Unicode Dataset Generator

Runs in browser Test Data

Datasets that exercise Unicode handling: surrogate pairs, combining sequences, NFC/NFD variants, bidi controls, homoglyphs and mixed-script strings.

TEST DATA ONLY — all generated names, contacts, identifiers and numbers are synthetic and do not belong to real people or accounts.
Privacy: This tool runs entirely in your browser. Your input never leaves your device.

What are Unicode and UTF-8?

Unicode assigns a number (a code point, written U+0627 for Arabic alef) to every character of every script; UTF-8 is the encoding that turns those numbers into one to four bytes and is the default on the web and in JSON. Most "garbled text" bugs are a UTF-8 byte sequence being read as another encoding, or a string counted in bytes where the program expected characters.

How to use Unicode Dataset Generator

  1. List the categories (or "all").
  2. Add random mixed strings if you want more volume.
  3. Press Generate and compare the length columns with what your system stores.

Unicode Dataset Generator features

  • 18 categories: Latin diacritics, Arabic, Devanagari, CJK, Cyrillic, Greek, Hebrew, Thai, emoji, combining marks, NFC/NFD pairs, surrogate pairs, bidi controls, zero-width, homoglyphs, mixed scripts, case mapping and fullwidth forms
  • Per row: escaped form, code points, UTF-16 units, UTF-8 bytes, graphemes and NFC = NFD flag
  • Extra random strings combining samples from the chosen categories
  • Counts of multi-unit graphemes in the summary
  • Seedable; JSON, CSV, SQL, XML, JSON Lines or table output

Unicode Dataset Generator example

Family emoji

Input:

Categories: emoji

Output:

text,codePoints,utf16Units,utf8Bytes,graphemes
👨‍👩‍👧‍👦,7,11,25,1

Frequently asked questions about Unicode Dataset Generator

What does each row tell me?

The text, its escaped form, the number of code points, UTF-16 units, UTF-8 bytes and user-perceived graphemes, and whether NFC and NFD forms are identical — everything you need to spot length and normalisation bugs.

Which scripts and cases are covered?

Latin with diacritics, Arabic, Devanagari, CJK, Cyrillic, Greek, Hebrew, Thai, emoji and ZWJ sequences, combining marks, NFC/NFD pairs, surrogate pairs, bidi controls, zero-width characters, homoglyphs, mixed scripts, case-mapping edge cases and fullwidth forms.

Can I generate random mixed strings too?

Yes — set how many extra strings to build from random samples of the selected categories, and how many samples each should combine.

References