Back to the tool 이 페이지를 한국어로 읽기
Where the data comes from
The character lists and names in the symbol catalogue, the emoji catalogue and the Unicode explorer all come from the data files the Unicode Consortium publishes. This page says which file each part came from, and which names are Unicode's and which are ours.
1. The files
- UnicodeData.txt (Unicode 17.0) - the formal English name and the general category of every character. The per-block counts in the catalogue and the names in the explorer are both read from this file. Original: unicode.org/Public/17.0.0/ucd/UnicodeData.txt
- emoji-test.txt (Emoji 17.0, UTS #51) - the emoji list, the groups they fall into, and their official English names. The emoji catalogue is this file. Original: unicode.org/Public/emoji/17.0/emoji-test.txt
Both files are read once, when the site is built. Nothing is fetched from unicode.org - or from anywhere else - while you are using the site.
2. Which names are official
- Symbol names are the formal Unicode names, without exception. Nothing here is a name we made up. The symbols have no Korean names, so searching by name works in English and by U+ code only.
- Emoji English names are Unicode's own.
- Emoji Korean labels are written by this project. They exist for some entries, not all; an emoji without one shows its official English name, and the screen says how many have a label.
- Block names ("Arrows", "Braille patterns") and group names ("Smileys & Emotion") follow Unicode in English; the Korean wording for them is ours.
3. What is left out
Inside a block, the unassigned holes are skipped, and so are the invisible things: controls, format characters, spaces, and combining marks that only exist attached to the character before them. Copying one of those gives you nothing, so they are not offered. That is why a block holds fewer characters than its code point range suggests. Every count on the page is counted from the data file.
For emoji, only the fully-qualified sequences are listed: the same picture written with fewer variation selectors would be a duplicate. Skin tone and hair variants are separate sequences and are kept as separate entries.
4. Licence
The data files are used under Unicode Licence v3. The full text is included here, and the original is at unicode.org/license.txt.
Unicode® is a registered trademark of Unicode, Inc. GlyPop is not affiliated with, or endorsed by, the Unicode Consortium.