[MPlayer-dev-eng] [PATCH] fix SAMI parsing

[Howard Chu]

Howard Chu wrote:
> Reimar Döffinger wrote:
>> On Sun, Jun 06, 2010 at 07:24:46PM -0400, hyc at highlandsun.com wrote:
>>> On Mon, Jun 07, 2010 at 12:26:33AM +0200, Reimar D?ffinger wrote:
>>>> On Sun, Jun 06, 2010 at 02:51:29PM -0700, Howard Chu wrote:
>>>>
>>>>> +		uint32_t c = strtol(s+2,&s, 0);
>>>>> +		uint8_t tmp;
>>>>> +		PUT_UTF8(c, tmp, *p++ = tmp;)
>>>>> +		if (*s == ';') s++; }
>>>>
>>>> I'm sorry, but I think you'll have to somehow "prove"
>>>> this is not a security issue.
>>>> Checking and documenting that PUT_UTF8 will never write
>>>> more than we read might be possible.
>>>> Or just "blindly" checking we still have at least 8 bytes
>>>> free should do as well.
>>>
>>> Not necessary. Decimal numbers encode only 3.25 bits per byte, while UTF-8 encodes 7 bits per byte. This conversion will always fit.
>>
>> Given that 3.25 bits just don't exist, it obviously is not that easy.
>> And you forgot another case as well: negative numbers.
>> This kind of thing just can't be done hand-wavy like that, or you'll
>> always miss a case.

> Of course, we could just check for a '-' in the string and drop this entity.
> Nobody should be using negative character codes.

Did that. Also had to check for " " in addition to " ".

-- 
   -- Howard Chu
   CTO, Symas Corp.           http://www.symas.com
   Director, Highland Sun     http://highlandsun.com/hyc/
   Chief Architect, OpenLDAP  http://www.openldap.org/project/
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: sub.txt
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20100607/281aafcb/attachment.txt>