Unicode Character Lookup

How to use it

1. Ask in whichever notation you have. U+2014 and 0x2014 are read as hexadecimal, a bare 8212 as decimal, and pasting the character works too. Press / to jump to the query field.
2. Read the facts panel: the Unicode name, the code point and its decimal value, the general category, the script, the containing block and plane, and the binary properties your browser reports.
3. Copy whichever spelling your format wants. Each notation has a button — the character, U+2014, 8212, the JavaScript escape, the numeric HTML reference, the UTF-8 bytes and the percent-encoded form.
4. Or browse by block. The directory below lists all 346 blocks with their ranges and assigned counts, and every row carries a stable anchor you can send to someone.

Type U+2014, paste a character, or search by name, and this page answers with that character's Unicode name, its block, script and general category, the plane it sits in, and every escape spelling you might need to paste somewhere else. The data is Unicode 17.0: 346 blocks and 40,470 stored names. Nothing you type is uploaded — the lookup runs on this device.

How it works

Three things get called a character, and separating them is most of what this page does. A code point is the number Unicode assigns, written U+2014. An encoding turns that number into bytes — three in UTF-8 for U+2014, and two UTF-16 units for anything above U+FFFF. A glyph is the shape a font draws, and it differs in every typeface you own.

The code space runs from U+0000 to U+10FFFF and is cut two ways. Planes hold 65,536 code points each, numbered 0 to 16; daily text sits in plane 0, the Basic Multilingual Plane, and emoji in plane 1. Blocks are named ranges inside those planes, 346 in Unicode 17.0, and a block is a filing decision, not a guarantee.

Names come from two places. 40,470 code points carry an explicit name in the Unicode Character Database, shipped here in 21 chunks. The rest are named by rule: CJK and Tangut ideographs take a fixed prefix plus their hexadecimal value, and each of the 11,172 Hangul syllables composes its name from the jamo it decomposes into — U+AC00 is HANGUL SYLLABLE GA. Those cost no download.

General category, script and the binary properties are not stored either. They are read live from your runtime's own Unicode property escapes, the \p{...} classes built into its regular-expression engine, which is how all 137,468 Private Use code points and all 2,048 surrogates get answered while shipping nothing about them.

The name index is the only heavy thing here, 1.16 MB across those 21 chunks, and it downloads only when you search by name — once per visit, not once per query. A single code-point lookup fetches at most one chunk, and a derived name fetches none. That budget is why the block directory is on the page and a 155,000-row character table is not.

Examples

U+2014

EM DASH · U+2014 · 8212 · Pd Dash Punctuation · General Punctuation · Common · BMP · — · \u{2014} · — · E2 80 94

8212

EM DASH · U+2014 · 8212 · Pd Dash Punctuation · General Punctuation · Common · BMP · — · \u{2014} · — · E2 80 94

😀

GRINNING FACE · U+1F600 · 128512 · So Other Symbol · Emoticons · Common · SMP · 😀 · \u{1F600} · 😀 · F0 9F 98 80

arrow

200 shown of 40,470 names searched, more matched · U+02C2 MODIFIER LETTER LEFT ARROWHEAD · U+02C3 MODIFIER LETTER RIGHT ARROWHEAD · U+02C4 MODIFIER LETTER UP ARROWHEAD

U+E000

U+E000 · 57344 · Co Private Use · Private Use Area · BMP · Private Use: no character is assigned here. Any glyph you see comes from a font that put its own artwork at this code point.

U+FDD0

U+FDD0 · 64976 · Cn Unassigned · Arabic Presentation Forms-A · BMP · Noncharacter: permanently reserved for internal use and never interchanged as text.

Every example on this page is executed against the same engine that powers the tool, on every build.

Common errors

U+110000 is past the last Unicode code point, U+10FFFF.

The notation parses but the value is too large. Unicode's code space ends at U+10FFFF because UTF-16 cannot address beyond it, so a bigger number is not a character at all.

U+GG isn't a code point — write it as U+2014 or 0x2014.

A U+ or 0x prefix followed by something that is not hexadecimal. Rather than guess between a typo and a name search, the page quotes the token back and states the rule.

Private Use: no character is assigned here. Any glyph you see comes from a font that put its own artwork at this code point.

Not an error, but the answer most often mistaken for one. Unicode assigns these code points to no character, so the page prints the class, block and plane and invents no name.

Character names couldn't be loaded, so name search is unavailable — code point and pasted-character lookups still work.

Shown when one of the 21 name chunks fails to arrive, usually a dropped connection. It is a notice, not an error page — everything computed from the code point keeps working.

Keyboard shortcuts

/

Focus the lookup box