[SatriaBagusNurhadianto]

What changed:

• Moved the named-entity table into src/lib/html-entities.ts and expanded it to cover HTML4 named entities.
• Updated decodeEntities regex to match entity names containing digits ([a-z][a-z0-9]*), so ½ and ² decode.
• Kept Object.hasOwn lookup for case-sensitive, prototype-safe decoding.

Why:

• Previously only ten named entities were known, so typographic and accented entities reached the preview checker as raw text.

Tests:

• npm run test (meta-fetcher suite, 54 tests) passes.
• npm run build succeeds.

Related: #396

[bhushan4work]

@SatriaBagusNurhadianto please read the contributing guidelines before opening a pr , u should get urself assigned first before writing a fix, dont spam prs. comment on the issue and wait to get assigned first , closing this u may reopen once u get assigned .

[anurag629]

Thanks @SatriaBagusNurhadianto. #396 wasn't assigned to you when you opened this, so per the Spam section of CONTRIBUTING.md this one gets the spam label. That counts as a strike, your first one. #396 is still free, so if you want it, comment there and I'll assign it. Keep in mind it asks for the 252 HTML 4 names rather than HTML5's full list of over 2,000 (this table has 2,125), plus a test that fails on the current code.