Validate the H5P title as Unicode characters, not bytes

[thisismyurl]

Hey folks,

A title made of non-Latin characters can be rejected even when it's well under the length limit. The validation pattern /^.{1,255}$/ counts bytes, not characters, so a multibyte title (Japanese or Arabic, say) hits the 255 cap at around 85 characters. Adding the u modifier makes the quantifier count Unicode characters, which is what the 255 limit is meant to mean. This matches the linked Moodle report (MDL-87358).

I applied it to both title patterns: the content title in $h5pRequired that the report points at, and the library title in $libraryRequired a few lines down, which carries the identical byte-counting pattern. One thing worth calling out: with /u, preg_match also rejects a title containing an invalid UTF-8 byte sequence, which the byte pattern silently accepted before. For JSON-sourced metadata that's the safer behaviour, but it is a change in behaviour.

There's no unit-test harness in the library, so here is the standalone check I used, which anyone can run in under a minute:

$title = str_repeat("\xC3\xA9", 200); // 200 e-acute characters = 400 bytes
var_dump(preg_match('/^.{1,255}$/',  $title)); // 0 - rejected on byte count
var_dump(preg_match('/^.{1,255}$/u', $title)); // 1 - accepted on character count

Closes #276

(full disclosure: AI helped me identify the issue and verify my work)

[Copilot]

Pull request overview

This PR updates H5P package and library metadata validation so the title length constraint is enforced in Unicode characters (not bytes), preventing multibyte titles (e.g., Japanese/Arabic) from being incorrectly rejected under the 255 limit.

Changes:

• Add the PCRE u modifier to the title validation regex in the h5p.json schema ($h5pRequired).
• Add the PCRE u modifier to the title validation regex in the library.json schema ($libraryRequired).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[thisismyurl]

Good catch — addressed in the follow-up commit (7672ced).

Changed === 0 to !== 1 in isValidRequirement(), consistent with the filter_xss() pattern already in this file at line 4911. This ensures that both a non-match (0) and a regex error on invalid UTF-8 (false) are treated as invalid — so a title with malformed byte sequences will now correctly fail validation rather than passing silently.