Skip to content
uencoder
TASK

Read HTML entities in pasted text

Turn & < ' and friends back into real characters — and know when NOT to decode.

  1. 1
    HTML Entity Decoder →

    Paste text containing entities (& < > " ' ' and numeric forms). The output replaces each entity with its actual character, including non-ASCII code points.

  2. 2

    Know your entities: named entities cover the XML/HTML specials plus a large catalog (© —  ); numeric forms are either decimal (') or hexadecimal (') and always map to a Unicode code point. If the output shows stray characters where text should be, the source was double-encoded — < decodes to < on the first pass; run it again only after confirming the intended data.

  3. 3

    When NOT to decode: text destined for HTML output should stay encoded — decoding user-supplied content and then rendering it unescaped is the classic XSS injection path. Decode for reading and analysis; keep encoding for storage and display.

Frequently asked questions

What is the difference between ' and '?
Both represent the apostrophe. ' is valid in XML and HTML5 but was not part of HTML 4; ' (numeric) works everywhere, which is why many encoders emit the numeric form.
Why does my text contain < instead of <?
Double encoding: the string was HTML-escaped twice (common when a template escapes already-escaped data). Decode twice to recover the original, then fix the escaping at the source.

Learn more

Want to understand why?

Related tools