databricks-cli/bundle/config/root.go

283 lines
7.0 KiB
Go
Raw Normal View History

package config
import (
"fmt"
"os"
"path/filepath"
"strings"
"github.com/databricks/cli/bundle/config/variable"
"github.com/databricks/databricks-sdk-go/service/jobs"
"github.com/ghodss/yaml"
"github.com/imdario/mergo"
)
type ConfigFileNames []string
// FileNames contains allowed names of bundle configuration files.
var FileNames = ConfigFileNames{"databricks.yml", "databricks.yaml", "bundle.yml", "bundle.yaml"}
func (c ConfigFileNames) FindInPath(path string) (string, error) {
result := ""
var firstErr error
for _, file := range c {
filePath := filepath.Join(path, file)
_, err := os.Stat(filePath)
if err == nil {
if result != "" {
return "", fmt.Errorf("multiple bundle root configuration files found in %s", path)
}
result = filePath
} else {
if firstErr == nil {
firstErr = err
}
}
}
if result == "" {
return "", firstErr
}
return result, nil
}
type Root struct {
// Path contains the directory path to the root of the bundle.
// It is set when loading `databricks.yml`.
Path string `json:"-" bundle:"readonly"`
// Contains user defined variables
Variables map[string]*variable.Variable `json:"variables,omitempty"`
// Bundle contains details about this bundle, such as its name,
// version of the spec (TODO), default cluster, default warehouse, etc.
Bundle Bundle `json:"bundle"`
// Include specifies a list of patterns of file names to load and
// merge into the this configuration. Only includes defined in the root
// `databricks.yml` are processed. Defaults to an empty list.
Include []string `json:"include,omitempty"`
// Workspace contains details about the workspace to connect to
// and paths in the workspace tree to use for this bundle.
Workspace Workspace `json:"workspace,omitempty"`
// Artifacts contains a description of all code artifacts in this bundle.
Artifacts Artifacts `json:"artifacts,omitempty"`
// Resources contains a description of all Databricks resources
// to deploy in this bundle (e.g. jobs, pipelines, etc.).
Resources Resources `json:"resources,omitempty"`
// Targets can be used to differentiate settings and resources between
// bundle deployment targets (e.g. development, staging, production).
// If not specified, the code below initializes this field with a
// single default-initialized target called "default".
Targets map[string]*Target `json:"targets,omitempty"`
// DEPRECATED. Left for backward compatibility with Targets
Environments map[string]*Target `json:"environments,omitempty"`
Added support for sync.include and sync.exclude sections (#671) ## Changes Added support for `sync.include` and `sync.exclude` sections ## Tests Added `sample-java` folder to gitignore ``` bundle: name: wheel-task sync: include: - "./sample-java/*.kts" ``` Kotlin files were correctly synced. ``` [DEBUG] Test execution command: /opt/homebrew/opt/go@1.21/bin/go test ./... -json -timeout 1h -coverpkg=./... -coverprofile=coverage.txt -run ^TestAcc [DEBUG] Test execution directory: /Users/andrew.nester/cli 2023/08/17 17:12:10 [INFO] ✅ TestAccAlertsCreateErrWhenNoArguments (2.320s) 2023/08/17 17:12:10 [INFO] ✅ TestAccApiGet (0.650s) 2023/08/17 17:12:12 [INFO] ✅ TestAccClustersList (1.060s) 2023/08/17 17:12:12 [INFO] ✅ TestAccClustersGet (0.760s) 2023/08/17 17:12:26 [INFO] ✅ TestAccFilerWorkspaceFilesReadWrite (13.270s) 2023/08/17 17:12:32 [INFO] ✅ TestAccFilerWorkspaceFilesReadDir (6.860s) 2023/08/17 17:12:46 [INFO] ✅ TestAccFilerDbfsReadWrite (13.380s) 2023/08/17 17:12:53 [INFO] ✅ TestAccFilerDbfsReadDir (7.460s) 2023/08/17 17:13:01 [INFO] ✅ TestAccFilerWorkspaceNotebookConflict (7.920s) 2023/08/17 17:13:10 [INFO] ✅ TestAccFilerWorkspaceNotebookWithOverwriteFlag (9.290s) 2023/08/17 17:13:10 [INFO] ✅ TestAccFilerLocalReadWrite (0.010s) 2023/08/17 17:13:11 [INFO] ✅ TestAccFilerLocalReadDir (0.010s) 2023/08/17 17:13:14 [INFO] ✅ TestAccFsCatForDbfs (3.180s) 2023/08/17 17:13:15 [INFO] ✅ TestAccFsCatForDbfsOnNonExistentFile (0.940s) 2023/08/17 17:13:15 [INFO] ✅ TestAccFsCatForDbfsInvalidScheme (0.560s) 2023/08/17 17:13:18 [INFO] ✅ TestAccFsCatDoesNotSupportOutputModeJson (2.910s) 2023/08/17 17:13:51 [INFO] ✅ TestAccFsCpDir (32.730s) 2023/08/17 17:14:06 [INFO] ✅ TestAccFsCpFileToFile (14.740s) 2023/08/17 17:14:20 [INFO] ✅ TestAccFsCpFileToDir (14.340s) 2023/08/17 17:14:53 [INFO] ✅ TestAccFsCpDirToDirFileNotOverwritten (32.710s) 2023/08/17 17:15:12 [INFO] ✅ TestAccFsCpFileToDirFileNotOverwritten (19.590s) 2023/08/17 17:15:32 [INFO] ✅ TestAccFsCpFileToFileFileNotOverwritten (19.950s) 2023/08/17 17:16:11 [INFO] ✅ TestAccFsCpDirToDirWithOverwriteFlag (38.970s) 2023/08/17 17:16:32 [INFO] ✅ TestAccFsCpFileToFileWithOverwriteFlag (21.040s) 2023/08/17 17:16:52 [INFO] ✅ TestAccFsCpFileToDirWithOverwriteFlag (19.670s) 2023/08/17 17:16:54 [INFO] ✅ TestAccFsCpErrorsWhenSourceIsDirWithoutRecursiveFlag (1.890s) 2023/08/17 17:16:54 [INFO] ✅ TestAccFsCpErrorsOnInvalidScheme (0.690s) 2023/08/17 17:17:10 [INFO] ✅ TestAccFsCpSourceIsDirectoryButTargetIsFile (15.810s) 2023/08/17 17:17:14 [INFO] ✅ TestAccFsLsForDbfs (4.000s) 2023/08/17 17:17:18 [INFO] ✅ TestAccFsLsForDbfsWithAbsolutePaths (4.000s) 2023/08/17 17:17:21 [INFO] ✅ TestAccFsLsForDbfsOnFile (3.140s) 2023/08/17 17:17:23 [INFO] ✅ TestAccFsLsForDbfsOnEmptyDir (2.030s) 2023/08/17 17:17:24 [INFO] ✅ TestAccFsLsForDbfsForNonexistingDir (0.840s) 2023/08/17 17:17:25 [INFO] ✅ TestAccFsLsWithoutScheme (0.590s) 2023/08/17 17:17:27 [INFO] ✅ TestAccFsMkdirCreatesDirectory (2.310s) 2023/08/17 17:17:30 [INFO] ✅ TestAccFsMkdirCreatesMultipleDirectories (2.800s) 2023/08/17 17:17:33 [INFO] ✅ TestAccFsMkdirWhenDirectoryAlreadyExists (2.700s) 2023/08/17 17:17:35 [INFO] ✅ TestAccFsMkdirWhenFileExistsAtPath (2.870s) 2023/08/17 17:17:40 [INFO] ✅ TestAccFsRmForFile (4.030s) 2023/08/17 17:17:43 [INFO] ✅ TestAccFsRmForEmptyDirectory (3.470s) 2023/08/17 17:17:46 [INFO] ✅ TestAccFsRmForNonEmptyDirectory (3.350s) 2023/08/17 17:17:47 [INFO] ✅ TestAccFsRmForNonExistentFile (0.940s) 2023/08/17 17:17:51 [INFO] ✅ TestAccFsRmForNonEmptyDirectoryWithRecursiveFlag (3.570s) 2023/08/17 17:17:52 [INFO] ✅ TestAccGitClone (0.890s) 2023/08/17 17:17:52 [INFO] ✅ TestAccGitCloneWithOnlyRepoNameOnAlternateBranch (0.730s) 2023/08/17 17:17:53 [INFO] ✅ TestAccGitCloneErrorsWhenRepositoryDoesNotExist (0.540s) 2023/08/17 17:18:02 [INFO] ✅ TestAccLock (8.800s) 2023/08/17 17:18:06 [INFO] ✅ TestAccLockUnlockWithoutAllowsLockFileNotExist (3.930s) 2023/08/17 17:18:09 [INFO] ✅ TestAccLockUnlockWithAllowsLockFileNotExist (3.320s) 2023/08/17 17:18:20 [INFO] ✅ TestAccSyncFullFileSync (10.570s) 2023/08/17 17:18:31 [INFO] ✅ TestAccSyncIncrementalFileSync (11.460s) 2023/08/17 17:18:42 [INFO] ✅ TestAccSyncNestedFolderSync (10.850s) 2023/08/17 17:18:53 [INFO] ✅ TestAccSyncNestedFolderDoesntFailOnNonEmptyDirectory (10.650s) 2023/08/17 17:19:04 [INFO] ✅ TestAccSyncNestedSpacePlusAndHashAreEscapedSync (10.930s) 2023/08/17 17:19:11 [INFO] ✅ TestAccSyncIncrementalFileOverwritesFolder (7.010s) 2023/08/17 17:19:18 [INFO] ✅ TestAccSyncIncrementalSyncPythonNotebookToFile (7.380s) 2023/08/17 17:19:24 [INFO] ✅ TestAccSyncIncrementalSyncFileToPythonNotebook (6.220s) 2023/08/17 17:19:30 [INFO] ✅ TestAccSyncIncrementalSyncPythonNotebookDelete (5.530s) 2023/08/17 17:19:32 [INFO] ✅ TestAccSyncEnsureRemotePathIsUsableIfRepoDoesntExist (2.620s) 2023/08/17 17:19:38 [INFO] ✅ TestAccSyncEnsureRemotePathIsUsableIfRepoExists (5.460s) 2023/08/17 17:19:40 [INFO] ✅ TestAccSyncEnsureRemotePathIsUsableInWorkspace (1.850s) 2023/08/17 17:19:40 [INFO] ✅ TestAccWorkspaceList (0.780s) 2023/08/17 17:19:51 [INFO] ✅ TestAccExportDir (10.350s) 2023/08/17 17:19:54 [INFO] ✅ TestAccExportDirDoesNotOverwrite (3.330s) 2023/08/17 17:19:58 [INFO] ✅ TestAccExportDirWithOverwriteFlag (3.770s) 2023/08/17 17:20:07 [INFO] ✅ TestAccImportDir (9.320s) 2023/08/17 17:20:24 [INFO] ✅ TestAccImportDirDoesNotOverwrite (16.950s) 2023/08/17 17:20:35 [INFO] ✅ TestAccImportDirWithOverwriteFlag (10.620s) 2023/08/17 17:20:35 [INFO] ✅ 68/68 passed, 0 failed, 3 skipped ```
2023-08-18 08:07:25 +00:00
// Sync section specifies options for files synchronization
Sync Sync `json:"sync"`
// RunAs section allows to define an execution identity for jobs and pipelines runs
RunAs *jobs.JobRunAs `json:"run_as,omitempty"`
}
func Load(path string) (*Root, error) {
var r Root
stat, err := os.Stat(path)
if err != nil {
return nil, err
}
// If we were given a directory, assume this is the bundle root.
if stat.IsDir() {
path, err = FileNames.FindInPath(path)
if err != nil {
return nil, err
}
}
if err := r.Load(path); err != nil {
return nil, err
}
return &r, nil
}
// SetConfigFilePath configures the path that its configuration
// was loaded from in configuration leafs that require it.
func (r *Root) SetConfigFilePath(path string) {
r.Resources.SetConfigFilePath(path)
if r.Artifacts != nil {
r.Artifacts.SetConfigFilePath(path)
}
if r.Targets != nil {
for _, env := range r.Targets {
if env == nil {
continue
}
if env.Resources != nil {
env.Resources.SetConfigFilePath(path)
}
if env.Artifacts != nil {
env.Artifacts.SetConfigFilePath(path)
}
}
}
}
// Initializes variables using values passed from the command line flag
// Input has to be a string of the form `foo=bar`. In this case the variable with
// name `foo` is assigned the value `bar`
func (r *Root) InitializeVariables(vars []string) error {
for _, variable := range vars {
parsedVariable := strings.SplitN(variable, "=", 2)
if len(parsedVariable) != 2 {
return fmt.Errorf("unexpected flag value for variable assignment: %s", variable)
}
name := parsedVariable[0]
val := parsedVariable[1]
if _, ok := r.Variables[name]; !ok {
return fmt.Errorf("variable %s has not been defined", name)
}
err := r.Variables[name].Set(val)
if err != nil {
return fmt.Errorf("failed to assign %s to %s: %s", val, name, err)
}
}
return nil
}
func (r *Root) Load(path string) error {
raw, err := os.ReadFile(path)
if err != nil {
return err
}
err = yaml.Unmarshal(raw, r)
if err != nil {
return fmt.Errorf("failed to load %s: %w", path, err)
}
if r.Environments != nil && r.Targets != nil {
return fmt.Errorf("both 'environments' and 'targets' are specified, only 'targets' should be used: %s", path)
}
if r.Environments != nil {
//TODO: add a command line notice that this is a deprecated option.
r.Targets = r.Environments
}
r.Path = filepath.Dir(path)
r.SetConfigFilePath(path)
_, err = r.Resources.VerifyUniqueResourceIdentifiers()
return err
}
func (r *Root) Merge(other *Root) error {
err := r.Sync.Merge(r, other)
if err != nil {
return err
}
other.Sync = Sync{}
// TODO: when hooking into merge semantics, disallow setting path on the target instance.
other.Path = ""
// Check for safe merge, protecting against duplicate resource identifiers
err = r.Resources.VerifySafeMerge(&other.Resources)
if err != nil {
return err
}
// TODO: define and test semantics for merging.
return mergo.Merge(r, other, mergo.WithOverride)
}
func (r *Root) MergeTargetOverrides(target *Target) error {
var err error
// Target may be nil if it's empty.
if target == nil {
2022-12-22 14:31:32 +00:00
return nil
}
if target.Bundle != nil {
err = mergo.Merge(&r.Bundle, target.Bundle, mergo.WithOverride)
if err != nil {
return err
}
}
if target.Workspace != nil {
err = mergo.Merge(&r.Workspace, target.Workspace, mergo.WithOverride)
if err != nil {
return err
}
}
if target.Artifacts != nil {
err = mergo.Merge(&r.Artifacts, target.Artifacts, mergo.WithOverride, mergo.WithAppendSlice)
if err != nil {
return err
}
}
if target.Resources != nil {
err = mergo.Merge(&r.Resources, target.Resources, mergo.WithOverride, mergo.WithAppendSlice)
if err != nil {
return err
}
err = r.Resources.MergeJobClusters()
if err != nil {
return err
}
}
if target.Variables != nil {
for k, v := range target.Variables {
variable, ok := r.Variables[k]
if !ok {
return fmt.Errorf("variable %s is not defined but is assigned a value", k)
}
// we only allow overrides of the default value for a variable
defaultVal := v
variable.Default = &defaultVal
}
}
if target.RunAs != nil {
r.RunAs = target.RunAs
}
if target.Mode != "" {
r.Bundle.Mode = target.Mode
Add development runs (#522) This implements the "development run" functionality that we desire for DABs in the workspace / IDE. ## bundle.yml changes In bundle.yml, there should be a "dev" environment that is marked as `mode: debug`: ``` environments: dev: default: true mode: development # future accepted values might include pull_request, production ``` Setting `mode` to `development` indicates that this environment is used just for running things for development. This results in several changes to deployed assets: * All assets will get '[dev]' in their name and will get a 'dev' tag * All assets will be hidden from the list of assets (future work; e.g. for jobs we would have a special job_type that hides it from the list) * All deployed assets will be ephemeral (future work, we need some form of garbage collection) * Pipelines will be marked as 'development: true' * Jobs can run on development compute through the `--compute` parameter in the CLI * Jobs get their schedule / triggers paused * Jobs get concurrent runs (it's really annoying if your runs get skipped because the last run was still in progress) Other accepted values for `mode` are `default` (which does nothing) and `pull-request` (which is reserved for future use). ## CLI changes To run a single job called "shark_sighting" on existing compute, use the following commands: ``` $ databricks bundle deploy --compute 0617-201942-9yd9g8ix $ databricks bundle run shark_sighting ``` which would deploy and run a job called "[dev] shark_sightings" on the compute provided. Note that `--compute` is not accepted in production environments, so we show an error if `mode: development` is not used. The `run --deploy` command offers a convenient shorthand for the common combination of deploying & running: ``` $ export DATABRICKS_COMPUTE=0617-201942-9yd9g8ix $ bundle run --deploy shark_sightings ``` The `--deploy` addition isn't really essential and I welcome feedback 🤔 I played with the idea of a "debug" or "dev" command but that seemed to only make the option space even broader for users. The above could work well with an IDE or workspace that automatically sets the target compute. One more thing I added is`run --no-wait` can now be used to run something without waiting for it to be completed (useful for IDE-like environments that can display progress themselves). ``` $ bundle run --deploy shark_sightings --no-wait ```
2023-07-12 06:51:54 +00:00
}
if target.ComputeID != "" {
r.Bundle.ComputeID = target.ComputeID
Add development runs (#522) This implements the "development run" functionality that we desire for DABs in the workspace / IDE. ## bundle.yml changes In bundle.yml, there should be a "dev" environment that is marked as `mode: debug`: ``` environments: dev: default: true mode: development # future accepted values might include pull_request, production ``` Setting `mode` to `development` indicates that this environment is used just for running things for development. This results in several changes to deployed assets: * All assets will get '[dev]' in their name and will get a 'dev' tag * All assets will be hidden from the list of assets (future work; e.g. for jobs we would have a special job_type that hides it from the list) * All deployed assets will be ephemeral (future work, we need some form of garbage collection) * Pipelines will be marked as 'development: true' * Jobs can run on development compute through the `--compute` parameter in the CLI * Jobs get their schedule / triggers paused * Jobs get concurrent runs (it's really annoying if your runs get skipped because the last run was still in progress) Other accepted values for `mode` are `default` (which does nothing) and `pull-request` (which is reserved for future use). ## CLI changes To run a single job called "shark_sighting" on existing compute, use the following commands: ``` $ databricks bundle deploy --compute 0617-201942-9yd9g8ix $ databricks bundle run shark_sighting ``` which would deploy and run a job called "[dev] shark_sightings" on the compute provided. Note that `--compute` is not accepted in production environments, so we show an error if `mode: development` is not used. The `run --deploy` command offers a convenient shorthand for the common combination of deploying & running: ``` $ export DATABRICKS_COMPUTE=0617-201942-9yd9g8ix $ bundle run --deploy shark_sightings ``` The `--deploy` addition isn't really essential and I welcome feedback 🤔 I played with the idea of a "debug" or "dev" command but that seemed to only make the option space even broader for users. The above could work well with an IDE or workspace that automatically sets the target compute. One more thing I added is`run --no-wait` can now be used to run something without waiting for it to be completed (useful for IDE-like environments that can display progress themselves). ``` $ bundle run --deploy shark_sightings --no-wait ```
2023-07-12 06:51:54 +00:00
}
Add validation for Git settings in bundles (#578) ## Changes This checks whether the Git settings are consistent with the actual Git state of a source directory. (This PR adds to https://github.com/databricks/cli/pull/577.) Previously, we would silently let users configure their Git branch to e.g. `main` and deploy with that metadata even if they were actually on a different branch. With these changes, the following config would result in an error when deployed from any other branch than `main`: ``` bundle: name: example workspace: git: branch: main environments: ... ``` > not on the right Git branch: > expected according to configuration: main > actual: my-feature-branch It's not very useful to set the same branch for all environments, though. For development, it's better to just let the CLI auto-detect the right branch. Therefore, it's now possible to set the branch just for a single environment: ``` bundle: name: example 2 environments: development: default: true production: # production can only be deployed from the 'main' branch git: branch: main ``` Adding to that, the `mode: production` option actually checks that users explicitly set the Git branch as seen above. Setting that branch helps avoid mistakes, where someone accidentally deploys to production from the wrong branch. (I could see us offering an escape hatch for that in the future.) # Testing Manual testing to validate the experience and error messages. Automated unit tests. --------- Co-authored-by: Fabian Jakobs <fabian.jakobs@databricks.com>
2023-07-30 12:44:33 +00:00
git := &r.Bundle.Git
if target.Git.Branch != "" {
git.Branch = target.Git.Branch
Add validation for Git settings in bundles (#578) ## Changes This checks whether the Git settings are consistent with the actual Git state of a source directory. (This PR adds to https://github.com/databricks/cli/pull/577.) Previously, we would silently let users configure their Git branch to e.g. `main` and deploy with that metadata even if they were actually on a different branch. With these changes, the following config would result in an error when deployed from any other branch than `main`: ``` bundle: name: example workspace: git: branch: main environments: ... ``` > not on the right Git branch: > expected according to configuration: main > actual: my-feature-branch It's not very useful to set the same branch for all environments, though. For development, it's better to just let the CLI auto-detect the right branch. Therefore, it's now possible to set the branch just for a single environment: ``` bundle: name: example 2 environments: development: default: true production: # production can only be deployed from the 'main' branch git: branch: main ``` Adding to that, the `mode: production` option actually checks that users explicitly set the Git branch as seen above. Setting that branch helps avoid mistakes, where someone accidentally deploys to production from the wrong branch. (I could see us offering an escape hatch for that in the future.) # Testing Manual testing to validate the experience and error messages. Automated unit tests. --------- Co-authored-by: Fabian Jakobs <fabian.jakobs@databricks.com>
2023-07-30 12:44:33 +00:00
git.Inferred = false
}
if target.Git.Commit != "" {
git.Commit = target.Git.Commit
Add validation for Git settings in bundles (#578) ## Changes This checks whether the Git settings are consistent with the actual Git state of a source directory. (This PR adds to https://github.com/databricks/cli/pull/577.) Previously, we would silently let users configure their Git branch to e.g. `main` and deploy with that metadata even if they were actually on a different branch. With these changes, the following config would result in an error when deployed from any other branch than `main`: ``` bundle: name: example workspace: git: branch: main environments: ... ``` > not on the right Git branch: > expected according to configuration: main > actual: my-feature-branch It's not very useful to set the same branch for all environments, though. For development, it's better to just let the CLI auto-detect the right branch. Therefore, it's now possible to set the branch just for a single environment: ``` bundle: name: example 2 environments: development: default: true production: # production can only be deployed from the 'main' branch git: branch: main ``` Adding to that, the `mode: production` option actually checks that users explicitly set the Git branch as seen above. Setting that branch helps avoid mistakes, where someone accidentally deploys to production from the wrong branch. (I could see us offering an escape hatch for that in the future.) # Testing Manual testing to validate the experience and error messages. Automated unit tests. --------- Co-authored-by: Fabian Jakobs <fabian.jakobs@databricks.com>
2023-07-30 12:44:33 +00:00
}
if target.Git.OriginURL != "" {
git.OriginURL = target.Git.OriginURL
Add validation for Git settings in bundles (#578) ## Changes This checks whether the Git settings are consistent with the actual Git state of a source directory. (This PR adds to https://github.com/databricks/cli/pull/577.) Previously, we would silently let users configure their Git branch to e.g. `main` and deploy with that metadata even if they were actually on a different branch. With these changes, the following config would result in an error when deployed from any other branch than `main`: ``` bundle: name: example workspace: git: branch: main environments: ... ``` > not on the right Git branch: > expected according to configuration: main > actual: my-feature-branch It's not very useful to set the same branch for all environments, though. For development, it's better to just let the CLI auto-detect the right branch. Therefore, it's now possible to set the branch just for a single environment: ``` bundle: name: example 2 environments: development: default: true production: # production can only be deployed from the 'main' branch git: branch: main ``` Adding to that, the `mode: production` option actually checks that users explicitly set the Git branch as seen above. Setting that branch helps avoid mistakes, where someone accidentally deploys to production from the wrong branch. (I could see us offering an escape hatch for that in the future.) # Testing Manual testing to validate the experience and error messages. Automated unit tests. --------- Co-authored-by: Fabian Jakobs <fabian.jakobs@databricks.com>
2023-07-30 12:44:33 +00:00
}
return nil
}