# Copyright(c) 1986 Association of Universities for Research in Astronomy Inc. include include include .help dtoc3 .nf ___________________________________________________________________________ This routine is based on the TOOLS routine of the same name by J. Chong. The major changes are translation to the IRAF pp language, implementation of the "G" format, and some restructuring to avoid duplicated code. This procedure is called by the more general routine DTOC, which adds the additional format type "H" (sexageximal, or hms format). Formats: f Fixed format. D == number of decimal places. If D is -1 (or less), the "." will not be printed. If the number being printed is too large to fit in F format, "E" format will be used instead. Nonsignificant digits are returned as zeros. e Exponential format. D == number of significant digits. g General format. D == number of significant digits. The actual format used may be either F or E, depending on which is smaller. If D is negative, the "." will be omitted if possible (abs(D) remains the precision). If the field width is too small to convert the number, the field will be filled with asterisks instead. If the number being converted is INDEF, the string "INDEF" will be returned. The number is rounded to the desired precision before being printed. .endhelp ______________________________________________________________________ # DTOC3 --- Convert double precision real to string. int procedure dtoc3 (val, out, maxch, decpl, a_fmt, width) double val # value to be encoded char out[ARB] # output string int maxch # max chars out int decpl # precision int a_fmt # type of encoding ('f', 'e', etc.) int width # field width double v char digits[MAX_DIGITS] bool neg, small, exp_format, squeeze int i, w, d, e, j, len, no_digits, max_size, e_size, f_size, fmt int itoc(), gstrcpy() begin # Set flags indicating whether the number is greater or less that zero, # and whether its absolute value is greater or less than 1. v = abs (val) w = abs (width) fmt = a_fmt if (IS_UPPER(a_fmt)) fmt = TO_LOWER(a_fmt) squeeze = (fmt == FMT_GENERAL) neg = (val < 0.0) small = (v < 0.1) if (squeeze) d = abs (decpl) else d = max (0, decpl) if (IS_INDEFD (val)) # INDEF is a special case return (gstrcpy ("INDEF", out, w)) # Scale number to 0.1 <= v < 1.0 call dtcscl (v, e, 1) # Start tally for the maximum size of the number to determine if an # error should be returned. if (neg) # 1 for neg, plus 1 for . max_size = 2 else max_size = 1 no_digits = min (MAX_DIGITS, d) # Determine exact format for printing. len = abs (e) # base size of E format e_size = 1 for (i=10; i <= 10000; i=i*10) { if (len < i) break e_size = e_size + 1 } e_size = e_size + max_size + 1 if (e < 0) e_size = e_size + 1 # allow space for leading '0' if (squeeze) { # G-format: find best format e_size = e_size + d if (e > 0) f_size = max (d, e + 1) else f_size = d - e f_size = f_size + max_size } else if (fmt == FMT_FIXED) f_size = max (e, 1) + max_size + d if (squeeze) { # 'G' format if (f_size <= e_size) { exp_format = false if (e > 0) no_digits = f_size - max_size max_size = f_size } else { exp_format = true max_size = e_size } d = w # deactivate dec-places count } else if (fmt == FMT_FIXED) { # Fortran 'F' format exp_format = f_size > w if (exp_format) { # is there too little space? no_digits = max (1, w - e_size) max_size = no_digits + e_size } else { no_digits = e + d + 1 # negative e is OK here max_size = f_size } } else { # Fortran 'E' format exp_format = true max_size = e_size + d d = w } # Round the number at digit (no_digits + 1). if (no_digits >= 0) v = v + 0.5 * 10. ** (-no_digits) # Be sure the number of digits is in range. no_digits = max(1, min(MAX_DIGITS, no_digits)) # Handle the unusual situation of rounding from .999.. up to 1.000. if (v >= 1.0) { v = v / 10.0 e = e + 1 if (!exp_format) { max_size = max_size + 1 no_digits = min (MAX_DIGITS, no_digits + 1) } } # See if the number will fit in 'w' characters if (max_size > w) { for (i=1; i <= w; i=i+1) out[i] = OVFL_CHAR out[i] = EOS return (w) } # Extract the first digits. At the start V is normalized # to a value less than 1.0. The algorithm is to multiply by ten and # take the integer part to get each digit. do i = 1, no_digits { v = v * 10.0d0 j = int (v + EPSILOND) # truncate to integer v = v - j # lop off the integral part # Make sure the next iteration will produce a decimal J in the # range 1-9. On some systems, due to precision problems J=int(V) # can be off by one compared to V-J and this will result in a J # of 10 in the next iteration. The expression below attempts to # look ahead for J>=10 and adjusts the J for the current iteration # up by one if this will occur. if (int (v * 10.0d0 + EPSILOND) >= 10) { j = j + 1 v = v - 1 if (v < 0) v = 0 } digits[i] = TO_DIGIT(j) } # Take digit string and exponent and arrange into desired format. len = 1 if (neg) { out[1] = '-' len = len + 1 } if (exp_format) { # set up exponential format out[len] = digits[1] out[len+1] = '.' len = len + 2 for (i=2; i <= no_digits; i=i+1) { out[len] = digits[i] len = len + 1 } out[len] = 'E' len = len + 1 if (e < 0) { out[len] = '-' len = len + 1 e = -e } len = len + itoc (e, out[len], w - len + 1) } else if (e >= no_digits) { # Handle numbers >= 1 with dp after figures. for (i=1; i <= no_digits; i=i+1) { out[len] = digits[i] len = len + 1 } for (i=no_digits; i <= e; i=i+1) { out[len] = '0' len = len + 1 } if (decpl > 0) { out[len] = '.' len = len + 1 if (!squeeze) { for (i=1; i <= d; i=i+1) { out[len] = '0' len = len + 1 } } } } else { if (e < 0) { # Handle fixed numbers < 1. if (d == 0 && e == -1 && digits[1] >= '5') out[len] = '1' else out[len] = '0' out[len + 1] = '.' len = len + 2 for (i=1; i < -e && d > 0; i=i+1) { out[len] = '0' len = len + 1 d = d - 1 } i = 1 } else { # Handle numbers > 1 with dp inside figures. e = e + 2 # one more zero below for (i=1; i < e; i=i+1) { out[len] = digits[i] len = len + 1 } if (decpl > 0) { out[len] = '.' len = len + 1 } } for (j=1; i <= no_digits && j <= d; j=j+1) { out[len] = digits[i] i = i + 1 len = len + 1 } if (squeeze) { while (len > 2) { # skip trailing zeroes len = len - 1 if (out[len] != '0') { len = len + 1 # non-digit -- keep it break } } } else { for (i=1; i < d+e-no_digits && i <= d; i=i+1) { out[len] = '0' len = len + 1 } } } out[len] = EOS return (len - 1) end