steve/databricks-cli
steve
/
databricks-cli
mirror of https://github.com/databricks/cli.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use `vfs.Path` for filesystem interaction (#1554) 

## Changes

Note: this doesn't cover _all_ filesystem interaction.

To intercept calls where read or stat files to determine their type, we
need a layer between our code and the `os` package calls that interact
with the local file system. Interception is necessary to accommodate
differences between a regular local file system and the FUSE-mounted
Workspace File System when running the CLI on DBR.

This change makes use of #1452 in the bundle struct.

It uses #1525 to access the bundle variable in path rewriting.

## Tests

* Unit tests pass.
* Integration tests pass.
This commit is contained in:
Pieter Noordhuis 2024-07-03 12:13:22 +02:00 committed by GitHub
parent 4787edba36
commit b3c044c461
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
12 changed files with 61 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
bundle/bundle.go
View File

@ -17,7 +17,6 @@ import (
"github.com/databricks/cli/bundle/env"
"github.com/databricks/cli/bundle/metadata"
"github.com/databricks/cli/libs/fileset"
"github.com/databricks/cli/libs/folders"
"github.com/databricks/cli/libs/git"
"github.com/databricks/cli/libs/locker"
"github.com/databricks/cli/libs/log"
@ -36,6 +35,10 @@ type Bundle struct {
// It is set when we instantiate a new bundle instance.
RootPath string
// BundleRoot is a virtual filesystem path to the root of the bundle.
// Exclusively use this field for filesystem operations.
BundleRoot vfs.Path
Config config.Root
// Metadata about the bundle deployment. This is the interface Databricks services
@ -74,6 +77,7 @@ type Bundle struct {
func Load(ctx context.Context, path string) (*Bundle, error) {
b := &Bundle{
RootPath: filepath.Clean(path),
BundleRoot: vfs.MustNew(path),
}
configFile, err := config.FileNames.FindInPath(path)
if err != nil {
@ -208,12 +212,12 @@ func (b *Bundle) GetSyncIncludePatterns(ctx context.Context) ([]string, error) {
}
func (b *Bundle) GitRepository() (*git.Repository, error) {
rootPath, err := folders.FindDirWithLeaf(b.RootPath, ".git")
_, err := vfs.FindLeafInTree(b.BundleRoot, ".git")
if err != nil {
return nil, fmt.Errorf("unable to locate repository root: %w", err)
}
return git.NewRepository(vfs.MustNew(rootPath))
return git.NewRepository(b.BundleRoot)
}
// AuthEnv returns a map with environment variables and their values

5
bundle/bundle_read_only.go
View File

@ -4,6 +4,7 @@ import (
"context"
"github.com/databricks/cli/bundle/config"
"github.com/databricks/cli/libs/vfs"
"github.com/databricks/databricks-sdk-go"
)
@ -23,6 +24,10 @@ func (r ReadOnlyBundle) RootPath() string {
return r.b.RootPath
}
func (r ReadOnlyBundle) BundleRoot() vfs.Path {
return r.b.BundleRoot
}
func (r ReadOnlyBundle) WorkspaceClient() *databricks.WorkspaceClient {
return r.b.WorkspaceClient()
}

3
bundle/config/mutator/load_git_details.go
View File

@ -8,7 +8,6 @@ import (
"github.com/databricks/cli/libs/diag"
"github.com/databricks/cli/libs/git"
"github.com/databricks/cli/libs/log"
"github.com/databricks/cli/libs/vfs"
)
type loadGitDetails struct{}
@ -23,7 +22,7 @@ func (m *loadGitDetails) Name() string {
func (m *loadGitDetails) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostics {
// Load relevant git repository
repo, err := git.NewRepository(vfs.MustNew(b.RootPath))
repo, err := git.NewRepository(b.BundleRoot)
if err != nil {
return diag.FromErr(err)
}

7
bundle/config/mutator/translate_paths.go
View File

@ -6,7 +6,6 @@ import (
"fmt"
"io/fs"
"net/url"
"os"
"path"
"path/filepath"
"strings"
@ -119,7 +118,7 @@ func (t *translateContext) rewritePath(
}
func (t *translateContext) translateNotebookPath(literal, localFullPath, localRelPath, remotePath string) (string, error) {
nb, _, err := notebook.Detect(localFullPath)
nb, _, err := notebook.DetectWithFS(t.b.BundleRoot, filepath.ToSlash(localRelPath))
if errors.Is(err, fs.ErrNotExist) {
return "", fmt.Errorf("notebook %s not found", literal)
}
@ -135,7 +134,7 @@ func (t *translateContext) translateNotebookPath(literal, localFullPath, localRe
}
func (t *translateContext) translateFilePath(literal, localFullPath, localRelPath, remotePath string) (string, error) {
nb, _, err := notebook.Detect(localFullPath)
nb, _, err := notebook.DetectWithFS(t.b.BundleRoot, filepath.ToSlash(localRelPath))
if errors.Is(err, fs.ErrNotExist) {
return "", fmt.Errorf("file %s not found", literal)
}
@ -149,7 +148,7 @@ func (t *translateContext) translateFilePath(literal, localFullPath, localRelPat
}
func (t *translateContext) translateDirectoryPath(literal, localFullPath, localRelPath, remotePath string) (string, error) {
info, err := os.Stat(localFullPath)
info, err := t.b.BundleRoot.Stat(filepath.ToSlash(localRelPath))
if err != nil {
return "", err
}

14
bundle/config/mutator/translate_paths_test.go
View File

@ -12,6 +12,7 @@ import (
"github.com/databricks/cli/bundle/config/mutator"
"github.com/databricks/cli/bundle/config/resources"
"github.com/databricks/cli/bundle/internal/bundletest"
"github.com/databricks/cli/libs/vfs"
"github.com/databricks/databricks-sdk-go/service/compute"
"github.com/databricks/databricks-sdk-go/service/jobs"
"github.com/databricks/databricks-sdk-go/service/pipelines"
@ -38,6 +39,7 @@ func TestTranslatePathsSkippedWithGitSource(t *testing.T) {
dir := t.TempDir()
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -108,6 +110,7 @@ func TestTranslatePaths(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -275,6 +278,7 @@ func TestTranslatePathsInSubdirectories(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -369,6 +373,7 @@ func TestTranslatePathsOutsideBundleRoot(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -402,6 +407,7 @@ func TestJobNotebookDoesNotExistError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Resources: config.Resources{
Jobs: map[string]*resources.Job{
@ -432,6 +438,7 @@ func TestJobFileDoesNotExistError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Resources: config.Resources{
Jobs: map[string]*resources.Job{
@ -462,6 +469,7 @@ func TestPipelineNotebookDoesNotExistError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Resources: config.Resources{
Pipelines: map[string]*resources.Pipeline{
@ -492,6 +500,7 @@ func TestPipelineFileDoesNotExistError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Resources: config.Resources{
Pipelines: map[string]*resources.Pipeline{
@ -523,6 +532,7 @@ func TestJobSparkPythonTaskWithNotebookSourceError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -557,6 +567,7 @@ func TestJobNotebookTaskWithFileSourceError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -591,6 +602,7 @@ func TestPipelineNotebookLibraryWithFileSourceError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -625,6 +637,7 @@ func TestPipelineFileLibraryWithNotebookSourceError(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Workspace: config.Workspace{
FilePath: "/bundle",
@ -660,6 +673,7 @@ func TestTranslatePathJobEnvironments(t *testing.T) {
b := &bundle.Bundle{
RootPath: dir,
BundleRoot: vfs.MustNew(dir),
Config: config.Root{
Resources: config.Resources{
Jobs: map[string]*resources.Job{

3
bundle/config/validate/validate_sync_patterns.go
View File

@ -8,7 +8,6 @@ import (
"github.com/databricks/cli/bundle"
"github.com/databricks/cli/libs/diag"
"github.com/databricks/cli/libs/fileset"
"github.com/databricks/cli/libs/vfs"
"golang.org/x/sync/errgroup"
)
@ -51,7 +50,7 @@ func checkPatterns(patterns []string, path string, rb bundle.ReadOnlyBundle) (di
index := i
p := pattern
errs.Go(func() error {
fs, err := fileset.NewGlobSet(vfs.MustNew(rb.RootPath()), []string{p})
fs, err := fileset.NewGlobSet(rb.BundleRoot(), []string{p})
if err != nil {
return err
}

3
bundle/deploy/files/sync.go
View File

@ -6,7 +6,6 @@ import (
"github.com/databricks/cli/bundle"
"github.com/databricks/cli/libs/sync"
"github.com/databricks/cli/libs/vfs"
)
func GetSync(ctx context.Context, rb bundle.ReadOnlyBundle) (*sync.Sync, error) {
@ -29,7 +28,7 @@ func GetSyncOptions(ctx context.Context, rb bundle.ReadOnlyBundle) (*sync.SyncOp
}
opts := &sync.SyncOptions{
LocalPath: vfs.MustNew(rb.RootPath()),
LocalPath: rb.BundleRoot(),
RemotePath: rb.Config().Workspace.FilePath,
Include: includes,
Exclude: rb.Config().Sync.Exclude,

10
bundle/deploy/state.go
View File

@ -6,7 +6,6 @@ import (
"fmt"
"io"
"io/fs"
"os"
"path/filepath"
"time"
@ -59,8 +58,8 @@ type entry struct {
info fs.FileInfo
}
func newEntry(path string) *entry {
info, err := os.Stat(path)
func newEntry(root vfs.Path, path string) *entry {
info, err := root.Stat(path)
if err != nil {
return &entry{path, nil}
}
@ -111,11 +110,10 @@ func FromSlice(files []fileset.File) (Filelist, error) {
return f, nil
}
func (f Filelist) ToSlice(basePath string) []fileset.File {
func (f Filelist) ToSlice(root vfs.Path) []fileset.File {
var files []fileset.File
root := vfs.MustNew(basePath)
for _, file := range f {
entry := newEntry(filepath.Join(basePath, file.LocalPath))
entry := newEntry(root, filepath.ToSlash(file.LocalPath))
// Snapshots created with versions <= v0.220.0 use platform-specific
// paths (i.e. with backslashes). Files returned by [libs/fileset] always

2
bundle/deploy/state_pull.go
View File

@ -85,7 +85,7 @@ func (s *statePull) Apply(ctx context.Context, b *bundle.Bundle) diag.Diagnostic
}
log.Infof(ctx, "Creating new snapshot")
snapshot, err := sync.NewSnapshot(state.Files.ToSlice(b.RootPath), opts)
snapshot, err := sync.NewSnapshot(state.Files.ToSlice(b.BundleRoot), opts)
if err != nil {
return diag.FromErr(err)
}

5
bundle/deploy/state_pull_test.go
View File

@ -17,6 +17,7 @@ import (
"github.com/databricks/cli/internal/testutil"
"github.com/databricks/cli/libs/filer"
"github.com/databricks/cli/libs/sync"
"github.com/databricks/cli/libs/vfs"
"github.com/databricks/databricks-sdk-go/service/iam"
"github.com/stretchr/testify/mock"
"github.com/stretchr/testify/require"
@ -59,8 +60,10 @@ func testStatePull(t *testing.T, opts statePullOpts) {
return f, nil
}}
tmpDir := t.TempDir()
b := &bundle.Bundle{
RootPath: t.TempDir(),
RootPath: tmpDir,
BundleRoot: vfs.MustNew(tmpDir),
Config: config.Root{
Bundle: config.Bundle{
Target: "default",

5
bundle/deploy/state_test.go
View File

@ -32,7 +32,8 @@ func TestFromSlice(t *testing.T) {
func TestToSlice(t *testing.T) {
tmpDir := t.TempDir()
fileset := fileset.New(vfs.MustNew(tmpDir))
root := vfs.MustNew(tmpDir)
fileset := fileset.New(root)
testutil.Touch(t, tmpDir, "test1.py")
testutil.Touch(t, tmpDir, "test2.py")
testutil.Touch(t, tmpDir, "test3.py")
@ -44,7 +45,7 @@ func TestToSlice(t *testing.T) {
require.NoError(t, err)
require.Len(t, f, 3)
s := f.ToSlice(tmpDir)
s := f.ToSlice(root)
require.Len(t, s, 3)
for _, file := range s {

2
cmd/sync/sync_test.go
View File

@ -9,6 +9,7 @@ import (
"github.com/databricks/cli/bundle"
"github.com/databricks/cli/bundle/config"
"github.com/databricks/cli/cmd/root"
"github.com/databricks/cli/libs/vfs"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
@ -17,6 +18,7 @@ func TestSyncOptionsFromBundle(t *testing.T) {
tempDir := t.TempDir()
b := &bundle.Bundle{
RootPath: tempDir,
BundleRoot: vfs.MustNew(tempDir),
Config: config.Root{
Bundle: config.Bundle{
Target: "default",