From: Peter Krefting <peter@softwolves.pp.se>
Date: Mon, 01 Jul 2013 07:14:02 GMT
Subject: Re: [PATCH 2/2] commit: reject overlong UTF-8 sequences
Message-ID: <alpine.DEB.2.00.1307010812470.21709@ds9.cixit.se>
In-Reply-To: <20130629174140.GC226907@vauxhall.crustytoothpaste.net>

brian m. carlson:

> 	int offset = 0;
> +	static const unsigned int max_codepoint[] = {
> +		0x7f, 0x7ff, 0xffff, 0x1fffff
> +	};

Since Unicode is not defined beyond U+10FFFF, you can easily make the 
last range end at U+10FFF. Doing that, ...

> 		if (codepoint > 0x10ffff)
> 			return bad_offset;

... you can also drop this test, as it will have been tested by the 
max test already.

-- 
\\// Peter - http://www.softwolves.pp.se/