Please enjoy using my FORTRAN library. It contains a whole load of random functions and subroutines that I have written over the years. Most of them are overloaded to allow their use by different kinds of arguments. To save repetitive typing you will find some Python scripts dotted around this repository which have been written to create all the different versions of each FORTRAN function or FORTRAN subroutine.
I find the GFortran Intrinsic Procedures Reference an excellent resource when determining what kind an argument must be. As an aside, it niggles me that a lot of intrinsic procedures, according to the FORTRAN standard, are "of default kind" (to me, these are the sort of problems that ISO_FORTRAN_ENV was supposed to fix).
Ditto, the MPI 3.1 Specification PDF and OpenMP 5.0 Specification PDF are excellent resources too.
All integers are declared as INT64 (from ISO_FORTRAN_ENV), except:
- when being saved to disk (depending on the situation); and
- when interacting with an intrinsic procedure (depending on the procedure's specification).
All reals are declared as REAL64 (from ISO_FORTRAN_ENV), except:
- when being saved to disk (depending on the situation); and
- when interacting with an intrinsic procedure (depending on the procedure's specification).
GOTO statements are never used because I do not want to be eaten by a velociraptor.
- mod_safe is a module that can safely be used (without any extra dependencies) containing:
- const_cm are constants containing colour maps
- const_factorial is a constant containing the answers to
x! - func_degrees is a function to convert from radians to degrees
- func_integrate_array is a function to integrate tabulated data
- func_interpolate_points is a function to interpolate between points
- func_overall_index is a function to return the overall index of an iteration inside many nested
DOloops - func_radians is a function to convert from degrees to radians
- sub_allocate_array is a subroutine to allocate an array after checking that the requested size makes sense
- sub_calc_loc_from_loc_and_bearing_and_dist is a subroutine which implements Vincenty's formula to calculate the finishing location from a starting location, a bearing and a distance
- sub_load_array_from_BIN is a subroutine to populate an array with data from a binary file
- sub_save_array_as_BIN is a subroutine to save an array's data to a binary file
- sub_save_array_as_PBM is a subroutine to save an array's data to a portable bit-map format (PBM) image file
- sub_save_array_as_PGM is a subroutine to save an array's data to a portable grey-map format (PGM) image file
- sub_save_array_as_PPM is a subroutine to save an array's data to a portable pix-map format (PPM) image file
- sub_solve_quadratic_equation is a subroutine to solve a quadratic equation of the form "ax2 + bx + c = 0"
- mod_safe_mpi is a module that can safely be used (without any extra dependencies) by a MPI compiler containing:
- sub_allreduce_array is a subroutine to reduce arrays that are larger than MPI's in-built limits
- sub_bcast_array is a subroutine to broadcast arrays that are larger than MPI's in-built limits
Additionally, mod_safe contains some other constants, such as const_c and const_kb. A full list can be found in consts.f90.
- This library is a consolidation of a whole load of functions and subroutines from a variety of different projects. It will take some time to pull them all in and make sure that they all work well together.
The directory tests contains some basic tests to (hopefully):
- find any simple bugs that I might have introduced; and
- demonstrate some simple programming techniques.
- There is a bug in Doxygen which results in FORTRAN INCLUDE statements not being parsed. Until this gets fixed then there is little point in maintaining the Doxygen comments in the source code here. The bug was reported on 2009-07-27 and it shows no sign of being fixed (I first commented on the bug report on 2013-01-11 and as of 2019-03-31 it is still open).