Dear All,
As a result of updating the CF checker for CF 1.3, I have a couple of
comments/proposals about flag variables......
1. One of the requirements for a CF compliant file states (Section 3.5)
"Variables with a flag_masks attribute must have a type that is
compatible with bit field expression (char, byte, short and int), not
floating-point (float, real, double), and the flag_masks attribute must
have the same type."
Taking into account the additional recommendation that
"When flag_masks and flag_values are both defined, the Boolean AND of
each entry in flag_values with its corresponding entry in flag_masks
should equal the flag_values entry, ie, the mask selects all the bits
required to express the value."
If, for example, flag_values = "A B C D" I don't think it makes much
sense to have a flag_masks attribute.
I wonder if the convention should explicitly state that "if a flag
variable is of type char then it should not have a flag_masks attribute." ?
2. I'd also like to propose the additional 2 requirements
** "If the flag_values attribute is present then the flag_meanings
attribute must be specified"
Having flag_values specified and attaching no meanings to them doesn't
really make much sense.
** "The type of the flag_meanings attribute is a string whose value is a
blank separated list of words or phrases (words connected by underscores)"
This is implied in the convention but is not explicitly stated in the
conformance document.
Regards,
Ros.
--
------------------------------------------------------------------------
Rosalyn Hatcher
NCAS Computational Modelling Services
Dept. of Meteorology, University of Reading,
Earley Gate, Reading. RG6 6BB
Email: r.s.hatcher at reading.ac.uk Tel: +44 (0) 118 378 6016
Received on Wed Nov 26 2008 - 02:44:35 GMT
