(j3.2006) Fwd: BOZ incompatiability in Fortran standards

[Bill Long]

> On Oct 6, 2017, at 5:45 PM, Van Snyder <Van.Snyder at jpl.nasa.gov> wrote:
> 
> On Fri, 2017-10-06 at 21:29 +0000, Bill Long wrote:
>> (Why haven?t we made DATA obsolescent, yet??)
> 
> Because it's a pain in the tusch to make a 600-line named-constant
> definition, no matter whether you're creating it as the ultimate
> objective or as the initializer for a variable, but DATA statements can
> do it a little bit at a time.

You can make a named constant with a DATA statement ?? 

For saved variables, you can declare the variable with the SAVE attribute and initialize sections of the object with a series of assignment statements, either in the main program or in a ?init? routine that you call only once. 

But if you really have an array initialization that takes 600 lines of code, I?d first suspect that the code design needs some work. 

> 
> To make a really large array named constant, you need to invent names
> for sub-arrays that will fit within the maximum statement size limits,
> and then combine those into bigger (sub) arrays, and ... and ultimately
> create your array.

That would work.  But does not involve DATA. 
> 
> I did propose during the F03 development that the statement line limit
> ought to be 9999, but that was shot down.

Vendor fatigue at constantly changing the limit, combined with some abusive user codes, have lead some implementations to give up and just allow unlimited continuations lines. 

Cheers,
Bill


> 
> 
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3

Bill Long                                                                       longb at cray.com
Principal Engineer, Fortran Technical Support &   voice:  651-605-9024
Bioinformatics Software Development                      fax:  651-605-9143
Cray Inc./ 2131 Lindau Lane/  Suite 1000/  Bloomington, MN  55425