Re: ISO time format for tsStampToText

[Jim Thomas <thomas@cfht.hawaii.edu>]

>>>>> "Peregrine" == Peregrine M McGehee <peregrine@lanl.gov> writes:
>>>>> "Peregrine" == Maren Purves <m.purves@jach.hawaii.edu> writes:

 Peregrine> *       For any of the formats, the fractional part of the time
 Peregrine> *       (represented as `nano-secs' above) is variable length.
 Peregrine> *       If the usec and nsec part of the time are both zero,
 Peregrine> *       only milli-seconds will be printed; or else if the nsec
 Peregrine> *       part is zero, only milli-seconds and micro-seconds will
 Peregrine> *       be printed.

and

 Peregrine> TS_TEXT_ISO        36   YYYY-MM-DDThh:mm:ss.nano-secsTZD
 Peregrine> TS_TEXT_ISOFITS    20   YYYY-MM-DDThh:mm:ss

1)  Why 36?  If we're only doing nano-seconds, shouldn't it be 30?

2)  The FITS standard (http://archive.stsci.edu/fits/fits_standard) does not
    limit the time fields to seconds.  From the definition for the DATE
    reserved keyword, "The decimal part of the seconds field is optional
    and may be arbitrarily long, so long as it is consistent with the rules
    for value formats (5.2)".  We happen to use hundredths of seconds.  The
    above just indicates how the string form will appear.  IMHO we should
    not have "ISOFITS" meaning seconds only.  I don't have a reasonable
    solution for Maren's problem though :-(

Jim