@turing-machine-js/builder

@turing-machine-js/builder

Status: not actively developed by the author. The package still works and existing tests pass — but no new features are planned. The same state-table construction pattern is shown as an inline example in @turing-machine-js/machine's README, so most users won't need this package as a separate dependency. Contributions are welcome if you'd like to extend it (e.g. multi-tape support, OR-patterns, a string-DSL parser shipped with the package itself).

What it does

Constructs a Turing machine from a declarative state-table object. Every transition is a single (state, currentSymbol) → (nextState, nextSymbol, movement) row — the simplest possible API surface, matching how state machines are typically presented in textbooks.

import { Tape } from '@turing-machine-js/machine';
import buildMachine from '@turing-machine-js/builder';

// Flip every bit on the tape; halt when the head reaches a blank.
const [machine, initialState] = buildMachine({
  alphabetString: ' 01',
  initialState: 'flip',
  finalStateList: ['DONE'],
  states: {
    flip: {
      '0': { state: 'flip', symbol: '1', movement: 'R' },
      '1': { state: 'flip', symbol: '0', movement: 'R' },
      ' ': { state: 'DONE', symbol: ' ', movement: 'S' },
    },
  },
});

machine.tapeBlock.replaceTape(new Tape({
  alphabet: machine.tapeBlock.alphabets[0],
  symbols: '0101'.split(''),
}));

await machine.run({ initialState, stepsLimit: 100 });
console.log(machine.tapeBlock.tapes[0].symbols.join('').trim()); // "1010"

See builder.spec.ts for a longer worked example — a 27-state binary-string-duplicator (input #011# → output #011#011#) — including a small parser that reads the textbook (state,symbol)→(state,symbol,movement); notation.

Limitations

The state-table format is intentionally minimal. It does not support:

• OR-patterns (matching multiple current symbols with one transition row). For tapeBlock.symbol('^10 style patterns, use the raw @turing-machine-js/machine API.
• Multi-tape machines (buildMachine is single-tape only).
• withOverriddenHaltState composition (the subroutine-call mechanism). For composed machines like library-binary-numbers's minusOne, use the raw API.

npm install @turing-machine-js/machine @turing-machine-js/builder

If you need any of the above, the inline state-table example in __INLINE_CODE_12__'s README shows how to write your own __INLINE_CODE_13__-equivalent in ~30 lines, and you can extend it to fit your case.

Install

npm install @turing-machine-js/machine @turing-machine-js/builder

__INLINE_CODE_14__ is a peer dependency (so consumer and library share the same singleton sentinels — __INLINE_CODE_15__, __INLINE_CODE_16__, etc.).

Links

• Turing Machine on Wikipedia
• __INLINE_CODE_17__ — the core engine, sufficient on its own for most use cases