Gade Architecture

Version: 1.0.0 Last Updated: February 3, 2026

This document describes the architecture of Gade (Groovy Analytics Development Environment), covering the major components, runtime execution models, communication protocols, and extension points.

System Overview
Component Architecture
Runtime Execution Models
JSON-RPC Protocol
Code Completion System
Extension Points
Build System
Data Flow

System Overview

High-Level Architecture

┌──────────────────────────────────────────────────────────────────┐
│                         Gade Main Process                         │
│  ┌────────────────────────────────────────────────────────────┐  │
│  │                    JavaFX UI Layer                          │  │
│  │  ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────────┐  │  │
│  │  │ Editor   │ │ Console  │ │ Environment│ │ Connections │  │  │
│  │  │ Tabs     │ │ Output   │ │ Variables │ │ Database    │  │  │
│  │  └──────────┘ └──────────┘ └──────────┘ └──────────────┘  │  │
│  └────────────────────────────────────────────────────────────┘  │
│  ┌────────────────────────────────────────────────────────────┐  │
│  │                  Application Logic                          │  │
│  │  ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────────┐  │  │
│  │  │ Code     │ │ Runtime  │ │ Git      │ │ File         │  │  │
│  │  │ Completion│ │ Manager  │ │ Integration│ │ Manager   │  │  │
│  │  └──────────┘ └──────────┘ └──────────┘ └──────────────┘  │  │
│  └────────────────────────────────────────────────────────────┘  │
│  ┌────────────────────────────────────────────────────────────┐  │
│  │                  Embedded Runtimes                          │  │
│  │  ┌──────────────────────────────────────────────────────┐  │  │
│  │  │ Gade Runtime (GroovyShell) - In-Process              │  │  │
│  │  └──────────────────────────────────────────────────────┘  │  │
│  └────────────────────────────────────────────────────────────┘  │
└──────────────────────────────────────────────────────────────────┘
           │ JSON-RPC                │ JSON-RPC
           │ (Socket)                │ (Socket)
           ↓                         ↓
┌─────────────────────────┐ ┌─────────────────────────┐
│  Gradle Runtime Process │ │  Maven Runtime Process  │
│  (Separate JVM)         │ │  (Separate JVM)         │
│  ┌───────────────────┐  │ │  ┌───────────────────┐  │
│  │ GradleRunner      │  │ │  │ MavenRunner       │  │
│  │ Groovy Engine     │  │ │  │ Groovy Engine     │  │
│  │ Gradle Tooling API│  │ │  │ Maven Embedder    │  │
│  └───────────────────┘  │ │  └───────────────────┘  │
└─────────────────────────┘ └─────────────────────────┘

Design Principles

Process Isolation: Gradle and Maven runtimes execute in separate JVMs to prevent classpath conflicts
Pluggable Architecture: Code completion, syntax highlighting, and runtimes are extensible
Reactive UI: JavaFX Platform.runLater() ensures UI remains responsive during long operations
Caching: Aggressive caching of classpaths, completions, and Git state for performance
Single Responsibility: Each component has a focused purpose (see Sprint 3 refactors)

Component Architecture

1. UI Layer (JavaFX)

Package: se.alipsa.gade

Main Window (Gade.java)

Application entry point
Menu bar, toolbar, tabs, console
Coordinates UI components

Editor Components

EditorTab - Base class for all editor tabs
GroovyTab - Groovy script editor
SqlTab - SQL script editor
JsTab - JavaScript editor
MarkdownTab - Markdown viewer/editor

Each tab extends ExecutableTab which provides:

Execute button management
TaskListener implementation (disable during execution)
Common file operations

See: docs/adr/002-executable-tab-pattern.md

Console Components

ConsoleTextArea - Output display with syntax coloring
GroovyEngine - Embedded Groovy execution engine
RuntimeProcessRunner - External runtime process manager

Refactored in Sprint 3 (Task #10): ConsoleComponent split into:

ConsoleButtonPanel - Button management
ConsoleEnvironmentPane - Variable display
ConsoleTextArea - Output display

2. Code Completion System

Package: se.alipsa.gade.code.completion

Core Classes

CompletionRegistry (Singleton, Thread-Safe)

Central registry for all completion engines
Language → Engine mapping
Cache invalidation coordination

CompletionEngine (Interface)

complete(CompletionContext) → List<CompletionItem>
supportedLanguages() → Set<String>
invalidateCache() - Called when classpath changes

CompletionContext (Immutable)

Full document text
Caret position
Utility methods: isMemberAccess(), isInsideString(), isInsideComment()

CompletionItem (Immutable)

Completion text, display text, kind (method, field, keyword, etc.)
Sort priority, cursor offset
Builder pattern for complex items

Engines

GroovyCompletionEngine

Uses ClassGraph for fast class scanning
Reflection for member access
Caches scanned classes (invalidated on classpath change)

SqlCompletionEngine

SQL keywords, functions
Table/column names from connected databases
JSqlParser for syntax analysis

JavaScriptCompletionEngine

JavaScript keywords
DOM API methods
Node.js completions

See: docs/adr/003-completion-engine-architecture.md

3. Runtime System

Package: se.alipsa.gade.runtime

Runtime Types

GadeRuntime (Embedded)

In-process GroovyShell execution
Fast, no IPC overhead
No external dependency management
Use case: Quick scripts, exploration

GradleRuntime (Subprocess)

Separate JVM process
Full Gradle Tooling API
Dependency resolution via build.gradle
Use case: Projects with dependencies

MavenRuntime (Subprocess)

Separate JVM process
maven-utils library for classpath resolution
Dependency resolution via pom.xml
Use case: Enterprise projects, POM inheritance

CustomRuntime (Subprocess)

User-specified Java executable
Custom classpath
Use case: Testing against specific JDK versions

See: docs/adr/001-separate-process-runtimes.md

RuntimeProcessRunner

Manages external runtime processes:

Responsibilities:

Start/stop runtime subprocess
JSON-RPC message handling
stdout/stderr forwarding to console
Script evaluation coordination
Graceful shutdown

Lifecycle:

Start subprocess with GadeRunnerMain
Wait for "hello" handshake
Send "eval" requests with script + bindings
Receive "result", "out", "err", "bindings" messages
On completion, send "shutdown" or kill process

Thread Safety: All public methods are synchronized

4. Gradle Integration

Package: se.alipsa.gade.utils.gradle

Refactored in Sprint 3 (Task #10): GradleUtils split into:

GradleUtils

Gradle Tooling API connection
Classpath resolution
Dependency introspection
Error detection with helpful hints

Key Methods:

resolveClasspath(File projectDir) → Set<File>
getCacheDir() → File
hasGradleProject(File dir) → boolean
getProjectFingerprint(File dir) → String

GradleCacheManager

Classpath fingerprinting (build.gradle + settings.gradle hash)
JSON-based cache storage
Cache invalidation on file changes

Cache Location: ~/.gade/gradle-cache/<project-hash>.json

GradleErrorDetector

Parse Gradle errors
Suggest fixes for common issues
Example: "Could not find X" → "Add repository: mavenCentral()"

Thread Safety: All classes are thread-safe

5. Git Integration

Package: se.alipsa.gade.inout.git

Library: JGit (Eclipse)

Features:

Clone, init, commit, push, pull
Branch management
Diff viewing
Stash operations
Tag management

Components:

GitUtils - JGit wrapper methods
GitStatusMonitor - Background thread monitoring .git directory
GitDialog - UI dialogs for Git operations

6. Database Integration

Package: se.alipsa.gade.environment.connections

Connection Types:

JDBC (standard databases)
BigQuery (Google Cloud Platform)

Components:

ConnectionHandler - JDBC connection management
ConnectionInfo - Connection metadata storage
ConnectionsTab - UI for managing connections

JDBC Driver Loading:

Dynamic classloading via data-utils library
No system classloader pollution
Add drivers via UI or Gradle dependencies

7. Build System

Tool: Gradle 9.3.1

Plugins:

application - Main class specification
groovy - Groovy compilation
org.javamodularity.moduleplugin - Java modules (JPMS)
org.beryx.runtime - Platform-specific distributions
org.openjfx.javafxplugin - JavaFX dependencies
jacoco - Code coverage
me.champeau.jmh - Performance benchmarks

Key Build Tasks:

./gradlew run             # Development run
./gradlew build           # Compile + test
./gradlew test            # Run all tests
./gradlew jacocoTestReport # Generate coverage report
./gradlew jmh             # Run benchmarks
./gradlew runtimeZip      # Create platform-specific distributions

Runtime Execution Models

Gade Runtime (Embedded)

┌─────────────────────────────────────────────┐
│         Gade Main Process                   │
│                                             │
│  User Script                                │
│      ↓                                      │
│  ┌────────────────────────────────────┐    │
│  │ GroovyShell.evaluate(script)       │    │
│  │   - Same classloader as Gade       │    │
│  │   - Access to all Gade classes     │    │
│  │   - stdout → ConsoleTextArea       │    │
│  └────────────────────────────────────┘    │
│      ↓                                      │
│  Result → Environment Tab                   │
└─────────────────────────────────────────────┘

Advantages:

⚡ Fast (no IPC overhead)
Simple (no subprocess management)

Limitations:

❌ No external dependency management
❌ Classpath shared with Gade (potential conflicts)

Gradle Runtime (Subprocess)

┌─────────────────────────────┐          ┌──────────────────────────┐
│   Gade Main Process         │          │  Gradle Runtime Process  │
│                             │          │                          │
│  User Script                │          │  GradleRunnerMain        │
│      ↓                      │          │                          │
│  ┌────────────────────┐    │ Socket   │  ┌────────────────────┐ │
│  │ RuntimeProcessRunner│◄───┼──────────┤►│ JSON-RPC Server    │ │
│  │  - Start process    │    │  JSON    │  │  - Receive "eval"  │ │
│  │  - Send "eval"      │    │  RPC     │  │  - Execute script  │ │
│  │  - Receive "result" │    │          │  │  - Send "result"   │ │
│  └────────────────────┘    │          │  └────────────────────┘ │
│      ↓                      │          │         ↓               │
│  Result → Environment Tab   │          │  GroovyShell.evaluate() │
│                             │          │  (with Gradle classpath)│
└─────────────────────────────┘          └──────────────────────────┘

Gradle Classpath Resolution:

Read build.gradle
Use Gradle Tooling API to resolve dependencies
Cache resolved classpath (fingerprinted by build.gradle hash)
Pass classpath to subprocess via command-line argument

Advantages:

✅ Full Gradle support (dependencies, plugins, etc.)
✅ Isolated classpath (no conflicts with Gade)

Limitations:

⏱️ Slower first run (dependency download)
⏱️ Subprocess startup overhead (~1-2 seconds)

Maven Runtime (Subprocess)

Similar to Gradle, but uses:

pom.xml instead of build.gradle
maven-utils library instead of Gradle Tooling API
MavenRunnerMain subprocess

JSON-RPC Protocol

Version: 1.0 (as of Sprint 3, Task #13)

See: src/main/java/se/alipsa/gade/runtime/ProtocolVersion.java

Message Types

1. Handshake (Runner → Gade)

Purpose: Subprocess signals readiness

{
  "type": "hello",
  "port": 12345,
  "protocolVersion": "1.0"
}

Gade Response: Stores port, marks runner as ready

2. Eval Request (Gade → Runner)

Purpose: Execute a script

{
  "cmd": "eval",
  "id": "uuid-1234",
  "script": "println 'hello'\n1 + 2",
  "bindings": {
    "customVar": "value"
  }
}

Fields:

id - Unique request ID (used to correlate response)
script - Groovy code to execute
bindings - Variables to inject into script context

3. Result (Runner → Gade)

Purpose: Return script result

{
  "type": "result",
  "id": "uuid-1234",
  "result": "3"
}

4. Error (Runner → Gade)

Purpose: Report script execution error

{
  "type": "error",
  "id": "uuid-1234",
  "error": "No such property: foo",
  "stacktrace": "groovy.lang.MissingPropertyException: ...\n  at ..."
}

5. Output (Runner → Gade)

Purpose: Forward stdout/stderr

{
  "type": "out",
  "text": "hello\n"
}

{
  "type": "err",
  "text": "Error: something failed\n"
}

6. Bindings (Runner → Gade)

Purpose: Update environment variables after script execution

{
  "type": "bindings",
  "bindings": {
    "x": "10",
    "y": "[1, 2, 3]"
  }
}

Gade Response: Updates Environment tab with new variables

7. Shutdown (Gade → Runner)

Purpose: Gracefully shut down subprocess

{
  "cmd": "shutdown"
}

Runner Response: Exits with code 0

Protocol Flow

Gade                                  Runner (Subprocess)
  │                                         │
  ├────────────── start process ──────────►│
  │                                         ├─ listen on random port
  │                                         │
  │◄────── hello (port, version) ──────────┤
  │                                         │
  ├────── eval (id, script, bindings) ────►│
  │                                         ├─ execute script
  │◄──────────── out ("hello") ────────────┤
  │◄──────────── result (id, "3") ─────────┤
  │◄──────────── bindings ─────────────────┤
  │                                         │
  ├────── eval (id2, script2, ...) ───────►│
  │◄──────────── error (id2, ...) ─────────┤
  │                                         │
  ├──────────── shutdown ──────────────────►│
  │                                         ├─ exit(0)

Error Handling

Connection Failures:

Gade waits 10 seconds for "hello" handshake
If timeout, kills process and shows error

Script Errors:

Runner catches all exceptions
Sends "error" message with stacktrace
Does NOT crash subprocess (can run next script)

Runner Crash:

Gade detects process exit
Shows "Runtime process terminated unexpectedly"
User can restart runtime

Code Completion System

Architecture

User Types → TextArea Event → CompletionPopup
                                     ↓
                            CompletionRegistry.getCompletions()
                                     ↓
                       ┌─────────────┴─────────────┐
                       ↓                           ↓
              GroovyCompletionEngine      SqlCompletionEngine
                       ↓                           ↓
         ┌─────────────┴──────────┐      ┌────────┴────────┐
         ↓                        ↓       ↓                 ↓
  ClassGraph Scan          Reflection   SQL Keywords   DB Metadata
  (all classes)           (members)                    (tables, columns)
         ↓                        ↓       ↓                 ↓
         └────────────┬───────────┘       └────────┬────────┘
                      ↓                            ↓
              List<CompletionItem>        List<CompletionItem>
                      └────────────┬───────────────┘
                                   ↓
                         Sort by Priority
                                   ↓
                        Display in Popup

Performance Optimizations

1. ClassGraph Caching

// Scan once, cache forever (until invalidateCache())
ClassGraph classGraph = new ClassGraph()
  .enableClassInfo()
  .enableMethodInfo()
  .scan();

cache.put("classGraph", classGraph);

2. Completion Item Pre-computation

// Convert classes to CompletionItems eagerly
List<CompletionItem> classCompletions =
  scanResult.getAllClasses()
    .stream()
    .map(cls -> CompletionItem.class(cls.getName()))
    .toList();

3. Prefix Filtering

// Filter by prefix on retrieval (not during scan)
String prefix = context.tokenPrefix();
return cachedItems.stream()
  .filter(item -> item.label().startsWith(prefix))
  .toList();

4. Lazy Loading

// Only scan when completion is actually requested
if (!context.isMemberAccess() && !userPressedCtrlSpace) {
  return Collections.emptyList();  // Don't scan yet
}

Target: < 100ms completion time (benchmarked in Task #27)

Extension Points

To add a new language completion:

Implement CompletionEngine:

public class MyLangCompletionEngine implements CompletionEngine {
  @Override
  public List<CompletionItem> complete(CompletionContext context) {
    // Your completion logic
  }

  @Override
  public Set<String> supportedLanguages() {
    return Set.of("mylang");
  }
}

Register on startup:

// In Gade.java or extension loader
CompletionRegistry.getInstance().register(new MyLangCompletionEngine());

Handle cache invalidation:

@Override
public void invalidateCache() {
  myCache.clear();
}

See: Javadoc for CompletionEngine, CompletionItem, CompletionContext

Extension Points

1. Code Completion Engines

Interface: CompletionEngine

Examples:

GroovyCompletionEngine
SqlCompletionEngine
JavaScriptCompletionEngine

Hook: CompletionRegistry.register(engine)

2. Syntax Highlighters

Library: RichTextFX

Current Implementations:

Groovy (keywords, strings, comments, numbers)
SQL (keywords, identifiers, operators)
JavaScript (ES6 keywords)
XML, HTML, Markdown

Extension: Add language-specific regex patterns in CodeTextArea.java

3. Custom Runtimes

Interface: RuntimeConfiguration

Steps:

Implement subprocess main class (like GradleRunnerMain)
Implement JSON-RPC protocol
Add UI for runtime selection
Register with RuntimeManager

4. Database Connectors

Interface: Standard JDBC

Current Support:

PostgreSQL, MySQL, H2, SQL Server, Oracle
BigQuery (custom connector)

Extension: Implement JDBC driver or custom ConnectionHandler

5. File Type Handlers

Pattern: Tab factories in FileOpener.java

Example:

if (file.getName().endsWith(".gmd")) {
  return new MarkdownTab(gui, file);
} else if (file.getName().endsWith(".sql")) {
  return new SqlTab(gui, file);
}

Extension: Add new tab type for your file extension

Build System

Module Structure (JPMS)

module se.alipsa.gade {
  // JavaFX
  requires javafx.controls;
  requires javafx.web;

  // Groovy
  requires org.apache.groovy;
  requires org.apache.groovy.sql;

  // Build tools
  requires org.gradle.tooling;
  requires se.alipsa.maven.utils;

  // UI libraries
  requires org.fxmisc.richtext;
  requires org.fxmisc.flowless;

  // Utilities
  requires io.github.classgraph;
  requires org.eclipse.jgit;
  requires com.fasterxml.jackson.databind;

  // Database
  requires java.sql;
  requires se.alipsa.groovy.datautil;

  // Matrix (data science)
  requires se.alipsa.groovy.matrix;
  requires se.alipsa.groovy.charts;

  // Exports (for extensions)
  exports se.alipsa.gade.code.completion;
  exports se.alipsa.gade.runtime;
}

Platform-Specific Distributions

Tool: org.beryx.runtime plugin (jlink)

Process:

Compile Java → .class files
Create modular JAR
jlink creates minimal JRE (Java 21 + JavaFX)
Bundle application + JRE → platform-specific package

Outputs:

gade-linux.zip (x64)
gade-macos.zip (x64 + ARM64)
gade-windows.zip (x64)

Advantages:

No JDK installation required
Smaller download (JRE includes only used modules)
Consistent Java version across platforms

Data Flow

Script Execution (Gade Runtime)

User clicks "Run"
     ↓
ExecutableTab.executeAction()
     ↓
ConsoleComponent.execute(script)
     ↓
GroovyEngine.eval(script)
     ↓
GroovyShell.evaluate(script)
     ↓
Result returned
     ↓
ConsoleComponent.displayResult()
     ↓
Environment tab updated (variables)

Script Execution (Gradle Runtime)

User clicks "Run"
     ↓
ExecutableTab.executeAction()
     ↓
ConsoleComponent.execute(script)
     ↓
RuntimeProcessRunner.execute(script)
     ↓
Send JSON: {"cmd":"eval","id":"...","script":"..."}
     ↓
[Process boundary - Socket IPC]
     ↓
GradleRunnerMain receives message
     ↓
GroovyShell.evaluate(script) [with Gradle classpath]
     ↓
stdout → Send JSON: {"type":"out","text":"..."}
     ↓
result → Send JSON: {"type":"result","id":"...","result":"..."}
     ↓
[Process boundary - Socket IPC]
     ↓
RuntimeProcessRunner.handleMessage()
     ↓
ConsoleComponent.displayResult()
     ↓
Environment tab updated

Code Completion Flow

User types "text."
     ↓
TextArea KeyTyped event
     ↓
CompletionPopup.show()
     ↓
CompletionRegistry.getCompletions(context)
     ↓
For each registered engine:
  engine.complete(context)
     ↓
Engines return List<CompletionItem>
     ↓
Merge + sort by priority
     ↓
Display top 10 in popup
     ↓
User presses Enter
     ↓
Insert item.insertText()
     ↓
Move cursor by item.cursorOffset()

Gradle Dependency Resolution

User opens project with build.gradle
     ↓
GradleUtils.resolveClasspath(projectDir)
     ↓
Calculate fingerprint (hash of build.gradle)
     ↓
Check cache: ~/.gade/gradle-cache/<fingerprint>.json
     ↓
Cache miss → Use Gradle Tooling API
     ↓
GradleConnector.forProjectDirectory(dir)
     ↓
connection.model(IdeaProject.class).get()
     ↓
Extract dependencies from model
     ↓
Convert to Set<File> (JAR paths)
     ↓
Save to cache (JSON)
     ↓
Return classpath
     ↓
Pass to RuntimeProcessRunner as -cp argument

Security Considerations

Script Execution

Gade Runtime:

⚠️ Runs in same process as UI
⚠️ Full access to Gade internals
⚠️ Can call System.exit(), modify files, etc.

Gradle/Maven Runtime:

✅ Separate process (limited blast radius)
⚠️ Still has full filesystem/network access
⚠️ No sandboxing (Groovy Security Manager deprecated)

Recommendation: Only run trusted scripts. Gade is not designed for untrusted code execution.

JDBC Connections

Passwords stored in plaintext in connection configuration
No encryption at rest
⚠️ Ensure .gade/connections.json has restricted permissions

Git Operations

SSH keys managed by OS (not Gade)
HTTPS credentials via Git credential manager
✅ No password storage in Gade

Performance Characteristics

Startup Time

Measured on 2023 MacBook Pro (M2):

Cold start: ~3-4 seconds
Warm start: ~2 seconds

Bottlenecks:

JavaFX initialization (1-1.5s)
Module loading (0.5s)
UI construction (0.5s)

Code Completion

Target: < 100ms (per roadmap Task #27)

Actual (benchmarked):

Simple string completion: ~42ms
Complex import completion: ~68ms
Member access: ~51ms

See: docs/improvements/task-27-performance-benchmarks.md

Gradle Classpath Resolution

First resolution (no cache):

Small project (~10 deps): 5-10 seconds
Medium project (~50 deps): 15-30 seconds
Large project (~100+ deps): 30-60 seconds

Cached resolution:

< 100ms (read JSON from disk)

Memory Usage

Typical usage:

Minimum: 200MB (empty project)
Normal: 500MB-1GB (with Gradle runtime)
Heavy: 2-4GB (large datasets in matrix)

Recommendation: -Xmx8G for data science workloads

Future Architecture Considerations

Potential Enhancements

Language Server Protocol (LSP)
- Replace custom completion with LSP
- Benefit: Reuse existing language servers
- Challenge: Integration with JavaFX UI
Plugin System
- Dynamic plugin loading
- Separate classloader per plugin
- Extension registry
Distributed Execution
- Remote script execution (SSH)
- Cluster support (Spark, etc.)
Scripting API
- Programmatic access to Gade features
- Automate workflows via scripts
Web-Based UI
- Browser-based alternative to JavaFX
- Better remote access
- Challenge: Requires complete rewrite

References

Document Version: 1.0.0 Last Updated: February 3, 2026 Maintained By: Gade Development Team

Uh oh!

FilesExpand file tree

ARCHITECTURE.md

Latest commit

History

ARCHITECTURE.md

File metadata and controls

Gade Architecture

Table of Contents

System Overview

High-Level Architecture

Design Principles

Component Architecture

1. UI Layer (JavaFX)

Main Window (Gade.java)

Editor Components

Console Components

2. Code Completion System

Core Classes

Engines

3. Runtime System

Runtime Types

RuntimeProcessRunner

4. Gradle Integration

GradleUtils

GradleCacheManager

GradleErrorDetector

5. Git Integration

6. Database Integration

7. Build System

Runtime Execution Models

Gade Runtime (Embedded)

Gradle Runtime (Subprocess)

Maven Runtime (Subprocess)

JSON-RPC Protocol

Message Types

1. Handshake (Runner → Gade)

2. Eval Request (Gade → Runner)

3. Result (Runner → Gade)

4. Error (Runner → Gade)

5. Output (Runner → Gade)

6. Bindings (Runner → Gade)

7. Shutdown (Gade → Runner)

Protocol Flow

Error Handling

Code Completion System

Architecture

Performance Optimizations

Extension Points

Extension Points

1. Code Completion Engines

2. Syntax Highlighters

3. Custom Runtimes

4. Database Connectors

5. File Type Handlers

Build System

Module Structure (JPMS)

Platform-Specific Distributions

Data Flow

Script Execution (Gade Runtime)

Script Execution (Gradle Runtime)

Code Completion Flow

Gradle Dependency Resolution

Security Considerations

Script Execution

JDBC Connections

Git Operations

Performance Characteristics

Startup Time

Code Completion

Gradle Classpath Resolution

Memory Usage

Future Architecture Considerations

Potential Enhancements

References