Why does a fullwidth space look empty on a webpage but still count as a character?

Question

Weirdo · Accepted Answer

It is not ASCII space (U+0020). It is U+3000 IDEOGRAPHIC SPACE. In CJK layout it is as wide as a Han character, so it looks like a gap; string length, JSON, and URLs still count it as one character.

The Unicode names list calls U+3000 IDEOGRAPHIC SPACE and notes it is compatible with U+0020 SPACE. Chapter 6 of the Standard is direct: in ideographic text, U+3000 is commonly used because its width matches the ideographs. EastAsianWidth.txt marks it F (Fullwidth). The page is showing a Han-width blank, not “no character”.

String length, JSON strings, and database columns treat U+3000 as one character. In a URL it encodes as %E3%80%80, not the ordinary space’s %20. Comparisons, search, and validation that only strip ASCII 0x20 will miss it. To drop it you have to handle U+3000; trimming halfwidth space is not enough.

Why does a fullwidth space look empty on a webpage but still count as a character?

A real code point that paints as a blank cell

Software counts code points, not how empty it looks