Compiler warnings

[florianhartig]

Not sure about you guys, but I get a bunch of compiler warnings such as

Warning: ‘countthinning’ may be used uninitialized in this function [-Wmaybe-uninitialized]
threePGN.f90:147:0:

     real(kind=8), allocatable, dimension(:,:) :: thinning

This is often not a problem, but maybe worth having a look at those, and ideally get rid of the warnings so that we don't miss important warnings during the compilation

[trotsiuk]

Those warnings are from the beginning and are mainly due to some undeclared or unused variables. Code need to be cleaned. We did part of it with special characters, etc.

[trotsiuk]

Moved to #17

[florianhartig]

@trotsiuk - this is not the same problem as the CRAN checks, is it?

[trotsiuk]

@florianhartig There still many issues with checks, and I feel it is too early to early to put main attention to the checks as the main code Is not ready. But you have more experience with building the packages, so we follow your advice.

[florianhartig]

But these compiler warnings are something completely different than the checks - these are simply unclean coding things that should be solved at some point

[florianhartig]

The CRAN checks don't flag the compiler warnings as warnings - I just see this in the compiler messages

[trotsiuk]

I think it will not pass the Trevis anyway with the compile warnings, or will it?