In Go, does accessing string elements as bytes involve a rune-to-byte conversion?

Accessing Elements of String as Byte: Conversion or Optimization?

When accessing elements of a string in Go, you can write str[i] to obtain individual characters. However, the underlying implementation reveals that str[i] actually returns a byte, not a character (rune). This begs the question: does Go perform a conversion from rune to byte during this operation?

The answer is no. Strings in Go store UTF-8 encoded bytes of the text, so indexing a string simply retrieves its bytes. However, when you use for ... range on a string, it iterates over the runes, not the bytes. This iteration requires an underlying conversion mechanism to extract characters from the bytes.

Performance-wise, iterating over the runes using for ... range is preferred over converting the string to a byte slice ([]byte) and iterating over the bytes. The conversion itself carries no performance penalty as it's optimized away by Go. However, iterating over the runes directly eliminates any potential overhead associated with the conversion.

In summary, accessing elements of a string as bytes does not require a conversion. Iterating over the runes of a string using for ... range is preferred over converting the string to bytes for performance reasons.

The above is the detailed content of In Go, does accessing string elements as bytes involve a rune-to-byte conversion?. For more information, please follow other related articles on the PHP Chinese website!