On Thu, Nov 23, 2017 at 7:58 PM, Thomas Koenig <tkoe...@netcologne.de> wrote:
> Hi Janne,
>
>> However, to continue my nitpicking (sorry!), it seems that in many
>> cases compare_fcn still takes an integer length argument. Could you
>> make that gfc_charlen_type as well? Or maybe size_t, since the
>> argument is passed straight to memcmp{_char4} anyway? Please consider
>> such a patch pre-approved. Thanks!
>
>
> Committed as r255109.
>
> I had missed out on the non-inlined maxval and maxloc versions...You still missed two places. Committed r255135 with the fixes. > The fun with max* and min* intrinsics is not yet over. Maxval and > Minval have yet to be implemented for character arguments, and then > there is the BACK argument to MAXLOC. > > Maybe (while we are breaking compatibility) we should just add BACK > to the front end, reject it whith a "not yet implemented" message, > add the argument to the library and worry about implementation > later. > > What do you think? Yeah, makes sense I suppose. -- Janne Blomqvist
