[J3] branch based on ieee class ?

Vipul Parekh parekhvs at gmail.com
Thu Nov 7 12:54:19 EST 2019

On Thu, Nov 7, 2019 at 10:38 AM Dan Nagle via J3
<j3 at mailman.j3-fortran.org> wrote:
> ..
> How to?
> C1145 removes the select case option (my first choice), and
> Table 10.2 eliminates if-blocks by not listing ieee_class_type
> among the allowed operands for == and /= (or .eqv. and .neqv.).
> ..

Dan and anyone interested in worklist item US-21 toward Fortran 202X:

By the way, this diagnostic routine you're trying to write for your
workshop is a good example of a use case, I think, for an enumeration
type.  See the paper I submitted at the last J3 meeting which got "No
Action" label: https://j3-fortran.org/doc/year/19/19-229.txt

Starting Fortran 2003, as you know, IEEE facilities as well as the
limited support for C enums were introduced.  Ideally the "proper"
enumeration types would have been added in that revision itself and
the IEEE facilities could have made use of such enumeration types
rather than the current form of IEEE_CLASS_TYPE, IEEE_ROUND_TYPE, etc.
with long list of named constants with long names such as
IEEE_NEGATIVE_NORMAL and so forth.  It would have made it a cleaner
support in the language for the use of SELECT CASE construct in the
diagnostic routine you seek to write or other code by the
practitioners of Fortran.


More information about the J3 mailing list