import { describe, expect, it } from 'vitest';
import { countChars, detectEncoding, segment } from './sms-segmenter';

describe('detectEncoding', () => {
  it('detects GSM-7 for basic Latin text', () => {
    expect(detectEncoding('Hello world!')).toBe('GSM-7');
  });

  it('detects GSM-7 for accented chars in the basic set', () => {
    expect(detectEncoding('Café à €')).toBe('GSM-7');
  });

  it('falls back to UCS-2 for emoji', () => {
    expect(detectEncoding('Hello 🎉')).toBe('UCS-2');
  });

  it('falls back to UCS-2 for non-GSM scripts', () => {
    expect(detectEncoding('こんにちは')).toBe('UCS-2');
  });

  it('treats empty string as GSM-7', () => {
    expect(detectEncoding('')).toBe('GSM-7');
  });
});

describe('countChars', () => {
  it('counts basic GSM-7 chars as one each', () => {
    expect(countChars('abc', 'GSM-7')).toBe(3);
  });

  it('counts GSM-7 extension chars as two each', () => {
    // '€' and '{' and '}' are extension-table characters.
    expect(countChars('€', 'GSM-7')).toBe(2);
    expect(countChars('{}', 'GSM-7')).toBe(4);
  });

  it('counts UCS-2 in UTF-16 code units', () => {
    expect(countChars('abc', 'UCS-2')).toBe(3);
    expect(countChars('🎉', 'UCS-2')).toBe(2); // surrogate pair
  });
});

describe('segment', () => {
  it('reports zero segments for empty text', () => {
    const r = segment('');
    expect(r.segments).toBe(0);
    expect(r.encoding).toBe('GSM-7');
    expect(r.remaining).toBe(160);
  });

  it('reports one GSM-7 segment under 160 chars', () => {
    const r = segment('a'.repeat(160));
    expect(r.encoding).toBe('GSM-7');
    expect(r.segments).toBe(1);
    expect(r.length).toBe(160);
    expect(r.remaining).toBe(0);
  });

  it('splits GSM-7 into two segments at 161 chars (153 each)', () => {
    const r = segment('a'.repeat(161));
    expect(r.segments).toBe(2);
    expect(r.maxMulti).toBe(153);
    expect(r.remaining).toBe(2 * 153 - 161);
  });

  it('reports one UCS-2 segment under 70 chars', () => {
    const r = segment('café 😀'); // contains emoji → UCS-2
    expect(r.encoding).toBe('UCS-2');
    expect(r.segments).toBe(1);
    expect(r.maxSingle).toBe(70);
  });

  it('splits UCS-2 at 71 chars (67 each)', () => {
    const r = segment('ñ'.repeat(71).concat('🎉')); // force UCS-2, > 70 units
    expect(r.encoding).toBe('UCS-2');
    expect(r.maxMulti).toBe(67);
    expect(r.segments).toBeGreaterThanOrEqual(2);
  });

  it('counts an extension char pair toward the segment length', () => {
    const r = segment('€'.repeat(80)); // 80 * 2 = 160 septets
    expect(r.encoding).toBe('GSM-7');
    expect(r.length).toBe(160);
    expect(r.segments).toBe(1);
  });
});