SLA_DFLTIN - Decode a Double Precision Number   

ACTION:

Convert free-format input into double precision floating point.

CALL:

CALL sla_DFLTIN (STRING, NSTRT, DRESLT, JFLAG)

GIVEN:

STRING string containing number to be decoded

NSTRT I pointer to where decoding is to commence

DRESLT D current value of result

RETURNED:

NSTRT advanced to next number

DRESLT D result

JFLAG I status: -1 = -OK, 0 = +OK, 1 = null result, 2 = error

NOTES:

1.

The reason sla_DFLTIN has separate `OK' status values for + and - is to enable minus zero to be detected. This is of crucial importance when decoding mixed-radix numbers. For example, an angle expressed as degrees, arcminutes and arcseconds may have a leading minus sign but a zero degrees field.

2.

A TAB is interpreted as a space, and lowercase characters are interpreted as uppercase. n.b. The test for TAB is ASCII-specific.

3.

The basic format is the sequence of fields $\pm n.n x \pm n$, where $\pm$ is a sign character `+' or `-', n means a string of decimal digits, `.' is a decimal point, and x, which indicates an exponent, means `D' or `E'. Various combinations of these fields can be omitted, and embedded blanks are permissible in certain places.

4.

Spaces:

• Leading spaces are ignored.
• Embedded spaces are allowed only after +, -, D or E, and after the decimal point if the first sequence of digits is absent.
• Trailing spaces are ignored; the first signifies end of decoding and subsequent ones are skipped.

5.

Delimiters:

• Any character other than +,-,0-9,.,D,E or space may be used to signal the end of the number and terminate decoding.
• Comma is recognized by sla_DFLTIN as a special case; it is skipped, leaving the pointer on the next character. See 13, below.
• Decoding will in all cases terminate if end of string is reached.

6.

Both signs are optional. The default is +.

7.

The mantissa n.n defaults to unity.

8.

The exponent $x\!\pm\!n$ defaults to `D0'.

9.

The strings of decimal digits may be of any length.

10.

The decimal point is optional for whole numbers.

11.

A null result occurs when the string of characters being decoded does not begin with +,-,0-9,.,D or E, or consists entirely of spaces. When this condition is detected, JFLAG is set to 1 and DRESLT is left untouched.

12.

NSTRT = 1 for the first character in the string.

13.

On return from sla_DFLTIN, NSTRT is set ready for the next decode - following trailing blanks and any comma. If a delimiter other than comma is being used, NSTRT must be incremented before the next call to sla_DFLTIN, otherwise all subsequent calls will return a null result.

14.

Errors (JFLAG=2) occur when:

• a +, -, D or E is left unsatisfied; or
• the decimal point is present without at least one decimal digit before or after it; or
• an exponent more than 100 has been presented.

15.

When an error has been detected, NSTRT is left pointing to the character following the last one used before the error came to light. This may be after the point at which a more sophisticated program could have detected the error. For example, sla_DFLTIN does not detect that `1D999' is unacceptable (on a computer where this is so) until the entire number has been decoded.

16.

Certain highly unlikely combinations of mantissa and exponent can cause arithmetic faults during the decode, in some cases despite the fact that they together could be construed as a valid number.

17.

Decoding is left to right, one pass.

18.

See also sla_FLOTIN and sla_INTIN.