oneocr-xhtml_controls_js

The viewer controls for the semantic XHTML output, written in Java and compiled to JavaScript by TeaVM.

Website · All repositories · Live output

---

Overview

This Maven project produces XHtmlOcrControls.js - the interactive JavaScript that powers OCR document visualization features like:

• Background image toggle
• Line and word bounding boxes
• Confidence-based color coding
• Layout mode switching
• Keyboard shortcuts (Ctrl+1-5)

Architecture

• Input: Java source code with TeaVM annotations
• Build Tool: Maven with TeaVM plugin
• Output: target/generated/js/XHtmlOcrControls.js
• Deployment Target: GitHub Pages CDN at https://oneocr.github.io/semantic_xhtml/

Build Requirements

• JDK: 11 or higher
• Maven: 3.6+
• TeaVM: Configured via Maven plugin
• IDE: NetBeans (recommended)

Building

mvn clean compile

The generated JavaScript will be available at:

target/generated/js/XHtmlOcrControls.js
target/generated/js/XHtmlOcrControls.js.map

Development Workflow

1. Write Java Code: Implement interactive features using TeaVM-compatible Java in NetBeans
2. Build: Run Maven to generate JavaScript
3. Deploy: Copy generated XHtmlOcrControls.js to the web repository for GitHub Pages hosting
4. Test: Verify functionality with XHTML OCR documents

Pure Java Approach

This project maintains our commitment to pure Java development:

• No Native Code: FFM eliminated the need for C/C++ bindings
• No Manual JavaScript: TeaVM transpiles Java to browser-compatible JavaScript
• No HTML Templating: XHTML documents generated using Java DSL
• Unified Skillset: Pure Java team can contribute across the entire stack

Why Java to JavaScript?

• Type Safety: Catch errors at compile time rather than runtime
• IDE Support: Full NetBeans support with code completion and refactoring
• Team Consistency: Pure Java team maintains productivity without context switching
• Maintainability: Structured OOP approach for complex interactive features

Integration

The generated JavaScript is served via GitHub Pages CDN and dynamically loaded by OCR documents, ensuring:

• Backward Compatibility: Existing OCR documents continue to work
• Independent Updates: Controls can be upgraded without touching OCR documents
• Clean Separation: Document generation and presentation remain decoupled
• Stable CDN URL: Consistent access via GitHub Pages hosting

Related Projects

• Web Repository: semantic_xhtml - GitHub Pages hosting for generated JavaScript
• Document Generator: oneocr-cli - Creates XHTML documents
• Core API: oneocr-api - Java FFM bindings for Windows OCR

---

Part of the OneOCR project (https://github.com/oneocr) - leveraging Windows 11 built-in OCR for Java applications.