RFR 8151788: NullPointerException from ntlm.Client.type3

[Wang Weijun]

Change looks fine to me.

Thanks
Max

> 在 2016年7月13日，14:52，Vyom Tewari <vyom.tewari at oracle.com> 写道：
> 
> Hi All,
> 
> Please find the updated webrev(http://cr.openjdk.java.net/%7Evtewari/8151788/webrev0.1/index.html). I addressed the review comments.
> 
> Thanks,
> Vyom
> 
> 
>> On Tuesday 12 July 2016 08:25 PM, Weijun Wang wrote:
>> 
>> 
>>> On 7/12/2016 22:34, Pavel Rappo wrote:
>>> What's the difference between no security buffer and an empty one (from the
>>> com.sun.security.ntlm.Client#type3's perspective)?
>> 
>> I quickly browse through the NTLM protocol and yes they look like the same in each case. (Except for one which I am not sure, is there any difference between no domain and empty domain?) In all cases where a security buffer is optional, there is a flag we can rely on, and no need to look at whether the offset of the security buffer is zero.
>> 
>> So it does look safer to return a new byte[0] right inside readSecurityBuffer(int offset) when the offset is zero.
>> 
>> Thanks
>> Max
>> 
>>> 
>>>> On 12 Jul 2016, at 15:25, Wang Weijun <weijun.wang at oracle.com> wrote:
>>>> 
>>>> When there is no offset, there is no security buffer at all. When the length is zero, the security buffer is an empty byte array.
>