chfev.f 4.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155
  1. *DECK CHFEV
  2. SUBROUTINE CHFEV (X1, X2, F1, F2, D1, D2, NE, XE, FE, NEXT, IERR)
  3. C***BEGIN PROLOGUE CHFEV
  4. C***PURPOSE Evaluate a cubic polynomial given in Hermite form at an
  5. C array of points. While designed for use by PCHFE, it may
  6. C be useful directly as an evaluator for a piecewise cubic
  7. C Hermite function in applications, such as graphing, where
  8. C the interval is known in advance.
  9. C***LIBRARY SLATEC (PCHIP)
  10. C***CATEGORY E3
  11. C***TYPE SINGLE PRECISION (CHFEV-S, DCHFEV-D)
  12. C***KEYWORDS CUBIC HERMITE EVALUATION, CUBIC POLYNOMIAL EVALUATION,
  13. C PCHIP
  14. C***AUTHOR Fritsch, F. N., (LLNL)
  15. C Lawrence Livermore National Laboratory
  16. C P.O. Box 808 (L-316)
  17. C Livermore, CA 94550
  18. C FTS 532-4275, (510) 422-4275
  19. C***DESCRIPTION
  20. C
  21. C CHFEV: Cubic Hermite Function EValuator
  22. C
  23. C Evaluates the cubic polynomial determined by function values
  24. C F1,F2 and derivatives D1,D2 on interval (X1,X2) at the points
  25. C XE(J), J=1(1)NE.
  26. C
  27. C ----------------------------------------------------------------------
  28. C
  29. C Calling sequence:
  30. C
  31. C INTEGER NE, NEXT(2), IERR
  32. C REAL X1, X2, F1, F2, D1, D2, XE(NE), FE(NE)
  33. C
  34. C CALL CHFEV (X1,X2, F1,F2, D1,D2, NE, XE, FE, NEXT, IERR)
  35. C
  36. C Parameters:
  37. C
  38. C X1,X2 -- (input) endpoints of interval of definition of cubic.
  39. C (Error return if X1.EQ.X2 .)
  40. C
  41. C F1,F2 -- (input) values of function at X1 and X2, respectively.
  42. C
  43. C D1,D2 -- (input) values of derivative at X1 and X2, respectively.
  44. C
  45. C NE -- (input) number of evaluation points. (Error return if
  46. C NE.LT.1 .)
  47. C
  48. C XE -- (input) real array of points at which the function is to be
  49. C evaluated. If any of the XE are outside the interval
  50. C [X1,X2], a warning error is returned in NEXT.
  51. C
  52. C FE -- (output) real array of values of the cubic function defined
  53. C by X1,X2, F1,F2, D1,D2 at the points XE.
  54. C
  55. C NEXT -- (output) integer array indicating number of extrapolation
  56. C points:
  57. C NEXT(1) = number of evaluation points to left of interval.
  58. C NEXT(2) = number of evaluation points to right of interval.
  59. C
  60. C IERR -- (output) error flag.
  61. C Normal return:
  62. C IERR = 0 (no errors).
  63. C "Recoverable" errors:
  64. C IERR = -1 if NE.LT.1 .
  65. C IERR = -2 if X1.EQ.X2 .
  66. C (The FE-array has not been changed in either case.)
  67. C
  68. C***REFERENCES (NONE)
  69. C***ROUTINES CALLED XERMSG
  70. C***REVISION HISTORY (YYMMDD)
  71. C 811019 DATE WRITTEN
  72. C 820803 Minor cosmetic changes for release 1.
  73. C 890411 Added SAVE statements (Vers. 3.2).
  74. C 890531 Changed all specific intrinsics to generic. (WRB)
  75. C 890703 Corrected category record. (WRB)
  76. C 890703 REVISION DATE from Version 3.2
  77. C 891214 Prologue converted to Version 4.0 format. (BAB)
  78. C 900315 CALLs to XERROR changed to CALLs to XERMSG. (THJ)
  79. C***END PROLOGUE CHFEV
  80. C Programming notes:
  81. C
  82. C To produce a double precision version, simply:
  83. C a. Change CHFEV to DCHFEV wherever it occurs,
  84. C b. Change the real declaration to double precision, and
  85. C c. Change the constant ZERO to double precision.
  86. C
  87. C DECLARE ARGUMENTS.
  88. C
  89. INTEGER NE, NEXT(2), IERR
  90. REAL X1, X2, F1, F2, D1, D2, XE(*), FE(*)
  91. C
  92. C DECLARE LOCAL VARIABLES.
  93. C
  94. INTEGER I
  95. REAL C2, C3, DEL1, DEL2, DELTA, H, X, XMI, XMA, ZERO
  96. SAVE ZERO
  97. DATA ZERO /0./
  98. C
  99. C VALIDITY-CHECK ARGUMENTS.
  100. C
  101. C***FIRST EXECUTABLE STATEMENT CHFEV
  102. IF (NE .LT. 1) GO TO 5001
  103. H = X2 - X1
  104. IF (H .EQ. ZERO) GO TO 5002
  105. C
  106. C INITIALIZE.
  107. C
  108. IERR = 0
  109. NEXT(1) = 0
  110. NEXT(2) = 0
  111. XMI = MIN(ZERO, H)
  112. XMA = MAX(ZERO, H)
  113. C
  114. C COMPUTE CUBIC COEFFICIENTS (EXPANDED ABOUT X1).
  115. C
  116. DELTA = (F2 - F1)/H
  117. DEL1 = (D1 - DELTA)/H
  118. DEL2 = (D2 - DELTA)/H
  119. C (DELTA IS NO LONGER NEEDED.)
  120. C2 = -(DEL1+DEL1 + DEL2)
  121. C3 = (DEL1 + DEL2)/H
  122. C (H, DEL1 AND DEL2 ARE NO LONGER NEEDED.)
  123. C
  124. C EVALUATION LOOP.
  125. C
  126. DO 500 I = 1, NE
  127. X = XE(I) - X1
  128. FE(I) = F1 + X*(D1 + X*(C2 + X*C3))
  129. C COUNT EXTRAPOLATION POINTS.
  130. IF ( X.LT.XMI ) NEXT(1) = NEXT(1) + 1
  131. IF ( X.GT.XMA ) NEXT(2) = NEXT(2) + 1
  132. C (NOTE REDUNDANCY--IF EITHER CONDITION IS TRUE, OTHER IS FALSE.)
  133. 500 CONTINUE
  134. C
  135. C NORMAL RETURN.
  136. C
  137. RETURN
  138. C
  139. C ERROR RETURNS.
  140. C
  141. 5001 CONTINUE
  142. C NE.LT.1 RETURN.
  143. IERR = -1
  144. CALL XERMSG ('SLATEC', 'CHFEV',
  145. + 'NUMBER OF EVALUATION POINTS LESS THAN ONE', IERR, 1)
  146. RETURN
  147. C
  148. 5002 CONTINUE
  149. C X1.EQ.X2 RETURN.
  150. IERR = -2
  151. CALL XERMSG ('SLATEC', 'CHFEV', 'INTERVAL ENDPOINTS EQUAL', IERR,
  152. + 1)
  153. RETURN
  154. C------------- LAST LINE OF CHFEV FOLLOWS ------------------------------
  155. END