Add Go-based entrypoint for Workbench session init (#871)

* add go program

* add multi-stage build dockerfile

* add positron

* fix path

* update err msgs and reuse file info

* merge comments

* enhance file copying with otiai10/copy package and update Dockerfile

* remove uneeded components

* add shared_run to common deps

* fix component filename

* update Dockerfile to use ubuntu:22.04 base image, as the scratch image doesn't work nicely with GOSS

* remove test var

---------

Co-authored-by: Ian Pittwood <ian.pittwood@posit.co>

workbench-session-init/Dockerfile.ubuntu2204

@@ -1,4 +1,4 @@
-FROM ubuntu:22.04 AS build
+FROM ubuntu:22.04 AS builder
 
 # Install required tools:
 # - ca-certificates installs necessary certificates to use cURL with HTTPS websites
@@ -9,17 +9,44 @@ RUN apt-get update && \
     rm -rf /var/lib/apt/lists/*
 
 ARG RSW_VERSION=2024.09.1+394.pro7
+ARG GO_VERSION=1.22.2
 
+# Download the RStudio Workbench session components and install Go
 SHELL ["/bin/bash", "-o", "pipefail", "-c"]
 RUN mkdir -p /pwb-staging && \
     RSW_VERSION_URL=$(echo -n "${RSW_VERSION}" | sed 's/+/-/g') && \
-    echo "Downloading https://s3.amazonaws.com/rstudio-ide-build/session/multi/x86_64/rsp-session-multi-linux-${RSW_VERSION_URL}-x86_64.tar.gz" && \
     curl -fsSL -o /pwb-staging/rsp-session-multi-linux.tar.gz "https://s3.amazonaws.com/rstudio-ide-build/session/multi/x86_64/rsp-session-multi-linux-${RSW_VERSION_URL}-x86_64.tar.gz" && \
     mkdir -p /opt/session-components && \
-    tar -C /opt/session-components -xf /pwb-staging/rsp-session-multi-linux.tar.gz && \
-    chmod -R 755 /opt/session-components && \
+    tar -C /opt/session-components -xpf /pwb-staging/rsp-session-multi-linux.tar.gz && \
+    chmod 755 /opt/session-components && \
+    curl -fsSL -o /pwb-staging/go.tar.gz "https://go.dev/dl/go${GO_VERSION}.linux-amd64.tar.gz" && \
+    tar -C /usr/local -xf /pwb-staging/go.tar.gz && \
     rm -rf /pwb-staging
 
-COPY --chmod=755 run.sh /usr/local/bin/run.sh
+# Add Go binary to PATH
+ENV PATH="/usr/local/go/bin:$PATH"
+
+# Set the Go workspace
+WORKDIR /workspace
+
+# Copy the Go source code and download dependencies
+COPY entrypoint/go.mod entrypoint/go.sum ./
+RUN go mod download
+
+# Copy the Go source code and build the binary
+COPY entrypoint/main.go .
+RUN CGO_ENABLED=0 GOOS=linux go build -ldflags '-s -w' -o entrypoint main.go
+
+# Create the final image
+FROM ubuntu:22.04 AS build
+
+ARG DEBIAN_FRONTEND=noninteractive
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends ca-certificates curl && \
+    rm -rf /var/lib/apt/lists/*
+
+# Copy the compiled Go binary and session components from the builder stage
+COPY --from=builder --chmod=755 /workspace/entrypoint /usr/local/bin/entrypoint
+COPY --from=builder --chmod=755 /opt/session-components /opt/session-components
 
-ENTRYPOINT ["/usr/local/bin/run.sh"]
+ENTRYPOINT ["/usr/local/bin/entrypoint"]

workbench-session-init/entrypoint/go.mod

@@ -0,0 +1,9 @@
+module entrypoint
+
+go 1.22.2
+
+require (
+	github.com/otiai10/copy v1.14.0 // indirect
+	golang.org/x/sync v0.3.0 // indirect
+	golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8 // indirect
+)

workbench-session-init/entrypoint/go.sum

@@ -0,0 +1,6 @@
+github.com/otiai10/copy v1.14.0 h1:dCI/t1iTdYGtkvCuBG2BgR6KZa83PTclw4U5n2wAllU=
+github.com/otiai10/copy v1.14.0/go.mod h1:ECfuL02W+/FkTWZWgQqXPWZgW9oeKCSQ5qVfSc4qc4w=
+golang.org/x/sync v0.3.0 h1:ftCYgMx6zT/asHUrPw8BLLscYtGznsLAnjq5RH9P66E=
+golang.org/x/sync v0.3.0/go.mod h1:FU7BRWz2tNW+3quACPkgCx/L+uEAv1htQ0V83Z9Rj+Y=
+golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8 h1:0A+M6Uqn+Eje4kHMK80dtF3JCXC4ykBgQG4Fe06QRhQ=
+golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=

workbench-session-init/entrypoint/main.go

@@ -0,0 +1,170 @@
+package main
+
+import (
+	"fmt"
+	"io"
+	"os"
+	"path/filepath"
+	"time"
+
+	cp "github.com/otiai10/copy"
+)
+
+const (
+	sourceDir = "/opt/session-components"
+	targetDir = "/mnt/init"
+)
+
+var (
+	// Read the PWB_SESSION_TYPE environment variable
+	sessionType = os.Getenv("PWB_SESSION_TYPE")
+
+	// Set the copy options.
+	// Preserve permissions, times, and owner.
+	opt = cp.Options{
+		PermissionControl: cp.PerservePermission,
+		PreserveTimes:     true,
+		PreserveOwner:     true,
+		NumOfWorkers:      20,
+	}
+
+	// List of dependencies common to all session types
+	commonDeps = []string{
+		"bin/git-credential-pwb",
+		"bin/focal",
+		"bin/jammy",
+		"bin/noble",
+		"bin/opensuse15",
+		"bin/postback",
+		"bin/pwb-supervisor",
+		"bin/quarto",
+		"bin/r-ldpath",
+		"bin/rhel8",
+		"bin/rhel9",
+		"bin/shared-run",
+		"R",
+		"resources",
+		"www",
+		"www-symbolmaps",
+	}
+
+	// Map of session-specific dependencies
+	sessionDeps = map[string][]string{
+		"jupyter": {
+			"bin/jupyter-session-run",
+			"bin/node",
+			"extras",
+		},
+		"positron": {
+			"bin/positron-server",
+			"bin/positron-session-run",
+			"extras",
+		},
+		"rstudio": {
+			"bin/node",
+			"bin/rsession-run",
+		},
+		"vscode": {
+			"bin/pwb-code-server",
+			"bin/vscode-session-run",
+			"extras",
+		},
+	}
+)
+
+func main() {
+	if sessionType == "" {
+		fmt.Println("PWB_SESSION_TYPE environment variable is not set")
+		os.Exit(1)
+	}
+
+	programStart := time.Now()
+	defer func() {
+		elapsed := time.Since(programStart)
+		fmt.Printf("Program took %s\n", elapsed)
+	}()
+
+	filesToCopy, err := getFilesToCopy(sessionType)
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+
+	err = validateTargetDir(targetDir)
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+
+	err = copyFiles(sourceDir, targetDir, filesToCopy)
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+	fmt.Println("Copy operation completed.")
+}
+
+// getFilesToCopy returns the list of files to copy based on the session type.
+func getFilesToCopy(sessionType string) ([]string, error) {
+	files := commonDeps
+	if deps, ok := sessionDeps[sessionType]; ok {
+		files = append(files, deps...)
+	} else {
+		return nil, fmt.Errorf("unknown session type: %s", sessionType)
+	}
+	return files, nil
+}
+
+// validateTargetDir checks if the target directory exists and is empty.
+func validateTargetDir(targetDir string) error {
+	if _, err := os.Stat(targetDir); os.IsNotExist(err) {
+		return fmt.Errorf("cannot find the copy target %s", targetDir)
+	}
+
+	isEmpty, err := isDirEmpty(targetDir)
+	if err != nil {
+		return fmt.Errorf("error checking if target directory is empty: %v", err)
+	}
+	if !isEmpty {
+		return fmt.Errorf("target directory %s is not empty", targetDir)
+	}
+
+	return nil
+}
+
+// isDirEmpty checks if a directory is empty.
+func isDirEmpty(dir string) (bool, error) {
+	f, err := os.Open(dir)
+	if err != nil {
+		return false, err
+	}
+	defer f.Close()
+
+	_, err = f.ReadDir(1)
+	if err == io.EOF {
+		return true, nil
+	}
+	return false, err
+}
+
+// copyFiles copies the files from the source directory to the target directory.
+// It uses the otiai10/copy package to copy files, with options to preserve
+// permissions, times, and owner.
+func copyFiles(src, dst string, filesToCopy []string) error {
+	fmt.Printf("Copying files from %s to %s\n", src, dst)
+	start := time.Now()
+
+	for _, file := range filesToCopy {
+		srcPath := filepath.Join(src, file)
+		dstPath := filepath.Join(dst, file)
+		err := cp.Copy(srcPath, dstPath, opt)
+		if err != nil {
+			return fmt.Errorf("error copying %s: %v", srcPath, err)
+		}
+	}
+
+	elapsed := time.Since(start)
+	fmt.Printf("Copy operation took %s\n", elapsed)
+
+	return nil
+}

workbench-session-init/entrypoint/main_test.go

@@ -0,0 +1,149 @@
+package main
+
+import (
+	"os"
+	"path/filepath"
+	"reflect"
+	"syscall"
+	"testing"
+)
+
+func TestGetFilesToCopy(t *testing.T) {
+	tests := []struct {
+		sessionType string
+		expected    []string
+		expectError bool
+	}{
+		{
+			sessionType: "jupyter",
+			expected:    append(commonDeps, sessionDeps["jupyter"]...),
+			expectError: false,
+		},
+		{
+			sessionType: "positron",
+			expected:    append(commonDeps, sessionDeps["positron"]...),
+			expectError: false,
+		},
+		{
+			sessionType: "rstudio",
+			expected:    append(commonDeps, sessionDeps["rstudio"]...),
+			expectError: false,
+		},
+		{
+			sessionType: "vscode",
+			expected:    append(commonDeps, sessionDeps["vscode"]...),
+			expectError: false,
+		},
+		{
+			sessionType: "unknown",
+			expected:    nil,
+			expectError: true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.sessionType, func(t *testing.T) {
+			files, err := getFilesToCopy(test.sessionType)
+			if test.expectError {
+				if err == nil {
+					t.Errorf("Expected error for session type %s, but got none", test.sessionType)
+				}
+			} else {
+				if err != nil {
+					t.Errorf("Did not expect error for session type %s, but got: %v", test.sessionType, err)
+				}
+				if !reflect.DeepEqual(files, test.expected) {
+					t.Errorf("Files do not match for session type %s. Expected: %v, Got: %v", test.sessionType, test.expected, files)
+				}
+			}
+		})
+	}
+}
+
+func TestCopy(t *testing.T) {
+	// Create temporary source and destination directories
+	srcDir, err := os.MkdirTemp("", "src")
+	if err != nil {
+		t.Fatalf("Failed to create temporary source directory: %v", err)
+	}
+	defer os.RemoveAll(srcDir)
+
+	dstDir, err := os.MkdirTemp("", "dst")
+	if err != nil {
+		t.Fatalf("Failed to create temporary destination directory: %v", err)
+	}
+	defer os.RemoveAll(dstDir)
+
+	// Create a sample directory structure in the source directory that looks like:
+	// srcDir
+	// ├── file1.txt
+	// └── subdir1
+	//     ├── file2.txt
+	//     └── subdir2
+	//         └── file3.txt
+	// |__ subdir3
+	err = os.MkdirAll(filepath.Join(srcDir, "subdir1"), 0755)
+	if err != nil {
+		t.Fatalf("Failed to create subdir1: %v", err)
+	}
+	err = os.WriteFile(filepath.Join(srcDir, "file1.txt"), []byte("file1 content"), 0644)
+	if err != nil {
+		t.Fatalf("Failed to create file1.txt: %v", err)
+	}
+	err = os.WriteFile(filepath.Join(srcDir, "subdir1", "file2.txt"), []byte("file2 content"), 0600)
+	if err != nil {
+		t.Fatalf("Failed to create file2.txt: %v", err)
+	}
+	err = os.MkdirAll(filepath.Join(srcDir, "subdir1", "subdir2"), 0755)
+	if err != nil {
+		t.Fatalf("Failed to create subdir2: %v", err)
+	}
+	err = os.WriteFile(filepath.Join(srcDir, "subdir1", "subdir2", "file3.txt"), []byte("file3 content"), 0644)
+	if err != nil {
+		t.Fatalf("Failed to create file3.txt: %v", err)
+	}
+	err = os.MkdirAll(filepath.Join(srcDir, "subdir3"), 0755)
+	if err != nil {
+		t.Fatalf("Failed to create subdir3: %v", err)
+	}
+
+	// Copy the directory structure from source to destination
+	// exclude subdir3
+	filesToCopy := []string{
+		"file1.txt",
+		"subdir1",
+	}
+	err = copyFiles(srcDir, dstDir, filesToCopy)
+	if err != nil {
+		t.Fatalf("Failed to copy files: %v", err)
+	}
+
+	// Verify that the directory structure and files are correctly copied
+	verifyFile(t, filepath.Join(dstDir, "file1.txt"), 0644, os.Getuid(), os.Getgid())
+	verifyFile(t, filepath.Join(dstDir, "subdir1", "file2.txt"), 0600, os.Getuid(), os.Getgid())
+	verifyFile(t, filepath.Join(dstDir, "subdir1", "subdir2", "file3.txt"), 0644, os.Getuid(), os.Getgid())
+	// Verify that subdir3 is not copied
+	if _, err := os.Stat(filepath.Join(dstDir, "subdir3")); !os.IsNotExist(err) {
+		t.Errorf("Directory subdir3 should not have been copied")
+	}
+}
+
+func verifyFile(t *testing.T, path string, mode os.FileMode, uid, gid int) {
+	info, err := os.Stat(path)
+	if err != nil {
+		t.Fatalf("Failed to stat file %s: %v", path, err)
+	}
+
+	if info.Mode() != mode {
+		t.Errorf("File %s has incorrect permissions: got %v, want %v", path, info.Mode(), mode)
+	}
+
+	stat, ok := info.Sys().(*syscall.Stat_t)
+	if !ok {
+		t.Fatalf("Failed to get file ownership for %s", path)
+	}
+
+	if int(stat.Uid) != uid || int(stat.Gid) != gid {
+		t.Errorf("File %s has incorrect ownership: got %d:%d, want %d:%d", path, stat.Uid, stat.Gid, uid, gid)
+	}
+}