> John Bray writes:
>
> Remember the old chestnut of wave recompilation when USEing
> subroutines contained in F90 modules. I need to revisit the problem,
> and have come up with a possible make based solution that relies on
> automatic generation of argument information from f90 source to
> produce seperate .interface files that can be INCLUDEd in calling
> routines.
>
> Before I fiddle around writing the Perl to do the generation of the
> interface files, I wondered if anyone else knew of a tool that would
> parse code and extract subroutine arguments and their associated
> declarations to a separate file. <snip>
>
> John
> --
> John Bray, Numerical Weather Prediction Tel: +44 (0) 1344 854035
> Room 337 [log in to unmask]
> The Met. Office http://www.met-office.gov.uk
> London Road, Bracknell, RG12 2SZ, UK http://www.jrbray.org.uk
>
The NAG F95 tools do an excellent job of this as well as many other
things including precision standardization, polishing of F95 code,
reading code and inserting ONLY clauses after USE, changing variable
names, generating call graphs and perhaps other functions that I have
forgotten. My group develops a lot of F95 code and uses these
tools extensively.
URLS for nag are: (UK)
www.nag.co.uk
(US)
www.nag.com
(I have no financial interest in NAG Ltd or NAG Inc.)
Barry W. Brown
Department of Biomathematics,
Box 237
University of Texas M. D.
Anderson Cancer Center
1515 Holcombe Blvd
Houston, TX 77030
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|