docs: fix punycode example and regex delimiter in plan/spec

- xn--80aafmakjldfe1b5h.xn--p1ai decodes to "алквзелчкоеиа.рф", not
  "вашаняня.рф". Correct Punycode for "вашаняня.рф" is
  xn--80aae0ca7d8bb.xn--p1ai. Updated plan Task 2 test and spec §9.
- DomainFormatter strip-path regex '#[/?#].*$#' had the '#' delimiter
  colliding with the literal '#' inside the character class. Switched
  to tilde delimiters: '~[/?#].*$~'.

Both caught by TDD RED/GREEN loop on Task 3.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Vladimir Bryzgalov
2026-04-17 22:04:15 +05:00
parent 6068a6a54f
commit 558483ef44
2 changed files with 3 additions and 3 deletions
@@ -207,7 +207,7 @@ final class DomainFormatterTest extends TestCase
{
self::assertSame(
'вашаняня.рф',
DomainFormatter::humanize('xn--80aafmakjldfe1b5h.xn--p1ai')
DomainFormatter::humanize('xn--80aae0ca7d8bb.xn--p1ai')
);
}
@@ -284,7 +284,7 @@ final class DomainFormatter
$host = $parsed['host'] ?? $host;
}
// Strip any remaining path/query
$host = preg_replace('#[/?#].*$#', '', $host) ?? $host;
$host = preg_replace('~[/?#].*$~', '', $host) ?? $host;
// Strip leading www.
$host = preg_replace('/^www\./i', '', $host) ?? $host;