diff --git a/.codegen/_openapi_sha b/.codegen/_openapi_sha index 4ceeab3d..ffd6f58d 100644 --- a/.codegen/_openapi_sha +++ b/.codegen/_openapi_sha @@ -1 +1 @@ -d05898328669a3f8ab0c2ecee37db2673d3ea3f7 \ No newline at end of file +6f6b1371e640f2dfeba72d365ac566368656f6b6 \ No newline at end of file diff --git a/.gitattributes b/.gitattributes index f35c4f81..2470eb33 100755 --- a/.gitattributes +++ b/.gitattributes @@ -6,6 +6,7 @@ cmd/account/cmd.go linguist-generated=true cmd/account/credentials/credentials.go linguist-generated=true cmd/account/csp-enablement-account/csp-enablement-account.go linguist-generated=true cmd/account/custom-app-integration/custom-app-integration.go linguist-generated=true +cmd/account/disable-legacy-features/disable-legacy-features.go linguist-generated=true cmd/account/encryption-keys/encryption-keys.go linguist-generated=true cmd/account/esm-enablement-account/esm-enablement-account.go linguist-generated=true cmd/account/groups/groups.go linguist-generated=true @@ -52,6 +53,7 @@ cmd/workspace/dashboard-widgets/dashboard-widgets.go linguist-generated=true cmd/workspace/dashboards/dashboards.go linguist-generated=true cmd/workspace/data-sources/data-sources.go linguist-generated=true cmd/workspace/default-namespace/default-namespace.go linguist-generated=true +cmd/workspace/disable-legacy-access/disable-legacy-access.go linguist-generated=true cmd/workspace/enhanced-security-monitoring/enhanced-security-monitoring.go linguist-generated=true cmd/workspace/experiments/experiments.go linguist-generated=true cmd/workspace/external-locations/external-locations.go linguist-generated=true @@ -108,6 +110,7 @@ cmd/workspace/storage-credentials/storage-credentials.go linguist-generated=true cmd/workspace/system-schemas/system-schemas.go linguist-generated=true cmd/workspace/table-constraints/table-constraints.go linguist-generated=true cmd/workspace/tables/tables.go linguist-generated=true +cmd/workspace/temporary-table-credentials/temporary-table-credentials.go linguist-generated=true cmd/workspace/token-management/token-management.go linguist-generated=true cmd/workspace/tokens/tokens.go linguist-generated=true cmd/workspace/users/users.go linguist-generated=true diff --git a/.github/workflows/push.yml b/.github/workflows/push.yml index 02bf7378..ee60da9d 100644 --- a/.github/workflows/push.yml +++ b/.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 @@ -100,7 +100,7 @@ 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 diff --git a/.github/workflows/release-snapshot.yml b/.github/workflows/release-snapshot.yml index defd1c53..6a601a5f 100644 --- a/.github/workflows/release-snapshot.yml +++ b/.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 diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml index 531fb39b..f9742a19 100644 --- a/.github/workflows/release.yml +++ b/.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 diff --git a/CHANGELOG.md b/CHANGELOG.md index d6383125..4f5f68ac 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,72 @@ # Version changelog +## [Release] Release v0.229.0 + +Bundles: + * Added support for creating all-purpose clusters ([#1698](https://github.com/databricks/cli/pull/1698)). + * Reduce time until the prompt is shown for bundle run ([#1727](https://github.com/databricks/cli/pull/1727)). + * Use Unity Catalog for pipelines in the default-python template ([#1766](https://github.com/databricks/cli/pull/1766)). + * Add verbose flag to the "bundle deploy" command ([#1774](https://github.com/databricks/cli/pull/1774)). + * Fixed full variable override detection ([#1787](https://github.com/databricks/cli/pull/1787)). + * Add sub-extension to resource files in built-in templates ([#1777](https://github.com/databricks/cli/pull/1777)). + * Fix panic in `apply_presets.go` ([#1796](https://github.com/databricks/cli/pull/1796)). + +Internal: + * Assert tokens are redacted in origin URL when username is not specified ([#1785](https://github.com/databricks/cli/pull/1785)). + * Refactor jobs path translation ([#1782](https://github.com/databricks/cli/pull/1782)). + * Add JobTaskClusterSpec validate mutator ([#1784](https://github.com/databricks/cli/pull/1784)). + * Pin Go toolchain to 1.22.7 ([#1790](https://github.com/databricks/cli/pull/1790)). + * Modify SetLocation test utility to take full locations as argument ([#1788](https://github.com/databricks/cli/pull/1788)). + * Simplified isFullVariableOverrideDef implementation ([#1791](https://github.com/databricks/cli/pull/1791)). + * Sort tasks by `task_key` before generating the Terraform configuration ([#1776](https://github.com/databricks/cli/pull/1776)). + * Trim trailing whitespace ([#1794](https://github.com/databricks/cli/pull/1794)). + * Move trampoline code into trampoline package ([#1793](https://github.com/databricks/cli/pull/1793)). + * Rename `RootPath` -> `BundleRootPath` ([#1792](https://github.com/databricks/cli/pull/1792)). + +API Changes: + * Changed `databricks apps delete` command to return . + * Changed `databricks apps deploy` command with new required argument order. + * Changed `databricks apps start` command to return . + * Changed `databricks apps stop` command to return . + * Added `databricks temporary-table-credentials` command group. + * Added `databricks serving-endpoints put-ai-gateway` command. + * Added `databricks disable-legacy-access` command group. + * Added `databricks account disable-legacy-features` command group. + +OpenAPI commit 6f6b1371e640f2dfeba72d365ac566368656f6b6 (2024-09-19) +Dependency updates: + * Upgrade to Go SDK 0.47.0 ([#1799](https://github.com/databricks/cli/pull/1799)). + * Upgrade to TF provider 1.52 ([#1781](https://github.com/databricks/cli/pull/1781)). + * Bump golang.org/x/mod from 0.20.0 to 0.21.0 ([#1758](https://github.com/databricks/cli/pull/1758)). + * Bump github.com/hashicorp/hc-install from 0.7.0 to 0.9.0 ([#1772](https://github.com/databricks/cli/pull/1772)). + +## [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: diff --git a/bundle/artifacts/expand_globs_test.go b/bundle/artifacts/expand_globs_test.go index c9c47844..dc7c77de 100644 --- a/bundle/artifacts/expand_globs_test.go +++ b/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" ) @@ -23,7 +24,7 @@ func TestExpandGlobs_Nominal(t *testing.T) { testutil.Touch(t, tmpDir, "bc.txt") b := &bundle.Bundle{ - RootPath: tmpDir, + BundleRootPath: tmpDir, Config: config.Root{ Artifacts: config.Artifacts{ "test": { @@ -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( @@ -62,7 +63,7 @@ func TestExpandGlobs_InvalidPattern(t *testing.T) { tmpDir := t.TempDir() b := &bundle.Bundle{ - RootPath: tmpDir, + BundleRootPath: tmpDir, Config: config.Root{ Artifacts: config.Artifacts{ "test": { @@ -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( @@ -110,7 +111,7 @@ func TestExpandGlobs_NoMatches(t *testing.T) { testutil.Touch(t, tmpDir, "b2.txt") b := &bundle.Bundle{ - RootPath: tmpDir, + BundleRootPath: tmpDir, Config: config.Root{ Artifacts: config.Artifacts{ "test": { @@ -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( diff --git a/bundle/artifacts/prepare.go b/bundle/artifacts/prepare.go index fb61ed9e..91e0bd09 100644 --- a/bundle/artifacts/prepare.go +++ b/bundle/artifacts/prepare.go @@ -47,7 +47,7 @@ func (m *prepare) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics // If artifact path is not provided, use bundle root dir if artifact.Path == "" { - artifact.Path = b.RootPath + artifact.Path = b.BundleRootPath } if !filepath.IsAbs(artifact.Path) { diff --git a/bundle/artifacts/whl/autodetect.go b/bundle/artifacts/whl/autodetect.go index 1601767f..88dc742c 100644 --- a/bundle/artifacts/whl/autodetect.go +++ b/bundle/artifacts/whl/autodetect.go @@ -35,21 +35,21 @@ func (m *detectPkg) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostic log.Infof(ctx, "Detecting Python wheel project...") // checking if there is setup.py in the bundle root - setupPy := filepath.Join(b.RootPath, "setup.py") + setupPy := filepath.Join(b.BundleRootPath, "setup.py") _, err := os.Stat(setupPy) if err != nil { log.Infof(ctx, "No Python wheel project found at bundle root folder") return nil } - log.Infof(ctx, fmt.Sprintf("Found Python wheel project at %s", b.RootPath)) + log.Infof(ctx, fmt.Sprintf("Found Python wheel project at %s", b.BundleRootPath)) module := extractModuleName(setupPy) if b.Config.Artifacts == nil { b.Config.Artifacts = make(map[string]*config.Artifact) } - pkgPath, err := filepath.Abs(b.RootPath) + pkgPath, err := filepath.Abs(b.BundleRootPath) if err != nil { return diag.FromErr(err) } diff --git a/bundle/bundle.go b/bundle/bundle.go index 8b5ff976..85625568 100644 --- a/bundle/bundle.go +++ b/bundle/bundle.go @@ -31,22 +31,26 @@ import ( const internalFolder = ".internal" type Bundle struct { - // RootPath contains the directory path to the root of the bundle. + // BundleRootPath is the local path to the root directory of the bundle. // It is set when we instantiate a new bundle instance. - RootPath string + BundleRootPath string - // BundleRoot is a virtual filesystem path to the root of the bundle. + // BundleRoot is a virtual filesystem path to [BundleRootPath]. // Exclusively use this field for filesystem operations. BundleRoot vfs.Path - // SyncRoot is a virtual filesystem path to the root directory of the files that are synchronized to the workspace. - // It can be an ancestor to [BundleRoot], but not a descendant; that is, [SyncRoot] must contain [BundleRoot]. - SyncRoot vfs.Path - // SyncRootPath is the local path to the root directory of files that are synchronized to the workspace. - // It is equal to `SyncRoot.Native()` and included as dedicated field for convenient access. + // By default, it is the same as [BundleRootPath]. + // If it is different, it must be an ancestor to [BundleRootPath]. + // That is, [SyncRootPath] must contain [BundleRootPath]. SyncRootPath string + // SyncRoot is a virtual filesystem path to [SyncRootPath]. + // Exclusively use this field for filesystem operations. + SyncRoot vfs.Path + + // Config contains the bundle configuration. + // It is loaded from the bundle configuration files and mutators may update it. Config config.Root // Metadata about the bundle deployment. This is the interface Databricks services @@ -84,14 +88,14 @@ type Bundle struct { func Load(ctx context.Context, path string) (*Bundle, error) { b := &Bundle{ - RootPath: filepath.Clean(path), - BundleRoot: vfs.MustNew(path), + BundleRootPath: filepath.Clean(path), + BundleRoot: vfs.MustNew(path), } configFile, err := config.FileNames.FindInPath(path) if err != nil { return nil, err } - log.Debugf(ctx, "Found bundle root at %s (file %s)", b.RootPath, configFile) + log.Debugf(ctx, "Found bundle root at %s (file %s)", b.BundleRootPath, configFile) return b, nil } @@ -160,7 +164,7 @@ func (b *Bundle) CacheDir(ctx context.Context, paths ...string) (string, error) if !exists || cacheDirName == "" { cacheDirName = filepath.Join( // Anchor at bundle root directory. - b.RootPath, + b.BundleRootPath, // Static cache directory. ".databricks", "bundle", @@ -212,7 +216,7 @@ func (b *Bundle) GetSyncIncludePatterns(ctx context.Context) ([]string, error) { if err != nil { return nil, err } - internalDirRel, err := filepath.Rel(b.RootPath, internalDir) + internalDirRel, err := filepath.Rel(b.BundleRootPath, internalDir) if err != nil { return nil, err } diff --git a/bundle/bundle_read_only.go b/bundle/bundle_read_only.go index 74b9d94d..ceab95c0 100644 --- a/bundle/bundle_read_only.go +++ b/bundle/bundle_read_only.go @@ -21,7 +21,7 @@ func (r ReadOnlyBundle) Config() config.Root { } func (r ReadOnlyBundle) RootPath() string { - return r.b.RootPath + return r.b.BundleRootPath } func (r ReadOnlyBundle) BundleRoot() vfs.Path { diff --git a/bundle/bundle_test.go b/bundle/bundle_test.go index a29aa024..1c310235 100644 --- a/bundle/bundle_test.go +++ b/bundle/bundle_test.go @@ -79,7 +79,7 @@ func TestBundleMustLoadSuccess(t *testing.T) { t.Setenv(env.RootVariable, "./tests/basic") b, err := MustLoad(context.Background()) require.NoError(t, err) - assert.Equal(t, "tests/basic", filepath.ToSlash(b.RootPath)) + assert.Equal(t, "tests/basic", filepath.ToSlash(b.BundleRootPath)) } func TestBundleMustLoadFailureWithEnv(t *testing.T) { @@ -98,7 +98,7 @@ func TestBundleTryLoadSuccess(t *testing.T) { t.Setenv(env.RootVariable, "./tests/basic") b, err := TryLoad(context.Background()) require.NoError(t, err) - assert.Equal(t, "tests/basic", filepath.ToSlash(b.RootPath)) + assert.Equal(t, "tests/basic", filepath.ToSlash(b.BundleRootPath)) } func TestBundleTryLoadFailureWithEnv(t *testing.T) { diff --git a/bundle/config/bundle.go b/bundle/config/bundle.go index 78648dfd..f533c4d1 100644 --- a/bundle/config/bundle.go +++ b/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"` diff --git a/bundle/config/loader/entry_point.go b/bundle/config/loader/entry_point.go index 2c73a582..d476cb22 100644 --- a/bundle/config/loader/entry_point.go +++ b/bundle/config/loader/entry_point.go @@ -20,7 +20,7 @@ func (m *entryPoint) Name() string { } func (m *entryPoint) Apply(_ context.Context, b *bundle.Bundle) diag.Diagnostics { - path, err := config.FileNames.FindInPath(b.RootPath) + path, err := config.FileNames.FindInPath(b.BundleRootPath) if err != nil { return diag.FromErr(err) } diff --git a/bundle/config/loader/entry_point_test.go b/bundle/config/loader/entry_point_test.go index 80271f0b..406b9b67 100644 --- a/bundle/config/loader/entry_point_test.go +++ b/bundle/config/loader/entry_point_test.go @@ -18,7 +18,7 @@ func TestEntryPointNoRootPath(t *testing.T) { func TestEntryPoint(t *testing.T) { b := &bundle.Bundle{ - RootPath: "testdata", + BundleRootPath: "testdata", } diags := bundle.Apply(context.Background(), b, loader.EntryPoint()) require.NoError(t, diags.Error()) diff --git a/bundle/config/loader/process_include_test.go b/bundle/config/loader/process_include_test.go index da4da9ff..2ccd84b3 100644 --- a/bundle/config/loader/process_include_test.go +++ b/bundle/config/loader/process_include_test.go @@ -14,7 +14,7 @@ import ( func TestProcessInclude(t *testing.T) { b := &bundle.Bundle{ - RootPath: "testdata", + BundleRootPath: "testdata", Config: config.Root{ Workspace: config.Workspace{ Host: "foo", @@ -22,7 +22,7 @@ func TestProcessInclude(t *testing.T) { }, } - m := loader.ProcessInclude(filepath.Join(b.RootPath, "host.yml"), "host.yml") + m := loader.ProcessInclude(filepath.Join(b.BundleRootPath, "host.yml"), "host.yml") assert.Equal(t, "ProcessInclude(host.yml)", m.Name()) // Assert the host value prior to applying the mutator diff --git a/bundle/config/loader/process_root_includes.go b/bundle/config/loader/process_root_includes.go index 25f284fd..c14fb7ce 100644 --- a/bundle/config/loader/process_root_includes.go +++ b/bundle/config/loader/process_root_includes.go @@ -47,7 +47,7 @@ func (m *processRootIncludes) Apply(ctx context.Context, b *bundle.Bundle) diag. } // Anchor includes to the bundle root path. - matches, err := filepath.Glob(filepath.Join(b.RootPath, entry)) + matches, err := filepath.Glob(filepath.Join(b.BundleRootPath, entry)) if err != nil { return diag.FromErr(err) } @@ -61,7 +61,7 @@ func (m *processRootIncludes) Apply(ctx context.Context, b *bundle.Bundle) diag. // Filter matches to ones we haven't seen yet. var includes []string for _, match := range matches { - rel, err := filepath.Rel(b.RootPath, match) + rel, err := filepath.Rel(b.BundleRootPath, match) if err != nil { return diag.FromErr(err) } @@ -76,7 +76,7 @@ func (m *processRootIncludes) Apply(ctx context.Context, b *bundle.Bundle) diag. slices.Sort(includes) files = append(files, includes...) for _, include := range includes { - out = append(out, ProcessInclude(filepath.Join(b.RootPath, include), include)) + out = append(out, ProcessInclude(filepath.Join(b.BundleRootPath, include), include)) } } diff --git a/bundle/config/loader/process_root_includes_test.go b/bundle/config/loader/process_root_includes_test.go index 737dbbef..27ff9b05 100644 --- a/bundle/config/loader/process_root_includes_test.go +++ b/bundle/config/loader/process_root_includes_test.go @@ -15,7 +15,7 @@ import ( func TestProcessRootIncludesEmpty(t *testing.T) { b := &bundle.Bundle{ - RootPath: ".", + BundleRootPath: ".", } diags := bundle.Apply(context.Background(), b, loader.ProcessRootIncludes()) require.NoError(t, diags.Error()) @@ -30,7 +30,7 @@ func TestProcessRootIncludesAbs(t *testing.T) { } b := &bundle.Bundle{ - RootPath: ".", + BundleRootPath: ".", Config: config.Root{ Include: []string{ "/tmp/*.yml", @@ -44,7 +44,7 @@ func TestProcessRootIncludesAbs(t *testing.T) { func TestProcessRootIncludesSingleGlob(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Include: []string{ "*.yml", @@ -52,9 +52,9 @@ func TestProcessRootIncludesSingleGlob(t *testing.T) { }, } - testutil.Touch(t, b.RootPath, "databricks.yml") - testutil.Touch(t, b.RootPath, "a.yml") - testutil.Touch(t, b.RootPath, "b.yml") + testutil.Touch(t, b.BundleRootPath, "databricks.yml") + testutil.Touch(t, b.BundleRootPath, "a.yml") + testutil.Touch(t, b.BundleRootPath, "b.yml") diags := bundle.Apply(context.Background(), b, loader.ProcessRootIncludes()) require.NoError(t, diags.Error()) @@ -63,7 +63,7 @@ func TestProcessRootIncludesSingleGlob(t *testing.T) { func TestProcessRootIncludesMultiGlob(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Include: []string{ "a*.yml", @@ -72,8 +72,8 @@ func TestProcessRootIncludesMultiGlob(t *testing.T) { }, } - testutil.Touch(t, b.RootPath, "a1.yml") - testutil.Touch(t, b.RootPath, "b1.yml") + testutil.Touch(t, b.BundleRootPath, "a1.yml") + testutil.Touch(t, b.BundleRootPath, "b1.yml") diags := bundle.Apply(context.Background(), b, loader.ProcessRootIncludes()) require.NoError(t, diags.Error()) @@ -82,7 +82,7 @@ func TestProcessRootIncludesMultiGlob(t *testing.T) { func TestProcessRootIncludesRemoveDups(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Include: []string{ "*.yml", @@ -91,7 +91,7 @@ func TestProcessRootIncludesRemoveDups(t *testing.T) { }, } - testutil.Touch(t, b.RootPath, "a.yml") + testutil.Touch(t, b.BundleRootPath, "a.yml") diags := bundle.Apply(context.Background(), b, loader.ProcessRootIncludes()) require.NoError(t, diags.Error()) @@ -100,7 +100,7 @@ func TestProcessRootIncludesRemoveDups(t *testing.T) { func TestProcessRootIncludesNotExists(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Include: []string{ "notexist.yml", diff --git a/bundle/config/mutator/apply_presets.go b/bundle/config/mutator/apply_presets.go index 28d015c1..1fd49206 100644 --- a/bundle/config/mutator/apply_presets.go +++ b/bundle/config/mutator/apply_presets.go @@ -35,8 +35,10 @@ func (m *applyPresets) Name() string { } func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { + var diags diag.Diagnostics + if d := validatePauseStatus(b); d != nil { - return d + diags = diags.Extend(d) } r := b.Config.Resources @@ -45,7 +47,11 @@ func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnos tags := toTagArray(t.Tags) // Jobs presets: Prefix, Tags, JobsMaxConcurrentRuns, TriggerPauseStatus - for _, j := range r.Jobs { + for key, j := range r.Jobs { + if j.JobSettings == nil { + diags = diags.Extend(diag.Errorf("job %s is not defined", key)) + continue + } j.Name = prefix + j.Name if j.Tags == nil { j.Tags = make(map[string]string) @@ -77,20 +83,27 @@ func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnos } // Pipelines presets: Prefix, PipelinesDevelopment - for i := range r.Pipelines { - r.Pipelines[i].Name = prefix + r.Pipelines[i].Name + for key, p := range r.Pipelines { + if p.PipelineSpec == nil { + diags = diags.Extend(diag.Errorf("pipeline %s is not defined", key)) + continue + } + p.Name = prefix + p.Name if config.IsExplicitlyEnabled(t.PipelinesDevelopment) { - r.Pipelines[i].Development = true + p.Development = true } if t.TriggerPauseStatus == config.Paused { - r.Pipelines[i].Continuous = false + p.Continuous = false } - // As of 2024-06, pipelines don't yet support tags } // Models presets: Prefix, Tags - for _, m := range r.Models { + for key, m := range r.Models { + if m.Model == nil { + diags = diags.Extend(diag.Errorf("model %s is not defined", key)) + continue + } m.Name = prefix + m.Name for _, t := range tags { exists := slices.ContainsFunc(m.Tags, func(modelTag ml.ModelTag) bool { @@ -104,7 +117,11 @@ func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnos } // Experiments presets: Prefix, Tags - for _, e := range r.Experiments { + for key, e := range r.Experiments { + if e.Experiment == nil { + diags = diags.Extend(diag.Errorf("experiment %s is not defined", key)) + continue + } filepath := e.Name dir := path.Dir(filepath) base := path.Base(filepath) @@ -128,39 +145,74 @@ func (m *applyPresets) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnos } // Model serving endpoint presets: Prefix - for i := range r.ModelServingEndpoints { - r.ModelServingEndpoints[i].Name = normalizePrefix(prefix) + r.ModelServingEndpoints[i].Name + for key, e := range r.ModelServingEndpoints { + if e.CreateServingEndpoint == nil { + diags = diags.Extend(diag.Errorf("model serving endpoint %s is not defined", key)) + continue + } + e.Name = normalizePrefix(prefix) + e.Name // As of 2024-06, model serving endpoints don't yet support tags } // Registered models presets: Prefix - for i := range r.RegisteredModels { - r.RegisteredModels[i].Name = normalizePrefix(prefix) + r.RegisteredModels[i].Name + for key, m := range r.RegisteredModels { + if m.CreateRegisteredModelRequest == nil { + diags = diags.Extend(diag.Errorf("registered model %s is not defined", key)) + continue + } + m.Name = normalizePrefix(prefix) + m.Name // As of 2024-06, registered models don't yet support tags } - // Quality monitors presets: Prefix + // Quality monitors presets: Schedule if t.TriggerPauseStatus == config.Paused { - for i := range r.QualityMonitors { + for key, q := range r.QualityMonitors { + if q.CreateMonitor == nil { + diags = diags.Extend(diag.Errorf("quality monitor %s is not defined", key)) + continue + } // Remove all schedules from monitors, since they don't support pausing/unpausing. // Quality monitors might support the "pause" property in the future, so at the // CLI level we do respect that property if it is set to "unpaused." - if r.QualityMonitors[i].Schedule != nil && r.QualityMonitors[i].Schedule.PauseStatus != catalog.MonitorCronSchedulePauseStatusUnpaused { - r.QualityMonitors[i].Schedule = nil + if q.Schedule != nil && q.Schedule.PauseStatus != catalog.MonitorCronSchedulePauseStatusUnpaused { + q.Schedule = nil } } } // Schemas: Prefix - for i := range r.Schemas { - r.Schemas[i].Name = normalizePrefix(prefix) + r.Schemas[i].Name + for key, s := range r.Schemas { + if s.CreateSchema == nil { + diags = diags.Extend(diag.Errorf("schema %s is not defined", key)) + continue + } + s.Name = normalizePrefix(prefix) + s.Name // HTTP API for schemas doesn't yet support tags. It's only supported in // the Databricks UI and via the SQL API. } - return nil + // Clusters: Prefix, Tags + for key, c := range r.Clusters { + if c.ClusterSpec == nil { + diags = diags.Extend(diag.Errorf("cluster %s is not defined", key)) + continue + } + 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 diags } func validatePauseStatus(b *bundle.Bundle) diag.Diagnostics { diff --git a/bundle/config/mutator/apply_presets_test.go b/bundle/config/mutator/apply_presets_test.go index ab2478ae..24295da4 100644 --- a/bundle/config/mutator/apply_presets_test.go +++ b/bundle/config/mutator/apply_presets_test.go @@ -251,3 +251,116 @@ func TestApplyPresetsJobsMaxConcurrentRuns(t *testing.T) { }) } } + +func TestApplyPresetsPrefixWithoutJobSettings(t *testing.T) { + b := &bundle.Bundle{ + Config: config.Root{ + Resources: config.Resources{ + Jobs: map[string]*resources.Job{ + "job1": {}, // no jobsettings inside + }, + }, + Presets: config.Presets{ + NamePrefix: "prefix-", + }, + }, + } + + ctx := context.Background() + diags := bundle.Apply(ctx, b, mutator.ApplyPresets()) + + require.ErrorContains(t, diags.Error(), "job job1 is not defined") +} + +func TestApplyPresetsResourceNotDefined(t *testing.T) { + tests := []struct { + resources config.Resources + error string + }{ + { + resources: config.Resources{ + Jobs: map[string]*resources.Job{ + "job1": {}, // no jobsettings inside + }, + }, + error: "job job1 is not defined", + }, + { + resources: config.Resources{ + Pipelines: map[string]*resources.Pipeline{ + "pipeline1": {}, // no pipelinespec inside + }, + }, + error: "pipeline pipeline1 is not defined", + }, + { + resources: config.Resources{ + Models: map[string]*resources.MlflowModel{ + "model1": {}, // no model inside + }, + }, + error: "model model1 is not defined", + }, + { + resources: config.Resources{ + Experiments: map[string]*resources.MlflowExperiment{ + "experiment1": {}, // no experiment inside + }, + }, + error: "experiment experiment1 is not defined", + }, + { + resources: config.Resources{ + ModelServingEndpoints: map[string]*resources.ModelServingEndpoint{ + "endpoint1": {}, // no CreateServingEndpoint inside + }, + RegisteredModels: map[string]*resources.RegisteredModel{ + "model1": {}, // no CreateRegisteredModelRequest inside + }, + }, + error: "model serving endpoint endpoint1 is not defined", + }, + { + resources: config.Resources{ + QualityMonitors: map[string]*resources.QualityMonitor{ + "monitor1": {}, // no CreateMonitor inside + }, + }, + error: "quality monitor monitor1 is not defined", + }, + { + resources: config.Resources{ + Schemas: map[string]*resources.Schema{ + "schema1": {}, // no CreateSchema inside + }, + }, + error: "schema schema1 is not defined", + }, + { + resources: config.Resources{ + Clusters: map[string]*resources.Cluster{ + "cluster1": {}, // no ClusterSpec inside + }, + }, + error: "cluster cluster1 is not defined", + }, + } + + for _, tt := range tests { + t.Run(tt.error, func(t *testing.T) { + b := &bundle.Bundle{ + Config: config.Root{ + Resources: tt.resources, + Presets: config.Presets{ + TriggerPauseStatus: config.Paused, + }, + }, + } + + ctx := context.Background() + diags := bundle.Apply(ctx, b, mutator.ApplyPresets()) + + require.ErrorContains(t, diags.Error(), tt.error) + }) + } +} diff --git a/bundle/config/mutator/compute_id_compat.go b/bundle/config/mutator/compute_id_compat.go new file mode 100644 index 00000000..3afe02e9 --- /dev/null +++ b/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 +} diff --git a/bundle/config/mutator/compute_id_compate_test.go b/bundle/config/mutator/compute_id_compate_test.go new file mode 100644 index 00000000..e59d37e3 --- /dev/null +++ b/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) +} diff --git a/bundle/config/mutator/default_workspace_paths.go b/bundle/config/mutator/default_workspace_paths.go index 71e562b5..02a1ddb3 100644 --- a/bundle/config/mutator/default_workspace_paths.go +++ b/bundle/config/mutator/default_workspace_paths.go @@ -29,6 +29,10 @@ func (m *defineDefaultWorkspacePaths) Apply(ctx context.Context, b *bundle.Bundl b.Config.Workspace.FilePath = path.Join(root, "files") } + if b.Config.Workspace.ResourcePath == "" { + b.Config.Workspace.ResourcePath = path.Join(root, "resources") + } + if b.Config.Workspace.ArtifactPath == "" { b.Config.Workspace.ArtifactPath = path.Join(root, "artifacts") } diff --git a/bundle/config/mutator/default_workspace_paths_test.go b/bundle/config/mutator/default_workspace_paths_test.go index 0ba20ea2..6779c373 100644 --- a/bundle/config/mutator/default_workspace_paths_test.go +++ b/bundle/config/mutator/default_workspace_paths_test.go @@ -22,6 +22,7 @@ func TestDefineDefaultWorkspacePaths(t *testing.T) { diags := bundle.Apply(context.Background(), b, mutator.DefineDefaultWorkspacePaths()) require.NoError(t, diags.Error()) assert.Equal(t, "/files", b.Config.Workspace.FilePath) + assert.Equal(t, "/resources", b.Config.Workspace.ResourcePath) assert.Equal(t, "/artifacts", b.Config.Workspace.ArtifactPath) assert.Equal(t, "/state", b.Config.Workspace.StatePath) } @@ -32,6 +33,7 @@ func TestDefineDefaultWorkspacePathsAlreadySet(t *testing.T) { Workspace: config.Workspace{ RootPath: "/", FilePath: "/foo/bar", + ResourcePath: "/foo/bar", ArtifactPath: "/foo/bar", StatePath: "/foo/bar", }, @@ -40,6 +42,7 @@ func TestDefineDefaultWorkspacePathsAlreadySet(t *testing.T) { diags := bundle.Apply(context.Background(), b, mutator.DefineDefaultWorkspacePaths()) require.NoError(t, diags.Error()) assert.Equal(t, "/foo/bar", b.Config.Workspace.FilePath) + assert.Equal(t, "/foo/bar", b.Config.Workspace.ResourcePath) assert.Equal(t, "/foo/bar", b.Config.Workspace.ArtifactPath) assert.Equal(t, "/foo/bar", b.Config.Workspace.StatePath) } diff --git a/bundle/config/mutator/expand_pipeline_glob_paths_test.go b/bundle/config/mutator/expand_pipeline_glob_paths_test.go index d1671c25..9f70b74a 100644 --- a/bundle/config/mutator/expand_pipeline_glob_paths_test.go +++ b/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" @@ -41,7 +42,7 @@ func TestExpandGlobPathsInPipelines(t *testing.T) { touchEmptyFile(t, filepath.Join(dir, "skip/test7.py")) b := &bundle.Bundle{ - RootPath: dir, + BundleRootPath: dir, Config: config.Root{ Resources: config.Resources{ Pipelines: map[string]*resources.Pipeline{ @@ -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) diff --git a/bundle/config/mutator/expand_workspace_root.go b/bundle/config/mutator/expand_workspace_root.go index 8954abd4..3f0547de 100644 --- a/bundle/config/mutator/expand_workspace_root.go +++ b/bundle/config/mutator/expand_workspace_root.go @@ -33,7 +33,7 @@ func (m *expandWorkspaceRoot) Apply(ctx context.Context, b *bundle.Bundle) diag. } if strings.HasPrefix(root, "~/") { - home := fmt.Sprintf("/Users/%s", currentUser.UserName) + home := fmt.Sprintf("/Workspace/Users/%s", currentUser.UserName) b.Config.Workspace.RootPath = path.Join(home, root[2:]) } diff --git a/bundle/config/mutator/expand_workspace_root_test.go b/bundle/config/mutator/expand_workspace_root_test.go index e6260dbd..40bf35ca 100644 --- a/bundle/config/mutator/expand_workspace_root_test.go +++ b/bundle/config/mutator/expand_workspace_root_test.go @@ -27,7 +27,7 @@ func TestExpandWorkspaceRoot(t *testing.T) { } diags := bundle.Apply(context.Background(), b, mutator.ExpandWorkspaceRoot()) require.NoError(t, diags.Error()) - assert.Equal(t, "/Users/jane@doe.com/foo", b.Config.Workspace.RootPath) + assert.Equal(t, "/Workspace/Users/jane@doe.com/foo", b.Config.Workspace.RootPath) } func TestExpandWorkspaceRootDoesNothing(t *testing.T) { diff --git a/bundle/config/mutator/load_git_details.go b/bundle/config/mutator/load_git_details.go index 9b1c963c..00e7f54d 100644 --- a/bundle/config/mutator/load_git_details.go +++ b/bundle/config/mutator/load_git_details.go @@ -56,7 +56,7 @@ func (m *loadGitDetails) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagn } // Compute relative path of the bundle root from the Git repo root. - absBundlePath, err := filepath.Abs(b.RootPath) + absBundlePath, err := filepath.Abs(b.BundleRootPath) if err != nil { return diag.FromErr(err) } diff --git a/bundle/config/mutator/mutator.go b/bundle/config/mutator/mutator.go index 0458beff..faf50ae6 100644 --- a/bundle/config/mutator/mutator.go +++ b/bundle/config/mutator/mutator.go @@ -23,6 +23,7 @@ func DefaultMutators() []bundle.Mutator { VerifyCliVersion(), EnvironmentsToTargets(), + ComputeIdToClusterId(), InitializeVariables(), DefineDefaultTarget(), LoadGitDetails(), diff --git a/bundle/config/mutator/override_compute.go b/bundle/config/mutator/override_compute.go index 73fbad36..5700cdf2 100644 --- a/bundle/config/mutator/override_compute.go +++ b/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 diff --git a/bundle/config/mutator/override_compute_test.go b/bundle/config/mutator/override_compute_test.go index 152ee543..369447d7 100644 --- a/bundle/config/mutator/override_compute_test.go +++ b/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{ diff --git a/bundle/config/mutator/paths/job_paths_visitor.go b/bundle/config/mutator/paths/job_paths_visitor.go new file mode 100644 index 00000000..275a8fa5 --- /dev/null +++ b/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 +} diff --git a/bundle/config/mutator/paths/job_paths_visitor_test.go b/bundle/config/mutator/paths/job_paths_visitor_test.go new file mode 100644 index 00000000..7f020157 --- /dev/null +++ b/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 +} diff --git a/bundle/config/mutator/paths/visitor.go b/bundle/config/mutator/paths/visitor.go new file mode 100644 index 00000000..40d1f14e --- /dev/null +++ b/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) diff --git a/bundle/config/mutator/populate_current_user.go b/bundle/config/mutator/populate_current_user.go index b5e0bd43..1e99b327 100644 --- a/bundle/config/mutator/populate_current_user.go +++ b/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, } diff --git a/bundle/config/mutator/prepend_workspace_prefix.go b/bundle/config/mutator/prepend_workspace_prefix.go new file mode 100644 index 00000000..dd467344 --- /dev/null +++ b/bundle/config/mutator/prepend_workspace_prefix.go @@ -0,0 +1,67 @@ +package mutator + +import ( + "context" + "fmt" + "strings" + + "github.com/databricks/cli/bundle" + "github.com/databricks/cli/libs/diag" + "github.com/databricks/cli/libs/dyn" +) + +type prependWorkspacePrefix struct{} + +// PrependWorkspacePrefix prepends the workspace root path to all paths in the bundle. +func PrependWorkspacePrefix() bundle.Mutator { + return &prependWorkspacePrefix{} +} + +func (m *prependWorkspacePrefix) Name() string { + return "PrependWorkspacePrefix" +} + +var skipPrefixes = []string{ + "/Workspace/", + "/Volumes/", +} + +func (m *prependWorkspacePrefix) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { + patterns := []dyn.Pattern{ + dyn.NewPattern(dyn.Key("workspace"), dyn.Key("root_path")), + dyn.NewPattern(dyn.Key("workspace"), dyn.Key("file_path")), + dyn.NewPattern(dyn.Key("workspace"), dyn.Key("artifact_path")), + dyn.NewPattern(dyn.Key("workspace"), dyn.Key("state_path")), + } + + err := b.Config.Mutate(func(v dyn.Value) (dyn.Value, error) { + var err error + for _, pattern := range patterns { + v, err = dyn.MapByPattern(v, pattern, func(p dyn.Path, pv dyn.Value) (dyn.Value, error) { + path, ok := pv.AsString() + if !ok { + return dyn.InvalidValue, fmt.Errorf("expected string, got %s", v.Kind()) + } + + for _, prefix := range skipPrefixes { + if strings.HasPrefix(path, prefix) { + return pv, nil + } + } + + return dyn.NewValue(fmt.Sprintf("/Workspace%s", path), v.Locations()), nil + }) + + if err != nil { + return dyn.InvalidValue, err + } + } + return v, nil + }) + + if err != nil { + return diag.FromErr(err) + } + + return nil +} diff --git a/bundle/config/mutator/prepend_workspace_prefix_test.go b/bundle/config/mutator/prepend_workspace_prefix_test.go new file mode 100644 index 00000000..287c694d --- /dev/null +++ b/bundle/config/mutator/prepend_workspace_prefix_test.go @@ -0,0 +1,79 @@ +package mutator + +import ( + "context" + "testing" + + "github.com/databricks/cli/bundle" + "github.com/databricks/cli/bundle/config" + "github.com/databricks/databricks-sdk-go/service/iam" + "github.com/stretchr/testify/require" +) + +func TestPrependWorkspacePrefix(t *testing.T) { + testCases := []struct { + path string + expected string + }{ + { + path: "/Users/test", + expected: "/Workspace/Users/test", + }, + { + path: "/Shared/test", + expected: "/Workspace/Shared/test", + }, + { + path: "/Workspace/Users/test", + expected: "/Workspace/Users/test", + }, + { + path: "/Volumes/Users/test", + expected: "/Volumes/Users/test", + }, + } + + for _, tc := range testCases { + b := &bundle.Bundle{ + Config: config.Root{ + Workspace: config.Workspace{ + RootPath: tc.path, + ArtifactPath: tc.path, + FilePath: tc.path, + StatePath: tc.path, + }, + }, + } + + diags := bundle.Apply(context.Background(), b, PrependWorkspacePrefix()) + require.Empty(t, diags) + require.Equal(t, tc.expected, b.Config.Workspace.RootPath) + require.Equal(t, tc.expected, b.Config.Workspace.ArtifactPath) + require.Equal(t, tc.expected, b.Config.Workspace.FilePath) + require.Equal(t, tc.expected, b.Config.Workspace.StatePath) + } +} + +func TestPrependWorkspaceForDefaultConfig(t *testing.T) { + b := &bundle.Bundle{ + Config: config.Root{ + Bundle: config.Bundle{ + Name: "test", + Target: "dev", + }, + Workspace: config.Workspace{ + CurrentUser: &config.User{ + User: &iam.User{ + UserName: "jane@doe.com", + }, + }, + }, + }, + } + diags := bundle.Apply(context.Background(), b, bundle.Seq(DefineDefaultWorkspaceRoot(), ExpandWorkspaceRoot(), DefineDefaultWorkspacePaths(), PrependWorkspacePrefix())) + require.Empty(t, diags) + require.Equal(t, "/Workspace/Users/jane@doe.com/.bundle/test/dev", b.Config.Workspace.RootPath) + require.Equal(t, "/Workspace/Users/jane@doe.com/.bundle/test/dev/artifacts", b.Config.Workspace.ArtifactPath) + require.Equal(t, "/Workspace/Users/jane@doe.com/.bundle/test/dev/files", b.Config.Workspace.FilePath) + require.Equal(t, "/Workspace/Users/jane@doe.com/.bundle/test/dev/state", b.Config.Workspace.StatePath) +} diff --git a/bundle/config/mutator/process_target_mode.go b/bundle/config/mutator/process_target_mode.go index 70382f05..9944f6ff 100644 --- a/bundle/config/mutator/process_target_mode.go +++ b/bundle/config/mutator/process_target_mode.go @@ -118,15 +118,18 @@ func findNonUserPath(b *bundle.Bundle) string { if b.Config.Workspace.RootPath != "" && !containsName(b.Config.Workspace.RootPath) { return "root_path" } - if b.Config.Workspace.StatePath != "" && !containsName(b.Config.Workspace.StatePath) { - return "state_path" - } if b.Config.Workspace.FilePath != "" && !containsName(b.Config.Workspace.FilePath) { return "file_path" } + if b.Config.Workspace.ResourcePath != "" && !containsName(b.Config.Workspace.ResourcePath) { + return "resource_path" + } if b.Config.Workspace.ArtifactPath != "" && !containsName(b.Config.Workspace.ArtifactPath) { return "artifact_path" } + if b.Config.Workspace.StatePath != "" && !containsName(b.Config.Workspace.StatePath) { + return "state_path" + } return "" } diff --git a/bundle/config/mutator/process_target_mode_test.go b/bundle/config/mutator/process_target_mode_test.go index 42f1929c..b0eb57ee 100644 --- a/bundle/config/mutator/process_target_mode_test.go +++ b/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) { diff --git a/bundle/config/mutator/python/python_mutator.go b/bundle/config/mutator/python/python_mutator.go index fbf3b7e0..da6c4d21 100644 --- a/bundle/config/mutator/python/python_mutator.go +++ b/bundle/config/mutator/python/python_mutator.go @@ -108,7 +108,7 @@ func (m *pythonMutator) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagno return dyn.InvalidValue, fmt.Errorf("failed to create cache dir: %w", err) } - rightRoot, diags := m.runPythonMutator(ctx, cacheDir, b.RootPath, pythonPath, leftRoot) + rightRoot, diags := m.runPythonMutator(ctx, cacheDir, b.BundleRootPath, pythonPath, leftRoot) mutateDiags = diags if diags.HasError() { return dyn.InvalidValue, mutateDiagsHasError @@ -228,12 +228,12 @@ func (m *pythonMutator) runPythonMutator(ctx context.Context, cacheDir string, r return output, pythonDiagnostics } -const installExplanation = `If using Python wheels, ensure that 'databricks-pydabs' is included in the dependencies, +const installExplanation = `If using Python wheels, ensure that 'databricks-pydabs' is included in the dependencies, and that the wheel is installed in the Python environment: $ .venv/bin/pip install -e . -If using a virtual environment, ensure it is specified as the venv_path property in databricks.yml, +If using a virtual environment, ensure it is specified as the venv_path property in databricks.yml, or activate the environment before running CLI commands: experimental: diff --git a/bundle/config/mutator/python/python_mutator_test.go b/bundle/config/mutator/python/python_mutator_test.go index bf12b249..7a419d79 100644 --- a/bundle/config/mutator/python/python_mutator_test.go +++ b/bundle/config/mutator/python/python_mutator_test.go @@ -570,12 +570,12 @@ func TestExplainProcessErr(t *testing.T) { Explanation: 'databricks-pydabs' library is not installed in the Python environment. -If using Python wheels, ensure that 'databricks-pydabs' is included in the dependencies, +If using Python wheels, ensure that 'databricks-pydabs' is included in the dependencies, and that the wheel is installed in the Python environment: $ .venv/bin/pip install -e . -If using a virtual environment, ensure it is specified as the venv_path property in databricks.yml, +If using a virtual environment, ensure it is specified as the venv_path property in databricks.yml, or activate the environment before running CLI commands: experimental: diff --git a/bundle/config/mutator/rewrite_sync_paths.go b/bundle/config/mutator/rewrite_sync_paths.go index 888714ab..f9a02369 100644 --- a/bundle/config/mutator/rewrite_sync_paths.go +++ b/bundle/config/mutator/rewrite_sync_paths.go @@ -45,15 +45,15 @@ func (m *rewriteSyncPaths) makeRelativeTo(root string) dyn.MapFunc { func (m *rewriteSyncPaths) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { err := b.Config.Mutate(func(v dyn.Value) (dyn.Value, error) { return dyn.Map(v, "sync", func(_ dyn.Path, v dyn.Value) (nv dyn.Value, err error) { - v, err = dyn.Map(v, "paths", dyn.Foreach(m.makeRelativeTo(b.RootPath))) + v, err = dyn.Map(v, "paths", dyn.Foreach(m.makeRelativeTo(b.BundleRootPath))) if err != nil { return dyn.InvalidValue, err } - v, err = dyn.Map(v, "include", dyn.Foreach(m.makeRelativeTo(b.RootPath))) + v, err = dyn.Map(v, "include", dyn.Foreach(m.makeRelativeTo(b.BundleRootPath))) if err != nil { return dyn.InvalidValue, err } - v, err = dyn.Map(v, "exclude", dyn.Foreach(m.makeRelativeTo(b.RootPath))) + v, err = dyn.Map(v, "exclude", dyn.Foreach(m.makeRelativeTo(b.BundleRootPath))) if err != nil { return dyn.InvalidValue, err } diff --git a/bundle/config/mutator/rewrite_sync_paths_test.go b/bundle/config/mutator/rewrite_sync_paths_test.go index fa7f124b..0e4dfc4c 100644 --- a/bundle/config/mutator/rewrite_sync_paths_test.go +++ b/bundle/config/mutator/rewrite_sync_paths_test.go @@ -9,12 +9,13 @@ 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" ) func TestRewriteSyncPathsRelative(t *testing.T) { b := &bundle.Bundle{ - RootPath: ".", + BundleRootPath: ".", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -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()) @@ -53,7 +54,7 @@ func TestRewriteSyncPathsRelative(t *testing.T) { func TestRewriteSyncPathsAbsolute(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/dir", + BundleRootPath: "/tmp/dir", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -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()) @@ -93,7 +94,7 @@ func TestRewriteSyncPathsAbsolute(t *testing.T) { func TestRewriteSyncPathsErrorPaths(t *testing.T) { t.Run("no sync block", func(t *testing.T) { b := &bundle.Bundle{ - RootPath: ".", + BundleRootPath: ".", } diags := bundle.Apply(context.Background(), b, mutator.RewriteSyncPaths()) @@ -102,7 +103,7 @@ func TestRewriteSyncPathsErrorPaths(t *testing.T) { t.Run("empty include/exclude blocks", func(t *testing.T) { b := &bundle.Bundle{ - RootPath: ".", + BundleRootPath: ".", Config: config.Root{ Sync: config.Sync{ Include: []string{}, diff --git a/bundle/config/mutator/rewrite_workspace_prefix.go b/bundle/config/mutator/rewrite_workspace_prefix.go new file mode 100644 index 00000000..8a39ee8a --- /dev/null +++ b/bundle/config/mutator/rewrite_workspace_prefix.go @@ -0,0 +1,72 @@ +package mutator + +import ( + "context" + "fmt" + "strings" + + "github.com/databricks/cli/bundle" + "github.com/databricks/cli/libs/diag" + "github.com/databricks/cli/libs/dyn" +) + +type rewriteWorkspacePrefix struct{} + +// RewriteWorkspacePrefix finds any strings in bundle configration that have +// workspace prefix plus workspace path variable used and removes workspace prefix from it. +func RewriteWorkspacePrefix() bundle.Mutator { + return &rewriteWorkspacePrefix{} +} + +func (m *rewriteWorkspacePrefix) Name() string { + return "RewriteWorkspacePrefix" +} + +func (m *rewriteWorkspacePrefix) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { + diags := diag.Diagnostics{} + paths := map[string]string{ + "/Workspace/${workspace.root_path}": "${workspace.root_path}", + "/Workspace${workspace.root_path}": "${workspace.root_path}", + "/Workspace/${workspace.file_path}": "${workspace.file_path}", + "/Workspace${workspace.file_path}": "${workspace.file_path}", + "/Workspace/${workspace.artifact_path}": "${workspace.artifact_path}", + "/Workspace${workspace.artifact_path}": "${workspace.artifact_path}", + "/Workspace/${workspace.state_path}": "${workspace.state_path}", + "/Workspace${workspace.state_path}": "${workspace.state_path}", + } + + err := b.Config.Mutate(func(root dyn.Value) (dyn.Value, error) { + // Walk through the bundle configuration, check all the string leafs and + // see if any of the prefixes are used in the remote path. + return dyn.Walk(root, func(p dyn.Path, v dyn.Value) (dyn.Value, error) { + vv, ok := v.AsString() + if !ok { + return v, nil + } + + for path, replacePath := range paths { + if strings.Contains(vv, path) { + newPath := strings.Replace(vv, path, replacePath, 1) + diags = append(diags, diag.Diagnostic{ + Severity: diag.Warning, + Summary: fmt.Sprintf("substring %q found in %q. Please update this to %q.", path, vv, newPath), + Detail: "For more information, please refer to: https://docs.databricks.com/en/release-notes/dev-tools/bundles.html#workspace-paths", + Locations: v.Locations(), + Paths: []dyn.Path{p}, + }) + + // Remove the workspace prefix from the string. + return dyn.NewValue(newPath, v.Locations()), nil + } + } + + return v, nil + }) + }) + + if err != nil { + return diag.FromErr(err) + } + + return diags +} diff --git a/bundle/config/mutator/rewrite_workspace_prefix_test.go b/bundle/config/mutator/rewrite_workspace_prefix_test.go new file mode 100644 index 00000000..d75ec89d --- /dev/null +++ b/bundle/config/mutator/rewrite_workspace_prefix_test.go @@ -0,0 +1,85 @@ +package mutator + +import ( + "context" + "testing" + + "github.com/databricks/cli/bundle" + "github.com/databricks/cli/bundle/config" + "github.com/databricks/cli/bundle/config/resources" + "github.com/databricks/cli/libs/diag" + "github.com/databricks/databricks-sdk-go/service/compute" + "github.com/databricks/databricks-sdk-go/service/jobs" + "github.com/stretchr/testify/require" +) + +func TestNoWorkspacePrefixUsed(t *testing.T) { + b := &bundle.Bundle{ + Config: config.Root{ + Workspace: config.Workspace{ + RootPath: "/Workspace/Users/test", + ArtifactPath: "/Workspace/Users/test/artifacts", + FilePath: "/Workspace/Users/test/files", + StatePath: "/Workspace/Users/test/state", + }, + + Resources: config.Resources{ + Jobs: map[string]*resources.Job{ + "test_job": { + JobSettings: &jobs.JobSettings{ + Tasks: []jobs.Task{ + { + SparkPythonTask: &jobs.SparkPythonTask{ + PythonFile: "/Workspace/${workspace.root_path}/file1.py", + }, + }, + { + NotebookTask: &jobs.NotebookTask{ + NotebookPath: "/Workspace${workspace.file_path}/notebook1", + }, + Libraries: []compute.Library{ + { + Jar: "/Workspace/${workspace.artifact_path}/jar1.jar", + }, + }, + }, + { + NotebookTask: &jobs.NotebookTask{ + NotebookPath: "${workspace.file_path}/notebook2", + }, + Libraries: []compute.Library{ + { + Jar: "${workspace.artifact_path}/jar2.jar", + }, + }, + }, + }, + }, + }, + }, + }, + }, + } + + diags := bundle.Apply(context.Background(), b, RewriteWorkspacePrefix()) + require.Len(t, diags, 3) + + expectedErrors := map[string]bool{ + `substring "/Workspace/${workspace.root_path}" found in "/Workspace/${workspace.root_path}/file1.py". Please update this to "${workspace.root_path}/file1.py".`: true, + `substring "/Workspace${workspace.file_path}" found in "/Workspace${workspace.file_path}/notebook1". Please update this to "${workspace.file_path}/notebook1".`: true, + `substring "/Workspace/${workspace.artifact_path}" found in "/Workspace/${workspace.artifact_path}/jar1.jar". Please update this to "${workspace.artifact_path}/jar1.jar".`: true, + } + + for _, d := range diags { + require.Equal(t, d.Severity, diag.Warning) + require.Contains(t, expectedErrors, d.Summary) + delete(expectedErrors, d.Summary) + } + + require.Equal(t, "${workspace.root_path}/file1.py", b.Config.Resources.Jobs["test_job"].JobSettings.Tasks[0].SparkPythonTask.PythonFile) + require.Equal(t, "${workspace.file_path}/notebook1", b.Config.Resources.Jobs["test_job"].JobSettings.Tasks[1].NotebookTask.NotebookPath) + require.Equal(t, "${workspace.artifact_path}/jar1.jar", b.Config.Resources.Jobs["test_job"].JobSettings.Tasks[1].Libraries[0].Jar) + require.Equal(t, "${workspace.file_path}/notebook2", b.Config.Resources.Jobs["test_job"].JobSettings.Tasks[2].NotebookTask.NotebookPath) + require.Equal(t, "${workspace.artifact_path}/jar2.jar", b.Config.Resources.Jobs["test_job"].JobSettings.Tasks[2].Libraries[0].Jar) + +} diff --git a/bundle/config/mutator/run_as_test.go b/bundle/config/mutator/run_as_test.go index e6cef9ba..abeea45d 100644 --- a/bundle/config/mutator/run_as_test.go +++ b/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", diff --git a/bundle/config/mutator/sync_default_path_test.go b/bundle/config/mutator/sync_default_path_test.go index a37e913d..c82a687b 100644 --- a/bundle/config/mutator/sync_default_path_test.go +++ b/bundle/config/mutator/sync_default_path_test.go @@ -15,8 +15,8 @@ import ( func TestSyncDefaultPath_DefaultIfUnset(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir", - Config: config.Root{}, + BundleRootPath: "/tmp/some/dir", + Config: config.Root{}, } ctx := context.Background() @@ -51,8 +51,8 @@ func TestSyncDefaultPath_SkipIfSet(t *testing.T) { for _, tcase := range tcases { t.Run(tcase.name, func(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir", - Config: config.Root{}, + BundleRootPath: "/tmp/some/dir", + Config: config.Root{}, } diags := bundle.ApplyFunc(context.Background(), b, func(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { diff --git a/bundle/config/mutator/sync_infer_root.go b/bundle/config/mutator/sync_infer_root.go index 012acf80..512adcdb 100644 --- a/bundle/config/mutator/sync_infer_root.go +++ b/bundle/config/mutator/sync_infer_root.go @@ -57,7 +57,7 @@ func (m *syncInferRoot) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagno var diags diag.Diagnostics // Use the bundle root path as the starting point for inferring the sync root path. - bundleRootPath := filepath.Clean(b.RootPath) + bundleRootPath := filepath.Clean(b.BundleRootPath) // Infer the sync root path by looking at each one of the sync paths. // Every sync path must be a descendant of the final sync root path. diff --git a/bundle/config/mutator/sync_infer_root_test.go b/bundle/config/mutator/sync_infer_root_test.go index 383e5676..f507cbc7 100644 --- a/bundle/config/mutator/sync_infer_root_test.go +++ b/bundle/config/mutator/sync_infer_root_test.go @@ -9,13 +9,14 @@ 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" ) func TestSyncInferRoot_NominalAbsolute(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir", + BundleRootPath: "/tmp/some/dir", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -46,7 +47,7 @@ func TestSyncInferRoot_NominalAbsolute(t *testing.T) { func TestSyncInferRoot_NominalRelative(t *testing.T) { b := &bundle.Bundle{ - RootPath: "./some/dir", + BundleRootPath: "./some/dir", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -77,7 +78,7 @@ func TestSyncInferRoot_NominalRelative(t *testing.T) { func TestSyncInferRoot_ParentDirectory(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir", + BundleRootPath: "/tmp/some/dir", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -108,7 +109,7 @@ func TestSyncInferRoot_ParentDirectory(t *testing.T) { func TestSyncInferRoot_ManyParentDirectories(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir/that/is/very/deeply/nested", + BundleRootPath: "/tmp/some/dir/that/is/very/deeply/nested", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -145,7 +146,7 @@ func TestSyncInferRoot_ManyParentDirectories(t *testing.T) { func TestSyncInferRoot_MultiplePaths(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/bundle/root", + BundleRootPath: "/tmp/some/bundle/root", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -172,7 +173,7 @@ func TestSyncInferRoot_MultiplePaths(t *testing.T) { func TestSyncInferRoot_Error(t *testing.T) { b := &bundle.Bundle{ - RootPath: "/tmp/some/dir", + BundleRootPath: "/tmp/some/dir", Config: config.Root{ Sync: config.Sync{ Paths: []string{ @@ -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()) diff --git a/bundle/config/mutator/translate_paths_jobs.go b/bundle/config/mutator/translate_paths_jobs.go index e34eeb2f..c29ff0ea 100644 --- a/bundle/config/mutator/translate_paths_jobs.go +++ b/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) } diff --git a/bundle/config/mutator/translate_paths_test.go b/bundle/config/mutator/translate_paths_test.go index 50fcd3b0..c03cee73 100644 --- a/bundle/config/mutator/translate_paths_test.go +++ b/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. diff --git a/bundle/config/resources.go b/bundle/config/resources.go index 3cee9abe..e932dab0 100644 --- a/bundle/config/resources.go +++ b/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 { diff --git a/bundle/config/resources/clusters.go b/bundle/config/resources/clusters.go new file mode 100644 index 00000000..63234566 --- /dev/null +++ b/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" +} diff --git a/bundle/config/root.go b/bundle/config/root.go index 46578769..4b146745 100644 --- a/bundle/config/root.go +++ b/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 } @@ -406,23 +406,45 @@ func (r *Root) MergeTargetOverrides(name string) error { return r.updateWithDynamicValue(root) } -var variableKeywords = []string{"default", "lookup"} +var allowedVariableDefinitions = []([]string){ + {"default", "type", "description"}, + {"default", "type"}, + {"default", "description"}, + {"lookup", "description"}, + {"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 } - for _, keyword := range variableKeywords { - if _, ok := mv.GetByString(keyword); ok { + for _, keys := range allowedVariableDefinitions { + if len(keys) != mv.Len() { + continue + } + + // Check if the keys are the same. + match := true + for _, key := range keys { + if _, ok := mv.GetByString(key); !ok { + match = false + break + } + } + + if match { return true } } diff --git a/bundle/config/root_test.go b/bundle/config/root_test.go index d2c7a9b1..9e612353 100644 --- a/bundle/config/root_test.go +++ b/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" ) @@ -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) + } + +} diff --git a/bundle/config/target.go b/bundle/config/target.go index fc6ba7b5..fae9c940 100644 --- a/bundle/config/target.go +++ b/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"` diff --git a/bundle/config/validate/job_task_cluster_spec.go b/bundle/config/validate/job_task_cluster_spec.go new file mode 100644 index 00000000..b80befcd --- /dev/null +++ b/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 +} diff --git a/bundle/config/validate/job_task_cluster_spec_test.go b/bundle/config/validate/job_task_cluster_spec_test.go new file mode 100644 index 00000000..a3a7ccf2 --- /dev/null +++ b/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, + }, + }, + } +} diff --git a/bundle/config/validate/validate.go b/bundle/config/validate/validate.go index b4da0bc0..79f42bd2 100644 --- a/bundle/config/validate/validate.go +++ b/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(), )) } diff --git a/bundle/config/workspace.go b/bundle/config/workspace.go index efc5caa6..e64808ab 100644 --- a/bundle/config/workspace.go +++ b/bundle/config/workspace.go @@ -47,13 +47,18 @@ type Workspace struct { // Remote workspace base path for deployment state, for artifacts, as synchronization target. // This defaults to "~/.bundle/${bundle.name}/${bundle.target}" where "~" expands to - // the current user's home directory in the workspace (e.g. `/Users/jane@doe.com`). + // the current user's home directory in the workspace (e.g. `/Workspace/Users/jane@doe.com`). RootPath string `json:"root_path,omitempty"` // Remote workspace path to synchronize local files to. // This defaults to "${workspace.root}/files". FilePath string `json:"file_path,omitempty"` + // Remote workspace path for resources with a presence in the workspace. + // These are kept outside [FilePath] to avoid potential naming collisions. + // This defaults to "${workspace.root}/resources". + ResourcePath string `json:"resource_path,omitempty"` + // Remote workspace path for build artifacts. // This defaults to "${workspace.root}/artifacts". ArtifactPath string `json:"artifact_path,omitempty"` diff --git a/bundle/deploy/files/upload.go b/bundle/deploy/files/upload.go index 2c126623..77b83611 100644 --- a/bundle/deploy/files/upload.go +++ b/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} } diff --git a/bundle/deploy/metadata/compute.go b/bundle/deploy/metadata/compute.go index 6ab997e2..bc8767de 100644 --- a/bundle/deploy/metadata/compute.go +++ b/bundle/deploy/metadata/compute.go @@ -40,7 +40,7 @@ func (m *compute) Apply(_ context.Context, b *bundle.Bundle) diag.Diagnostics { // Compute config file path the job is defined in, relative to the bundle // root l := b.Config.GetLocation("resources.jobs." + name) - relativePath, err := filepath.Rel(b.RootPath, l.File) + relativePath, err := filepath.Rel(b.BundleRootPath, l.File) if err != nil { return diag.Errorf("failed to compute relative path for job %s: %v", name, err) } diff --git a/bundle/deploy/metadata/compute_test.go b/bundle/deploy/metadata/compute_test.go index 6d43f845..2c2c7237 100644 --- a/bundle/deploy/metadata/compute_test.go +++ b/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, diff --git a/bundle/deploy/state_pull_test.go b/bundle/deploy/state_pull_test.go index f7519306..42701eb2 100644 --- a/bundle/deploy/state_pull_test.go +++ b/bundle/deploy/state_pull_test.go @@ -62,8 +62,8 @@ func testStatePull(t *testing.T, opts statePullOpts) { tmpDir := t.TempDir() b := &bundle.Bundle{ - RootPath: tmpDir, - BundleRoot: vfs.MustNew(tmpDir), + BundleRootPath: tmpDir, + BundleRoot: vfs.MustNew(tmpDir), SyncRootPath: tmpDir, SyncRoot: vfs.MustNew(tmpDir), @@ -259,7 +259,7 @@ func TestStatePullNoState(t *testing.T) { }} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "default", @@ -447,7 +447,7 @@ func TestStatePullNewerDeploymentStateVersion(t *testing.T) { }} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "default", diff --git a/bundle/deploy/state_push.go b/bundle/deploy/state_push.go index 176a907c..6912414c 100644 --- a/bundle/deploy/state_push.go +++ b/bundle/deploy/state_push.go @@ -10,6 +10,8 @@ import ( "github.com/databricks/cli/libs/log" ) +const MaxStateFileSize = 10 * 1024 * 1024 // 10MB + type statePush struct { filerFactory FilerFactory } @@ -35,6 +37,17 @@ func (s *statePush) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostic } defer local.Close() + if !b.Config.Bundle.Force { + state, err := local.Stat() + if err != nil { + return diag.FromErr(err) + } + + if state.Size() > MaxStateFileSize { + return diag.Errorf("Deployment state file size exceeds the maximum allowed size of %d bytes. Please reduce the number of resources in your bundle, split your bundle into multiple or re-run the command with --force flag.", MaxStateFileSize) + } + } + log.Infof(ctx, "Writing local deployment state file to remote state directory") err = f.Write(ctx, DeploymentStateFileName, local, filer.CreateParentDirectories, filer.OverwriteIfExists) if err != nil { diff --git a/bundle/deploy/state_push_test.go b/bundle/deploy/state_push_test.go index 39e4d13a..038b7534 100644 --- a/bundle/deploy/state_push_test.go +++ b/bundle/deploy/state_push_test.go @@ -45,7 +45,7 @@ func TestStatePush(t *testing.T) { }} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "default", diff --git a/bundle/deploy/state_update_test.go b/bundle/deploy/state_update_test.go index 72096d14..1f5010b5 100644 --- a/bundle/deploy/state_update_test.go +++ b/bundle/deploy/state_update_test.go @@ -27,7 +27,7 @@ func setupBundleForStateUpdate(t *testing.T) *bundle.Bundle { require.NoError(t, err) return &bundle.Bundle{ - RootPath: tmpDir, + BundleRootPath: tmpDir, Config: config.Root{ Bundle: config.Bundle{ Target: "default", diff --git a/bundle/deploy/terraform/convert.go b/bundle/deploy/terraform/convert.go index f13c241c..b8993c03 100644 --- a/bundle/deploy/terraform/convert.go +++ b/bundle/deploy/terraform/convert.go @@ -4,6 +4,7 @@ import ( "context" "encoding/json" "fmt" + "sort" "github.com/databricks/cli/bundle/config" "github.com/databricks/cli/bundle/config/resources" @@ -82,6 +83,10 @@ func BundleToTerraform(config *config.Root) *schema.Root { conv(src, &dst) if src.JobSettings != nil { + sort.Slice(src.JobSettings.Tasks, func(i, j int) bool { + return src.JobSettings.Tasks[i].TaskKey < src.JobSettings.Tasks[j].TaskKey + }) + for _, v := range src.Tasks { var t schema.ResourceJobTask conv(v, &t) @@ -231,6 +236,13 @@ 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 +406,16 @@ 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 +465,11 @@ 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 } diff --git a/bundle/deploy/terraform/convert_test.go b/bundle/deploy/terraform/convert_test.go index e4ef6114..4c6866d9 100644 --- a/bundle/deploy/terraform/convert_test.go +++ b/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) } diff --git a/bundle/deploy/terraform/init_test.go b/bundle/deploy/terraform/init_test.go index 450e7eb6..e3621c6c 100644 --- a/bundle/deploy/terraform/init_test.go +++ b/bundle/deploy/terraform/init_test.go @@ -33,7 +33,7 @@ func TestInitEnvironmentVariables(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -60,7 +60,7 @@ func TestSetTempDirEnvVarsForUnixWithTmpDirSet(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -88,7 +88,7 @@ func TestSetTempDirEnvVarsForUnixWithTmpDirNotSet(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -114,7 +114,7 @@ func TestSetTempDirEnvVarsForWindowWithAllTmpDirEnvVarsSet(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -144,7 +144,7 @@ func TestSetTempDirEnvVarsForWindowWithUserProfileAndTempSet(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -174,7 +174,7 @@ func TestSetTempDirEnvVarsForWindowsWithoutAnyTempDirEnvVarsSet(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -202,7 +202,7 @@ func TestSetTempDirEnvVarsForWindowsWithoutAnyTempDirEnvVarsSet(t *testing.T) { func TestSetProxyEnvVars(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -250,7 +250,7 @@ func TestSetProxyEnvVars(t *testing.T) { func TestSetUserAgentExtraEnvVar(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Experimental: &config.Experimental{ PyDABs: config.PyDABs{ @@ -333,7 +333,7 @@ func TestFindExecPathFromEnvironmentWithWrongVersion(t *testing.T) { ctx := context.Background() m := &initialize{} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -357,7 +357,7 @@ func TestFindExecPathFromEnvironmentWithCorrectVersionAndNoBinary(t *testing.T) ctx := context.Background() m := &initialize{} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -380,7 +380,7 @@ func TestFindExecPathFromEnvironmentWithCorrectVersionAndBinary(t *testing.T) { ctx := context.Background() m := &initialize{} b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", diff --git a/bundle/deploy/terraform/interpolate.go b/bundle/deploy/terraform/interpolate.go index faa098e1..12894c68 100644 --- a/bundle/deploy/terraform/interpolate.go +++ b/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) diff --git a/bundle/deploy/terraform/interpolate_test.go b/bundle/deploy/terraform/interpolate_test.go index 5ceb243b..630a904a 100644 --- a/bundle/deploy/terraform/interpolate_test.go +++ b/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) diff --git a/bundle/deploy/terraform/load_test.go b/bundle/deploy/terraform/load_test.go index c6221718..b7243ca1 100644 --- a/bundle/deploy/terraform/load_test.go +++ b/bundle/deploy/terraform/load_test.go @@ -17,7 +17,7 @@ func TestLoadWithNoState(t *testing.T) { } b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", diff --git a/bundle/deploy/terraform/state_pull_test.go b/bundle/deploy/terraform/state_pull_test.go index 39937a3c..c4798c57 100644 --- a/bundle/deploy/terraform/state_pull_test.go +++ b/bundle/deploy/terraform/state_pull_test.go @@ -32,7 +32,7 @@ func mockStateFilerForPull(t *testing.T, contents map[string]any, merr error) fi func statePullTestBundle(t *testing.T) *bundle.Bundle { return &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "default", diff --git a/bundle/deploy/terraform/state_push.go b/bundle/deploy/terraform/state_push.go index 6cdde137..84d8e767 100644 --- a/bundle/deploy/terraform/state_push.go +++ b/bundle/deploy/terraform/state_push.go @@ -47,6 +47,17 @@ func (l *statePush) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostic } defer local.Close() + if !b.Config.Bundle.Force { + state, err := local.Stat() + if err != nil { + return diag.FromErr(err) + } + + if state.Size() > deploy.MaxStateFileSize { + return diag.Errorf("Terraform state file size exceeds the maximum allowed size of %d bytes. Please reduce the number of resources in your bundle, split your bundle into multiple or re-run the command with --force flag", deploy.MaxStateFileSize) + } + } + // Upload state file from local cache directory to filer. cmdio.LogString(ctx, "Updating deployment state...") log.Infof(ctx, "Writing local state file to remote state directory") diff --git a/bundle/deploy/terraform/state_push_test.go b/bundle/deploy/terraform/state_push_test.go index ac74f345..4cc52b7a 100644 --- a/bundle/deploy/terraform/state_push_test.go +++ b/bundle/deploy/terraform/state_push_test.go @@ -3,6 +3,7 @@ package terraform import ( "context" "encoding/json" + "fmt" "io" "testing" @@ -29,7 +30,7 @@ func mockStateFilerForPush(t *testing.T, fn func(body io.Reader)) filer.Filer { func statePushTestBundle(t *testing.T) *bundle.Bundle { return &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "default", @@ -59,3 +60,29 @@ func TestStatePush(t *testing.T) { diags := bundle.Apply(ctx, b, m) assert.NoError(t, diags.Error()) } + +func TestStatePushLargeState(t *testing.T) { + mock := mockfiler.NewMockFiler(t) + m := &statePush{ + identityFiler(mock), + } + + ctx := context.Background() + b := statePushTestBundle(t) + + largeState := map[string]any{} + for i := 0; i < 1000000; i++ { + largeState[fmt.Sprintf("field_%d", i)] = i + } + + // Write a stale local state file. + writeLocalState(t, ctx, b, largeState) + diags := bundle.Apply(ctx, b, m) + assert.ErrorContains(t, diags.Error(), "Terraform state file size exceeds the maximum allowed size of 10485760 bytes. Please reduce the number of resources in your bundle, split your bundle into multiple or re-run the command with --force flag") + + // Force the write. + b = statePushTestBundle(t) + b.Config.Bundle.Force = true + diags = bundle.Apply(ctx, b, m) + assert.NoError(t, diags.Error()) +} diff --git a/bundle/deploy/terraform/tfdyn/convert_cluster.go b/bundle/deploy/terraform/tfdyn/convert_cluster.go new file mode 100644 index 00000000..f25f09ea --- /dev/null +++ b/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{}) +} diff --git a/bundle/deploy/terraform/tfdyn/convert_cluster_test.go b/bundle/deploy/terraform/tfdyn/convert_cluster_test.go new file mode 100644 index 00000000..e7d2542f --- /dev/null +++ b/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"]) + +} diff --git a/bundle/deploy/terraform/tfdyn/convert_job.go b/bundle/deploy/terraform/tfdyn/convert_job.go index d1e7e73e..8948e3ba 100644 --- a/bundle/deploy/terraform/tfdyn/convert_job.go +++ b/bundle/deploy/terraform/tfdyn/convert_job.go @@ -3,6 +3,7 @@ package tfdyn import ( "context" "fmt" + "sort" "github.com/databricks/cli/bundle/internal/tf/schema" "github.com/databricks/cli/libs/dyn" @@ -19,8 +20,38 @@ func convertJobResource(ctx context.Context, vin dyn.Value) (dyn.Value, error) { log.Debugf(ctx, "job normalization diagnostic: %s", diag.Summary) } + // Sort the tasks of each job in the bundle by task key. Sorting + // the task keys ensures that the diff computed by terraform is correct and avoids + // recreates. For more details see the NOTE at + // https://registry.terraform.io/providers/databricks/databricks/latest/docs/resources/job#example-usage + // and https://github.com/databricks/terraform-provider-databricks/issues/4011 + // and https://github.com/databricks/cli/pull/1776 + vout := vin + var err error + tasks, ok := vin.Get("tasks").AsSequence() + if ok { + sort.Slice(tasks, func(i, j int) bool { + // We sort the tasks by their task key. Tasks without task keys are ordered + // before tasks with task keys. We do not error for those tasks + // since presence of a task_key is validated for in the Jobs backend. + tk1, ok := tasks[i].Get("task_key").AsString() + if !ok { + return true + } + tk2, ok := tasks[j].Get("task_key").AsString() + if !ok { + return false + } + return tk1 < tk2 + }) + vout, err = dyn.Set(vin, "tasks", dyn.V(tasks)) + if err != nil { + return dyn.InvalidValue, err + } + } + // Modify top-level keys. - vout, err := renameKeys(vin, map[string]string{ + vout, err = renameKeys(vout, map[string]string{ "tasks": "task", "job_clusters": "job_cluster", "parameters": "parameter", diff --git a/bundle/deploy/terraform/tfdyn/convert_job_test.go b/bundle/deploy/terraform/tfdyn/convert_job_test.go index b9e1f967..695b9ba2 100644 --- a/bundle/deploy/terraform/tfdyn/convert_job_test.go +++ b/bundle/deploy/terraform/tfdyn/convert_job_test.go @@ -42,8 +42,8 @@ func TestConvertJob(t *testing.T) { }, Tasks: []jobs.Task{ { - TaskKey: "task_key", - JobClusterKey: "job_cluster_key", + TaskKey: "task_key_b", + JobClusterKey: "job_cluster_key_b", Libraries: []compute.Library{ { Pypi: &compute.PythonPyPiLibrary{ @@ -55,6 +55,17 @@ func TestConvertJob(t *testing.T) { }, }, }, + { + TaskKey: "task_key_a", + JobClusterKey: "job_cluster_key_a", + }, + { + TaskKey: "task_key_c", + JobClusterKey: "job_cluster_key_c", + }, + { + Description: "missing task key 😱", + }, }, }, Permissions: []resources.Permission{ @@ -100,8 +111,15 @@ func TestConvertJob(t *testing.T) { }, "task": []any{ map[string]any{ - "task_key": "task_key", - "job_cluster_key": "job_cluster_key", + "description": "missing task key 😱", + }, + map[string]any{ + "task_key": "task_key_a", + "job_cluster_key": "job_cluster_key_a", + }, + map[string]any{ + "task_key": "task_key_b", + "job_cluster_key": "job_cluster_key_b", "library": []any{ map[string]any{ "pypi": map[string]any{ @@ -113,6 +131,10 @@ func TestConvertJob(t *testing.T) { }, }, }, + map[string]any{ + "task_key": "task_key_c", + "job_cluster_key": "job_cluster_key_c", + }, }, }, out.Job["my_job"]) diff --git a/bundle/deploy/terraform/util_test.go b/bundle/deploy/terraform/util_test.go index 251a7c25..74b32925 100644 --- a/bundle/deploy/terraform/util_test.go +++ b/bundle/deploy/terraform/util_test.go @@ -13,7 +13,7 @@ import ( func TestParseResourcesStateWithNoFile(t *testing.T) { b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -31,7 +31,7 @@ func TestParseResourcesStateWithNoFile(t *testing.T) { func TestParseResourcesStateWithExistingStateFile(t *testing.T) { ctx := context.Background() b := &bundle.Bundle{ - RootPath: t.TempDir(), + BundleRootPath: t.TempDir(), Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", diff --git a/bundle/internal/bundletest/location.go b/bundle/internal/bundletest/location.go index 380d6e17..2ffd621b 100644 --- a/bundle/internal/bundletest/location.go +++ b/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. diff --git a/bundle/internal/tf/codegen/generator/generator.go b/bundle/internal/tf/codegen/generator/generator.go index 86d76243..b31fdf15 100644 --- a/bundle/internal/tf/codegen/generator/generator.go +++ b/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", diff --git a/bundle/internal/tf/codegen/schema/version.go b/bundle/internal/tf/codegen/schema/version.go index efb29724..b71ea7d1 100644 --- a/bundle/internal/tf/codegen/schema/version.go +++ b/bundle/internal/tf/codegen/schema/version.go @@ -1,3 +1,3 @@ package schema -const ProviderVersion = "1.50.0" +const ProviderVersion = "1.52.0" diff --git a/bundle/internal/tf/schema/data_source_clusters.go b/bundle/internal/tf/schema/data_source_clusters.go index 7a5f3053..8c5f9578 100644 --- a/bundle/internal/tf/schema/data_source_clusters.go +++ b/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"` } diff --git a/bundle/internal/tf/schema/data_source_external_location.go b/bundle/internal/tf/schema/data_source_external_location.go index a3e78cbd..e1ad9dc3 100644 --- a/bundle/internal/tf/schema/data_source_external_location.go +++ b/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"` diff --git a/bundle/internal/tf/schema/data_source_share.go b/bundle/internal/tf/schema/data_source_share.go index 3b40fbb5..da9afaae 100644 --- a/bundle/internal/tf/schema/data_source_share.go +++ b/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"` } diff --git a/bundle/internal/tf/schema/resource_automatic_cluster_update_workspace_setting.go b/bundle/internal/tf/schema/resource_automatic_cluster_update_workspace_setting.go index e95639de..5d7f6a14 100644 --- a/bundle/internal/tf/schema/resource_automatic_cluster_update_workspace_setting.go +++ b/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"` } diff --git a/bundle/internal/tf/schema/resource_cluster.go b/bundle/internal/tf/schema/resource_cluster.go index e4106d04..4ae063c8 100644 --- a/bundle/internal/tf/schema/resource_cluster.go +++ b/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"` diff --git a/bundle/internal/tf/schema/resource_compliance_security_profile_workspace_setting.go b/bundle/internal/tf/schema/resource_compliance_security_profile_workspace_setting.go index 50815f75..8265adae 100644 --- a/bundle/internal/tf/schema/resource_compliance_security_profile_workspace_setting.go +++ b/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 { diff --git a/bundle/internal/tf/schema/resource_enhanced_security_monitoring_workspace_setting.go b/bundle/internal/tf/schema/resource_enhanced_security_monitoring_workspace_setting.go index 2f552402..e9c3b0ab 100644 --- a/bundle/internal/tf/schema/resource_enhanced_security_monitoring_workspace_setting.go +++ b/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 { diff --git a/bundle/internal/tf/schema/resource_model_serving.go b/bundle/internal/tf/schema/resource_model_serving.go index 379807a5..29d55cd5 100644 --- a/bundle/internal/tf/schema/resource_model_serving.go +++ b/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 { diff --git a/bundle/internal/tf/schema/resource_share.go b/bundle/internal/tf/schema/resource_share.go index e531e777..37f4d454 100644 --- a/bundle/internal/tf/schema/resource_share.go +++ b/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"` } diff --git a/bundle/internal/tf/schema/resource_sql_table.go b/bundle/internal/tf/schema/resource_sql_table.go index 51fb3bc0..4f305c52 100644 --- a/bundle/internal/tf/schema/resource_sql_table.go +++ b/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"` diff --git a/bundle/internal/tf/schema/root.go b/bundle/internal/tf/schema/root.go index ebdb7f09..5fc34d6b 100644 --- a/bundle/internal/tf/schema/root.go +++ b/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{ diff --git a/bundle/libraries/expand_glob_references_test.go b/bundle/libraries/expand_glob_references_test.go index e7f2e169..2dfbddb7 100644 --- a/bundle/libraries/expand_glob_references_test.go +++ b/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) diff --git a/bundle/phases/deploy.go b/bundle/phases/deploy.go index 49544227..cb0ecf75 100644 --- a/bundle/phases/deploy.go +++ b/bundle/phases/deploy.go @@ -15,9 +15,10 @@ import ( "github.com/databricks/cli/bundle/deploy/terraform" "github.com/databricks/cli/bundle/libraries" "github.com/databricks/cli/bundle/permissions" - "github.com/databricks/cli/bundle/python" "github.com/databricks/cli/bundle/scripts" + "github.com/databricks/cli/bundle/trampoline" "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( @@ -156,8 +157,8 @@ func Deploy() bundle.Mutator { artifacts.CleanUp(), libraries.ExpandGlobReferences(), libraries.Upload(), - python.TransformWheelTask(), - files.Upload(), + trampoline.TransformWheelTask(), + files.Upload(outputHandler), deploy.StateUpdate(), deploy.StatePush(), permissions.ApplyWorkspaceRootPermissions(), diff --git a/bundle/phases/initialize.go b/bundle/phases/initialize.go index 8039a4f1..a41819c7 100644 --- a/bundle/phases/initialize.go +++ b/bundle/phases/initialize.go @@ -9,8 +9,8 @@ import ( "github.com/databricks/cli/bundle/deploy/metadata" "github.com/databricks/cli/bundle/deploy/terraform" "github.com/databricks/cli/bundle/permissions" - "github.com/databricks/cli/bundle/python" "github.com/databricks/cli/bundle/scripts" + "github.com/databricks/cli/bundle/trampoline" ) // The initialize phase fills in defaults and connects to the workspace. @@ -39,9 +39,16 @@ func Initialize() bundle.Mutator { mutator.MergePipelineClusters(), mutator.InitializeWorkspaceClient(), mutator.PopulateCurrentUser(), + mutator.DefineDefaultWorkspaceRoot(), mutator.ExpandWorkspaceRoot(), mutator.DefineDefaultWorkspacePaths(), + mutator.PrependWorkspacePrefix(), + + // This mutator needs to be run before variable interpolation because it + // searches for strings with variable references in them. + mutator.RewriteWorkspacePrefix(), + mutator.SetVariables(), // Intentionally placed before ResolveVariableReferencesInLookup, ResolveResourceReferences, // ResolveVariableReferencesInComplexVariables and ResolveVariableReferences. @@ -66,7 +73,7 @@ func Initialize() bundle.Mutator { mutator.ConfigureWSFS(), mutator.TranslatePaths(), - python.WrapperWarning(), + trampoline.WrapperWarning(), permissions.ApplyBundlePermissions(), permissions.FilterCurrentUser(), metadata.AnnotateJobs(), diff --git a/bundle/render/render_text_output.go b/bundle/render/render_text_output.go index 7cca8689..70c1d681 100644 --- a/bundle/render/render_text_output.go +++ b/bundle/render/render_text_output.go @@ -167,7 +167,7 @@ func renderDiagnosticsOnly(out io.Writer, b *bundle.Bundle, diags diag.Diagnosti // Make location relative to bundle root if d.Locations[i].File != "" { - out, err := filepath.Rel(b.RootPath, d.Locations[i].File) + out, err := filepath.Rel(b.BundleRootPath, d.Locations[i].File) // if we can't relativize the path, just use path as-is if err == nil { d.Locations[i].File = out diff --git a/bundle/schema/jsonschema.json b/bundle/schema/jsonschema.json index 2db1a5ab..afdf9fb9 100644 --- a/bundle/schema/jsonschema.json +++ b/bundle/schema/jsonschema.json @@ -59,6 +59,127 @@ "cli": { "bundle": { "config": { + "resources.Cluster": { + "anyOf": [ + { + "type": "object", + "properties": { + "apply_policy_default_values": { + "description": "When set to true, fixed and default values from the policy will be used for fields that are omitted. When set to false, only fixed values from the policy will be applied.", + "$ref": "#/$defs/bool" + }, + "autoscale": { + "description": "Parameters needed in order to automatically scale clusters up and down based on load.\nNote: autoscaling works best with DB runtime versions 3.0 or later.", + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.AutoScale" + }, + "autotermination_minutes": { + "description": "Automatically terminates the cluster after it is inactive for this time in minutes. If not set,\nthis cluster will not be automatically terminated. If specified, the threshold must be between\n10 and 10000 minutes.\nUsers can also set this value to 0 to explicitly disable automatic termination.", + "$ref": "#/$defs/int" + }, + "aws_attributes": { + "description": "Attributes related to clusters running on Amazon Web Services.\nIf not specified at cluster creation, a set of default values will be used.", + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.AwsAttributes" + }, + "azure_attributes": { + "description": "Attributes related to clusters running on Microsoft Azure.\nIf not specified at cluster creation, a set of default values will be used.", + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.AzureAttributes" + }, + "cluster_log_conf": { + "description": "The configuration for delivering spark logs to a long-term storage destination.\nTwo kinds of destinations (dbfs and s3) are supported. Only one destination can be specified\nfor one cluster. If the conf is given, the logs will be delivered to the destination every\n`5 mins`. The destination of driver logs is `$destination/$clusterId/driver`, while\nthe destination of executor logs is `$destination/$clusterId/executor`.", + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.ClusterLogConf" + }, + "cluster_name": { + "description": "Cluster name requested by the user. This doesn't have to be unique.\nIf not specified at creation, the cluster name will be an empty string.\n", + "$ref": "#/$defs/string" + }, + "custom_tags": { + "description": "Additional tags for cluster resources. Databricks will tag all cluster resources (e.g., AWS\ninstances and EBS volumes) with these tags in addition to `default_tags`. Notes:\n\n- Currently, Databricks allows at most 45 custom tags\n\n- Clusters can only reuse cloud resources if the resources' tags are a subset of the cluster tags", + "$ref": "#/$defs/map/string" + }, + "data_security_mode": { + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.DataSecurityMode" + }, + "docker_image": { + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.DockerImage" + }, + "driver_instance_pool_id": { + "description": "The optional ID of the instance pool for the driver of the cluster belongs.\nThe pool cluster uses the instance pool with id (instance_pool_id) if the driver pool is not\nassigned.", + "$ref": "#/$defs/string" + }, + "driver_node_type_id": { + "description": "The node type of the Spark driver. Note that this field is optional;\nif unset, the driver node type will be set as the same value\nas `node_type_id` defined above.\n", + "$ref": "#/$defs/string" + }, + "enable_elastic_disk": { + "description": "Autoscaling Local Storage: when enabled, this cluster will dynamically acquire additional disk\nspace when its Spark workers are running low on disk space. This feature requires specific AWS\npermissions to function correctly - refer to the User Guide for more details.", + "$ref": "#/$defs/bool" + }, + "enable_local_disk_encryption": { + "description": "Whether to enable LUKS on cluster VMs' local disks", + "$ref": "#/$defs/bool" + }, + "gcp_attributes": { + "description": "Attributes related to clusters running on Google Cloud Platform.\nIf not specified at cluster creation, a set of default values will be used.", + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.GcpAttributes" + }, + "init_scripts": { + "description": "The configuration for storing init scripts. Any number of destinations can be specified. The scripts are executed sequentially in the order provided. If `cluster_log_conf` is specified, init script logs are sent to `\u003cdestination\u003e/\u003ccluster-ID\u003e/init_scripts`.", + "$ref": "#/$defs/slice/github.com/databricks/databricks-sdk-go/service/compute.InitScriptInfo" + }, + "instance_pool_id": { + "description": "The optional ID of the instance pool to which the cluster belongs.", + "$ref": "#/$defs/string" + }, + "node_type_id": { + "description": "This field encodes, through a single value, the resources available to each of\nthe Spark nodes in this cluster. For example, the Spark nodes can be provisioned\nand optimized for memory or compute intensive workloads. A list of available node\ntypes can be retrieved by using the :method:clusters/listNodeTypes API call.\n", + "$ref": "#/$defs/string" + }, + "num_workers": { + "description": "Number of worker nodes that this cluster should have. A cluster has one Spark Driver\nand `num_workers` Executors for a total of `num_workers` + 1 Spark nodes.\n\nNote: When reading the properties of a cluster, this field reflects the desired number\nof workers rather than the actual current number of workers. For instance, if a cluster\nis resized from 5 to 10 workers, this field will immediately be updated to reflect\nthe target size of 10 workers, whereas the workers listed in `spark_info` will gradually\nincrease from 5 to 10 as the new nodes are provisioned.", + "$ref": "#/$defs/int" + }, + "permissions": { + "$ref": "#/$defs/slice/github.com/databricks/cli/bundle/config/resources.Permission" + }, + "policy_id": { + "description": "The ID of the cluster policy used to create the cluster if applicable.", + "$ref": "#/$defs/string" + }, + "runtime_engine": { + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.RuntimeEngine" + }, + "single_user_name": { + "description": "Single user name if data_security_mode is `SINGLE_USER`", + "$ref": "#/$defs/string" + }, + "spark_conf": { + "description": "An object containing a set of optional, user-specified Spark configuration key-value pairs.\nUsers can also pass in a string of extra JVM options to the driver and the executors via\n`spark.driver.extraJavaOptions` and `spark.executor.extraJavaOptions` respectively.\n", + "$ref": "#/$defs/map/string" + }, + "spark_env_vars": { + "description": "An object containing a set of optional, user-specified environment variable key-value pairs.\nPlease note that key-value pair of the form (X,Y) will be exported as is (i.e.,\n`export X='Y'`) while launching the driver and workers.\n\nIn order to specify an additional set of `SPARK_DAEMON_JAVA_OPTS`, we recommend appending\nthem to `$SPARK_DAEMON_JAVA_OPTS` as shown in the example below. This ensures that all\ndefault databricks managed environmental variables are included as well.\n\nExample Spark environment variables:\n`{\"SPARK_WORKER_MEMORY\": \"28000m\", \"SPARK_LOCAL_DIRS\": \"/local_disk0\"}` or\n`{\"SPARK_DAEMON_JAVA_OPTS\": \"$SPARK_DAEMON_JAVA_OPTS -Dspark.shuffle.service.enabled=true\"}`", + "$ref": "#/$defs/map/string" + }, + "spark_version": { + "description": "The Spark version of the cluster, e.g. `3.3.x-scala2.11`.\nA list of available Spark versions can be retrieved by using\nthe :method:clusters/sparkVersions API call.\n", + "$ref": "#/$defs/string" + }, + "ssh_public_keys": { + "description": "SSH public key contents that will be added to each Spark node in this cluster. The\ncorresponding private keys can be used to login with the user name `ubuntu` on port `2200`.\nUp to 10 keys can be specified.", + "$ref": "#/$defs/slice/string" + }, + "workload_type": { + "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/compute.WorkloadType" + } + }, + "additionalProperties": false + }, + { + "type": "string", + "pattern": "\\$\\{(var(\\.[a-zA-Z]+([-_]?[a-zA-Z0-9]+)*(\\[[0-9]+\\])*)+)\\}" + } + ] + }, "resources.Grant": { "anyOf": [ { @@ -109,7 +230,7 @@ "$ref": "#/$defs/github.com/databricks/databricks-sdk-go/service/jobs.JobEmailNotifications" }, "environments": { - "description": "A list of task execution environment specifications that can be referenced by tasks of this job.", + "description": "A list of task execution environment specifications that can be referenced by serverless tasks of this job.\nAn environment is required to be present for serverless tasks.\nFor serverless notebook tasks, the environment is accessible in the notebook environment panel.\nFor other serverless tasks, the task environment is required to be specified using environment_key in the task settings.", "$ref": "#/$defs/slice/github.com/databricks/databricks-sdk-go/service/jobs.JobEnvironment" }, "format": { @@ -293,7 +414,7 @@ "$ref": "#/$defs/slice/github.com/databricks/cli/bundle/config/resources.Permission" }, "rate_limits": { - "description": "Rate limits to be applied to the serving endpoint. NOTE: only external and foundation model endpoints are supported as of now.", + "description": "Rate limits to be applied to the serving endpoint. NOTE: this field is deprecated, please use AI Gateway to manage rate limits.", "$ref": "#/$defs/slice/github.com/databricks/databricks-sdk-go/service/serving.RateLimit" }, "route_optimized": { @@ -747,6 +868,9 @@ { "type": "object", "properties": { + "cluster_id": { + "$ref": "#/$defs/string" + }, "compute_id": { "$ref": "#/$defs/string" }, @@ -923,6 +1047,9 @@ { "type": "object", "properties": { + "clusters": { + "$ref": "#/$defs/map/github.com/databricks/cli/bundle/config/resources.Cluster" + }, "experiments": { "$ref": "#/$defs/map/github.com/databricks/cli/bundle/config/resources.MlflowExperiment" }, @@ -990,6 +1117,9 @@ "bundle": { "$ref": "#/$defs/github.com/databricks/cli/bundle/config.Bundle" }, + "cluster_id": { + "$ref": "#/$defs/string" + }, "compute_id": { "$ref": "#/$defs/string" }, @@ -2028,7 +2158,7 @@ }, "compute.RuntimeEngine": { "type": "string", - "description": "Decides which runtime engine to be use, e.g. Standard vs. Photon. If unspecified, the runtime\nengine is inferred from spark_version.", + "description": "Determines the cluster's runtime engine, either standard or Photon.\n\nThis field is not compatible with legacy `spark_version` values that contain `-photon-`.\nRemove `-photon-` from the `spark_version` and set `runtime_engine` to `PHOTON`.\n\nIf left unspecified, the runtime engine defaults to standard unless the spark_version\ncontains -photon-, in which case Photon will be used.\n", "enum": [ "NULL", "STANDARD", @@ -2610,7 +2740,7 @@ "anyOf": [ { "type": "object", - "description": "Write-only setting, available only in Create/Update/Reset and Submit calls. Specifies the user or service principal that the job runs as. If not specified, the job runs as the user who created the job.\n\nOnly `user_name` or `service_principal_name` can be specified. If both are specified, an error is thrown.", + "description": "Write-only setting. Specifies the user, service principal or group that the job/pipeline runs as. If not specified, the job/pipeline runs as the user who created the job/pipeline.\n\nExactly one of `user_name`, `service_principal_name`, `group_name` should be specified. If not, an error is thrown.", "properties": { "service_principal_name": { "description": "Application ID of an active service principal. Setting this field requires the `servicePrincipal/user` role.", @@ -4904,6 +5034,20 @@ "cli": { "bundle": { "config": { + "resources.Cluster": { + "anyOf": [ + { + "type": "object", + "additionalProperties": { + "$ref": "#/$defs/github.com/databricks/cli/bundle/config/resources.Cluster" + } + }, + { + "type": "string", + "pattern": "\\$\\{(var(\\.[a-zA-Z]+([-_]?[a-zA-Z0-9]+)*(\\[[0-9]+\\])*)+)\\}" + } + ] + }, "resources.Job": { "anyOf": [ { diff --git a/bundle/scripts/scripts.go b/bundle/scripts/scripts.go index 629b3a8a..f9e1541e 100644 --- a/bundle/scripts/scripts.go +++ b/bundle/scripts/scripts.go @@ -30,7 +30,7 @@ func (m *script) Name() string { } func (m *script) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics { - executor, err := exec.NewCommandExecutor(b.RootPath) + executor, err := exec.NewCommandExecutor(b.BundleRootPath) if err != nil { return diag.FromErr(err) } diff --git a/bundle/scripts/scripts_test.go b/bundle/scripts/scripts_test.go index 1bc216b6..0c92bc2c 100644 --- a/bundle/scripts/scripts_test.go +++ b/bundle/scripts/scripts_test.go @@ -23,7 +23,7 @@ func TestExecutesHook(t *testing.T) { }, } - executor, err := exec.NewCommandExecutor(b.RootPath) + executor, err := exec.NewCommandExecutor(b.BundleRootPath) require.NoError(t, err) _, out, err := executeHook(context.Background(), executor, b, config.ScriptPreBuild) require.NoError(t, err) diff --git a/bundle/tests/clusters/databricks.yml b/bundle/tests/clusters/databricks.yml new file mode 100644 index 00000000..1074462a --- /dev/null +++ b/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" diff --git a/bundle/tests/clusters_test.go b/bundle/tests/clusters_test.go new file mode 100644 index 00000000..def8a2a3 --- /dev/null +++ b/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) +} diff --git a/bundle/tests/complex_variables_test.go b/bundle/tests/complex_variables_test.go index 6371071c..7a9a53a7 100644 --- a/bundle/tests/complex_variables_test.go +++ b/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) +} diff --git a/bundle/tests/pipeline_glob_paths_test.go b/bundle/tests/pipeline_glob_paths_test.go index c1c62cfb..d17f0404 100644 --- a/bundle/tests/pipeline_glob_paths_test.go +++ b/bundle/tests/pipeline_glob_paths_test.go @@ -11,7 +11,7 @@ func TestExpandPipelineGlobPaths(t *testing.T) { require.NoError(t, diags.Error()) require.Equal( t, - "/Users/user@domain.com/.bundle/pipeline_glob_paths/default/files/dlt/nyc_taxi_loader", + "/Workspace/Users/user@domain.com/.bundle/pipeline_glob_paths/default/files/dlt/nyc_taxi_loader", b.Config.Resources.Pipelines["nyc_taxi_pipeline"].Libraries[0].Notebook.Path, ) } diff --git a/bundle/tests/relative_path_translation_test.go b/bundle/tests/relative_path_translation_test.go index 199871d2..0f553ac3 100644 --- a/bundle/tests/relative_path_translation_test.go +++ b/bundle/tests/relative_path_translation_test.go @@ -12,9 +12,9 @@ func TestRelativePathTranslationDefault(t *testing.T) { require.NoError(t, diags.Error()) t0 := b.Config.Resources.Jobs["job"].Tasks[0] - assert.Equal(t, "/remote/src/file1.py", t0.SparkPythonTask.PythonFile) + assert.Equal(t, "/Workspace/remote/src/file1.py", t0.SparkPythonTask.PythonFile) t1 := b.Config.Resources.Jobs["job"].Tasks[1] - assert.Equal(t, "/remote/src/file1.py", t1.SparkPythonTask.PythonFile) + assert.Equal(t, "/Workspace/remote/src/file1.py", t1.SparkPythonTask.PythonFile) } func TestRelativePathTranslationOverride(t *testing.T) { @@ -22,7 +22,7 @@ func TestRelativePathTranslationOverride(t *testing.T) { require.NoError(t, diags.Error()) t0 := b.Config.Resources.Jobs["job"].Tasks[0] - assert.Equal(t, "/remote/src/file2.py", t0.SparkPythonTask.PythonFile) + assert.Equal(t, "/Workspace/remote/src/file2.py", t0.SparkPythonTask.PythonFile) t1 := b.Config.Resources.Jobs["job"].Tasks[1] - assert.Equal(t, "/remote/src/file2.py", t1.SparkPythonTask.PythonFile) + assert.Equal(t, "/Workspace/remote/src/file2.py", t1.SparkPythonTask.PythonFile) } diff --git a/bundle/tests/variables/complex/databricks.yml b/bundle/tests/variables/complex/databricks.yml index ca27f606..3b32a7c8 100644 --- a/bundle/tests/variables/complex/databricks.yml +++ b/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" diff --git a/bundle/python/conditional_transform_test.go b/bundle/trampoline/conditional_transform_test.go similarity index 92% rename from bundle/python/conditional_transform_test.go rename to bundle/trampoline/conditional_transform_test.go index 1d397f7a..57aa9aac 100644 --- a/bundle/python/conditional_transform_test.go +++ b/bundle/trampoline/conditional_transform_test.go @@ -1,4 +1,4 @@ -package python +package trampoline import ( "context" @@ -17,8 +17,8 @@ func TestNoTransformByDefault(t *testing.T) { tmpDir := t.TempDir() b := &bundle.Bundle{ - RootPath: filepath.Join(tmpDir, "parent", "my_bundle"), - SyncRootPath: filepath.Join(tmpDir, "parent"), + BundleRootPath: filepath.Join(tmpDir, "parent", "my_bundle"), + SyncRootPath: filepath.Join(tmpDir, "parent"), Config: config.Root{ Bundle: config.Bundle{ Target: "development", @@ -66,8 +66,8 @@ func TestTransformWithExperimentalSettingSetToTrue(t *testing.T) { tmpDir := t.TempDir() b := &bundle.Bundle{ - RootPath: filepath.Join(tmpDir, "parent", "my_bundle"), - SyncRootPath: filepath.Join(tmpDir, "parent"), + BundleRootPath: filepath.Join(tmpDir, "parent", "my_bundle"), + SyncRootPath: filepath.Join(tmpDir, "parent"), Config: config.Root{ Bundle: config.Bundle{ Target: "development", diff --git a/bundle/python/warning.go b/bundle/trampoline/python_dbr_warning.go similarity index 99% rename from bundle/python/warning.go rename to bundle/trampoline/python_dbr_warning.go index 0e9d8bef..f62e9eab 100644 --- a/bundle/python/warning.go +++ b/bundle/trampoline/python_dbr_warning.go @@ -1,4 +1,4 @@ -package python +package trampoline import ( "context" diff --git a/bundle/python/warning_test.go b/bundle/trampoline/python_dbr_warning_test.go similarity index 99% rename from bundle/python/warning_test.go rename to bundle/trampoline/python_dbr_warning_test.go index a5ab7563..d293c947 100644 --- a/bundle/python/warning_test.go +++ b/bundle/trampoline/python_dbr_warning_test.go @@ -1,4 +1,4 @@ -package python +package trampoline import ( "context" diff --git a/bundle/python/transform.go b/bundle/trampoline/python_wheel.go similarity index 94% rename from bundle/python/transform.go rename to bundle/trampoline/python_wheel.go index 9d3b1ab6..8e309a62 100644 --- a/bundle/python/transform.go +++ b/bundle/trampoline/python_wheel.go @@ -1,4 +1,4 @@ -package python +package trampoline import ( "context" @@ -69,7 +69,7 @@ func TransformWheelTask() bundle.Mutator { res := b.Config.Experimental != nil && b.Config.Experimental.PythonWheelWrapper return res, nil }, - mutator.NewTrampoline( + NewTrampoline( "python_wheel", &pythonTrampoline{}, NOTEBOOK_TEMPLATE, @@ -94,9 +94,9 @@ func (t *pythonTrampoline) CleanUp(task *jobs.Task) error { return nil } -func (t *pythonTrampoline) GetTasks(b *bundle.Bundle) []mutator.TaskWithJobKey { +func (t *pythonTrampoline) GetTasks(b *bundle.Bundle) []TaskWithJobKey { r := b.Config.Resources - result := make([]mutator.TaskWithJobKey, 0) + result := make([]TaskWithJobKey, 0) for k := range b.Config.Resources.Jobs { tasks := r.Jobs[k].JobSettings.Tasks for i := range tasks { @@ -110,7 +110,7 @@ func (t *pythonTrampoline) GetTasks(b *bundle.Bundle) []mutator.TaskWithJobKey { continue } - result = append(result, mutator.TaskWithJobKey{ + result = append(result, TaskWithJobKey{ JobKey: k, Task: task, }) diff --git a/bundle/python/transform_test.go b/bundle/trampoline/python_wheel_test.go similarity index 98% rename from bundle/python/transform_test.go rename to bundle/trampoline/python_wheel_test.go index c7bddca1..517be35e 100644 --- a/bundle/python/transform_test.go +++ b/bundle/trampoline/python_wheel_test.go @@ -1,4 +1,4 @@ -package python +package trampoline import ( "context" @@ -115,7 +115,7 @@ func TestTransformFiltersWheelTasksOnly(t *testing.T) { func TestNoPanicWithNoPythonWheelTasks(t *testing.T) { tmpDir := t.TempDir() b := &bundle.Bundle{ - RootPath: tmpDir, + BundleRootPath: tmpDir, Config: config.Root{ Bundle: config.Bundle{ Target: "development", diff --git a/bundle/config/mutator/trampoline.go b/bundle/trampoline/trampoline.go similarity index 99% rename from bundle/config/mutator/trampoline.go rename to bundle/trampoline/trampoline.go index dcca5014..1dc1c446 100644 --- a/bundle/config/mutator/trampoline.go +++ b/bundle/trampoline/trampoline.go @@ -1,4 +1,4 @@ -package mutator +package trampoline import ( "context" @@ -23,6 +23,7 @@ type TrampolineFunctions interface { GetTasks(b *bundle.Bundle) []TaskWithJobKey CleanUp(task *jobs.Task) error } + type trampoline struct { name string functions TrampolineFunctions diff --git a/bundle/config/mutator/trampoline_test.go b/bundle/trampoline/trampoline_test.go similarity index 94% rename from bundle/config/mutator/trampoline_test.go rename to bundle/trampoline/trampoline_test.go index 08d3c822..4682d8fa 100644 --- a/bundle/config/mutator/trampoline_test.go +++ b/bundle/trampoline/trampoline_test.go @@ -1,4 +1,4 @@ -package mutator +package trampoline import ( "context" @@ -56,8 +56,8 @@ func TestGenerateTrampoline(t *testing.T) { } b := &bundle.Bundle{ - RootPath: filepath.Join(tmpDir, "parent", "my_bundle"), - SyncRootPath: filepath.Join(tmpDir, "parent"), + BundleRootPath: filepath.Join(tmpDir, "parent", "my_bundle"), + SyncRootPath: filepath.Join(tmpDir, "parent"), Config: config.Root{ Workspace: config.Workspace{ FilePath: "/Workspace/files", diff --git a/cmd/account/disable-legacy-features/disable-legacy-features.go b/cmd/account/disable-legacy-features/disable-legacy-features.go new file mode 100755 index 00000000..6d25b943 --- /dev/null +++ b/cmd/account/disable-legacy-features/disable-legacy-features.go @@ -0,0 +1,215 @@ +// Code generated from OpenAPI specs by Databricks SDK Generator. DO NOT EDIT. + +package disable_legacy_features + +import ( + "fmt" + + "github.com/databricks/cli/cmd/root" + "github.com/databricks/cli/libs/cmdio" + "github.com/databricks/cli/libs/flags" + "github.com/databricks/databricks-sdk-go/service/settings" + "github.com/spf13/cobra" +) + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var cmdOverrides []func(*cobra.Command) + +func New() *cobra.Command { + cmd := &cobra.Command{ + Use: "disable-legacy-features", + Short: `Disable legacy features for new Databricks workspaces.`, + Long: `Disable legacy features for new Databricks workspaces. + + For newly created workspaces: 1. Disables the use of DBFS root and mounts. 2. + Hive Metastore will not be provisioned. 3. Disables the use of ‘No-isolation + clusters’. 4. Disables Databricks Runtime versions prior to 13.3LTS.`, + + // This service is being previewed; hide from help output. + Hidden: true, + } + + // Add methods + cmd.AddCommand(newDelete()) + cmd.AddCommand(newGet()) + cmd.AddCommand(newUpdate()) + + // Apply optional overrides to this command. + for _, fn := range cmdOverrides { + fn(cmd) + } + + return cmd +} + +// start delete command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var deleteOverrides []func( + *cobra.Command, + *settings.DeleteDisableLegacyFeaturesRequest, +) + +func newDelete() *cobra.Command { + cmd := &cobra.Command{} + + var deleteReq settings.DeleteDisableLegacyFeaturesRequest + + // TODO: short flags + + cmd.Flags().StringVar(&deleteReq.Etag, "etag", deleteReq.Etag, `etag used for versioning.`) + + cmd.Use = "delete" + cmd.Short = `Delete the disable legacy features setting.` + cmd.Long = `Delete the disable legacy features setting. + + Deletes the disable legacy features setting.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(0) + return check(cmd, args) + } + + cmd.PreRunE = root.MustAccountClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + a := root.AccountClient(ctx) + + response, err := a.Settings.DisableLegacyFeatures().Delete(ctx, deleteReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range deleteOverrides { + fn(cmd, &deleteReq) + } + + return cmd +} + +// start get command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var getOverrides []func( + *cobra.Command, + *settings.GetDisableLegacyFeaturesRequest, +) + +func newGet() *cobra.Command { + cmd := &cobra.Command{} + + var getReq settings.GetDisableLegacyFeaturesRequest + + // TODO: short flags + + cmd.Flags().StringVar(&getReq.Etag, "etag", getReq.Etag, `etag used for versioning.`) + + cmd.Use = "get" + cmd.Short = `Get the disable legacy features setting.` + cmd.Long = `Get the disable legacy features setting. + + Gets the value of the disable legacy features setting.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(0) + return check(cmd, args) + } + + cmd.PreRunE = root.MustAccountClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + a := root.AccountClient(ctx) + + response, err := a.Settings.DisableLegacyFeatures().Get(ctx, getReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range getOverrides { + fn(cmd, &getReq) + } + + return cmd +} + +// start update command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var updateOverrides []func( + *cobra.Command, + *settings.UpdateDisableLegacyFeaturesRequest, +) + +func newUpdate() *cobra.Command { + cmd := &cobra.Command{} + + var updateReq settings.UpdateDisableLegacyFeaturesRequest + var updateJson flags.JsonFlag + + // TODO: short flags + cmd.Flags().Var(&updateJson, "json", `either inline JSON string or @path/to/file.json with request body`) + + cmd.Use = "update" + cmd.Short = `Update the disable legacy features setting.` + cmd.Long = `Update the disable legacy features setting. + + Updates the value of the disable legacy features setting.` + + cmd.Annotations = make(map[string]string) + + cmd.PreRunE = root.MustAccountClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + a := root.AccountClient(ctx) + + if cmd.Flags().Changed("json") { + err = updateJson.Unmarshal(&updateReq) + if err != nil { + return err + } + } else { + return fmt.Errorf("please provide command input in JSON format by specifying the --json flag") + } + + response, err := a.Settings.DisableLegacyFeatures().Update(ctx, updateReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range updateOverrides { + fn(cmd, &updateReq) + } + + return cmd +} + +// end service DisableLegacyFeatures diff --git a/cmd/account/settings/settings.go b/cmd/account/settings/settings.go index a750e81e..9a9cd44b 100755 --- a/cmd/account/settings/settings.go +++ b/cmd/account/settings/settings.go @@ -6,6 +6,7 @@ import ( "github.com/spf13/cobra" csp_enablement_account "github.com/databricks/cli/cmd/account/csp-enablement-account" + disable_legacy_features "github.com/databricks/cli/cmd/account/disable-legacy-features" esm_enablement_account "github.com/databricks/cli/cmd/account/esm-enablement-account" personal_compute "github.com/databricks/cli/cmd/account/personal-compute" ) @@ -27,6 +28,7 @@ func New() *cobra.Command { // Add subservices cmd.AddCommand(csp_enablement_account.New()) + cmd.AddCommand(disable_legacy_features.New()) cmd.AddCommand(esm_enablement_account.New()) cmd.AddCommand(personal_compute.New()) diff --git a/cmd/bundle/deploy.go b/cmd/bundle/deploy.go index a0ef99c5..a25e02f6 100644 --- a/cmd/bundle/deploy.go +++ b/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), )), ) } diff --git a/cmd/bundle/generate/generate_test.go b/cmd/bundle/generate/generate_test.go index 7de6805f..943f721c 100644 --- a/cmd/bundle/generate/generate_test.go +++ b/cmd/bundle/generate/generate_test.go @@ -24,7 +24,7 @@ func TestGeneratePipelineCommand(t *testing.T) { root := t.TempDir() b := &bundle.Bundle{ - RootPath: root, + BundleRootPath: root, } m := mocks.NewMockWorkspaceClient(t) @@ -122,7 +122,7 @@ func TestGenerateJobCommand(t *testing.T) { root := t.TempDir() b := &bundle.Bundle{ - RootPath: root, + BundleRootPath: root, } m := mocks.NewMockWorkspaceClient(t) diff --git a/cmd/bundle/run.go b/cmd/bundle/run.go index 63458f85..9ef5eb8f 100644 --- a/cmd/bundle/run.go +++ b/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 diff --git a/cmd/sync/sync.go b/cmd/sync/sync.go index 23a4c018..2092d9e3 100644 --- a/cmd/sync/sync.go +++ b/cmd/sync/sync.go @@ -6,7 +6,6 @@ import ( "fmt" "io" "path/filepath" - stdsync "sync" "time" "github.com/databricks/cli/bundle" @@ -46,6 +45,21 @@ func (f *syncFlags) syncOptionsFromArgs(cmd *cobra.Command, args []string) (*syn return nil, flag.ErrHelp } + var outputFunc func(context.Context, <-chan sync.Event, io.Writer) + switch f.output { + case flags.OutputText: + outputFunc = sync.TextOutput + case flags.OutputJSON: + outputFunc = sync.JsonOutput + } + + var outputHandler sync.OutputHandler + if outputFunc != nil { + outputHandler = func(ctx context.Context, events <-chan sync.Event) { + outputFunc(ctx, events, cmd.OutOrStdout()) + } + } + opts := sync.SyncOptions{ LocalRoot: vfs.MustNew(args[0]), Paths: []string{"."}, @@ -62,6 +76,8 @@ func (f *syncFlags) syncOptionsFromArgs(cmd *cobra.Command, args []string) (*syn // exist and add it to the `.gitignore` file in the root. SnapshotBasePath: filepath.Join(args[0], ".databricks"), WorkspaceClient: root.WorkspaceClient(cmd.Context()), + + OutputHandler: outputHandler, } return &opts, nil } @@ -118,23 +134,7 @@ func New() *cobra.Command { if err != nil { return err } - - var outputFunc func(context.Context, <-chan sync.Event, io.Writer) - switch f.output { - case flags.OutputText: - outputFunc = textOutput - case flags.OutputJSON: - outputFunc = jsonOutput - } - - var wg stdsync.WaitGroup - if outputFunc != nil { - wg.Add(1) - go func() { - defer wg.Done() - outputFunc(ctx, s.Events(), cmd.OutOrStdout()) - }() - } + defer s.Close() if f.watch { err = s.RunContinuous(ctx) @@ -142,8 +142,6 @@ func New() *cobra.Command { _, err = s.RunOnce(ctx) } - s.Close() - wg.Wait() return err } diff --git a/cmd/sync/sync_test.go b/cmd/sync/sync_test.go index bd03eec9..8f65aedb 100644 --- a/cmd/sync/sync_test.go +++ b/cmd/sync/sync_test.go @@ -17,10 +17,10 @@ import ( func TestSyncOptionsFromBundle(t *testing.T) { tempDir := t.TempDir() b := &bundle.Bundle{ - RootPath: tempDir, - BundleRoot: vfs.MustNew(tempDir), - SyncRootPath: tempDir, - SyncRoot: vfs.MustNew(tempDir), + BundleRootPath: tempDir, + BundleRoot: vfs.MustNew(tempDir), + SyncRootPath: tempDir, + SyncRoot: vfs.MustNew(tempDir), Config: config.Root{ Bundle: config.Bundle{ Target: "default", diff --git a/cmd/workspace/apps/apps.go b/cmd/workspace/apps/apps.go index bc3fbe92..baec6d03 100755 --- a/cmd/workspace/apps/apps.go +++ b/cmd/workspace/apps/apps.go @@ -75,8 +75,8 @@ func newCreate() *cobra.Command { var createSkipWait bool var createTimeout time.Duration - cmd.Flags().BoolVar(&createSkipWait, "no-wait", createSkipWait, `do not wait to reach IDLE state`) - cmd.Flags().DurationVar(&createTimeout, "timeout", 20*time.Minute, `maximum amount of time to reach IDLE state`) + cmd.Flags().BoolVar(&createSkipWait, "no-wait", createSkipWait, `do not wait to reach ACTIVE state`) + cmd.Flags().DurationVar(&createTimeout, "timeout", 20*time.Minute, `maximum amount of time to reach ACTIVE state`) // TODO: short flags cmd.Flags().Var(&createJson, "json", `either inline JSON string or @path/to/file.json with request body`) @@ -130,13 +130,13 @@ func newCreate() *cobra.Command { } spinner := cmdio.Spinner(ctx) info, err := wait.OnProgress(func(i *apps.App) { - if i.Status == nil { + if i.ComputeStatus == nil { return } - status := i.Status.State + status := i.ComputeStatus.State statusMessage := fmt.Sprintf("current status: %s", status) - if i.Status != nil { - statusMessage = i.Status.Message + if i.ComputeStatus != nil { + statusMessage = i.ComputeStatus.Message } spinner <- statusMessage }).GetWithTimeout(createTimeout) @@ -198,11 +198,11 @@ func newDelete() *cobra.Command { deleteReq.Name = args[0] - err = w.Apps.Delete(ctx, deleteReq) + response, err := w.Apps.Delete(ctx, deleteReq) if err != nil { return err } - return nil + return cmdio.Render(ctx, response) } // Disable completions since they are not applicable. @@ -240,35 +240,23 @@ func newDeploy() *cobra.Command { // TODO: short flags cmd.Flags().Var(&deployJson, "json", `either inline JSON string or @path/to/file.json with request body`) + cmd.Flags().StringVar(&deployReq.DeploymentId, "deployment-id", deployReq.DeploymentId, `The unique id of the deployment.`) cmd.Flags().Var(&deployReq.Mode, "mode", `The mode of which the deployment will manage the source code. Supported values: [AUTO_SYNC, SNAPSHOT]`) + cmd.Flags().StringVar(&deployReq.SourceCodePath, "source-code-path", deployReq.SourceCodePath, `The workspace file system path of the source code used to create the app deployment.`) - cmd.Use = "deploy APP_NAME SOURCE_CODE_PATH" + cmd.Use = "deploy APP_NAME" cmd.Short = `Create an app deployment.` cmd.Long = `Create an app deployment. Creates an app deployment for the app with the supplied name. Arguments: - APP_NAME: The name of the app. - SOURCE_CODE_PATH: The workspace file system path of the source code used to create the app - deployment. This is different from - deployment_artifacts.source_code_path, which is the path used by the - deployed app. The former refers to the original source code location of - the app in the workspace during deployment creation, whereas the latter - provides a system generated stable snapshotted source code path used by - the deployment.` + APP_NAME: The name of the app.` cmd.Annotations = make(map[string]string) cmd.Args = func(cmd *cobra.Command, args []string) error { - if cmd.Flags().Changed("json") { - err := root.ExactArgs(1)(cmd, args) - if err != nil { - return fmt.Errorf("when --json flag is specified, provide only APP_NAME as positional arguments. Provide 'source_code_path' in your JSON input") - } - return nil - } - check := root.ExactArgs(2) + check := root.ExactArgs(1) return check(cmd, args) } @@ -284,9 +272,6 @@ func newDeploy() *cobra.Command { } } deployReq.AppName = args[0] - if !cmd.Flags().Changed("json") { - deployReq.SourceCodePath = args[1] - } wait, err := w.Apps.Deploy(ctx, deployReq) if err != nil { @@ -759,8 +744,8 @@ func newStart() *cobra.Command { var startSkipWait bool var startTimeout time.Duration - cmd.Flags().BoolVar(&startSkipWait, "no-wait", startSkipWait, `do not wait to reach SUCCEEDED state`) - cmd.Flags().DurationVar(&startTimeout, "timeout", 20*time.Minute, `maximum amount of time to reach SUCCEEDED state`) + cmd.Flags().BoolVar(&startSkipWait, "no-wait", startSkipWait, `do not wait to reach ACTIVE state`) + cmd.Flags().DurationVar(&startTimeout, "timeout", 20*time.Minute, `maximum amount of time to reach ACTIVE state`) // TODO: short flags cmd.Use = "start NAME" @@ -794,14 +779,14 @@ func newStart() *cobra.Command { return cmdio.Render(ctx, wait.Response) } spinner := cmdio.Spinner(ctx) - info, err := wait.OnProgress(func(i *apps.AppDeployment) { - if i.Status == nil { + info, err := wait.OnProgress(func(i *apps.App) { + if i.ComputeStatus == nil { return } - status := i.Status.State + status := i.ComputeStatus.State statusMessage := fmt.Sprintf("current status: %s", status) - if i.Status != nil { - statusMessage = i.Status.Message + if i.ComputeStatus != nil { + statusMessage = i.ComputeStatus.Message } spinner <- statusMessage }).GetWithTimeout(startTimeout) @@ -838,6 +823,11 @@ func newStop() *cobra.Command { var stopReq apps.StopAppRequest + var stopSkipWait bool + var stopTimeout time.Duration + + cmd.Flags().BoolVar(&stopSkipWait, "no-wait", stopSkipWait, `do not wait to reach STOPPED state`) + cmd.Flags().DurationVar(&stopTimeout, "timeout", 20*time.Minute, `maximum amount of time to reach STOPPED state`) // TODO: short flags cmd.Use = "stop NAME" @@ -863,11 +853,30 @@ func newStop() *cobra.Command { stopReq.Name = args[0] - err = w.Apps.Stop(ctx, stopReq) + wait, err := w.Apps.Stop(ctx, stopReq) if err != nil { return err } - return nil + if stopSkipWait { + return cmdio.Render(ctx, wait.Response) + } + spinner := cmdio.Spinner(ctx) + info, err := wait.OnProgress(func(i *apps.App) { + if i.ComputeStatus == nil { + return + } + status := i.ComputeStatus.State + statusMessage := fmt.Sprintf("current status: %s", status) + if i.ComputeStatus != nil { + statusMessage = i.ComputeStatus.Message + } + spinner <- statusMessage + }).GetWithTimeout(stopTimeout) + close(spinner) + if err != nil { + return err + } + return cmdio.Render(ctx, info) } // Disable completions since they are not applicable. diff --git a/cmd/workspace/clusters/clusters.go b/cmd/workspace/clusters/clusters.go index a64a6ab7..b36102d9 100755 --- a/cmd/workspace/clusters/clusters.go +++ b/cmd/workspace/clusters/clusters.go @@ -217,7 +217,7 @@ func newCreate() *cobra.Command { cmd.Flags().StringVar(&createReq.NodeTypeId, "node-type-id", createReq.NodeTypeId, `This field encodes, through a single value, the resources available to each of the Spark nodes in this cluster.`) cmd.Flags().IntVar(&createReq.NumWorkers, "num-workers", createReq.NumWorkers, `Number of worker nodes that this cluster should have.`) cmd.Flags().StringVar(&createReq.PolicyId, "policy-id", createReq.PolicyId, `The ID of the cluster policy used to create the cluster if applicable.`) - cmd.Flags().Var(&createReq.RuntimeEngine, "runtime-engine", `Decides which runtime engine to be use, e.g. Supported values: [NULL, PHOTON, STANDARD]`) + cmd.Flags().Var(&createReq.RuntimeEngine, "runtime-engine", `Determines the cluster's runtime engine, either standard or Photon. Supported values: [NULL, PHOTON, STANDARD]`) cmd.Flags().StringVar(&createReq.SingleUserName, "single-user-name", createReq.SingleUserName, `Single user name if data_security_mode is SINGLE_USER.`) // TODO: map via StringToStringVar: spark_conf // TODO: map via StringToStringVar: spark_env_vars @@ -236,6 +236,12 @@ func newCreate() *cobra.Command { If Databricks acquires at least 85% of the requested on-demand nodes, cluster creation will succeed. Otherwise the cluster will terminate with an informative error message. + + Rather than authoring the cluster's JSON definition from scratch, Databricks + recommends filling out the [create compute UI] and then copying the generated + JSON definition from the UI. + + [create compute UI]: https://docs.databricks.com/compute/configure.html Arguments: SPARK_VERSION: The Spark version of the cluster, e.g. 3.3.x-scala2.11. A list of @@ -463,7 +469,7 @@ func newEdit() *cobra.Command { cmd.Flags().StringVar(&editReq.NodeTypeId, "node-type-id", editReq.NodeTypeId, `This field encodes, through a single value, the resources available to each of the Spark nodes in this cluster.`) cmd.Flags().IntVar(&editReq.NumWorkers, "num-workers", editReq.NumWorkers, `Number of worker nodes that this cluster should have.`) cmd.Flags().StringVar(&editReq.PolicyId, "policy-id", editReq.PolicyId, `The ID of the cluster policy used to create the cluster if applicable.`) - cmd.Flags().Var(&editReq.RuntimeEngine, "runtime-engine", `Decides which runtime engine to be use, e.g. Supported values: [NULL, PHOTON, STANDARD]`) + cmd.Flags().Var(&editReq.RuntimeEngine, "runtime-engine", `Determines the cluster's runtime engine, either standard or Photon. Supported values: [NULL, PHOTON, STANDARD]`) cmd.Flags().StringVar(&editReq.SingleUserName, "single-user-name", editReq.SingleUserName, `Single user name if data_security_mode is SINGLE_USER.`) // TODO: map via StringToStringVar: spark_conf // TODO: map via StringToStringVar: spark_env_vars diff --git a/cmd/workspace/cmd.go b/cmd/workspace/cmd.go index 11be8077..3fe5b268 100755 --- a/cmd/workspace/cmd.go +++ b/cmd/workspace/cmd.go @@ -76,6 +76,7 @@ import ( system_schemas "github.com/databricks/cli/cmd/workspace/system-schemas" table_constraints "github.com/databricks/cli/cmd/workspace/table-constraints" tables "github.com/databricks/cli/cmd/workspace/tables" + temporary_table_credentials "github.com/databricks/cli/cmd/workspace/temporary-table-credentials" token_management "github.com/databricks/cli/cmd/workspace/token-management" tokens "github.com/databricks/cli/cmd/workspace/tokens" users "github.com/databricks/cli/cmd/workspace/users" @@ -165,6 +166,7 @@ func All() []*cobra.Command { out = append(out, system_schemas.New()) out = append(out, table_constraints.New()) out = append(out, tables.New()) + out = append(out, temporary_table_credentials.New()) out = append(out, token_management.New()) out = append(out, tokens.New()) out = append(out, users.New()) diff --git a/cmd/workspace/disable-legacy-access/disable-legacy-access.go b/cmd/workspace/disable-legacy-access/disable-legacy-access.go new file mode 100755 index 00000000..fea2b3c4 --- /dev/null +++ b/cmd/workspace/disable-legacy-access/disable-legacy-access.go @@ -0,0 +1,217 @@ +// Code generated from OpenAPI specs by Databricks SDK Generator. DO NOT EDIT. + +package disable_legacy_access + +import ( + "fmt" + + "github.com/databricks/cli/cmd/root" + "github.com/databricks/cli/libs/cmdio" + "github.com/databricks/cli/libs/flags" + "github.com/databricks/databricks-sdk-go/service/settings" + "github.com/spf13/cobra" +) + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var cmdOverrides []func(*cobra.Command) + +func New() *cobra.Command { + cmd := &cobra.Command{ + Use: "disable-legacy-access", + Short: `'Disabling legacy access' has the following impacts: 1.`, + Long: `'Disabling legacy access' has the following impacts: + + 1. Disables direct access to the Hive Metastore. However, you can still access + Hive Metastore through HMS Federation. 2. Disables Fallback Mode (docs link) + on any External Location access from the workspace. 3. Alters DBFS path access + to use External Location permissions in place of legacy credentials. 4. + Enforces Unity Catalog access on all path based access.`, + + // This service is being previewed; hide from help output. + Hidden: true, + } + + // Add methods + cmd.AddCommand(newDelete()) + cmd.AddCommand(newGet()) + cmd.AddCommand(newUpdate()) + + // Apply optional overrides to this command. + for _, fn := range cmdOverrides { + fn(cmd) + } + + return cmd +} + +// start delete command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var deleteOverrides []func( + *cobra.Command, + *settings.DeleteDisableLegacyAccessRequest, +) + +func newDelete() *cobra.Command { + cmd := &cobra.Command{} + + var deleteReq settings.DeleteDisableLegacyAccessRequest + + // TODO: short flags + + cmd.Flags().StringVar(&deleteReq.Etag, "etag", deleteReq.Etag, `etag used for versioning.`) + + cmd.Use = "delete" + cmd.Short = `Delete Legacy Access Disablement Status.` + cmd.Long = `Delete Legacy Access Disablement Status. + + Deletes legacy access disablement status.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(0) + return check(cmd, args) + } + + cmd.PreRunE = root.MustWorkspaceClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + w := root.WorkspaceClient(ctx) + + response, err := w.Settings.DisableLegacyAccess().Delete(ctx, deleteReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range deleteOverrides { + fn(cmd, &deleteReq) + } + + return cmd +} + +// start get command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var getOverrides []func( + *cobra.Command, + *settings.GetDisableLegacyAccessRequest, +) + +func newGet() *cobra.Command { + cmd := &cobra.Command{} + + var getReq settings.GetDisableLegacyAccessRequest + + // TODO: short flags + + cmd.Flags().StringVar(&getReq.Etag, "etag", getReq.Etag, `etag used for versioning.`) + + cmd.Use = "get" + cmd.Short = `Retrieve Legacy Access Disablement Status.` + cmd.Long = `Retrieve Legacy Access Disablement Status. + + Retrieves legacy access disablement Status.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(0) + return check(cmd, args) + } + + cmd.PreRunE = root.MustWorkspaceClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + w := root.WorkspaceClient(ctx) + + response, err := w.Settings.DisableLegacyAccess().Get(ctx, getReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range getOverrides { + fn(cmd, &getReq) + } + + return cmd +} + +// start update command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var updateOverrides []func( + *cobra.Command, + *settings.UpdateDisableLegacyAccessRequest, +) + +func newUpdate() *cobra.Command { + cmd := &cobra.Command{} + + var updateReq settings.UpdateDisableLegacyAccessRequest + var updateJson flags.JsonFlag + + // TODO: short flags + cmd.Flags().Var(&updateJson, "json", `either inline JSON string or @path/to/file.json with request body`) + + cmd.Use = "update" + cmd.Short = `Update Legacy Access Disablement Status.` + cmd.Long = `Update Legacy Access Disablement Status. + + Updates legacy access disablement status.` + + cmd.Annotations = make(map[string]string) + + cmd.PreRunE = root.MustWorkspaceClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + w := root.WorkspaceClient(ctx) + + if cmd.Flags().Changed("json") { + err = updateJson.Unmarshal(&updateReq) + if err != nil { + return err + } + } else { + return fmt.Errorf("please provide command input in JSON format by specifying the --json flag") + } + + response, err := w.Settings.DisableLegacyAccess().Update(ctx, updateReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range updateOverrides { + fn(cmd, &updateReq) + } + + return cmd +} + +// end service DisableLegacyAccess diff --git a/cmd/workspace/pipelines/pipelines.go b/cmd/workspace/pipelines/pipelines.go index f1cc4e3f..5b4d9645 100755 --- a/cmd/workspace/pipelines/pipelines.go +++ b/cmd/workspace/pipelines/pipelines.go @@ -935,6 +935,7 @@ func newUpdate() *cobra.Command { cmd.Flags().Var(&updateJson, "json", `either inline JSON string or @path/to/file.json with request body`) cmd.Flags().BoolVar(&updateReq.AllowDuplicateNames, "allow-duplicate-names", updateReq.AllowDuplicateNames, `If false, deployment will fail if name has changed and conflicts the name of another pipeline.`) + cmd.Flags().StringVar(&updateReq.BudgetPolicyId, "budget-policy-id", updateReq.BudgetPolicyId, `Budget policy of this pipeline.`) cmd.Flags().StringVar(&updateReq.Catalog, "catalog", updateReq.Catalog, `A catalog in Unity Catalog to publish data from this pipeline to.`) cmd.Flags().StringVar(&updateReq.Channel, "channel", updateReq.Channel, `DLT Release Channel that specifies which version to use.`) // TODO: array: clusters diff --git a/cmd/workspace/serving-endpoints/serving-endpoints.go b/cmd/workspace/serving-endpoints/serving-endpoints.go index b92f824d..0837652d 100755 --- a/cmd/workspace/serving-endpoints/serving-endpoints.go +++ b/cmd/workspace/serving-endpoints/serving-endpoints.go @@ -53,6 +53,7 @@ func New() *cobra.Command { cmd.AddCommand(newLogs()) cmd.AddCommand(newPatch()) cmd.AddCommand(newPut()) + cmd.AddCommand(newPutAiGateway()) cmd.AddCommand(newQuery()) cmd.AddCommand(newSetPermissions()) cmd.AddCommand(newUpdateConfig()) @@ -151,6 +152,7 @@ func newCreate() *cobra.Command { // TODO: short flags cmd.Flags().Var(&createJson, "json", `either inline JSON string or @path/to/file.json with request body`) + // TODO: complex arg: ai_gateway // TODO: array: rate_limits cmd.Flags().BoolVar(&createReq.RouteOptimized, "route-optimized", createReq.RouteOptimized, `Enable route optimization for the serving endpoint.`) // TODO: array: tags @@ -754,8 +756,9 @@ func newPut() *cobra.Command { cmd.Short = `Update rate limits of a serving endpoint.` cmd.Long = `Update rate limits of a serving endpoint. - Used to update the rate limits of a serving endpoint. NOTE: only external and - foundation model endpoints are supported as of now. + Used to update the rate limits of a serving endpoint. NOTE: Only foundation + model endpoints are currently supported. For external models, use AI Gateway + to manage rate limits. Arguments: NAME: The name of the serving endpoint whose rate limits are being updated. This @@ -800,6 +803,79 @@ func newPut() *cobra.Command { return cmd } +// start put-ai-gateway command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var putAiGatewayOverrides []func( + *cobra.Command, + *serving.PutAiGatewayRequest, +) + +func newPutAiGateway() *cobra.Command { + cmd := &cobra.Command{} + + var putAiGatewayReq serving.PutAiGatewayRequest + var putAiGatewayJson flags.JsonFlag + + // TODO: short flags + cmd.Flags().Var(&putAiGatewayJson, "json", `either inline JSON string or @path/to/file.json with request body`) + + // TODO: complex arg: guardrails + // TODO: complex arg: inference_table_config + // TODO: array: rate_limits + // TODO: complex arg: usage_tracking_config + + cmd.Use = "put-ai-gateway NAME" + cmd.Short = `Update AI Gateway of a serving endpoint.` + cmd.Long = `Update AI Gateway of a serving endpoint. + + Used to update the AI Gateway of a serving endpoint. NOTE: Only external model + endpoints are currently supported. + + Arguments: + NAME: The name of the serving endpoint whose AI Gateway is being updated. This + field is required.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(1) + return check(cmd, args) + } + + cmd.PreRunE = root.MustWorkspaceClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + w := root.WorkspaceClient(ctx) + + if cmd.Flags().Changed("json") { + err = putAiGatewayJson.Unmarshal(&putAiGatewayReq) + if err != nil { + return err + } + } + putAiGatewayReq.Name = args[0] + + response, err := w.ServingEndpoints.PutAiGateway(ctx, putAiGatewayReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range putAiGatewayOverrides { + fn(cmd, &putAiGatewayReq) + } + + return cmd +} + // start query command // Slice with functions to override default command behavior. diff --git a/cmd/workspace/settings/settings.go b/cmd/workspace/settings/settings.go index 214986c7..aaeecf41 100755 --- a/cmd/workspace/settings/settings.go +++ b/cmd/workspace/settings/settings.go @@ -8,6 +8,7 @@ import ( automatic_cluster_update "github.com/databricks/cli/cmd/workspace/automatic-cluster-update" compliance_security_profile "github.com/databricks/cli/cmd/workspace/compliance-security-profile" default_namespace "github.com/databricks/cli/cmd/workspace/default-namespace" + disable_legacy_access "github.com/databricks/cli/cmd/workspace/disable-legacy-access" enhanced_security_monitoring "github.com/databricks/cli/cmd/workspace/enhanced-security-monitoring" restrict_workspace_admins "github.com/databricks/cli/cmd/workspace/restrict-workspace-admins" ) @@ -31,6 +32,7 @@ func New() *cobra.Command { cmd.AddCommand(automatic_cluster_update.New()) cmd.AddCommand(compliance_security_profile.New()) cmd.AddCommand(default_namespace.New()) + cmd.AddCommand(disable_legacy_access.New()) cmd.AddCommand(enhanced_security_monitoring.New()) cmd.AddCommand(restrict_workspace_admins.New()) diff --git a/cmd/workspace/tables/tables.go b/cmd/workspace/tables/tables.go index 4564b4fe..ec297f29 100755 --- a/cmd/workspace/tables/tables.go +++ b/cmd/workspace/tables/tables.go @@ -220,6 +220,7 @@ func newGet() *cobra.Command { cmd.Flags().BoolVar(&getReq.IncludeBrowse, "include-browse", getReq.IncludeBrowse, `Whether to include tables in the response for which the principal can only access selective metadata for.`) cmd.Flags().BoolVar(&getReq.IncludeDeltaMetadata, "include-delta-metadata", getReq.IncludeDeltaMetadata, `Whether delta metadata should be included in the response.`) + cmd.Flags().BoolVar(&getReq.IncludeManifestCapabilities, "include-manifest-capabilities", getReq.IncludeManifestCapabilities, `Whether to include a manifest containing capabilities the table has.`) cmd.Use = "get FULL_NAME" cmd.Short = `Get a table.` @@ -299,6 +300,7 @@ func newList() *cobra.Command { cmd.Flags().BoolVar(&listReq.IncludeBrowse, "include-browse", listReq.IncludeBrowse, `Whether to include tables in the response for which the principal can only access selective metadata for.`) cmd.Flags().BoolVar(&listReq.IncludeDeltaMetadata, "include-delta-metadata", listReq.IncludeDeltaMetadata, `Whether delta metadata should be included in the response.`) + cmd.Flags().BoolVar(&listReq.IncludeManifestCapabilities, "include-manifest-capabilities", listReq.IncludeManifestCapabilities, `Whether to include a manifest containing capabilities the table has.`) cmd.Flags().IntVar(&listReq.MaxResults, "max-results", listReq.MaxResults, `Maximum number of tables to return.`) cmd.Flags().BoolVar(&listReq.OmitColumns, "omit-columns", listReq.OmitColumns, `Whether to omit the columns of the table from the response or not.`) cmd.Flags().BoolVar(&listReq.OmitProperties, "omit-properties", listReq.OmitProperties, `Whether to omit the properties of the table from the response or not.`) @@ -366,6 +368,7 @@ func newListSummaries() *cobra.Command { // TODO: short flags + cmd.Flags().BoolVar(&listSummariesReq.IncludeManifestCapabilities, "include-manifest-capabilities", listSummariesReq.IncludeManifestCapabilities, `Whether to include a manifest containing capabilities the table has.`) cmd.Flags().IntVar(&listSummariesReq.MaxResults, "max-results", listSummariesReq.MaxResults, `Maximum number of summaries for tables to return.`) cmd.Flags().StringVar(&listSummariesReq.PageToken, "page-token", listSummariesReq.PageToken, `Opaque pagination token to go to next page based on previous query.`) cmd.Flags().StringVar(&listSummariesReq.SchemaNamePattern, "schema-name-pattern", listSummariesReq.SchemaNamePattern, `A sql LIKE pattern (% and _) for schema names.`) diff --git a/cmd/workspace/temporary-table-credentials/temporary-table-credentials.go b/cmd/workspace/temporary-table-credentials/temporary-table-credentials.go new file mode 100755 index 00000000..8718f7ba --- /dev/null +++ b/cmd/workspace/temporary-table-credentials/temporary-table-credentials.go @@ -0,0 +1,122 @@ +// Code generated from OpenAPI specs by Databricks SDK Generator. DO NOT EDIT. + +package temporary_table_credentials + +import ( + "github.com/databricks/cli/cmd/root" + "github.com/databricks/cli/libs/cmdio" + "github.com/databricks/cli/libs/flags" + "github.com/databricks/databricks-sdk-go/service/catalog" + "github.com/spf13/cobra" +) + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var cmdOverrides []func(*cobra.Command) + +func New() *cobra.Command { + cmd := &cobra.Command{ + Use: "temporary-table-credentials", + Short: `Temporary Table Credentials refer to short-lived, downscoped credentials used to access cloud storage locationswhere table data is stored in Databricks.`, + Long: `Temporary Table Credentials refer to short-lived, downscoped credentials used + to access cloud storage locationswhere table data is stored in Databricks. + These credentials are employed to provide secure and time-limitedaccess to + data in cloud environments such as AWS, Azure, and Google Cloud. Each cloud + provider has its own typeof credentials: AWS uses temporary session tokens via + AWS Security Token Service (STS), Azure utilizesShared Access Signatures (SAS) + for its data storage services, and Google Cloud supports temporary + credentialsthrough OAuth 2.0.Temporary table credentials ensure that data + access is limited in scope and duration, reducing the risk ofunauthorized + access or misuse. To use the temporary table credentials API, a metastore + admin needs to enable the external_access_enabled flag (off by default) at the + metastore level, and user needs to be granted the EXTERNAL USE SCHEMA + permission at the schema level by catalog admin. Note that EXTERNAL USE SCHEMA + is a schema level permission that can only be granted by catalog admin + explicitly and is not included in schema ownership or ALL PRIVILEGES on the + schema for security reason.`, + GroupID: "catalog", + Annotations: map[string]string{ + "package": "catalog", + }, + } + + // Add methods + cmd.AddCommand(newGenerateTemporaryTableCredentials()) + + // Apply optional overrides to this command. + for _, fn := range cmdOverrides { + fn(cmd) + } + + return cmd +} + +// start generate-temporary-table-credentials command + +// Slice with functions to override default command behavior. +// Functions can be added from the `init()` function in manually curated files in this directory. +var generateTemporaryTableCredentialsOverrides []func( + *cobra.Command, + *catalog.GenerateTemporaryTableCredentialRequest, +) + +func newGenerateTemporaryTableCredentials() *cobra.Command { + cmd := &cobra.Command{} + + var generateTemporaryTableCredentialsReq catalog.GenerateTemporaryTableCredentialRequest + var generateTemporaryTableCredentialsJson flags.JsonFlag + + // TODO: short flags + cmd.Flags().Var(&generateTemporaryTableCredentialsJson, "json", `either inline JSON string or @path/to/file.json with request body`) + + cmd.Flags().Var(&generateTemporaryTableCredentialsReq.Operation, "operation", `The operation performed against the table data, either READ or READ_WRITE. Supported values: [READ, READ_WRITE]`) + cmd.Flags().StringVar(&generateTemporaryTableCredentialsReq.TableId, "table-id", generateTemporaryTableCredentialsReq.TableId, `UUID of the table to read or write.`) + + cmd.Use = "generate-temporary-table-credentials" + cmd.Short = `Generate a temporary table credential.` + cmd.Long = `Generate a temporary table credential. + + Get a short-lived credential for directly accessing the table data on cloud + storage. The metastore must have external_access_enabled flag set to true + (default false). The caller must have EXTERNAL_USE_SCHEMA privilege on the + parent schema and this privilege can only be granted by catalog owners.` + + cmd.Annotations = make(map[string]string) + + cmd.Args = func(cmd *cobra.Command, args []string) error { + check := root.ExactArgs(0) + return check(cmd, args) + } + + cmd.PreRunE = root.MustWorkspaceClient + cmd.RunE = func(cmd *cobra.Command, args []string) (err error) { + ctx := cmd.Context() + w := root.WorkspaceClient(ctx) + + if cmd.Flags().Changed("json") { + err = generateTemporaryTableCredentialsJson.Unmarshal(&generateTemporaryTableCredentialsReq) + if err != nil { + return err + } + } + + response, err := w.TemporaryTableCredentials.GenerateTemporaryTableCredentials(ctx, generateTemporaryTableCredentialsReq) + if err != nil { + return err + } + return cmdio.Render(ctx, response) + } + + // Disable completions since they are not applicable. + // Can be overridden by manual implementation in `override.go`. + cmd.ValidArgsFunction = cobra.NoFileCompletions + + // Apply optional overrides to this command. + for _, fn := range generateTemporaryTableCredentialsOverrides { + fn(cmd, &generateTemporaryTableCredentialsReq) + } + + return cmd +} + +// end service TemporaryTableCredentials diff --git a/go.mod b/go.mod index ba41ef3a..9141274c 100644 --- a/go.mod +++ b/go.mod @@ -1,16 +1,18 @@ module github.com/databricks/cli -go 1.22 +go 1.22.0 + +toolchain go1.22.7 require ( github.com/Masterminds/semver/v3 v3.3.0 // MIT github.com/briandowns/spinner v1.23.1 // Apache 2.0 - github.com/databricks/databricks-sdk-go v0.46.0 // Apache 2.0 + github.com/databricks/databricks-sdk-go v0.47.0 // Apache 2.0 github.com/fatih/color v1.17.0 // MIT github.com/ghodss/yaml v1.0.0 // MIT + NOTICE github.com/google/uuid v1.6.0 // BSD-3-Clause github.com/hashicorp/go-version v1.7.0 // MPL 2.0 - github.com/hashicorp/hc-install v0.7.0 // MPL 2.0 + github.com/hashicorp/hc-install v0.9.0 // MPL 2.0 github.com/hashicorp/terraform-exec v0.21.0 // MPL 2.0 github.com/hashicorp/terraform-json v0.22.1 // MPL 2.0 github.com/manifoldco/promptui v0.9.0 // BSD-3-Clause @@ -22,7 +24,7 @@ require ( github.com/spf13/pflag v1.0.5 // BSD-3-Clause github.com/stretchr/testify v1.9.0 // MIT golang.org/x/exp v0.0.0-20240222234643-814bf88cf225 - golang.org/x/mod v0.20.0 + golang.org/x/mod v0.21.0 golang.org/x/oauth2 v0.23.0 golang.org/x/sync v0.8.0 golang.org/x/term v0.24.0 @@ -49,6 +51,7 @@ require ( github.com/google/s2a-go v0.1.7 // indirect github.com/googleapis/enterprise-certificate-proxy v0.3.2 // indirect github.com/hashicorp/go-cleanhttp v0.5.2 // indirect + github.com/hashicorp/go-retryablehttp v0.7.7 // indirect github.com/inconshreveable/mousetrap v1.1.0 // indirect github.com/mattn/go-colorable v0.1.13 // indirect github.com/pmezard/go-difflib v1.0.0 // indirect diff --git a/go.sum b/go.sum index 3d4a2cdc..177707a5 100644 --- a/go.sum +++ b/go.sum @@ -32,8 +32,8 @@ github.com/cncf/udpa/go v0.0.0-20191209042840-269d4d468f6f/go.mod h1:M8M6+tZqaGX github.com/cpuguy83/go-md2man/v2 v2.0.4/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o= github.com/cyphar/filepath-securejoin v0.2.4 h1:Ugdm7cg7i6ZK6x3xDF1oEu1nfkyfH53EtKeQYTC3kyg= github.com/cyphar/filepath-securejoin v0.2.4/go.mod h1:aPGpWjXOXUn2NCNjFvBE6aRxGGx79pTxQpKOJNYHHl4= -github.com/databricks/databricks-sdk-go v0.46.0 h1:D0TxmtSVAOsdnfzH4OGtAmcq+8TyA7Z6fA6JEYhupeY= -github.com/databricks/databricks-sdk-go v0.46.0/go.mod h1:ds+zbv5mlQG7nFEU5ojLtgN/u0/9YzZmKQES/CfedzU= +github.com/databricks/databricks-sdk-go v0.47.0 h1:eE7dN9axviL8+s10jnQAayOYDaR+Mfu7E9COGjO4lrQ= +github.com/databricks/databricks-sdk-go v0.47.0/go.mod h1:ds+zbv5mlQG7nFEU5ojLtgN/u0/9YzZmKQES/CfedzU= github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38= github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c= github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38= @@ -99,10 +99,14 @@ github.com/googleapis/gax-go/v2 v2.12.4 h1:9gWcmF85Wvq4ryPFvGFaOgPIs1AQX0d0bcbGw github.com/googleapis/gax-go/v2 v2.12.4/go.mod h1:KYEYLorsnIGDi/rPC8b5TdlB9kbKoFubselGIoBMCwI= github.com/hashicorp/go-cleanhttp v0.5.2 h1:035FKYIWjmULyFRBKPs8TBQoi0x6d9G4xc9neXJWAZQ= github.com/hashicorp/go-cleanhttp v0.5.2/go.mod h1:kO/YDlP8L1346E6Sodw+PrpBSV4/SoxCXGY6BqNFT48= +github.com/hashicorp/go-hclog v1.6.3 h1:Qr2kF+eVWjTiYmU7Y31tYlP1h0q/X3Nl3tPGdaB11/k= +github.com/hashicorp/go-hclog v1.6.3/go.mod h1:W4Qnvbt70Wk/zYJryRzDRU/4r0kIg0PVHBcfoyhpF5M= +github.com/hashicorp/go-retryablehttp v0.7.7 h1:C8hUCYzor8PIfXHa4UrZkU4VvK8o9ISHxT2Q8+VepXU= +github.com/hashicorp/go-retryablehttp v0.7.7/go.mod h1:pkQpWZeYWskR+D1tR2O5OcBFOxfA7DoAO6xtkuQnHTk= github.com/hashicorp/go-version v1.7.0 h1:5tqGy27NaOTB8yJKUZELlFAS/LTKJkrmONwQKeRZfjY= github.com/hashicorp/go-version v1.7.0/go.mod h1:fltr4n8CU8Ke44wwGCBoEymUuxUHl09ZGVZPK5anwXA= -github.com/hashicorp/hc-install v0.7.0 h1:Uu9edVqjKQxxuD28mR5TikkKDd/p55S8vzPC1659aBk= -github.com/hashicorp/hc-install v0.7.0/go.mod h1:ELmmzZlGnEcqoUMKUuykHaPCIR1sYLYX+KSggWSKZuA= +github.com/hashicorp/hc-install v0.9.0 h1:2dIk8LcvANwtv3QZLckxcjyF5w8KVtiMxu6G6eLhghE= +github.com/hashicorp/hc-install v0.9.0/go.mod h1:+6vOP+mf3tuGgMApVYtmsnDoKWMDcFXeTxCACYZ8SFg= github.com/hashicorp/terraform-exec v0.21.0 h1:uNkLAe95ey5Uux6KJdua6+cv8asgILFVWkd/RG0D2XQ= github.com/hashicorp/terraform-exec v0.21.0/go.mod h1:1PPeMYou+KDUSSeRE9szMZ/oHf4fYUmB923Wzbq1ICg= github.com/hashicorp/terraform-json v0.22.1 h1:xft84GZR0QzjPVWs4lRUwvTcPnegqlyS7orfb5Ltvec= @@ -180,8 +184,8 @@ golang.org/x/exp v0.0.0-20240222234643-814bf88cf225/go.mod h1:CxmFvTBINI24O/j8iY golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE= golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU= golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc= -golang.org/x/mod v0.20.0 h1:utOm6MM3R3dnawAiJgn0y+xvuYRsm1RKM/4giyfDgV0= -golang.org/x/mod v0.20.0/go.mod h1:hTbmBsO62+eylJbnUtE2MGJUyE7QWk4xUqPFrRgJ+7c= +golang.org/x/mod v0.21.0 h1:vvrHzRwRfVKSiLrG+d4FMl/Qi4ukBCE6kZlTUkDYRT0= +golang.org/x/mod v0.21.0/go.mod h1:6SkKJ3Xj0I0BrPOZoBy3bdMptDDU9oJrpohJ3eWZ1fY= golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4= golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4= golang.org/x/net v0.0.0-20190213061140-3a22650c66bd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4= diff --git a/internal/bundle/artifacts_test.go b/internal/bundle/artifacts_test.go index fa052e22..775327e1 100644 --- a/internal/bundle/artifacts_test.go +++ b/internal/bundle/artifacts_test.go @@ -36,8 +36,8 @@ func TestAccUploadArtifactFileToCorrectRemotePath(t *testing.T) { wsDir := internal.TemporaryWorkspaceDir(t, w) b := &bundle.Bundle{ - RootPath: dir, - SyncRootPath: dir, + BundleRootPath: dir, + SyncRootPath: dir, Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -101,8 +101,8 @@ func TestAccUploadArtifactFileToCorrectRemotePathWithEnvironments(t *testing.T) wsDir := internal.TemporaryWorkspaceDir(t, w) b := &bundle.Bundle{ - RootPath: dir, - SyncRootPath: dir, + BundleRootPath: dir, + SyncRootPath: dir, Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", @@ -171,8 +171,8 @@ func TestAccUploadArtifactFileToCorrectRemotePathForVolumes(t *testing.T) { touchEmptyFile(t, whlPath) b := &bundle.Bundle{ - RootPath: dir, - SyncRootPath: dir, + BundleRootPath: dir, + SyncRootPath: dir, Config: config.Root{ Bundle: config.Bundle{ Target: "whatever", diff --git a/internal/bundle/bundles/clusters/databricks_template_schema.json b/internal/bundle/bundles/clusters/databricks_template_schema.json new file mode 100644 index 00000000..c1c5cf12 --- /dev/null +++ b/internal/bundle/bundles/clusters/databricks_template_schema.json @@ -0,0 +1,16 @@ +{ + "properties": { + "unique_id": { + "type": "string", + "description": "Unique ID for job name" + }, + "spark_version": { + "type": "string", + "description": "Spark version used for job cluster" + }, + "node_type_id": { + "type": "string", + "description": "Node type id for job cluster" + } + } +} diff --git a/internal/bundle/bundles/clusters/template/databricks.yml.tmpl b/internal/bundle/bundles/clusters/template/databricks.yml.tmpl new file mode 100644 index 00000000..e0d6320a --- /dev/null +++ b/internal/bundle/bundles/clusters/template/databricks.yml.tmpl @@ -0,0 +1,24 @@ +bundle: + name: basic + +workspace: + root_path: "~/.bundle/{{.unique_id}}" + +resources: + clusters: + test_cluster: + cluster_name: "test-cluster-{{.unique_id}}" + spark_version: "{{.spark_version}}" + node_type_id: "{{.node_type_id}}" + num_workers: 2 + spark_conf: + "spark.executor.memory": "2g" + + jobs: + foo: + name: test-job-with-cluster-{{.unique_id}} + tasks: + - task_key: my_notebook_task + existing_cluster_id: "${resources.clusters.test_cluster.cluster_id}" + spark_python_task: + python_file: ./hello_world.py diff --git a/internal/bundle/bundles/clusters/template/hello_world.py b/internal/bundle/bundles/clusters/template/hello_world.py new file mode 100644 index 00000000..f301245e --- /dev/null +++ b/internal/bundle/bundles/clusters/template/hello_world.py @@ -0,0 +1 @@ +print("Hello World!") diff --git a/internal/bundle/clusters_test.go b/internal/bundle/clusters_test.go new file mode 100644 index 00000000..a961f3ea --- /dev/null +++ b/internal/bundle/clusters_test.go @@ -0,0 +1,56 @@ +package bundle + +import ( + "fmt" + "testing" + + "github.com/databricks/cli/internal" + "github.com/databricks/cli/internal/acc" + "github.com/databricks/cli/internal/testutil" + "github.com/databricks/cli/libs/env" + "github.com/databricks/databricks-sdk-go/service/compute" + "github.com/google/uuid" + "github.com/stretchr/testify/require" +) + +func TestAccDeployBundleWithCluster(t *testing.T) { + ctx, wt := acc.WorkspaceTest(t) + + if testutil.IsAWSCloud(wt.T) { + t.Skip("Skipping test for AWS cloud because it is not permitted to create clusters") + } + + nodeTypeId := internal.GetNodeTypeId(env.Get(ctx, "CLOUD_ENV")) + uniqueId := uuid.New().String() + root, err := initTestTemplate(t, ctx, "clusters", map[string]any{ + "unique_id": uniqueId, + "node_type_id": nodeTypeId, + "spark_version": defaultSparkVersion, + }) + require.NoError(t, err) + + t.Cleanup(func() { + err = destroyBundle(t, ctx, root) + require.NoError(t, err) + + cluster, err := wt.W.Clusters.GetByClusterName(ctx, fmt.Sprintf("test-cluster-%s", uniqueId)) + if err != nil { + require.ErrorContains(t, err, "does not exist") + } else { + require.Contains(t, []compute.State{compute.StateTerminated, compute.StateTerminating}, cluster.State) + } + + }) + + err = deployBundle(t, ctx, root) + require.NoError(t, err) + + // Cluster should exists after bundle deployment + cluster, err := wt.W.Clusters.GetByClusterName(ctx, fmt.Sprintf("test-cluster-%s", uniqueId)) + require.NoError(t, err) + require.NotNil(t, cluster) + + out, err := runResource(t, ctx, root, "foo") + require.NoError(t, err) + require.Contains(t, out, "Hello World!") +} diff --git a/internal/bundle/deploy_test.go b/internal/bundle/deploy_test.go index 736c880d..88543585 100644 --- a/internal/bundle/deploy_test.go +++ b/internal/bundle/deploy_test.go @@ -236,7 +236,7 @@ func TestAccDeployBasicBundleLogs(t *testing.T) { stdout, stderr := blackBoxRun(t, root, "bundle", "deploy") assert.Equal(t, strings.Join([]string{ - fmt.Sprintf("Uploading bundle files to /Users/%s/.bundle/%s/files...", currentUser.UserName, uniqueId), + fmt.Sprintf("Uploading bundle files to /Workspace/Users/%s/.bundle/%s/files...", currentUser.UserName, uniqueId), "Deploying resources...", "Updating deployment state...", "Deployment complete!\n", diff --git a/internal/bundle/helpers.go b/internal/bundle/helpers.go index 3547c175..b8c81a8d 100644 --- a/internal/bundle/helpers.go +++ b/internal/bundle/helpers.go @@ -114,7 +114,7 @@ func getBundleRemoteRootPath(w *databricks.WorkspaceClient, t *testing.T, unique // Compute root path for the bundle deployment me, err := w.CurrentUser.Me(context.Background()) require.NoError(t, err) - root := fmt.Sprintf("/Users/%s/.bundle/%s", me.UserName, uniqueId) + root := fmt.Sprintf("/Workspace/Users/%s/.bundle/%s", me.UserName, uniqueId) return root } diff --git a/internal/init_test.go b/internal/init_test.go index c3cb0127..d1a89f7b 100644 --- a/internal/init_test.go +++ b/internal/init_test.go @@ -126,7 +126,7 @@ func TestAccBundleInitHelpers(t *testing.T) { }{ { funcName: "{{short_name}}", - expected: auth.GetShortUserName(me.UserName), + expected: auth.GetShortUserName(me), }, { funcName: "{{user_name}}", diff --git a/internal/testutil/cloud.go b/internal/testutil/cloud.go index e547069f..ba5b75ec 100644 --- a/internal/testutil/cloud.go +++ b/internal/testutil/cloud.go @@ -49,3 +49,7 @@ func GetCloud(t *testing.T) Cloud { } return -1 } + +func IsAWSCloud(t *testing.T) bool { + return GetCloud(t) == AWS +} diff --git a/libs/auth/user.go b/libs/auth/user.go index 8eaa8763..c6aa974f 100644 --- a/libs/auth/user.go +++ b/libs/auth/user.go @@ -4,12 +4,17 @@ import ( "strings" "github.com/databricks/cli/libs/textutil" + "github.com/databricks/databricks-sdk-go/service/iam" ) // Get a short-form username, based on the user's primary email address. // We leave the full range of unicode letters in tact, but remove all "special" characters, // including dots, which are not supported in e.g. experiment names. -func GetShortUserName(emailAddress string) string { - local, _, _ := strings.Cut(emailAddress, "@") +func GetShortUserName(user *iam.User) string { + name := user.UserName + if IsServicePrincipal(user.UserName) && user.DisplayName != "" { + name = user.DisplayName + } + local, _, _ := strings.Cut(name, "@") return textutil.NormalizeString(local) } diff --git a/libs/auth/user_test.go b/libs/auth/user_test.go index 62b2d29a..24b61464 100644 --- a/libs/auth/user_test.go +++ b/libs/auth/user_test.go @@ -3,70 +3,111 @@ package auth import ( "testing" + "github.com/databricks/databricks-sdk-go/service/iam" "github.com/stretchr/testify/assert" ) func TestGetShortUserName(t *testing.T) { tests := []struct { name string - email string + user *iam.User expected string }{ { - email: "test.user.1234@example.com", + user: &iam.User{ + UserName: "test.user.1234@example.com", + }, expected: "test_user_1234", }, { - email: "tést.üser@example.com", + user: &iam.User{ + UserName: "tést.üser@example.com", + }, expected: "tést_üser", }, { - email: "test$.user@example.com", + user: &iam.User{ + UserName: "test$.user@example.com", + }, expected: "test_user", }, { - email: `jöhn.dœ@domain.com`, // Using non-ASCII characters. + user: &iam.User{ + UserName: `jöhn.dœ@domain.com`, // Using non-ASCII characters. + }, expected: "jöhn_dœ", }, { - email: `first+tag@email.com`, // The plus (+) sign is used for "sub-addressing" in some email services. + user: &iam.User{ + UserName: `first+tag@email.com`, // The plus (+) sign is used for "sub-addressing" in some email services. + }, expected: "first_tag", }, { - email: `email@sub.domain.com`, // Using a sub-domain. + user: &iam.User{ + UserName: `email@sub.domain.com`, // Using a sub-domain. + }, expected: "email", }, { - email: `"_quoted"@domain.com`, // Quoted strings can be part of the local-part. + user: &iam.User{ + UserName: `"_quoted"@domain.com`, // Quoted strings can be part of the local-part. + }, expected: "quoted", }, { - email: `name-o'mally@website.org`, // Single quote in the local-part. + user: &iam.User{ + UserName: `name-o'mally@website.org`, // Single quote in the local-part. + }, expected: "name_o_mally", }, { - email: `user%domain@external.com`, // Percent sign can be used for email routing in legacy systems. + user: &iam.User{ + UserName: `user%domain@external.com`, // Percent sign can be used for email routing in legacy systems. + }, expected: "user_domain", }, { - email: `long.name.with.dots@domain.net`, // Multiple dots in the local-part. + user: &iam.User{ + UserName: `long.name.with.dots@domain.net`, // Multiple dots in the local-part. + }, expected: "long_name_with_dots", }, { - email: `me&you@together.com`, // Using an ampersand (&) in the local-part. + user: &iam.User{ + UserName: `me&you@together.com`, // Using an ampersand (&) in the local-part. + }, expected: "me_you", }, { - email: `user!def!xyz@domain.org`, // The exclamation mark can be valid in some legacy systems. + user: &iam.User{ + UserName: `user!def!xyz@domain.org`, // The exclamation mark can be valid in some legacy systems. + }, expected: "user_def_xyz", }, { - email: `admin@ιντερνετ.com`, // Domain in non-ASCII characters (IDN or Internationalized Domain Name). + user: &iam.User{ + UserName: `admin@ιντερνετ.com`, // Domain in non-ASCII characters (IDN or Internationalized Domain Name). + }, expected: "admin", }, + { + user: &iam.User{ + UserName: `1706906c-c0a2-4c25-9f57-3a7aa3cb8123`, + DisplayName: "my-service-principal", + }, + expected: "my_service_principal", + }, + { + user: &iam.User{ + UserName: `1706906c-c0a2-4c25-9f57-3a7aa3cb8123`, + // This service princpal has DisplayName (it's an optional property) + }, + expected: "1706906c_c0a2_4c25_9f57_3a7aa3cb8123", + }, } for _, tt := range tests { - assert.Equal(t, tt.expected, GetShortUserName(tt.email)) + assert.Equal(t, tt.expected, GetShortUserName(tt.user)) } } diff --git a/libs/git/repository_test.go b/libs/git/repository_test.go index a28038ee..93d9a03d 100644 --- a/libs/git/repository_test.go +++ b/libs/git/repository_test.go @@ -209,7 +209,26 @@ func TestRepositoryGitConfigWhenNotARepo(t *testing.T) { } func TestRepositoryOriginUrlRemovesUserCreds(t *testing.T) { - repo := newTestRepository(t) - repo.addOriginUrl("https://username:token@github.com/databricks/foobar.git") - repo.assertOriginUrl("https://github.com/databricks/foobar.git") + tcases := []struct { + url string + expected string + }{ + { + url: "https://username:token@github.com/databricks/foobar.git", + expected: "https://github.com/databricks/foobar.git", + }, + { + // Note: The token is still considered and parsed as a username here. + // However credentials integrations by Git providers like GitHub + // allow for setting a PAT token as a username. + url: "https://token@github.com/databricks/foobar.git", + expected: "https://github.com/databricks/foobar.git", + }, + } + + for _, tc := range tcases { + repo := newTestRepository(t) + repo.addOriginUrl(tc.url) + repo.assertOriginUrl(tc.expected) + } } diff --git a/cmd/sync/output.go b/libs/sync/output.go similarity index 83% rename from cmd/sync/output.go rename to libs/sync/output.go index 2785343f..c01b25ef 100644 --- a/cmd/sync/output.go +++ b/libs/sync/output.go @@ -5,12 +5,10 @@ import ( "context" "encoding/json" "io" - - "github.com/databricks/cli/libs/sync" ) // Read synchronization events and write them as JSON to the specified writer (typically stdout). -func jsonOutput(ctx context.Context, ch <-chan sync.Event, w io.Writer) { +func JsonOutput(ctx context.Context, ch <-chan Event, w io.Writer) { enc := json.NewEncoder(w) for { select { @@ -31,7 +29,7 @@ func jsonOutput(ctx context.Context, ch <-chan sync.Event, w io.Writer) { } // Read synchronization events and write them as text to the specified writer (typically stdout). -func textOutput(ctx context.Context, ch <-chan sync.Event, w io.Writer) { +func TextOutput(ctx context.Context, ch <-chan Event, w io.Writer) { bw := bufio.NewWriter(w) for { diff --git a/libs/sync/sync.go b/libs/sync/sync.go index 9eaebf2a..cc9c7394 100644 --- a/libs/sync/sync.go +++ b/libs/sync/sync.go @@ -3,6 +3,7 @@ package sync import ( "context" "fmt" + stdsync "sync" "time" "github.com/databricks/cli/libs/filer" @@ -15,6 +16,8 @@ import ( "github.com/databricks/databricks-sdk-go/service/iam" ) +type OutputHandler func(context.Context, <-chan Event) + type SyncOptions struct { LocalRoot vfs.Path Paths []string @@ -34,6 +37,8 @@ type SyncOptions struct { CurrentUser *iam.User Host string + + OutputHandler OutputHandler } type Sync struct { @@ -49,6 +54,10 @@ type Sync struct { // Synchronization progress events are sent to this event notifier. notifier EventNotifier seq int + + // WaitGroup is automatically created when an output handler is provided in the SyncOptions. + // Close call is required to ensure the output handler goroutine handles all events in time. + outputWaitGroup *stdsync.WaitGroup } // New initializes and returns a new [Sync] instance. @@ -106,31 +115,41 @@ func New(ctx context.Context, opts SyncOptions) (*Sync, error) { return nil, err } + var notifier EventNotifier + var outputWaitGroup = &stdsync.WaitGroup{} + if opts.OutputHandler != nil { + ch := make(chan Event, MaxRequestsInFlight) + notifier = &ChannelNotifier{ch} + outputWaitGroup.Add(1) + go func() { + defer outputWaitGroup.Done() + opts.OutputHandler(ctx, ch) + }() + } else { + notifier = &NopNotifier{} + } + return &Sync{ SyncOptions: &opts, - fileSet: fileSet, - includeFileSet: includeFileSet, - excludeFileSet: excludeFileSet, - snapshot: snapshot, - filer: filer, - notifier: &NopNotifier{}, - seq: 0, + fileSet: fileSet, + includeFileSet: includeFileSet, + excludeFileSet: excludeFileSet, + snapshot: snapshot, + filer: filer, + notifier: notifier, + outputWaitGroup: outputWaitGroup, + seq: 0, }, nil } -func (s *Sync) Events() <-chan Event { - ch := make(chan Event, MaxRequestsInFlight) - s.notifier = &ChannelNotifier{ch} - return ch -} - func (s *Sync) Close() { if s.notifier == nil { return } s.notifier.Close() s.notifier = nil + s.outputWaitGroup.Wait() } func (s *Sync) notifyStart(ctx context.Context, d diff) { diff --git a/libs/template/helpers.go b/libs/template/helpers.go index 1dfe74d7..88c73cc4 100644 --- a/libs/template/helpers.go +++ b/libs/template/helpers.go @@ -119,7 +119,7 @@ func loadHelpers(ctx context.Context) template.FuncMap { return "", err } } - return auth.GetShortUserName(cachedUser.UserName), nil + return auth.GetShortUserName(cachedUser), nil }, // Get the default workspace catalog. If there is no default, or if // Unity Catalog is not enabled, return an empty string. diff --git a/libs/template/templates/dbt-sql/template/{{.project_name}}/README.md.tmpl b/libs/template/templates/dbt-sql/template/{{.project_name}}/README.md.tmpl index dbf8a8d8..cd4c29a7 100644 --- a/libs/template/templates/dbt-sql/template/{{.project_name}}/README.md.tmpl +++ b/libs/template/templates/dbt-sql/template/{{.project_name}}/README.md.tmpl @@ -121,7 +121,7 @@ You can find that job by opening your workpace and clicking on **Workflows**. You can also deploy to your production target directly from the command-line. The warehouse, catalog, and schema for that target are configured in databricks.yml. -When deploying to this target, note that the default job at resources/{{.project_name}}_job.yml +When deploying to this target, note that the default job at resources/{{.project_name}}.job.yml has a schedule set that runs every day. The schedule is paused when deploying in development mode (see https://docs.databricks.com/dev-tools/bundles/deployment-modes.html). diff --git a/libs/template/templates/dbt-sql/template/{{.project_name}}/databricks.yml.tmpl b/libs/template/templates/dbt-sql/template/{{.project_name}}/databricks.yml.tmpl index f96ce4fe..5594749a 100644 --- a/libs/template/templates/dbt-sql/template/{{.project_name}}/databricks.yml.tmpl +++ b/libs/template/templates/dbt-sql/template/{{.project_name}}/databricks.yml.tmpl @@ -24,8 +24,8 @@ targets: mode: production workspace: host: {{workspace_host}} - # We explicitly specify /Users/{{user_name}} to make sure we only have a single copy. - root_path: /Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} + # We explicitly specify /Workspace/Users/{{user_name}} to make sure we only have a single copy. + root_path: /Workspace/Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} permissions: - {{if is_service_principal}}service_principal{{else}}user{{end}}_name: {{user_name}} level: CAN_MANAGE diff --git a/libs/template/templates/dbt-sql/template/{{.project_name}}/resources/{{.project_name}}_job.yml.tmpl b/libs/template/templates/dbt-sql/template/{{.project_name}}/resources/{{.project_name}}.job.yml.tmpl similarity index 100% rename from libs/template/templates/dbt-sql/template/{{.project_name}}/resources/{{.project_name}}_job.yml.tmpl rename to libs/template/templates/dbt-sql/template/{{.project_name}}/resources/{{.project_name}}.job.yml.tmpl diff --git a/libs/template/templates/default-python/template/__preamble.tmpl b/libs/template/templates/default-python/template/__preamble.tmpl index a919a269..69b769cd 100644 --- a/libs/template/templates/default-python/template/__preamble.tmpl +++ b/libs/template/templates/default-python/template/__preamble.tmpl @@ -18,7 +18,7 @@ This file only template directives; it is skipped for the actual output. {{if $notDLT}} {{skip "{{.project_name}}/src/dlt_pipeline.ipynb"}} - {{skip "{{.project_name}}/resources/{{.project_name}}_pipeline.yml"}} + {{skip "{{.project_name}}/resources/{{.project_name}}.pipeline.yml"}} {{end}} {{if $notNotebook}} @@ -26,7 +26,7 @@ This file only template directives; it is skipped for the actual output. {{end}} {{if (and $notDLT $notNotebook $notPython)}} - {{skip "{{.project_name}}/resources/{{.project_name}}_job.yml"}} + {{skip "{{.project_name}}/resources/{{.project_name}}.job.yml"}} {{else}} {{skip "{{.project_name}}/resources/.gitkeep"}} {{end}} diff --git a/libs/template/templates/default-python/template/{{.project_name}}/README.md.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/README.md.tmpl index 5adade0b..53847a9c 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/README.md.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/README.md.tmpl @@ -29,7 +29,7 @@ The '{{.project_name}}' project was generated by using the default-python templa ``` Note that the default job from the template has a schedule that runs every day - (defined in resources/{{.project_name}}_job.yml). The schedule + (defined in resources/{{.project_name}}.job.yml). The schedule is paused when deploying in development mode (see https://docs.databricks.com/dev-tools/bundles/deployment-modes.html). diff --git a/libs/template/templates/default-python/template/{{.project_name}}/databricks.yml.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/databricks.yml.tmpl index 8544dc83..c42b822a 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/databricks.yml.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/databricks.yml.tmpl @@ -21,8 +21,8 @@ targets: mode: production workspace: host: {{workspace_host}} - # We explicitly specify /Users/{{user_name}} to make sure we only have a single copy. - root_path: /Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} + # We explicitly specify /Workspace/Users/{{user_name}} to make sure we only have a single copy. + root_path: /Workspace/Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} permissions: - {{if is_service_principal}}service_principal{{else}}user{{end}}_name: {{user_name}} level: CAN_MANAGE diff --git a/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_job.yml.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.job.yml.tmpl similarity index 97% rename from libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_job.yml.tmpl rename to libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.job.yml.tmpl index d2100e90..5211e389 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_job.yml.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.job.yml.tmpl @@ -40,7 +40,7 @@ resources: - task_key: notebook_task {{- end}} pipeline_task: - {{- /* TODO: we should find a way that doesn't use magics for the below, like ./{{project_name}}_pipeline.yml */}} + {{- /* TODO: we should find a way that doesn't use magics for the below, like ./{{project_name}}.pipeline.yml */}} pipeline_id: ${resources.pipelines.{{.project_name}}_pipeline.id} {{end -}} {{- if (eq .include_python "yes") }} diff --git a/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_pipeline.yml.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.pipeline.yml.tmpl similarity index 50% rename from libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_pipeline.yml.tmpl rename to libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.pipeline.yml.tmpl index 4b8f74d1..50e5ad97 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}_pipeline.yml.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/resources/{{.project_name}}.pipeline.yml.tmpl @@ -3,10 +3,16 @@ resources: pipelines: {{.project_name}}_pipeline: name: {{.project_name}}_pipeline + {{- if eq default_catalog ""}} + ## Specify the 'catalog' field to configure this pipeline to make use of Unity Catalog: + # catalog: catalog_name + {{- else}} + catalog: {{default_catalog}} + {{- end}} target: {{.project_name}}_${bundle.environment} libraries: - notebook: path: ../src/dlt_pipeline.ipynb configuration: - bundle.sourcePath: /Workspace/${workspace.file_path}/src + bundle.sourcePath: ${workspace.file_path}/src diff --git a/libs/template/templates/default-python/template/{{.project_name}}/src/dlt_pipeline.ipynb.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/src/dlt_pipeline.ipynb.tmpl index b152e9a3..253ed321 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/src/dlt_pipeline.ipynb.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/src/dlt_pipeline.ipynb.tmpl @@ -14,7 +14,7 @@ "source": [ "# DLT pipeline\n", "\n", - "This Delta Live Tables (DLT) definition is executed using a pipeline defined in resources/{{.project_name}}_pipeline.yml." + "This Delta Live Tables (DLT) definition is executed using a pipeline defined in resources/{{.project_name}}.pipeline.yml." ] }, { diff --git a/libs/template/templates/default-python/template/{{.project_name}}/src/notebook.ipynb.tmpl b/libs/template/templates/default-python/template/{{.project_name}}/src/notebook.ipynb.tmpl index a228f8d1..6782a053 100644 --- a/libs/template/templates/default-python/template/{{.project_name}}/src/notebook.ipynb.tmpl +++ b/libs/template/templates/default-python/template/{{.project_name}}/src/notebook.ipynb.tmpl @@ -14,7 +14,7 @@ "source": [ "# Default notebook\n", "\n", - "This default notebook is executed using Databricks Workflows as defined in resources/{{.project_name}}_job.yml." + "This default notebook is executed using Databricks Workflows as defined in resources/{{.project_name}}.job.yml." ] }, { diff --git a/libs/template/templates/default-sql/template/{{.project_name}}/databricks.yml.tmpl b/libs/template/templates/default-sql/template/{{.project_name}}/databricks.yml.tmpl index 55c1aae4..51d03e99 100644 --- a/libs/template/templates/default-sql/template/{{.project_name}}/databricks.yml.tmpl +++ b/libs/template/templates/default-sql/template/{{.project_name}}/databricks.yml.tmpl @@ -41,8 +41,8 @@ targets: mode: production workspace: host: {{workspace_host}} - # We explicitly specify /Users/{{user_name}} to make sure we only have a single copy. - root_path: /Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} + # We explicitly specify /Workspace/Users/{{user_name}} to make sure we only have a single copy. + root_path: /Workspace/Users/{{user_name}}/.bundle/${bundle.name}/${bundle.target} variables: warehouse_id: {{index ((regexp "[^/]+$").FindStringSubmatch .http_path) 0}} catalog: {{.default_catalog}} diff --git a/libs/template/templates/default-sql/template/{{.project_name}}/resources/{{.project_name}}_sql_job.yml.tmpl b/libs/template/templates/default-sql/template/{{.project_name}}/resources/{{.project_name}}_sql.job.yml.tmpl similarity index 100% rename from libs/template/templates/default-sql/template/{{.project_name}}/resources/{{.project_name}}_sql_job.yml.tmpl rename to libs/template/templates/default-sql/template/{{.project_name}}/resources/{{.project_name}}_sql.job.yml.tmpl diff --git a/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_daily.sql.tmpl b/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_daily.sql.tmpl index e5ceb77a..444ae4e0 100644 --- a/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_daily.sql.tmpl +++ b/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_daily.sql.tmpl @@ -1,4 +1,4 @@ --- This query is executed using Databricks Workflows (see resources/{{.project_name}}_sql_job.yml) +-- This query is executed using Databricks Workflows (see resources/{{.project_name}}_sql.job.yml) USE CATALOG {{"{{"}}catalog{{"}}"}}; USE IDENTIFIER({{"{{"}}schema{{"}}"}}); diff --git a/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_raw.sql.tmpl b/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_raw.sql.tmpl index c73606ef..80f6773c 100644 --- a/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_raw.sql.tmpl +++ b/libs/template/templates/default-sql/template/{{.project_name}}/src/orders_raw.sql.tmpl @@ -1,4 +1,4 @@ --- This query is executed using Databricks Workflows (see resources/{{.project_name}}_sql_job.yml) +-- This query is executed using Databricks Workflows (see resources/{{.project_name}}_sql.job.yml) -- -- The streaming table below ingests all JSON files in /databricks-datasets/retail-org/sales_orders/ -- See also https://docs.databricks.com/sql/language-manual/sql-ref-syntax-ddl-create-streaming-table.html