php-ml/tests/Tokenization/WordTokenizerTest.php
2018-10-28 07:44:52 +01:00

42 lines
1.6 KiB
PHP

<?php
declare(strict_types=1);
namespace Phpml\Tests\Tokenization;
use Phpml\Tokenization\WordTokenizer;
use PHPUnit\Framework\TestCase;
class WordTokenizerTest extends TestCase
{
public function testTokenizationOnAscii(): void
{
$tokenizer = new WordTokenizer();
$text = 'Lorem ipsum-dolor sit amet, consectetur/adipiscing elit.
Cras consectetur, dui et lobortis;auctor.
Nulla vitae ,.,/ congue lorem.';
$tokens = ['Lorem', 'ipsum', 'dolor', 'sit', 'amet', 'consectetur', 'adipiscing', 'elit',
'Cras', 'consectetur', 'dui', 'et', 'lobortis', 'auctor',
'Nulla', 'vitae', 'congue', 'lorem', ];
self::assertEquals($tokens, $tokenizer->tokenize($text));
}
public function testTokenizationOnUtf8(): void
{
$tokenizer = new WordTokenizer();
$text = '鋍鞎 鳼 鞮鞢騉 袟袘觕, 炟砏 蒮 謺貙蹖 偢偣唲 蒛 箷箯緷 鑴鱱爧 覮轀,
剆坲 煘煓瑐 鬐鶤鶐 飹勫嫢 銪 餀 枲柊氠 鍎鞚韕 焲犈,
殍涾烰 齞齝囃 蹅輶 鄜, 孻憵 擙樲橚 藒襓謥 岯岪弨 蒮 廞徲 孻憵懥 趡趛踠 槏';
$tokens = ['鋍鞎', '鞮鞢騉', '袟袘觕', '炟砏', '謺貙蹖', '偢偣唲', '箷箯緷', '鑴鱱爧', '覮轀',
'剆坲', '煘煓瑐', '鬐鶤鶐', '飹勫嫢', '枲柊氠', '鍎鞚韕', '焲犈',
'殍涾烰', '齞齝囃', '蹅輶', '孻憵', '擙樲橚', '藒襓謥', '岯岪弨', '廞徲', '孻憵懥', '趡趛踠', ];
self::assertEquals($tokens, $tokenizer->tokenize($text));
}
}