Why Do You See “锟斤拷”? Unraveling Unicode, UTF‑8 and GBK Encoding Mysteries

What is the mysterious character?

In computing, every visible symbol is represented by a binary code. The simplest example is ASCII, where the binary 0100 0001 (decimal 65) maps to the letter A. Unicode also defines a special placeholder character � (code point 0xFFFD, decimal 65533) that is used when a decoder encounters an unknown or unrepresentable byte sequence.

When a program cannot map a byte sequence to a known character, it substitutes this placeholder, which often shows up as a garbled box or the string “�”.

Why does “锟斤拷” appear?

Consider the byte array new byte[] {-25, -119, -25, -116}. In UTF‑8 this sequence does not correspond to any valid character, so the decoder replaces it with the Unicode placeholder �. The placeholder itself is encoded in UTF‑8 as the three‑byte sequence 0xEFBFBD (decimal [-17, -65, -67]).

When the same three‑byte sequence is interpreted using the GBK encoding, which is a double‑byte scheme, it is split into three two‑byte units: 0xEFBF, 0xBDEF, and 0xBFBD. In GBK these correspond to the Chinese characters 锟, 斤, and 拷 respectively. Thus the placeholder that should appear as “�” is rendered as the visible string “锟斤拷”.

The phenomenon is common when text encoded in one charset (e.g., UTF‑8) is mistakenly read as another (e.g., GBK) without proper conversion, leading to the infamous “锟斤拷” garble.

Understanding the mapping between binary representations, Unicode code points, and specific legacy encodings like GBK helps prevent such mis‑display issues.