If you read an UTF-8 string from somewhere, you should validate it, or risk random segmentation faults

from_unichar 0xiii converts a code point iii (usually in hexadecimal form) into a string containing the UTF-8 encoded character 0xiii. See unicode.org for charmaps. Does not check that the given code point is a valid unicode point.

to_unichar_validated decodes an UTF-8 encoded code point and checks for incomplete characters, invalid characters and overlong encodings.

• raises Convert.Error

  if invalid

to_unichar decodes an UTF-8 encoded code point. Result is undefined if pos does not point to a valid UTF-8 encoded character.

to_unistring decodes an UTF-8 encoded string into an array of unichar. The string must be valid.