How to search for Unicode code points to check for garbled characters and normalization

When this guide is useful

This is for users who want to check the Unicode code table, Unicode character search, code point search, UTF-8/UTF-16 conversion, garbled character investigation, and NFC/NFD normalization in a browser.

• Check code points for emoji and alternate glyph selectors
• Find out why accented characters aren't matching in search
• Identify zero-width and control characters in logs
• Create Unicode escape notation to paste into the program

Information that can be checked in the Unicode code table

• Unicode 17.0.0 code points, official names, aliases, blocks, Script
• General Category, Additional Versions, East Asian Width, Combined Class
• Encoding notation of UTF-8/UTF-16/UTF-32
• Escape notation for JavaScript, Java, C#, Python, CSS, and HTML
• Normalization results and number of grapheme clusters for NFC, NFD, NFKC, and NFKD

Steps to search and parse Unicode characters

1. Input
2. Check Unicode names, categories, blocks, Script, and UTF notation.
3. Additional information for this feature.
4. Compare the results of NFC, NFD, NFKC, and NFKD and select the format required by the user.

Notes on glyph shapes, invisible characters, and normalization

Even if a character is registered in Unicode, if the terminal's font does not support it, it will be displayed as a rectangle. Control characters, zero-width characters, and combining characters may not be visible by themselves.

NFKC/NFKD replaces compatible characters, so do not unconditionally apply them to identifiers or data to be signed that need to retain their original notation.

Specific example: Find out why the same “é” does not match

"é" is displayed the same way at one code point, U+00E9, or at two code points, U+0065 and U+0301. When a string comparison compares code point columns, the two do not match out of the box.

In NFC, it is decomposed into synthesized characters, and in NFD, it is decomposed into base characters and combining symbols, so both must be aligned to the same normalized form before comparison.

1. Enter two types of "é" into the text analysis field.
2. Check the code point sequence and number of grapheme clusters for each character.
3. Compare the results of NFC and NFD.
4. Match the normalization rules of the system that performs searches, saves, and comparisons.

For user IDs, file names, digital signatures, etc., the presence or absence of normalization affects specifications and security. Please check compatibility with existing data.

Frequently asked questions

Are UTF-8 and Unicode code points the same?

It's different. A code point is a number assigned to a character, and UTF-8 is a method for encoding that number into 1 to 4 bytes.

Why are emojis displayed as squares?

The browser or OS font may not have the glyph shape for that character. You can check the code point information, but the displayed glyph depends on the terminal environment.