some errors in analyitical.f90 when compiling build.script

[leo_lee]

Hi, ROMS users:
I'm a novice in learning ROMS. Recently I follow the example of Coriolis in the ROMS Tutorials, but when I compile the build.script, there are some errors in the analytical.f90, the errors are shown like these:

analytical.f90(893): error #6052: This label has not been defined as a FORMAT label. [10]
WRITE (stdout,10) TRIM(Vname(2,idUbar))//': '// &
----------------------^
analytical.f90(900): error #6052: This label has not been defined as a FORMAT label. [10]
WRITE (stdout,10) TRIM(Vname(2,idVbar))//': '// &
----------------------^
analytical.f90(920): error #6052: This label has not been defined as a FORMAT label. [10]
WRITE (stdout,10) TRIM(Vname(2,idFsur))//': '// &
----------------------^
analytical.f90(949): error #6052: This label has not been defined as a FORMAT label. [10]
WRITE (stdout,10) TRIM(Vname(2,idUvel))//': '// &
----------------------^
analytical.f90(956): error #6052: This label has not been defined as a FORMAT label. [10]
WRITE (stdout,10) TRIM(Vname(2,idVvel))//': '// &
----------------------^
analytical.f90(973): error #6404: This name does not have a type, and must have an explicit type. [TEMP]
t(i,j,k,2,itemp)=t(i,j,k,1,temp)
---------------------------------------^
analytical.f90(971): error #6321: An unterminated block exists.
DO i=IstrT,IendT
^
analytical.f90(970): error #6321: An unterminated block exists.
DO j=JstrT,JendT
^
analytical.f90(969): error #6321: An unterminated block exists.
DO k=1,N(ng)
^
analytical.f90(893): error #6323: This label is not defined in this scoping unit. [10]
WRITE (stdout,10) TRIM(Vname(2,idUbar))//': '// &
----------------------^
analytical.f90(982): error #6911: The syntax of this substring is invalid. [T]
& LBi, UBi, LBj, UBj, 1, N(ng), t(:,:,:,1,itrc))
--------------------------------------------------------^
analytical.f90(980): warning #5427: Program may contain only one main entry routine
DO itrc=1,NAT
------^
analytical.f90(1376): error #6790: This is an invalid statement; an END [PROGRAM] statement is required.
END MODULE analytical_mod
------^
analytical.f90(1376): error #6785: This name does not match the unit name. [ANALYTICAL_MOD]
END MODULE analytical_mod

------^

I don't know how to deal with these errors in the analytical.f90, can anyone help me solve these? Thanks so much !

[kate]

The thing about analytical.f90 is that I have no idea what's in yours. It is created by running analytical.F through the C preprocessor and the results are wildly different for each application. I am unfamiliar with the Coriolis problem and it's not one of the supported ones the ROMS/Include directory. As ROMS evolves, it's best to stick with a supported test case so you know you have a consistent set of files. Once you get one or three of those running, you might learn a lot by figuring out for yourself why the Coriolis files don't compile - starting by looking at analytical.f90. "Use the source, Luke."

[leo_lee]

The thing about analytical.f90 is that I have no idea what's in yours. It is created by running analytical.F through the C preprocessor and the results are wildly different for each application. I am unfamiliar with the Coriolis problem and it's not one of the supported ones the ROMS/Include directory. As ROMS evolves, it's best to stick with a supported test case so you know you have a consistent set of files. Once you get one or three of those running, you might learn a lot by figuring out for yourself why the Coriolis files don't compile - starting by looking at analytical.f90. "Use the source, Luke."

Hi, Kate:
Thanks a lot for your suggestions in learning ROMS, I found out that these errors were due to my mistake syntax rules in adding the ana_initial.h for Coriolis case. I just followed the upwelling example, so i just want to take coriolis example as a simple self-defined case to compile and run ROMS. Since i corrected the syntax, i got the outputs of ocean_his.nc and ocean_rst.nc. However i dind't get the output of ocean_flt.nc, even though i put the float.in in my project directory. Should i input float.in to oceanM solely just like coriolis.in or there are any other solutions?

[kate]

You need to #define FLOATS as well.