[PATCH] [WIP] fdo#45747 - [EasyHack] remove the limitation to 3 sort entries in calc

[Michael Stahl]

On 24/03/12 01:26, Albert Thuswaldner wrote:
> Hi Markus,
> Thanks for your review.
> 
> On Thu, Mar 22, 2012 at 01:14, Markus Mohrhard
> <markus.mohrhard at googlemail.com> wrote:
>>
>> Ok. So as promised I reviewed the patch now. Sadly there are some more
>> problems than I thought after my first quick look:

also, this thing crashes in sc_unoapi with a STL assertion, which i've
just pushed an awful workaround for that is likely wrong and incomplete;

>> - Can you recheck ScSortParam constructos that already fill
>> maKeyState? std::vector[] does not increase the vector size so you are
>> only allowed to access existing entries. I think we should use
>> push_back there.

my fix actually tried to do this for one of them, but i've noticed that
in table3.cxx there is code that hard-codes sorting related arrays to
size 3, which crashes when you add back the code that i've commented
out; at that point i thought this really needs a calc developer to look
at :)

>> - personally I prefer static_cast in c++ over the old c-cast but I
>> have no strong opinion there

c-style casts are evil.