[FLASH-BUGS] new python incompatibility

[Alvaro Caceres]

Hi Sean,

Thanks for pointing this out, it's indeed due to changes in the new
Python. Here's a setup.py (based on the FLASH2.2 release) which seems to
work properly (the changes are backwards-compatible).

http://flash.uchicago.edu/~caceres/setup.py

Alvaro 

Sean Matt wrote:
> 
> Hello again,
> 
>         We've just noticed a possible compatibility problem with the
> latest version of Python.  When we setup the briowu problem on our alpha
> machines (running OSF1) using
> 
> ./setup briowu -auto -maxblocks=1000
> 
> and using Python version 2.3a2, we get the following lines in the
> "common-defines.fh" file:
> 
> #define COMMON_VAR_ADVECT   True, True, True, True, False, True, False,
> False, False, True, True, True, True
> #define COMMON_VAR_RENORM   False, False, False, False, False, False,
> False, False, False, False, False, False, False
> #define COMMON_VAR_CONSERVE True, True, True, False, False, True, False,
> False, False, False, False, False, False
> 
> If we use the slightly older Python version 2.2.1, the same lines read:
> 
> #define COMMON_VAR_ADVECT   1, 1, 1, 1, 0, 1, 0, 0, 0, 1, 1, 1, 1
> #define COMMON_VAR_RENORM   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
> #define COMMON_VAR_CONSERVE 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0
> 
> The former (with "True, False...") is a problem for our Compaq Fortran
> compiler (V5.5-1877), and gives a compile time error that looks like this:
> 
> f95 -cpp  -c -fast  -r8 -i4 -I/usr/mpi/include  -DN_DIM=2 -DMAXBLOCKS=1000
> -DNXB=8 -DNYB=8 -DNZB=1  dBaseDeclarations.F90
> f90: Error: dBaseDeclarations.F90, line 126: This name does not have a
> type, and must have an explicit type.   [TRUE]
>     / True, True, True, True, False, True, False, False, False, True,
> True, True, True /
> ------^
> f90: Error: dBaseDeclarations.F90, line 126: A symbol must be a defined
> parameter in this context.   [TRUE]
>     / True, True, True, True, False, True, False, False, False, True,
> True, True, True /
> ------^
> 
> etc... etc...
> 
> We are able to compile without problems if we simply edit the
> common-defines.fh file (replacing "True" with "1" and "False" with "0"),
> but a more permanent fix would be favorable.  Would this be a simple
> change in the setup.py script?  We admit to knowing little/nothing about
> Python scripting.  Thanks.
> 
>         -Sean