InputRelay

InputRelay is a Linux utility that bridges physical input devices to a virtual keyboard. It intercepts events from hardware controllers (via evdev), maps them to macros or keystrokes, and injects them back into the system using the uinput kernel module.

This tool is useful for:

• Mapping game controller buttons to keyboard shortcuts.
• Creating macro sequences (Key press -> Wait -> Key press).
• Using controllers in applications that only support keyboard input.

Architecture

InputRelay migrates away from libusb (used in previous versions) to the standard Linux Input Subsystem.

graph LR

    subgraph Input Layer
        A[Physical Controller]
        B[InputRelay Scanner]
    end

    subgraph Processing
        C[Mapper Engine]
        D[Macro Processor]
    end

    subgraph Output Layer
        E[Virtual Keyboard]
        F[OS / Application]
    end

    A -->|evdev| B
    B --> C
    C -->|config.ini| D
    D -->|uinput| E
    E --> F

Loading

Prerequisites

• Linux Kernel: Requires uinput support enabled
• Libraries:
  • ncurses (for the TUI)
• Permissions: Root access (sudo) is required to grab exclusive access to input devices

Building

To build the project run:

make

This will compile the source files located in src/ and generate the inputrelay binary.

To clean build artifacts:

make clean

Usage

InputRelay must be run as root to access hardware devices

1. Basic Execution

   To start the relay, provide the name (or partial name) of the input device you wish to capture.

   sudo ./inputrelay "xbox"

   • The program will scan /dev/input/ for a device matching the string "xbox"
   • It grabs the device exclusively (the OS will stop seeing the original controller)
   • Input events are translated according to config.ini config.ini

2. Daemon Mode

   To tun the relay in the background:

   sudo ./inputrelay --daemon

3. Configure Interface (TUI)

   To launch the ncurses-based configurator:

   sudo ./inputrelay --configure "sony"

   • Note: The configurator currently only supports capturing inputs and generating basic configuration entries. Manual editing of config.ini is recommended for complex macros.

Configuration

Mappings are defined in config.ini in the root directory

Syntax:

[MAPPINGS]
TRIGGER_BUTTON = ACTION_1, ACTION_2, ...

• Triggers are the standard Linux event names (e.g., BTN_SOUTH)
• ACtions:
  • Keys: KEY_A, KEY_ENTER, KEY_SPACE
  • Delays: WAIT_ms (e.g., WAIT_100 for 100ms of delay)

Example:

[MAPPINGS]
# Remap 'A' button (South) to type "HI" and press Enter
BTN_SOUTH = KEY_H, WAIT_50, KEY_I, WAIT_50, KEY_ENTER

# Remap 'B' button (East) to Spacebar
BTN_EAST = KEY_SPACE

Project Structure

• src/main.c: Entry point and main event loop
• src/input/: Handles scanning /dev/input/event*` and device grabbing
• src/mapping/: Config parsing and macro translation logic.
• src/output/: Virtual device creation via ioctl on /dev/uinput.
• src/ui/: Ncurses interface implementation.