths:     []dyn.Path{taskPath},
+			})
+		}
+	}
+
+	return diags
+}
+
+// isComputeTask returns true if the task runs on a cluster or serverless GC
+func isComputeTask(task jobs.Task) bool {
+	if task.NotebookTask != nil {
+		// if warehouse_id is set, it's SQL notebook that doesn't need cluster or serverless GC
+		if task.NotebookTask.WarehouseId != "" {
+			return false
+		} else {
+			// task settings don't require specifying a cluster/serverless GC, but task itself can run on one
+			// we handle that case separately in validateJobTask
+			return true
+		}
+	}
+
+	if task.PythonWheelTask != nil {
+		return true
+	}
+
+	if task.DbtTask != nil {
+		return true
+	}
+
+	if task.SparkJarTask != nil {
+		return true
+	}
+
+	if task.SparkSubmitTask != nil {
+		return true
+	}
+
+	if task.SparkPythonTask != nil {
+		return true
+	}
+
+	if task.SqlTask != nil {
+		return false
+	}
+
+	if task.PipelineTask != nil {
+		// while pipelines use clusters, pipeline tasks don't, they only trigger pipelines
+		return false
+	}
+
+	if task.RunJobTask != nil {
+		return false
+	}
+
+	if task.ConditionTask != nil {
+		return false
+	}
+
+	// for each task doesn't use clusters, underlying task(s) can though
+	if task.ForEachTask != nil {
+		return false
+	}
+
+	return false
+}

bundle/config/validate/job_task_cluster_spec_test.go

@@ -0,0 +1,203 @@
+package validate
+
+import (
+	"context"
+	"testing"
+
+	"github.com/databricks/cli/bundle"
+	"github.com/databricks/cli/bundle/config"
+	"github.com/databricks/cli/bundle/config/resources"
+	"github.com/databricks/databricks-sdk-go/service/compute"
+	"github.com/databricks/databricks-sdk-go/service/jobs"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestJobTaskClusterSpec(t *testing.T) {
+	expectedSummary := "Missing required cluster or environment settings"
+
+	type testCase struct {
+		name         string
+		task         jobs.Task
+		errorPath    string
+		errorDetail  string
+		errorSummary string
+	}
+
+	testCases := []testCase{
+		{
+			name: "valid notebook task",
+			task: jobs.Task{
+				// while a cluster is needed, it will use notebook environment to create one
+				NotebookTask: &jobs.NotebookTask{},
+			},
+		},
+		{
+			name: "valid notebook task (job_cluster_key)",
+			task: jobs.Task{
+				JobClusterKey: "cluster1",
+				NotebookTask:  &jobs.NotebookTask{},
+			},
+		},
+		{
+			name: "valid notebook task (new_cluster)",
+			task: jobs.Task{
+				NewCluster:   &compute.ClusterSpec{},
+				NotebookTask: &jobs.NotebookTask{},
+			},
+		},
+		{
+			name: "valid notebook task (existing_cluster_id)",
+			task: jobs.Task{
+				ExistingClusterId: "cluster1",
+				NotebookTask:      &jobs.NotebookTask{},
+			},
+		},
+		{
+			name: "valid SQL notebook task",
+			task: jobs.Task{
+				NotebookTask: &jobs.NotebookTask{
+					WarehouseId: "warehouse1",
+				},
+			},
+		},
+		{
+			name: "valid python wheel task",
+			task: jobs.Task{
+				JobClusterKey:   "cluster1",
+				PythonWheelTask: &jobs.PythonWheelTask{},
+			},
+		},
+		{
+			name: "valid python wheel task (environment_key)",
+			task: jobs.Task{
+				EnvironmentKey:  "environment1",
+				PythonWheelTask: &jobs.PythonWheelTask{},
+			},
+		},
+		{
+			name: "valid dbt task",
+			task: jobs.Task{
+				JobClusterKey: "cluster1",
+				DbtTask:       &jobs.DbtTask{},
+			},
+		},
+		{
+			name: "valid spark jar task",
+			task: jobs.Task{
+				JobClusterKey: "cluster1",
+				SparkJarTask:  &jobs.SparkJarTask{},
+			},
+		},
+		{
+			name: "valid spark submit",
+			task: jobs.Task{
+				NewCluster:      &compute.ClusterSpec{},
+				SparkSubmitTask: &jobs.SparkSubmitTask{},
+			},
+		},
+		{
+			name: "valid spark python task",
+			task: jobs.Task{
+				JobClusterKey:   "cluster1",
+				SparkPythonTask: &jobs.SparkPythonTask{},
+			},
+		},
+		{
+			name: "valid SQL task",
+			task: jobs.Task{
+				SqlTask: &jobs.SqlTask{},
+			},
+		},
+		{
+			name: "valid pipeline task",
+			task: jobs.Task{
+				PipelineTask: &jobs.PipelineTask{},
+			},
+		},
+		{
+			name: "valid run job task",
+			task: jobs.Task{
+				RunJobTask: &jobs.RunJobTask{},
+			},
+		},
+		{
+			name: "valid condition task",
+			task: jobs.Task{
+				ConditionTask: &jobs.ConditionTask{},
+			},
+		},
+		{
+			name: "valid for each task",
+			task: jobs.Task{
+				ForEachTask: &jobs.ForEachTask{
+					Task: jobs.Task{
+						JobClusterKey: "cluster1",
+						NotebookTask:  &jobs.NotebookTask{},
+					},
+				},
+			},
+		},
+		{
+			name: "invalid python wheel task",
+			task: jobs.Task{
+				PythonWheelTask: &jobs.PythonWheelTask{},
+				TaskKey:         "my_task",
+			},
+			errorPath: "resources.jobs.job1.tasks[0]",
+			errorDetail: `Task "my_task" requires a cluster or an environment to run.
+Specify one of the following fields: job_cluster_key, environment_key, existing_cluster_id, new_cluster.`,
+			errorSummary: expectedSummary,
+		},
+		{
+			name: "invalid for each task",
+			task: jobs.Task{
+				ForEachTask: &jobs.ForEachTask{
+					Task: jobs.Task{
+						PythonWheelTask: &jobs.PythonWheelTask{},
+						TaskKey:         "my_task",
+					},
+				},
+			},
+			errorPath: "resources.jobs.job1.tasks[0].for_each_task.task",
+			errorDetail: `Task "my_task" requires a cluster or an environment to run.
+Specify one of the following fields: job_cluster_key, environment_key, existing_cluster_id, new_cluster.`,
+			errorSummary: expectedSummary,
+		},
+	}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			job := &resources.Job{
+				JobSettings: &jobs.JobSettings{
+					Tasks: []jobs.Task{tc.task},
+				},
+			}
+
+			b := createBundle(map[string]*resources.Job{"job1": job})
+			diags := bundle.ApplyReadOnly(context.Background(), bundle.ReadOnly(b), JobTaskClusterSpec())
+
+			if tc.errorPath != "" || tc.errorDetail != "" || tc.errorSummary != "" {
+				assert.Len(t, diags, 1)
+				assert.Len(t, diags[0].Paths, 1)
+
+				diag := diags[0]
+
+				assert.Equal(t, tc.errorPath, diag.Paths[0].String())
+				assert.Equal(t, tc.errorSummary, diag.Summary)
+				assert.Equal(t, tc.errorDetail, diag.Detail)
+			} else {
+				assert.ElementsMatch(t, []string{}, diags)
+			}
+		})
+	}
+}
+
+func createBundle(jobs map[string]*resources.Job) *bundle.Bundle {
+	return &bundle.Bundle{
+		Config: config.Root{
+			Resources: config.Resources{
+				Jobs: jobs,
+			},
+		},
+	}
+}

bundle/config/validate/validate.go

@@ -34,6 +34,7 @@ func (v *validate) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics
 		JobClusterKeyDefined(),
 		FilesToSync(),
 		ValidateSyncPatterns(),
+		JobTaskClusterSpec(),
 	))
 }
 

bundle/config/variable/variable.go

@@ -16,6 +16,11 @@ const (
 	VariableTypeComplex VariableType = "complex"
 )
 
+// We alias it here to override the JSON schema associated with a variable value
+// in a target override. This is because we allow for directly specifying the value
+// in addition to the variable.Variable struct format in a target override.
+type TargetVariable Variable
+
 // An input variable for the bundle config
 type Variable struct {
 	// A type of the variable. This is used to validate the value of the variable

bundle/deploy/files/upload.go

@@ -8,9 +8,12 @@ import (
 	"github.com/databricks/cli/libs/cmdio"
 	"github.com/databricks/cli/libs/diag"
 	"github.com/databricks/cli/libs/log"
+	"github.com/databricks/cli/libs/sync"
 )
 
-type upload struct{}
+type upload struct {
+	outputHandler sync.OutputHandler
+}
 
 func (m *upload) Name() string {
 	return "files.Upload"
@@ -18,11 +21,18 @@ func (m *upload) Name() string {
 
 func (m *upload) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics {
 	cmdio.LogString(ctx, fmt.Sprintf("Uploading bundle files to %s...", b.Config.Workspace.FilePath))
-	sync, err := GetSync(ctx, bundle.ReadOnly(b))
+	opts, err := GetSyncOptions(ctx, bundle.ReadOnly(b))
 	if err != nil {
 		return diag.FromErr(err)
 	}
 
+	opts.OutputHandler = m.outputHandler
+	sync, err := sync.New(ctx, *opts)
+	if err != nil {
+		return diag.FromErr(err)
+	}
+	defer sync.Close()
+
 	b.Files, err = sync.RunOnce(ctx)
 	if err != nil {
 		return diag.FromErr(err)
@@ -32,6 +42,6 @@ func (m *upload) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics {
 	return nil
 }
 
-func Upload() bundle.Mutator {
-	return &upload{}
+func Upload(outputHandler sync.OutputHandler) bundle.Mutator {
+	return &upload{outputHandler}
 }

bundle/deploy/metadata/compute_test.go

@@ -9,6 +9,7 @@ import (
 	"github.com/databricks/cli/bundle/config/resources"
 	"github.com/databricks/cli/bundle/internal/bundletest"
 	"github.com/databricks/cli/bundle/metadata"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/databricks/databricks-sdk-go/service/jobs"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
@@ -55,9 +56,9 @@ func TestComputeMetadataMutator(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, "resources.jobs.my-job-1", "a/b/c")
-	bundletest.SetLocation(b, "resources.jobs.my-job-2", "d/e/f")
-	bundletest.SetLocation(b, "resources.pipelines.my-pipeline", "abc")
+	bundletest.SetLocation(b, "resources.jobs.my-job-1", []dyn.Location{{File: "a/b/c"}})
+	bundletest.SetLocation(b, "resources.jobs.my-job-2", []dyn.Location{{File: "d/e/f"}})
+	bundletest.SetLocation(b, "resources.pipelines.my-pipeline", []dyn.Location{{File: "abc"}})
 
 	expectedMetadata := metadata.Metadata{
 		Version: metadata.Version,

bundle/deploy/terraform/convert.go

@@ -231,6 +231,13 @@ func BundleToTerraform(config *config.Root) *schema.Root {
 		tfroot.Resource.QualityMonitor[k] = &dst
 	}
 
+	for k, src := range config.Resources.Clusters {
+		noResources = false
+		var dst schema.ResourceCluster
+		conv(src, &dst)
+		tfroot.Resource.Cluster[k] = &dst
+	}
+
 	// We explicitly set "resource" to nil to omit it from a JSON encoding.
 	// This is required because the terraform CLI requires >= 1 resources defined
 	// if the "resource" property is used in a .tf.json file.
@@ -394,6 +401,16 @@ func TerraformToBundle(state *resourcesState, config *config.Root) error {
 				}
 				cur.ID = instance.Attributes.ID
 				config.Resources.Schemas[resource.Name] = cur
+			case "databricks_cluster":
+				if config.Resources.Clusters == nil {
+					config.Resources.Clusters = make(map[string]*resources.Cluster)
+				}
+				cur := config.Resources.Clusters[resource.Name]
+				if cur == nil {
+					cur = &resources.Cluster{ModifiedStatus: resources.ModifiedStatusDeleted}
+				}
+				cur.ID = instance.Attributes.ID
+				config.Resources.Clusters[resource.Name] = cur
 			case "databricks_permissions":
 			case "databricks_grants":
 				// Ignore; no need to pull these back into the configuration.
@@ -443,6 +460,11 @@ func TerraformToBundle(state *resourcesState, config *config.Root) error {
 			src.ModifiedStatus = resources.ModifiedStatusCreated
 		}
 	}
+	for _, src := range config.Resources.Clusters {
+		if src.ModifiedStatus == "" && src.ID == "" {
+			src.ModifiedStatus = resources.ModifiedStatusCreated
+		}
+	}
 
 	return nil
 }

bundle/deploy/terraform/convert_test.go

@@ -663,6 +663,14 @@ func TestTerraformToBundleEmptyLocalResources(t *testing.T) {
 					{Attributes: stateInstanceAttributes{ID: "1"}},
 				},
 			},
+			{
+				Type: "databricks_cluster",
+				Mode: "managed",
+				Name: "test_cluster",
+				Instances: []stateResourceInstance{
+					{Attributes: stateInstanceAttributes{ID: "1"}},
+				},
+			},
 		},
 	}
 	err := TerraformToBundle(&tfState, &config)
@@ -692,6 +700,9 @@ func TestTerraformToBundleEmptyLocalResources(t *testing.T) {
 	assert.Equal(t, "1", config.Resources.Schemas["test_schema"].ID)
 	assert.Equal(t, resources.ModifiedStatusDeleted, config.Resources.Schemas["test_schema"].ModifiedStatus)
 
+	assert.Equal(t, "1", config.Resources.Clusters["test_cluster"].ID)
+	assert.Equal(t, resources.ModifiedStatusDeleted, config.Resources.Clusters["test_cluster"].ModifiedStatus)
+
 	AssertFullResourceCoverage(t, &config)
 }
 
@@ -754,6 +765,13 @@ func TestTerraformToBundleEmptyRemoteResources(t *testing.T) {
 					},
 				},
 			},
+			Clusters: map[string]*resources.Cluster{
+				"test_cluster": {
+					ClusterSpec: &compute.ClusterSpec{
+						ClusterName: "test_cluster",
+					},
+				},
+			},
 		},
 	}
 	var tfState = resourcesState{
@@ -786,6 +804,9 @@ func TestTerraformToBundleEmptyRemoteResources(t *testing.T) {
 	assert.Equal(t, "", config.Resources.Schemas["test_schema"].ID)
 	assert.Equal(t, resources.ModifiedStatusCreated, config.Resources.Schemas["test_schema"].ModifiedStatus)
 
+	assert.Equal(t, "", config.Resources.Clusters["test_cluster"].ID)
+	assert.Equal(t, resources.ModifiedStatusCreated, config.Resources.Clusters["test_cluster"].ModifiedStatus)
+
 	AssertFullResourceCoverage(t, &config)
 }
 
@@ -888,6 +909,18 @@ func TestTerraformToBundleModifiedResources(t *testing.T) {
 					},
 				},
 			},
+			Clusters: map[string]*resources.Cluster{
+				"test_cluster": {
+					ClusterSpec: &compute.ClusterSpec{
+						ClusterName: "test_cluster",
+					},
+				},
+				"test_cluster_new": {
+					ClusterSpec: &compute.ClusterSpec{
+						ClusterName: "test_cluster_new",
+					},
+				},
+			},
 		},
 	}
 	var tfState = resourcesState{
@@ -1020,6 +1053,22 @@ func TestTerraformToBundleModifiedResources(t *testing.T) {
 					{Attributes: stateInstanceAttributes{ID: "2"}},
 				},
 			},
+			{
+				Type: "databricks_cluster",
+				Mode: "managed",
+				Name: "test_cluster",
+				Instances: []stateResourceInstance{
+					{Attributes: stateInstanceAttributes{ID: "1"}},
+				},
+			},
+			{
+				Type: "databricks_cluster",
+				Mode: "managed",
+				Name: "test_cluster_old",
+				Instances: []stateResourceInstance{
+					{Attributes: stateInstanceAttributes{ID: "2"}},
+				},
+			},
 		},
 	}
 	err := TerraformToBundle(&tfState, &config)
@@ -1081,6 +1130,13 @@ func TestTerraformToBundleModifiedResources(t *testing.T) {
 	assert.Equal(t, "", config.Resources.Schemas["test_schema_new"].ID)
 	assert.Equal(t, resources.ModifiedStatusCreated, config.Resources.Schemas["test_schema_new"].ModifiedStatus)
 
+	assert.Equal(t, "1", config.Resources.Clusters["test_cluster"].ID)
+	assert.Equal(t, "", config.Resources.Clusters["test_cluster"].ModifiedStatus)
+	assert.Equal(t, "2", config.Resources.Clusters["test_cluster_old"].ID)
+	assert.Equal(t, resources.ModifiedStatusDeleted, config.Resources.Clusters["test_cluster_old"].ModifiedStatus)
+	assert.Equal(t, "", config.Resources.Clusters["test_cluster_new"].ID)
+	assert.Equal(t, resources.ModifiedStatusCreated, config.Resources.Clusters["test_cluster_new"].ModifiedStatus)
+
 	AssertFullResourceCoverage(t, &config)
 }
 

bundle/deploy/terraform/interpolate.go

@@ -58,6 +58,8 @@ func (m *interpolateMutator) Apply(ctx context.Context, b *bundle.Bundle) diag.D
 				path = dyn.NewPath(dyn.Key("databricks_quality_monitor")).Append(path[2:]...)
 			case dyn.Key("schemas"):
 				path = dyn.NewPath(dyn.Key("databricks_schema")).Append(path[2:]...)
+			case dyn.Key("clusters"):
+				path = dyn.NewPath(dyn.Key("databricks_cluster")).Append(path[2:]...)
 			default:
 				// Trigger "key not found" for unknown resource types.
 				return dyn.GetByPath(root, path)

bundle/deploy/terraform/interpolate_test.go

@@ -31,6 +31,7 @@ func TestInterpolate(t *testing.T) {
 								"other_model_serving":    "${resources.model_serving_endpoints.other_model_serving.id}",
 								"other_registered_model": "${resources.registered_models.other_registered_model.id}",
 								"other_schema":           "${resources.schemas.other_schema.id}",
+								"other_cluster":          "${resources.clusters.other_cluster.id}",
 							},
 							Tasks: []jobs.Task{
 								{
@@ -67,6 +68,7 @@ func TestInterpolate(t *testing.T) {
 	assert.Equal(t, "${databricks_model_serving.other_model_serving.id}", j.Tags["other_model_serving"])
 	assert.Equal(t, "${databricks_registered_model.other_registered_model.id}", j.Tags["other_registered_model"])
 	assert.Equal(t, "${databricks_schema.other_schema.id}", j.Tags["other_schema"])
+	assert.Equal(t, "${databricks_cluster.other_cluster.id}", j.Tags["other_cluster"])
 
 	m := b.Config.Resources.Models["my_model"]
 	assert.Equal(t, "my_model", m.Model.Name)

bundle/deploy/terraform/tfdyn/convert_cluster.go

@@ -0,0 +1,52 @@
+package tfdyn
+
+import (
+	"context"
+	"fmt"
+
+	"github.com/databricks/cli/bundle/internal/tf/schema"
+	"github.com/databricks/cli/libs/dyn"
+	"github.com/databricks/cli/libs/dyn/convert"
+	"github.com/databricks/cli/libs/log"
+	"github.com/databricks/databricks-sdk-go/service/compute"
+)
+
+func convertClusterResource(ctx context.Context, vin dyn.Value) (dyn.Value, error) {
+	// Normalize the output value to the target schema.
+	vout, diags := convert.Normalize(compute.ClusterSpec{}, vin)
+	for _, diag := range diags {
+		log.Debugf(ctx, "cluster normalization diagnostic: %s", diag.Summary)
+	}
+
+	return vout, nil
+}
+
+type clusterConverter struct{}
+
+func (clusterConverter) Convert(ctx context.Context, key string, vin dyn.Value, out *schema.Resources) error {
+	vout, err := convertClusterResource(ctx, vin)
+	if err != nil {
+		return err
+	}
+
+	// We always set no_wait as it allows DABs not to wait for cluster to be started.
+	vout, err = dyn.Set(vout, "no_wait", dyn.V(true))
+	if err != nil {
+		return err
+	}
+
+	// Add the converted resource to the output.
+	out.Cluster[key] = vout.AsAny()
+
+	// Configure permissions for this resource.
+	if permissions := convertPermissionsResource(ctx, vin); permissions != nil {
+		permissions.JobId = fmt.Sprintf("${databricks_cluster.%s.id}", key)
+		out.Permissions["cluster_"+key] = permissions
+	}
+
+	return nil
+}
+
+func init() {
+	registerConverter("clusters", clusterConverter{})
+}

bundle/deploy/terraform/tfdyn/convert_cluster_test.go

@@ -0,0 +1,97 @@
+package tfdyn
+
+import (
+	"context"
+	"testing"
+
+	"github.com/databricks/cli/bundle/config/resources"
+	"github.com/databricks/cli/bundle/internal/tf/schema"
+	"github.com/databricks/cli/libs/dyn"
+	"github.com/databricks/cli/libs/dyn/convert"
+	"github.com/databricks/databricks-sdk-go/service/compute"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestConvertCluster(t *testing.T) {
+	var src = resources.Cluster{
+		ClusterSpec: &compute.ClusterSpec{
+			NumWorkers:   3,
+			SparkVersion: "13.3.x-scala2.12",
+			ClusterName:  "cluster",
+			SparkConf: map[string]string{
+				"spark.executor.memory": "2g",
+			},
+			AwsAttributes: &compute.AwsAttributes{
+				Availability: "ON_DEMAND",
+			},
+			AzureAttributes: &compute.AzureAttributes{
+				Availability: "SPOT",
+			},
+			DataSecurityMode: "USER_ISOLATION",
+			NodeTypeId:       "m5.xlarge",
+			Autoscale: &compute.AutoScale{
+				MinWorkers: 1,
+				MaxWorkers: 10,
+			},
+		},
+
+		Permissions: []resources.Permission{
+			{
+				Level:    "CAN_RUN",
+				UserName: "jack@gmail.com",
+			},
+			{
+				Level:                "CAN_MANAGE",
+				ServicePrincipalName: "sp",
+			},
+		},
+	}
+
+	vin, err := convert.FromTyped(src, dyn.NilValue)
+	require.NoError(t, err)
+
+	ctx := context.Background()
+	out := schema.NewResources()
+	err = clusterConverter{}.Convert(ctx, "my_cluster", vin, out)
+	require.NoError(t, err)
+
+	cluster := out.Cluster["my_cluster"]
+	assert.Equal(t, map[string]any{
+		"num_workers":   int64(3),
+		"spark_version": "13.3.x-scala2.12",
+		"cluster_name":  "cluster",
+		"spark_conf": map[string]any{
+			"spark.executor.memory": "2g",
+		},
+		"aws_attributes": map[string]any{
+			"availability": "ON_DEMAND",
+		},
+		"azure_attributes": map[string]any{
+			"availability": "SPOT",
+		},
+		"data_security_mode": "USER_ISOLATION",
+		"no_wait":            true,
+		"node_type_id":       "m5.xlarge",
+		"autoscale": map[string]any{
+			"min_workers": int64(1),
+			"max_workers": int64(10),
+		},
+	}, cluster)
+
+	// Assert equality on the permissions
+	assert.Equal(t, &schema.ResourcePermissions{
+		JobId: "${databricks_cluster.my_cluster.id}",
+		AccessControl: []schema.ResourcePermissionsAccessControl{
+			{
+				PermissionLevel: "CAN_RUN",
+				UserName:        "jack@gmail.com",
+			},
+			{
+				PermissionLevel:      "CAN_MANAGE",
+				ServicePrincipalName: "sp",
+			},
+		},
+	}, out.Permissions["cluster_my_cluster"])
+
+}

bundle/internal/bundle/schema/main.go

@@ -1,42 +0,0 @@
-package main
-
-import (
-	"encoding/json"
-	"fmt"
-	"log"
-	"os"
-
-	"github.com/databricks/cli/bundle/schema"
-)
-
-func main() {
-	if len(os.Args) != 2 {
-		fmt.Println("Usage: go run main.go <output-file>")
-		os.Exit(1)
-	}
-
-	// Output file, to write the generated schema descriptions to.
-	outputFile := os.Args[1]
-
-	// Input file, the databricks openapi spec.
-	inputFile := os.Getenv("DATABRICKS_OPENAPI_SPEC")
-	if inputFile == "" {
-		log.Fatal("DATABRICKS_OPENAPI_SPEC environment variable not set")
-	}
-
-	// Generate the schema descriptions.
-	docs, err := schema.UpdateBundleDescriptions(inputFile)
-	if err != nil {
-		log.Fatal(err)
-	}
-	result, err := json.MarshalIndent(docs, "", "  ")
-	if err != nil {
-		log.Fatal(err)
-	}
-
-	// Write the schema descriptions to the output file.
-	err = os.WriteFile(outputFile, result, 0644)
-	if err != nil {
-		log.Fatal(err)
-	}
-}

bundle/internal/bundletest/location.go

@@ -8,15 +8,13 @@ import (
 // SetLocation sets the location of all values in the bundle to the given path.
 // This is useful for testing where we need to associate configuration
 // with the path it is loaded from.
-func SetLocation(b *bundle.Bundle, prefix string, filePath string) {
+func SetLocation(b *bundle.Bundle, prefix string, locations []dyn.Location) {
 	start := dyn.MustPathFromString(prefix)
 	b.Config.Mutate(func(root dyn.Value) (dyn.Value, error) {
 		return dyn.Walk(root, func(p dyn.Path, v dyn.Value) (dyn.Value, error) {
 			// If the path has the given prefix, set the location.
 			if p.HasPrefix(start) {
-				return v.WithLocations([]dyn.Location{{
-					File: filePath,
-				}}), nil
+				return v.WithLocations(locations), nil
 			}
 
 			// The path is not nested under the given prefix.

bundle/internal/schema/main.go

@@ -0,0 +1,109 @@
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"log"
+	"os"
+	"reflect"
+
+	"github.com/databricks/cli/bundle/config"
+	"github.com/databricks/cli/bundle/config/variable"
+	"github.com/databricks/cli/libs/jsonschema"
+)
+
+func interpolationPattern(s string) string {
+	return fmt.Sprintf(`\$\{(%s(\.[a-zA-Z]+([-_]?[a-zA-Z0-9]+)*(\[[0-9]+\])*)+)\}`, s)
+}
+
+func addInterpolationPatterns(typ reflect.Type, s jsonschema.Schema) jsonschema.Schema {
+	if typ == reflect.TypeOf(config.Root{}) || typ == reflect.TypeOf(variable.Variable{}) {
+		return s
+	}
+
+	// The variables block in a target override allows for directly specifying
+	// the value of the variable.
+	if typ == reflect.TypeOf(variable.TargetVariable{}) {
+		return jsonschema.Schema{
+			AnyOf: []jsonschema.Schema{
+				// We keep the original schema so that autocomplete suggestions
+				// continue to work.
+				s,
+				// All values are valid for a variable value, be it primitive types
+				// like string/bool or complex ones like objects/arrays. Thus we override
+				// the schema to allow all valid JSON values.
+				{},
+			},
+		}
+	}
+
+	switch s.Type {
+	case jsonschema.ArrayType, jsonschema.ObjectType:
+		// arrays and objects can have complex variable values specified.
+		return jsonschema.Schema{
+			AnyOf: []jsonschema.Schema{
+				s,
+				{
+					Type:    jsonschema.StringType,
+					Pattern: interpolationPattern("var"),
+				}},
+		}
+	case jsonschema.IntegerType, jsonschema.NumberType, jsonschema.BooleanType:
+		// primitives can have variable values, or references like ${bundle.xyz}
+		// or ${workspace.xyz}
+		return jsonschema.Schema{
+			AnyOf: []jsonschema.Schema{
+				s,
+				{Type: jsonschema.StringType, Pattern: interpolationPattern("resources")},
+				{Type: jsonschema.StringType, Pattern: interpolationPattern("bundle")},
+				{Type: jsonschema.StringType, Pattern: interpolationPattern("workspace")},
+				{Type: jsonschema.StringType, Pattern: interpolationPattern("artifacts")},
+				{Type: jsonschema.StringType, Pattern: interpolationPattern("var")},
+			},
+		}
+	default:
+		return s
+	}
+}
+
+func main() {
+	if len(os.Args) != 2 {
+		fmt.Println("Usage: go run main.go <output-file>")
+		os.Exit(1)
+	}
+
+	// Output file, where the generated JSON schema will be written to.
+	outputFile := os.Args[1]
+
+	// Input file, the databricks openapi spec.
+	inputFile := os.Getenv("DATABRICKS_OPENAPI_SPEC")
+	if inputFile == "" {
+		log.Fatal("DATABRICKS_OPENAPI_SPEC environment variable not set")
+	}
+
+	p, err := newParser(inputFile)
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	// Generate the JSON schema from the bundle Go struct.
+	s, err := jsonschema.FromType(reflect.TypeOf(config.Root{}), []func(reflect.Type, jsonschema.Schema) jsonschema.Schema{
+		p.addDescriptions,
+		p.addEnums,
+		addInterpolationPatterns,
+	})
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	b, err := json.MarshalIndent(s, "", "  ")
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	// Write the schema descriptions to the output file.
+	err = os.WriteFile(outputFile, b, 0644)
+	if err != nil {
+		log.Fatal(err)
+	}
+}

bundle/internal/schema/parser.go

@@ -0,0 +1,123 @@
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"os"
+	"path"
+	"reflect"
+	"strings"
+
+	"github.com/databricks/cli/libs/jsonschema"
+)
+
+type Components struct {
+	Schemas map[string]jsonschema.Schema `json:"schemas,omitempty"`
+}
+
+type Specification struct {
+	Components Components `json:"components"`
+}
+
+type openapiParser struct {
+	ref map[string]jsonschema.Schema
+}
+
+func newParser(path string) (*openapiParser, error) {
+	b, err := os.ReadFile(path)
+	if err != nil {
+		return nil, err
+	}
+
+	spec := Specification{}
+	err = json.Unmarshal(b, &spec)
+	if err != nil {
+		return nil, err
+	}
+
+	p := &openapiParser{}
+	p.ref = spec.Components.Schemas
+	return p, nil
+}
+
+// This function checks if the input type:
+// 1. Is a Databricks Go SDK type.
+// 2. Has a Databricks Go SDK type embedded in it.
+//
+// If the above conditions are met, the function returns the JSON schema
+// corresponding to the Databricks Go SDK type from the OpenAPI spec.
+func (p *openapiParser) findRef(typ reflect.Type) (jsonschema.Schema, bool) {
+	typs := []reflect.Type{typ}
+
+	// Check for embedded Databricks Go SDK types.
+	if typ.Kind() == reflect.Struct {
+		for i := 0; i < typ.NumField(); i++ {
+			if !typ.Field(i).Anonymous {
+				continue
+			}
+
+			// Deference current type if it's a pointer.
+			ctyp := typ.Field(i).Type
+			for ctyp.Kind() == reflect.Ptr {
+				ctyp = ctyp.Elem()
+			}
+
+			typs = append(typs, ctyp)
+		}
+	}
+
+	for _, ctyp := range typs {
+		// Skip if it's not a Go SDK type.
+		if !strings.HasPrefix(ctyp.PkgPath(), "github.com/databricks/databricks-sdk-go") {
+			continue
+		}
+
+		pkgName := path.Base(ctyp.PkgPath())
+		k := fmt.Sprintf("%s.%s", pkgName, ctyp.Name())
+
+		// Skip if the type is not in the openapi spec.
+		_, ok := p.ref[k]
+		if !ok {
+			continue
+		}
+
+		// Return the first Go SDK type found in the openapi spec.
+		return p.ref[k], true
+	}
+
+	return jsonschema.Schema{}, false
+}
+
+// Use the OpenAPI spec to load descriptions for the given type.
+func (p *openapiParser) addDescriptions(typ reflect.Type, s jsonschema.Schema) jsonschema.Schema {
+	ref, ok := p.findRef(typ)
+	if !ok {
+		return s
+	}
+
+	s.Description = ref.Description
+	for k, v := range s.Properties {
+		if refProp, ok := ref.Properties[k]; ok {
+			v.Description = refProp.Description
+		}
+	}
+
+	return s
+}
+
+// Use the OpenAPI spec add enum values for the given type.
+func (p *openapiParser) addEnums(typ reflect.Type, s jsonschema.Schema) jsonschema.Schema {
+	ref, ok := p.findRef(typ)
+	if !ok {
+		return s
+	}
+
+	s.Enum = append(s.Enum, ref.Enum...)
+	for k, v := range s.Properties {
+		if refProp, ok := ref.Properties[k]; ok {
+			v.Enum = append(v.Enum, refProp.Enum...)
+		}
+	}
+
+	return s
+}

bundle/internal/schema/testdata/fail/basic.yml

@@ -0,0 +1,3 @@
+bundle:
+  # expected type is 'string'
+  name: 1234

bundle/internal/schema/testdata/fail/invalid_enum_value_in_job.yml

@@ -0,0 +1,4 @@
+resources:
+  jobs:
+    myjob:
+      format: INVALID_VALUE

bundle/internal/schema/testdata/fail/invalid_enum_value_in_model.yml

@@ -0,0 +1,6 @@
+resources:
+  models:
+    mymodel:
+      latest_versions:
+        - creation_timestamp: 123
+          status: INVALID_VALUE

bundle/internal/schema/testdata/fail/invalid_reference_in_job.yml

@@ -0,0 +1,8 @@
+resources:
+  jobs:
+    outer:
+      name: outer job
+      tasks:
+        - task_key: run job task 1
+          run_job_task:
+            job_id: ${invalid.reference}

bundle/internal/schema/testdata/fail/invalid_reference_in_model.yml

@@ -0,0 +1,5 @@
+resources:
+  models:
+    mymodel:
+      latest_versions:
+        - creation_timestamp: ${invalid.reference}

bundle/internal/schema/testdata/fail/required_field_missing_in_job.yml

@@ -0,0 +1,9 @@
+resources:
+  jobs:
+    foo:
+      name: my job
+      tasks:
+        # All tasks need to have a task_key.
+        - notebook_task:
+            notebook_path: /Users/abc/notebooks/inner
+          existing_cluster_id: abcd

bundle/internal/schema/testdata/fail/unknown_field_in_job.yml

@@ -0,0 +1,5 @@
+resources:
+  jobs:
+    myjob:
+      # unknown fields should cause schema failure.
+      unknown_field: "value"

bundle/internal/schema/testdata/fail/unknown_field_in_model.yml

@@ -0,0 +1,6 @@
+resources:
+  models:
+    mymodel:
+      creation_timestamp: 123
+      description: "my model"
+      unknown: "value"

bundle/internal/schema/testdata/fail/unknown_top_level_field.yml

@@ -0,0 +1 @@
+unknown: value

bundle/internal/schema/testdata/pass/artifact_references.yml

@@ -0,0 +1,11 @@
+artifacts:
+  abc:
+    path: /Workspace/a/b/c
+    type: wheel
+    files:
+      - source: ./x.whl
+
+resources:
+  jobs:
+    foo:
+      name: ${artifacts.abc.type}

bundle/internal/schema/testdata/pass/basic.yml

@@ -0,0 +1,2 @@
+bundle:
+  name: basic

bundle/internal/schema/testdata/pass/direct_value_in_target.yml

@@ -0,0 +1,4 @@
+targets:
+  development:
+    variables:
+      myvar: value

bundle/internal/schema/testdata/pass/job.yml

@@ -0,0 +1,63 @@
+bundle:
+  name: a job
+
+workspace:
+  host: "https://myworkspace.com"
+  root_path: /abc
+
+presets:
+  name_prefix: "[DEV]"
+  jobs_max_concurrent_runs: 10
+
+variables:
+  simplevar:
+    default: true
+    description: "simplevar description"
+
+  complexvar:
+    default:
+      key1: value1
+      key2: value2
+      key3:
+        - value3
+        - value4
+    description: "complexvar description"
+
+run_as:
+  service_principal_name: myserviceprincipal
+
+resources:
+  jobs:
+    myjob:
+      name: myjob
+      continuous:
+        pause_status: PAUSED
+      edit_mode: EDITABLE
+      max_concurrent_runs: 10
+      description: "my job description"
+      email_notifications:
+        no_alert_for_skipped_runs: true
+      environments:
+        - environment_key: venv
+          spec:
+            dependencies:
+              - python=3.7
+            client: "myclient"
+      format: MULTI_TASK
+      tags:
+        foo: bar
+        bar: baz
+      tasks:
+        - task_key: mytask
+          notebook_task:
+            notebook_path: ${var.simplevar}
+          existing_cluster_id: abcd
+        - task_key: mytask2
+          for_each_task:
+            inputs: av
+            concurrency: 10
+            task:
+              task_key: inside_for_each
+              notebook_task:
+                notebook_path: ${var.complexvar.key3[0]}
+        - ${var.complexvar}

bundle/internal/schema/testdata/pass/ml.yml

@@ -0,0 +1,72 @@
+bundle:
+  name: ML
+
+workspace:
+  host: "https://myworkspace.com"
+  root_path: /abc
+
+presets:
+  name_prefix: "[DEV]"
+  jobs_max_concurrent_runs: 10
+
+variables:
+  simplevar:
+    default: "true"
+    description: "simplevar description"
+
+  complexvar:
+    default:
+      key1: value1
+      key2: value2
+      key3:
+        - value3
+        - value4
+    description: "complexvar description"
+
+resources:
+  models:
+    mymodel:
+      creation_timestamp: 123
+      description: "my model"
+      latest_versions:
+        - creation_timestamp: 123
+          tags: ${var.complexvar.key1}
+          status: READY
+      permissions:
+        - service_principal_name: myserviceprincipal
+          level: CAN_MANAGE
+
+  experiments:
+    myexperiment:
+      artifact_location: /dbfs/myexperiment
+      last_update_time: ${var.complexvar.key2}
+      lifecycle_stage: ${var.simplevar}
+      permissions:
+        - service_principal_name: myserviceprincipal
+          level: CAN_MANAGE
+
+  model_serving_endpoints:
+    myendpoint:
+      config:
+        served_models:
+          - model_name: ${resources.models.mymodel.name}
+            model_version: abc
+            scale_to_zero_enabled: true
+            workload_size: Large
+      name: myendpoint
+
+  schemas:
+    myschema:
+      catalog_name: mycatalog
+      name: myschema
+
+  registered_models:
+    myregisteredmodel:
+      catalog_name: mycatalog
+      name: myregisteredmodel
+      schema_name: ${resources.schemas.myschema.name}
+      grants:
+        - principal: abcd
+          privileges:
+            - SELECT
+            - INSERT

bundle/internal/schema/testdata/pass/pipeline.yml

@@ -0,0 +1,54 @@
+bundle:
+  name: a pipeline
+
+workspace:
+  host: "https://myworkspace.com"
+  root_path: /abc
+
+presets:
+  name_prefix: "[DEV]"
+  jobs_max_concurrent_runs: 10
+
+variables:
+  simplevar:
+    default: true
+    description: "simplevar description"
+
+  complexvar:
+    default:
+      key1: value1
+      key2: value2
+      key3:
+        - value3
+        - value4
+    description: "complexvar description"
+
+artifacts:
+  mywheel:
+    path: ./mywheel.whl
+    type: WHEEL
+
+run_as:
+  service_principal_name: myserviceprincipal
+
+resources:
+  jobs:
+    myjob:
+      name: myjob
+      tasks:
+        - task_key: ${bundle.name} pipeline trigger
+          pipeline_task:
+            pipeline_id: ${resources.mypipeline.id}
+
+  pipelines:
+    mypipeline:
+      name: mypipeline
+      libraries:
+        - whl: ./mywheel.whl
+      catalog: 3{var.complexvar.key2}
+      development: true
+      clusters:
+        - autoscale:
+            mode: ENHANCED
+            max_workers: 10
+            min_workers: 1

bundle/internal/schema/testdata/pass/quality_monitor.yml

@@ -0,0 +1,16 @@
+bundle:
+  name: quality_monitor
+
+resources:
+  quality_monitors:
+    myqualitymonitor:
+      inference_log:
+        granularities:
+          - a
+          - b
+        model_id_col: a
+        prediction_col: b
+        timestamp_col: c
+        problem_type: PROBLEM_TYPE_CLASSIFICATION
+      assets_dir: /dbfs/mnt/abc
+      output_schema_name: default

bundle/internal/schema/testdata/pass/run_job_task.yml

@@ -0,0 +1,56 @@
+bundle:
+  name: a run job task
+  databricks_cli_version: 0.200.0
+  compute_id: "mycompute"
+
+
+variables:
+  simplevar:
+    default: 5678
+    description: "simplevar description"
+
+  complexvar:
+    default:
+      key1: 1234
+      key2: value2
+      key3:
+        - value3
+        - 9999
+    description: "complexvar description"
+
+resources:
+  jobs:
+    inner:
+      permissions:
+        - user_name: user1
+          level: CAN_MANAGE
+
+      name: inner job
+      tasks:
+        - task_key: inner notebook task
+          notebook_task:
+            notebook_path: /Users/abc/notebooks/inner
+          existing_cluster_id: abcd
+
+    outer:
+      name: outer job
+      tasks:
+        - task_key: run job task 1
+          run_job_task:
+            job_id: 1234
+
+        - task_key: run job task 2
+          run_job_task:
+            job_id: ${var.complexvar.key1}
+
+        - task_key: run job task 3
+          run_job_task:
+            job_id: ${var.simplevar}
+
+        - task_key: run job task 4
+          run_job_task:
+            job_id: ${resources.inner.id}
+
+        - task_key: run job task 5
+          run_job_task:
+            job_id: ${var.complexvar.key3[1]}

bundle/internal/schema/testdata/pass/schema.yml

@@ -0,0 +1,24 @@
+bundle:
+  name: basic
+
+variables:
+  complexvar:
+    default:
+      key1: 1234
+      key2: value2
+      key3:
+        - value3
+        - 9999
+    description: complexvar description
+
+resources:
+  schemas:
+    myschema:
+      name: myschema
+      catalog_name: main
+      grants:
+        - ${var.complexvar}
+        - principal: ${workspace.current_user.me}
+          privileges:
+            - ${var.complexvar.key3[0]}
+            - ${var.complexvar.key2}

bundle/internal/tf/codegen/generator/generator.go

@@ -51,9 +51,15 @@ func (r *root) Generate(path string) error {
 }
 
 func Run(ctx context.Context, schema *tfjson.ProviderSchema, path string) error {
-	// Generate types for resources.
+	// Generate types for resources
 	var resources []*namedBlock
 	for _, k := range sortKeys(schema.ResourceSchemas) {
+		// Skipping all plugin framework struct generation.
+		// TODO: This is a temporary fix, generation should be fixed in the future.
+		if strings.HasSuffix(k, "_pluginframework") {
+			continue
+		}
+
 		v := schema.ResourceSchemas[k]
 		b := &namedBlock{
 			filePattern:    "resource_%s.go",
@@ -71,6 +77,12 @@ func Run(ctx context.Context, schema *tfjson.ProviderSchema, path string) error
 	// Generate types for data sources.
 	var dataSources []*namedBlock
 	for _, k := range sortKeys(schema.DataSourceSchemas) {
+		// Skipping all plugin framework struct generation.
+		// TODO: This is a temporary fix, generation should be fixed in the future.
+		if strings.HasSuffix(k, "_pluginframework") {
+			continue
+		}
+
 		v := schema.DataSourceSchemas[k]
 		b := &namedBlock{
 			filePattern:    "data_source_%s.go",

bundle/internal/tf/codegen/schema/version.go

@@ -1,3 +1,3 @@
 package schema
 
-const ProviderVersion = "1.50.0"
+const ProviderVersion = "1.52.0"

bundle/internal/tf/schema/data_source_clusters.go

@@ -2,8 +2,16 @@
 
 package schema
 
-type DataSourceClusters struct {
-	ClusterNameContains string   `json:"cluster_name_contains,omitempty"`
-	Id                  string   `json:"id,omitempty"`
-	Ids                 []string `json:"ids,omitempty"`
+type DataSourceClustersFilterBy struct {
+	ClusterSources []string `json:"cluster_sources,omitempty"`
+	ClusterStates  []string `json:"cluster_states,omitempty"`
+	IsPinned       bool     `json:"is_pinned,omitempty"`
+	PolicyId       string   `json:"policy_id,omitempty"`
+}
+
+type DataSourceClusters struct {
+	ClusterNameContains string                      `json:"cluster_name_contains,omitempty"`
+	Id                  string                      `json:"id,omitempty"`
+	Ids                 []string                    `json:"ids,omitempty"`
+	FilterBy            *DataSourceClustersFilterBy `json:"filter_by,omitempty"`
 }

bundle/internal/tf/schema/data_source_external_location.go

@@ -19,6 +19,7 @@ type DataSourceExternalLocationExternalLocationInfo struct {
 	CreatedBy         string                                                           `json:"created_by,omitempty"`
 	CredentialId      string                                                           `json:"credential_id,omitempty"`
 	CredentialName    string                                                           `json:"credential_name,omitempty"`
+	Fallback          bool                                                             `json:"fallback,omitempty"`
 	IsolationMode     string                                                           `json:"isolation_mode,omitempty"`
 	MetastoreId       string                                                           `json:"metastore_id,omitempty"`
 	Name              string                                                           `json:"name,omitempty"`

bundle/internal/tf/schema/data_source_share.go

@@ -18,12 +18,14 @@ type DataSourceShareObject struct {
 	AddedBy                  string                           `json:"added_by,omitempty"`
 	CdfEnabled               bool                             `json:"cdf_enabled,omitempty"`
 	Comment                  string                           `json:"comment,omitempty"`
+	Content                  string                           `json:"content,omitempty"`
 	DataObjectType           string                           `json:"data_object_type"`
 	HistoryDataSharingStatus string                           `json:"history_data_sharing_status,omitempty"`
 	Name                     string                           `json:"name"`
 	SharedAs                 string                           `json:"shared_as,omitempty"`
 	StartVersion             int                              `json:"start_version,omitempty"`
 	Status                   string                           `json:"status,omitempty"`
+	StringSharedAs           string                           `json:"string_shared_as,omitempty"`
 	Partition                []DataSourceShareObjectPartition `json:"partition,omitempty"`
 }
 

bundle/internal/tf/schema/resource_automatic_cluster_update_workspace_setting.go

@@ -2,20 +2,14 @@
 
 package schema
 
-type ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceEnablementDetails struct {
-	ForcedForComplianceMode           bool `json:"forced_for_compliance_mode,omitempty"`
-	UnavailableForDisabledEntitlement bool `json:"unavailable_for_disabled_entitlement,omitempty"`
-	UnavailableForNonEnterpriseTier   bool `json:"unavailable_for_non_enterprise_tier,omitempty"`
-}
-
 type ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceMaintenanceWindowWeekDayBasedScheduleWindowStartTime struct {
-	Hours   int `json:"hours,omitempty"`
-	Minutes int `json:"minutes,omitempty"`
+	Hours   int `json:"hours"`
+	Minutes int `json:"minutes"`
 }
 
 type ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceMaintenanceWindowWeekDayBasedSchedule struct {
-	DayOfWeek       string                                                                                                                             `json:"day_of_week,omitempty"`
-	Frequency       string                                                                                                                             `json:"frequency,omitempty"`
+	DayOfWeek       string                                                                                                                             `json:"day_of_week"`
+	Frequency       string                                                                                                                             `json:"frequency"`
 	WindowStartTime *ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceMaintenanceWindowWeekDayBasedScheduleWindowStartTime `json:"window_start_time,omitempty"`
 }
 
@@ -25,9 +19,9 @@ type ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspa
 
 type ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspace struct {
 	CanToggle                       bool                                                                                            `json:"can_toggle,omitempty"`
-	Enabled                         bool                                                                                            `json:"enabled,omitempty"`
+	Enabled                         bool                                                                                            `json:"enabled"`
+	EnablementDetails               []any                                                                                           `json:"enablement_details,omitempty"`
 	RestartEvenIfNoUpdatesAvailable bool                                                                                            `json:"restart_even_if_no_updates_available,omitempty"`
-	EnablementDetails               *ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceEnablementDetails `json:"enablement_details,omitempty"`
 	MaintenanceWindow               *ResourceAutomaticClusterUpdateWorkspaceSettingAutomaticClusterUpdateWorkspaceMaintenanceWindow `json:"maintenance_window,omitempty"`
 }
 

bundle/internal/tf/schema/resource_cluster.go

@@ -176,6 +176,7 @@ type ResourceCluster struct {
 	IdempotencyToken          string                            `json:"idempotency_token,omitempty"`
 	InstancePoolId            string                            `json:"instance_pool_id,omitempty"`
 	IsPinned                  bool                              `json:"is_pinned,omitempty"`
+	NoWait                    bool                              `json:"no_wait,omitempty"`
 	NodeTypeId                string                            `json:"node_type_id,omitempty"`
 	NumWorkers                int                               `json:"num_workers,omitempty"`
 	PolicyId                  string                            `json:"policy_id,omitempty"`

bundle/internal/tf/schema/resource_compliance_security_profile_workspace_setting.go

@@ -3,8 +3,8 @@
 package schema
 
 type ResourceComplianceSecurityProfileWorkspaceSettingComplianceSecurityProfileWorkspace struct {
-	ComplianceStandards []string `json:"compliance_standards,omitempty"`
-	IsEnabled           bool     `json:"is_enabled,omitempty"`
+	ComplianceStandards []string `json:"compliance_standards"`
+	IsEnabled           bool     `json:"is_enabled"`
 }
 
 type ResourceComplianceSecurityProfileWorkspaceSetting struct {

bundle/internal/tf/schema/resource_enhanced_security_monitoring_workspace_setting.go

@@ -3,7 +3,7 @@
 package schema
 
 type ResourceEnhancedSecurityMonitoringWorkspaceSettingEnhancedSecurityMonitoringWorkspace struct {
-	IsEnabled bool `json:"is_enabled,omitempty"`
+	IsEnabled bool `json:"is_enabled"`
 }
 
 type ResourceEnhancedSecurityMonitoringWorkspaceSetting struct {

bundle/internal/tf/schema/resource_model_serving.go

@@ -95,14 +95,16 @@ type ResourceModelServingConfigServedEntities struct {
 }
 
 type ResourceModelServingConfigServedModels struct {
-	EnvironmentVars    map[string]string `json:"environment_vars,omitempty"`
-	InstanceProfileArn string            `json:"instance_profile_arn,omitempty"`
-	ModelName          string            `json:"model_name"`
-	ModelVersion       string            `json:"model_version"`
-	Name               string            `json:"name,omitempty"`
-	ScaleToZeroEnabled bool              `json:"scale_to_zero_enabled,omitempty"`
-	WorkloadSize       string            `json:"workload_size"`
-	WorkloadType       string            `json:"workload_type,omitempty"`
+	EnvironmentVars          map[string]string `json:"environment_vars,omitempty"`
+	InstanceProfileArn       string            `json:"instance_profile_arn,omitempty"`
+	MaxProvisionedThroughput int               `json:"max_provisioned_throughput,omitempty"`
+	MinProvisionedThroughput int               `json:"min_provisioned_throughput,omitempty"`
+	ModelName                string            `json:"model_name"`
+	ModelVersion             string            `json:"model_version"`
+	Name                     string            `json:"name,omitempty"`
+	ScaleToZeroEnabled       bool              `json:"scale_to_zero_enabled,omitempty"`
+	WorkloadSize             string            `json:"workload_size,omitempty"`
+	WorkloadType             string            `json:"workload_type,omitempty"`
 }
 
 type ResourceModelServingConfigTrafficConfigRoutes struct {

bundle/internal/tf/schema/resource_share.go

@@ -18,20 +18,27 @@ type ResourceShareObject struct {
 	AddedBy                  string                         `json:"added_by,omitempty"`
 	CdfEnabled               bool                           `json:"cdf_enabled,omitempty"`
 	Comment                  string                         `json:"comment,omitempty"`
+	Content                  string                         `json:"content,omitempty"`
 	DataObjectType           string                         `json:"data_object_type"`
 	HistoryDataSharingStatus string                         `json:"history_data_sharing_status,omitempty"`
 	Name                     string                         `json:"name"`
 	SharedAs                 string                         `json:"shared_as,omitempty"`
 	StartVersion             int                            `json:"start_version,omitempty"`
 	Status                   string                         `json:"status,omitempty"`
+	StringSharedAs           string                         `json:"string_shared_as,omitempty"`
 	Partition                []ResourceShareObjectPartition `json:"partition,omitempty"`
 }
 
 type ResourceShare struct {
-	CreatedAt int                   `json:"created_at,omitempty"`
-	CreatedBy string                `json:"created_by,omitempty"`
-	Id        string                `json:"id,omitempty"`
-	Name      string                `json:"name"`
-	Owner     string                `json:"owner,omitempty"`
-	Object    []ResourceShareObject `json:"object,omitempty"`
+	Comment         string                `json:"comment,omitempty"`
+	CreatedAt       int                   `json:"created_at,omitempty"`
+	CreatedBy       string                `json:"created_by,omitempty"`
+	Id              string                `json:"id,omitempty"`
+	Name            string                `json:"name"`
+	Owner           string                `json:"owner,omitempty"`
+	StorageLocation string                `json:"storage_location,omitempty"`
+	StorageRoot     string                `json:"storage_root,omitempty"`
+	UpdatedAt       int                   `json:"updated_at,omitempty"`
+	UpdatedBy       string                `json:"updated_by,omitempty"`
+	Object          []ResourceShareObject `json:"object,omitempty"`
 }

bundle/internal/tf/schema/resource_sql_table.go

@@ -15,6 +15,7 @@ type ResourceSqlTable struct {
 	ClusterKeys           []string                 `json:"cluster_keys,omitempty"`
 	Comment               string                   `json:"comment,omitempty"`
 	DataSourceFormat      string                   `json:"data_source_format,omitempty"`
+	EffectiveProperties   map[string]string        `json:"effective_properties,omitempty"`
 	Id                    string                   `json:"id,omitempty"`
 	Name                  string                   `json:"name"`
 	Options               map[string]string        `json:"options,omitempty"`

bundle/internal/tf/schema/root.go

@@ -21,7 +21,7 @@ type Root struct {
 
 const ProviderHost = "registry.terraform.io"
 const ProviderSource = "databricks/databricks"
-const ProviderVersion = "1.50.0"
+const ProviderVersion = "1.52.0"
 
 func NewRoot() *Root {
 	return &Root{

bundle/libraries/expand_glob_references_test.go

@@ -10,6 +10,7 @@ import (
 	"github.com/databricks/cli/bundle/config/resources"
 	"github.com/databricks/cli/bundle/internal/bundletest"
 	"github.com/databricks/cli/internal/testutil"
+	"github.com/databricks/cli/libs/dyn"
 	"github.com/databricks/databricks-sdk-go/service/compute"
 	"github.com/databricks/databricks-sdk-go/service/jobs"
 	"github.com/stretchr/testify/require"
@@ -61,7 +62,7 @@ func TestGlobReferencesExpandedForTaskLibraries(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, ExpandGlobReferences())
 	require.Empty(t, diags)
@@ -146,7 +147,7 @@ func TestGlobReferencesExpandedForForeachTaskLibraries(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, ExpandGlobReferences())
 	require.Empty(t, diags)
@@ -221,7 +222,7 @@ func TestGlobReferencesExpandedForEnvironmentsDeps(t *testing.T) {
 		},
 	}
 
-	bundletest.SetLocation(b, ".", filepath.Join(dir, "resource.yml"))
+	bundletest.SetLocation(b, ".", []dyn.Location{{File: filepath.Join(dir, "resource.yml")}})
 
 	diags := bundle.Apply(context.Background(), b, ExpandGlobReferences())
 	require.Empty(t, diags)

bundle/phases/deploy.go

@@ -18,6 +18,7 @@ import (
 	"github.com/databricks/cli/bundle/python"
 	"github.com/databricks/cli/bundle/scripts"
 	"github.com/databricks/cli/libs/cmdio"
+	"github.com/databricks/cli/libs/sync"
 	terraformlib "github.com/databricks/cli/libs/terraform"
 	tfjson "github.com/hashicorp/terraform-json"
 )
@@ -128,7 +129,7 @@ properties such as the 'catalog' or 'storage' are changed:`
 }
 
 // The deploy phase deploys artifacts and resources.
-func Deploy() bundle.Mutator {
+func Deploy(outputHandler sync.OutputHandler) bundle.Mutator {
 	// Core mutators that CRUD resources and modify deployment state. These
 	// mutators need informed consent if they are potentially destructive.
 	deployCore := bundle.Defer(
@@ -157,7 +158,7 @@ func Deploy() bundle.Mutator {
 				libraries.ExpandGlobReferences(),
 				libraries.Upload(),
 				python.TransformWheelTask(),
-				files.Upload(),
+				files.Upload(outputHandler),
 				deploy.StateUpdate(),
 				deploy.StatePush(),
 				permissions.ApplyWorkspaceRootPermissions(),

bundle/schema/README.md

@@ -1,18 +0,0 @@
-### Overview
-
-`docs/bundle_descriptions.json` contains both autogenerated as well as manually written
-descriptions for the json schema. Specifically
-1. `resources` : almost all descriptions are autogenerated from the OpenAPI spec
-2. `targets` : almost all descriptions are copied over from root level entities (eg: `bundle`, `artifacts`)
-3. `bundle` : manually editted
-4. `include` : manually editted
-5. `workspace` : manually editted
-6. `artifacts` : manually editted
-
-These descriptions are rendered in the inline documentation in an IDE
-
-### SOP: Add schema descriptions for new fields in bundle config
-
-Manually edit bundle_descriptions.json to add your descriptions. Note that the
-descriptions in `resources` block is generated from the OpenAPI spec, and thus
-any changes there will be overwritten.

bundle/schema/docs.go

@@ -1,109 +0,0 @@
-package schema
-
-import (
-	_ "embed"
-	"encoding/json"
-	"fmt"
-	"os"
-	"reflect"
-
-	"github.com/databricks/cli/bundle/config"
-	"github.com/databricks/cli/libs/jsonschema"
-)
-
-// A subset of Schema struct
-type Docs struct {
-	Description          string           `json:"description"`
-	Properties           map[string]*Docs `json:"properties,omitempty"`
-	Items                *Docs            `json:"items,omitempty"`
-	AdditionalProperties *Docs            `json:"additionalproperties,omitempty"`
-}
-
-//go:embed docs/bundle_descriptions.json
-var bundleDocs []byte
-
-func (docs *Docs) refreshTargetsDocs() error {
-	targetsDocs, ok := docs.Properties["targets"]
-	if !ok || targetsDocs.AdditionalProperties == nil ||
-		targetsDocs.AdditionalProperties.Properties == nil {
-		return fmt.Errorf("invalid targets descriptions")
-	}
-	targetProperties := targetsDocs.AdditionalProperties.Properties
-	propertiesToCopy := []string{"artifacts", "bundle", "resources", "workspace"}
-	for _, p := range propertiesToCopy {
-		targetProperties[p] = docs.Properties[p]
-	}
-	return nil
-}
-
-func LoadBundleDescriptions() (*Docs, error) {
-	embedded := Docs{}
-	err := json.Unmarshal(bundleDocs, &embedded)
-	return &embedded, err
-}
-
-func UpdateBundleDescriptions(openapiSpecPath string) (*Docs, error) {
-	embedded, err := LoadBundleDescriptions()
-	if err != nil {
-		return nil, err
-	}
-
-	// Generate schema from the embedded descriptions, and convert it back to docs.
-	// This creates empty descriptions for any properties that were missing in the
-	// embedded descriptions.
-	schema, err := New(reflect.TypeOf(config.Root{}), embedded)
-	if err != nil {
-		return nil, err
-	}
-	docs := schemaToDocs(schema)
-
-	// Load the Databricks OpenAPI spec
-	openapiSpec, err := os.ReadFile(openapiSpecPath)
-	if err != nil {
-		return nil, err
-	}
-	spec := &Specification{}
-	err = json.Unmarshal(openapiSpec, spec)
-	if err != nil {
-		return nil, err
-	}
-	openapiReader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-
-	// Generate descriptions for the "resources" field
-	resourcesDocs, err := openapiReader.ResourcesDocs()
-	if err != nil {
-		return nil, err
-	}
-	resourceSchema, err := New(reflect.TypeOf(config.Resources{}), resourcesDocs)
-	if err != nil {
-		return nil, err
-	}
-	docs.Properties["resources"] = schemaToDocs(resourceSchema)
-	docs.refreshTargetsDocs()
-	return docs, nil
-}
-
-// *Docs are a subset of *Schema, this function selects that subset
-func schemaToDocs(jsonSchema *jsonschema.Schema) *Docs {
-	// terminate recursion if schema is nil
-	if jsonSchema == nil {
-		return nil
-	}
-	docs := &Docs{
-		Description: jsonSchema.Description,
-	}
-	if len(jsonSchema.Properties) > 0 {
-		docs.Properties = make(map[string]*Docs)
-	}
-	for k, v := range jsonSchema.Properties {
-		docs.Properties[k] = schemaToDocs(v)
-	}
-	docs.Items = schemaToDocs(jsonSchema.Items)
-	if additionalProperties, ok := jsonSchema.AdditionalProperties.(*jsonschema.Schema); ok {
-		docs.AdditionalProperties = schemaToDocs(additionalProperties)
-	}
-	return docs
-}

bundle/schema/docs_test.go

@@ -1,62 +0,0 @@
-package schema
-
-import (
-	"encoding/json"
-	"testing"
-
-	"github.com/databricks/cli/libs/jsonschema"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-)
-
-func TestSchemaToDocs(t *testing.T) {
-	jsonSchema := &jsonschema.Schema{
-		Type:        "object",
-		Description: "root doc",
-		Properties: map[string]*jsonschema.Schema{
-			"foo": {Type: "number", Description: "foo doc"},
-			"bar": {Type: "string"},
-			"octave": {
-				Type:                 "object",
-				AdditionalProperties: &jsonschema.Schema{Type: "number"},
-				Description:          "octave docs",
-			},
-			"scales": {
-				Type:        "object",
-				Description: "scale docs",
-				Items:       &jsonschema.Schema{Type: "string"},
-			},
-		},
-	}
-	docs := schemaToDocs(jsonSchema)
-	docsJson, err := json.MarshalIndent(docs, "		", "	")
-	require.NoError(t, err)
-
-	expected :=
-		`{
-			"description": "root doc",
-			"properties": {
-				"bar": {
-					"description": ""
-				},
-				"foo": {
-					"description": "foo doc"
-				},
-				"octave": {
-					"description": "octave docs",
-					"additionalproperties": {
-						"description": ""
-					}
-				},
-				"scales": {
-					"description": "scale docs",
-					"items": {
-						"description": ""
-					}
-				}
-			}
-		}`
-	t.Log("[DEBUG] actual: ", string(docsJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(docsJson))
-}

bundle/schema/embed.go

@@ -0,0 +1,6 @@
+package schema
+
+import _ "embed"
+
+//go:embed jsonschema.json
+var Bytes []byte

bundle/schema/embed_test.go

@@ -0,0 +1,71 @@
+package schema_test
+
+import (
+	"encoding/json"
+	"testing"
+
+	"github.com/databricks/cli/bundle/schema"
+	"github.com/databricks/cli/libs/jsonschema"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func walk(defs map[string]any, p ...string) jsonschema.Schema {
+	v, ok := defs[p[0]]
+	if !ok {
+		panic("not found: " + p[0])
+	}
+
+	if len(p) == 1 {
+		b, err := json.Marshal(v)
+		if err != nil {
+			panic(err)
+		}
+		res := jsonschema.Schema{}
+		err = json.Unmarshal(b, &res)
+		if err != nil {
+			panic(err)
+		}
+		return res
+	}
+
+	return walk(v.(map[string]any), p[1:]...)
+}
+
+func TestJsonSchema(t *testing.T) {
+	s := jsonschema.Schema{}
+	err := json.Unmarshal(schema.Bytes, &s)
+	require.NoError(t, err)
+
+	// Assert job fields have their descriptions loaded.
+	resourceJob := walk(s.Definitions, "github.com", "databricks", "cli", "bundle", "config", "resources.Job")
+	fields := []string{"name", "continuous", "deployment", "tasks", "trigger"}
+	for _, field := range fields {
+		assert.NotEmpty(t, resourceJob.AnyOf[0].Properties[field].Description)
+	}
+
+	// Assert descriptions were also loaded for a job task definition.
+	jobTask := walk(s.Definitions, "github.com", "databricks", "databricks-sdk-go", "service", "jobs.Task")
+	fields = []string{"notebook_task", "spark_jar_task", "spark_python_task", "spark_submit_task", "description", "depends_on", "environment_key", "for_each_task", "existing_cluster_id"}
+	for _, field := range fields {
+		assert.NotEmpty(t, jobTask.AnyOf[0].Properties[field].Description)
+	}
+
+	// Assert descriptions are loaded for pipelines
+	pipeline := walk(s.Definitions, "github.com", "databricks", "cli", "bundle", "config", "resources.Pipeline")
+	fields = []string{"name", "catalog", "clusters", "channel", "continuous", "deployment", "development"}
+	for _, field := range fields {
+		assert.NotEmpty(t, pipeline.AnyOf[0].Properties[field].Description)
+	}
+
+	// Assert enum values are loaded
+	schedule := walk(s.Definitions, "github.com", "databricks", "databricks-sdk-go", "service", "catalog.MonitorCronSchedule")
+	assert.Contains(t, schedule.AnyOf[0].Properties["pause_status"].Enum, "PAUSED")
+	assert.Contains(t, schedule.AnyOf[0].Properties["pause_status"].Enum, "UNPAUSED")
+
+	providers := walk(s.Definitions, "github.com", "databricks", "databricks-sdk-go", "service", "jobs.GitProvider")
+	assert.Contains(t, providers.Enum, "gitHub")
+	assert.Contains(t, providers.Enum, "bitbucketCloud")
+	assert.Contains(t, providers.Enum, "gitHubEnterprise")
+	assert.Contains(t, providers.Enum, "bitbucketServer")
+}

bundle/schema/openapi.go

@@ -1,293 +0,0 @@
-package schema
-
-import (
-	"encoding/json"
-	"fmt"
-	"strings"
-
-	"github.com/databricks/cli/libs/jsonschema"
-)
-
-type OpenapiReader struct {
-	// OpenAPI spec to read schemas from.
-	OpenapiSpec *Specification
-
-	// In-memory cache of schemas read from the OpenAPI spec.
-	memo map[string]jsonschema.Schema
-}
-
-const SchemaPathPrefix = "#/components/schemas/"
-
-// Read a schema directly from the OpenAPI spec.
-func (reader *OpenapiReader) readOpenapiSchema(path string) (jsonschema.Schema, error) {
-	schemaKey := strings.TrimPrefix(path, SchemaPathPrefix)
-
-	// return early if we already have a computed schema
-	memoSchema, ok := reader.memo[schemaKey]
-	if ok {
-		return memoSchema, nil
-	}
-
-	// check path is present in openapi spec
-	openapiSchema, ok := reader.OpenapiSpec.Components.Schemas[schemaKey]
-	if !ok {
-		return jsonschema.Schema{}, fmt.Errorf("schema with path %s not found in openapi spec", path)
-	}
-
-	// convert openapi schema to the native schema struct
-	bytes, err := json.Marshal(*openapiSchema)
-	if err != nil {
-		return jsonschema.Schema{}, err
-	}
-	jsonSchema := jsonschema.Schema{}
-	err = json.Unmarshal(bytes, &jsonSchema)
-	if err != nil {
-		return jsonschema.Schema{}, err
-	}
-
-	// A hack to convert a map[string]interface{} to *Schema
-	// We rely on the type of a AdditionalProperties in downstream functions
-	// to do reference interpolation
-	_, ok = jsonSchema.AdditionalProperties.(map[string]interface{})
-	if ok {
-		b, err := json.Marshal(jsonSchema.AdditionalProperties)
-		if err != nil {
-			return jsonschema.Schema{}, err
-		}
-		additionalProperties := &jsonschema.Schema{}
-		err = json.Unmarshal(b, additionalProperties)
-		if err != nil {
-			return jsonschema.Schema{}, err
-		}
-		jsonSchema.AdditionalProperties = additionalProperties
-	}
-
-	// store read schema into memo
-	reader.memo[schemaKey] = jsonSchema
-
-	return jsonSchema, nil
-}
-
-// Resolve all nested "$ref" references in the schema. This function unrolls a single
-// level of "$ref" in the schema and calls into traverseSchema to resolve nested references.
-// Thus this function and traverseSchema are mutually recursive.
-//
-// This function is safe against reference loops. If a reference loop is detected, an error
-// is returned.
-func (reader *OpenapiReader) safeResolveRefs(root *jsonschema.Schema, tracker *tracker) (*jsonschema.Schema, error) {
-	if root.Reference == nil {
-		return reader.traverseSchema(root, tracker)
-	}
-	key := *root.Reference
-
-	// HACK to unblock CLI release (13th Feb 2024). This is temporary until proper
-	// support for recursive types is added to the docs generator. PR: https://github.com/databricks/cli/pull/1204
-	if strings.Contains(key, "ForEachTask") {
-		return root, nil
-	}
-
-	if tracker.hasCycle(key) {
-		// self reference loops can be supported however the logic is non-trivial because
-		// cross refernce loops are not allowed (see: http://json-schema.org/understanding-json-schema/structuring.html#recursion)
-		return nil, fmt.Errorf("references loop detected")
-	}
-	ref := *root.Reference
-	description := root.Description
-	tracker.push(ref, ref)
-
-	// Mark reference nil, so we do not traverse this again. This is tracked
-	// in the memo
-	root.Reference = nil
-
-	// unroll one level of reference.
-	selfRef, err := reader.readOpenapiSchema(ref)
-	if err != nil {
-		return nil, err
-	}
-	root = &selfRef
-	root.Description = description
-
-	// traverse again to find new references
-	root, err = reader.traverseSchema(root, tracker)
-	if err != nil {
-		return nil, err
-	}
-	tracker.pop(ref)
-	return root, err
-}
-
-// Traverse the nested properties of the schema to resolve "$ref" references. This function
-// and safeResolveRefs are mutually recursive.
-func (reader *OpenapiReader) traverseSchema(root *jsonschema.Schema, tracker *tracker) (*jsonschema.Schema, error) {
-	// case primitive (or invalid)
-	if root.Type != jsonschema.ObjectType && root.Type != jsonschema.ArrayType {
-		return root, nil
-	}
-	// only root references are resolved
-	if root.Reference != nil {
-		return reader.safeResolveRefs(root, tracker)
-	}
-	// case struct
-	if len(root.Properties) > 0 {
-		for k, v := range root.Properties {
-			childSchema, err := reader.safeResolveRefs(v, tracker)
-			if err != nil {
-				return nil, err
-			}
-			root.Properties[k] = childSchema
-		}
-	}
-	// case array
-	if root.Items != nil {
-		itemsSchema, err := reader.safeResolveRefs(root.Items, tracker)
-		if err != nil {
-			return nil, err
-		}
-		root.Items = itemsSchema
-	}
-	// case map
-	additionalProperties, ok := root.AdditionalProperties.(*jsonschema.Schema)
-	if ok && additionalProperties != nil {
-		valueSchema, err := reader.safeResolveRefs(additionalProperties, tracker)
-		if err != nil {
-			return nil, err
-		}
-		root.AdditionalProperties = valueSchema
-	}
-	return root, nil
-}
-
-func (reader *OpenapiReader) readResolvedSchema(path string) (*jsonschema.Schema, error) {
-	root, err := reader.readOpenapiSchema(path)
-	if err != nil {
-		return nil, err
-	}
-	tracker := newTracker()
-	tracker.push(path, path)
-	resolvedRoot, err := reader.safeResolveRefs(&root, tracker)
-	if err != nil {
-		return nil, tracker.errWithTrace(err.Error(), "")
-	}
-	return resolvedRoot, nil
-}
-
-func (reader *OpenapiReader) jobsDocs() (*Docs, error) {
-	jobSettingsSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "jobs.JobSettings")
-	if err != nil {
-		return nil, err
-	}
-	jobDocs := schemaToDocs(jobSettingsSchema)
-	// TODO: add description for id if needed.
-	// Tracked in https://github.com/databricks/cli/issues/242
-	jobsDocs := &Docs{
-		Description:          "List of Databricks jobs",
-		AdditionalProperties: jobDocs,
-	}
-	return jobsDocs, nil
-}
-
-func (reader *OpenapiReader) pipelinesDocs() (*Docs, error) {
-	pipelineSpecSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "pipelines.PipelineSpec")
-	if err != nil {
-		return nil, err
-	}
-	pipelineDocs := schemaToDocs(pipelineSpecSchema)
-	// TODO: Two fields in resources.Pipeline have the json tag id. Clarify the
-	// semantics and then add a description if needed. (https://github.com/databricks/cli/issues/242)
-	pipelinesDocs := &Docs{
-		Description:          "List of DLT pipelines",
-		AdditionalProperties: pipelineDocs,
-	}
-	return pipelinesDocs, nil
-}
-
-func (reader *OpenapiReader) experimentsDocs() (*Docs, error) {
-	experimentSpecSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "ml.Experiment")
-	if err != nil {
-		return nil, err
-	}
-	experimentDocs := schemaToDocs(experimentSpecSchema)
-	experimentsDocs := &Docs{
-		Description:          "List of MLflow experiments",
-		AdditionalProperties: experimentDocs,
-	}
-	return experimentsDocs, nil
-}
-
-func (reader *OpenapiReader) modelsDocs() (*Docs, error) {
-	modelSpecSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "ml.Model")
-	if err != nil {
-		return nil, err
-	}
-	modelDocs := schemaToDocs(modelSpecSchema)
-	modelsDocs := &Docs{
-		Description:          "List of MLflow models",
-		AdditionalProperties: modelDocs,
-	}
-	return modelsDocs, nil
-}
-
-func (reader *OpenapiReader) modelServingEndpointsDocs() (*Docs, error) {
-	modelServingEndpointsSpecSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "serving.CreateServingEndpoint")
-	if err != nil {
-		return nil, err
-	}
-	modelServingEndpointsDocs := schemaToDocs(modelServingEndpointsSpecSchema)
-	modelServingEndpointsAllDocs := &Docs{
-		Description:          "List of Model Serving Endpoints",
-		AdditionalProperties: modelServingEndpointsDocs,
-	}
-	return modelServingEndpointsAllDocs, nil
-}
-
-func (reader *OpenapiReader) registeredModelDocs() (*Docs, error) {
-	registeredModelsSpecSchema, err := reader.readResolvedSchema(SchemaPathPrefix + "catalog.CreateRegisteredModelRequest")
-	if err != nil {
-		return nil, err
-	}
-	registeredModelsDocs := schemaToDocs(registeredModelsSpecSchema)
-	registeredModelsAllDocs := &Docs{
-		Description:          "List of Registered Models",
-		AdditionalProperties: registeredModelsDocs,
-	}
-	return registeredModelsAllDocs, nil
-}
-
-func (reader *OpenapiReader) ResourcesDocs() (*Docs, error) {
-	jobsDocs, err := reader.jobsDocs()
-	if err != nil {
-		return nil, err
-	}
-	pipelinesDocs, err := reader.pipelinesDocs()
-	if err != nil {
-		return nil, err
-	}
-	experimentsDocs, err := reader.experimentsDocs()
-	if err != nil {
-		return nil, err
-	}
-	modelsDocs, err := reader.modelsDocs()
-	if err != nil {
-		return nil, err
-	}
-	modelServingEndpointsDocs, err := reader.modelServingEndpointsDocs()
-	if err != nil {
-		return nil, err
-	}
-	registeredModelsDocs, err := reader.registeredModelDocs()
-	if err != nil {
-		return nil, err
-	}
-
-	return &Docs{
-		Description: "Collection of Databricks resources to deploy.",
-		Properties: map[string]*Docs{
-			"jobs":                    jobsDocs,
-			"pipelines":               pipelinesDocs,
-			"experiments":             experimentsDocs,
-			"models":                  modelsDocs,
-			"model_serving_endpoints": modelServingEndpointsDocs,
-			"registered_models":       registeredModelsDocs,
-		},
-	}, nil
-}

bundle/schema/openapi_test.go

@@ -1,493 +0,0 @@
-package schema
-
-import (
-	"encoding/json"
-	"testing"
-
-	"github.com/databricks/cli/libs/jsonschema"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-)
-
-func TestReadSchemaForObject(t *testing.T) {
-	specString := `
-	{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "number"
-				},
-				"fruits": {
-					"type": "object",
-					"description": "fruits that are cool",
-					"properties": {
-						"guava": {
-							"type": "string",
-							"description": "a guava for my schema"
-						},
-						"mango": {
-							"type": "object",
-							"description": "a mango for my schema",
-							"$ref": "#/components/schemas/mango"
-						}
-					}
-				},
-				"mango": {
-					"type": "object",
-					"properties": {
-						"foo": {
-							"$ref": "#/components/schemas/foo"
-						}
-					}
-				}
-			}
-		}
-	}
-	`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "fruits that are cool",
-			"properties": {
-				"guava": {
-					"type": "string",
-					"description": "a guava for my schema"
-				},
-				"mango": {
-					"type": "object",
-					"description": "a mango for my schema",
-					"properties": {
-						"foo": {
-							"type": "number"
-						}
-					}
-				}
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestReadSchemaForArray(t *testing.T) {
-	specString := `
-	{
-		"components": {
-			"schemas": {
-				"fruits": {
-					"type": "object",
-					"description": "fruits that are cool",
-					"items": {
-						"description": "some papayas, because papayas are fruits too",
-						"$ref": "#/components/schemas/papaya"
-					}
-				},
-				"papaya": {
-					"type": "number"
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "fruits that are cool",
-			"items": {
-				"type": "number",
-				"description": "some papayas, because papayas are fruits too"
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestReadSchemaForMap(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"fruits": {
-					"type": "object",
-					"description": "fruits that are meh",
-					"additionalProperties": {
-						"description": "watermelons. watermelons.",
-						"$ref": "#/components/schemas/watermelon"
-					}
-				},
-				"watermelon": {
-					"type": "number"
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "fruits that are meh",
-			"additionalProperties": {
-				"type": "number",
-				"description": "watermelons. watermelons."
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestRootReferenceIsResolved(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "object",
-					"description": "this description is ignored",
-					"properties": {
-						"abc": {
-							"type": "string"
-						}
-					}
-				},
-				"fruits": {
-					"type": "object",
-					"description": "foo fighters fighting fruits",
-					"$ref": "#/components/schemas/foo"
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	schema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-	fruitsSchemaJson, err := json.MarshalIndent(schema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "foo fighters fighting fruits",
-			"properties": {
-				"abc": {
-					"type": "string"
-				}
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestSelfReferenceLoopErrors(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "object",
-					"description": "this description is ignored",
-					"properties": {
-						"bar": {
-							"type": "object",
-							"$ref": "#/components/schemas/foo"
-						}
-					}
-				},
-				"fruits": {
-					"type": "object",
-					"description": "foo fighters fighting fruits",
-					"$ref": "#/components/schemas/foo"
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	_, err = reader.readResolvedSchema("#/components/schemas/fruits")
-	assert.ErrorContains(t, err, "references loop detected. traversal trace:  -> #/components/schemas/fruits -> #/components/schemas/foo")
-}
-
-func TestCrossReferenceLoopErrors(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "object",
-					"description": "this description is ignored",
-					"properties": {
-						"bar": {
-							"type": "object",
-							"$ref": "#/components/schemas/fruits"
-						}
-					}
-				},
-				"fruits": {
-					"type": "object",
-					"description": "foo fighters fighting fruits",
-					"$ref": "#/components/schemas/foo"
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	_, err = reader.readResolvedSchema("#/components/schemas/fruits")
-	assert.ErrorContains(t, err, "references loop detected. traversal trace:  -> #/components/schemas/fruits -> #/components/schemas/foo")
-}
-
-func TestReferenceResolutionForMapInObject(t *testing.T) {
-	specString := `
-	{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "number"
-				},
-				"fruits": {
-					"type": "object",
-					"description": "fruits that are cool",
-					"properties": {
-						"guava": {
-							"type": "string",
-							"description": "a guava for my schema"
-						},
-						"mangos": {
-							"type": "object",
-							"description": "multiple mangos",
-							"$ref": "#/components/schemas/mango"
-						}
-					}
-				},
-				"mango": {
-					"type": "object",
-					"additionalProperties": {
-						"description": "a single mango",
-						"$ref": "#/components/schemas/foo"
-					}
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "fruits that are cool",
-			"properties": {
-				"guava": {
-					"type": "string",
-					"description": "a guava for my schema"
-				},
-				"mangos": {
-					"type": "object",
-					"description": "multiple mangos",
-					"additionalProperties": {
-						"type": "number",
-						"description": "a single mango"
-					}
-				}
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestReferenceResolutionForArrayInObject(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "number"
-				},
-				"fruits": {
-					"type": "object",
-					"description": "fruits that are cool",
-					"properties": {
-						"guava": {
-							"type": "string",
-							"description": "a guava for my schema"
-						},
-						"mangos": {
-							"type": "object",
-							"description": "multiple mangos",
-							"$ref": "#/components/schemas/mango"
-						}
-					}
-				},
-				"mango": {
-					"type": "object",
-					"items": {
-						"description": "a single mango",
-						"$ref": "#/components/schemas/foo"
-					}
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"description": "fruits that are cool",
-			"properties": {
-				"guava": {
-					"type": "string",
-					"description": "a guava for my schema"
-				},
-				"mangos": {
-					"type": "object",
-					"description": "multiple mangos",
-					"items": {
-						"type": "number",
-						"description": "a single mango"
-					}
-				}
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}
-
-func TestReferenceResolutionDoesNotOverwriteDescriptions(t *testing.T) {
-	specString := `{
-		"components": {
-			"schemas": {
-				"foo": {
-					"type": "number"
-				},
-				"fruits": {
-					"type": "object",
-					"properties": {
-						"guava": {
-							"type": "object",
-							"description": "Guava is a fruit",
-							"$ref": "#/components/schemas/foo"
-						},
-						"mango": {
-							"type": "object",
-							"description": "What is a mango?",
-							"$ref": "#/components/schemas/foo"
-						}
-					}
-				}
-			}
-		}
-	}`
-	spec := &Specification{}
-	reader := &OpenapiReader{
-		OpenapiSpec: spec,
-		memo:        make(map[string]jsonschema.Schema),
-	}
-	err := json.Unmarshal([]byte(specString), spec)
-	require.NoError(t, err)
-
-	fruitsSchema, err := reader.readResolvedSchema("#/components/schemas/fruits")
-	require.NoError(t, err)
-
-	fruitsSchemaJson, err := json.MarshalIndent(fruitsSchema, "		", "	")
-	require.NoError(t, err)
-
-	expected := `{
-			"type": "object",
-			"properties": {
-				"guava": {
-					"type": "number",
-					"description": "Guava is a fruit"
-				},
-				"mango": {
-					"type": "number",
-					"description": "What is a mango?"
-				}
-			}
-		}`
-
-	t.Log("[DEBUG] actual: ", string(fruitsSchemaJson))
-	t.Log("[DEBUG] expected: ", expected)
-	assert.Equal(t, expected, string(fruitsSchemaJson))
-}

bundle/schema/schema.go

@@ -1,287 +0,0 @@
-package schema
-
-import (
-	"container/list"
-	"fmt"
-	"reflect"
-	"strings"
-
-	"github.com/databricks/cli/libs/dyn/dynvar"
-	"github.com/databricks/cli/libs/jsonschema"
-)
-
-// Fields tagged "readonly" should not be emitted in the schema as they are
-// computed at runtime, and should not be assigned a value by the bundle author.
-const readonlyTag = "readonly"
-
-// Annotation for internal bundle fields that should not be exposed to customers.
-// Fields can be tagged as "internal" to remove them from the generated schema.
-const internalTag = "internal"
-
-// Annotation for bundle fields that have been deprecated.
-// Fields tagged as "deprecated" are removed/omitted from the generated schema.
-const deprecatedTag = "deprecated"
-
-// This function translates golang types into json schema. Here is the mapping
-// between json schema types and golang types
-//
-//   - GolangType               ->    Javascript type / Json Schema2
-//
-//   - bool                     ->    boolean
-//
-//   - string                   ->    string
-//
-//   - int (all variants)       ->    number
-//
-//   - float (all variants)     ->    number
-//
-//   - map[string]MyStruct      ->   { type: object, additionalProperties: {}}
-//     for details visit: https://json-schema.org/understanding-json-schema/reference/object.html#additional-properties
-//
-//   - []MyStruct               ->   {type: array, items: {}}
-//     for details visit: https://json-schema.org/understanding-json-schema/reference/array.html#items
-//
-//   - []MyStruct               ->   {type: object, properties: {}, additionalProperties: false}
-//     for details visit: https://json-schema.org/understanding-json-schema/reference/object.html#properties
-func New(golangType reflect.Type, docs *Docs) (*jsonschema.Schema, error) {
-	tracker := newTracker()
-	schema, err := safeToSchema(golangType, docs, "", tracker)
-	if err != nil {
-		return nil, tracker.errWithTrace(err.Error(), "root")
-	}
-	return schema, nil
-}
-
-func jsonSchemaType(golangType reflect.Type) (jsonschema.Type, error) {
-	switch golangType.Kind() {
-	case reflect.Bool:
-		return jsonschema.BooleanType, nil
-	case reflect.String:
-		return jsonschema.StringType, nil
-	case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64,
-		reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64,
-		reflect.Float32, reflect.Float64:
-
-		return jsonschema.NumberType, nil
-	case reflect.Struct:
-		return jsonschema.ObjectType, nil
-	case reflect.Map:
-		if golangType.Key().Kind() != reflect.String {
-			return jsonschema.InvalidType, fmt.Errorf("only strings map keys are valid. key type: %v", golangType.Key().Kind())
-		}
-		return jsonschema.ObjectType, nil
-	case reflect.Array, reflect.Slice:
-		return jsonschema.ArrayType, nil
-	default:
-		return jsonschema.InvalidType, fmt.Errorf("unhandled golang type: %s", golangType)
-	}
-}
-
-// A wrapper over toSchema function to:
-//  1. Detect cycles in the bundle config struct.
-//  2. Update tracker
-//
-// params:
-//
-//   - golangType: Golang type to generate json schema for
-//
-//   - docs: Contains documentation to be injected into the generated json schema
-//
-//   - traceId: An identifier for the current type, to trace recursive traversal.
-//     Its value is the first json tag in case of struct fields and "" in other cases
-//     like array, map or no json tags
-//
-//   - tracker: Keeps track of types / traceIds seen during recursive traversal
-func safeToSchema(golangType reflect.Type, docs *Docs, traceId string, tracker *tracker) (*jsonschema.Schema, error) {
-	// HACK to unblock CLI release (13th Feb 2024). This is temporary until proper
-	// support for recursive types is added to the schema generator. PR: https://github.com/databricks/cli/pull/1204
-	if traceId == "for_each_task" {
-		return &jsonschema.Schema{
-			Type: jsonschema.ObjectType,
-		}, nil
-	}
-
-	// WE ERROR OUT IF THERE ARE CYCLES IN THE JSON SCHEMA
-	// There are mechanisms to deal with cycles though recursive identifiers in json
-	// schema. However if we use them, we would need to make sure we are able to detect
-	// cycles where two properties (directly or indirectly) pointing to each other
-	//
-	// see: https://json-schema.org/understanding-json-schema/structuring.html#recursion
-	// for details
-	if tracker.hasCycle(golangType) {
-		return nil, fmt.Errorf("cycle detected")
-	}
-
-	tracker.push(golangType, traceId)
-	props, err := toSchema(golangType, docs, tracker)
-	if err != nil {
-		return nil, err
-	}
-	tracker.pop(golangType)
-	return props, nil
-}
-
-// This function returns all member fields of the provided type.
-// If the type has embedded (aka anonymous) fields, this function traverses
-// those in a breadth first manner
-func getStructFields(golangType reflect.Type) []reflect.StructField {
-	fields := []reflect.StructField{}
-	bfsQueue := list.New()
-
-	for i := 0; i < golangType.NumField(); i++ {
-		bfsQueue.PushBack(golangType.Field(i))
-	}
-	for bfsQueue.Len() > 0 {
-		front := bfsQueue.Front()
-		field := front.Value.(reflect.StructField)
-		bfsQueue.Remove(front)
-
-		if !field.Anonymous {
-			fields = append(fields, field)
-			continue
-		}
-
-		fieldType := field.Type
-		if fieldType.Kind() == reflect.Pointer {
-			fieldType = fieldType.Elem()
-		}
-
-		for i := 0; i < fieldType.NumField(); i++ {
-			bfsQueue.PushBack(fieldType.Field(i))
-		}
-	}
-	return fields
-}
-
-func toSchema(golangType reflect.Type, docs *Docs, tracker *tracker) (*jsonschema.Schema, error) {
-	// *Struct and Struct generate identical json schemas
-	if golangType.Kind() == reflect.Pointer {
-		return safeToSchema(golangType.Elem(), docs, "", tracker)
-	}
-	if golangType.Kind() == reflect.Interface {
-		return &jsonschema.Schema{}, nil
-	}
-
-	rootJavascriptType, err := jsonSchemaType(golangType)
-	if err != nil {
-		return nil, err
-	}
-	jsonSchema := &jsonschema.Schema{Type: rootJavascriptType}
-
-	// If the type is a non-string primitive, then we allow it to be a string
-	// provided it's a pure variable reference (ie only a single variable reference).
-	if rootJavascriptType == jsonschema.BooleanType || rootJavascriptType == jsonschema.NumberType {
-		jsonSchema = &jsonschema.Schema{
-			AnyOf: []*jsonschema.Schema{
-				{
-					Type: rootJavascriptType,
-				},
-				{
-					Type:    jsonschema.StringType,
-					Pattern: dynvar.VariableRegex,
-				},
-			},
-		}
-	}
-
-	if docs != nil {
-		jsonSchema.Description = docs.Description
-	}
-
-	// case array/slice
-	if golangType.Kind() == reflect.Array || golangType.Kind() == reflect.Slice {
-		elemGolangType := golangType.Elem()
-		elemJavascriptType, err := jsonSchemaType(elemGolangType)
-		if err != nil {
-			return nil, err
-		}
-		var childDocs *Docs
-		if docs != nil {
-			childDocs = docs.Items
-		}
-		elemProps, err := safeToSchema(elemGolangType, childDocs, "", tracker)
-		if err != nil {
-			return nil, err
-		}
-		jsonSchema.Items = &jsonschema.Schema{
-			Type:                 elemJavascriptType,
-			Properties:           elemProps.Properties,
-			AdditionalProperties: elemProps.AdditionalProperties,
-			Items:                elemProps.Items,
-			Required:             elemProps.Required,
-		}
-	}
-
-	// case map
-	if golangType.Kind() == reflect.Map {
-		if golangType.Key().Kind() != reflect.String {
-			return nil, fmt.Errorf("only string keyed maps allowed")
-		}
-		var childDocs *Docs
-		if docs != nil {
-			childDocs = docs.AdditionalProperties
-		}
-		jsonSchema.AdditionalProperties, err = safeToSchema(golangType.Elem(), childDocs, "", tracker)
-		if err != nil {
-			return nil, err
-		}
-	}
-
-	// case struct
-	if golangType.Kind() == reflect.Struct {
-		children := getStructFields(golangType)
-		properties := map[string]*jsonschema.Schema{}
-		required := []string{}
-		for _, child := range children {
-			bundleTag := child.Tag.Get("bundle")
-			// Fields marked as "readonly", "internal" or "deprecated" are skipped
-			// while generating the schema
-			if bundleTag == readonlyTag || bundleTag == internalTag || bundleTag == deprecatedTag {
-				continue
-			}
-
-			// get child json tags
-			childJsonTag := strings.Split(child.Tag.Get("json"), ",")
-			childName := childJsonTag[0]
-
-			// skip children that have no json tags, the first json tag is ""
-			// or the first json tag is "-"
-			if childName == "" || childName == "-" {
-				continue
-			}
-
-			// get docs for the child if they exist
-			var childDocs *Docs
-			if docs != nil {
-				if val, ok := docs.Properties[childName]; ok {
-					childDocs = val
-				}
-			}
-
-			// compute if the child is a required field. Determined by the
-			// presence of "omitempty" in the json tags
-			hasOmitEmptyTag := false
-			for i := 1; i < len(childJsonTag); i++ {
-				if childJsonTag[i] == "omitempty" {
-					hasOmitEmptyTag = true
-				}
-			}
-			if !hasOmitEmptyTag {
-				required = append(required, childName)
-			}
-
-			// compute Schema.Properties for the child recursively
-			fieldProps, err := safeToSchema(child.Type, childDocs, childName, tracker)
-			if err != nil {
-				return nil, err
-			}
-			properties[childName] = fieldProps
-		}
-
-		jsonSchema.AdditionalProperties = false
-		jsonSchema.Properties = properties
-		jsonSchema.Required = required
-	}
-
-	return jsonSchema, nil
-}

bundle/schema/spec.go

@@ -1,11 +0,0 @@
-package schema
-
-import "github.com/databricks/cli/libs/jsonschema"
-
-type Specification struct {
-	Components *Components `json:"components"`
-}
-
-type Components struct {
-	Schemas map[string]*jsonschema.Schema `json:"schemas,omitempty"`
-}

bundle/schema/tracker.go

@@ -1,53 +0,0 @@
-package schema
-
-import (
-	"container/list"
-	"fmt"
-)
-
-type tracker struct {
-	// Nodes encountered in current path during the recursive traversal. Used to
-	// check for cycles
-	seenNodes map[interface{}]struct{}
-
-	// List of node names encountered in order in current path during the recursive traversal.
-	// Used to hydrate errors with path to the exact node where error occured.
-	//
-	// NOTE: node and node names can be the same
-	listOfNodes *list.List
-}
-
-func newTracker() *tracker {
-	return &tracker{
-		seenNodes:   map[interface{}]struct{}{},
-		listOfNodes: list.New(),
-	}
-}
-
-func (t *tracker) errWithTrace(prefix string, initTrace string) error {
-	traceString := initTrace
-	curr := t.listOfNodes.Front()
-	for curr != nil {
-		if curr.Value.(string) != "" {
-			traceString += " -> " + curr.Value.(string)
-		}
-		curr = curr.Next()
-	}
-	return fmt.Errorf(prefix + ". traversal trace: " + traceString)
-}
-
-func (t *tracker) hasCycle(node interface{}) bool {
-	_, ok := t.seenNodes[node]
-	return ok
-}
-
-func (t *tracker) push(node interface{}, name string) {
-	t.seenNodes[node] = struct{}{}
-	t.listOfNodes.PushBack(name)
-}
-
-func (t *tracker) pop(nodeType interface{}) {
-	back := t.listOfNodes.Back()
-	t.listOfNodes.Remove(back)
-	delete(t.seenNodes, nodeType)
-}

bundle/tests/clusters/databricks.yml

@@ -0,0 +1,36 @@
+bundle:
+  name: clusters
+
+workspace:
+  host: https://acme.cloud.databricks.com/
+
+resources:
+  clusters:
+    foo:
+      cluster_name: foo
+      num_workers: 2
+      node_type_id: "i3.xlarge"
+      autoscale:
+        min_workers: 2
+        max_workers: 7
+      spark_version: "13.3.x-scala2.12"
+      spark_conf:
+        "spark.executor.memory": "2g"
+
+targets:
+  default:
+
+  development:
+    resources:
+      clusters:
+        foo:
+          cluster_name: foo-override
+          num_workers: 3
+          node_type_id: "m5.xlarge"
+          autoscale:
+            min_workers: 1
+            max_workers: 3
+          spark_version: "15.2.x-scala2.12"
+          spark_conf:
+            "spark.executor.memory": "4g"
+            "spark.executor.memory2": "4g"

bundle/tests/clusters_test.go

@@ -0,0 +1,36 @@
+package config_tests
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestClusters(t *testing.T) {
+	b := load(t, "./clusters")
+	assert.Equal(t, "clusters", b.Config.Bundle.Name)
+
+	cluster := b.Config.Resources.Clusters["foo"]
+	assert.Equal(t, "foo", cluster.ClusterName)
+	assert.Equal(t, "13.3.x-scala2.12", cluster.SparkVersion)
+	assert.Equal(t, "i3.xlarge", cluster.NodeTypeId)
+	assert.Equal(t, 2, cluster.NumWorkers)
+	assert.Equal(t, "2g", cluster.SparkConf["spark.executor.memory"])
+	assert.Equal(t, 2, cluster.Autoscale.MinWorkers)
+	assert.Equal(t, 7, cluster.Autoscale.MaxWorkers)
+}
+
+func TestClustersOverride(t *testing.T) {
+	b := loadTarget(t, "./clusters", "development")
+	assert.Equal(t, "clusters", b.Config.Bundle.Name)
+
+	cluster := b.Config.Resources.Clusters["foo"]
+	assert.Equal(t, "foo-override", cluster.ClusterName)
+	assert.Equal(t, "15.2.x-scala2.12", cluster.SparkVersion)
+	assert.Equal(t, "m5.xlarge", cluster.NodeTypeId)
+	assert.Equal(t, 3, cluster.NumWorkers)
+	assert.Equal(t, "4g", cluster.SparkConf["spark.executor.memory"])
+	assert.Equal(t, "4g", cluster.SparkConf["spark.executor.memory2"])
+	assert.Equal(t, 1, cluster.Autoscale.MinWorkers)
+	assert.Equal(t, 3, cluster.Autoscale.MaxWorkers)
+}

bundle/tests/complex_variables_test.go

@@ -88,3 +88,21 @@ func TestComplexVariablesOverrideWithMultipleFiles(t *testing.T) {
 		require.Equalf(t, "false", cluster.NewCluster.SparkConf["spark.speculation"], "cluster: %v", cluster.JobClusterKey)
 	}
 }
+
+func TestComplexVariablesOverrideWithFullSyntax(t *testing.T) {
+	b, diags := loadTargetWithDiags("variables/complex", "dev")
+	require.Empty(t, diags)
+
+	diags = bundle.Apply(context.Background(), b, bundle.Seq(
+		mutator.SetVariables(),
+		mutator.ResolveVariableReferencesInComplexVariables(),
+		mutator.ResolveVariableReferences(
+			"variables",
+		),
+	))
+	require.NoError(t, diags.Error())
+	require.Empty(t, diags)
+
+	complexvar := b.Config.Variables["complexvar"].Value
+	require.Equal(t, map[string]interface{}{"key1": "1", "key2": "2", "key3": "3"}, complexvar)
+}

bundle/tests/variables/complex/databricks.yml

@@ -35,6 +35,13 @@ variables:
       - jar: "/path/to/jar"
       - egg: "/path/to/egg"
       - whl: "/path/to/whl"
+  complexvar:
+    type: complex
+    description: "A complex variable"
+    default:
+      key1: "value1"
+      key2: "value2"
+      key3: "value3"
 
 
 targets:
@@ -49,3 +56,9 @@ targets:
         spark_conf:
           spark.speculation: false
           spark.databricks.delta.retentionDurationCheck.enabled: false
+      complexvar:
+        type: complex
+        default:
+          key1: "1"
+          key2: "2"
+          key3: "3"

cmd/bundle/deploy.go

@@ -10,6 +10,7 @@ import (
 	"github.com/databricks/cli/cmd/bundle/utils"
 	"github.com/databricks/cli/cmd/root"
 	"github.com/databricks/cli/libs/diag"
+	"github.com/databricks/cli/libs/sync"
 	"github.com/spf13/cobra"
 )
 
@@ -23,13 +24,19 @@ func newDeployCommand() *cobra.Command {
 	var force bool
 	var forceLock bool
 	var failOnActiveRuns bool
-	var computeID string
+	var clusterId string
 	var autoApprove bool
+	var verbose bool
 	cmd.Flags().BoolVar(&force, "force", false, "Force-override Git branch validation.")
 	cmd.Flags().BoolVar(&forceLock, "force-lock", false, "Force acquisition of deployment lock.")
 	cmd.Flags().BoolVar(&failOnActiveRuns, "fail-on-active-runs", false, "Fail if there are running jobs or pipelines in the deployment.")
-	cmd.Flags().StringVarP(&computeID, "compute-id", "c", "", "Override compute in the deployment with the given compute ID.")
+	cmd.Flags().StringVar(&clusterId, "compute-id", "", "Override cluster in the deployment with the given compute ID.")
+	cmd.Flags().StringVarP(&clusterId, "cluster-id", "c", "", "Override cluster in the deployment with the given cluster ID.")
 	cmd.Flags().BoolVar(&autoApprove, "auto-approve", false, "Skip interactive approvals that might be required for deployment.")
+	cmd.Flags().MarkDeprecated("compute-id", "use --cluster-id instead")
+	cmd.Flags().BoolVar(&verbose, "verbose", false, "Enable verbose output.")
+	// Verbose flag currently only affects file sync output, it's used by the vscode extension
+	cmd.Flags().MarkHidden("verbose")
 
 	cmd.RunE = func(cmd *cobra.Command, args []string) error {
 		ctx := cmd.Context()
@@ -42,7 +49,10 @@ func newDeployCommand() *cobra.Command {
 				b.AutoApprove = autoApprove
 
 				if cmd.Flag("compute-id").Changed {
-					b.Config.Bundle.ComputeID = computeID
+					b.Config.Bundle.ClusterId = clusterId
+				}
+				if cmd.Flag("cluster-id").Changed {
+					b.Config.Bundle.ClusterId = clusterId
 				}
 				if cmd.Flag("fail-on-active-runs").Changed {
 					b.Config.Bundle.Deployment.FailOnActiveRuns = failOnActiveRuns
@@ -51,11 +61,18 @@ func newDeployCommand() *cobra.Command {
 				return nil
 			})
 
+			var outputHandler sync.OutputHandler
+			if verbose {
+				outputHandler = func(ctx context.Context, c <-chan sync.Event) {
+					sync.TextOutput(ctx, c, cmd.OutOrStdout())
+				}
+			}
+
 			diags = diags.Extend(
 				bundle.Apply(ctx, b, bundle.Seq(
 					phases.Initialize(),
 					phases.Build(),
-					phases.Deploy(),
+					phases.Deploy(outputHandler),
 				)),
 			)
 		}

cmd/bundle/generate/generate_test.go

@@ -152,6 +152,12 @@ func TestGenerateJobCommand(t *testing.T) {
 					},
 				},
 			},
+			Parameters: []jobs.JobParameterDefinition{
+				{
+					Name:    "empty",
+					Default: "",
+				},
+			},
 		},
 	}, nil)
 
@@ -198,6 +204,9 @@ func TestGenerateJobCommand(t *testing.T) {
         - task_key: notebook_task
           notebook_task:
             notebook_path: %s
+      parameters:
+        - name: empty
+          default: ""
 `, filepath.Join("..", "src", "notebook.py")), string(data))
 
 	data, err = os.ReadFile(filepath.Join(srcDir, "notebook.py"))

cmd/bundle/run.go

@@ -55,13 +55,7 @@ task or a Python wheel task, the second example applies.
 			return diags.Error()
 		}
 
-		diags = bundle.Apply(ctx, b, bundle.Seq(
-			phases.Initialize(),
-			terraform.Interpolate(),
-			terraform.Write(),
-			terraform.StatePull(),
-			terraform.Load(terraform.ErrorOnEmptyState),
-		))
+		diags = bundle.Apply(ctx, b, phases.Initialize())
 		if err := diags.Error(); err != nil {
 			return err
 		}
@@ -84,6 +78,16 @@ task or a Python wheel task, the second example applies.
 			return fmt.Errorf("expected a KEY of the resource to run")
 		}
 
+		diags = bundle.Apply(ctx, b, bundle.Seq(
+			terraform.Interpolate(),
+			terraform.Write(),
+			terraform.StatePull(),
+			terraform.Load(terraform.ErrorOnEmptyState),
+		))
+		if err := diags.Error(); err != nil {
+			return err
+		}
+
 		runner, err := run.Find(b, args[0])
 		if err != nil {
 			return err