Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enable offline install of labs projects #2049

Open
wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Enable offline install of labs projects #2049

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
98bdd8d
commit for offline install
HariGS-DB Dec 27, 2024
c514cca
commit for offlineinstall
HariGS-DB Dec 27, 2024
41313db
resolving review comments
HariGS-DB Jan 27, 2025
870003d
review comments
HariGS-DB Jan 27, 2025
4ed82de
changing the logic for offline install
HariGS-DB Jan 27, 2025
d846c91
added check for local dir exists in offline mode
HariGS-DB Jan 27, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 0 additions & 21 deletions .vscode/settings.json
View file
Open in desktop
HariGS-DB marked this conversation as resolved.
Show resolved Hide resolved

This file was deleted.

11 changes: 8 additions & 3 deletions cmd/labs/github/releases.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,24 +16,29 @@ const cacheTTL = 1 * time.Hour
func NewReleaseCache(org, repo, cacheDir string) *ReleaseCache {
pattern := fmt.Sprintf("%s-%s-releases", org, repo)
return &ReleaseCache{
cache: localcache.NewLocalCache[Versions](cacheDir, pattern, cacheTTL),
Cache: localcache.NewLocalCache[Versions](cacheDir, pattern, cacheTTL),
Org: org,
Repo: repo,
}
}

type ReleaseCache struct {
cache localcache.LocalCache[Versions]
Cache localcache.LocalCache[Versions]
Org string
Repo string
}

func (r *ReleaseCache) Load(ctx context.Context) (Versions, error) {
return r.cache.Load(ctx, func() (Versions, error) {
return r.Cache.Load(ctx, func() (Versions, error) {
return getVersions(ctx, r.Org, r.Repo)
})
}

func (r *ReleaseCache) LoadCache(ctx context.Context) (Versions, error) {
cached, err := r.Cache.LoadCache()
return cached.Data, err
}

HariGS-DB marked this conversation as resolved.
Show resolved Hide resolved
// getVersions is considered to be a private API, as we want the usage go through a cache
func getVersions(ctx context.Context, org, repo string) (Versions, error) {
var releases Versions
Expand Down
26 changes: 15 additions & 11 deletions cmd/labs/install.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,16 +7,20 @@ import (
)

func newInstallCommand() *cobra.Command {
return &cobra.Command{
Use: "install NAME",
Args: root.ExactArgs(1),
Short: "Installs project",
RunE: func(cmd *cobra.Command, args []string) error {
inst, err := project.NewInstaller(cmd, args[0])
if err != nil {
return err
}
return inst.Install(cmd.Context())
},
cmd := &cobra.Command{}
var offlineInstall bool

cmd.Flags().BoolVar(&offlineInstall, "offline-install", offlineInstall, `If installing in offline mode, set this flag to true.`)
HariGS-DB marked this conversation as resolved.
Show resolved Hide resolved

cmd.Use = "install NAME"
cmd.Args = root.ExactArgs(1)
cmd.Short = "Installs project"
cmd.RunE = func(cmd *cobra.Command, args []string) error {
inst, err := project.NewInstaller(cmd, args[0], offlineInstall)
if err != nil {
return err
}
return inst.Install(cmd.Context(), offlineInstall)
}
return cmd
}
4 changes: 2 additions & 2 deletions cmd/labs/localcache/jsonfile.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ type LocalCache[T any] struct {
}

func (r *LocalCache[T]) Load(ctx context.Context, refresh func() (T, error)) (T, error) {
cached, err := r.loadCache()
cached, err := r.LoadCache()
if errors.Is(err, fs.ErrNotExist) {
return r.refreshCache(ctx, refresh, r.zero)
} else if err != nil {
Expand Down Expand Up @@ -96,7 +96,7 @@ func (r *LocalCache[T]) FileName() string {
return filepath.Join(r.dir, fmt.Sprintf("%s.json", r.name))
}

func (r *LocalCache[T]) loadCache() (*cached[T], error) {
func (r *LocalCache[T]) LoadCache() (*cached[T], error) {
jsonFile := r.FileName()
raw, err := os.ReadFile(r.FileName())
if err != nil {
Expand Down
43 changes: 32 additions & 11 deletions cmd/labs/project/fetcher.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,15 +15,15 @@ import (
)

type installable interface {
Install(ctx context.Context) error
Install(ctx context.Context, offlineInstall bool) error
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This should be a property on the installable, not a function argument.

}

type devInstallation struct {
*Project
*cobra.Command
}

func (d *devInstallation) Install(ctx context.Context) error {
func (d *devInstallation) Install(ctx context.Context, offlineInstall bool) error {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For example, not needed here.

if d.Installer == nil {
return nil
}
Expand Down Expand Up @@ -54,7 +54,7 @@ func (d *devInstallation) Install(ctx context.Context) error {
return d.Installer.runHook(d.Command)
}

func NewInstaller(cmd *cobra.Command, name string) (installable, error) {
func NewInstaller(cmd *cobra.Command, name string, offlineInstall bool) (installable, error) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The bool should be a property on the installer struct.

if name == "." {
wd, err := os.Getwd()
if err != nil {
Expand All @@ -75,14 +75,17 @@ func NewInstaller(cmd *cobra.Command, name string) (installable, error) {
version = "latest"
}
f := &fetcher{name}
version, err := f.checkReleasedVersions(cmd, version)

version, err := f.checkReleasedVersions(cmd, version, offlineInstall)
if err != nil {
return nil, fmt.Errorf("version: %w", err)
}
prj, err := f.loadRemoteProjectDefinition(cmd, version)

prj, err := f.loadRemoteProjectDefinition(cmd, version, offlineInstall)
if err != nil {
return nil, fmt.Errorf("remote: %w", err)
}

return &installer{
Project: prj,
version: version,
Expand All @@ -92,11 +95,11 @@ func NewInstaller(cmd *cobra.Command, name string) (installable, error) {

func NewUpgrader(cmd *cobra.Command, name string) (*installer, error) {
f := &fetcher{name}
version, err := f.checkReleasedVersions(cmd, "latest")
version, err := f.checkReleasedVersions(cmd, "latest", false)
if err != nil {
return nil, fmt.Errorf("version: %w", err)
}
prj, err := f.loadRemoteProjectDefinition(cmd, version)
prj, err := f.loadRemoteProjectDefinition(cmd, version, false)
if err != nil {
return nil, fmt.Errorf("remote: %w", err)
}
Expand All @@ -115,15 +118,21 @@ type fetcher struct {
name string
}

func (f *fetcher) checkReleasedVersions(cmd *cobra.Command, version string) (string, error) {
func (f *fetcher) checkReleasedVersions(cmd *cobra.Command, version string, offlineInstall bool) (string, error) {
ctx := cmd.Context()
cacheDir, err := PathInLabs(ctx, f.name, "cache")
if err != nil {
return "", err
}
// `databricks labs isntall X` doesn't know which exact version to fetch, so first
// we fetch all versions and then pick the latest one dynamically.
versions, err := github.NewReleaseCache("databrickslabs", f.name, cacheDir).Load(ctx)
var versions github.Versions
if offlineInstall {
versions, err = github.NewReleaseCache("databrickslabs", f.name, cacheDir).LoadCache(ctx)
} else {
versions, err = github.NewReleaseCache("databrickslabs", f.name, cacheDir).Load(ctx)
}

if err != nil {
return "", fmt.Errorf("versions: %w", err)
}
Expand All @@ -140,9 +149,21 @@ func (f *fetcher) checkReleasedVersions(cmd *cobra.Command, version string) (str
return version, nil
}

func (i *fetcher) loadRemoteProjectDefinition(cmd *cobra.Command, version string) (*Project, error) {
func (i *fetcher) loadRemoteProjectDefinition(cmd *cobra.Command, version string, offlineInstall bool) (*Project, error) {
ctx := cmd.Context()
raw, err := github.ReadFileFromRef(ctx, "databrickslabs", i.name, version, "labs.yml")
var raw []byte
var err error
if !offlineInstall {
raw, err = github.ReadFileFromRef(ctx, "databrickslabs", i.name, version, "labs.yml")
} else {
libDir, file_err := PathInLabs(ctx, i.name, "lib")
if file_err != nil {
return nil, file_err
}
fileName := filepath.Join(libDir, "labs.yml")
raw, err = os.ReadFile(fileName)
}

if err != nil {
return nil, fmt.Errorf("read labs.yml from GitHub: %w", err)
}
Expand Down
10 changes: 6 additions & 4 deletions cmd/labs/project/installer.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,7 +79,7 @@ type installer struct {
cmd *cobra.Command
}

func (i *installer) Install(ctx context.Context) error {
func (i *installer) Install(ctx context.Context, offlineInstall bool) error {
err := i.EnsureFoldersExist()
if err != nil {
return fmt.Errorf("folders: %w", err)
Expand All @@ -101,9 +101,11 @@ func (i *installer) Install(ctx context.Context) error {
} else if err != nil {
return fmt.Errorf("login: %w", err)
}
err = i.downloadLibrary(ctx)
if err != nil {
return fmt.Errorf("lib: %w", err)
if !offlineInstall {
err = i.downloadLibrary(ctx)
if err != nil {
return fmt.Errorf("lib: %w", err)
}
HariGS-DB marked this conversation as resolved.
Show resolved Hide resolved
}
err = i.setupPythonVirtualEnvironment(ctx, w)
if err != nil {
Expand Down
34 changes: 34 additions & 0 deletions cmd/labs/project/installer_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -240,6 +240,40 @@ func TestInstallerWorksForReleases(t *testing.T) {
r.RunAndExpectOutput("setting up important infrastructure")
}

func TestOfflineInstallerWorksForReleases(t *testing.T) {
server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if r.URL.Path == "/api/2.1/clusters/get" {
respondWithJSON(t, w, &compute.ClusterDetails{
State: compute.StateRunning,
})
return
}
t.Logf("Requested: %s", r.URL.Path)
t.FailNow()
}))
defer server.Close()

ctx := installerContext(t, server)
newHome := copyTestdata(t, "testdata/installed-in-home")
ctx = env.WithUserHomeDir(ctx, newHome)

ctx, stub := process.WithStub(ctx)
stub.WithStdoutFor(`python[\S]+ --version`, "Python 3.10.5")
// on Unix, we call `python3`, but on Windows it is `python.exe`
stub.WithStderrFor(`python[\S]+ -m venv .*/.databricks/labs/blueprint/state/venv`, "[mock venv create]")
stub.WithStderrFor(`python[\S]+ -m pip install --upgrade --upgrade-strategy eager .`, "[mock pip install]")
stub.WithStdoutFor(`python[\S]+ install.py`, "setting up important infrastructure")

// simulate the case of GitHub Actions
ctx = env.Set(ctx, "DATABRICKS_HOST", server.URL)
ctx = env.Set(ctx, "DATABRICKS_TOKEN", "...")
ctx = env.Set(ctx, "DATABRICKS_CLUSTER_ID", "installer-cluster")
ctx = env.Set(ctx, "DATABRICKS_WAREHOUSE_ID", "installer-warehouse")

r := testcli.NewRunner(t, ctx, "labs", "install", "blueprint", "--offline-install=true", "--debug")
r.RunAndExpectOutput("setting up important infrastructure")
}

func TestInstallerWorksForDevelopment(t *testing.T) {
defer func() {
if !t.Failed() {
Expand Down
Loading