1      subroutine profil(iopt,tx,nx,ty,ny,c,kx,ky,u,nu,cu,ier)
2c  if iopt=0 subroutine profil calculates the b-spline coefficients of
3c  the univariate spline f(y) = s(u,y) with s(x,y) a bivariate spline of
4c  degrees kx and ky, given in the b-spline representation.
5c  if iopt = 1 it calculates the b-spline coefficients of the univariate
6c  spline g(x) = s(x,u)
7c
8c  calling sequence:
9c     call profil(iopt,tx,nx,ty,ny,c,kx,ky,u,nu,cu,ier)
10c
11c  input parameters:
12c   iopt  : integer flag, specifying whether the profile f(y) (iopt=0)
13c           or the profile g(x) (iopt=1) must be determined.
14c   tx    : real array, length nx, which contains the position of the
15c           knots in the x-direction.
16c   nx    : integer, giving the total number of knots in the x-direction
17c   ty    : real array, length ny, which contains the position of the
18c           knots in the y-direction.
19c   ny    : integer, giving the total number of knots in the y-direction
20c   c     : real array, length (nx-kx-1)*(ny-ky-1), which contains the
21c           b-spline coefficients.
22c   kx,ky : integer values, giving the degrees of the spline.
23c   u     : real value, specifying the requested profile.
24c           tx(kx+1)<=u<=tx(nx-kx), if iopt=0.
25c           ty(ky+1)<=u<=ty(ny-ky), if iopt=1.
26c   nu    : on entry nu must specify the dimension of the array cu.
27c           nu >= ny if iopt=0, nu >= nx if iopt=1.
28c
29c  output parameters:
30c   cu    : real array of dimension (nu).
31c           on successful exit this array contains the b-spline
32c   ier   : integer error flag
33c    ier=0 : normal return
34c    ier=10: invalid input data (see restrictions)
35c
36c  restrictions:
37c   if iopt=0 : tx(kx+1) <= u <= tx(nx-kx), nu >=ny.
38c   if iopt=1 : ty(ky+1) <= u <= ty(ny-ky), nu >=nx.
39c
40c  other subroutines required:
41c    fpbspl
42c
43c  author :
44c    p.dierckx
45c    dept. computer science, k.u.leuven
46c    celestijnenlaan 200a, b-3001 heverlee, belgium.
47c    e-mail : Paul.Dierckx@cs.kuleuven.ac.be
48c
49c  latest update : march 1987
50c
51c  ..scalar arguments..
52      integer iopt,nx,ny,kx,ky,nu,ier
53      real*8 u
54c  ..array arguments..
55      real*8 tx(nx),ty(ny),c((nx-kx-1)*(ny-ky-1)),cu(nu)
56c  ..local scalars..
57      integer i,j,kx1,ky1,l,l1,m,m0,nkx1,nky1
58      real*8 sum
59c  ..local array
60      real*8 h(6)
61c  ..
62c  before starting computations a data check is made. if the input data
63c  are invalid control is immediately repassed to the calling program.
64      kx1 = kx+1
65      ky1 = ky+1
66      nkx1 = nx-kx1
67      nky1 = ny-ky1
68      ier = 10
69      if(iopt.ne.0) go to 200
70      if(nu.lt.ny) go to 300
71      if(u.lt.tx(kx1) .or. u.gt.tx(nkx1+1)) go to 300
72c  the b-splinecoefficients of f(y) = s(u,y).
73      ier = 0
74      l = kx1
75      l1 = l+1
76 110  if(u.lt.tx(l1) .or. l.eq.nkx1) go to 120
77      l = l1
78      l1 = l+1
79      go to 110
80 120  call fpbspl(tx,nx,kx,u,l,h)
81      m0 = (l-kx1)*nky1+1
82      do 140 i=1,nky1
83        m = m0
84        sum = 0.
85        do 130 j=1,kx1
86          sum = sum+h(j)*c(m)
87          m = m+nky1
88 130    continue
89        cu(i) = sum
90        m0 = m0+1
91 140  continue
92      go to 300
93 200  if(nu.lt.nx) go to 300
94      if(u.lt.ty(ky1) .or. u.gt.ty(nky1+1)) go to 300
95c  the b-splinecoefficients of g(x) = s(x,u).
96      ier = 0
97      l = ky1
98      l1 = l+1
99 210  if(u.lt.ty(l1) .or. l.eq.nky1) go to 220
100      l = l1
101      l1 = l+1
102      go to 210
103 220  call fpbspl(ty,ny,ky,u,l,h)
104      m0 = l-ky
105      do 240 i=1,nkx1
106        m = m0
107        sum = 0.
108        do 230 j=1,ky1
109          sum = sum+h(j)*c(m)
110          m = m+1
111 230    continue
112        cu(i) = sum
113        m0 = m0+nky1
114 240  continue
115 300  return
116      end
117
118