[iaufwg] FITS panel recommendations; ready for review?

[William Thompson]

Item #6 recommends that keyword order be preserved in an HDU.  The software most 
commonly used within the solar physics community routinely violates this 
recommendation, because of the conversion from string header to data structure. 
  I already explained this in part in the fitsbits list.  I'll make a few 
additional comments:

The order of COMMENT and HISTORY keywords are maintained within each subgroup, 
but not as how they relate to the FITS header as a whole.  All the COMMENT lines 
are gathered together into one array, and all the HISTORY lines are gathered in 
another array.  For example, if a COMMENT line follows a certain keyword, and is 
meant as a comment on that keyword, that association is lost.

Lines with all spaces in the keyword field, and any in-line commentation after 
any keyword value, are lost.

Bill Thompson

-- 
William Thompson
NASA Goddard Space Flight Center
Code 671
Greenbelt, MD  20771
USA

301-286-2040
William.T.Thompson.1 at gsfc.nasa.gov