Data Types
interface	cfp_sort_float_int_1d
interface	sort_int_1d

Functions/Subroutines
subroutine, public	sort_float (n, d2, arr)
	todo OpenMP parallelize the routines below
subroutine, public	heap_sort_int_float (n, arr, brr)
	Heap-sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:) is the list of integers and the array brr(:) is the list of floats.
subroutine, public	sort_int_float (n, d2, arr, brr)
	Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of integers and the array brr(:,d2) is the list of floats.
subroutine, public	sort_float_int (n, d2, arr, brr)
	Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of floats and the array brr(:,d2) is the list of integers.
subroutine, public	sort_int_int (n, d2, arr, brr)
	Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of integers and the array brr(:,d2) is the list of integers.
subroutine, public	multiway_merge_float_int (nelem, reals, ints, sizes, scratch_prefix_opt)
	Merge sorted arrays.

Function/Subroutine Documentation

◆ heap_sort_int_float()

subroutine, public sort_gbl::heap_sort_int_float	(	integer	n,
		integer, dimension(:), intent(inout)	arr,
		real(kind=cfp), dimension(:), intent(inout)	brr )

Heap-sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:) is the list of integers and the array brr(:) is the list of floats.

◆ multiway_merge_float_int()

subroutine, public sort_gbl::multiway_merge_float_int	(	integer, intent(in)	nelem,
		real(cfp), dimension(:), intent(inout)	reals,
		integer, dimension(:), intent(inout)	ints,
		integer, dimension(:), intent(in)	sizes,
		character(len=*), intent(in), optional	scratch_prefix_opt )

Merge sorted arrays.

Authors: J Benda

Date: 2025

On input, this subroutine expect real and integer arrays reals and ints of length n that consist of consecutive chunks presorted (in acending way) on ints. The subroutine will make a copy of both these arrays and then perform multiway merge of the presorted chunks into the original storage.

If non-empty scratch_prefix_opt is provided on input, the copy of the arrays will be mapped to disk prefixed with the given location. Anonymous mapping (i.e., plain allocation) is used otherwise.

Parameters

[in]	nelem	Length of the arrays floats and ints.
[in,out]	reals	Array of reals.
[in,out]	ints	Array if integers.
[in]	sizes	Lengths of presorted chunks (according to ints).
[in]	scratch_prefix_opt	Optional prefix of temporary scratch files.

◆ sort_float()

subroutine, public sort_gbl::sort_float	(	integer	n,
		integer	d2,
		real(kind=cfp), dimension(:,:), intent(inout)	arr )

todo OpenMP parallelize the routines below

Sorts the first n elements of the floating point array arr(:,d2).

◆ sort_float_int()

subroutine, public sort_gbl::sort_float_int	(	integer	n,
		integer	d2,
		real(kind=cfp), dimension(:,:), intent(inout)	arr,
		integer, dimension(:,:), intent(inout)	brr )

Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of floats and the array brr(:,d2) is the list of integers.

◆ sort_int_float()

subroutine, public sort_gbl::sort_int_float	(	integer	n,
		integer	d2,
		integer, dimension(:,:), intent(inout)	arr,
		real(kind=cfp), dimension(:,:), intent(inout)	brr )

Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of integers and the array brr(:,d2) is the list of floats.

◆ sort_int_int()

subroutine, public sort_gbl::sort_int_int	(	integer	n,
		integer	d2,
		integer, dimension(:,:), intent(inout)	arr,
		integer, dimension(:,:), intent(inout)	brr )

Sorts the first n elements of the array arr and simultaneously the array brr. Array arr(:,d2) is the list of integers and the array brr(:,d2) is the list of integers.

Data Types