[mythtv] regression with 7f00642b

Gary Buhrmaster gary.buhrmaster at gmail.com
Fri May 11 15:05:01 UTC 2018


On Fri, May 11, 2018 at 8:04 AM, Jean-Yves Avenard <jyavenard at gmail.com> wrote:
> On 11 May 2018 at 09:53, Jean-Yves Avenard <jyavenard at gmail.com> wrote:
>> reverse that change and pass &gdesc.front() instead
>>
>> (and &pdesc.front())
>
> actually, &*gdesc.begin() may be better.

My recollection (of the standard) is that if the
vector is empty, neither begin() nor front() are
defined and/or can be dereferenced.  And these
are cases where the vector is empty.

But I would certainly defer to someone who
knows the standards better than I.


More information about the mythtv-dev mailing list