System Design: Input Processing and Text Reconstruction

Context

You are designing a cross-platform input processing layer for a text editor that reconstructs exactly what a user typed from raw keyboard and mouse events. The system must support standard keyboards, mice/trackpads, and IME/composition for non-Latin input. Assume a single-user session per editor instance, possibly with multiple input devices (e.g., external keyboard + laptop keyboard + mouse).

Functional Requirements

1. Event capture and handling
   • Keyboard: keydown/keyup, key repeat, modifiers (Shift/Ctrl/Alt/Meta/CapsLock), backspace/delete.
   • Pointer: cursor movement, mouse down/up/move, selection via drag, double/triple click.
   • Clipboard: copy, cut, paste.
   • IME/composition: compositionstart/update/end and pre-edit text.
2. Ordering and time
   • Maintain precise timestamps and a total order across devices (keyboard, mouse).
3. APIs
   • recordEvent(event)
   • getCurrentText()
   • getSelectionRange()
   • undo()
   • redo()
4. Persistence and replay
   • Persist an event log and support deterministic replay to derive final text.
5. Reliability and resilience
   • Debounce where appropriate; handle lost/duplicated events.
   • Define data models for events and sessions.

Deliverables

• High-level components and their responsibilities.
• Core data structures (document, selection, history, logging).
• Recovery strategy after crashes.
• Testing strategy (functional, property-based, fuzzing, IME).
• Performance considerations (latency, throughput, memory).