Re: [Quantlib-dev] FW: SABR regression

[Peter C. <pca...@gm...>]

fine for me. Would we need something like

QL_REQUIRE(seed == (sizeof(seed) == 4 ? 12345 : 6789), "...")

then ? Are 4 and 8 bytes then only possible values on the myriads of
platforms that are supported ?

On 19 November 2014 17:13, Luigi Ballabio <lui...@gm...> wrote:
> Nope, the first arg to boost::hash_combine must be a size_t.
> At this point I'd go for choosing the expected result based on the size of
> size_t. What do you think?
>
>
> On Wed, Nov 19, 2014 at 10:44 AM, Luigi Ballabio <lui...@gm...>
> wrote:
>>
>> It looks like we crossed answers :)
>>
>> I'll check what happens using long instead of size_t.