Stateless UTF-8 decoder helpers. More...

#include <stdint.h>
#include <stddef.h>

Include dependency graph for font_utf8.h:

Macros
#define	NANO_UTF8_INVALID 0xFFFFFFFFu
	Sentinel value returned by nano_utf8_decode() when the input is malformed, truncated, or contains an invalid sequence.

Functions
static uint32_t	nano_utf8_decode (const char *p, const char end)
	Decode a single UTF-8 codepoint from `p` in a stateless way. More...

Detailed Description

Stateless UTF-8 decoder helpers.

Definition in file font_utf8.h.

Function Documentation

◆ nano_utf8_decode()

static uint32_t nano_utf8_decode	(	const char **	p,
		const char *	end
	)

inlinestatic

Decode a single UTF-8 codepoint from p in a stateless way.

Unlike NanoFont::unicode16FromUtf8(), which keeps internal static state and returns a 16-bit truncated codepoint, this helper decodes a full 21-bit Unicode codepoint and never mutates global state, so it is safe to use from re-entrant contexts.

On entry p must point at the first byte of a UTF-8 sequence. On successful return p is advanced past the decoded sequence and the codepoint is returned. The decoder stops at the NUL byte; if end is non-NULL the decoder never reads past end.

The following inputs are reported as invalid (return value NANO_UTF8_INVALID, p advanced by one byte to allow resynchronisation):

lone continuation bytes (0x80..0xBF) at the start of a sequence;
truncated multi-byte sequences (NUL or end before all trailing bytes have been seen);
bytes 0xC0, 0xC1, 0xF5..0xFF (always invalid in UTF-8);
overlong encodings (e.g. U+0000 encoded as two bytes);
codepoints in the surrogate range U+D800..U+DFFF;
codepoints above U+10FFFF.

Parameters

p	in/out pointer to the next byte to decode. Must not be NULL.
end	optional end-of-buffer guard, may be NULL for NUL-terminated input.

Returns: decoded codepoint, or NANO_UTF8_INVALID.

Definition at line 73 of file font_utf8.h.

Macros

Functions

Detailed Description

Function Documentation

◆ nano_utf8_decode()