Convert HTML Entities to UTF8
Decode HTML entities back into UTF-8 text, named and numeric, with unknown names reported.
What this page computes
Paste markup and the page turns every HTML entity back into the character it names: the named forms in its table, such as é, & and <, and every numeric form, decimal as é or hexadecimal as é. A stray & with no semicolon is left alone, because ordinary prose is allowed to contain one and refusing it would make the page useless on a normal sentence.
Where it stops
A numeric reference above U+10FFFF, or one that names a surrogate, is refused instead of being turned into a replacement character. An unknown named entity such as ⊂ is left exactly as written and named in the second line, because the table is finite and silently deleting text would lose data; the count tells you it happened, so the output is not quietly wrong. Nested or double-escaped entities are decoded once, not repeatedly.
The scan runs in this tab over the string you pasted. Nothing is uploaded, nothing is stored, and the same markup always decodes to the same text.
`<p>Café & crème</p>` decodes to `<p>Café & crème</p>`: the named forms in the table and every numeric form, decimal or hexadecimal, are turned back into characters.
An unknown name is left exactly as written and reported, while a malformed numeric reference is refused rather than replaced.
How to use it
Paste the markup and press Run. Each `&…;` is looked up once; a stray `&` without a semicolon is left alone, since ordinary prose contains one.
The table
The decoder knows the same named table the encoder writes, plus `'`, and accepts every numeric form: `é` and `é` both become é. A reference is only read when a semicolon follows within a short distance, so `AT&T` is untouched.
Where it stops
A malformed numeric reference is refused with the reference quoted, and an unknown name is passed through and counted, so the output never quietly loses text. The page decodes one level only: a double-escaped `&lt;` becomes `<`, not `<`.
What to Expect
Best for
- `<p>Café & crème</p>` decodes to `<p>Café & crème</p>`: the named forms in the table and every numeric form, decimal or hexadecimal, are turned back into characters.
- Seeing the result before copying it.
What this tool keeps
- Your input is converted in the browser and never posted, so the page works offline and retains nothing.
Sample input
<p>Café & crème</p>
Sample output
<p>Café & crème</p> 7 entities decoded → 19 characters
Frequently asked questions
How do I use it?
Paste the markup and press Run. The decoded text is printed first, with a count of entities decoded and characters produced on the second line.
Is anything uploaded?
No. The text is converted in this tab with JavaScript: no request is made, no account is needed, nothing is stored, and the page keeps working offline once loaded.
What are its limits?
A numeric reference above U+10FFFF or one that names a surrogate is refused, because no character exists for it. An unknown named entity must be left as written and named in the report rather than deleted.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.